Volume 47, Issue 2. April-june. Pgs. 101-210.
Transcripción
Volume 47, Issue 2. April-june. Pgs. 101-210.
REVISTA de la SOCIEDAD QUÍMICA d e M É X I C O (Rev. Soc. Quím. Méx.) Páginas 101-210 Vol. 47, Núm. 2, abril-junio del 2003 Fecha de publicación: Julio 2003 REVISTA de la SOCIEDAD QUÍMICA de MÉXICO (Rev. Soc. Quím. Méx.) ISSN 0583-7693 Publicación trimestral editada y distribuida por la Sociedad Química de México, A.C., Barranca del Muerto 26 (esq. Hércules). Col. Crédito Constructor, Delegación Benito Juárez, C.P. 03940, México, D.F. Tels.: 5662-6823 y 5662-6837. Fax: 5662-6823. Editor: Guillermo Delgado Lamas (E-mail: [email protected]) Editor Técnico: Arturo Sánchez y Gándara (E-mail: [email protected]) D.R. © Sociedad Química de México, A.C. Se prohíbe la reproducción o impresión parcial o total sin la autorización por escrito del titular de los derechos. Reserva del título número 158-67 (mayo de 1967) otorgado por la Dirección General de Derechos de Autor, SEP. Certificado de licitud número 3565 y de contenido número 3867 otorgados por la Comisión Calificadora de Publicaciones y Revistas Ilustradas de la Secretaría de Gobernación. Publicación periódica. Registro número 0790 790. Características 2294 5112, autorizado por SEPOMEX, 23 de julio de 1990. Oficio número 317 Exp. 091.70/2485. Autorizada como correspondencia de segunda clase por la Dirección General de Correos con fecha 25 de agosto de 1967. Edición e impresión: S y G Editores S.A. de C.V., Calle Cuapinol 52, Col. Santo Domingo, Delegación Coyoacán, 04369 México, D.F. Tels.: 5619-5293, 5617-5610, E-mail: [email protected]. Editorial La investigación científica es una actividad de gran importancia, particularmente en los países en vías de desarrollo, debido, entre muchos otros aspectos, a su incidencia en la implementación de tecnologías propias que permiten la conservación y aprovechamiento racional de los recursos naturales en beneficio de la sociedad. La vegetación constituye un importante recurso natural en nuestro país, y existe una amplia tradición en el manejo y utilización de las especies vegetales que, en algunos casos, puede remontarse prácticamente a los inicios de nuestra era [1]. Así, desde el inicio de la institucionalización de la investigación química en México [2], se abordaron temas referentes a la química de productos naturales. Es, por lo tanto, oportuno reconocer la trayectoria de investigación del doctor Alfonso Romo de Vivar, quien constituye una figura señera en la investigación de la composición química de la vegetación de nuestro país. Este fascículo de la Revista de la Sociedad Química de México está integrado por trabajos de investigación dedicados al distinguido académico, en reconocimiento a su labor científica desarrollada durante cinco décadas. La biografía del doctor Romo de Vivar permite atestiguar numerosos cambios y vicisitudes [3]. Nació en San Francisco de los Romo, Aguascalientes, una región donde la familia Romo de Vivar tiene antecedentes centenarios. La familia emigró a Aguascalientes, donde concluyó los estudios primarios e inició la secundaria en el Instituto de Ciencias Autónomo de Aguascalientes. En 1945, parte de la familia se trasladó a la Ciudad de México, y en esta ciudad el futuro químico concluyó su educación media, en la Secundaria 15, ubicada en Tacuba, donde el doctor Humberto Estrada, de grata memoria para los profesionales de la química, impartía clases. Después de cursar la Preparatoria en el Colegio de San Ildefonso, Alfonso Romo de Vivar ingresó a la entonces Escuela Nacional de Ciencias Químicas, y concluyó los estudios de químico en 1952. En 1953 ingresó al Instituto de Química, donde desarrolló sus tesis de licenciatura y doctorado bajo la dirección del doctor Jesús Romo Armería (1922-1977), distinguido pionero de la investigación química en México, y también originario de Aguascalientes. Así, Alfonso Romo de Vivar ingresó al grupo de investigación del Instituto de Química, formado entonces por Alberto Sandoval, José Francisco Herrán, Octavio Mancera, José Luis Mateos, Javier Padilla, Fernando Walls, Jesús Romo, José Iriarte, entre otros distinguidos académicos (Foto p. 104). Sus primeros trabajos fueron sobre la química de esteroides, entre otros temas [4], dada la estrecha relación que sostenían, durante la década de 1950, el Instituto de Química y la empresa farmacéutica Syntex, que comercializaba por aquellos años, la progesterona, sintetizada a partir de la diosgenina, un producto natural obtenido a partir de la raíz del barbasco. La tesis doctoral de Alfonso Romo de Vivar versó sobre el análisis químico de la especie vegetal Helenium mexicanum, una planta que había llamado su atención desde su niñez, ya que había observado que las vacas que consumían este vegetal producían leche amarga. Efectivamente, los principios amargos aislados de este vegetal, conocido como rosilla o chapuz, fueron un grupo de substancias conocidas como lactonas sesquiterpénicas, y denominadas trivialmente como mexicaninas A, B, C..., entre otras [5]. De 1962 a 1963, el Dr. Romo de Vivar realizó una estancia con el Dr. Werner Herz, en Tallahassee, en la Universidad de Florida. Desde los sesentas, y a lo largo de tres décadas, el grupo del doctor Alfonso Romo de Vivar llevó a cabo importantes contribuciones al conocimiento de los constituyentes químicos de varios grupos de plantas, entre los cuales destacan los géneros Iva, Ambrosia, Chrysanthemum, Zaluzania, Artemisia, Zinnia, Parthenium, Yucca, Pluchea, Viguiera, Tithonia, entre otros. Así, se configuraron y consolidaron las actividades de una importante línea de investigación, iniciada por los doctores Herrán [6], Sandoval [7], Romo [8], Iriarte [9], entre otros, la cual consiste, en términos generales, en la generación de conocimiento científico mediante el estudio químico sistemático de la flora nacional. Es pertinente mencionar que actualmente, en una sociedad de economía de mercado, donde se requiere asignar precio para que algo sea valorado, la biodiversidad de nuestro país, que es considerada una de las mayores a nivel mundial, es un patrimonio que no puede ponderarse mediante criterios utilitarios, ya que aún no está completamente descrita y apenas se conocen, de manera muy fragmentaria, las estructuras moleculares de los metabolitos secundarios presentes en las diversas fuentes naturales. Sin embargo, sí conocemos los 102 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) efectos devastadores de la erosión y la degradación de la biodiversidad. Hace casi dos décadas Alfonso Romo de Vivar publicó el libro “Productos Naturales de la Flora Mexicana” [10], el cual compila parte de las investigaciones realizadas en el Instituto de Química de la UNAM, y que actualmente constituye una referencia clásica sobre el tema. Más recientemente, durante la última década, el grupo del doctor Romo de Vivar ha incidido en el estudio químico del género Senecio [11] y taxa afines [12], los cuales constituyen un grupo importante de especies vegetales de notable complejidad taxonómica. A lo largo de su carrera científica el doctor Alfonso Romo de Vivar ha recibido numerosas distinciones, tales como el Premio de la Academia de la Investigación Científica (1968), el Premio Banamex de Ciencia y Tecnología (1975); el Premio Nacional de Química Andrés Manuel del Río de la Sociedad Química de México, y el Premio Nacional de Ciencias Farmacéuticas (ambos en 1977). Es Investigador Nacional desde 1984; fue acreedor del Premio Universidad Nacional (1987); en 1990 se le otorgó el Premio Syntex-IOCD, por la Sociedad Química Americana; en 1991 recibió la distinción de Investigador Emérito de la Universidad Nacional, y el Premio Aguascalientes de Ciencia y Tecnología, y es Investigador Emérito del Sistema Nacional de Investigadores. Los profesionales de la química, los miembros de la Sociedad Química de México, colegas, amigos y alumnos nos enorgullecemos de la figura señera que representa Alfonso Romo de Vivar, y le agradecemos su papel como forjador de la joven tradición científica de nuestro país. Guillermo D. L. Referencias 1. Rius, M.; Galdeano, C. La Química Prehispánica. En: Química en México. Ayer, Hoy y Mañana. Garritz, A., Comp., Edición de la Facultad de Química. Universidad Nacional Autónoma de México. 1991. pp. 23-52. 2. Walls, F. El Instituto de Química. Inicio de la Investigación. En: Química en México. Ayer, Hoy y Mañana. Garritz, A., Comp., Edición de la Facultad de Química. Universidad Nacional Autónoma de México. 1991. pp. 109-121. 3. (a) Romo de Vivar, A. Familia Romo de Vivar: 345 años en Aguascalientes, 50 años en el Instituto de Química. Serie: Forjadores de la Ciencia en la UNAM. Coordinación de la Investigación Científica de la UNAM. 2003. pp. 7-39. (b) Delgado, G. Alfonso Romo de Vivar. Reseña Biográfica. En: Nuestros Maestros. Edición de la Universidad Nacional Autónoma de México. Tomo III. 1996. pp 17-26. 4. (a) Romo, J.; Romo de Vivar, A. J. Org. Chem. 1956, 79, 902909. (b) Romo, J.; Romo de Vivar, A. Bol. Inst. Quím. Univ. Nac. Autón. Méx. 1956, 8, 10-16. 5. (a) Romo, J.; Romo de Vivar, A. Chem. and Ind. 1959, 882. (b) Romo de Vivar, A.; Romo, J. J. Am. Chem. Soc. 1961, 83, 2326 2328. (c) Romo de Vivar, A.; Romo, J. Ciencia (Méx.) 1961, 21, 33-35. 6. Herrán, J. Anuario de la Comisión Impulsora y Coordinadora de la Investigación Científica 1943, 217-221. 7. (a) Zechmeister, L.; Sandoval, A. Science 1945, 101, 585. (b) Zechmeister, L.; Sandoval, A. Arch. Biochem. 1945, 8, 425. 8. Romo, J. Bol. Inst. Quím. Univ. Nac. Autón. Méx. 1945, 1, 67-74 9. Iriarte, J. Bol. Inst. Quím. Univ. Nac. Autón. Méx. 1945, 1, 80-87. 10. Romo de Vivar, A. Productos Naturales de la Flora Mexicana. Ed. Limusa, México. 1985. 220 pp. 11. Romo de Vivar, A.; Pérez-Castorena, A. L.; Arciniegas, A.; Villaseñor, J. L. Rec. Res. Devel. Phytochem. 2000, 4, 61-74. 12. Arciniegas, A.; Pérez-Castorena, A. L.; Reyes, S.; Contreras, J. L.; Romo de Vivar, A. J. Nat. Prod. 2003, 66, 225-229. Guillermo Delgado Lamas Editorial 103 Jacobo Gómez Lara (1935-1999), Derek H.R. Barton (1918-1998) y Alfonso Romo de Vivar (1928). Vestíbulo del Instituto de Química, UNAM, julio de 1997. 104 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Guillermo D. L. Personal del Instituto de Química en 1953. Abajo: Maya, Isaac Lerner, Jesús Reynoso, José Luis Mateos, Jesús Romo Armería, Fernando Walls, José Iriarte y Alfonso Romo de Vivar. En medio: Nemorio Reynoso, Cristina Pérez-Amador, Pascual Aguinaco y José F. Herrán (agachado). Atrás: Visitante, Armando Manjarréz, Javier Padilla, Catalina Vélez, Ana Villanueva, Harry Miller (Fundación Rockefeller) y Octavio Mancera. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 105-106 Scientific Contributions of Dr. Alfonso Romo de Vivar Nikolaus H. Fischer Department of Pharmacognosy and Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of Mississippi, University, MS 38677, USA I consider it a great honor to be asked to present the “Scientific Laudatio” of Dr. Alfonso Romo de Vivar and it is a pleasant duty to write a summary of Dr. Romo de Vivar’s accomplishments in natural products chemistry. It speaks for itself, that his scientific contributions span over a time period of nearly one half century. Alfonso can with pride look back on a highly productive scientific career with impressive accomplishments and significant contributions to organic chemistry, in general, and natural products chemistry, in particular. Alfonso was born in Mexico on April 30, 1928 in San Francisco de los Romo, Aguascalientes. He received the B.Sc. degree in Chemistry at the School of Chemistry, Universidad Nacional Autonoma de Mexico (UNAM) in Mexico City. He subsequently entered the Postgraduate Program in Organic Chemistry and completed his Ph.D. in 1959. His dissertation research was directed toward the isolation and chemistry of sesquiterpene lactones from Helenium mexicanum. Alfonso spent his whole research career at the Institute of Chemistry at UNAM, which began in 1958, and in 1992 he received the status of Professor Emeritus at the Institute of Chemistry at UNAM. Since 1994 he is also a National Emeritus Research Scientist of the National System of Scientific Research. His early research interests were shared with his mentor, the late Professor Jesús Romo Armería, the “father” of the Institute of Chemistry. Their main focus was on the chemistry of plant steroids as well as sesquiterpene lactones from Helenium mexicanum. It is most impressive that between 1956 and 1961 eight publications appeared with these two individuals as sole authors. The papers were published in highly prestigious international journals including three publications in the Journal of the American Chemical Society and two in the Journal of Organic Chemistry [1-6]. His increasing interest in sesquiterpene lactones led to a one-year sabbatical leave (1962-1963) in the laboratory of Professor Werner Herz at Florida State University in Tallahassee, Florida, USA. This highly productive collaboration led to a series of pioneering studies on the structure and chemistry of pseudoguaianolides from the genera Helenium and related taxa. Six papers [7-12] resulted from this collaboration and appeared in the Journal of the American Chemical Society [7], the Journal of Organic Chemistry [10, 11] and Tetrahedron [8, 9, 12], which again speaks for the high productivity and caliber of these publications. Dr. Romo de Vivar’s teaching career in the School of Chemistry at UNAM began in 1957. He was the mentor of about 40 undergraduate students and eleven students received the Ph.D. degree under his direction. Many of his publications on multiple structural types of natural products, ranging from sesquiterpene lactones to diterpenes, to triterpenes, were coauthored with his dedicated students. Many of them have subsequently developed their own highly successful, internationally recognized research programs at UNAM and other prestigious academic institutions. Other collaborators include his colleagues at the Institute of Chemistry and natural product chemists from academic institutions in Mexico and other countries. His high productivity in research publications continued for over three decades [13-16], with a total of about150 peerreviewed papers and reviews being published. He continues to do research and publish in top international journals on various aspects of natural products chemistry. Since the beginning of the year 2000 alone, nearly ten papers have appeared or are in press. This seems to be a good example of a chemist, who “never stops to react”. Alfonso was the recipient of a number of significant academic awards for his scientific work. In 1968, he received the Science Award of the Mexican Academy of Sciences and in 1977 was awarded the National Chemistry Award “Andres Manuel del Rio” by the Mexican Chemical Society. In 1987 he was the recipient of the National University Science Award, and in 1990, the IOCD-Syntex Award by the American Chemical Society. I wish to conclude with a personal note. My first correspondence with Dr. Romo de Vivar goes back to October 1976. At that time Alfonso informed me in a short letter, that Dr. Ronald Hartman had visited and stored our plant collection at a safe place in the Institute. In a handwritten footnote, he pointed out that a young scientist in the Institute, Dr. Leovigildo Quijano, would send his application for an advertised postdoctoral position. Leo joined my research group shortly thereafter. This was the beginning of a life-long scientific collaboration and close friendship with Leo, that my wife Helga and I cherish very much. My association and interaction with many members of the faculty in the Institute of Chemistry and the School of Chemistry continues to this date. Thank you, Alfonso. You started it all! 106 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Nikolaus H. Fischer References 1. Some experiments with 16β-bromo-17α-acetoxy-20-keto steroids. Synthesis of 16α-17α-dihydroxysteroids and related compounds. Romo, J.; Romo de Vivar, A. J. Org. Chem., 1956, 21, 902-909. 2. The Favorskii rearrangement in the pregnane series. cis-trans Isomerism in some 17,20-dehydro derivatives. Romo, J.; Romo de Vivar, A. J. Am. Chem. Soc., 1957, 79, 1118-1123. 3. Constituents of Helenium mexicanum H.B.K. Romo de Vivar, A.; Romo, J. Chem. and Ind. 1959, 882. 4. The Beckmann rearrangements of the acetoxime of 5,16-pregnadien-3β-ol-20-one acetate with boron trifluoride. Romo, J.; Romo de Vivar, A. J. Am. Chem. Soc. 1959, 81, 3446-3452. 5. Mexicanin E, a norsesquiterpenoid lactone. Romo de Vivar, A.; Romo, J. J. Am. Chem. Soc., 1961, 83, 2326-2328. 6. Las lactonas de Helenium mexicanum H.B.K. Romo de Vivar, A.; Romo, J. Ciencia (Méx.), 1961, 21, 33-35. 7. Constituents of Helenium species, XIII. The structure of helenalin and mexicanin A. Herz, W.; Romo de Vivar, A.; Romo, J.; Viswanathan, N. J. Am. Chem. Soc., 1963, 85, 19-26. 8. Constituents of Helenium species, XV. The structure of mexicanin C. Relative stereochemistry of its congeners. Herz, W.; Romo de Vivar, A.; Romo, J.; Viswanathan, N. Tetrahedron, 1963, 19, 1359-1369. 9. Constituents of Helenium species, XIV. The structure of mexicanin E. Romo, J.; Romo de Vivar, A.; Herz, W. Tetrahedron, 1963, 19, 2717-2322. 10. Constituents of Iva species, III. Structure of microcephalin, a new sesquiterpenic lactone. Herz, W.; Hogenauer, G.; Romo de Vivar, A. J. Org. Chem., 1964, 29, 1700-1703. 11. Constituents of Iva species, IV. Structure of pseudoivalin, a new guaianolide. Herz, W.; Romo de Vivar, A.; Lakshmikantham, M. V. J. Org. Chem., 1965, 30, 118. 12. Further transformations of Helenalin and its congeners: the 1-epihelenalin and 1-epiambrosin series. Romo de Vivar, A.; Rodriguez-Hahn, L.; Lakshmikantham, M.V.; Mirrington, R.N.; Kagan, J.; Herz, W. Tetrahedron, 1966, 22, 3279. 13. The constituents of Zalazania augusta. The structure of zaluzanin A and B. Romo, J.; Romo de Vivar, A.; Joseph-Nathan, P. Tetrahedron 1967, 23, 29. 14. The constituents of Zalazania H. Structure of zaluzanins C and D. Romo de Vivar, A.; Cabrera, A.; Ortega A.; Romo, J. Tetrahedron 1967, 23, 3903. 15. Stevin, a new pseudoguaianolide isolated from Stevia rhombifolia H.B.K. Rios, T.; Romo de Vivar, A.; Romo, J. Tetrahedron, 1967, 23, 4265. 16. The Pseudoguaianolides. Romo, J.; Romo de Vivar, A. Fortschritte der Chemie Org. Naturstoffe 1967, 25, 90-130. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 107 El aporte a la Química Iberoamericana del doctor Alfonso Romo de Vivar. Asegurar la inmortalidad de la ciencia para así perpetuar la humanidad Mario Silva, Ph.D. (London) Laboratorio de Química de Productos Naturales, Departamento de Botánica, Facultad de Ciencias, Universidad de Concepción, Chile. Concepción, Chile. E-mail: [email protected] El doctor Alfonso Romo de Vivar, hombre de ciencia e investigador emérito, pero también formador de hombres de ciencia e investigadores, ha sido un auténtico líder en el ideal del siglo XX: un científico que labora para encaminar al mundo y a los hombres hacia la verdad, que no es otra cosa que aquello que el hombre siempre ha anhelado y buscado con denuedo, por medios disímiles como lo son el arte, la reflexión, o la ciencia. Si se puede afirmar que la época renacentista busca a través del arte y que los siglos XVIII y XIX lo hacen a través de la reflexión, se puede postular, también, que el siglo XX ha acudido a la técnica y a la ciencia para encontrar las respuestas a la gran pregunta: ¿qué somos? La segunda mitad del siglo XX ha sido obviamente el inicio de la era de la ciencia, en la que estamos profundamente sumergidos y en la que vemos aparecer sus frutos, a veces espectaculares, pero, las más de las veces, como cuestiones más bien banales, sin tener conciencia plena del esfuerzo y la inteligencia invertida por las personas en su realización. En este sentido, la labor del doctor Romo de Vivar es inconmensurable, al igual que la tarea de quizás cuántos científicos que, en el aislamiento del laboratorio, buscan y encuentran respuestas a las inquietudes que les despiertan las incógnitas del mundo. La ciencia y su apoyo técnico han sido los motores del proceso de globalización que se vive en nuestros tiempos, donde lo que les sucede a nuestros congéneres lejanos nos afecta sobre manera y donde lo que hagamos o dejemos de hacer compromete no sólo a nuestros contemporáneos sino, más aún, a los descendientes nuestros y ajenos. La globalización obedece a criterios de crecimiento económico originados en intereses que tienen como objetivo el éxito de la empresa sin que la felicidad o bienestar humano sea considerado y que requiere anular o por lo menos mitigar todas aquellas circunstancias que pudiesen alterar ese buen éxito. Frente a este proceso de globalización, que ha sido disparado por la ciencia, pero llevado a cabo por organismos extra científicos, se alzan visiones, propuestas y actos que desean preservar algunos tipos de identidad, creencias, prácticas y destinos: sin negarse a la inserción en el mundo: “quiero ser yo en mi grupo, con mis divinidades, mis ritos y mi vida después de mi vida”. Nota: Agradezco al Dr. Guillermo Delgado y por su intermedio a la Sociedad Química de México la oportunidad que me brindaron para escribir esta nota sobre el Dr. Alfonso Romo de Vivar. El doctor Romo de Vivar ha realizado una labor extensa, densa y magnífica en pro de la ciencia y de la humanidad. Ha formado discípulos y con ello ha cumplido con la exigencia que se le hace al hombre de ciencia: asegurar la inmortalidad de la ciencia para así perpetuar la humanidad. Aunque, para bien o para mal, el hombre sea diferente en diferentes latitudes, y pese a las diversidades que se ofrecen en cuanto a culturas, idiomas y religiones, así como también en cuanto a tipos humanos, conductas sociales y cosmovisiones, la ciencia trata de suputar la esencia del hombre y la del mundo, pero, más allá, la esencia del hombre en el mundo y la del mundo en el hombre. En esta perspectiva, no es muy atrevido afirmar que el doctor Romo de Vivar ha indagado en estas dos últimas áreas, buscando y encontrando elementos, procedimientos y modos de acción de aquello que ofrece el mundo y que puede ponerse al servicio del hombre para su bienestar y su felicidad. Como breve resumen de la labor del doctor Romo de Vivar podemos decir que ha dictado conferencias en diversos centros de alto nivel mundial, ha entregado sus conocimientos a nivel de difusión de la Química a estudiantes de pregrado y para el público en general; ha publicado brillantes trabajos en revistas científicas del más alto impacto y participado con sus alumnos en un número muy importante de Congresos Mexicanos e Internacionales. Además, el Dr. Romo de Vivar ha dirigido numerosas tesis de licenciatura, de maestría y de doctorado. Cabe destacar que este hombre de ciencia ha recibido innumerables distinciones en México, tanto en su país, como a nivel internacional, entre las cuales destacan: Premio Universidad Nacional, Premio IOCD-Syntex para la Excelencia de la Química, Investigador Emérito Universitario, Investigador Nacional Emérito, entre otras distinciones. Finalmente es importante señalar que el doctor Romo de Vivar está inscrito en el Cuadro de Honor de la Química, de México y Latinoamérica, junto a los Químicos que han sido destacados por la relevancia de su labor científica. El doctor Alfonso Romo de Vivar, ¿un hombre de dos vidas paralelas, dos hombres en uno, un hombre en el mundo y otro o el mismo en la ciencia? Lo más cierto es que no hay respuesta, pero sólo él mismo puede percibir un bosquejo de sentimiento frente al tema, no como una reflexión acerca de su ego, sino como apreciación de la proyección de su yo y como complacencia desde donde puede mirar hacia atrás. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 108-116 Investigación Synthesis, Structural, and Theoretical Study of New β-Heterosubstituted Captodative Olefins 1-Acetylvinyl Arenecarboxylates Jorge A. Mendoza,1 Hugo A. Jiménez-Vázquez,1 Rafael Herrera,2 Jide Liu,1 and Joaquín Tamariz1* Departamento de Química Orgánica, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Prol. de Carpio y Plan de Ayala, 11340 México, D.F. México. Tel: (+5255) 5729-6300 / 62411; Fax: (+5255) 5396-3503; E-mail: [email protected] 2 Instituto de Investigaciones Quimicobiológicas, Universidad Michoacana de San Nicolás de Hidalgo, Edif. B-1, Ciudad Universitaria, Francisco J. Mujica S/N, 58066 Morelia, Mich., México. 1 Dedicated to Dr. Alfonso Romo de Vivar on the occasion of his 50th anniversary of relevant contributions in natural products chemistry at the Instituto de Química,UNAM Recibido el 10 de diciembre del 2002; aceptado el 6 de febrero del 2003 Abstract. A new series of β-heterosubstituted captodative olefins 1acetylvinyl arenecarboxylates (7a and 10a-10h) has been prepared, by introducing nitrogen or sulphur as heteroatoms substituted by alkyl and aryl groups. Three preparation methods were evaluated by modifying the leaving group of the starting material, as well as the nucleophilic character of the adding thiols. All of them were efficient and stereoselective, providing the desired alkenes in good yields and with the Z configuration of the double bond. Ab initio calculations (HF/6-31G*) of the FMOs, of some of the beta amino and bromo olefins, explained their experimental reactivity in Diels-Alder additions with respect to the unsubstituted olefin 1a. It also appears that the HOMO and LUMO energies of the beta sulphur analogues are governed by the particular electronic features of the sulphur atom, and that their very low reactivity before a diene is due to steric hindrance. A comparison between bond distances obtained by X-ray crystallography of different β-substituted and unsubstituted olefins seems to correlate with the delocalization effect of the heteroatom lone electron pair for the bromo and amino β-substituted olefins. Keywords: Captodative olefin, structure, reactivity. Resumen. Se describe la preparación de una nueva serie de olefinas captodativas β-heterosustituidas 1-acetilvinil arencarboxilatos (7a and 10a-10h), donde el heteroátomo es nitrógeno o azufre, sustituido por grupos alquilo y arilo. Se evaluaron tres métodos para su preparación, modificando el grupo saliente en el sustrato, y el carácter nucleofílico de los tioles que se adicionaron. Los tres métodos fueron eficientes y estereoselectivos, proporcionando los alquenos deseados en buenos rendimientos y con la configuración Z del doble enlace. El cálculo de orbitales moleculares frontera (ab initio, HF / 6-31G*) de algunas de las olefinas preparadas permitió explicar sus diferencias en reactividad experimental con respecto a la olefina no sustituida 1a. Así, se sugiere que las energías de los orbitales HOMO y LUMO de los análogos azufrados están gobernadas por las propiedades electrónicas particulares del átomo de azufre, y que su baja reactividad ante un dieno depende también del efecto estérico. Una comparación de los datos de cristalografía de rayos X entre distancias de enlace de diferentes olefinas β-sustituidas y la no sustituida parece correlacionarse con el efecto de deslocalización del par de electrones no compartidos del heteroátomo para las olefinas β-bromo y β-amino sustituidas. Palabras clave: Olefina captodativa, estructura, reactividad. Introduction enaminones are important organic intermediates [7], and have potential biological activity [8]. Moreover, other β-substituted captodative olefins have been prepared, showing interesting pericyclic behavior and synthetic usefulness [9]. Attempts to carry out the Diels-Alder cycloaddition of alkenes 3a-3d with dienes such as cyclopentadiene (4) and isoprene (5) were, however, unsuccessful, except for derivative 3a. The latter yielded adducts with diene 4 in an unexpectedly high Captodative olefins have attracted particular attention in recent years, due to the opposite electronic demand and to the synthetic potential displayed by their geminally substituted functional groups [1]. We have shown that 1-acetylvinyl parenecarboxylates 1a-1c were highly reactive and selective in Diels-Alder [2] and 1,3-dipolar cycloadditions [3], and they also proved to be very useful synthons in natural product synthesis [4]. More recently, the alkyl 2-aroyloxy acrylates 2a-2b were prepared, showing also high reactivity and selectivity in Diels-Alder reactions [5]. With the aim of evaluating the effect of a third substituent in the double bond on the reactivity in [4+2] additions, compound 3a was prepared through a stereoselective synthetic route starting from 1a [6]. A series of amines 3b and thiols 3c and 3d were synthesized by treatment of 3a with the corresponding amines and thiols. In particular, Synthesis, Structural, and Theoretical Study of New β-Heterosubstituted Captodative Olefins 1-Acetylvinyl Arenecarboxylates Table 1. Preparation of the β-sulfanyl substituted captodative olefins 10a-10h from olefin 3a. Entrya RSX Baseb T (°C) 1 2 3 4 5 6 7 8 9 10 9a (R = Me, X = Na) 9b (R = Et, X = Na) 9c (R = i-Pr, X = Na) 9d (R = Et, X = H) 9e (R = i-Pr, X = H) 9f (R = t-Bu, X = H) 9g (R = Bn, X = H) 9h (R = C6H4p-Br, X = H) 9i (R = C6H4p-OMe, X = H) 9j (R = C6H4p-Me, X = H) ___ ___ ___ Et3N Et3N Et3N Et3N Et3N Et3N Et3N 25 25 25 0-120d 0-120d 0-120d 0-120d 0-120d 0-120d 0-120d 109 t (h) 10 (%)c 0.5 0.5 0.5 3.0 3.0 3.0 3.0 3.0 3.0 3.0 10a (85) 10b (90) 10c (90) 10b (78) 10c (81) 10d (82) 10e (90) 10f (96) 10g (80) 10h (81) All under N2 atmosphere, with 1.3 mol equiv. of RSX, in DMF as solvent. With 1.3 mol equiv. of Et3N. c After column and radial chromatography. d The reaction was succesively maintained at 0 °C for 1 h, 20 °C for 1 h, and at 120 °C for 1 h. a b Fig. 1. Fig. 2. exo stereoselectivity, and in comparable para/meta regioselectivity to that observed for olefin 1a with diene 5 [6]. The high reactivity and selectivity of captodative olefins in cycloaddition reactions is rather unexpected, since the electron-releasing effect of the aroyloxy group should decrease their reactivity in comparison with a dienophile or dipolarophile bearing only an electron-withdrawing group, such as methyl vinyl ketone (6) [10]. Structural and theoretical studies of olefin 1a revealed that the delocalization of the oxygen lone pair of the electron-donor group toward the π-sytem was inhibited by conformational restrictions [11]. In addition, FMO calculations suggested a dominant effect of the acetyl electron-withdrawing group on the polarization of the olefin [11]. However, the high regioselectivity shown by olefins 1 in 1,3-dipolar additions toward nitrones and nitrile oxides was better rationalized by DFT/HSAB theory [3c], showing the relevance of the electron-donor group in controlling the interaction of the cycloaddends. Therefore, electronic and structural factors should be taken into account to explain the reactivity and regiochemistry observed in both Diels-Alder and 1,3dipolar reactions. It is then relevant to evaluate the perturbation of the beta substituent in olefins 3 on the electronic and structural properties of these molecules and, in particular, on the double bond. Accordingly, we hereby report the preparation of a large series of new captodative olefins β-substituted with a new amine, and alkyl and aryl thiols. MO calculations were also carried out to assess the effect of the third substituent on the FMO energies and coefficients. Results and discussion Synthesis of β-heterosubstituted captodative olefins The β-amino substituted olefin 7a was prepared by treating αbromoalkene 3a with amine 8a, as an extension of the method used for the preparation of amino derivatives 3b [6] (Fig. 1). Thus, addition of N,N-methylphenyl amine (8a), in methylene chloride at 10 °C for 30 min, satisfactorily led to the desired substituted olefin 7a in 62 % yield. The Z configuration of the double bond was established through NOE experiments. Irradiation of proton H-4 resulted in enhancement of the signal for the acetyl group. In the previous report, the synthesis of β-thio (β-sulfanyl) olefins was limited to 3c and 3d by treatment of 3a with the corresponding thiol in the presence of triethylamine in DMF at room temperature [6]. The use of some other alkyl mercaptanes and aryl thiophenols under these conditions was not as efficient. Therefore, three additional methodologies were investigated to improve the yields. The first method involved the addition of the sodium salt of the thiol (9a-9c) to 3a (Fig. 2), leading to products 10a-10c in high yields and under mild conditions (Table 1, entries 1-3). When thiols 9d-9g were used, more severe conditions were applied in order to improve the preparation of olefins 10b-10e (entries 4-7). The yields were high even for the bulky thiol 9f. Under the same reaction conditions, substituted thiophenols 9h-9j reacted with 3a to give alkenes 10f-10h also in good yields (entries 8-10). The leaving group Y at the starting alkenes was also evaluated. Instead of the bromine atom of olefin 3a, the dimeth- 110 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Jorge A. Mendoza et al. Table 2. Preparation of the β-sulfanyl substituted captodative olefins 10b-10h from olefin 7b. Entrya 1 2 3 4 5 6 7 RSH 10 (%)b 9d (R = Et) 9e (R = i-Pr) 9f (R = t-Bu) 9g (R = Bn) 9h (R = C6H4p-Br) 9i (R = C6H4p-OMe) 9j (R = C6H4p-Me) 10b (74) 10c (77) 10d (70) 10e (87) 10f (91) 10g (93) 10h (90) a All under N2 atmosphere, with 1.3 mol equiv. of RSH and with 1.3 mol equiv. of Et3N, in DMF as solvent. The reaction was succesively maintained at 0 °C for 1 h, 120 °C for 1 h, and at 20 °C for 1 h. b After column and radial chromatography. ylamino group was used, due to its known aptitude as a leaving group [12]. In these cases the optimum reaction conditions were found to be similar to those employed for olefin 3a, that is, reacting olefin 7b in the presence of thiols 9d-9j and triethylamine as base, and heating from 0 °C up to 120 °C for 3 h (Table 2). The aliphatic thiols 9d-9g provided compounds 10b-10e in quite lower yields than those obtained when olefin 3a was used (Table 2, entries 1-4 vs. Table 1, entries 4-7); however, the yields increased for the preparation of 10g-10h when 7b was treated with thiophenols 9i-9j (Table 2, entries 6-7 vs. Table 1, entries 9-10). For both olefins, 3a and 7b, there were no significant differences in reactivity or yields to furnish the β-substituted alkenes 10 with the para substituted thiophenols 9h-9j (Table 1, entries 8-10 and Table 2, entries 5-7), considering that the temperature and the reaction time were comparable, regardless of the substituent. It is interesting that sodium thiolates 9a-9c reacted with olefin 3a to give the 1,4-addition products 10a-10c in good yields, instead of providing the hydrolisis products by addition to the p-nitrobenzoyloxy group, as observed when the corresponding alcohols were used. As expected, this behavior illustrates again the known greater softness of the sulphur atom with respect to the oxygen atom. The Z configuration of the double bond is maintained in all of the new olefins 10a-10h, as observed for the β-amino alkene 7a, indicating that it is largely more stable than the E configuration [6]. It is likely that the higher stability of the Z configuration is associated to destabilizing steric interactions found between the acetyl group and the beta substituent in the opposite E configuration. Such interactions may inhibit the effective conjugation of the enone moiety, and the delocalization of the heteroatom lone electron pair through the π system [6, 13]. The configuration of the double bond and the planar conformation of the enone moiety was confirmed by single crystal X-ray crystallography of olefin 10f (Fig. 3). It shows that both conjugated moieties, the enone and p-nitrobenzoyl groups, are in a quasi-orthogonal conformation. This agrees with those structures obtained for other analogues [5, 6, 11], as well as in the s-trans conformation of the enone. Interestingly, the p-bromophenyl ring lies out of the plane formed by the enone π-system. FMO calculations of β-heterosubstituted captodative olefins Frontier molecular orbital (FMO) theory has proven to be a reliable model to predict reactivity and regioselectivity in Diels-Alder [14] and 1,3-dipolar cycloadditions [15]. It has been able to explain the behavior of substituted olefins as dienophiles in Diels-Alder additions. For instance, the rate increases when the dienophile bears electron-withdrawing groups, whereas it decreases with dienophiles bearing electron-donating groups [16]. FMO theory has also been useful in explaining the reactivity and regioselectivity of captodative olefins with dienes such as isoprene (5) [2a, 11], showing that the interaction was controlled by normal electron demand (NED), i.e. the HOMOdiene/LUMO-dienophile interaction was the energetically most favorable. Furthermore, these results were supported by experimental measurement of ionization energies (IEs) and vertical attachment energies (VAEs) of 1a, whose relative values fit well with the calculated HOMO and LUMO energies, respectively [11]. Table 3. Ab initio HF/6-31G* Frontier Molecular Orbitals energies, and energy gaps (eV) of olefins 3a, 7a, 7b, 10a, 10f, and 10g, and isoprene (5). Entry 1 2 3 4 5 6 7 8 a Fig. 3. X-ray structure of captodative olefin 10f (ellipsoids with 30 % probability). Compounda HOMO LUMO HOMO-LUMOb 1ac 3a 7a 7b 10a 10f 10g 5c –11.0460 –10.4288 –8.4516 –8.8119 –9.1412 –9.2764 –10.2014 –8.6193 2.4588 2.1015 2.9375 3.1767 2.4681 2.3162 2.1900 3.5337 11.0781 10.7208 11.5568 11.7960 11.0874 10.9355 10.8093 Of the non-planar s-trans conformation for olefins 1a, 3a, 7a, 7b, 10f, and 10g, and of the s-cis conformation for olefin 10a and for the diene. b Energy gaps for the energetically more favorable HOMO-diene/LUMO-dienophile interaction. c Ref. [11]. Synthesis, Structural, and Theoretical Study of New β-Heterosubstituted Captodative Olefins 1-Acetylvinyl Arenecarboxylates 111 Table 4. Comparison of X-ray selected bond distances (Å) (estimated standard deviations) of the crystal structure of 10f with those of olefins 1a, 3a, and 7b, and with the average lenghts of bonds of the enone, enamine, vinyl ester and bromo vinyl systems. Bond a 10f 1aa 3ab 7bb Average Lengthc O(4)-C(3) C(2)-C(3) C(1)-C(2) C(1)-Y(5) 1.208 (4) 1.470 (5) 1.320 (4) 1.725 (4) 1.215 (5) 1.490 (5) 1.306 (6) 1.20 (3) 1.40 (3) 1.42 (3) 1.79 (3) 1.231 (7) 1.41 (1) 1.366 (9) 1.32 (1) C(2)-O(6) O(6)-C(7) C(7)-O(8) C(7)-C(9) 1.412 (3) 1.355 (4) 1.195 (4) 1.486 (4) 1.398 (4) 1.353 (4) 1.200 (4) 1.488 (5) 1.41 (3) 1.38 (2) 1.21 (3) 1.50 (3) 1.423 (7) 1.354 (6) 1.188 (6) 1.481 (7) 1.222 1.462 1.340 1.712 (S); 1.881 (Br); 1.358 (Nsp2); 1.418 (Nsp3) 1.353 1.359 1.201 1.481 Ref. [11]. b Ref. [6]. c Ref. [19] Table 5. Ab initio HF/6-31G* calculations of coefficientes (Ci) of the frontier molecular orbitals for olefins 1a, 3a, 7a, 7b, 10a, 10f, and 10g, and diene 5.a HOMO Compdb 3a 7a 7b 10a 10f 10g 1ad 5d LUMO C1 C2 C3 C4 ∆Cic C1 C2 C3 C4 ∆Cιc –0.2456 –0.1192 –0.1711 –0.1848 –0.1625 –0.2082 –0.3593 0.3247 –0.2839 –0.3110 –0.3702 –0.3135 –0.2694 –0.2400 –0.3565 0.2523 0.0095 –0.0215 –0.0190 –0.0038 –0.0019 0.0042 0.0236 –0.2180 0.1277 0.1414 0.1718 0.1475 0.1277 0.1174 0.1676 –0.2857 –0.0383 –0.1918 –0.1991 –0.1287 –0.1069 –0.0318 0.0028 0.0390 0.3188 0.2701 0.2799 0.3189 0.2872 0.2843 0.2940 0.2591 –0.2499 –0.1547 –0.1433 –0.2200 –0.2073 –0.2404 –0.2386 –0.2236 –0.2783 –0.2314 –0.2564 –0.2807 –0.2506 –0.2579 –0.2889 –0.2306 0.2759 0.2143 0.2316 0.2501 0.2263 0.2574 0.2800 0.2793 0.0689 0.1154 0.1366 0.0989 0.0799 0.0439 0.0554 –0.0202 These are the values of the 2pz coefficients, the relative 2pz’ contributions and their ∆Ci are analogous. The FMOs of the non-planar s-trans conformation for olefins 1a 3a, 7a, 7b, and 10g, and s-cis for 5, 10a, and 10f. c Carbon 1 - carbon 2 for the olefins; carbon 1 - carbon 4 for the diene. d Ref. [11] a b Therefore, the electronic effect produced by the substituent in the beta position of the captodative olefins 3, 7, and 10 on their behavior in Diels-Alder reactions could be evaluated by calculating the FMO energies of these molecules, and correlating them with those of the corresponding FMOs of a diene such as 5. Table 3 summarizes the calculated (HF/6-31G*) FMO energies of bromo and amino alkenes 3a, 7a, and 7b, as well as the thio alkenes 10a, 10f, and 10g. The geometries were optimized with the same basis set, showing that the most stable geometry for 3a, 7a, and 7b, corresponded to the s-trans conformation of the enone moiety, and the non-planar conformation of the p-nitrobenzoyloxy group. However, only for thioether 10g, the enone s-trans conformation was more stable, while olefins 10a and 10f were the exception, since the scis conformer for the enone conjugate system was slightly more stable (0.72 kcal/mol for 10a, and 0.24 kcal/mol for 10f), which is not in agreement with the X-ray structure of 10f. Even though these energy differences are negligible, and, practically, both conformations are isoenergetic, the FMO energy values listed in Table 3 corresponded to those obtained for the most stable confomation in every molecule. The co- 112 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) efficient differences follow a similar trend in the other conformers. As expected for the amino substituted olefins 7a and 7b, when an electron-releasing group is introduced into the double bond of the captodative olefin, both HOMO and LUMO should increase in energy (Table 3, entry 1 vs. entries 3 and 4). However, for the rest of the olefins, only the HOMO was energetically destabilized, since the LUMO energy of olefins 3a, 10f, and 10g was lower than that of the unsubstituted olefin 1a. It is likely that the effect of the heteroatom on the energy of the FMOs be the result of an interplay of different factors. The delocalization of the lone electron pair of bromine and sulphur toward the π-conjugated enone system will be less efficient than that of the nitrogen atom, due to the differences in electronic configuration. The inductive effect of these electronegative heteroatoms may increase the electronwithdrawing effect of the beta substituent on the electron density of the double bond [17]. The delocalization of the lone electron pair in enaminone 7b was reflected in a shortening of the C(1)-N and C(2)-C(3) bonds, and by an increase of the C(3)=O(4) and C(1)=C(2) bond lengths [6], as observed for other analogues by X-ray diffraction [18]. Similarly, significant delocalization appears in the bromo olefin 3a, since the C(2)-C(3) bond length is shorter and the C(1)=C(2) bond is longer than the corresponding bond lengths of 1a and of the average values taken from X-ray data of similar functional groups [19] (Table 4). In contrast, the Xray structure of 10f (Fig. 2) provides bond distances similar to those for the β-unsubstituted olefin and for the non-delocalized enone [19] (Table 4). It is noteworthy that the bond distances between the atoms of the aroyloxy group are not really perturbed by changes in the heteroatom at the beta position. This supports the idea of a non significant interaction between the lone pair of the oxygen atom of the electron donor group and the double bond, at least, in the crystalline state. According to data in Table 3, one could anticipate a higher reactivity of olefins 3a, 10f, and 10g with respect to 1a in Diels-Alder cycloadditions with diene 5, since the HOMOdiene / LUMO-dienophile energy gaps for the former dienophiles are smaller than that found for the latter. This expectation is only partially in agreement with the observed reactivity, since olefin 3a added to 5 under thermal and Lewis acid catalytic conditions similar to those used for 1a in shorter reaction times [6]. However, thioalkene 3d, which is analogous to olefins 10, failed to react even under catalysis. As predicted, the beta amino substituted alkene 7b was less reactive than 1a, being unable to react with cyclopentadiene (12) [6], which is considered a very reactive diene. The unreliable prediction of the reactivity of olefins 10 by FMO theory may be attributed to steric hindrance [20], which would counterbalance the electronic effect, as suggested previously by establishing a correlation between the reaction rate and LUMO energies of olefins 1a and 6 [11]. In addition, it has been found that the FMO model fails to account for the regioselectivity in Diels-Alder reactions with trisusbtituted dienophiles or with phenylthiosubstituted dienes [21], or in Jorge A. Mendoza et al. 1,3-dipolar cycloadditions of nitrile oxides and nitrones with olefins such as 1a [3c]. It is noteworthy that, under thermal and catalyzed conditions, the regioselectivity found in the Diels-Alder addition of dienophile 3a with isoprene (5) was similar to that observed with the β-unsubstituted olefin 1a [2a, 6]. From the FMO viewpoint, regioselectivity can be predicted on the basis of the atomic coefficient differences for the appropriate frontier orbital interaction: HOMO-diene/LUMO-dienophile, under NED control [14a, 14b]. It can be observed from Table 5 that, for olefin 3a, the relative magnitude of the LUMO coefficient in the monosubstituted terminus of the double bond is larger than that of the geminally disubstituted carbon. In comparison with the LUMO of olefin 1a, the difference in coefficients C1 and C2 for this olefin and bromo olefin 3a are analogous, hence a comparable regioselectivity should be observed, which is, indeed, experimentally found. Thus, for diene 5, which has the HOMO largest coefficient in carbon C-1, a preferred interaction with the largest coefficient on the double bond of the dienophile agrees with the para orientation as the major regioisomer. Polarization of the π-system in the HOMO for the amino olefins 7a and 7b, and for the thio olefins 10 is towards the substituted captodative carbon of the double bond (Table 5). The larger difference in coefficients (∆Ci) of these alkenes with respect to olefin 1a reflects the electron-donor effect of the heteroatom in beta position. In contrast, the opposite polarization is found for the LUMO, where the larger coefficient is located in the monosubstituted carbon atom (C1) of the double bond. Conclusions The stereoselective synthesis of new β-heteroatom substituted captodative olefins, including amino compound 7a, and sulphur derivatives 10a-10h, was feasible through three analogous routes. The common feature among them involved the replacement of a leaving group at the beta position of the olefin by the corresponding amino or thio compound. Both bromo and dimethylamino were efficient as the leaving groups in the starting activated substrates 3a and 7b, respectively. The former underwent nucleophilic attack of alkyl and aryl thiols or the sodium salt of some of them in good yields. The Z configuration of the double bond was established by NMR and X-ray crystallography. A comparison between bond distances by X-ray crystallography of different β-substituted and unsubstituted olefins seems to correlate with the delocalization effect of the heteroatom lone electron pair for the bromo and amino β-substituted olefins. Ab initio calculations of FMOs of trisubstituted amino olefins 7a and 7b showed an increase of HOMO and LUMO energies with respect to the unsubstituted 1a, as expected for the perturbation of the π-orbital by an electron-donating group. The presence of a bromine atom and alkyl and aryl thio groups in olefins 3a, and 10a, 10f, and 10g, however, pro- Synthesis, Structural, and Theoretical Study of New β-Heterosubstituted Captodative Olefins 1-Acetylvinyl Arenecarboxylates duced an increase of the HOMO energy and a stabilization of the LUMO. Based on these results, the predicted reactivity of the Diels-Alder additions with diene 5 agrees with experiment only for alkenes 3a, 7a and 7b. The regioselectivity observed for the cycloaddition of olefin 3a is also explained by FMO theory. Therefore, these calculations clearly indicate a significant perturbation of the double bond of the captodative olefin by the third substituent in the beta position. Experimental section General. Melting points (uncorrected) were determined with an Electrothermal capillary melting point apparatus. IR spectra were recorded on a Perkin-Elmer 1600 spectrophotometer. 1H and 13C NMR spectra were obtained on a Varian Gemini300 (300 MHz and 75.4 MHz), and Brucker DMX-500 (500 MHz and 125 MHz) instruments, with CDCl3 as solvent and TMS as internal standard. The mass spectra (MS) were taken on a Hewlett-Packard 5971A spectrometer. X-Ray analyses were collected using Mo Kα radiation (graphite crystal monochromator, λ = 0.71073 Å). Microanalyses were performed by M-H-W Laboratories (Phoenix, AZ). Analytical thin-layer chromatography was carried out using E. Merck silica gel 60 F254 coated 0.25 plates, visualizing by long- and short-wavelength UV lamp. All air moisture sensitive reactions were carried out under nitrogen using oven-dried glassware. DMF was freshly distilled and received on molecular sieves (4 Å), and methylene chloride from calcium hydride, prior to use. Triethylamine was freshly distilled from NaOH. All other reagents were used without further purification. Compounds 3a and 7b were prepared as described previously [6]. (Z)-4-(N,N-Methylphenylamino)-3-(4-nitrobenzoyloxy)-3buten-2-one (7a). To a solution of 1.0 g (3.18 mmol) of 3a in CH2Cl2 (25 mL), at 10 °C, 0.443 g (4.14 mmol) of 8a were added, and the mixture was stirred for 30 min. The reaction mixture was diluted with CH2Cl2 (50 mL), and was washed with a cold 5 % aqueous solution of HCl (2 × 25 mL), and a cold saturated solution of NaCl (2 × 30 mL). The organic layer was dried (MgSO4), and the solvent was evaporated under vacuum. The residue was successively purified by flash column chromatography on silica gel treated with 10 % of triethylamine (20 g, hexane/EtOAc, 90:10), and by radial chromatography (hexane/CH2Cl2, 90:10). The solid was recrystallized (hexane/CH2Cl2, 20:80) to give 0.67 g (62 %) of 7a as pale brown crystals: Rf 0.14 (hexane/EtOAc, 8:2); mp 141-143 °C; IR (CH2Cl2) 1741, 1621, 1591, 1529, 1495, 1350, 1316, 1244, 1096, 896 cm–1; 1H NMR (300 MHz, CDCl3) δ 2.24 (br s, 3H, CH3CO), 3.45 (s, 3H, CH3N), 6.99-7.04 (m, 1H, Ph-H), 7.13-7.16 (m, 2H, Ph-H), 7.22-7.27 (m, 2H, Ph-H), 7.48 (br s, 1H, HC=), 8.03-8.06 (m, 2H, Ar-H), 8.21-8.24 (m, 2H, Ar-H); 13C NMR (75.4 MHz, CDCl ) δ 24.3 (CH CO), 42.1 (CH N), 3 3 3 122.7, 123.2, 125.8, 126.8 (C-3), 129.2, 130.9 (C-4), 131.1, 134.3, 145.7, 150.6, 162.7 (ArCO2), 187.0 (CH3CO). Anal. 113 Calcd for C18H16N2O5: C, 63.52; H, 4.74; N, 8.23. Found: C, 62.75; H, 4.58; N, 7.71. General Procedures for the Preparation of Olefins 10a10h. Method A. To a mixture of 1.0 g (3.18 mmol) of 3a and 1.31 mol equiv. of the sodium salt of the corresponding thiol 9a-9c, at 0 °C, anhydrous DMF (10 mL) was added, and the mixture was stirred for 30 min. The solution was concentrated under vacuum, the reaction crude was diluted with CH2Cl2 (50 mL), and was washed with a cold 5 % aqueous solution of HCl (2 × 25 mL), and a cold saturated solution of NaCl (2 × 30 mL). The organic layer was dried (MgSO4), and the solvent was evaporated under vacuum. The residue was successively purified by column chromatography on silica gel treated with 10 % of triethylamine (30 g/1 g of crude, hexane/ EtOAc, 90:10), and by radial chromatography (hexane/CH2Cl2, 80:20). Method B. To a solution of 1.0 g (3.18 mmol) of 3a in anhydrous DMF (20 mL) 1.31 mol equiv. of the corresponding thiol 9d-9j were added at 20 °C. The mixture was stirred and cooled down to 0 °C, and a solution of 0.42 g (4.16 mmol) of triethylamine in anhydrous DMF (3 mL) was added dropwise. The mixture was maintained at the same temperature for 1 h, then warmed up to 120 °C for 1 h, and cooled down to 20 °C for 1 h. The solution was concentrated under vacuum, the reaction crude was diluted with CH2Cl2 (50 mL), and washed with a cold 5 % aqueous solution of HCl (2 × 25 mL), and a cold saturated solution of NaCl (2 × 30 mL). The organic layer was dried (MgSO4), and the solvent was evaporated under vacuum. The residue was successively purified by column chromatography on silica gel treated with 10 % of triethylamine (30 g/1 g of crude, hexane/EtOAc, 90:10), and by radial chromatography (hexane/CH2Cl2, 90:10). For the solid products, the recrystallization was carried out from hexane/CH2Cl2, 10:90. Method C. To a solution of 1.0 g (3.59 mmol) of 7b in anhydrous DMF (25 mL) 1.3 mol equiv. of the corresponding thiol 9d-9j were added at 20 °C. The mixture was stirred and cooled down to 0 °C, and a solution of 0.472 g (4.67 mmol) of triethylamine in anhydrous DMF (3 mL) was added dropwise. The mixture was maintained at the same temperature for 1 h, then warmed up to 120 °C for 1 h, and cooled down to 20 °C for 1 h. The solution was concentrated under vacuum, the reaction crude was diluted with CH2Cl2 (50 mL), and was washed with a cold 5 % aqueous solution of HCl (2 × 25 mL), and a cold saturated solution of NaCl (2 × 30 mL). The organic layer was dried (MgSO4), and the solvent was evaporated under vacuum. The residue was successively purified by column chromatography on silica gel treated with 10 % of triethylamine (30 g/1 g of crude, hexane/EtOAc, 80:20), and by radial chromatography (hexane/CH2Cl2, 80:20). For the solid products, the recrystallization was carried out from hexane/CH2Cl2, 10:90. 114 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) (Z)-4-Methylsulfanyl-3-(p-nitrobenzoyloxy)-3-buten-2-one (10a). According to method A with 0.291 g (4.16 mmol) of 9a, afforded 0.76 g (85 %) of 10a as a pale yellow oil: Rf 0.21 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1746, 1675, 1589, 1530, 1347, 1248, 1093, 895 cm–1; 1H NMR (300 MHz, CDCl3) δ 2.34 (s, 3H, CH3CO), 2.50 (s, 3H, CH3S), 7.39 (s, 1H, HC=), 8.32-8.34 (m, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 17.3 (CH3S), 24.7 (CH3CO), 123.7, 131.4, 134.1, 136.9 (C-4), 142.3 (C-3), 151.1, 161.7 (ArCO2), 187.1 (CH3CO); MS (70 eV) 281 (M+, 4), 150 (100), 134 (22), 120 (15), 104 (22), 92 (11), 76 (14). (Z)-4-Ethylsulfanyl-3-(p-nitrobenzoyloxy)-3-buten-2-one (10b). According to method A with 0.349 g (4.16 mmol) of 9b, afforded 0.84 g (90 %) of 10b as a pale yellow oil. According to method B with 0.258 g (4.16 mmol) of 9d, gave 0.73 g (78 %) of 10b. According to method C with 0.289 g (4.67 mmol) of 9d, furnished 0.83 g (74%) of 10b: Rf 0.22 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1746, 1675, 1588, 1529, 1423, 1348, 1275, 1246, 1217, 1049, 1013 cm–1; 1H NMR (300 MHz, CDCl3) δ 1.43 (t, J = 7.1 Hz, 3H, CH3CH2S), 2.35 (s, 3H, CH3CO), 2.92 (q, J = 7.1 Hz, 3H, CH3CH2S), 7.49 (s, 1H, HC=), 8.45-8.47 (m, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 15.4 (CH3CH2S), 24.7 (CH3CO), 28.4 (CH3CH2S), 123.6, 131.4, 134.1, 135.1 (C-4), 142.5 (C-3), 151.0, 161.7 (ArCO2), 187.1 (CH3CO). Anal. Calcd for C13H13NO5S: C, 52.87; H, 4.44. Found: C, 52.80; H, 4.68. Jorge A. Mendoza et al. Anal. Calcd for C15H17NO5S: C, 55.71; H, 5.30. Found: C, 55.50; H, 5.30. (Z)-4-Benzylsulfanyl-3-(p-nitrobenzoyloxy)-3-buten-2-one (10e). According to method B with 0.516 g (4.13 mmol) of 9g, afforded 1.02 g (90 %) of 10e as a pale yellow oil. According to method C with 0.579 g (4.67 mmol) of 9g, furnished 1.18 g (87 %) of 10e: Rf 0.28 (hexane / EtOAc, 8:2); IR (CH2Cl2) 1740, 1677, 1606, 1530, 1348, 1248, 1093 cm–1; 1H NMR (300 MHz, CDCl3) δ 2.25 (s, 3H, CH3CO), 4.08 (s, 2H, CH2Ph), 7.27-7.46 (m, 6H, HC=, Ph-H), 8.28-8.39 (m, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 24.7 (CH3CO), 38.0 (CH2Ph), 123.7, 128.2, 129.0, 129.1, 131.4, 134.0, 134.2 (C4), 136.7, 142.4 (C-3), 150.9, 161.7 (ArCO2), 187.1 (CH3CO). (Z)-4-(4-Bromophenylsulfanyl)-3-(4-nitrobenzoyloxy)-3buten-2-one (10f). According to method B with 0.786 g (4.16 mmol) of 9h, afforded 1.29 g (96 %) of 10f as pale yellow crystals. According to method C with 0.883 g (4.67 mmol) of 9h, furnished 1.45 g (91 %) of 10f: mp 174-175 °C; Rf 0.16 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1747, 1679, 1531, 1419, 1348, 1243, 1090, 1008 cm–1; 1H NMR (300 MHz, CDCl3) δ 2.35 (s, 3H, CH3CO), 7.36-7.40 (m, 2H, Ar-H), 7.46 (s, 1H, HC=), 7.51-7.55 (m, 2H, Ar-H), 8.30-8.38 (m, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 24.9 (CH3CO), 123.6, 123.7, 130.8, 131.5, 132.9, 133.1, 133.7, 134.1 (C-4), 142.6 (C-3), 151.2, 161.7 (ArCO 2 ), 187.3 (CH 3 CO). Anal. Calcd for C17H12BrNO5S: C, 48.36; H, 2.86; N 3.32; S, 7.58. Found: C, 48.04; H, 2.80; N, 3.29; S, 7.41. (Z)-4-Isopropylsulfanyl-3-(p-nitrobenzoyloxy)-3-buten-2one (10c). According to method A with 0.408 g (4.16 mmol) of 9c, afforded 0.88 g (90 %) of 10c as a pale yellow oil. According to method B with 0.358 g (4.16 mmol) of 9e, gave 0.80 g (81 %) of 10c. According to method C with 0.402 g (4.67 mmol) of 9e, furnished 0.90 g (77 %) of 10c: Rf 0.30 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1746, 1674, 1587, 1530, 1453, 1427, 1348, 1314, 1218, 1157, 1092, 1015, 958, 898, 870, 847 cm–1; 1H NMR (300 MHz, CDCl3) δ 1.38 (d, J = 7.3 Hz, 6H, (CH3)2CHS), 2.35 (s, 3H, CH3CO), 3.33 (sept, J = 7.3 Hz, 3H, (CH3)2CHS), 7.50 (s, 1H, HC=), 8.38-8.36 (m, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 23.4 ((CH3)2CHS), 24.5 (CH3CO), 38.4 ((CH3)2CHS), 123.4, 131.1, 133.9, 134.1 (C-4), 141.9 (C-3), 150.7, 161.5 (ArCO2), 187.0 (CH3CO). Anal. Calcd for C14H15NO5S: C, 54.36; H, 4.89. Found: C, 54.30; H, 5.00. (Z)-4-(4-Methoxyphenylsulfanyl)-3-(4-nitrobenzoyloxy)-3buten-2-one (10g). According to method B with 0.582 g (4.16 mmol) of 9i, afforded 0.95 g (80 %) of 10g as pale yellow crystals. According to method C with 0.654 g (4.67 mmol) of 9i, furnished 1.25 g (93 %) of 10g: mp 122-123 °C; R f 0.20 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1743, 1674, 1592, 1546, 1531, 1492, 1421, 1346, 1264, 1093 cm–1; 1H NMR (300 MHz, CDCl3) δ 2.33 (s, 3H, CH3CO), 3.82 (s, 3H, MeO), 6.88-6.94 (m, 2H, Ar-H), 7.44-7.49 (m, 2H, Ar-H), 7.45 (s, 1H, HC=), 8.34 (s, 4H, Ar-H); 13C NMR (75.4 MHz, CDCl3) δ 24.8 (CH3CO), 55.4 (CH3O), 115.3, 122.0, 123.6, 131.5, 134.0, 134.2, 137.2 (C-4), 141.5 (C-3), 150.9, 160.7, 161.7 (ArCO2), 187.2 (CH3CO). Anal. Calcd for C18H15NO6S: C, 57.90; H, 4.05; N, 3.75; S, 8.59. Found: C, 58.08; H, 4.12; N 3.74; S, 8.70. (Z)-4-tert-Butylsulfanyl-3-(p-nitrobenzoyloxy)-3-buten-2one (10d). According to method B with 0.375 g (4.13 mmol) of 9f, afforded 0.84 g (82 %) of 10d as a pale yellow oil. According to method C with 0.514 g (4.67 mmol) of 9f, furnished 0.86 g (70%) of 10d: Rf 0.26 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1744, 1673, 1584, 1529, 1349, 1248, 1093, 769, 751 cm–1; 1H NMR (300 MHz, CDCl3) δ 1.48 (s, 9H, (CH3)3CS), 2.35 (s, 3H, CH3CO), 7.60 (s, 1H, HC=), 8.33 (br s, 4H, ArH); 13C NMR (75.4 MHz, CDCl 3) δ 24.8 (CH 3CO), 31.1 ((CH 3) 3CS), 45.7 ((CH 3) 3CS), 123.7, 131.4, 131.6 (C-4), 134.2, 142.3 (C-3), 151.0, 162.4 (ArCO2), 198.8 (CH3CO). (Z)-4-(4-Methylphenylsulfanyl)-3-(4-nitrobenzoyloxy)-3buten-2-one (10h). According to method B with 0.516 g (4.16 mmol) of 9j, afforded 0.92 g (81 %) of 10h as pale yellow crystals. According to method C with 0.58 g (4.67 mmol) of 9j, furnished 1.15 g (90 %) of 10h: mp 113-115 °C; Rf 0.22 (hexane/EtOAc, 8:2); IR (CH2Cl2) 1747, 1677, 1589, 1531, 1348, 1091, 897, 847, 807 cm–1; 1H NMR (500 MHz, CDCl3) δ 2.33 (s, 3H, CH3CO), 2.37 (s, 3H, MeAr), 7.20-7.22 (m, 2H, Ar-H), 7.39-7.41 (m, 2H, Ar-H), 7.50 (s, 1H, HC=), 8.34 (s, 4H, Ar-H); 13C NMR (125 MHz, CDCl3) δ 21.1 (CH3Ar), 24.8 (CH3CO), 123.7, 128.2, 130.5, 131.5, 131.9, 134.0, 136.3 Synthesis, Structural, and Theoretical Study of New β-Heterosubstituted Captodative Olefins 1-Acetylvinyl Arenecarboxylates (C-4), 138.5, 141.9 (C-3), 151.0, 161.7 (ArCO 2 ), 187.3 (CH3CO). Anal. Calcd for C18H15NO5S: C, 60.49; H, 4.23. Found: C, 60.27; H, 4.50. Single-Crystal X-Ray Crystallography [22]. Olefin 10f was obtained as pale yellow crystals. These were mounted in glass fibers. Crystallographic measurements were performed on a Siemens P4 diffractometer with Mo Kα radiation (λ = 0.7107 Å; graphite monochromator) at room temperature. Two standard reflections were monitored periodically; they showed no change during data collection. Unit cell parameters were obtained from least-squares refinement of 26 reflections in the range 2 < 2Θ < 20°. Intensities were corrected for Lorentz and polarization effects. No absorption correction was applied. Anisotropic temperature factors were introduced for all nonhydrogen atoms. Hydrogen atoms were placed in idealized positions and their atomic coordinates refined. Unit weights were used in the refinement. The structure was solved using SHELXTL on a personal computer [23]. Data of 10f: Formula: C17H12BrNO5S; molecular weight: 422.25; cryst. syst.: monoclinic; space group: P21/c; unit cell parameters: a, 6.4121 (9), b, 12.8041 (11), c, 20.998 (2) (Å); α, 90, β, 94.884 (9), γ, 90 (deg); temp. (°K): 293 (2); Z: 4; No. of reflections collected: 4641; no. of independent reflections: 3335; no. of observed reflections: 3290; R: 0.0429; GOF: 1.012. Calculations. The ab initio SCF/HF calculations were carried out with the 6-31G* basis sets using Gaussian 94 [24] and MacSpartan [24]. Geometries were fully optimized by the AM1 semiempirical method [25] and these were employed as starting point for optimization, at the 6-31G* level. Acknowledgments We thank Fernando Labarrios for his help in spectrometric measurements. J.T. would like to acknowledge DEPI/IPN (Grants 921769 and 200410) and CONACyT (Grants 1570P and 32273-E) for financial support. H.A.J.-V. thanks CONACyT (Grant 3251P) for financial support. J.M. and R.H. are grateful to CONACyT (Grants 86038 and 91187) for graduate fellowships, to PIFI-IPN program for a scholarship, and to the Ludwig K. Hellweg Foundation for a partial scholarship. J.L. is grateful to Secretaría de Relaciones Exteriores, México (Grant DAC-III 811.5/(510)/137) for a research fellowship. J.T. and H.A.J.-V. are fellows of the EDD/IPN and COFAA/IPN programs. References and notes 1. (a) Viehe, H. G.; Janousek, Z.; Merényi, R.; Stella, L. Acc. Chem. Res. 1985, 18, 148-154. (b) Cativiela, C.; Fraile, J. M.; García, J. I.; Mayoral, J. A.; Pires, E.; Royo, A. J.; Figueras, F.; de Ménorval, L. C. Tetrahedron 1993, 49, 4073-4084. (c) Boucher, J.-L.; Stella, L. Tetrahedron 1986, 42, 3871-3885. (d) Seerden, J.-P. G.; Scheeren, H. W. Tetrahedron Lett. 1993, 34, 2669-2672. (e) Döpp, D.; Libera, H. Tetrahedron Lett. 1983, 24, 885-888. (f) Rulev, A. Y. Russ. Chem. Rev. 2002, 71, 195-221. (g) Seneci, P.; 115 Leger, I.; Souchet, M.; Nadler, G. Tetrahedron 1997, 53, 1709717114. (h) Moody, C. J.; Hughes, R. A.; Thompson, S. P.; Alcaraz, L. Chem. Commun. 2002, 1760-1761. (i) Kozmin, S. A.; Iwama, T.; Huang, Y.; Rawal, V. H. J. Am. Chem. Soc. 2002, 104, 4628-4641. (j) Ferreira, P. M. T.; Maia, H. L. S.; Monteiro, L. S. Tetrahedron Lett. 2002, 43, 4491-4493. (k) Ferreira, P. M. T.; Maia, H. L. S.; Monteiro, L. S. Tetrahedron Lett. 2002, 43, 4495-4497. (l) Huang, C.-G.; Chang, B.-R.; Chang, N.-C. Tetrahedron Lett. 2002, 43, 2721-2723. 2. (a) Reyes, A.; Aguilar, R.; Muñoz, A. H.; Zwick, J.-C.; Rubio, M.; Escobar, J.-L.; Soriano, M.; Toscano, R.; Tamariz, J. J. Org. Chem. 1990, 55, 1024-1034. (b) Aguilar. R.; Reyes. A.; Tamariz, J.; Birbaum, J.-L. Tetrahedron Lett. 1987. 28. 865-868. (c) García de Alba, O.; Chanona, J.; Delgado, F.; Zepeda, G.; Labarrios, F.; Bates, R. W.; Bott, S.; Juaristi, E.; Tamariz, J. Anal. Quím., Int. Ed. 1996, 92, 108-117. 3. (a) Nagarajan, A.; Zepeda, G.; Tamariz, J. Tetrahedron Lett. 1996, 37, 6835-6838. (b) Jiménez, R.; Pérez, L.; Tamariz, J.; Salgado, H. Heterocycles 1993, 35, 591-598. (c) Herrera, R.; Nagarajan, A.; Morales, M. A.; Méndez, F.; Jiménez-Vázquez, H. A.; Zepeda, L. G.; Tamariz, J. J. Org. Chem. 2001, 66, 12521263. 4. (a) Andrade, R. M.; Muñóz, A. H.; Tamariz, J. Synth. Commun. 1992, 22, 1603-1609. (b) Orduña, A.; Zepeda, G.; Tamariz, J. Synthesis 1993, 375-377. (c) Dienes, Z.; Vogel, P. J. Org. Chem. 1996, 61, 6958-6970. (d) Ochoa, M. E.; Arias, A. S.; Aguilar, R.; Delgado, F.; Tamariz, J. Tetrahedron 1999, 55, 14535-14546. (e) Aguilar, R.; Reyes, A.; Orduña, A.; Zepeda, G.; Bates, R. W.; Tamariz, J. Rev. Soc. Quím. Méx. 2000, 44, 91-96. 5. Herrera, R.; Jiménez-Vázquez, H. A.; Modelli, A.; Jones, D.; Söderberg, B. C.; Tamariz, J. Eur. J. Org. Chem. 2001, 46574669. 6. Peralta, J.; Bullock, J. P.; Bates, R. W.; Bott, S.; Zepeda, G.; Tamariz, J. Tetrahedron 1995, 51, 3979-3996. 7. (a) Rappoport, Z., Ed., The Chemistry of Enamines. Wiley, Chichester, 1994. (b) Greenhill, J. V. Chem. Soc. Rev. 1977, 6, 277-294. (c) Granik, V. G. Russ. Chem. Rev. 1984, 53, 383-404. 8. (a) Scott, K. R.; Edafiogho, I. O.; Richardson, E. L.; Farrar, V. A.; Moore, J. A.; Tietz, E. I.; Hinko, C. N.; Chang, H.; El-Assadi, A.; Nicholson, J. M. J. Med. Chem. 1993, 36, 1947-1955. (b) Kesten, S. J.; Degnan, M. J.; Hung, J.; McNamara, D. J.; Ortwine, D. F.; Uhlendorf, S. E.; Werbel, L. M. J. Med. Chem. 1992, 35, 3429-3447. 9. For some recent examples: (a) Bhathia, G. S.; Lowe, R. F.; Pritchard, R. G.; Stoodley, R. J. Chem. Commun. 1997, 19811982. (b) Abbiati, G.; Clerici, F.; Gelmi, M. L.; Gambini, A.; Pilati, T. J. Org. Chem. 2001, 66, 6299-6304. (c) Clerici, F.; Gelmi, M. L.; Pocar, D.; Pilati, T. Tetrahedron: Asymmetry 2001, 12, 2663-2669. (d) Buñuel, E.; Gil, A. M.; Díaz-de-Villegas, M. D.; Cativiela, C. Tetrahedron 2001, 57, 6417-6427. (e) Clerici, F.; Gelmi, M. L.; Gambini, A.; Nava, D. Tetrahedron 2001, 57, 6429-6438. (f) Maekawa, K.; Igarashi, T.; Kubo, K.; Sakurai, T. Tetrahedron 2001, 57, 5515-5526. (g) Caine, D. Tetrahedron 2001, 57, 2643-2684. (h) Stanovnik, B.; Svete, J. Synlett 2000, 1077-1091. (i) Yonehara, K.; Ohe, K.; Uemura, S. J. Org. Chem. 1999, 64, 9381-9385. 10. (a) Sustmann, R. Tetrahedron Lett. 1971, 2721-2724. (b) Houk, K. N.; Sims, J.; Watts, C. R.; Luskus, L. J. J. Am. Chem. Soc. 1973, 95, 7301-7315. 11. Jiménez-Vázquez, H. A.; Ochoa, M. E.; Zepeda, G.; Modelli, A.; Jones, D.; Mendoza, J. A.; Tamariz, J. J. Phys. Chem. A 1997, 101, 10082-10089. 12. (a) Guseinov, F. I.; Yudina, N. A.; Burangulova, R. N.; Ryzhikova, T. Ya.; Valliullina, R. Zh. Chem. Heterocyclic Comp. 2002, 38, 496-497. (b) Toplak, R.; Svete, J.; Stanovnik, B.; Grdadolnik, S. G. J. Heterocyclic Chem. 1999, 36, 225-235. (c) Pizzioli, L.; Ornik, B.; Svete, J.; Stanovnik, B. Helv. Chim. Acta 1998, 81, 231-235. 116 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) 13. Zhuo, J.-C. Magn. Reson. Chem. 1998, 36, 565-572. 14. (a) Fleming, I. Frontier Orbitals and Organic Chemical Reactions. John Wiley & Sons, Chichester, 1976. (b) Houk, K. N. Acc. Chem. Res. 1975, 8, 361-369. (c) Eisenstein, O.; Lefour, J. M.; Anh, N. T.; Hudson, R. F. Tetrahedron 1977, 33, 523-531. (d) Spíno, C.; Pesant, M. Dory, Y. Angew. Chem., Int. Ed. 1998, 37, 3262-3265. 15. (a) Carruthers, W. Cycloaddition Reactions in Organic Synthesis. Pergamon Press, Oxford, 1990, 269-331. (b) Houk, K. N.; Chang, Y.-M.; Strozier, R. W.; Caramella, P. Heterocycles 1977, 7, 793799. (c) Huisgen, R., in: 1,3-Dipolar Cycloaddition Chemistry, Vol. 1, Padwa, A., Ed., Wiley-Interscience, New York, 1984. (d) Jarosková, L.; Fisera, L.; Matejková, I.; Ertl, P.; Prónayová, N. Monatsh. Chem. 1994, 125, 1413-1425. (e) Houk, K. N.; Sims, J.; Duke, Jr., R. E.; Strozier, R. W.; George, J. K. J. Am. Chem. Soc. 1973, 95, 7287-7301. 16. (a) Sauer, J.; Sustmann, R. Angew. Chem., Int. Ed. Engl. 1980, 19, 779-807. (b) Anh, N. T.; Canadell, E.; Eisenstein, O. Tetrahedron 1978, 34, 2283-2288. 17. (a) Exner, O. J. Phys. Org. Chem. 1999, 12, 265-274. (b) Charton, M. J. Phys. Org. Chem. 1999, 12, 275-282. (c) Galkin, V. I. J. Phys. Org. Chem. 1999, 12, 283-288. 18. Zhuo, J.-C.; Schenk, K. Helv. Chim. Acta 1997, 80, 2137-2147. 19. Allen, F. H.; Kennard, E.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1-S19. 20. (a) Fox, M. A.; Cardona, R.; Kiwiet, N. J. J. Org. Chem. 1987, 52, 1469-1474. (b) Bachler, V.; Mark, F. Theoret. Chim. Acta Jorge A. Mendoza et al. 1976, 43, 121-135. (c) Tripathy, R.; Franck, R. W.; Onan, K. D. J. Am. Chem. Soc. 1988, 110, 3257-3262. (d) Padwa, A.; Kline, D. N.; Koehler, K. F.; Matzinger, M.; Venkatramanan, M. K. J. Org. Chem. 1987, 52, 3909-3917. 21. Kahn, S. D.; Pau, C. F.; Overman, L. E.; Hehre, W. J. J. Am. Chem. Soc. 1986, 108, 7381-7396. 22. The authors have deposited the atomic coordinates for this structure with the Cambridge Crystallographic Data Centre. The coordinates can be obtained, on request, from the Director Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK. 23. SHELXTL, v. 5.03, Siemens Energy & Automation, Germany, 1995. 24. Calculated with Gaussian 94, Revision E.2: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian, Inc., Pittsburgh, PA, 1995; and MacSpartan, v. 1.0, WaveFunction Inc., 18401 VonKarman, Suite 370, Irvine, CA 92715. 25. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J. Am. Chem. Soc. 1985, 107, 3902-3909. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 117-123 Investigación 2D 1H and 13C NMR from the adducts of the dichloro carbene addition to β-ionone. The role of the catalyst on the phase transfer reaction Eduardo Díaz,*a José Luis Nava,a Héctor Barrios,a David Corona,a Ángel Guzmán,a Ma. de Lourdes Muciño,b and Aydeé Fuentesb Instituto de Química, Universidad Nacional Autónoma de México. Circuito Exterior, Ciudad Universitaria, Coyoacán 04510 México D.F. b Facultad de Química, Universidad Autónoma del Estado de México, Toluca, Estado de México. a Recibido el 9 de diciembre del 2002; aceptado el 26 de febrero del 2003 En homenaje al Dr. Alfonso Romo de Vivar Abstract. Several haloderivatives were synthesized from β-ionone using CHCl3, NaOH and selected tetralkyl ammonium halide. 2D NMR and X-ray single crystal analysis of the products are reported. Keywords: Addition of dichlorocarbenes to β-Ionone, phase transfer addition, formation of 1,1 dichlorocyclopropanes. 1H and 13C NMR. Resumen. Algunos haloderivados fueron sintetizados a partir de βionona usando cloroformo, hidróxido de sodio y halogenuros de tetralquilamonio selectos. Se informan los análisis por 2D RMN y rayos X de los productos. Palabras clave: Adición de diclorocarbenos a β-ionona, adición de transferencia de fase, formación de 1,1 diclorociclopropanos. RMN de 1H y 13C. Introduction It has been reported that in phase transfer dihalocarbene addition, stereochemistry and regioselectivity can be controlled by varying the catalyst [8-11]. Phase transfer reactions are known to be somewhat dependent on the exact reaction conditions and the formation of products apparently derived from trihalogenomethyl anion or dihalogen carbene can be controlled by varying the catalyst [9]. Some years ago [12], our group reported β-Ionone when treated under CHCl3, NaOH and triethylbenzyl ammonium chloride (TEBAC) yield two furenones 2 and 3 which were isolated in low yield. Their structures were well supported by 2D NMR as well as x-ray crystallographic analysis [13]. On the mechanism of formation of these compounds it was assumed an initial chemoselective adduct formation on the β-ionone to generate the epoxide, which is rapidly transformed to an intermediate dichloro ether which is hydrolized to a γ-lactone where the presence of atmospheric oxygen during the reaction induced free radical dimerization [12]. The low yield observed in this reaction as well as the new findings about the phase transfer reactions encourage us to undertake the study of the dichlorocarbene addition to βionone using different quaternary ammonium catalyst. Under the new catalyst selection, the usual γ-lactone derivatives (2 and 3) were isolated together with several new compounds which display structure versatility and in addition allowed us to improve yields, as well as products ratio of the obtained adducts. Structures of compounds 2 and 3 were well discussed in our previous report and they will not be focused here. A large number of C13-compounds formed by oxidative cleavage of carotenoids have been isolated from tobacco [1] and marine sponges [2] or their derivatives, originated from α and β ionones (1' and 1). In the same way, some other secondary metabolites as 4-oxomegastigmenos, important compounds for their excellent flavor characteristics are also formed in the tobacco plant from α and β ionones. Other derivatives from ionones, isolated from rabbit urine and from the secretion of the anal gland of the red fox have been obtained when α-ionone was photo-oxygenated [3, 4]. The synthesis of several damascones [5] and abscisic acids [6, 7] using readily available and inexpensive ionones as raw material encourages us to perform a reaction leading to some new functionalized and useful unsaturated compounds. Results and discussion The present paper deals with the synthesis and structure elucidation of new representatives of this group of compounds. We describe here the products obtained when β-ionone was treated with CHCl3, NaOH, using as catalyst different tetralkyl quaternary ammonium salts looking to improve yields and versatility of the reaction under phase transfer conditions. Likewise, the reaction of carbenes with terpenes provides a simple means of examining the stereochemistry and regioselectivity of addition of such species to a variety of double bonds. 118 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) O Eduardo Díaz et al. O O 1 O 2 1' O O Cl O Cl H O O Cl H Cl O 3 4 5 Cl O O Cl O O 6 7 6a O O O O O OH O CH 3 8 9 ** * ** * 9a O O O O O O O Cl 10 11 Cl Cl Cl Cl Scheme 1 Scheme 1 shows the structures of the new isolated derivatives. The IR spectrum of 4 showed an absorption at 1717 cm–1 which clearly indicates an unsaturated ketone group. The proton NMR spectrum showed a one proton doublet and one proton multiplet at δ = 3.04 and 2.83 respectively. The latter was converted to a doublet of triplet when the methyl protons (C7) were decoupled. A methyl ketone group appeared at δ = 2.45 as a singlet. These findings suggest that in the structure of compound 4 was lost the α,β-unsaturated double bond observed in the starting material 1. The mass spectrum of compound 4 shows the molecular ions at m/z 274 (M+), m/z 276 (M++ 2) and m/z 278 (M+ + 4) corresponding to the molecular formula of C14H20OCl2. It suggests an increase in molecular weight of the starting ionone 1 by 82 atomic mass units (CCl2). The NOESY experiment enabled us to establish the key proton vicinities. There exist a strong correlations between methine doublet at δ = 3.04 and methyl signals δ = 2.45 (C13), 1.00 (C-9) and a weaker one between methyne multiplet at =2.83 and the methyl singlet at δ = 1.75. The 13C 1D NMR and DEPT edited spectra [14] allowed the assignment of the carbon atoms (protonated carbons), and the long range heterocorrelated spectra [15] enabled us to establish connectivities between protons and carbons (see experimental). Because the unusual trans cyclopropane vicinal coupling between HC10-HC11 (J = 9.0 Hz) and in order to confirm this adduct structure we carried out an x-ray crystallographic study of compound 4. Fig. 1 display the Ortep plot of compound 4. On the other hand, compound 5 shows a molecular weight as was identified from MS as 274, 276, 278 (C14H20O Cl2). The proton NMR presents an AX pattern (δA = 6.63 (H10) and δX = 6.17 (H-11, J=16.0 Hz), four singlets at δ = 2.30, 1.25, 1.20 and 1.00 for the methyls at C-13, C-7, C-8 and C-9 and the signals for six protons on C-4, C-5 and C-6 methylenes. The structure of compound 5 supported by the assignments of 1H and 13C (CDCl3, 500 MHz , 125 MHz) spectra and confirmed by HMQC, HMBC along with HOMOCOSY and the mass spectrum fragmentation pattern matching the expected molecular weight for such a structure. The structure of the monochlorine furane derivative 6 was supported by its spectroscopic features. The molecular ion M+ was observed in mass spectrometry (EI) at m/z 238 (M+), m/z 240 (M+ + 2) in agreement with the molecular weight for a furane derivative. Tentatively, two isomeric structures (6 and 6a) emerge to be considered under mechanistic approach (Scheme 2). The 1H and 13C NMR (CDCl3, 300 MHz, 75.0 MHz), together with HETCOR [16, 17] and COLOC [18] spectra were recorded in order to probe the structure 6. At first, since 13CNMR chemical shifts observed (and calculated) [19] for 6 (or 6a) did not enable unambiguously to differentiate between 6 or 6a, we performed a COLOC experiment in order to overcome such drawback. In this experiment we were able to observe the long range proton-carbon correlation between the furane methyl protons (δ = 2.20, C-13) with the methine C-11 (δ = 111.0) and with the nonprotonated carbon at δ = 149.9 (C-12). It suggests unambiguously structure 6 for this furane derivative. In the structure 6a the 3σ bond correlation between the methyl group with the carbon at δ = 149.9 is not permissible (Fig. 2). The remaining signals that enabled us the elucidation of structure 6 are described in experimental. On the other hand, the formation of the dimeric compound 9 (or 9a, 9a' or 9') can be explained by two different mechanism approach. At first, if we assume that compound 9a (or 9a') could be formed by a concerted opening and dimerization of the epoxide 7 (Scheme 3) obtained in a small amount under phase transfer conditions. This assumption was considered because the previously mentioned opening was for Cl1 C7 O1 C1 0 C1 4 C1 C12 C2 C6 C1 1 C1 3 Cl2 C5 C4 C9 C3 O 11 7 C8 1 2 6 4 Fig. 1. X-ray ortep plot of compound 4. 10 3 5 12 13 14 Cl 9 8 Cl 2D 1H and 13C NMR from the adducts of the dichloro carbene addition to β-ionone Cl O Cl Cl Cl O O H CCl2 1 OH CCl2 Cl - Cl Cl Cl OH O O O 6a O Cl Cl O O O Cl H 6 2 Scheme 2 us described and probed, since we were able to isolate the alcohol-methoxyether derivative 8, whose structure and stereochemistry were confirmed by an x-ray crystallographic study [12]. The 13C NMR chemical shifts were a key element in order to rule out the structures of the dioxine derivatives 9a, 9a' or the oxide 9'. The calculated chemical shifts [19] for the non protonated carbons attached to both oxygen atoms (** and * in schemes 1 and 3) should to display chemical shifts at δ = 90.0 and 71.0, respectively. However, compound 9 showed in its 13C NMR chemical shifts at δ =71.0 and 65.8 ppm respectively, which definitely support the structure 9 (instead 9a, 9a' or 9') for this compound. About the configuration of the stereogenic centers, they were established by the energetic calculation [20] of the isomeric structures shown in scheme 4 (9, 9' and 9") being 9 those having the lesser energy (Scheme 4). Fig. 2. 119 Additional spectroscopic features of compound 9 also support such structure. For example the MS of the oxidized derivative 9 shows a molecular ion at m/z 416 (FAB). The 1H NMR shown the usual doublets for the vinylic protons at δ = 6.97 and 6.22 (J = 16.0 Hz) and the corresponding methyl singlets at δ = 2.22 (3H), 1.08 (6H) and 0.86 (3H). The 13C 1D and DEPT edited spectra allowed the partial assignment of the chemical shift of the remaining carbon atoms (protonated carbons) at δ = 35.4, 33.5, 29.7, 28.1, 25.8 (2C), 20.8 and 16.8. The singlets at δ = 197.5 (C-12), 70.5 (C-2), 65.8 (C-1) and 33.5 (C-3) matching the spectrum for such a structure. The tentative approach of the mechanism involved in the above mentioned dimerization is displayed in scheme 3. On the other hand, the new furenone 10 displayed at IR spectrum an absorption at 1762 cm–1, which clearly indicates an α, β unsaturated γ-lactone moiety. The absorbance at UV spectrum λmax = 295 nm, reflects an α, β, γ cromophore [21]. The proton spectrum showed a one proton quartet at δ = 7.13 4J = 1.2 Hz and a doublet methyl group at δ = 2.05. The former was converted to a singlet when the methyl protons at δ = 2.05 was decoupled. The 300 MHz 1H NMR showed in addition methyl singlets at δ = 1.58 (C-7), 1.30 (C-8) and 1.29 (C9). Both, the strong IR band at 1762 cm–1 and the 13CNMR singlet at δ = 170.4 indicate the presence of an enol lactone carbonyl. The DEPT edited spectrum allowed the observation of two methines at δ = 137.8 (C-11) and 36.2 (C-6). Also, two methylene carbons at δ = 37.7 and 17.3 assigned for C-4 and C-5 respectively. Finally, four methyl signals at δ = 10.7, 29.2, 28.1 and 29.6 for C-13, C-7, C-9 and C-8 respectively. The 1D NMR enabled us the assignment of the nonprotonated carbons as δ = 149.1 (C-10), 128.3 (C-12), 106.1 (C-2), 71.2 (C-14), 35.0 (C-3), and 29.6 (C-1). A new unexpected trichloro derivative 11 whose structure emerge from its observed MS molecular ions [m/z (M+) 120 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Eduardo Díaz et al. Fig. 3. O O O O O O ** 7 9a *O O O O O O 7 9a ' ** * O O O O O O O O * O ** O 1 O O 9 O O 9' 1 O ** O *O O O O Scheme 3 O O O O O O 11 6.0 Kcal m ol Scheme 4 Conclusions O O O O O -1 135.0 Kcal mo l -1 309, (M + 2) m/z 311, (M + 4) m/z 313, and (M + 6) m/z 316]. The proton NMR spectrum showed a one proton doublet of doublet δ = 7.12 (J = 15.5, 10.5 Hz) and an one proton doublet of doublet at δ = 6.41 (J = 0.5, 15.5) for CH-10 and CH-11 respectively. A new methylene signal appeared at δ = 4.22 instead of the usual methyl ketone singlet (δ = 2.31) observed in the 1HNMR spectrum of starting β-ionone 1 (Fig. 3). Additionally, we were able to observe a doublet of doublet at δ = 2.40 (J = 10.5, 0.5, CH-2) and three methyl singlets at δ = 1.30, 0.95 and 0.78 for C-7, C-8, and C-9 respectively. The 1D 13C NMR and DEPT edited spectra allowed the assignment of protonated carbons as four methines at δ = 148.0 (C-10), 129.1 (C-11), 48.1 (C-2) and 32.8 (C-6) respectively. Three methylenes at δ = 47.2, (C-13), 35.1 (C-4) and 15.9 (C-5). Three methyl signals were also observed at δ = 29.4, 22.5, 19.7 for C-8, C-7 and C-9 respectively. The assignment of the non protonated carbons at δ = 190.5 (C-12), 72.8 (C-14), 32.0 (C-3) and 29.8 (C-1) was performed using the COLOC [18] spectrum. The formation of compounds 10 and 11 may be inferred through isomerization [22-26] of the bond C6-C1 from compound 1' to 1 respectively and then the dichlorocarbene addition to the new formed double bond. On the other hand, the role of the catalyst used in this work, play an important role in the ratio of the products obtained as well as the versatility of the structures and yield. Graphics 1 and 2 are self explained and show the behaviour of the different catalyst and ratio of products. O 1 35.5 Kcal mo l -1 The successful use of several ammonium quaternary catalyst in the phase transfer reactions enabled us to obtain some interesting derivatives from the starting material (β-Ionone). The structures of the formed compounds where fully elucidated 2D 1H and 13C NMR from the adducts of the dichloro carbene addition to β-ionone Product2 Product3 Product4 121 Product5 Product6 Product9 80 70 60 Y i 50 e l d 40 30 % 20 10 0 N one I II III IV V VI V II V III IX X XI X II X III X IV XV X V I X V II X V III X IX XX XXI C atalystType Graphic 1 100 90 80 70 60 % 50 40 30 20 10 0 I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX XX XXI XXII C atalyst type Reaction Breakdow n Recovered Graphic 2. Reaction and Breakdown-recovered substrate. using 2DNMR, X-ray crystallography, chemical shifts and molecular modeling calculations. Experimental Melting points were determined with a Kofler Hot Stage apparatus and were not corrected. The NMR 1H and 13C spectra were recorded using Varian Unity 300 spectrometer operating at observation frequency of 300.0 MHz for 1H and 75.0 MHz for 13C. The 1H and 13C chemical shifts (δ) are given in ppm relative to tetramethyl silane (TMS). The COSY, NOESY, HETCOR, DEPT and COLOC spectra were recorded using the usual Varian Unity software. High resolution spectra were recorded on a Varian Unity 500 operating at 500.3 MHz for 1H and 125.0 MHz for 13C. The experiments were performed using an inverse detection 5 mm probe. The COSY, NOESY, HMQC and HMBC experiments were performed using the usual Varian Unity software. Mass spectra were recorded on instruments using CI/EI sources on a JEOL-JMS-AX505 HA and JEOL-JMS-10217. The IR spectra were performed on Nicolet FX-sx and Nicolet 55-X in film mode. The β-ionona was purchased from Aldrich Chemical and used as received. The ammonium quaternary catalyst were kindly provided by Akzo Chemical Chicago Ill. and they received the trade name usually used. Benzalconium Chloride 122 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) (1); Arquad HT-50 (II); Arquad 2C-75 (III); Arquad S-50 (IV); Arquad 316 (V); Ethoquad C-12 CB75 (VI); Propoquad C-12 O2A (VII); Arquad M2H TB80 (VIII); Arquad 16-25W (IX); Propoquad T-12 O2A (X); Tetramethyl ammonium acetate (XI); Arquad 2HT-75 (XII); Ethoquad O-12 (XIII); Arquad 12/50 (XIV); Arquad 16/29 (XV); Arquad T-50 (XVI); Arquad 2HT50 (XVII); Arquad 16/50 (XVIII); TEBAC (XIX); Dimethyldidecyl ammonium chloride (XX); Tetrabutyl ammonium chloride (XXI). 2,6,6-trimethyl-1-(2-acetyl-3,3-dichlorocyclopropyl) cyclohexene 4. Colorless crystals, mp 54-56 °C. C14H20OCl2. M.W. 274. 1HNMR. δ 0.96 2 (CH3-8); 1.01 s (CH3-9; 1.37 m (CH24); 1.60 m (CH2-5); 1.75 s (CH3-7); 2.05 m (CH2-6); 2.44 s (CH3-13); 2.86 (CH-10), 3J = 9.0, 5J = 1.3; 3.10 d J = 9 (CH11). 13C NMR. δ 196.3 O=C-12; 136.1, C-2; 128.6, C-1; 64.4, C-14; 44.1 ,C-11; 41.2, CH2-4; 38.3, C-10; 34.5, C-3; 32.8, CH2-6; 32.1, CH3-13; 28.7, CH3-8; 28.6, CH3-9; 21.1, CH3-7; 19.5, CH2-5. IR νmax cm–1 2967, 2934, 2869, 1718, 1170, 842. MS EI m/z M+ 274 (13); M++2, 276 (7); M++4, 278 (3); 231 (87); 195 (49); 43 (100). 7,7-dichloro-2,5,6-trimethyl-1-(3-oxo-1-butenyl)bicyclo[4,1,0] heptane. 5. Colorless liquid C14H20OCl2 MW 274. 1HNMR δ 1.01, s CH -9; 1.22, s CH -8; 1.23, s CH -7; 1.42, 3 3 3 m CH2-4; 1.84, m CH2-5; 2.02, m CH2-6; 2.28, s CH3-13; 6.16 d, CH-11 3J = 16.5; 6.63, d CH-10 3J = 16.5. 13C NMR δ. 197.7 O=C-12; 143.2, C-10; 136.1, C-11; 76.3, C-14; 40.9, C3; 35.9, C-4; 33.8, C-2; 32.1, C-1; 30.2, CH3-8; 28.2, C-6; 27.4, CH3-13; 25.7, CH3-7; 22.7, CH3-9; 17.9, C-5. IR νmax cm–1; 2959, 2935, 2870, 1678, 1253, 837. MS. EI m/z M+ 274 (3); M++2, 276 (1); M++4, 278; 259 (27), 196 (38); 181 (88); 161 (53); 123 (100); 43 (97). 2-chloro-5-methyl-3-(2,6,6-trimethyl-1-cyclohexenyl) furane. 6. Colorless liquid. C14H19OCl, MW 238 1HNMR. δ 0.91, s CH3-8; 1.00, s CH3-9; 1.43, s CH3-7; 1.53, m CH2-4; 1.77, m CH2-5; 2.04, m CH2-6; 2.25, d , 4J = 1.0, CH3-13; 5.80, q, 4J = 1.0, CH-11. 13C NMR δ 149.9, C-12; 132, C-2; 131.1, C-14; 130.1, C-1; 119.5, C-10; 111.0, C-11; 39.1, C-4; 34.9, C-3; 32.0, C-6; 29.1, C-8; 28.3, C-9; 20.8, C-7; 19.3, C5; 13.6, C-13. IR νmax cm–1. 2960, 2931, 2867, 2832, 1614, 1236. MS EI m/z M+ 238 (38), M++2, 240 (15); 223, (100); 187 (34); 159, (24); 145, (31); 129, (53); 115, (40), 91, (50); 77, (49); 65, (35). 4-[4,4,4b, 8,8, 10a-hexamethyl-4a (3-oxo-but-1-enyl)-decahydro-9,10- dioxa-phenanthren-8a-yl]-but-3-en-2-one. 9. Colorless liquid C26H40O4 MW 416. 1H NMR δppm 0.86, s CH3-9; 1.08, s CH3-8; 1.00 m, CH2-4; 1.08 s CH3-7; 1.37 m CH2-5; 1.77, CH2-6; 2.22, s CH3-13; 6.22 d 3J = 16.0 CH-11; 6.97 d 3J = 16.0 CH-10. 13C NMR δ 197.5 C-12; 142.6 C-10; 132.4 C-11; 70.5 C-2,65.8 C-1; 35.4 C-4; 33.5 C-3; 29.7 C-6; 28.1 CH3-13; 25.8 (2C) CH3-8, CH3-9; 20.8 CH3-7; 16.8 C-5. IR νmax cm–1 2960, 2935, 2872, 1677, 1627, 1255. MS FAB m/z M+ 416, 400. 383, 355, 341, 327, 209, 191, 123(100), 69, 43. Eduardo Díaz et al. 5-(7,7-dichloro-1,3,3-trimethylbicyclo[4,1,0]heptyl-2-iden)3-methyl-Furan-2-one. 10. Colorless solid m. p. 80-81 ºC C15H18Cl2O2 MW 300. UV λmaxnm 295. 1H NMR δppm 1.29, s CH3-9; 1.30, s CH3-8; 1.43, m CH2-4; 1.58, s CH3-7; 1.68, q CH2-6; 2.10, m CH2-5; 2.05, q CH3-13; 7.13, q H-11. 13C NMR δ 170.4 C-15; 149.1 C-10; 137.8 C-11; 128.3 C-12; 106.1 C-2; 71.2 C-14; 37.7 C-4; 36.2 C-6; 35.0 C-3; 29.6 C-8; 29.2 CH3-7; 28.1 CH3-9; 27.2 CH3-7; 17.3 C-5; 10.7 CH3-13. IR νmax cm–1 2932, 2867, 1762,1207, 802, 741. MS EI, m/z M+ 300 (51), M++ 2, 302 (35); M++ 4 304 (7); 285, 265, 244, 217, 169 (100); 142 (78), 138 (60). 7,7-dichloro-1,3,3-trimethyl-2-(3-oxo-4-chloro-1butenyl)bicyclo[4,1,0] heptane. 11. 1H NMR δppm. 0.78 (s) CH3-9; 0.95 (s) CH3-8; 1.30 (s) CH3-7; 1.23, (m) CH2-4; 1.40 dd, J=10.0, 8.0, CH-6; 1.93 (m) CH2-5; 2.38 (m) CH-2; 4.22 CH2-13; 6.41, dd J=0.5, 15.5, CH-11; 7.12, dd J=10.5, 15.5, CH-10. 13C NMR δ 190.5 (C-12), 148.0, (C-10); 129.1 (CH11); 72.8 (C-14); 48.1 (CH-2); 47.2 (CH2-13); 37.8 (CH-6); 35.1 (CH 2-4); 32.1 (C-3); 29.8 (C-1); 29.4 (CH 3-8); 22.5 (CH3-7); 19.7 (CH3-9); 15.9 (CH2-5). MS CI M+ m/z 309 (51), M + 2 m/z 311 (42); M + 4 m/z 313 (17); M + 6 m/z 316 (6); 273 (86); 237 (70); 217 (80); 169 (100); 123 (58). IR νmax cm–1. 2957, 2930, 2867, 1715, 1697, 838. Acknowledgements We thank M. I. Chávez and B. Quiroz for the NMR determinations. We also thank R. Patiño, L. Velasco and R. A. Toscano for the IR, MS and X-ray determinations, respectively. L. Muciño thanks to CGI y EA of UAEM and SNIConacyt for partial financial support. We thank also E. Rivera of Akzo Chem. Chicago Ill. For the samples of ammonium quaternary catalyst used in this work. References 1. a) Aasen, A. J.; Kimland, B.; Almquist, S.D.; Enzell, C.R. Acta Chem. Scand. 1972, 26, 2573-2576. b) Wahlberg, I.; Enzell, C. R. Nat. Prod. Rep. 1987, 4, 237-276. 2. Kernan, M. R.; Faulkner, D. J.; Jacobs, R. S. J. Org. Chem. 1987, 52, 3081-3083. 3. Demole, E.; Enggest, P.; Winter, M.; Furrer, A.; Sculte-Elte, K. H.; Egger B.; Ohloff, G.; Helv. Chem. Acta 1979, 62, 67-75. 4. Behr, D.; Wahlemberg, I.; Nishida, T.; Enzell, C. R. Acta Chem. Scand. Ser. B 1977, B31, 609-613. 5. a) Buchi, G.; Vederas, J. C. J. Amer. Chem. Soc. 1972, 94, 91289132. b) Demole, E.; Berthet, D. Helv. Chim. Acta 1981, 54, 681-686. c) Snowden, R. L.; Linder, S. M.; Muller, M.L.; Shulte-Elte, K .H. Helv. Chim. Acta 1987, 70, 1858-1878. 6. Findlay, J. A.; MacKay, W.D. Can. J. Chem. 1971, 49, 23692371. 7. a) Roberts, D. L.; Heckman, R. A.; Hege, B. P.; Bellin, S. A. J. Org. Chem. 1968, 33, 3566-3569. b) Ontani, T.; Yamashita, K. Tetrahedron Lett. 1972, 2521-2524. 8. Demlov, E. V.; Prashad, M. J. Chem. Res. 1982, 354. 2D 1H and 13C NMR from the adducts of the dichloro carbene addition to β-ionone 9. Baird, M. S., Baxter, A. G. W.; Devling, B. R. J.; Searle, R. J .G. J. Chem. Soc. Chem. Commun. 1979, 210-211. 10. Sydnes, L. K. Acta Chem. Scand. Ser B 1977, 31, 823-828. 11. Sydnes, L. K.; Skattebold, L. Tetrahedron Lett. 1975, 4603-4606. 12. Díaz, E.; Toscano, R. A.; Alvarez A.; Shoolery, J. N.; Jankowski: Can. J. Chem. 1990, 68, 701-704. 13. Díaz, E.; Fuentes, A.; Villafranca, E.V.; Jankowski, K Acta Crystallographica 1994, C50, 2030-3032. 14. Doddrell, D. M.; Pegg, D.T.; Bendall, M. R. J. Magn. Reson. 1982, 48, 323-327. 15. a) Bax, A.; Subramanian, S. J. Magn. Reson. 1986, 67, 565-569. b) Bax, A.; Freeman, R. J. Magn. Reson. 1981, 44, 542-561. 16. a) Derome, A. E. in: Modern Techniques for Chemistry Research, Pergamon Press, N.Y. 1987, p 240. b) Martin, G. E.; Sektzer, A. S. in: Two Dimensional NMR Methods for Establishing Molecular Connectivity, VCH Publishers, N.Y. 1988, p. 219. 17. a) Bax, A.; Davies, D. G. J. Magn. Reson. 1985, 63, 207-213. b) Bax, A.; Summers, M. F. J. Amer. Chem. Soc. 1986, 108, 2093-2094. 123 18. Kessler, H.; M. Gehrke, M.; Griesinger, C. Angew. Chem. Int. Ed. Engl., 1988, 27, 490-536. 19. a) Furst, A.; Pretsch, E.; Robien, W. Ann. Chem. Acta 1990, 233, 213. b) Furst, A.; Pretsch, E. Ann. Chim. Acta 1990, 229, 17. c) Furst, A.; Pretsch, E.; Robien, W. Ann. Chim. Acta 1991, 248, 415. 20. Calculations were first minimized to 0.1 Kcalmol–1, using semiempirical method CS Chem 3D version 5 for MacIntosh. 21. Silverstein, R. M.; Bassler, G. C.; Merril, T. C. Identificación Espectroscópica de Compuestos Orgánicos, Editorial Diana. México 1980 pp 17-85. 22. Wahlemberg, Enzell, C. R. Nat. Prod. Reports. 1987, 4, 237-276. 23. Bucherer, R.; Hamm, P.; Eugster, C. H. Helv. Chim. Acta 1974, 57, 631-656. 24. Molnar, P.; Szaboles, J. Acta Chim. Acad. Sci. 1979, 99, 155. 25. Bischofberger, N.; Frei, B.; Wirz, J. Helv. Chim. Acta. 1983, 66, 2489. 26. Horspool, W. M. Photochem. 1985, V16, 248. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 124-126 Investigación Reactivity of IrH2{C6H3-2,6-(CH2PBut2)2} towards alkene compounds Valente Gómez-Benítez, Rocío Redón, and David Morales-Morales* Instituto de Química, Universidad Nacional Autónoma de México, Cd. Universitaria, Circuito Exterior, Coyoacán, 04510 México D. F. *E-mail: [email protected] (D. Morales-Morales) Dedicated to Dr. Alfonso Romo de Vivar Recibido el 22 de noviembre del 2002; aceptado el 20 de marzo del 2003 Abstract. The reactions of IrH2{C6H3-2,6-(CH2PBut2)2} with ethylene and 1-octene have been carried out, products containing the olefin compounds coordinated in a η2 fashion have been obtained. Both complexes have been characterized by multinuclear NMR probing unequivocally the proposed formulations. Examination of the thermal stability of both complexes under catalytic conditions shows the 1-octene adduct to be more stable. Keywords: PCP pincer ligands, iridium complexes, olefin complexes, dehydrogenation, catalysis. Resumen. Se llevaron a cabo reacciones del complejo IrH2{C6H32,6-(CH2PBut2)2} con etileno y 1-octeno, obteniéndose los derivados olefínicos coordinados de forma η2. Ambos complejos fueron caracterizados por RMN multinuclear, probando las formulaciones propuestas. El examen de la estabilidad térmica de ambos complejos bajo condiciones catalíticas mostró que el aducto 1-octano es más estable. Palabras clave: ligandos tipo pinza PCP, complejos de iridio, complejos olefínicos, deshidrogenación, catálisis. Introduction lived, reactions with simple alkenes might shed light in both the initial stages of the olefin isomerization process and the deactivation of the catalyst by product inhibition reactions which have not yet been explored. Thus, in this paper we wish to present the results obtained from the reactions of IrH2 {C6H3-2,6-(CH2PBut2)2} with ethylene and 1-octene. Terminal alkenes (alpha-olefins) are a major feedstock for the production of plastics, detergents, and lubricants. Their production through the selective dehydrogenation of linear alkanes would be an attractive alternative to the present commercial processes based on hydrogen, ethylene, and trialkylaluminum catalysts.[1] The iridium PCP pincer complexes: IrH2{C6H3-2,6-(CH2PR2)2} (R = But, 1; Pri, 2) are extraordinarily active and robust catalysts for aliphatic dehydrogenation reactions.[2-5] Recent studies of this reactivity lead to the discovery of the first efficient catalytic system for the selective dehydrogenation of n-alkanes to alpha-olefins [6]. While this system serves to validate the concept of producing alphaolefins through this method, it suffers from several practical limitations. The concentrations of the dehydrogenated products were found to quickly attain a low (1.5-4.0 %), constant value. Furthermore, the high selectivity for alpha-olefins was found to be short lived as the complexes show secondary catalytic activity for alkene isomerization and the alkene distribution rapidly shifted towards the internal isomers.[6] Finally, the requisite consumption of a stoichiometric amount of a sacrificial hydrogen acceptor used in the reported selective dehydrogenation of n-alkanes is economically and environmentally unattractive. Liu and Goldman had previously achieved the thermochemical dehydrogenation of linear alkanes by 2 without the use of a hydrogen acceptor. However, only internal alkenes were produced in their experiments [4]. As a part of our continuous interest in the reaction mechanism ruling in this process we have tried to identify the probable olefin intermediate species during the reaction, although some of these complexes are extremely reactive and short- Experimental Materials and methods Unless stated otherwise, all reactions were carried out under an atmosphere of argon using conventional Schlenk glassware and Young NMR tubes. Solvents were degassed and dried using standard procedures. The 1H NMR spectra were recorded on a Varian Unity Inova 400 spectrometer. Chemical shifts are reported in ppm down field of TMS using the solvent as internal standard (cyclohexane-d12, δ 1.38). 13C and 31P NMR spectra were recorded with complete proton decoupling and are reported in ppm downfield of TMS with solvent as internal standard (cyclohexane-d12, δ 26.45) and external 85 % H3PO4 respectively. Elemental analyses were determined on a PerkinElmer 240. Positive-ion FAB mass spectra were recorded on a JEOL JMS-SX102A mass spectrometer operated at an accelerating voltage of 10 Kv. Samples were desorbed from a nitrobenzyl alcohol (NOBA) matrix using 3 KeV xenon atoms. Mass measurements in FAB are performed at a resolution of 3000 using magnetic field scans and the matrix ions as the reference material or, alternatively, by electric field scans with the sample peak bracketed by two (polyethylene glycol or cesium iodide) reference ions. The 1-octene was purchased Reactivity of IrH2{C6H3-2,6-(CH2PBut2)2} towards alkene compounds 125 from Aldrich Chemicals Co. and used without further purification. The complex, IrH2{C6H3-2,6-(CH2PBut2)2} (1) was synthesized by the literature method [2]. (bs, ArC), 32.79 (bs, CH2P), 32.40 (bs, CH2=CH-R), 32.24 (bs, PC(CH3)2), 30.29 (s, PC(CH3)3), 23.42 (bs, -(CH2)n-), 14.41 (bs, -(CH2)n-CH3); 31P NMR (161.93 MHz, cyclohexane-d12) δ = 54.97 (s, 1P). Anal calcd for C32H59P2Ir1 (697.98) C, 55.07 %; H, 8.52 %. Found: C, 54.94 %; H, 8.47 %. Synthesis of Ir(η2-CH2=CH2){C6H3-2,6-(CH2PBut2)2} (3) A solution consisting of 5 mg (8.5 × 10 –3 mmol) of IrH2{C6H3-2,6-(CH2PBut2)2} (1) and 1 mL of cyclohexaned12, was freeze-pump-thaw degassed 3 times. The solution was then treated with excess of ethylene at room temperature. An immediate change from orange to deep red-brown is observed. Removal of the solvent in vacuo affords Ir(η2CH2=CH2){C6H3-2,6-(CH2PBut2)2} as a deep red-brown solid in nearly quantitative yield (based upon 31P NMR). 1H NMR (400.03 MHz, cyclohexane-d12) δ = 1.253 (vt, JHP = 5.8 Hz, 36H, PC(CH3)3), 3.049 (vt, JHP = 3.2 Hz, 4H, CH2PC(CH3)3), 3.287 (s, 4H, CH2=CH2), 6.825 (t, JHH = 7.4 Hz, 1H, arom), 7.026 (d, JHH = 7.2 Hz, 2H, arom); 13C NMR (100.59 MHz, cyclohexane-d12) δ = 178.28 (s, ArC), 153.84 (vt, J = 9.15 Hz, ArC), 122.81 (s, ArC), 119.77 (vt, JPC = 7.7 Hz, ArC), 40.84 (vt, JPC = 13.53 Hz, CH2P), 37.968 (s, CH2=CH2), 36.91 (vt, J PC =8.4 Hz, PC(CH 3 ) 2 ), 31.02 (s, PC(CH 3 ) 3 ); 31 P NMR (161.93 MHz, cyclohexane-d12) δ = 54.68 (s, 1P). Anal calcd for C26H47P2Ir1 (613.82) C, 50.87 %; H, 7.72 %. Found: C, 50.76 %; H, 7.70 %. Synthesis of Ir(η2-CH2=CH(CH2)5CH3){C6H3-2,6(CH2PBut2)2} (4) IrH2{C6H3-2,6-(CH2PBut2)2} (1) (5 mg, 8.5 × 10–3 mmol) was dissolved in 1 mL of 1-octene, an immediate release of hydrogen was observed. A change in color from deep orange to bright yellow was also observed, the reaction was stirred at room temperature for further 5 min after this time the excess of 1-octene is evaporated under vacuo for 36 h to yield a Ir(η 2 -CH 2 =CH(CH 2 ) 5 CH 3 ){C 6 H 3 -2,6-(CH 2 PBu t 2 ) 2 } as a bright yellow microcrystalline powder in nearly quantitative yield (based upon 31P NMR). 1H NMR (400.03 MHz, cyclohexane-d12) δ = 0.895 (bs, 10H, -(CH2)n-), 1.30 (bs, 36H, PC(CH 3 ) 3 ), 1.56 (bs, 3H, -(CH 2 ) n -CH 3 ), 2.00 (bs, 4H, CH 2 PC(CH 3 ) 3 ), 2.18 (bs, 2H, CH 2 =CH-R), 2.27 (s, 1H, CH2=CH-R), 6.50 (t, JHH= 7.3 Hz, 1H, arom), 6.64 (d, JHH= 7.2 Hz, 2H, arom); 13C NMR (100.59 MHz, cyclohexane-d12) δ = 150.0 (bs, ArC), 148.0 (bs, ArC), 122.8 (bs, ArC), 121.0 PBut2 Ir PBut2 H2 H Ir + H t PBut2 PBu 2 PBut H2 H Ir + PBut Ir H PBut Scheme 1 PBut Results and discussion Synthesis and characterization of Ir(η2-CH2=CH2) {C6H3-2,6-(CH2PBut2)2} (3) and Ir(η2CH2=CH(CH2)5CH3){C6H3-2,6-(CH2PBut2)2} (4) The reaction of the PCP pincer complex IrH 2{C 6H 3-2,6(CH2PBut2)2} (1) with excess of ethylene or 1-octene affords complexes Ir(η2-CH2=CH2){C6H3-2,6-(CH2PBut2)2} (3) and Ir(η 2-CH 2=CH(CH 2) 5CH 3){C 6H 3-2,6-(CH 2PBu t2) 2} (4) as unique products (Scheme 1) in quantitative yields as red brown or bright yellow powders respectively. The 1H NMR spectrum of 3 clearly shows a singlet at 3.29 ppm corresponding to the ethylene molecule coordinated in a η2 fashion to the metal center, moreover the fact that only one signal is observed for this molecule implies that is placed in a highly symmetrical environment. Besides the presence of the coordination of the ethylene, the signals corresponding to the presence of the PCP pincer ligand can also be clearly identified. Thus, a signal corresponding to the methyl groups in the PC(CH3)3 can be observed at 1.25 ppm. A virtual triplet due to the CH2 group can be observed at 3.5 ppm, the multiplicity of the signal being due to the coupling of the protons on the CH2 group with the phosphorous nuclei in CH2PC(CH3)3). A signal corresponding to the aromatic proton 4-H can be observed as a triplet centered at 6.83 ppm, while that corresponding to the aromatic protons 3,5-H is located at 7.03 ppm, no signals corresponding to the presence of metal-hydrides where detected at higher field. Analogously, signals in the 1H NMR spectrum of 4 corresponding to the presence of the 1octene coordinated in a η2 fashion can be observed as broad singlets in 2.18 and 2.27 ppm, other signals corresponding to the rest of the coordinated 1-octene molecule can be observed at 0.9 ppm for the -(CH2)n- groups and a broad singlet at 1.56 ppm which can be assigned to the terminal -(CH2)n-CH3, signals due to the presence of the PCP pincer ligand can be observed at similar chemical shifts as those observed for the analogous complex with ethylene. As is the case for 3, no signals for the presence of hydride ligands were detected at higher field. The 13C NMR spectrum of 3 exhibits all the signals expected for the proposed formulation, it is noteworthy that the signal at 37.97 ppm corresponding to the coordinated ethylene is a singlet, thus the same conclusion regarding the molecule to be in a highly symmetrical environment can be deducted. The 13C NMR spectrum of 4 exhibits signals corresponding to the presence of the aliphatic moiety and the tertbutyl groups in the PCP pincer ligand as well as those for the 126 Rev. Soc. Quím. Méx. Vol. 47, Núm. 1 (2003) CH2 directly coordinated to the P centers. It is noteworthy the presence of a signal at 32.40 ppm which evidences the presence of the coordinated olefin, the shift to higher field of this particular signal clearly illustrates the protecting effects of the aliphatic chain moiety in the case of the 1-octene adduct 4. In both cases, the 31P NMR spectra shows a unique signal at 54.7 and 54.97 ppm for 3 and 4 respectively which is consistent with a trans configuration for both phosphorus nuclei in the two molecules. Elemental analysis are also consistent with the proposed formulations. The FAB+-Mass spectra of 3 exhibits a peak at 585 M/z [M+-H2C=CH2] corresponding to the loss of the ethylene molecule while that of 4 shows the molecular ion at [M+ = 698 M/z]. Both complexes have been exposed to catalytic conditions where the reaction temperature reaches 200 °C, at this temperature the complex containing ethylene, releases the ethylene molecule which makes this complex an excellent candidate for further studies oriented to the possible functionalization of this molecule, however the complex containing the 1octene it is resistant even at this temperatures, therefore it can be conclude that it is precisely this stability the problem (product inhibition) we have to go against in order to optimize the present system for the dehydrogenation process, other alternatives will involve the design of new ligands where electronic and steric factors could be tuned in such way that the elimination of the alkene molecule could be carried out easier and faster. Efforts aimed to achieve this goals are currently under investigation in our laboratory. Valente Gómez-Benítez et al. Acknowledgements V. G.-B. would like to thank CONACyT and DGAPA for financial support. We would like to thank Chem. Eng. Luis Velasco Ibarra and M. Sc. Francisco Javier Perez Flores for their invaluable help in the running of the FAB-Mass Spectra. The support of this research by CONACyT (J41206-Q) and DGAPA-UNAM (IN116001) is gratefully acknowledged. References 1. Behr, A. Ullmann's Encyclopedia of Industrial Chemistry, 5th edn.; Elvers, B.,Hawkins, S., Russey, W., Eds.; VCH Verlagsgesellschaft: Weinheim, 1994, pp 242-249. 2. (a) Gupta, M.; Hagen, C.; Kaska, W. C.; Flesher, R.; Jensen, C. M. Chem. Commun., 1996, 2083-2084. (b) Gupta, M.; Hagen, C.; Kaska, W. C.; Cramer, V.; Jensen, C. M. J. Am. Chem. Soc., 1997, 119, 840-841. (c) Gupta, M.; Kaska, W. C.; Jensen, C. M. Chem. Commun., 1997, 461-462. 3. Xu, W. -W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C. Krough-Jespersen, K.; Goldman, A. S. Chem. Commun., 1997, 2273-2274. 4. Liu, F.; Goldman, A. S. Chem. Commun., 1999, 655-656. 5. Jensen, C. M. Chem. Commun., 1999, 2443-2449. 6. Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086-4087. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 127-129 Investigación α-phenylethyl]-diamines as Phosphorylated Chiral The Use of N,N’-Di[α Derivatizing Agents for the Determination of the Enantiomeric Purity of Chiral Secondary Alcohols Gloria E. Moreno,1,2 Virginia M. Mastranzo,1,2 Leticia Quintero,2 Cecilia Anaya de Parrodi,*,1 and Eusebio Juaristi*, 3 Centro de Investigaciones Químico Biológicas, Universidad de las Américas-Puebla, Santa Catarina Mártir, Cholula, 72820 Puebla, México. E-mail: [email protected] 2 Centro de Investigación de la Facultad de Ciencias Químicas, Universidad Autónoma de Puebla, 72570 Puebla, México. 3 Departamento de Química, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, 07000 México, D. F. E-mail: [email protected] 1 Recibido el 17 de febrero del 2003; aceptado el 26 de marzo del 2003 This paper is dedicated to Dr. Alfonso Romo de Vivar in appreciation of his contributions to chemistry Abstract: The influence of the framework structure in C2-symmetric N,N’-di[α-phenylethyl]diamines as chiral derivatizing agents, on their effectiveness for the determination of the enantiomeric composition of chiral secondary alcohols by 31P-NMR spectroscopy of derived diastereomeric phosphonamides, is described. Key words: chiral derivatizing agents, enantiomeric composition, C2-symmetric diamines, 31P-NMR spectroscopy. Resumen: Se describe la influencia de la estructura de N,N’-di[αfeniletil]-diaminas con eje de simetría C2, como agentes derivatizantes quirales en la determinación de la composición enantiomérica de alcoholes secundarios quirales empleando RMN de 31P. Palabras clave: Agente derivatizante quiral, composición enantiomérica, diaminas con simetría C2, RMN de 31P. Introduction 1,2-diamine, A, and N,N’-di[(S)-α-phenylethyl]propane-1,3diamine, B, (Fig. 1) have been reported as effective and inexpensive CDAs [6]. In addition, the use of chiral diamines based on enantiopure trans-1,2-diaminocyclohexane as CDAs, such as C has been demonstrated [7]. Furthermore, our group also reported the use of trans-N,N’-di-[(S)-α-phenylethyl]cyclohexane-1,2-diamines, D and E as convenient CDAs [8]. Chiral derivatizing agents (CDAs) for NMR spectroscopy represent one of the most effective tools to satisfy the great demand for rapid and reliable methods for the determination of the enantiomeric composition of chiral substrates [1]. Their use involves the derivatization reaction of an enantiomerically pure chiral auxiliary with the substrates to be analyzed, in order to obtain diastereoisomeric products presenting anisochronous absorptions in their NMR spectra. Efforts directed to the development of fast and reliable CDAs continue [2]. In this context, the high sensitivity afforded by 31P NMR spectroscopic methods makes attractive the use of phosphorus-containing derivatives towards this goal [3]. Chiral diamines have been shown to be useful chiral reagents and ligands for chemical catalysis, with especial application in asymmetric synthesis [4]. By the same token, (R)- and (S)-α-phenylethylamine are simple, yet powerful stereodifferentiating auxiliaries in organic transformations [5]. Presently, there exist several reports in the literature describing the use of diamines containing diverse N-substituents as CDAs, which are based on the synthesis of chiral phospholidines for the determination of enantiomeric composition of chiral alcohols, amines, carboxylic acids, halohydrins and thiols. In particular, N,N’-di[(S)-α-phenylethyl]ethane- Ph S NH HN S Ph Ph S A NH HN B R R CH3 NH NH CH3 C S S Ph S NH S NH Ph D Figure 1. S Ph S R R Ph Ph NH NH S E 128 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Ph S NH HN 1 Ph R HN Ph R S Ph Gloria E. Moreno et al. Table 1. Synthesis of P-chloro-1,3-diazaphospholidines. Ph R NH R Ph 2 S Ph Entry S NH NH S Ph S S R S NH NH R Ph Ph S 4a 1 2 3 4 5 S 3 S Ph Diamine Ph NH a NH S 1 2 3 4a 4b Reaction time, h 0.5 0.5 24 0.5 0.5 Conversion %a δ (ppm) 31P-NMR of Phospholidine 100 100 23 100 100 165.21 168.93 166.50 177.40 175.25 The percent of conversion were determined from the 31P-NMR spectra in relation to the amount of PCl3. vorable, because of strain in the trans-fused five-membered bicyclic phospholidine formed. Gratifyingly, the P-Cl bond of the phospholidines 2, 4a and 4b were readily cleaved, upon treatment with the carbinols of interest. So, derivatization was carried out by addition of 0.8 equiv of the racemic secondary alcohols, 5a-d, to afford the respective P-alkoxy-1,3-diazaphosphonamides (Table 2). The all-(S) diamine 4a afforded the largest differences of chemical shifts (∆δ) in the 31P NMR spectra of the diastereomeric phosphonamides. Nevertheless, the P-Cl bond of the P-chloro-1,3-diazaphospholidine derived from 1 was specially resistant upon treatment with carbinols. Thus, diamines 1 and 3 were not useful as CDAs. In summary, the P-chloro-1,3-diazaphospholidines derived from chiral diamines 2, 4a, and 4b, incorporating N(α-phenylethyl) substituents, are convenient chiral derivatizing agents for the determination of the enantiomeric purity of chiral alcohols. The quantitative and fast P-Cl bond cleavage upon alcoholysis leads to phosphonamide formation, directly in the NMR tube prior to measurement. Large differences in 4b Figure 2. Results and discussion The use of C2-symmetric N,N’-di[α-phenylethyl]-diamines 1, 2, 3, 4a and 4b as chiral ligands (Fig. 2) will be reported [9]. In the present work their application as chiral derivatizing agents (CDAs), via the formation of the corresponding P-chloro-1,3diazaphospholidines is described. These CDAs can be prepared directly in the NMR tube employed for 31P NMR analysis [8]. To this end, each diamine in CDCl3 solvent was treated with one equivalent of PCl3 in CH2Cl2 at room temperature to give the corresponding phospholidines as intermediates. The formation of the P-chloro-1,3-diazaphospholidines of interest was very fast and quantitative with diamines 1, 2, 4a, and 4b, but rather slow and incomplete with diamine 3 (Table 1). Apparently, the reaction of diamine 3 with PCl3 is also unfa- Table 2. 31P NMR data of P-alkoxy-1,3-diazaphospholidines derived from secondary alcohols 5a-d recorded in CDCl3. Entry 1 2 3 4 5 6 7 8 9 Diamine R*OH Diastereoisomeric P-OR, ∆δ A lit.6 ∆δ B lit.6 ∆δ C lit.7a ∆δ D lit.8 ∆δ E lit.8 ∆δ 4a 4b 4a 4b 4a 4b 2 4a 4b 5a 1.48 0.44 1.61 0.37 2.79 0.21 0.44 1.21 0.18 — 2.73 0.27 1.07 0.38 — — 0.27 0.98 0.31 0.39 3.69 — 6.34 1.01 0.20 1.38 0.40 4.71 0.58 5b 5c 5d The Use of N,N’-Di[α-phenylethyl]-diamines as Phosphorylated... the 31P NMR chemical shifts (∆δ) for the diastereomeric phosphonamides were observed, allowing accurate integration and quantitative determination of the diastereomeric ratios. Experimental section 31P NMR spectra were measured on a Varian Mercury-200 MHz spectrometer. Chemical shifts are given as δ values (ppm). All reagents were purchased from Aldrich Chemical Co. General Procedure for Chiral Alcohol Derivatization In an NMR tube are placed with vigorous stirring 0.16 mmol of free diamine, 0.5 mL of CDCl3, 119 mg (0.80 mmol) of diethylaniline, and 23 mg (0.16 mmol) of PCl3 previously dissolved in 50 µL of CH2Cl2, affording the chlorodiazaphospholidine and the 31P NMR spectra are recorded (Table I). Immediately after, 0.13 mmol of racemic secondary alcohols is added, and the resulting mixture is stirred for 30 minutes before 31P NMR spectra are recorded at room temperature (Table 2). Acknowledgment We thank CONACyT for financial support (Projects No. 32202-E and 33023-E, and Grants No. 91275 and 144937). 129 References 1. (a) For a detailed review, see: Parker, D. Chem. Rev. 1991, 91, 1441-1457. See, also: (b) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-519. (c) Pirkle, W. H.; Hoover, D. J. Top. Stereochem. 1982, 13, 263-331. (d) Benson, S. C.; Cai, P.; Colon, M.; Haiza, M. A.; Tokles, M.; Snyder, J. K. J. Org. Chem. 1988, 53, 5335-5341. (e) Jursic, B. S.; Zdravkovski, Z.; Zuanic, M. Tetrahedron: Asymmetry 1995, 5, 1711-1716. 2. (a) Uccello-Barreta, G.; Bernardini, R.; Lazzaroni, R.; Salvadori, P. Org. Lett. 2000, 2, 1795-1798. (b) Reymond, S.; Brunel, J. M.; Buono, G. Tetrahedron: Asymmetry 2000, 11, 1273-1278. (c) Alexakis, A.; Chauvin, A.-S. Tetrahedron: Asymmetry 2001, 12, 1411-1416. 3. (a) Verkade, L. D.; Quin, L. D. Phosphorus-31 NMR Spectroscopy in Stereochemical Analysis; VCH Publishers: Deerfield Beach, 1987. (b) See, also: Juaristi, E. Introduction to Stereochemistry and Conformational Analysis; Wiley: New York, 1991; pp 137-138. (c) Johnson, C. R.; Elliott, R. C.; Penning, T. D. J. Am. Chem. Soc. 1984, 106, 5019-5020. (d) Kato, N. J. Am. Chem. Soc. 1990, 112, 254-257. (e) Anderson, R. C.; Shapiro, M. J. J. Org. Chem. 1984, 49, 1304-1305. f) Hulst, R.; Kellogg, R. M.; Feringa, B. L. Rec. Trav. Chim. Pays-Bas 1995, 114, 115-138. 4. For excellent reviews, see: (a) Togni, A.; Venanzi, L. M. Angew. Chem., Int. Ed. 1994, 33, 497-526. (b) Bennani, Y. L.; Hanessian, S. Chem. Rev. 1997, 97, 3161. (c) Lucet, D.; Le Gall, T.; Miokowski, Ch. Angewandte Chem., Int. Ed. 1998, 37, 25802627. 5. (a) Jaen, J. In Encyclopedia of Reagents for Organic Synthesis, Paquette, L. A., Ed.; Wiley: New York, 1995, Vol. 5, pp 34273431. (b) Juaristi, E.; Escalante, J.; León-Romo, J. L.; Reyes, A. Tetrahedron: Asymmetry 1998, 9, 715-740. (c) Juaristi, E.; LeónRomo, J. L.; Reyes, A.; Escalante, J. Tetrahedron: Asymmetry 1999, 10, 2441-2495. 6. Hulst, R.; de Vries, K.; Feringa, B. L. Tetrahedron: Asymmetry 1994, 5, 699-708. 7. (a) Alexakis, A.; Mutti, S.; Mangeney, P. J. Org. Chem. 1992, 57, 1224-1237. (b) Alexakis, A.; Frutos, J. C.; Mutti, S.; Mangeney, P. J. Org. Chem. 1994, 59, 3326-3334. For other chiral derivatizing agents based on enantiopure trans-1,2-diaminocyclohexane, see: (c) Resch, J. F.; Meinwald, J. Tetrahedron Lett. 1981, 22, 3159-3162. (d) Staubach, B.; Buddrus, J. Angew. Chem., Int. Ed. 1996, 35, 1344-1346. 8. Anaya de Parrodi, C.; Moreno, G. E.; Quintero L.; Juaristi, E. Tetrahedron: Asymmetry 1998, 9, 2093-2099. 9. Mastranzo, V. M.; Quintero, L.; Anaya de Parrodi, C.; Juaristi, E.; Walsh, P. J. Unpublished results. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 130-131 Investigación Terpenoids and Flavones from Achillea falcata (Asteraceae) Maurizio Bruno,1 Sergio Rosselli,1 Rosa Angela Raccuglia,1 Antonella Maggio,1 Felice Senatore,2 Nelly Apostolides Arnold,3 Claire A. Griffin4 and Werner Herz4 Dipartimento di Chimica Organica, Universitá di Palermo, Viale de Scienze, Pardo d'Orleans II, 90128 Palermo, Italy Dipartimento Chimica Sostanze Naturali, Universitá Federico II, via D. Montesano, 49-80131 Napoli, Italy 3 Faculté des Sciences Agronomiques, Université Saint Esprit, Kaslik (Beirut), Lebanon. 4 Department of Chemistry and Biochemistry, The Florida State University, Tallahassee, FL 32306-4390, USA. Tel: (1)-850-644-2774; Fax: (1)-850-644-8281; E-mail: [email protected] 1 2 Recibido el 25 de febrero del 2003; aceptado el 2 de abril del 2003 Dedicated to Professor Alfonso Romo de Vivar, a valued collaborator during the early stages of his career Abstract. Aerial parts of Achillea falcata L. furnished the monoterpenes 3,7-dihydroxy-3,7-dimethyl-1,5-octadiene and 3,6-dihydroxy3,7-dimethyl-1,7-octadiene, the sesquiterpene lactone sintenin and the flavonoids 5-hydroxy-6,7,3', 4'-tetramethoxyflavone (6-hydroxyluteolin-6,7,3',4'-tetramethyl ether) and 5-hydroxy-6,7,8,3',4'-pentamethoxyflavone (desmethoxynobiletin). Keywords: Achillea falcata, Asteraceae, monoterpenes, sesquiterpene lactone, sintenin, flavonoids. Resumen. El análisis químico de las partes aéreas de Achillea falcata permitió la caracterización de los monoterpenos 3,7-dihidroxi-3,7dimetil-1,5-octadieno y 3,6-dihidroxi-3,7-dimetil-1,7-octadieno, la lactona sesquiterpénica sintenina y los flavonoides 5-hidroxi-6,7,3',4'tetrametoxi-flavona (6,7,3',4'-tetrametil éter de 6-hidroxi-luteolina) y 5-hidroxi-6,7,8,3',4'-pentametoxiflavona (desmetoxinobiletina). Palabras clave: Achillea falcata, asteraceae, monoterpenos, lactona sesquiterpénica, sintenina, flavonas. Our groups have previously described the chemistry of two Achillea species, A. ligustica All. from Sicily [1] and A. cretica L. from Cyprus [2]. We now report the results of our study of Achillea falcata L. from Lebanon. Aerial parts of A. falcata L. (syns. A. damascene DC, A. sulfurea Boiss.) were extracted at room temperature with acetone; the extract was purified by silica gel chromatography and radial chromatography to afford five compounds. Of these, 5-hydroxy-6,7,3',4'-tetramethoxyflavone(6-hydroxyluteolin-6,7,3',4'-tetramethyl ether) and 5-hydroxy-6,7,8,3'4'-pentamethoxyflavone (desmethoxynobiletin) were identified by MS and comparison of their 1H-NMR spectra with spectra in our files. Two others, the monoterpenes 3,7-dihydroxy-3,7dimethyl-1,5-octadiene (1) and its isomer 3,6-dihydroxy-3,7dimethyl-1,7-octadiene (2), have been previously reported from Cinnamum camphora [3]; diene 1 has also been isolated in our laboratories from Achillea ligustica All. [1] where its high resolution 1H NMR spectrum was reported. Doubling of the signals of H-1a, H-1b and H-2 in our 500 MHz 1H NMR spectrum of 2 (see Experimental section) indicated that it was a 1:1 mixture of C-3 epimers. The remaining constituent was the germacradienolide sintenin (3) first reported with incorrect C-9 stereochemistry from Achillea sintenisii Hub.-Mor. [4], a matter subsequently corrected with material from Achillea biebersteinii Afran (as A. micrantha Willd.) [5]. Sintenin has also been isolated from the near Eastern species A. aleppica DC. and pseudoaleppica Hub. Mor. [6], A. cucullata (Hausskn.) Bornm., A. goniocephala Boiss. et Bal. and A. vermicularis Trin. [7] as well as from A. teretifolia Willd. [8], all, like A. sintenisii, A. biebersteinii and now A. falcata, members of Achillea sect. Santolinoidea C. Koch [9] which suggests that sintenin might be a marker for the section. An exception is the Balkan species A. crithmifolia Waldst. et Kit. several collections of which [10-13] yielded a variety of sesquiterpene lactone types among which sintenin appeared only once [12]. OAc OH OH OH AcO OH O O 1 2 3 Experimental section General experimental procedures. Column chromatography was performed using Merck Si gel (No. 7734). 1H NMR spectra were obtained on a Varian Inova 500 MHz NMR spectrometer in CDCl3, whereas 13C NMR spectra were run on an IBM/Bruker WP27OSY NMR spectrometer at 67.5 MHz in CDCl3. Mass spectra were acquired on a JEOL MS Route 600 H instrument. Plant material. Aerial parts of Achillea falcata L. were collected at Jab. Kneissé, Lebanon at 1700 m s / l in July 2000. Reactivity of IrH2{C6H3-2,6-(CH2PBut2)2} towards alkene compounds 131 A voucher specimen (leg., det. and confirmed by N. Arnold s.n. is deposited in the herbarium of the Botanical Garden and the Botanische Museum, Freie Universität Berlin, Germany. C10H18O2- 2H2O + H, 135.1174; 1H NMR (CDCl3) δ 5.89 and 5.88 (both dd, J = 17.3, 10.8 Hz, H-2 of epimers A and B), 5.22 and 5.21 (both dd, J = 17.3, 1.4 Hz, H-1a of epimers A and B), 4.94 and 4.93 (both q, J = 4, 1 Hz, H-8a of both epimers), 4.83 and 4.82 (both q, 4 Hz, H-8b of both epimers), 4.04 (brq, 6.3 Hz, H-6 of both epimers, 1.70 (brs, 3H, H-8), 1.64-1.53 (c, 4H, H-4a,b H-5a,b), 1.28 s (3H, H-10). Extraction and isolation. Dried and powdered aerial parts (750 g) were extracted with acetone (3 × 5 l) at room temperature for one week each time. The extracts were combined and evaporated at reduced pressure and low temperature (35 °C) to give 58 g of residue. The residue was subjected to dry column chromatography over Si gel with a solvent gradient ranging from petroleum ether (bp 50-70 °C) to EtOAc (100 %) and finally with EtOAc-MeOH (19:1 and 9:1). The fraction eluted with petroleum ether-EtOAc (2:3) was resubmitted to chromatography using petroleum ether-EtOAc (4:1, 3:7 and 1:1) as eluent to afford several subfractions. The subfraction eluted with petroleum ether-EtOAc (3:70 weighing 250 mg was subjected to radial chromatography using CH2Cl2-MeOH (99:1) as eluent to afford, in order of increasing polarity, desmethokynobiletin (20 mg) identified by MS and 1H NMR spectrometry, sintenin (10 mg), identified by MS, 1H and 13C NMR spectrometry [5], and 10 mg of 2. The subfraction eluted with petroleum ether-ethyl acetate (1:1) weighing 200 mg was subjected to radial chromatography using CH2Cl2-MeOH (49:1) as eluent to afford in order of increasing polarity 60 mg of 5hydroxy-6,7,3',4'-tetramethoxyflavone and 45 mg of 1. 3,7-Dihydroxy-3,7-dimethyl-1,5-octadiene (1): Mass and 1H NMR spectra corresponded to data reported earlier. 3,6-Dihydroxy-3,7-dimethyl-1,7-octadiene (2): 1:1 mixture of C-3 epimers; oil, MS CI (isobutene) 153.1279 (25), 135.1174 (21.9); calcd for C10H18-O2H2O + H 153.1279; for References 1. Bruno, M.; Herz, W. Phytochemistry 1988, 27, 1871-1872. 2. Bruno, M.; Bondi, M. L.; Paternostro, M. P.; Arnold, N. A.; Diaz, J. G.; Herz, W. Phytochemistry 1996, 42, 737-740. 3. Takaoka, D.; Hiroi, M. Phytochemistry 1976, 15, 330-331. 4. Gören, N.; Öksüz, S.; Ulubelen, A. Phytochemistry 1988, 27, 2346-2347. 5. Hatam, N. A. R.; Yousif, N. J.; Porzel, A.; Seifert, K. Phytochemistry 1992, 31, 2160-2162. 6. Appendino, G.; Jakupovic, J.; Özen, A. C.; Schuster, A. Phytochemistry 1993, 34, 1171-1172. 7. Öksüz, S.; Gümüs, S.; Alpinar, K. Biochem. Syst. Ecol. 1991, 19, 439. 8. Öksüz, S.; Ulubelen, A.; Tuslaci, E. Fitoterapia 1990, 61, 283. 9. Davis, P. H. Ed., Flora of Turkey, Vol. 5, pp. 224-251, 1975. Edinburgh University Press. 10. Miloslavljevic, S.; Aljancic, I.; Macura, S.; Milinkovic, D.; Stefanovic, M. Phytochemistry 1991, 30, 3464-3466. 11. Miloslavljevic, S.; Macura, S.; Stefanovic, M.; Aljancic, I.; Milinkovic, D. J. Nat. Prod. 1994, 57, 64-67. 12. Todorova, M. N.; Markova, M. M.; Tsankova, E. T. Phytochemistry 1998, 49, 2429-2432. 13. Todorova, M. N.; Vogler, B.; Tsankova, E. T. Natural Prod. Lett. 2000, 14, 463-468. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 132-138 Investigación Preparation of 11-Hydroxylated 11,13-Dihydrosesquiterpene Lactones Howard G. Pentes,1 Francisco A. Macias2 and Nikolaus H. Fischer*1,a Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803, USA Present address: Department of Pharmacognosy, School of Pharmacy, University of Mississippi, University, MS 38677, USA. Tel: (662) 915-7026; Fax: (662)-915-6975; E-mail: [email protected] 2 Departamento de Química Orgánica, Facultad de Ciencias, Universidad de Cádiz, Apdo. 40,11080 Puerto Real, Cádiz, Spain 1 a Recibido el 4 de febrero del 2003; aceptado el 24 de abril del 2003 Dedicated to Professor Alfonso Romo de Vivar Abstract. Hydroxylations of the α-position of lactonic carbonyl groups of four different skeletal types (germacranolides, eudesmanolides, guaianolides, and elemanolides) of 11,13-dihydrosesquiterpene lactones were achieved by LDA-mediated generation of the corresponding lactone enolates and trapping with gaseous oxygen or with a chiral oxidizing agent, (camphorylsulfonyl)oxaziridine. The oxidations with oxygen were non-stereospecific and generated both, the 11 α- and 11β-hydroxylactones in combined yields ranging from 13-47 % along with norsesquiterpene ketones which are most likely formed by decomposition of the hydroperoxide anion intermediates. Hydroxylation of the germacranolide-type 11,13-dihydroparthenolide with either (+)- or (-)-(camphorylsulfonyl)oxaziridine gave exclusively the 11β-hydroxylactone (66-72 %) with no detection of the norsesquiterpene ketone. Keywords: Sesquiterpene lactones, hydroxylation, LDA, enolates, oxidations, nor-sesquiterpene ketones, germacranolides, eudesmanolides, guaranolides, elemanolides. Resumen. Se llevaron a cabo hidroxilaciones de las posiciones α- del grupo carbonilo lactónico en cuatro esqueletos diferentes de 11,13dihidro- derivados de lactonas sesquiterpénicas (germacranólidas, eudesmanólidas, guayanólidas y elemanólidas), mediante la generación del enolato con LDA y su atrapamiento con oxígeno gaseoso o con un agente oxidante quiral, (canforilsulfonil)aziridina. Las oxidaciones con oxígeno no fueron estereo-específicas y generaron las hidroxi-lactonas 11α- y 11β- en rendimientos combinados que fluctúan entre 13 al 47 %, junto con cetonas nor-sesquiterpénicas, que se forman probablemente por la descomposición de los aniones hidroperóxidos intermediarios. La hidroxilación de la germacranólida 11,13-dihidropartenólida, con (+)- o (-)- (canforilsulfonil)-aziridina produjo la 11β-hidroxi-lactona exclusivamente (66-72 %), sin detectarse la cetona nor-sesquiterpénica. Palabras clave: Lactonas sesquiterpénicas, hidroxilaciones, LDA, enolatos, oxidaciones, cetonas nor-sesquiterpénicas, germacranólidas, eudesmanólidas, guayanólidas, elemanólidas. Introduction tainly enhances the inhibition of PFK, Vargas et al. [4] showed that a hydroxyl group located in proximity to the lactone functionality of sesquiterpene lactones also enhances inhibition of PFK. The hydroxyl group of 7-hydroxysesquiterpene lactones is possibly enhancing PFK inhibition by hydrogen bonding to the active site of the enzyme. With the assumption that 11-hydroxysesquiterpene lactones might show biological activities similar to their 7-hydroxy analogs, the synthesis of a series of 11-hydroxylated sesquiterpene lactones as synthetic models for the study PFK inhibition was desired. In this paper, we describe the preparation of 11-hydroxysesquiterpene lactones from the corresponding 11,13-dihydrosesquiterpene lactones by reaction of the lactone enolates with oxygen (Scheme 1) [7]. Transformations of four skeletal types of 11, 13-dihydrosesquiterpene lactones (germacrolides, eudesmanolides, guaianolides, and elemanolides) were carried out. 7-Hydroxyl-bearing sesquiterpene lactones are uncommon in nature [1]. However, they show very interesting biological activities. For example, 7α-hydroxydehydrocostus lactone (21a) exhibits molluscicidal activity against Biomphalaria glabrata snails [2], that are hosts in the life cycle of the blood fluke which is responsible for human Schistosomiasis (bilharzia), a disease which affects more than 200 million people in Africa, Asia, and South America [3]. In contrast, dehydrocostus lactone (21b) is not active against Biomphalaria [1]. 7α-Hydroxydehydrocostus lactone (21a) has been shown to inhibit the in vitro activity of mammalian phosphofructokinase (PFK), and exhibits a twenty-fold higher in vitro inhibitory activity towards PFK than dehydrocostus lactone (21b) [4]. While there is no direct correlation of molluscicidal activity and PFK inhibition by sesquiterpene lactones, it is interesting to note that the most potent molluscicidal sesquiterpene lactone is also the most active PFK inhibitor [4]. Most biological activities of sesquiterpene lactones seem to depend on the presence of the α-methylene-γ-lactone moiety which is a receptor of biological nucleophiles such as essential thiol groups present in a number of enzymes and proteins [5, 6]. While the presence of the α-methylene-γ-lactone moiety cer- Results and discussion Dihydroparthenolide (4) was oxidized as outlined in Scheme 2. The enolate of 4 was generated at –70 °C in THF by deprotonation with LDA under argon atmosphere. Subsequently Preparation of 11-Hydroxylated 11,13-Dihydrosesquiterpene Lactones Table 1. Selected 1H NMR dataa. Sesquiterpene lactone Dihydroparthenolide (4) 11α-Hydroxyhydroparthenolide (5) 11β-Hydroxydihydroparthenolide (6) a CH3-13 1.26 (d) H-6 3.80 (dd) 1.31 (s) 3.79 (dd) 1.39 (s) 4.12 (dd) Chemical shifts in ppm, multiplicity in parenthesis, s = singlet, d = doublet. oxygen, dried over P2O5, was bubbled through the solution for about 20 minutes. The reaction was quenched by the addition of 3-4 mL of distilled water. The solution was then carefully neutralized with 5 % HCl and extracted with diethyl ether. Sesquiterpene lactones 1, 10, 15, 21, and 24 were reacted under similar conditions. The products were separated using silica gel column chromatography, preparative thin-layer chromatography, or reversed-phase HPLC. The 11-hydroxylactones were analyzed by application of IR, 1H and 13C NMR, and MS methods. The IR spectra of the CH3 LDA O O o-Li+ O O2 CH3 OH O H2 O CH3 O O - O O - O A O - O 1. H2O 2. Neutralize CH3 O HO B O O O - OH O O CH3 HO CH 3 - CO32- O Scheme 1. Proposed mechanism of lactone enolate oxidations. Scheme 2 133 derivatives clearly showed a broad absorption signal near 3400 cm–1 due to the lactonic 11-hydroxyl group. The 1H NMR data also indicated hydroxylation at C-11 by collapse to a methyl singlet of the dihydrolactone C-11-methyl doublets (C-13). The 1 H NMR data was also used to distinguish between the 11α- and 11β-hydroxy-derivatives. Due to the through-space deshielding effect of the C-11β hydroxyl group, the chemical shift of the lactonic signal (H-6β) for all 11βhydroxy-derivatives had shifted downfield by approximately 0.3-0.5ppm, when compared to the corresponding nonhydroxylated starting compounds. In contrast, the chemical shifts of the H-6β signals for all 11α-hydroxyderivatives remained about the same as those of the corresponding dihydroprecursors (Table 1). The total yield of the 11-hydroxylactones in these reactions ranged from 13-47 % with no apparent trends in stereoselectivity (Table 2). Norsesquiterpene ketones 13, 14, 20, and 28 were obtained as minor products of the reactions of 4, 10, 15, and 24, respectively, and in some cases they represented the only product. The IR spectra of these compounds showed absorptions near 3400 cm –1 due to the C-6 hydroxyl group and another at about 1710 cm–1 due to the C-11-ketone carbonyl stretch absorption. The 1H NMR data also showed methyl singlets near 2.10-2.20 ppm, indicative of a methyl ketone. The 13C NMR spectra of these compounds indicated the presence of only 14-carbons. Based on the above data, structures 13, 14, 20, and 28 were proposed. Table 3 summarizes the 13C NMR assignments of compounds 1-16, 20, and 21. A possible mechanism for the formation of the norsesquiterpene ketones may involve the decarboxylation of a hydroperoxide anion intermediate (Scheme 1). Hydroperoxides have been reported as the major products in reactions of ester enolates with t-BuOK instead of LDA [9]. The existence of lactonic hydroperoxide intermediates was supported by the isolation of 18 and 27 from their respective product mixtures. The hydroperoxide intermediates (Scheme 1, A) are then reduced to the alcohols, probably by the conjugate acid, diisopropylamine, generated from LDA during formation of the enolate [10]. 1,2-Dioxetane formation could arise following hydrolysis of the hydroperoxide anion (Scheme 1, B). 1,2dioxetanes have been observed to decompose cleanly to carbonyl compounds which would generate the decomposition products isolated [11]. The respective 1,10-epoxyderivatives 8, 9, 10 and 12 were obtained by stereo- and regiospecific epoxidations of the 1,10-double bond of the 11-hydroxy-derivatives 2, 3, 4, and 6 with m-chloroperbenzoic acid (m-CPBA) in the presence of sodium acetate as a buffer to prevent further cyclizations [12]. 134 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Howard G. Pentes et al. Oxidation of the enolate anion of dihydroparthenolide (4) with (-)-(camphorylsulfonyl)oxaziridine provided 11β-hydroxydihydroparthenolide (6) in 66 % yield. Neither the 11αhydroxydihydroparthenolide (5) nor the norsesquiterpene ketone (13) were detected (Table 2). The same results were observed for the oxidation of the enolate anion of 4 with (+)(camphorylsulfonyl)oxaziridine, except that the yield of 11βhydroxy-11,13-dihydroparthenolide (6) was slightly higher (72 %). When compared to the enolate oxidation with oxygen, the (camphorylsulfonyl)oxaziridine oxidizing agents are clear- ly superior due the higher yields and the regio- and stereospecificity of the reactions. Apparently, the frozen solute conformation of the 12,6trans-lactone 4, favors a β-attack by the (camphorylsulfonyl)oxaziridine oxidizing agents from the β-face of the enolate intermediate. This is in analogy to protonations that follow NaBH4 reductions in methanol of the α-methylene-γ-lactone group in similar sesquiterpene lactones such as parthenolide (4a) and costunolide (1a). Enolate oxidations with (camphorylsulfonyl)oxaziridines may not be stereospecific with conformationally more flexible sesquiterpene lactones such as 12,8-lactonized or 12,6-cis-lactonized germacranolides. 9 29 8a 4 N 2 3 S O O2 1 30 N O S O2 Fig. 1. (+)-(2R, 8aS)-(Camphorylsulfonyl)oxaziridine (29) and (-)(2S, 8aR)-(Camphorylsulfonyl)oxaziridine (30). Conclusions In summary, four skeletal types of 11,13-dihydrosesquiterpene-γ-lactones (germacrolides, eudesmanolides, guaianolides, and elemanolides) were transformed into 11-hydroxy- Preparation of 11-Hydroxylated 11,13-Dihydrosesquiterpene Lactones 135 Table 2. Yield (%) of Products from Enolate Oxidation of Sesquiterpene Lactonesa. Sesquiterpene lactone 1 4 4 4 10 15 21 24 aYields Oxidizing Agent Total Yield of 11-OH-products 11α-OH 11β-OH Norketone oxygen oxygen (–)-oxaziridine (+)-oxaziridine oxygen oxygen oxygen oxygen 37 47 66 72 — 24 29 13 15 29 — — — 15 12 5 22 18 66 72 — 9 17 8 — 16 — — 37 14 — 13 are based on recovered starting materials. Tabla 3a. 13C NMR Data for Compounds 1-11a. Carbon 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 127.0 25.7 40.6 136.5 126.4 80.9 54.2 28.0 39.1 139.6 41.7 178.0 12.8 15.6 127.0 24.0 40.9 136.9 127.0 79.5 55.9 26.1 39.5 140.6 75.3 179.6 19.2 16.1 127.1b 23.2 41.2 137.2 126.7b 81.1 56.4 25.9 39.5 140.7 75.5 177.8 22.0 16.1 15 16.7 17.1 17.1 4 [13] 5 6 7 8 9 125.1 24.0 36.6 61.4 66.3 82.1 51.9 29.7 41.1 134.4 42.4 177.3 13.2 16.8 125.0 23.9 36.6 62.0 66.2 80.8 53.2 24.7 41.0 134.5 75.5 178.8 19.0 16.7b 124.5 23.9b 36.8 61.7 66.5 82.4 53.7 24.1b 41.2 135.1 75.3 176.9 21.7 16.9c 67.4 24.5 35.9 143.0 123.9 80.2 54.9 25.6 39.2 61.1 42.1 178.0 12.7 17.3b 67.7 24.6b 36.1 143.7 123.9 78.4 56.5 25.0b 39.2 61.3 75.2 178.2 19.0 17.6c 67.5 21.8 36.3 143.4 124.0 79.9 56.9 24.6 39.6 61.4 75.3 177.2 20.6 17.5b 17.1 16.9b 17.0c 16.9b 17.2c 17.2b Table 3b. 13C NMR Data of Compounds 12-16, 20 and 21a. Carbon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 10 11 12 13 14 15 16 20 21 64.1b 23.6 34.9 60.2c 63.4b 81.2 51.2 25.0 39.7 60.3c 42.3 176.8 12.6 16.6d 64.3b 20.8 35.3 60.5c 64.2c 80.0 53.3 23.8 39.9 60.8c 75.3 178.2 19.2 16.9d 17.1d 17.5d 64.8b 21.7 35.5 60.5c 63.4b 81.6 53.7 23.7 40.3 60.9c 74.9 176.2 19.5 16.9d 17.4d 125.2 23.7 37.2 60.2 64.5 69.5 71.0 28.1 40.1 135.0 209.8 — 29.6 17.2 17.2 64.9 23.5b 36.1 59.3 63.4 67.5 71.2 23.7b 39.1 60.6 210.0 — 30.6 17.1c 17.1c 23.0e 35.7d 122.1 133.0 50.5b 81.8 53.9b 23.5e 37.6c 39.1 74.0 179.6 12.3f 17.2f 22.7f 22.8e 35.8c 122.6 132.7 51.0b 79.6 56.1b 23.5e 37.6c 39.1 74.0 180.4 19.0f 18.1f 17.3f 23.3c 34.6b 123.5 134.4 51.0 69.2 60.8 24.5c 38.0b 39.5 212.7 — 29.4 16.6 22.9 46.9 41.9 28.6 151.6 51.8 85.1 49.7 32.4c 38.6c 149.8 42.1b 178.5 13.1 111.7 109.0 a = Spectra were determined in CDCl3 at 200 MHz with Me4Si as internal standard. Chemical shifts are in ppm. Assignments were made (except for 4) by comparison with 13C NMR data of similar known compounds. b-f = Assignments are interchangeable. 136 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) lactone analogs by LDA-mediated generation of the corresponding lactone enolates followed by trapping with gaseous oxygen or chiral oxidizing agent, (camphorylsulfonyl) oxaziridines. The oxidations with oxygen were non-specific, resulting in low to moderate yields (13-47 %) of mixtures of 11α- and 11β-hydroxylactone derivatives plus norsesquiterpene ketones formed as degradation products of the hydroperoxide intermediates. Improved yields (66-72 %) and stereoselectivity were observed for enolate oxidations of 11,13-dihydrosesquiterpene lactones with the respective (+)- and (-)(2S,8aR)-(camphorylsulfonyl)oxaziridine [8], providing the 11β-hydroxylactones exclusively. Experimental section 1H and 13C NMR spectra were recorded on a Bruker-AC200 spectrometer in CDCl3 using SiMe4 as an internal standard. Mass spectra were obtained on a HP5985 spectrometer. IR spectra were recorded either on a Perkin-Elmer 257 or 1760x spectrometer as a film on NaCl plates. (-)-(2S,8aR)-(Camphorylsulfonyl)oxaziridine and (+)(2R,8aS)-(camphorylsulfonyl) oxaziridine (Aldrich) were used without further purification. Reagent grade THF was freshly distilled over Li metal before use to remove any traces of water. A 1.5 M solution of LDA in cyclohexane (Aldrich) was used without further purification. Chromatographic separations were made on silica gel (60-200 mesh, J.T.Baker Chemical Co.). HPLC separations were carried out on a Milton-Roy HPLC using RSIL-C18-10 µ semi-preparative column (Alltech/Applied Science). Dihydroparthenolide (4) was isolated from the dichloromethane (DCM) extract of the aerial parts of Ambrosia artimisiifolia [13,14]. Costunolide and dehydrocostus lactone (1a and 21b) were isolated by vacuum liquid chromatography [15] from Costus Resinoid (Pierre Chauvet, S.A.). The exocyclic methylene groups of costunolide and dehydrocostus lactone were reduced with NaBH4 in methanol at 0 °C [16] to give 1 and 21 respectively. α-Cyclodihydrocostunolide (15) was prepared via acidic transannular cyclization of 1 [16]. Saussurea lactone (24) was prepared by thermolysis of 1 [17]. Spectroscopic and physical data for compounds 1, 4, 15, 21 and 24 are consistent with those previously reported in the literature. 11α-Hydroxydihydrocostunolide (2) and 11β-Hydroxydihydrocostunolide (3). Compound 1 (325 mg, 1.39 mmol), dissolved in 5 mL of dry THF, was added slowly over 15 min by syringe to a stirred solution of 1.2 mL of LDA in 5 mL of THF under argon at –70 °C. After an additional 15 min., dry oxygen was bubbled through the solution for 20 min at 0 °C. The reaction was then quenched with 5 mL of water. The solution was neutralized with 5 % aq. HCl and extracted with diethyl ether. The ether solution was dried over anhydrous Na2SO4, filtered, and the solvent evaporated. Column chromatography on silica gel using DCM / acetone (95:5) yielded Howard G. Pentes et al. 21 mg (15 %) of 2 and 31 mg (22 %) of 3. Lactone 2 was isolated as a colorless powder: IR 3434, 1773, 1668 cm–1; 1H NMR: δ 4.80 (m, 1H, C1-H); 4.60 (dd, 1H, C6-H); 1.69 (s, 3H, C15-CH3); 1.40 (s, 3H, C14-CH3); 1.33 (s, 3H, C13-CH3); MS m/z (relative intensity) 250 (M+) (1.2), 232 (M-18+) (0.4), 222 (M-28+) (2.6), 207 (M-43+) (2.3). Compound 3 was isolated as a colorless powder: IR 3435, 1754 cm–1; 1H NMR: δ 4.94 (dd, 1H, C6-H); 4.80 (m, 1H, C1-H); 4.60 (d, 1H, C5-H, J = 10 Hz); 1.77 (s, 3H, C15-CH3); 1.45 (s, 3H ,C13- or C14CH3); 1.42 (s, 3H, C13- or C14-CH3); MS m/z (relative intensity) 250 (M+) (0.7), 222 (M-28+) (2.8), 207 (M-43+) (0.7). 11 α-Hydroxydihydroparthenolide (5), 11 β-Hydroxydihydroparthenolide (6), and Ketone (13). Compound 4 (372 mg) was reacted with LDA and oxygen as described above. Column chromatography on silica gel with hexane / EtOAc (1:1) yielded 8 8mg (29%) of 5, 55 mg (18%) of 6, and 45 mg (16%) of 13. Compound 5 was isolated as a white powder: IR 3412, 1784 cm–1; 1H NMR: δ 5.18 (dd, 1H, C1-H, J= 10 Hz); 3.79 (dd, 1H, C6-H, J = 9 Hz); 2.76 (d, 1H, C5-H, J = 9 Hz); 1.70 (s, 3H, C14-CH3); 1.31 (s, 6H, C13- and C15-CH3); MS m/z (relative intensity) 266 (M+) (0.02), 223 (M-43+) (0.07), 207 (M59+) (0.08), 43 (C2H3O+) (100). Lactone 6 was obtained as a white powder: IR 3443, 1753 cm–1; 1H NMR: δ 5.15 (dd, 1H,C1-H, J = 2, 9 Hz); 4.12 (dd, 1H, C6-H, J = 9 Hz); 2.66 (d, 1H, C5-H, J = 9 Hz); 1.69 (s, 3H, C14-CH3); 1.39 (s, 3H, C13-CH3); 1.28 (s, 3H, C15-CH3); MS m/z (relative intensity) 266 (M+) (0.03), 231 (M-35+) (0.04), 223 (M-43+) (0.02), 207 (M-59+) (0.14). Compound 13 was isolated as a colorless gum: IR 3438, 1761 cm–1; 1H NMR: δ 5.14 (dd, 1H, C1-H, J = 4, 7 Hz); 3.56 (dd, 1H, C6-H, J = 9 Hz); 2.75 (d, 1H, C5-H); 2.19 (s, 3H, C13CH3); 1.65 (s, 3H, C14-CH3); 1.27 (s, 3H, C15-CH3); MS m/z (relative intensity) 238 (M+) (0.1), 223 (M-15+) (0.2), 220 (M18+) (0.7), 195 (M-43+) (0.4), 177 (M-61+) (6.1). 1,10-Epoxydihydrocostunolide (7). Compound 1 (200 mg) was dissolved in 10 mL of DCM and stirred at room temp. Sodium acetate (200 mg) was added to the solution to buffer the epoxidation and prevent possible acid-catalyzed transannular cyclization [12]. m-CPBA (220 mg) was added to the suspension. After stirring at room temp. for 1 h, the solution was filtered and washed with 5 % Na2CO3 (2 × 50 mL) and H2O (3 × 50 mL). The DCM solution was dried over anhydrous Na2SO4, filtered, and the solvent evaporated yielding 181 mg (85 %) of 7: IR 1771, 1672 cm–1; 1H NMR: δ 5.12 (d, 1H, C5-H, J = 10 Hz); 4.54 (dd, 1H, C6-H, J = 10 Hz); 2.61 (dd, 1H, C1-H, J = 2, 11 Hz); 1.75 (s, 3H, C15-CH3); 1.14 (d, 3H, C13-CH3, J=7 Hz); 1.06 (s, 3H, C14-CH3); MS m/z (relative intensity) 250 (M+) (0.9), 235 (M-15+) (0.3), 232 (M-18+) (0.3), 207 (M-43+) (0.6), 193 (M-57+) (1.8). Preparation of 11-Hydroxylated 11,13-Dihydrosesquiterpene Lactones 137 1,10-epoxy-11α-hydroxydihydrocostunolide (8). Compound 2 (7 mg) was epoxidized as described above yielding 4 mg (54 %) of 8: IR 3418, 1775, 1671 cm–1; 1H NMR: δ 5.20 (d, 1H, C5-H, J = 10 Hz); 4.60 (dd, 1H, C6-H, J=10 Hz); 2.68 (dd, 1H, C1-H); 1.83 (s, 3H, C15-CH3); 1.33 (s, 3H, C13-CH3); 1.12 (s, 3H, C14-CH3); MS m/z (relative intensity) 266 (M+) (0.3), 221 (M-45+) (0.1), 210 (M-56+) (0.1), 189 (M-77+) (0.7). 16: IR 3449, 1770 cm–1; 1H NMR: δ 5.37 (s, br, 1H, C3-H); 3.92 (dd, 1H, C6-H, J = 11 Hz); 2.75 (s, br, OH); 1.76 (s, 3H, C15-CH3); 1.36 (s, 3H, C13-CH3); 0.90 (s, 3H, C14-CH3); MS m/z (relative intensity) 250 (M+) (1.3), 207 (M-43+) (0.4), 191 (M-59+) (0.7). 1,10-epoxy-11β-hydroxydihydrocostunolide (9). Compound 3 (9 mg) was epoxidized as described above yielding 7 mg of 9: IR 3443, 1773, 1674 cm–1; 1H NMR: δ 5.15 (d, 1H, C5-H, J = 10 Hz); 4.97 (dd, 1H, C6-H, J = 10 Hz); 2.66 (dd, 1H, C1-H, J = 2, 11 Hz); 1.80 (s, 3H, C15-CH3); 1.43 (s, 3H ,C13-CH3); 1.14 (s, 3H, C14-CH3). MS m/z (relative intensity) 266 (M+) (0.1), 244 (M-22+) (0.1), 222 (M-44+) (0.2), 207 (M-59+) (0.2), 189 (M-77+) (0.3). 1,10-epoxydihydroparthenolide (10). Compound 4 (150 mg) was epoxidized as described above yielding 151 mg (95 %) of 10. Spectroscopic and physical data for the title compound are consistent with those reported in the literature [18]. 1,10-epoxy-11α-hydroxydihydroparthenolide (11). Compound 5 (31 mg) was epoxidized as described above yielding 4 mg (10 %) of 11: IR 3422, 1782 cm–1; 1H NMR: δ 3.86 (dd, 1H, C6-H, J = 10 Hz); 2.87 (d, 1H, C5-H, J = 10 Hz); 2.80 (dd, 1H, C1-H); 1.40 (s, 3H, C13-CH3); 1.33 (s, 6H, C14and C15-CH3); MS m/z (relative intensity) 282 (M+) (0.03), 257 (M-25+) (0.6), 219 (M-63+) (0.3), 211 (M-71+) (0.5), 197 (M-85+) (0.9). 1,10-epoxy-11β-hydroxydihydroparthenolide (12). Compound 6 (25 mg) was epoxidized as described above yielding 25 mg (95 %) of 12: IR 3391,1781 cm–1; 1H NMR: δ 4.20 (dd, 1H, C6-H, J = 9 Hz); 2.81 (d, 1H, C1-H); 2.80 (d, 1H, C5-H, J = 9 Hz); 1.46 (s, 3H, C13-,C14-, or C15-CH3); 1.43 (s, 3H, C13-, C14-, or C15-CH3); 1.36 (s, 3H, C13-,C14-, or C15CH3); MS m/z (relative intensity) 282 (M+) (0.1), 210 (M-72+) (0.1), 195 (M-87+) (0.1). Ketone 14. Compound 10 (114 mg) was reacted with LDA and O2 as described before yielding 40 mg of ketone 14: IR 3449, 1711 cm–1; 1H NMR: δ 3.65 (dd, 1H, C6-H, J = 9 Hz); 2.86 (d, 1H, C5-H); 2.24 (s, 3H, C13-CH3); 1.38 (s, 3H, C14- or C15CH3); 1.30 (s, 3H, C14- or C15-CH3); MS m/z (relative intensity) 193 (M-61+) (0.1), 179 (M-75+) (1.4), 161 (M-93+) (1.2). 11α-Hydroxy-α-cyclodihydrocostunolide (16), 11βHydroxy-α-cyclodihydrocostunolide (17), 11α-Hydroperoxy-α-cyclodihydrocostunolide (18), 7,11-Dehydro-α-cyclodihydrocostunolide (19), and Ketone 20. Compound 15 (102 mg) was reacted with LDA and O2 as described before yielding 16 mg (15 %) of 16, 10 mg (9 %) of 17, 14 mg (14 %) of 20, 1 mg of 18, and 1 mg of 19. Compounds 17, 18, and 19 were isolated by HPLC following column chromatography. 17: IR 3458, 1761 cm–1; 1H NMR: δ 5.38 (s, br, 1H,C3-H); 4.36 (dd, 1H, C6-H, J = 5 Hz); 1.82 (s, 3H, C15-CH3); 1.45 (s, 3H, C13-CH3); 0.92 (s, 3H, C14-CH3); MS m/z (relative intensity) 250 (M+) (2.0), 207 (M-43+) (1.2). 18: IR 3414, 1778 cm–1; 1H NMR: δ 8.73 (s, 1H, OOH); 5.39 (s, br, 1H,C3-H); 3.96 (dd, 1H, C6-H, J = 10 Hz); 1.80 (s, 3H, C15-CH3); 1.37 (s, 3H, C13-CH3); 0.91 (s, 3H ,C14-CH3); MS m/z (relative intensity) 266 (M+) (0.04), 223 (M-43+) (0.5), 220 (M-46+) (0.2), 216 (M-50+) (0.3). 19: IR 1752, 1682 cm–1; 1H NMR: δ 5.43 (s, br, 1H, C3-H); 4.67 (d, 1H, C6-H, J = 11 Hz); 1.89 (s, 3H, C15-CH3); 1.83 (s, 3H, C13-CH3); 0.99 (s, 3H, C14-CH3); MS m/z (relative intensity) 232 (M+) (2.7), 217 (M-15+) (7.3), 207 (M-25+) (7.4). 20: IR 3449, 1700 cm–1; 1H NMR: δ 5.35 (s, br, 1H, C3-H); 4.02 (ddd, 1H, C6-H, J = 5, 11 Hz); 2.21 (s, 3H, C13-CH3); 1.83 (s, 3H, C15-CH3); 0.81 (s, 3H, C14-CH3); MS m/z (relative intensity) 222 (M+) (0.6), 123 (M-99+) (12.7), 121 (M-101+) (17.1). 11α-Hydroxydihydrodehydrocostuslactone (22) and 11βHydroxydihydrodehydrocostuslactone (23). Compound 21 (235 mg) was reacted with LDA and O2 as described before yielding 30 mg (12 %) of 22 and 42 mg (17 %) of 23. 22: IR 3467, 1770, 1638 cm–1; 1H NMR: δ 5.16 (s, 1H, C15H); 5.05 (s, 1H, C15-H); 4.87 (s, 1H, C14-H); 4.77 (s, 3H, C14H); 3.87 (dd, 1H, C6-H, J = 9 Hz); 1.30 (s, 3H, C13-CH3); MS m/z (relative intensity) 248 (M+) (1.7), 220 (M-28+) (1.5), 202 (M-46+) (0.3), 192 (M-56+) (0.2). 23: IR 3423, 1761, 1630 cm–1; 1H NMR: δ 5.20 (s, 1H, C15H); 5.05 (s, 1H, C15-H); 4.88 (s, 1H, C14-H); 4.80 (s, 1H, C14H); 4.20 (dd, 1H, C6-H, J = 9 Hz); 1.43 (s, 3H, C13-CH3); MS m/z (relative intensity) 248 (M+) (11.0), 204 (M-44+) (2.5), 191 (M-57+) (2.3), 189 (M-59+) (2.3). 11α-Hydroxysaussurea lactone (25), 11β-Hydroxysaussurea lactone (26), 11α-Hydroperoxysaussurea lactone (27), and Ketone 28. Compound 24 (93 mg) was reacted with LDA and O2 as described before yielding 4 mg (5 %) of 25, 6 mg (8 %) of 26, 9 mg (13 %) of 28, and less than 1 mg of 27. 25: IR 3440, 1778, 1638 cm–1; 1H NMR: δ 5.79 (dd, 1H, C1H, J = 11, 17 Hz); 5.04 (m, 4H, C2-Ha,b, C3-Ha,b); 4.14 (dd, 1H, C6-H, J = 11 Hz); 2.27 (d, 1H, C5-H, J = 9 Hz); 1.79 (s, 3H, C15-CH3); 1.38 (s, 3H, C13-CH3); 1.08 (s, 3H, C14-CH3); MS m/z (relative intensity) 250 (M+) (0.2), 223 (M-28+) (1.7), 207 (M-43+) (1.3), 189 (M-61+) (1.0). 138 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) 26: IR 3449, 1752, 1638 cm–1; 1H NMR: δ 5.80 (dd, 1H, C1H, J = 11, 17 Hz); 5.00 (m, 4H, C2-Ha,b, C3-Ha,b); 4.60 (dd, 1H, C6-H, J = 10, 11Hz); 2.20 (d, 1H, C5-H, J = 12 Hz); 1.79 (s, 3H, C15-CH3); 1.46 (s, 3H, C13-CH3); 1.10 (s, 3H, C14CH3); MS m/z (relative intensity) 250 (M+) (0.3), 204 (M-46+) (0.4), 121 (M-129+) (3.2). 27: IR 3353, 1770, 1638 cm–1; 1H NMR: 8.68 (s, 1H, OOH); 5.80 (dd, 1H, C1-H, J = 11, 17 Hz); 5.00 (m, 4H, C2-Ha,b, C3Ha,b); 4.17 (dd, 1H, C6-H, J = 11 Hz); 2.32 (d, 1H, C5-H, J = 11 Hz); 1.78 (s, 3H, C15-CH3); 1.39 (s, 3H, C13-CH3); 1.08 (s, 3H, C14-CH3); MS m/z (relative intensity) 266 (M+) (0.5), 216 (M-50+) (0.5), 166 M-100+) (0.9). 28: IR 3466, 1708, 1638 cm–1; 1H NMR: δ 5.76 (dd, 1H, C1-H, J = 11, 17 Hz); 4.90 (m, 4H, C2-Ha,b, C3-Ha,b); 4.10 (dd, 1H, C6-H, J = 11 Hz); 2.25 (s, 3H, C13-CH3); 1.78 (s, 3H, C15-CH3); 1.04 (s, 3H, C14-CH3); MS m/z (relative intensity) 222 (M+) (2.0), 204 (M-18+) (1.5), 189 (M-33+) (1.0), 161 (M-61+) (3.6). Oxidation of the enolate anion of dihydroparthenolide (4) with (-)-(2S,8aR)-(camphorylsulfonyl)oxaziridine. Dihydroparthenolide (4) (200 mg, 0.8 mmol) dissolved in 5 mL of dry THF was added slowly over 15 min by syringe to a stirred solution of 0.7 mL (1.04 mmol) of LDA in 5 mL of THF under argon at –70 °C. After stirring the solution for an additional 15 min, a THF solution of (-)-(2S,8aR)-(camphorylsulfonyl)oxaziridine (30, 370 mg, 1.6 mmol) was added to the reaction flask by syringe over a 5 min period at –70 °C. After 5 more min, the reaction was quenched with the addition of 5 mL of a saturated aqueous NH4Cl solution. The reaction mixture was extracted with diethyl ether (6 × 10 mL). The ether solution was dried over anhydrous Na2SO4, filtered, and the solvent was evaporated. Attempted precipitation of the unreacted oxaziridine and its reduced form, the imine, at –78 °C in diethyl ether only removed 60-70 % of these reagents. Repeated precipitations did not further purify the product. Dry column (silica gel) chromatography [19] was used to separate the product mixture eluting with DCM/acetone (9:1). The oxaziridine and the imine eluted in the very early fractions. Dihydroparthenolide (4) (58 mg) was recovered and 100 mg (66 %) of 11β-hydroxydihydroparthenolide (6) was isolated. The 1H NMR data of compound 6 was identical with the data for the product isolated from the reaction of oxygen with the enolate anion of dihydroparthenolide (4). The norsesquiterpene ketone 13 and 11αhydroxydihydroparthenolide (5) were not detected. Oxidation of enolate anion of dihydroparthenolide (4) with (+)-(2R,8aS)-(camphorylsulfonyl)oxaziridine. Dihydroparthenolide (4) (200 mg) was oxidized as described above with (+)(2R,8aS)-(camphorylsulfonyl)oxaziridine 29. The product mixture was separated by dry column (silica gel) chromatography [19] eluting with DCM/acetone (9:1). Dihydroparthenolide (4) (56 mg) was recovered and 110 mg (72 %) of 11β-hydroxydihydroparthenolide (6) was isolated as the only product. Howard G. Pentes et al. References 1. Fronczek, F.R.; Vargas, D.; Fischer, N.H.; Hostettmann, K. J. Nat. Prod. 1984, 47, 1036-1039. 2. Vargas, D.; Fronczek, F.R.; Fischer, N.H.; Hostettmann, K. J. Nat. Prod. 1986, 49, 133-138. 3. Hostettmann, K.; Marston, A. "Plants Used in African Traditional Medicines." in Folk Medicine. The Art and the Science, editor R.P. Steiner, 1986, ACS, Wash., D.C. pp. 111-124. 4. Vargas, D.; Younathan, E.S.; Fischer, N.H. Rev. Soc. Quím. Méx. 2001, 45, 159-162. 5. Picman, A.K. Biochem. System. Ecol. 1986, 14, 255-281. 6. Rodriguez, E.; Towers, G.H.N.; Mitchell, J.C. Phytochemistry 1976, 15, 1573-1580. 7. Collado, I.G.; Macías, F.A.; Massanet, G.M.; Molinillo, J.M.G.; R.-Luis, F. J. Org. Chem. 1987, 52, 3323-3326. 8. Davis, F.A.; Hague, M.J.; Ulatowski, T.G.; Towson, J.C. J. Org. Chem. 1986, 51, 2402-2404. 9. Gersmann, H.R.; Bickel, A.F. J. Chem. Soc. (B) 1971, 11, 22302237. 10. Biloski, A.; Ganem, B. Synthesis 1983, 7, 537-538. 11. Bartlett, P.D.; Schaap, A.P. J. Amer. Chem. Soc. 1970, 92, 32233226. 12. Rodriguez, A.A.S.; Garcia, M.; Rabi, J. Phytochemistry 1978, 17, 953-954. 13. Lu, T.; Fischer, N.H. Spectroscopy Letters 1996, 29, 437-448; references therein. 14. Parodi, F.J.; Fronczek, F.R.; Fischer, N.H. J. Nat. Prod. 1989, 52, 554-566. 15. Coll, J.C.; Bowden, B.F. J. Nat. Prod. 1986, 49, 934-936. 16. Lee, I.-Y. "New Sesquiterpene Lactones from the Genera Calea berlandiera (Asteraceae) and the Fragmentation Reactions of 1,3Dihydroxyeudesmanolide Derivatives." Dissertation, Louisiana State University, 1983. 17. Rao, A.S.; Sadgopal, A.P.; Bhattacharyya, S.C. Tetrahedron 1961, 13, 319. 18. Govindachari, T.R.; Joshi, B.S.; Kamat, V.N. Tetrahedron 1965, 21, 1509-1519. 19. Loev, B.; Goodman, M. M. Chem. and Ind. 1967, 48, 2026-2032. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 139-142 Investigación Chemical Composition and Antimicrobial Activity of the Essential Oils from Annona cherimola (Annonaceae) María Yolanda Ríos,1* Federico Castrejón,2 Norma Robledo,2 Ismael León,1 Gabriela Rojas,3 and Víctor Navarro3 Centro de Investigaciones Químicas de la Universidad Autónoma del Estado de Morelos. Av. Universidad 1001, Col. Chamilpa, Cuernavaca, Morelos, México, 62210. Phone +52 (777) 329-997 ext. 6024; Fax: +52 (777) 329-7997. E-mail: [email protected] 2 Centro de Desarrollo de Productos Bióticos del Instituto Politécnico Nacional. Carretera Yautepec-Jojutla Km. 8, Yautepec, Morelos, México, 62731. 3 Centro de Investigación Biomédica del Sur, Instituto Mexicano del Seguro Social. Argentina No. 1, Col. Centro, Xochitepec, Morelos, México, 62790. 1 Recibido el 25 de febrero del 2003; aceptado el 23 de mayo del 2003 This paper is dedicated to Professor Alfonso Romo de Vivar Abstract. The chemical composition of the essential oils obtained by steam distillation of the fresh leaves, flowers and fruits from Annona cherimola was analyzed by means of Gas Chromatography-Mass Spectrometry (GC/MS). Sixty constituents were identified from the oils. While bicyclogermacrene, trans-caryophyllene and δ-amorphene were found to be the major constituents in the oil of the leaves; bicyclogermacrene, α-terpinolene and germacrene D were the major constituents in the oil of the flowers and β-pinene, α-terpinolene, βfenchyl alcohol and α-pinene were the major constituents in the oil of the fruits. The in vitro antimicrobial activity of the three essential oils and of some of their major constituents against five Gram (±) bacteria and one fungus is reported. Keywords: Annona cherimola, essential oil, Gas ChromatographyMass Spectrometry, bicyclogermacrene, trans-caryophyllene, αamorphene, α-copaene, α-terpinolene, germacrene D, linalool, βfenchyl alcohol, β-pinene, α-pinene. Resumen. La composición química de los aceites esenciales obtenidos por arrastre de vapor de las hojas, flores y frutos frescos de Annona cherimola fue analizada por Cromatografía de Gases-Espectrometría de Masas (GC/MS). Sesenta componentes fueron identificados en los aceites esenciales. Mientras que el biciclogermacreno, el trans-cariofileno y el δ-amorfeno se identificaron como los constituyentes mayoritarios en el aceite esencial de las hojas; el biciclogermacreno, el α-terpinoleno y el germacreno D fueron los constituyentes mayoritarios del aceite esencial de las flores, y el β-pineno, el α-terpinoleno, el alcohol β-fenchílico y el α-pineno fueron los principales componentes del aceite esencial de los frutos. La actividad antimicrobiana in vitro de los tres aceites esenciales y de algunos de sus contituyentes mayoritarios fue evaluada contra cinco bacterias Gram (+), (-) y un hongo. Palabras clave: Annona cherimola, aceite esencial, Cromatografía de Gases-Espectrometría de Masas, biciclogermacreno, trans-cariofileno, α-amorfeno, α-copaeno, α-terpinoleno, germacreno D, linalool, alcohol β-fenchílico, β-pineno, α-pineno. Introduction Results and discussion The Annonaceae family includes 80 genera and about 850 species distributed in tropical and subtropical areas of America, Africa and Asia. Only four genera of this family are of economic importance, and the genus Annona is one of them. Annona cherimola is highly appreciated for its exquisite fruits and for its use in traditional medicine in the treatment of skin diseases [1], tumors and cancer [2], and is reported to have antimicrobial and insecticidal properties [3,4]. Although several reports on the chemical composition of A. cherimola have been published [1-9], to date there are no reports on the chemical analysis of the essential oil of this species. We present here the chemical composition and the antimicrobial activity of the essential oils from the leaves, flowers and fruits of A. cherimola. CG/MS analysis of the three oils led to the identification of sixty constituents, which are listed in Table 1 along with their quantitative data. The identification of each component was based on a comparison of its mass spectrum with those contained in the HP CHEMSTATION-Wiley275.L Library. A high proportion of the essential oils is constituted by four main compounds: more than 40 % of the essential oil from the leaves is composed by bicyclogermacrene (18.20 %), transcaryophyllene (11.50 %), α-amorphene (7.57 %) and αcopaene (5.63 %); similarly, 34.02 % of the essential oil from the flowers corresponds to bicyclogermacrene (11.73 %), αterpinolene (9.75 %), germacrene D (7.01 %) and linalool (5.53 %); finally, almost 45 % of the essential oil from the fruits corresponds to β-pinene (15.48 %), α-terpinolene (13.59 140 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Ríos Gómez et al. Table 1. Chemical composition of the essential oils of leaves, flowers and fruits from A. cherimola. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Compounda rtb KIc Isoprene α-Pinene Camphene β-Terpinene β-Pinene Mircene β-Phellandrene 1,8-Cineole trans-β-Ocimene δ-3-Carene trans-Linalool oxide α-Terpinolene Linalool Borneol α-Terpinene-4-ol β-Fenchyl alcohol α-Terpineol Borneyl Acetate α-Bisabolene α-Cubebene α-Copaene β-Cubebene β-Elemeno Unknown cis-Caryophyllene trans-Caryophyllene (-)Caryophyllene oxide β-gurjunene α-gurjunene (-)Isoledene α-Humulene Aromadendrene α-Amorphene Germacrene D Alloaromadendrene Bicyclogermacrene 10-Hydroxy-α-gurjunene δ-Amorphene γ-Cadinene Unknown δ-Cadinene Cadina-1,4-diene 9-Aromadendrene β-Elemol Nerolidol 1(10),4,11-Germacratrien-9-ol γ-Cadinol Lauric acid T-cadinol α-Cadinol Muurolol Unknown Azuleno Unknown Dehydroaromadendrene 9,10-Dehydroisolongifolene Cycloisolongifolene Unknown Unknown Palmitic acid (8 β,13β)-Kaur-16-ene Oleic acid Kauran-16-ol Stearic acid Kaur-16-en-19-ol Kaur-16-en-18-oic acid Total of compounds identified (%) 2.41 3.29 3.41 3.67 3.73 3.85 4.02 4.30 4.53 4.67 4.79 5.13 5.27 6.00 6.21 6.31 6.66 7.60 8.51 8.67 9.04 9.13 9.19 9.35 9.40 9.60 9.68 9.70 9.82 9.90 9.97 10.05 10.26 10.31 10.40 10.46 10.55 10.61 10.69 10.77 10.81 10.93 10.99 11.11 11.19 11.28 11.50 11.66 12.25 12.39 12.58 12.76 13.27 13.79 14.63 15.14 15.20 16.34 16.92 18.30 18.57 20.84 21.56 21.62 21.84 24.52 421 800 804 813 815 819 825 837 846 853 868 876 885 939 959 970 982 989 1085 1109 1170 1189 1202 1204 1205 1207 1209 1210 1213 1215 1217 1220 1226 1228 1231 1233 1236 1239 1242 1244 1247 1251 1256 1259 1264 1269 1281 1287 1407 1410 1415 1430 1434 1452 1493 1607 1610 1674 1732 1801 1811 1886 1905 1908 1913 2170 dGC Area % Ae 0.35 1.60 — — 4.11 0.23 — 0.32 0.15 — — — 3.06 — 0.17 — — — 1.69 0.89 5.63 — 3.57 — 1.34 11.50 — 0.30 1.37 0.60 3.05 0.39 2.20 3.75 2.36 18.20 4.47 7.57 2.50 0.60 4.63 1.27 0.70 — — 2.99 0.24 — 0.42 0.70 — — — 0.39 0.23 1.43 1.39 0.87 1.40 0.27 0.49 0.57 — — — — 96.70 Be Ce 0.08 0.14 — 0.08 1.32 2.75 — 0.30 0.08 — — 9.75 5.53 — 1.67 1.70 0.15 — 1.74 0.11 1.04 — 1.82 — — 2.30 0.16 — — — 0.24 0.18 — 7.01 — 11.73 1.01 1.08 0.14 — 1.98 0.12 — 0.40 — 4.57 0.86 0.58 3.40 3.30 2.39 — 1.51 2.90 — — — — — 12.97 — 6.64 — 6.25 — — 97.08 — 6.37 0.45 — 15.48 — 0.20 4.02 0.30 0.19 1.18 13.59 0.29 2.02 2.04 8.81 0.83 0.21 1.38 5.88 3.28 1.09 — 1.79 3.60 — 0.81 1.40 — 0.33 1.48 0.43 — 1.44 — 1.49 0.69 — — — 3.01 — — — 0.81 1.01 2.00 0.27 2.03 1.48 2.48 1.21 — — — — — — — — 0.44 — 0.99 — 0.64 1.90 96.25 a Composition listed in order of elution from a HP-1 column. bRetention times (rt) in minutes. cKovats Indices (KI) on HP-1 capillary dGas Chromatography. eA, B and C represent the essential oil of leaves, flowers and fruits of Annona cherimola, respectively - no detected column Chemical Composition and Antimicrobial Activity of the Essential Oils from Annona cherimola (Annonaceae) 141 Table 2. Proportion (%) of mono- and sesquiterpenes in the essential oils from A. cherimola. Sample hydrocarbons Monoterpenes alcohols oxides total hydrocarbons Sesquiterpenes alcohols oxides total Leaves Flowers Fruits 6.09 14.12 36.58 3.23 9.05 13.99 0.32 0.3 5.41 9.64 23.47 55.98 76.56 31.00 24.81 8.82 15.93 10.50 0.00 0.16 0.81 85.38 47.09 36.12 Table 3. Principal skeleta (%) of mono- and sesquiterpenes in the essential oils from A. cherimola. Sample acyclic monoterpenes Leaves Flowers Fruits 3.21 5.61 1.77 Monoterpenes pinene p-menthane 5.71 1.46 21.85 0.72 14.70 20.68 %), β-fenchyl alcohol (8.81 %) and α-pinene (6.37 %). The identified components represent between 96-97% of the total composition of the oils. The monoterpenes and sesquiterpenes are the main type of compounds in the three essential oils (Table 2). The essential oil of the leaves have a high proportion of sesquiterpenes (85.38 %) and showed a weak antimicrobial activity against the assayed microorganisms (Tabla 4). In the essential oil of the flowers the proportion of sesquiterpenes was lowest (47.09 %), increasing the proportion of monoterpenes (23.47 %), and for this essential oil the MIC values observed were minor in all the assayed microorganisms, with the exception of E. faecalis. In the essential oil of the fruits the monoterpenes are the major constituents (55.98 %) being this essential oil the most active against S. aureous and P. mirabilis. These results indicate that the antimicrobial activity of these essential oils could be associated to the presence and amount of the monoterpenic compounds. The mono- and sesquiterpenes isolated could be classified as hydrocarbons, alcohols and oxides, being the hydrocarbons the major components of the three essential oils (Table 2). In the three essential oils more than 79 % of the monoterpenes belongs to acyclic monoterpenes, and monoterpenes with pinene and p-menthane skeleton. In the essential oils from the leaves and flowers 60 % and 49 %, respectively, of the sesquiterpenes have the skeleton of caryophyllene, germacrene and aromadendrene, while in the essential oil from fruits the proportion of this sesquiterpenes is very lowest (16 %) (Table 3). The three essential oils of A. cherimola showed a significant activity against Gram-positive, Gram-negative bacteria and one fungus (Table 4). Although no previous reports on the antimicrobial activity of the major constituents of the essential oil from the leaves were found in the literature, trans-caryophyllene showed moderate activity against Staphylococcus aureus, Enterococcus faecalis, Escherichia coli and Shigella sonei. The essential oil from the flowers was the most active against all the microorganisms tested, and the second most active against S. aureus, this activity could be associated with total caryophyllene 100.00 92.70 79.10 12.84 2.46 0.81 Sesquiterpenes germacrene aromadendrene 21.95 18.74 2.93 17.00 2.09 2.09 total 60.65 49.45 16.14 its high concentration of linalool, that was very active against all the microorganisms included in the Table 2, and which is known to possess antimicrobial and antifungal activity [10,11]. Previous reports have been carried out of the antimicrobial activity for α-pinene [10,12,13], who in our hands showed moderate activity, however, a very important activity was observed for β-fenchyl alcohol, both presents in high proportion in the essential oil of the fruits, which shows the best effect against the bacteria S. aureus, E. faecalis and Proteous mirabilis. Although at least one of the major constituents of each essential oil showed antimicrobial activity against the tested microorganisms, the MIC values obtained in each case are biggest that those of their corresponding essential oils, this suggest that the antimicrobial activity could be due to a synergistic effect between the constituents of each essential oil. Experimental section Plant material: The leaves (442 g), flowers (309 g) and green fruits with an average size of 2.5 cm (206 g) were collected from 10 individuals of a wild population of A. cherimola. The plant material was collected during the flowering and fruiting stage in April-May of 2002. A specimen (voucher No. 18854) was deposited at the Herbarium of the Universidad Autónoma del Estado de Morelos (HUMO), Cuernavaca, Morelos, México. Chemical analysis: The leaves, flowers and fruits of A. cherimola were finely cuted and subjected to steam distillation (1.5 h) using a modified Clevenger-type apparatus, to yield 0.63 %, 0.39 %, and 0.83 % of a yellow oil, respectively. The physical properties for each sample were: leaves ([α]D25 + 16.1 (CHCl3, c = 1.1), d25 0.83), flowers ([α]D25 + 6.4 (CHCl3, c = 0.97), d25 0.87) and fruits ([α]D25 + 8.3 (CHCl3, c = 0.92), d25 0.84). The oils were subjected to GC / MS analysis in a Hewlett Packard 6890 GC / 5972 MSD chromatograph equipped with a HP-1 capillary column (length 30 m, id 0.25 mm, ft 0.25 µm). The carrier gas was helium and the linear gas velocity was 36 cm/s. The injector temperature was 250 142 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Ríos Gómez et al. Table 4. Antimicrobial activity of the essential oils of leaves, flowers and fruits from A. cherimola. Sample Leaves Flowers Fruits trans-caryophyllene Terpinolene linalool β-Pinene α-Pinene β-fenchyl alcohol Gentamicin Nystatin Staphylococcus aureus Enterococcus faecalis 0.25 0.125 0.06 8.0 > 16 2 > 16 16 4 0.004 — 0.5 0.5 0.5 16 > 16 4 > 16 16 2 0.004 — MIC (mg / mL) Escherichia coli ºC and the column temperature, initially at 60 ºC, was gradually increased at a rate of 10 ºC/min up to 160 ºC and then gradually increased at a rate of 5º C/min up to 220 ºC and kept at 220 ºC for 5 min. For detection, a flame ionization detector at 280 ºC, IE (Scan 30-550 uma) was used. Standards of pure metabolites. trans-caryophyllene (Aldrich, C-9653), terpinolene (Fluka, 86485), linalool (Aldrich, L2602), β-pinene (Fluka, 80608), α-pinene (Aldrich, 26,807-0) and β-fenchyl alcohol (Aldrich, 19,644-4) were obtained from commercial sources. Antimicrobial Activity: The bacteria Staphylococcus aureus (ATCC 25213), Enterococcus faecalis (ATCC 29212), Escherichia coli (ATCC 25922), Proteus mirabilis (ATCC 12453) and Shigella sonei (ATCC 11060) were maintained on Trypticase soya agar, while Candida albicans (ATCC 10231) was maintained on Sabouraud 4 % dextrose agar. The inoculum for each organism was 104 colony forming units (CFU)/ mL. The minimum inhibitory concentrations (MICs) were measured as described previously for essential oils [12]. Initial emulsions of the oils (20 mg/mL) and the standards of pure metabolites (16 mg/mL) were prepared in sterile distilled water with 10 % DMSO. Serial dilutions of the stock solutions in both media (100 µL of Muller Hinton broth or Sabouraud broth) were prepared in a microtiter plate and 2 µL of microbial suspension was added to each well. For each strain, the growth conditions and the sterility of the medium were proved and the plates were incubated 24 h at 37 °C for the bacteria, and 48 h at 28 °C for the yeast. Standard antibiotics (gentamicin and nystatin) were used as positive controls, and MICs were determined as the lowest concentrations preventing visible growth. To indicate the bacterial growth, p-Iodonitrotetrazolium violet (SIGMA I-8377) was added to the microplate wells, as described by Eloff [14]. Copies of the original GC and GC-MS chromatographs and spectra can be obtained from the author of correspondence. 10 2 2 16 > 16 2 > 16 4 2 0.008 — Shigella sonei Proteous mirabilis Candida albicans 5 2.5 2.5 16 > 16 2 > 16 16 2 0.008 — 5 2 1 > 16 > 16 2 > 16 16 2 0.008 — 5 0.5 2 > 16 > 16 4 > 16 8 8 — 0.004 Acknowledgements We thank Enrique Salazar Leyva for technical assistance. References 1. Chen, Ch.Y.; Chang, F.R.; Pan, W.B.; Wu, Y.C. Phytochemistry 2001, 56, 753-757. 2. Chen, Ch.Y.; Chang, F.R.; Chiu, H.F.; Wu, M.J.; Wu, Y.C. Phytochemistry 1999, 51, 429-433. 3. Cortes, D.; Myint, S.H.; Dupont, B.; Davoust, D. Phytochemistry 1993, 32, 1475-1482. 4. Simeon, S.J.; Ríos, L.; Villar, A. Pharmazie 1990, 45, 442-443. 5. Woo, M.H.; Kim, D.H.; Fotopoulos, S.S.; McLaughlin, J.L. J. Nat. Prod. 1999, 62, 1250-1255. 6. Chen, Ch.Y.; Chang, F.R.; Yen, H.F.; Wu, Y.Ch. Phytochemistry 1998, 49, 1443-1447. 7. Chen, Ch.Y.; Chang, F.R.; Wu, Y.Ch. Tetrahedron Letters 1997, 38, 6247-6248. 8. Sahpaz, S.; González, M.C.; Hocquemiller, R.; Zafra-Polo, M.C.; Cortes, D. Phytochemistry 1996, 42, 103-107. 9. Cortes, D.; Myint, S.H.; Leboeuf, M.; Cavé, A. Tetrahedron Letters 1991, 32, 6133-6134. 10. Adam, K.; Sivropoulou, A.; Kokkini, S.; Lanaras, T.; Arsenakis, M. J. Agric. Food Chem. 1998, 46, 1739-1745. 11. Skaltsa, H.; Lazari, D.; Mavromati, A.; Tiligada, E.; Constantinidis T. Planta Medica 2000, 66, 672-674. 12. Aligiannis, N.; Kalpoutzakis, E.; Chinou, I.; Mitakou, S. J. Agric. Food Chem. 2001, 49, 811-815. 13. Tzakou, O.; Pitarokili, D.; Chinou, I.; Harvala C. Planta Medica 2001, 67, 81-83. 14. Eloff, J.N. Planta Medica 1998, 64, 711-713. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 143-145 Investigación Sesquiterpene Lactone Sequestration by the Tortoise Beetle Physonota arizonae (Cassidinae) Manuel Aregullín and Eloy Rodríguez* Natural Products Laboratory, Biotechnology 259, College of Agriculture and Life Sciences, Cornell University, Ithaca, NY 14853-4301, USA In honor of Dr. Alfonso Romo de Vivar for his contributions to the field of phytochemistry the chemistry of sesquiterpene lactones Recibido el 24 de marzo del 2003; aceptado el 28 de mayo del 2003 Abstract. The phenomenon of sequestration of plant secondary metabolites by herbivorous arthropods and its importance as an arthropod defense strategy is well documented in chemical ecology studies. Damsin (1), and the related terpene damsinic acid (2), are sesquiterpene lactones sequestered by the tortoise beetle Physonota arizonae (Cassidinae) from its host plant Ambrosia ambrosioides (Asteraceae) for possible chemical protection. Keywords: Sequestration, sesquiterpene lactones, Asteraceae, Ambrosia ambrosioides, Cassidinae, Physonota arizonae, Coleoptera, tortoise beetles. Resumen. El fenómeno de secuestro de metabolitos secundarios de plantas por artrópodos hervíboros, y su importancia como estrategia de defensa artrópoda, se encuentra bien documentado en estudios de ecología química. Damsina (1), y el terpeno ácido damsinico (2), son sesquiterpen lactonas secuestradas por el escarabajo tortuga Physonota arizonae (Cassidinae) de su planta anfitrión Ambrosia ambrosioides (Asteraceae) para su posible protección química. Palabras clave: Secuestro, sesquiterpen lactonas, Asteraceae, Ambrosia ambrosioides, Cassidinae, Physonota arizonae, Coleoptera, escarabajos tortuga. Introducción Because of the external adult morphology (i.e., analogous to the reptilian Chelonia), the beetles in the family Cassidinae are commonly referred to as tortoise beetles (Fig. 1a). The pronotum and elytra in these beetles extend to cover completely the margins of the body, and is an excellent example of mechanical defense for the adult life stage. However, the nymphal stages possess a very different morphology (i.e., platyform with lateral segmental appendages) that cannot be used effectively for protection. The nymphal stages possess a caudal bifurcated process (i.e., urogomphi) where exuvia are accumulated, as the beetle develops, and covered with fecal matter, this assemblage can be best described as a small green caudal globe that can represent from 1 / 5 to 1 / 3 of the overall size of the beetle (Fig. 1b). Because the urogomphi are articulated with muscles, the beetle can raise this caudal globe over the body in an umbrella or shield like fashion and conspicuously display it. Moreover, in the event of a threat, this display is accompanied by a typical defensive behavior in which the caudal globe is actually pointed in the direction of the threat (scorpion syndrome). Thus, we propose that the nymphal stages are protected chemically by these accumulations of fecal matter in combination with a behavioral response, and that this protection represents a case of external sequestration of plant derived chemicals. We have now shown that P. arizonae covers its exuvia with resinous fecal deposits that comprise up to 90 % (w/w) of a mixture of two sesquiterpene lactones, Damsin (1) and Damsinic acid (2), sequestered from A. ambrosioides. Flavonoids were also present in the mixture but were not characterized. The phenomenon of sequestration of natural products is a well documented biological event occurring extensively in the Arthropoda [1, 2]. The uptake of toxic substances of exogenous origin, for purposes of defense, has been dramatically demonstrated in several plant-insect interactions [3, 4]. In the class Insecta, plant-derived chemicals are often stored internally in the hemolymph or in specialized glands, and used in acts of reflex bleeding or active secretion triggered by the encounter with a potential predatorial threat [1]. However, less common are the cases of sequestration in which the toxic chemicals are stored externally (i.e., appendages or whole body coatings). The possible advantage of such a defense posture is its utilization as visual or olfactory warning of chemical protection. It is only recently that the chemistry of these defense strategies or “shields” has been recognized [5-7]. In this study, we investigated the chemistry of an interesting case of external sequestration in the family Cassidinae within the Coleoptera. One such tortoise beetle species, Physonota arizonae [8], is endemic to the southwest United States and northern Mexico and it uses as host plant a commonly occurring shrub from the family Asteraceae, Ambrosia ambrosioides (canyon ragweed). We became very interested in learning the chemical composition of the caudal globe the nymphal stages carry, in an attempt to infer from this composition any potential defensive value. 144 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Manuel Aregullin and Eloy Rodriguez Fig. 1. 1a (left) Adult tortoise beetle Physonota arizonae with typical Cassidinae morphology. 1b (right) Larval stages of Physonota arizonae feeding on host plant Ambrosia ambrosioides and displaying defensive fecal shields (FS). Experimental Collection of Plant Material and Beetles. Ambrosia ambrosioides leaf material containing feeding Physonota arizonae beetles were collected from wild populations in washes, in the Coyote Mountains off of Hwy 86, approximately 20 mi. west of Tucson, Arizona during the months of July and August. The leaf plant material was brought to the laboratory, and the beetles were removed manually and processed separately. The leaf plant material was air-dried and ground prior to solvent extraction, and from the beetles the exuviae and its resinous coating were collected and extracted fresh with organic solvents. Ambrosia ambrosioides Extraction and Chemical Analysis. Dried and ground leaf plant material (250 g) was extracted with 1500 mL of chloroform and magnetic stirring overnight. The chloroformic extract was filtered and the solvent evaporated to dryness in a rotavapor under vacuum. The residue was retaken in methanol to remove most of the hydrocarbons and filtered. The filtrate was evaporated to dryness to yield 24 g (9.6 %) of methanol soluble crude extract. The crude extract was dissolved in methanol and applied to a Sephadex LH-20 chromatographic column and eluted with methanol. Fractions from the column were collected according to their fluorescence under longwave light from a portable ultraviolet lamp. Fractions 1 and 2 showed to contain a mixture of hydrocarbons that was not further characterized. Fraction 5 was shown to contain the sesquiterpene lactone Damsin (1). Fraction 6 was shown to contain the sesquiterpene lactone Damsinic acid (2). The structures of the two sesquiterpene lactones were established by spectroscopic methods (i.e., UV, IR, 1H-NMR and 13H-NMR, and MS), and comparison with authentic samples. Fractions 7-10 were shown to contain a mixture of flavonoids that was not further characterized. Beetle Resinous Coating Extraction and Chemical Analysis. Approximately 50 to 100 exuvia covered with feces were removed from the beetles and extracted fresh with chloroform. The chloroformic extract was filtered, and the solvent evaporated to dryness in a rotavapor under vacuum. The residue was redissolved in methanol, the insoluble fraction was discarded, and the filtrate evaporated to dryness to yield a gummy residue. The residue was dissolved in methanol and applied to a Sephadex LH-20 chromatographic column and eluted with methanol. Several fractions from the column were collected according to their different fluorescence under longwave ultraviolet light. Two fractions showed to contain each one a discrete sesquiterpene lactone that upon spectral analysis were shown to be Damsin (1) and Damsinic acid (2). The structures of (1) and (2) were established by spectroscopic analysis and comparison with authentic samples. Results and discussion Field observations have revealed the occurrence of the tortoise beetle Physonota arizonae on wild populations of canyon ragweed (Ambrosia ambrosioides), suggesting high host-plant specificity for this Asteraceae. All four life stages of the beetle (i.e., eggs, larvae, pupae and adults) were present on the host plant. Field observations also suggested that the appearance of the beetle on the plant is closely associated with the initiation of the regional monsoon season that usually starts in JulyAugust [9]. We suspected that the highly conspicuous accumulation of fecal matter (primarily derived from plant material), in combination with a discrete defensive behavior by the nymphal stages, could be implicated in the chemical protection of the beetle from potential predators (i.e., birds, lizards, etc.). P. arizonae larvae, feeding on the plants, were collected from wild populations of A. ambrosioides to obtain enough material for chemical analysis. It was found that the fecal excretions covering the urogomphi were essentially lipophilic and that dissolved in chloroform rather easily. Thus, the chloroformic solution of the excretions was analyzed preliminary by tlc using a standard solvent system for terpenic chemicals (i.e., chloroform-acetone, 9:1) [10], the tlc plates were sprayed with a vanillin spray reagent highly specific for terpenoids [11]. This preliminary analysis revealed the presence of two major components in the excretion. Sesquiterpene Lactone Sequestration by the Tortoise Beetle Physonota arizonae (Cassidinae) In further analysis, the chloroform washings were combined and the solvent evaporated to yield a green resinous material. The green resinous material was dissolved in methanol to remove the lipid fraction, and filtered. The filtrate was concentrated and applied to a Sephadex LH-20 chromatographic column and eluted with methanol. Several fractions were collected and tlc analysis revealed that fractions 5 and 6 contained the two major constituents of the fecal excretions identical to the ones originally detected by tlc. Final purification of the constituents of fractions 5 and 6 was achieved by preparative tlc. The 1H-NMR spectra of fractions 5 and 6 had a very important diagnostic value in determining that fraction 5 contained a sesquiterpene lactone (i.e., doublets at 6.27 and 5.57 ppm with a coupling constant of approximately 3 Hz corresponding to the two protons in the exocyclic double bond of the γ-butyrolactone), and that fraction 6 contained a structurally related compound. This initial finding suggested that the origin of these sesquiterpene lactones was actually the host-plant. The sesquiterpene lactone chemistry of A. ambrosioides has been previously reported [12-14] and it is known that there are geographical variations in the chemistry of A. ambrosioides [15]. Following the same isolation procedure as the one used in the case of the beetle excretions our study revealed that the population sampled, contained the sesquiterpene lactones Damsin (1) and Damsinic acid (2). Moreover, we determined by 1H- and 13C-NMR that these sesquiterpene lactones are identical to the compounds in fractions 5 and 6 from the beetle secretions. It is well known that sesquiterpene lactones are plant secondary metabolites with a wide array of very important biological activities (i.e., antifungal, insecticidal, allergenic, antitumoral, etc.) on different biological systems [16-18]. It is reasonable to suggest that for purposes of insect chemical protection, the sesquiterpene lactones occurring naturally in plants should be effective deterrents of predators. We are currently conducting experiments to demonstrate that arthropods that sequester and use sequiterpene lactones as chemical defenses are capable of deterring more effectively potential predators (i.e., birds, lizards, and other arthropods), and parasites (i.e., wasps). In summary, this is the first instance in which sesquiterpene lactones are shown to be in the repertoire of plant chemicals that are of defensive value to arthropods. Moreover, the study of sequestration as presented by the tortoise beetles should provide us with better insights into the evolution of sequestration. Further study of tortoise beetles species within the genus Physonota, its related genera, and their associations H H O O O O OH O (1 ) Damsin D am sin (2 ) Damsinic D am sinicacid a cid 145 with other plant species will shed light on the plant-insect coevolutionary and adaptive aspects and the insect-predator and insect-parasite relationships. It is now necessary to evaluate the defensive value to the insect of this new type of sequestered substance in order to understand better the efficacy of its deterring activity. Acknowledgements The authors wish to thank Dr. William S. Bowers and Dr. Floyd G. Werner for their support and technical assistance. The authors also wish to thank NIH grant TW000076-0751, and Cornell Hatch funds for financial support. References 1. Blum, M.S. Chemical Defenses of Arthropods. Academic Press, London, 1981. 2. Duffey, S.S. Sequestration of Plant Natural Products by Insects. Annual Review of Entomology 1980, 25, 447. 3. Rothschild, M. Secondary plant substances and warning colourations in insects, In H.F. Van Emden (ed.). Insect / Plant Relationships, Symp. R. Entomol. Soc. London 6, Blackwell Sci., Oxford, 1972. Pg. 59-83. 4. Schilknecht, H. Endeavour 1970, 30, 136. 5. Olmstead, K.L.; Denno, R.F. Ecology 1993, 74, 1394. 6. Gomez-Nelida, E.; Witte, L.; Hartmann, T. J. Chem. Ecol. 1999, 25, 1007. 7. Vencl, F.V.; Morton, T.C.; Mumma, R.O.; Schultz, J.C. J. Chem. Ecol. 1999, 25, 549. 8. Sanderson, M.W. Ann. Entomol. Soc. Am. 1948, XLI, 468. 9. Werner, F. Personal communication. 10. Yoshioka, H.; Mabry, T.J.; Timmermann, B.N. Sesquiterpene Lactones. Chemistry, NMR and Plant Distribution. University of Tokyo Press, Japan, 1973. 11. Picman, A.K.; Ranieri, R.L.; Towers, G.H.N.; Lam, J. J. Chrom. 1980, 189, 187. 12. Doskotch, R.W. and Hufford, C.D. J. Org. Chem. 1970, 35, 486. 13. Higo, A.; Hammam, Z.; Timmermann, B.N.; Yoshioka, H.; Lee, J.; Mabry, T.J. Phytochemistry 1971, 10, 2241. 14. Romo, J.; Romo de Vivar, A.; Velez, A.; Urbina, E. Can. J. Chem. 1968, 46, 1535. 15. Seaman, F.C. Bot. Rev. 1982, 48, 121. 16. Rodríguez, E.; Towers, G.H.N.; Mitchell, J.C. Phytochemistry 1976, 15, 1573. 17. Picman, A.K. Biochem. Syst. Ecol. 1986, 14, 255. 18. Robles, M., Aregullin, M., West, J., and Rodriguez, E. Planta Medica 1995, 61, 199. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 146-150 Investigación Mechanism of Glutamate Neurochemistry: Electron Transfer and Reactive Oxygen Species Peter Kovacic1, Ratnasamy Somanathan2* and Michelle Inzunza1 1 2 Department of Chemistry, San Diego State University, San Diego, CA 92182-1030, USA. Centro de Graduados e Investigación del Instituto Tecnológico de Tijuana, Apdo Postal 1166, 22000 Tijuana, B.C. México. E-mail: [email protected] Recibido el 8 de abril del 2003; Aceptado el 3 de junio del 2003 Dedicated to Professor Alfonso Romo de Vivar Abstract. Glutamate (Glu) undergoes metabolism to an imine derivative. We propose involvement of the conjugated α-iminocarboxylic acid in neurotoxicity and possibly in neurotransmission. Electrochemistry, captodative effect and bioactivity of related cyclic α-imino acids are relevant. There is also consistency with background literature of Glu indicating participation of oxidative stress, reactive oxygen species, and electron transfer. Alternatively, metal chelates of Glu and Glu imine may play a role. Various analogs of Glu imine were synthesized, namely, oxime and a cyclic model derived from cyclization of the intermediate hydrazone. Keywords: Glutamate, neurotransmission, toxicity, imine, electron transfer, reactive oxygen species. Resumen. El glutamato (Glu) se metaboliza a un derivado de imina. Se propone la inclusión del ácido iminocarboxílico conjugado en la neurotoxocidad y posiblemente en la neurotransmisión. La electroquímica, el efecto captodativo y la bioactividad del los ácidos αimino-cíclicos son relevantes. Los hallazgos son consistentes con los antecedentes de la literatura que indican la participación de Glu en el estrés oxidativo, en la química de las especies reactivas de oxígeno, y en la transferencia electrónica. Alternativamente, los quelatos metálicos de Glu y de iminas Glu, juegan un papel importante. Varios análogos de imina Glu fueron sintetizados, en particular la oxima, y un modelo cíclico derivado de la ciclización de la hidrazona intermediaria. Palabras clave: Glutamato, neurotransmisión, toxicidad, iminas, transferencia electrónica, especies reactivas de oxígeno. Introduction lend support to the thesis. Several synthetic analogs, both acyclic and cyclic, of the α-imino metabolite were synthesized. During the last score of years extensive evidence has accumulated in support of involvement of oxidative stress (OS) with both endogenous and exogenous agents, including anti-infective drugs [1], anticancer agents [2], carcinogens [3], reproductive toxins [4], nephrotoxins [5], hepatotoxins [6], and various others [7a]. The most common reactive oxygen species (ROS) are superoxide, hydrogen peroxide, peroxyl radicals, and the important hydroxyl radical, that are generated by electron transfer (ET). ET functionalities comprise quinones (or precursors), metal complexes (or chelators), aromatic nitro compounds (or reduced products), and imines (or iminiums), which on redox cycling with oxygen give rise to ROS. The present article will focus on the conjugated imine category. In some cases, ET occurs without oxygen participation. Very many bioactive substances or their metabolites incorporate ET groups. OS can produce beneficial results, as with drugs, or unwanted side effects in toxicity. The mode of action is usually complex and probably multifaceted in many instances. This report deals with the mode of action of glutamate (Glu), both in neurotransmission and in toxicity. We propose that the α-imino metabolite may play a role as an ET agent in these processes. Glu metabolism, generation of ROS and prior literature on bioactivity of related α-iminocarboxylic acids Metabolism In relation to mechanism, Glu metabolism has attracted scant attention even though metabolites often play an important role in physiological activity. It is significant that Glu can serve as substrate for an enzyme that effects conversion to the imine derivative (1a), particularly since Glu dehydrogenase plays a central role in amino acid deamination because in most organisms Glu is the only amino acid that has an active dehydrogenase [8]. The labile product (1a) can undergo several subsequent conversions. Nucleophilic attack by a basic primary amino NR O O HO OH 1 a.) R= H b) R= Substituent Mechanism of Glutamate Neurochemistry: Electron Transfer and Reactive Oxygen Species entity, such as protein lysine, with elimination of ammonia, gives rise to the more stable imine 1b possessing a substituent on nitrogen. Alternatively, 1a could undergo hydrolysis to the α-keto acid which also serves as precursor of (1b) by reaction with pri-amine [8]. We will explore possible ramifications relative to mode of action resulting from generation of conjugated imine (1b). Several common routes are generally available for imine and iminium synthesis in biological systems, including oxidation of aliphatic amines and nonenzymatic condensation of carbonyl compounds with the pri-amino moiety of basic amino acids in protein. Hence, these transformations are applicable to Glu, making for relevance to our report. Regarding the iminium types, they can also be derived simply by alkylation or protonation of imine. Synthesis Cyclic iminocarboxylates were synthesized from α-ketoglutaric acid (2a), by condensing with hydrazine in methanol to yield cyclic ester 2 which on hydrolysis gave the cyclic iminocarboxylic acid (2c) in good yield. Similarly, acyclic types were prepared by condensing 2a with hydroxylamine in methanol to provide the corresponding methyl iminocarboxylate 2d and the carboxylic acid 2e, which were separated by preparative thin layer chromatography (Scheme-1). Neurotoxicity Even before a decade ago, substantial data had built up pointing to OS as a factor in neuropathology by excitory amino acids, with Glu as the major effector [9]. O O O OH 2a OH NH2NH2 . HCl NH2OH.HCl CH3OH CH3OH OH O N OCH3 O N H N O HO n+ O O O 3 Prominent evidence includes generation of ROS and subsequent oxidative damage; radical scavengers and inhibitors prevent neuronal degradation. Other suspected sources of OS are nitric oxide, peroxynitrite, calcium, iron, and activated ROS-generating enzymes. Two broad mechanisms were offered to account for cell vulnerability, based on OS and excessive activation of Glu receptors. In one case, a sequential process pertains, whereas the other, of particular interest, entails interaction of the two. Four years later, increasing evidence supported the claim that excitotoxicity and oxidative stress play important roles in pathogenesis of both acute and chronic neurologic diseases [10]. The view was reiterated that the two effects may cooperate to induce neuronal degeneration. Hydroxyl radical levels and the volume of lesions were attenuated by spin trapping, pointing to radical scavenging. From our standpoint, the cooperative effect consists of ET entailing the receptor bound species. On the other hand, ET by imine might occur at another site in keeping with the sequential scheme. The literature provides little discussion of specific, detailed pathways for OS generation from Glu, partly due to lack of working hypotheses. We suggest that the α-imine metabolite may play a role in redox cycling to produce ROS, at least in toxicity. This approach is in harmony with various observations, “Although the activation of Glu receptors is a key step in the sequence of events leading to neuronal degeneration, it is by no means all that is necessary… thus delayed neurotoxicity has been effectively dissociated from neuronal excitation...” [9]. A goodly number of other investigations document involvement of OS, of which several will be cited, including generation of ROS [11-13], induction of lipid peroxidation [13a], DNA fragmentation [13b] and protection by antioxidants [13-15]. In addition, involving OS from Glu imine in neurotoxocity represents yet another example in the widespread documentation of ROS from conjugated imine or iminium in toxic manifestations [3-7a]. R N M HO OCH3 2d M H2N O OCH3 2b 147 n+ O O O 4 O N OH O N H 2c Neurotransmission OH O O N OH OH 2e Glu and related amino acids appear to be the major excitatory neurotransmitters in the brain with involvement in 40 % of the synapses [9]. The importance of Glu is reflected by its presence in the CNS in relatively large quantities. An hypotheti- 148 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) cal role for ET by Glu imine might also be invoked in this category. Although ROS at high concentrations induce adverse effects, OS at low levels may contribute in the transmission process. By analogy, ROS have both beneficial and damaging effects, depending on various factors, on sperm in the fertilization process [16]. There is general consensus that neurotransmission occurs by ionic pathways [17]. It should be recognized that ET is not necessarily incompatible with polar processes. Movement of negative electrons (ET) induces an electric field which could affect the migration of negative and positive ions, e.g. Cl, Ca, Na and K, in the vicinity. Electrons have been shown to migrate over substantial distances [18]. Another analogy might comprise movement of electrons in a conducting copper wire. The ET framework may also be applicable to other neurotransmitters, such as, nitric oxide [19] and catacholamines [7b]. Metal complexes Another plausible scenario for ET by Glu involves the corresponding metal complexes. This general class of ET agents is omnipresent in the major drug and toxin categories [1-7]. There is the favorable feature generally of quite positive reduction potentials which energetically favor ET in vivo. Since the α-aminoacid moiety is a facile chelator, metal derivatives (3) are well documented [20]. Similarly metal chelates (4) of the α-imino metabolites should also be considered. Hence, it is conceivable that such complexes 3 and 4 might participate in toxicity and/or transmission. Possible involvement of cyclic analogs in physiological activity is discussed in the subsequent section. Cyclic α-iminocarboxylic acids Prior literature contains various reports on participation of these species in physiological activity from the ET viewpoint. ß-Lactams inactivate bacterial cell wall enzyme by covalent binding. Little attention has been given to the fact that aiminocarboxylic acids, e.g., 5 from cephalosporins, arise in the process [21]. Peter Kovacic et al. N N S HO2C HO2C 7 6 In relation to bioactivity of ∆2-thiazoline-2-carboxylate (8), this compound in the category has been the object of most attention [22]. Hypotheses have been advanced that it is an intracellular messenger for insulin, and an effector of diverse metabolic activities, such as diuretic response and cell growth [23]. Also, it is a potent inhibitor of dopamine-ß-hydroxylase [24], a metalloenzyme responsible for producing norepinephrine. Compound 8 possessed electron affinity compatible with ET in vivo [21]. Alternatively, all ligands (6-8) in this category are expected to be avid chelators of metal ions in the biological milieu, forming complexes of type (9). Cu and Fe chelates with (6) and (8) exhibit quite positive reduction potentials [22]. The role of ET-OS in eliciting a variety of physiological responses from metal complexes is discussed elsewhere [1-7a]. N S N n+ M O -O O O 8 9 A related chelator is pyridine-2-carboxylic acid [25], whose anticancer activity has been attributed to redox cycling entailing OS by a derived metal complex [26]. Conformational Restriction and Bioactivity The relatively high activities of kainic (10) [27] and domoic (11) [28] acids in neurotransmission have been attributed to conformational restriction [29] imposed by the cyclic structures. Just how this property translates into improved activity is unknown. NHCOR S EO N O CO2H 5 Model compounds, such as 6 and 7, displayed favorable reduction potentials, raising the possibility of ET participation by 5 in vivo subsequent to site binding, both in toxicity and in antibacterial action. The values increased with decreasing pH in line with formation of conjugated iminium which conceivably might be an actor in the physiological manifestations. It is indicative that thiazolidine-4-carboxylic acid undergoes dehydrogenation in vivo to the corresponding α-imino acid (7) [22]. By analogy, 10 and 11 would be converted into derivatives of 6 which has favorable properties for ET in biosystems. Mechanism of Glutamate Neurochemistry: Electron Transfer and Reactive Oxygen Species 149 Although the synthetic counterpart (2c) fits into the cyclic imino acid category, it is deficient in lacking an N-alkyl containing substituent and a second carboxyl group. The favorable influence shown by 10 and 11 may result from assistance in coplanarity for the radical anion formed from the proposed imine metabolite on electron uptake, thus promoting resonance stabilization. There is widespread presence of a γ-carboxyl group in the various neurotransmitters, whose role has not been delineated, perhaps as a site binder. In the acyclic structures, the γ-carboxyl may decrease adverse steric interaction with the other carboxyl by association with iminium nitrogen, as depicted in 12, but not as effectively as for the covalent cyclic category. anhydrous sodium sulfate, removal of solvent gave a colorless solid (1.37 g, 80 % yield). mp. 120-125 °C; IR (KBr) 3138, 2938, 1714, 1622, 1435, 1287, 1206, 1000 cm–1; 1H NMR (200 MHz, CDCl3) δ 3.91( s, 3H), 2.86 ( t, 3H, J=8.0Hz), 2.60 (t, 3H, J=8.0Hz) ppm; EIMs: 156 (100 %), 124. . Synthesis of 2-(Hydroxyimino)pentanedioic acid dimethyl ester (2d) and 2-(Hydroxyimino)pentanedioic acid (2e) [35]. α-Ketoglutaric acid (1.61 g, 11 mmol) and hydroxylamine hydrochloride (0.77 g, 11 mmol) were dissolved in methanol (20 mL) and stirred overnight at room temperature. After solvent was removed under reduced pressure, the residue was dissolved in dichloromethane and the organic phase was washed with water. The organic phase was dried over anhydrous sodium sulfate and removal of solvent gave a viscous liquid which on triturating with diethyl ether gave a crystalline solid (1.76g, 85 % yield). The crude solid was separated on Chromatotron using dichloromethane: methanol (3 %) as eluting solvent to give (2d) and (2e) in 30 and 45 % yields, respectively, which were characterized as follows: HN .. δ+ HO δ- O 12 Captodative effect Various reports have shown that carbon radicals benefit from enhanced stabilization when attached to both an electron withdrawing and electron donating substituent [30]. As a result, the combined influence is synergistic. Investigators predicted that this effect would find application in vivo [30]. The following examples can be cited: paraquat, flavins and quinones [22]. The captodative radical formed from one- electron reduction of α-iminium carboxylic acid is depicted in 13. H O R + H N - O O O 13 Future studies with model compounds 2c and 2e might provide further insight into mechanistic aspects. The unifying theme of ET and ROS in neurochemistry is further elaborated in a review [31] dealing with nitric oxide, catecholamines, and Glu. Experimental Synthesis of 1,4,5,6-tetrahydro-6-oxo-3-pyridazine carboxylic acid methyl ester (2b). α-Ketoglutaric acid (1.61g, 11 mmol) and hydrazine dihydrochloride (1.16g, 11 mmol) were stirred together in methanol at room temperature overnight. Following removal of solvent under reduced pressure, the residue was dissolved in dichloromethane and washed with water. After the organic phase was dried over 1,4,5,6-Tetrahydro-6-oxo-3-pyradazinecarboxylic acid (2c). The above ester (2b) (0.5g) was hydrolyzed with 10 % methanolic sodium hydroxide and neutralized with DOWEX cation exchange resin to give the acid (2c) in quantitative yield. mp. 194-196 °C; IR ( KBr) 3382, 1720; 1H NMR ( 200 MHz, CDCl3) δ 10.85 (s, 1H), 2.84 (t, 2H, J= 8.0 Hz), 2.48 ( t, 2H, J=8 Hz) ppm. Prior synthesis: α-Ketoglutaric acid and hydrazine [32,33], lit. [34] mp.194 °C for 2c. 2-(Hydroxyimino)pentanedioic acid dimethyl ester (2d). [35]. Colorless solid (30 %); mp 113-115 °C; IR(KBr) 3382, 2956, 1733, 1442, 1127, 1035, 801 cm–1; 1H NMR (200 MHz, CDCl3) δ 3.70 (s, 3H ), 3.60 (s, 3H), 2.82 ( t, 2H, J= 8.0 Hz), 2.52 ( t, 2H, J= 8.0 Hz) ppm; 13C (50 MHz, CDCl3) δ 172.92, 163.75, 150.87, 52.40, 51.58, 29,59, 20.03 ppm; MS (EI) 189, 156 (100 %). 2-(Hydroxyimino)pentanedioic acid (2e) [36]. Colorless solid (45 %); mp. 152-153 °C; lit. [36] mp. 152 °C. IR (KBr) 3422, 1734, 1666, 1448, 1029, 635 cm–1; 1H NMR (200 MHz, CDCl3) δ 12.16 ( 2H), 2.83 (t, 3H J= 8.00 Hz), 2.52 ( t, 2H, J= 8.0 Hz) ppm. References 1. Kovacic, P.; Becvar, L.E. Curr. Pharmaceut. Des. 2000, 6, 143167. 2. Kovacic, P.; Osuna, J.A. Curr. Pharmaceut. Des. 2000, 6, 277309. 3. Kovacic, P.; Jacintho, J.D. Curr. Med. Chem. 2001, 8, 773-796. 4. Kovacic, K.; Jacintho, J.D. Curr. Med. Chem. 2001, 8, 863-892. 5. Kovacic, P.; Sacman, A.; Wu-Weis, M. Curr. Med. Chem. 2002, 9, 823-847. 6. Poli, G.; Cheeseman, K.H.; Bianzani, M.U.; Slater, T.F. Eds. Free Radicals in the Pathogenesis of Liver Injury. Pergamon Press, New York. 1989. 150 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) 7. Halliwell, B.; Gutteridge, J.M.C. Free Radicals in Biology and Medicine. Oxford University Press, New York, 1999, (a) pp. 1859. (b) p. 570. 8. Lehninger, A. Biochemistry. Worth, New York, 1975, pp. 565566. 9. Coyle, J.T.; Puttfarcken, P. Science 1993, 262, 689-695. 10. Lancelot, E.; Revaud, M.L.; Boulu, R.G.; Plokine, M.; Callebert, J. Free Rad. Biol. Med. 1997, 23, 1031-1034. 11. Said, S.I.; Pakbaz, H.; Berisha, H.I.; Raza, S. Free Rad. Biol. Med. 2000, 28, 1300-1302. 12. Loikkanen, J.J.; Naarala, J.; Savolainen, K.M. Free. Rad. Biol. Med. 1998, 24, 377-384. 13. Yasui, Y.; Mawatari, K.; Higuchi, Y.; Tanu, H.; Kato, S. In: Free Radicals in Brain Physiology and Disorders, Packer, L.; Hiramatsu, M.; Yoshikawa, T. Eds., Academic Press, New York, 1996, (a) p.58, (b) pp. 60, 6 14. Kobayashi, M.S.; Han, D.; Packer, L. Free Rad. Res. 2000, 32, 115-124. 15. Hampson, A.J.; Grimaldi, M.; Lolic, M.; Wink, D.; Rosenthal, R.; Axelrod, J.; Ann. N.Y. Acad. Sci. 2000, 899, 274-282. 16. Sikka, S.C. Curr. Med. Chem. 2001, 8, 851-862. 17. Voet, D.; Voet, J.G. Biochemistry, Wiley, New York. 1995, Sect. 34C. 18. McLendon, G. Acc. Chem. Res. 1988, 21,160-167. 19. Kovacic, P. Bioelectrochem. Bioenerg. 1996, 39, 155-159. 20. Lipschitz, J.; Schouteden, F.L.M. Rec. Trav. Chim. 1939, 58, 411-422. 21. Kovacic, P.; Jawdosiuk, M.; Ames, J.R.; Ryan, M.D. Bioorg. Chem. 1987, 15, 423. Peter Kovacic et al. 22. Kovacic, P.; Popp, W.J.; Timberlake, J.W.; Ryan, M.D. Chem. Biol. Interact. 1989, 69, 235- 244 and references therein. 23. Hamilton, G.A. Adv. Enzymol. 1985, 57, 85-178. 24. Naber, N.; Venkatesan, P.P.; Hamilton, G.A. Biochem. Biophys. Res. Commun. 1982, 107, 374-380. 25. Schneider, F.; Schaeg, W. Z. Physiol. Chem. 1962, 327, 74 26. Kovacic, P.; Ames, J.R.; Lumme, P.; Elo, H.; Cox, O.; Jackson, H.; Rivera, L.A.; Ramirez, L.; Ryan, M.D. Anti- Canc. Drug. Des. 1986, 1, 197-214. 27. Cooper, J.R.; Bloom, F.E.; Roth, R.H. The Biochemical Basis of Neuropharmacology, Oxford University Press, New York, 1996, p. 174. 28. June, D.E.; Hoo, K.; Kamboj.; Deversill, M.; Blackman, D.; Mandelzys, A.J. Med. Chem. 1997, 40, 3645-3650. 29. Guroff, G. Molecular Neurobiology, Marcel Dekker, New York, 1980, p. 430. 30. Viehe, H.G.; Janousek, Z.; Merenyi, R.; Stella, L. Acc. Chem. Res. 1985, 18, 148-154. 31. Jacintho, J. D.; Kovacic, P. Curr. Med. Chem. in press. 32. Lagna, W.M.; Callery, P.S. J. Labelled Compd. Radiopharm. 1984, 21, 337. 33. Kline, G.B.; Cox, S.H. J. Org. Chem. 1961, 26, 1854-1856. 34. Kaupp, G.; Schmeyer, J. J. Phys. Org. Chem. 2000, 13, 388. 35. Hoffman, C.; Tanke, R.S.; Miller, M. J. Org. Chem. 1989, 54, 3750-3751. 36. Ahmad, A.; Spenser, I. D. Can. J. Chem. 1961, 39, 1340-1359. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 151-154 Investigación Cumarinas presentes en especies del género Casimiroa Aída N. García-Argáez,1 Nadia M. González-Lugo,2 Carmen Márquez2 y Mariano Martínez-Vázquez2* Departamento de Ecología y Recursos Naturales, Facultad de Ciencias, Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México D. F. 2 Instituto de Química, Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México D. F. Tel: +(52) 56224403; Fax: + (52) 56162203; E-mail: [email protected] 1 Recibido el 21 de abril del 2003; aceptado el 13 de junio del 2003 Dedicado al Dr. Alfonso Romo de Vivar Resumen. Los extractos orgánicos de hojas y semillas de Casimiroa pubescens, C. edulis y C. calderoniae se analizaron mediante cromatografía líquida de alta resolución; los resultados mostraron la presencia de las cumarinas felopterina (6), isopimpinelina (9), heraclenol (15) y heraclenina (16) en las tres especies estudiadas. En promedio la mayor concentración de este tipo de metabolitos se presentó en C. calderoniae. Estos resultados, además de los previamente reportados en la literatura indican la presencia de 16 cumarinas en el género. Aun cuando se ha postulado que las cumarinas aisladas del género Casimiroa tienen como precursor común a la umbeliferona, existe una clara diferencia entre las cumarinas sintetizadas por C. greggii y al resto de las especies estudiadas. Así la seselina y el Ogeranil-ostenol, ambas provenientes de la umbeliferona prenilada en C-8, son sintetizadas por C. greggii, mientras que las furanocumarinas sintetizadas por las demás especies tienen como precursor a la umbeliferona prenilada en C-6. El presente trabajo constituye el primer estudio fitoquímico de C. pubescens y C. calderoniae. Palabras clave: Casimiroa, cumarinas, Rutaceae, umbeliferona. Abstract. Leaf and seeds extracts of Casimiroa pubescens, C. edulis and C. calderoniae were analyzed by HPLC. The results showed that coumarins phellopterin (6), isopimpinellin (9), heraclenol (15) and heraclenin (16) were present in all species studied. Of all species, C. calderoniae showed the highest concentration of this type of metabolites. These results, in addition to those previously reported, indicate that 16 different coumarins occur in the genus. Even though it has been proposed that the umbelliferone is a common precursor in the biogenesis of the Casimiroa coumarins, there is a clear difference between the coumarins synthesized by C. greggii and those present in the rest of the species. Thus, seselin and O-geranyl-osthenol, both synthesized from an umbelliferone prenylated at C-8, are present in C. greggii, while the coumarins present in the other species originate from umbelliferone prenylated at C-6. This is the first study of C. pubescens and C. calderoniae. Keywords: Casimiroa, coumarins, Rutaceae, umbelliferone. Introducción Se ha postulado que las furano- y pirano-cumarinas se forman biogenéticamente cuando al núcleo cumarínico se adiciona un grupo prenilo y éste interacciona con un grupo ortofenólico, de tal modo que se generan diferentes estructuras con un anillo heterocíclico adicional. Con base a lo anterior se ha propuesto que una prenilación en la posición 6 de la umbeliferona da origen a las furanocumarinas lineales como el psoraleno, mientras que una prenilación en la posición 8 da origen a las furanocumarinas angulares como la angelicina (3). Algunos autores argumentan que las pirano- y furanocumarinas, tanto lineales como angulares, comparten caminos biosintéticos y solo se diferencian en la etapa final de ciclización; lo anterior se ha demostrado experimentalmente, utilizando marcadores radioactivos, en la elucidación de la transformación biogenética de la demetilsuberosina a psoraleno en Ruta graveolens (3). En el presente trabajo se dan a conocer los resultados del análisis por cromatografía de líquidos de alta resolución, de las cumarinas presentes en extractos de hojas y semillas de Casimiroa pubescens, C. edulis y C. calderoniae. Adicionalmente se comparan los resultados obtenidos, con los diferentes tipos de cumarinas presentes en C. greggii. Las cumarinas son probablemente los metabolitos más comunes derivados de la ruta biosintética del shikimato-corismato (1). En miembros de la familia Rutaceae se han encontrado aproximadamente 200 cumarinas y las evidencias experimentales han demostrado que éstas se sintetizan por las mismas rutas biosintéticas observadas en otras familias de plantas, donde la umbeliferona se considera el intermediario común para la biosíntesis de cumarinas lineales y angulares (2). El principal factor de diversificación estructural de las cumarinas en las Rutaceae es la amplia incorporación de unidades prenilo al núcleo cumarínico; en algunas especies de esta familia se ha demostrado que la prenilación ocurre cuando se ha formado la umbeliferona (1, 2). Las modificaciones secundarias sobre los grupos prenilo, usualmente iniciadas por epoxidación del doble enlace, contribuyen a esta diversificación estructural de manera importante. La transformación del doble enlace al diol respectivo en la cadena prenilada lateral, teniendo al epóxido como intermediario, ha sido demostrada con marcadores radioactivos (1). 152 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Aída N. García-Argáez et al. Tabla 1. Cumarinas presentes en el género Casimiroa. Compuestos Especie C. greggii Seselina (1) O-Geranil-ostenol (2) 8-Geraniloxipsoraleno (3) Bergapteno (4) Xantotoxol (5) Felopterina (6) Cumarina 7 Cumarina 8 Isopimpinelina (9) Escopoletina (10) Ester metílico de escopoletina (11) 5-Geranil-oxipsoraleno (12) 8-Geranil-5 metoxi-oxipsoraleno (13) 9-Hidroxi-4-metoxi-furano-(3,2,6) benzopiran-7-ona (14) Heraclenol (15) Heraclenina (16) √ √ Resultados y discusión Los resultados del análisis cromatográfico indicaron la presencia de las cumarinas felopterina (6), isopimpinelina (9), heraclenol (15) y heraclenina (16), tanto en hojas como en semillas de las tres especies estudiadas (Tabla 1). La presencia de estos compuestos se comprobó mediante procedimientos de co-cromatografía. Sin embargo, no se detectó la presencia de la seselina (1), cumarina previamente aislada de C. greggii (5). El análisis de resultados para cada especie se presenta por separado. C. pubescens. El extracto de las hojas de esta especie mostró a la heraclenina como la única y más abundante cumarina presente en los cinco individuos estudiados; la felopterina se detectó en cuatro individuos, mientras que la isopimpinelina se encontró en dos individuos y el heraclenol solamente en uno. La misma tendencia se observó en el análisis del extracto de semillas. Estos resultados indicaron que en esta etapa de crecimiento de las plantas, y en este sitio de colecta, la especie acumula y/o sintetiza preferentemente heraclenina y felopterina (Fig. 1). C. edulis. El estudio del extracto de las hojas de esta especie mostró una presencia abundante de las cuatro cumarinas de referencia (6, 9, 15 y 16). Por otra parte, en el extracto de las semillas, la presencia de las cuatro cumarinas fue escasa; la única cumarina presente en todos los individuos fue la felopterina mientras que el heraclenol solamente se detectó en un solo individuo (Fig. 1). C. calderoniae. Los resultados del análisis de esta especie muestran un patrón similar al observado en C. pubescens, i.e. la presencia de 6, 9, 15 y 16, tanto en frecuencia como en con- C. pringlei C. edulis √ √ √ √ √ √ √ √ √ C. pubescens C. calderoniae √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ centración, es mayor en las hojas que en las semillas. Sin embargo son notables las altas concentraciones de heraclenina (20.9 mg / g) y de felopterina (23.8 mg / g) en el extracto de hojas (Fig. 1). Estos resultados combinados con los reportados previamente, demuestran la presencia de 16 cumarinas en el género (Tabla 1) (2, 5-10). La presencia de 10 y 11 en el género se puede considerar un hecho poco frecuente ya que solo la C. edulis sintetiza este tipo de cumarinas. La distribución del resto de las cumarinas en el género indican claramente que difieren en el sitio de prenilación; en C-8 en C. greggii como en 1 y en C-6 en C. edulis, C. pubescens, C. pringlei y C. calderoniae, como en 3-9. Conclusión Los resultados de las tres especies estudiadas indican que la presencia de 6, 9, 15 y 16, pero no de 1, previamente aislada de C. greggii. No obstante existen diferencias notables en los resultados de los análisis de hojas, así en C. pubescens y C. calderoniae las cumarinas con mayor presencia fueron heraclenina (16) y felopterina (6). Sin embargo, la concentración de estas cumarinas fue del orden de cien veces más en C. calderoniae que en C. pubescens. Hasta el momento, las cumarinas aisladas del género Casimiroa indican que tienen como precursor común a la umbeliferona, propuesta biogenética que es general a la familia Rutaceae (2). Sin embargo, tomando en cuenta nuestros resultados y lo reportado hasta el momento, existe una clara diferencia de la distribución entre las cumarinas sintetizadas por la C. greggii y las especies C. edulis, C. pringlei, C. pubescens y C. calderoniae. Cumarinas presentes en especies del género Casimiroa O O O Geranil O O 1 153 O 2 R1 MeO O O R O O RO O 10 R=H 11 R=Me 3 R = O-geranil R1 = H 4R=H R1 = OMe 5 R = OH R1 = H 6 R = isoprenil R1 = OMe 7 R = 6,7-dihidroxi-3,7-dimetil-2-octenil R1 = H 8 R = 4-acetoxi-3-metil-butil R1 = H 9 R = R1 = OMe 12 R = H R1 = O-geranil 13 R = O-geranil R1 = OMe 14 R = OH R1 = OMe 15 R = O-(2,3-dihidroxi-3-metil)butil R1 = H 16 R = O-(2,3-en-3-metil)butil R1 = H Esquema 1 (15) y heraclenina (16), aislados previamente de Decatropis bicolor (4). De cada cumarina se prepararon disoluciones en acetato de etilo a concentraciones de 2.9, 2.6, 2.6, 2.7 y 3.6 mg / mL, respectivamente. Las soluciones estándar se utilizaron para preparar cinco diluciones de cada cumarina en el rango de 0.013 a 0.13 mg / mL. Las curvas de calibración se prepararon inyectando, por triplicado, las diferentes diluciones estándar. Los coeficientes de correlación (r2) para cada gráfica de cada cumarina se calcularon teniendo valores para r2 mayores a 0.997. Obtención y análisis de los extractos hexánico y metanólico. Hojas o semillas secas y molidas de cada individuo se extrajeron tres veces con hexano, por maceración a temperatura ambiente durante 24 h. Los extractos se combinaron y después de evaporar el disolvente a presión reducida se obtuvo el extracto hexánico. Este procedimiento se repitió utilizando metanol para obtener el extracto metanólico correspondiente a cada muestra. De cada extracto se prepararon disoluciones de concentración conocida y se inyectaron en el cromatógrafo de líquidos. Con el fin de identificar la presencia de las cumarinas de referencia a las muestras analizadas se les adicionaron cantidades conocidas de las soluciones estándar. Para el procesamiento de los datos se empleó el software Millenium (Waters), que calcula la cantidad de cada componente presente en cada mililitro de disolución inyectada, Parte experimental Se realizaron los análisis a cinco individuos de cada especie, recolectados en el mismo sitio y en etapa de fructificación. Las cumarinas 1, 6, 9, 15 y 16, utilizadas como metabolitos secundarios de referencia, se aislaron previamente de Decatropis bicolor (Rutaceae) (4). Estándares y curvas de calibración. Se utilizaron estándares de seselina (1), felopterina (6), isopimpinelina (9), heraclenol mg/g de compuesto mg/g de compuesto 0.3 0.2 0.1 1.5 1.0 0.5 0.0 15 9 16 15 6 1.00 16 6 1.50 mg/g de compuesto mg/g de compuesto 9 Casimiroa pubescens 0.80 0.60 0.40 0.20 0.00 15 1.00 0.50 0.00 9 16 6 15 C. edulis 25.0 9 16 6 4.0 mg/g de compuesto Cromatografía Líquida de Alta Resolución. La cromatografía líquida de alta resolución (CLAR) se efectuó en un cromatógrafo de líquidos Waters modelo Delta PREP 4000, equipado con detector UV modelo 486 que se mantuvo a una longitud de onda de 310 nm. Se utiliizó 1 mL / min de flujo del disolvente. Se utilizó MeOH / H2O 50/50 en gradiente hasta MeOH / H2O 90/10 en 20 min. Para el procesamiento de los datos se utilizó el software Millenium (Waters). 0.4 0.0 mg/g de compuesto Material biológico. Se colectaron cinco individuos de cada especie, y los ejemplares de herbario se depositaron en el Herbario de la Facultad de Ciencias de la Universidad Nacional de México, (FCME). Casimiroa edulis se colectó en el Mpio. de Comala, Colima, en julio de 2000 (Nos. de registro del herbario 84847, 84849-84852). C. pubescens se colectó en Ixmiquilpan, Hidalgo, en junio de 2000 (Nos. de registro 84835-84839), y C. calderoniae en la zona árida oaxaqueñopoblana, Oaxaca, en octubre de 2000, (Nos. de registro 848768480). 2.0 0.5 20.0 15.0 10.0 5.0 0.0 3.0 2.0 1.0 0.0 15 9 Hojas 16 6 C. calderoniae 15 9 16 6 Semillas 15 = heraclenol, 9= isopimpinelina, 16= heraclenina felopterina, 6= felopterina Fig. 1. 154 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) tomando como base las curvas de calibración preparadas con los estándares. La cantidad de cada cumarina presente en el extracto hexánico se suma a la de la cumarina correspondiente presente en el extracto metanólico y de esta manera se obtiene el rendimiento de cada producto por gramo de planta seca. Agradecimientos Los autores agradecen el apoyo en el financiamiento parcial del CONACyT (proyecto No. 34992-N) y la Beca de PASPA, DGAPA para los estudios de doctorado de la M. en C. GarcíaArgáez. Aída N. García-Argáez et al. Referencias 1. Dewick, P.M. Nat. Prod. Rep. 1994, 11, 173-203. 2. Gray, A. I.; Waterman, P.G. Phytochem. 1978, 17, 845-864. 3. Murray, R.; Méndez, J.; Brown, S. The Natural Coumarins, John Wiley & Sons Ltd.; Norwich; 1982; 163-185. 4. García-Argáez, A. N.; Ramírez, A. T. O.; Parra, D. H.; Velázquez, G.; Martínez-Vázquez, M. Planta Medica 2000, 66, 279-281. 5. Meyer, B. N., Wall, M.E., Wani, M.C., Taylor, H.L. J. Nat. Prod. 1985, 48, 952-956. 6. Castellanos, S. V. Tesis FES Zaragoza, UNAM, México, 1998. 124 pp. 7. Rizvi, S. H.; Kapil, R.S.; Shoe, A. J. Nat. Prod. 1985, 48, 146. 8. Iriarte, J.; Kincl, F.A.; Rosenkranz, G.; Sondheimer, F. J. Chem. Soc. 1956, 4170-4173. 9. Kincl, F.; Romo, J.; Rosenkranz, G.; Sondheimer, F. J. Chem. Soc., 1956, 4163-4169. 10. Enríquez, R. G.; Romero, M. L.; Escobar, L. I.; Joseph-Nathan, P.; Reynolds, W. F. J. Chrom., 1984, 287, 209-214. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 155-159 Investigación Preparación de materiales mesoporosos tipo Ti-MCM-41 y su uso en la apertura nucleófilica de epiclorhidrina con L-prolinol Deyanira Ángeles Beltrán,1 Ana Marisela Maubert Franco,1 Leticia Lomas,2 Victor Hugo Lara Corona,2 Jorge Cárdenas3 y Guillermo Negrón1 Área de Química Aplicada. Universidad Autónoma Metropolitana Azcapotzalco, México 02200, D.F. E-mail: [email protected] 2 Departamento de Química. Universidad Autónoma Metropolitana Iztapalapa, México 09340, D.F. 3 Instituto de Química, Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México D. F. 1 Recibido el 7 de abril del 2003; aceptado el 23 de junio del 2003 En homenaje al Dr. Alfonso Romo de Vivar Resumen. Materiales mesoporosos del tipo Ti-MCM-41 preparados por tratamiento hidrotérmico, se caracterizaron por DRX, SEM, FTIR, Si29-RMN y análisis textural, fueron utilizados en la reacción de apertura nucleofílica de epiclorhidrina con L-prolinol. Palabras clave: Ti-MCM-41, tratamiento térmico, β-aminoalcoholes. Abstract. Mesoporous materials Ti-MCM-41 were prepared by hydrothermal treatment then characterized by DRX, SEM, FT-IR, Si29-NMR and textural analysis to be used in a nucleophilic opening reaction of epichlorydrine with L-prolinol. Keywords: Ti-MCM-41, hydrothermal treatment, β-aminoalcohols. Las características del MCM-41 (Mobil Composition of Matter41) así como los demás miembros de la familia M41S que fueron sintetizados por primera vez en 1992 [1,2]. Pueden clasificarse como materiales ordenados o semicristalinos, intermedio en la clasificación de cristalinidad existente entre los geles porosos y los silicatos laminares [3]. Particularmente el MCM-41, posee canales unidimensionales en forma de panal de abeja y una composición química modificable, mediante la adición de cationes metálicos o variaciones en las condiciones de síntesis. Una de las características más importante del MCM-41 es el tamaño de sus poros, los cuales pueden variarse, modificando el surfactante que se utilice en su síntesis. En general los MCM-41, pueden usarse para la adsorción de moléculas orgánicas grandes, en separaciones cromatográficas, como anfitrión para confinar moléculas huésped y arreglos atómicos, así como también en catálisis de selectividad de forma [4]. La adición de aluminio al MCM-41 silícico, ha permitido otorgarle acidez al material, aunque siendo de naturaleza débil, tiene aplicación catalítica en reacciones de síntesis química debido a su gran área superficial [5]. Son variadas las técnicas de obtención del MCM-41, ya que su morfología, propiedades y uso son función de las condiciones experimentales de preparación [6]. La incorporación de titanio al MCM-41 convencional estudiada desde 1993, ha generado importante información sobre los efectos de dicho elemento en la red original del material, demostrándose que el titanio se incorpora tetrahédricamente en ésta, para constituir un catalizador útil en reacciones de oxidación selectiva de moléculas orgánicas de gran tamaño, hidroxilación de compuestos aromáticos [7], epoxidación de olefinas [8], oxidación de sulfuros a sulfóxidos [9], epoxidación de α-terpineol al epóxido correspondiente [10] y oxidación de aminas, esta última reacción con interés por parte de las industrias farmacéuticas [11]. La presencia de grupos Ti-OH en el Ti-MCM-41 permite la adsorción de epóxidos, haciendo posible la apertura del anillo oxiránico dando origen a dioles. Sobre estos antecedentes, se consideró pertinente que estos materiales podrían ser útiles en la obtención de pirrolidinas quirales, al usar como materia prima un epóxido racémico barato, tal como la epiclorhidrina y un nucleófilo quiral nitrogenado de cinco miembros como el L-prolinol, seguida de una separación diastereoisomérica de los productos pirrolidínicos. La obtención de pirrolidinas ópticamente activas son de gran importancia, pues, además de ser subunidades estructurales de substancias naturales y sintéticas con actividad biológica, también pueden ser utilizadas como catalizadores quirales en adiciones enantioselectivas de dietilzinc a aldehídos [12]. Resultados y discusión El patrón de difracción de rayos X del Ti-MCM-41 muestra tres picos .que corresponden a los planos 100, 110 y 200 de la estructura hexagonal del material característicos de los materiales MCM-41 [13]. El primer pico, que usualmente se encuentra en 2θ = 2.5, en el MCM-41 se observa desplazado a la izquierda, comportamiento que se explica por la presencia de titanio en la red del silicato mesoporoso. La distancia interplanar calculada es de 38.65 Å (Fig. 1). La gráfica de comparación de la función de distribución radial de Ti-MCM-41 y un patrón de sílice nos demuestran que el Titanio se encuentra dentro de la red (Fig. 2). 156 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Deyanira Ángeles Beltrán et al. Fig. 1. Tabla 1. Longitud ideal (Å) Longitud por FDR (Å) Átomos vecinos 1.7 2.55 3.10 3.90 4.05 4.90 5.10 8.30 9.20 Si-O O-O Si-Si Ti-Ti Si-Si-O-O Si-Ti Si-Si Ti-O Ti-O 1.62 2.65 3.24 4.10 5.10 Fig. 2. Tabla 2. Oxígeno Silicio Titanio 57.25 60.09 58.11 65.03 41.49 37.79 39.93 33.55 1.24 2.11 1.92 1.41 O Cl H N OH OH O O Cl OH N N OH OH N N O OH 1 Esquema 1 2 3 4 5 En la Tabla 1 se comparan los valores de longitud (Å) entre átomos vecinos de los elementos componentes de la muestra de Ti-MCM-41 (titanio, silicio y oxígeno) calculados por función de distribución radial, FDR y los valores teóricos de distancia entre vecinos de los mismos elementos en la muestra de referencia de sílice. Las distancias entre vecinos Si-O así como Si-Si, corresponden a las teóricas, por lo que no se aprecia alteración en la estructura de silicato del material mesoporoso con la adición de titanio a la red. Además los valores de proximidad entre átomos de Si y titanio son similares a los valores de referencia. Mediante el análisis semicuantitativo por microscopía electrónica de barrido, se calculó el porcentaje atómico de titanio, silicio y oxígeno presentes en la muestra en distintos puntos del analito. Se encontró una distribución promedio de titanio igual a 1.67 en el material como se ilustra en la Tabla 2. La muestra presenta un área superficial de 950 m2 / g, con una isoterma de adsorción-desorción de la forma IV de BET y posee un ciclo de histéresis indicativo de la condensación capilar, que ocurre frecuentemente en materiales clasificados como mesoporosos [14] (Fig. 3). La gráfica de distribución de tamaño de poro tiene tendencia unimodal con diámetro de poros de 39.5255 Å y 0.9458 cc / g (Fig. 4). Por análisis de RMN de Si29 de sólidos se asignaron las siguientes señales; una pequeña cercana a los –120 ppm asociada con las uniones de enlace Si-Ti-O y un hombro en –110 ppm debido a interacciones [15] del tipo Q4 como los Si(4Si), otros más cercanos a –100 ppm de silicios Q3 sobre sitios Si(OSi)3OH (Fig. 5). En el espectro de infrarrojo se observa un hombro en 957 cm–1, que es atribuido al titanio incorporado a la red de los silicatos mesoporos, la banda que aparece en 3746 cm–1 se relaciona con los grupos silanol y las demás entre 1200 y 1300 cm –1 son vibraciones de enlace entre átomos de silicio y oxígeno [16] (Fig. 6). Al hacer reaccionar a temperatura ambiente dos milimoles de epiclorhidrina (1) con dos milimoles de L-prolinol (2), en presencia de 400 mg de Ti-MCM-41 calcinado, se obtienen los productos 3, 4 y 5 con rendimientos del 84 %, 9 % y trazas, respectivamente. El primero es resultado de un ataque nucleofílico del L-prolinol (2) sobre el oxiránico y el segundo mediante una reacción intramolecular del producto 3 (Esquema 1). La caracterización de los productos, así como los rendimientos de la reacción fueron calculados de los crudos de reacción usando la cromatografía de gases acoplado a un detector de masas. En el espectro de masas del producto mayoritario del cromatograma con tiempo de retención de 12.69 minutos se observan los iones moleculares a m/z 194 [M+1]+ que es también el pico base; 222 [M+29]+ y 234 [M+41]+; todos estos iones con la contribución isotópica debida al cloro. También se observan los iones a m/z 176 [M+-OH]; 158 [M+Cl] y 114 [M + -CHOH-CH 2 Cl] que corresponden con la estructura (Espectro 1). Preparación de materiales mesoporosos tipo Ti-MCM-41 y su uso en la apertura nucleofílica de epiclorhidrina con L-prolinol 157 -110ppm Isoterma de adsorción-desorción de Nitrógeno muestra D19 800 Volumen adsorbido (cc/g) 700 600 -122ppm 500 400 300 200 100 0 0 0,2 0,4 0,6 0,8 1 Presión reducida, (P/Po) Fig. 3. Fig. 5. 100 98 96 1371.54 94 92 816.20 84 3746.45 80 1235.79 82 78 957.84 86 1626.76 88 1708.21 %Transmittance 90 76 74 UAMAZCAPOTZALCO *D19 72 70 3500 3000 2500 2000 1500 1000 500 Wavenumbers (cm-1) Fig. 4. Fig. 6. Ab u n d a n c e Abunda nce Sca n 1553 (12.699 min): 110.D 194 S c a n 7 1 0 ( 8 .0 2 0 m in ) : 1 1 0 .D 140 260000 3500000 240000 OH 3000000 N 2500000 O 158 220000 Cl 200000 N 180000 OH OH 160000 2000000 140000 158 120000 1500000 100000 176 1000000 80000 114 60000 84 222 500000 40000 234 70 0 60 84 80 102 100 140 126 120 140 160 180 206 200 220 250 262 240 260 186 70 20000 m/z--> 102 114 198 126 278 292 280 300 170 0 60 80 100 120 140 1 60 180 200 212 223 220 241 240 255 267 281 260 280 297 300 m /z- - > Espectro 1 Espectro 2 El pico del cromatograma con tiempo de retención 8.02 min., presenta un espectro de masas con los iones moleculares a m/z 158 [M+1]+; 186 [M+29]+ y 198 [M+41]+; el pico base se encuentra a m/z 140 [M-OH]+ (Espectro 2). El producto 5 podría resultar de la dimerización de 3 o 4 y la estructura que se propone se hace sobre la base de la existencia de los iones moleculares a m/z 315 [M+1]+, siendo también el pico base; 343 [M+29]+ y 356 [M+41]+. El tiempo de retención es de 17.72 min (Espectro 3). Parte experimental Para la preparación del material sólido catalítico se usó un reactor Parr 4243. La caracterización de los sólidos se hizo por difracción de rayos X de polvos, en un difractómetro Siemens con radiación monocromática de CuK en un rango 2θ = 0-10. Las determinaciones de adsorción-desorción de nitrógeno, se llevaron a cabo en un equipo Micrometrics ASAP 2000 a una temperatura de desgasificación de 400 °C. El análisis semicuantitativo elemental se realizó con un microscopio electrónico de barrido Leica-Zeiss LEO 440. Los 158 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Deyanira Ángeles Beltrán et al. Abunda nce S c a n 1 1 2 4 ( 1 7 .7 4 5 m in ) : 1 7 6 .D 50000 315 45000 40000 35000 OH 30000 O 299 N 25000 N 20000 O OH 15000 10000 343 5000 0 242 240 269 260 355 283 280 300 329 320 340 357 373 389 415 429 445 458 475 489 503 516 530 552565 360 380 400 420 440 460 480 500 520 540 m /z- - > Espectro 3 espectros de infrarrojo se realizaron en un instrumento Magna Nicolet Spectrometer modelo 750 por la técnica de reflectancia difusa de polvo. Los productos de reacción de apertura nucleofílica de la epiclorhidrina fueron analizados en un cromatografo de gases con detector de masas, en el modo de ionización química, HP 5890 Serie II Plus con columna HP 5 de 60 m × 0.25 µm × 0.25 mm con detector de masas HP modelo 5973. Preparación del Ti-MCM-41 A una solución acuosa de bromuro de cetiltrimetilamonio, agitanda vigorosamente, se agregó etilamina e hidróxido de tetrametilamonio al 10 %. Posteriormente se adicionó tetracloruro de titanio y tetraetilortosilicato. El gel obtenido (TEOS:0.2CTMABr:0.02Ti:0.6EA:0.20TMAOH:150H2O) se trató a 140 °C bajo presión autógena en un reactor de acero inoxidable con recubrimiento de teflón por 68 h. Una vez terminado el tratamiento hidrotérmico, se recuperó por filtración el sólido obtenido y se lavó con suficiente agua desionizada para eliminar el exceso de surfactante. Se secó a temperatura ambiente y se calcinó a 600 °C por 7 h en flujo moderado de aire cromatográfico, obteniéndose un sólido blanco. Reacción de apertura nucleofílica de la epiclorhidrina con L-prolinol (0.156 mL, 2 mmol) de epiclorhidrina y (0.196 mL), 2 mmol de (S)-(+)-pirrolidinametanol, se hicieron reaccionar en presencia de 400 mg de TiMCM-41 a temperatura ambiente, bajo atmósfera de nitrógeno, en 3 mL de acetonitrilo recién destilado, en un matraz de 50 mL durante 8 h, tiempo en el que se logra la máxima conversión El crudo de reacción se evaporó a sequedad bajo presión reducida y se analizó en un cromatógrafo de gases con detector de masas. Conclusiones Se prepararon materiales mesoporosos tipo Ti-MCM-41 usando la técnica de hidrólisis de sales de Titanio por tratamiento hidrotérmico. Mediante su caracterización se demostró la presencia de titanio incorporado a la red del silicato mesoporoso, sin que se vieran afectadas sus propiedades de superficie, es decir, se obtuvo un material con amplios poros de tamaño regular y gran área superficial con la incorporación de titanio en su estructura. Se discutió la dispersión del metal en el material mesoporoso y se comprobó su interacción con los átomos de silicio y oxígeno presentes en la red inicial. Se utilizó el catalizador preparado en la apertura nucleofílica de la epiclorhidrina, notándose la influencia positiva del uso del material Ti-MCM-41 en la formación de un β-aminoalcohol clorado 3 como producto principal de la reacción, el cual se transforma al epóxido 4 en el seno de la reacción. Tanto 3 como 4 pueden dar origen al sistema cíclico 5. Agradecimientos Al CONACyT por el apoyo financiero mediante el proyecto no. 33366-E. A la M. en C. María Isabel Chávez por los espectros de RMN de sólidos. Referencias 1. Beck, J.S.; Vartulli, J.C.; Roth, W.J.; Leonowicz, M.E.; Kresge,C.I.; Schimitt, K.D.; Chu, C. T. W.; Olson, D. H.; Sheppard, E. W.; McCullen, S. B.; Higgins J.B.; J. L. Schlenker. J. Am. Chem. Soc. 1992 114, 10834-10843. 2. Tuel. A. Microporous and Mesoporous Materials 1999, 27, 151169. 3. Corma, A.; Iglesias, M.; Sánchez, F. Catalysis Letters 1996, 39, 153. Preparación de materiales mesoporosos tipo Ti-MCM-41 y su uso en la apertura nucleofílica de epiclorhidrina con L-prolinol 4. Tuel, A. Mesoporous molecular sieves Studies in Surface Science and Catalysis 1998, 117, 159-170. 5. Zholobenko, V.L.; Plant, D.; Evans, A.J.; Holms, S.M. Microporous and Mesoporous Materials 2001, 44-45, 793-799. 6. Corma and D. Kumar. Mesoporous Molecular Sieves 1998. Studies in Surfaces Science and Catalysis 1998, 117, 201-222. 7. He, J.; Xu, W.; Evans, D.G.; Duan, X.; Li, C. Microporous and Mesoporous Materials 2001, 44-45, 581-586. 8. Laha, S.C.; Kumar, R. Microporous and Mesoporous Materials 2002, 53, 163-177. 9. Corma, A.; Domine, M.; Gaona, J.A.; Jordá, J.L.; Navarro, M.T.; Rey, F.; Pérez-Pariente, J.; Tsuji, J.; McCulloch, B.; Nemeth, L.T. Chem. Commun. 1998, 2211. 159 10. Blasco, T.; Corma, A.; Navarro, M.T.; Pérez, P. Journal of Catalysis 1995, 156, 65-74. 11. Berlini, C.; Ferraris, G.; Uiotti, M.; Moretti, G.; Psaro, R.; Ravasio, N. Microporous and Mesoporous Materials 2001, 4445, 595-602 12. Soai, K.; Konishi, T.; Shibata, T. Heterocycles, 1998, 51, 6. 13. Ahn, W.S.; Lee, D.E.; Kim, T.J.; Kim, J.H.; Seo, G; Ryoo, R. Applied Catalysis 1999, 181, 39-49. 14. Thieme, M.; Schüth, F. Microporous and Mesoporous Materials 1999, 27, 193-2001. 15. Alba M.D.; Zhaohua Luan, Z.; Klinowski, J. J. Phys. Chem. 1996, 100, 2178-2182. 16. Bhaumik, A.; Tatsumi, T. Journal of Catalysis 2000, 189, 31-39. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 160-166 Investigación New Eremophilanoids from the Roots of Psacalium radulifolium. Hypoglycemic, Antihyperglycemic and Anti-Oxidant Evaluations María Luisa Garduño-Ramírez1 and Guillermo Delgado2,* Centro de Investigaciones Químicas de la Universidad Autónoma del Estado de Morelos, Avenida Universidad 1001, Chamilpa 62210, Cuernavaca, Morelos, México. 2 Instituto de Química de la Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México, D.F. E-mail: [email protected] 1 Recibido el 21 de abril del 2003; aceptado el 23 de junio del 2003 Dedicated to Dr. Alfonso Romo de Vivar Abstract. The investigation of the chemical constituents from the roots of Psacalium radulifolium (Compositae), a member of the matarique complex of medicinal plants, resulted in the isolation of four additional new eremophilanoids: radulifolin D, radulifolin E (ketodecompostin), radulifolin F (3β-hydroxy-cacalone-3-O-β-D-glucopyranoside) and epi-radulifolin F (3β-hydroxy-6-epi-cacalone-3-Oβ-D-glucopyranoside), together with the known compounds maturinone, acetylmaturine, dimaturine, triacontanol, hydroxycacalolide, epi-hydroxycacalolide, β-sitosteryl-3-O-β-D-glucopyranoside, β-Dglucopyranose and saccharose. The methanol extract from the roots of this species displayed hypoglycemic activity, but cacalol, cacalone, epi-cacalone, O-methyl-1,2-dehydrocacalol and decompostin did not exhibit activity. The antihyperglycemic evaluation of the extract demontrated that it was inactive. Some isolated compounds were also tested for antioxidant activity, and cacalol was found to be active. Keywords: Psacalium radulifolium, Compositae, matarique, radulifolin D, radulifolin E, radulifolin F, epi-radulifolin F, eremophilanoids, hypoglycemic activity, antihyperglycemic activity, anti-oxidant activity. Resumen. La investigación de los constituyentes químicos de las raíces de Psacalium radulifolium (Compositae), una especie perteneciente al complejo matarique de plantas medicinales, resultó en el aislamiento de cuatro nuevos eremofilanoides: radulifolina D, radulifolin E (cetodecompostina), radulifolina F (3-O-β-D-glucopiranósido de 3β-hidroxicacalona) y epi-radulifolina F (3-O-β-D-glucopiranósido de 3β-hidroxi-6-epi-cacalona), junto con las substancias conocidas maturinona, acetil maturina, dimaturina, triacontanol, hidroxicacalólida, epi-hidroxicacalólida y 3-O-β-D-glucopiranósido de β-sitosterilo, β-D-glucopiranosa y sacarosa. El extracto metanólico de las raices de esta especie mostró actividad hipoglucémica, pero cacalol, cacalona, epi-cacalona, el éter metílico de 1,2-deshidrocacalol y la decompostina no mostraron actividad. La evaluación antihiperglucémica del extracto demostró su inactividad. La actividad anti-oxidante fue ensayada para algunas substancias, y se encontró que el cacalol es activo. Palabras clave: Psacalium radulifolium, Compositae, matarique, radulifolina D, radulifolina E, radulifolina F, epi-radulifolina F, eremofilanoides, actividad hipoglicémica, actividad antihiperglicémica, actividad anti-oxidante. Matarique is the common name for a group of plants used in Mexican traditional medicine for the treatment of diabetes, kidney pains, infections, and general body pains, among other ailments [1-4]. This group includes Psacalium decompositum, P. palmeri, P. peltatum, P. sinuatum, and A. thurberi. Psacalium belongs to the Tussilaginoid genera of the Senecioneae (Compositae), and includes ca. 40 species which are located chiefly in Mexico [5]. The structures and chemistry of the secondary metabolites isolated from P. decompositum (syn: Cacalia decomposita) have been the subject of several investigations [6], and the synthesis of the main constituents, cacalol (1), cacalone (2) and structural analogs have been achieved [7]. Cacalol (1) has been found as the bioactive constituent in anti-microbial [8], antioxidant [9], allelopatic and phytopathogenic assays [10]. The hypoglycemic activity of extracts of P. decompositum and P. peltatum in mice have been evaluated [11,12], and the antihyperglycemic activity of aqueous extracts and some constituents from P. decompositum using diabetic mice have been determined [13]. P. radulifolium is considered a substitute for the preferred P. decompositum in the matarique complex of medicinal plants, and previous examination of the less polar constituents of the roots of this species allowed the isolation of 1, 2, epicacalone (3), radulifolin A (4), epi-radulifolin A (5), radulifolin B (6), radulifolin C (7), O-methyl-1,2-dehydrocacalol (8), adenostin A (9), decompostin (10) and neoadenostylone (11), from which 1 displayed major antimicrobial activity [14]. Here we report the hypoglycemic and antihyperglycemic evaluations of the methanol extract of the roots of P. radulifolium, and the isolation of the polar constituents, which resulted in the characterization of four new metabolites: radulifolin D (12), radulifolin E (ketodecompostin, 13), radulifolin F (3β-hydroxy-cacalone-3-O-β-D-glucopyranoside, 14) and epi-radulifolin F (3β-hydroxy-6-epi-cacalone-3-O-β-D-glucopyranoside, 15), together with the known compounds maturinone (16), acetylmaturine (17), triacontanol, dimaturine (18), hydroxycacalolide (19), epi-hydroxy-cacalolide (20), βsitosteryl-β-D-glucopyranoside, β-D-glucopyranose and sac- New Eremophilanoids from the Roots of Psacalium radulifolium. Hypoglycemic, antihyperglycemic and anti-oxidant evaluations charose. The anti-oxidant activity of some isolated compounds is also reported. avoid interactions with the methyls at C-13 and C-15), the hydroxyl group at C-6 of radulifolin D could be tentatively proposed with the β- configuration (12), to explain the opposite specific rotation to that observed for radulifolin C (7). Radulifolin E (13) was isolated as a UV active solid (λmax 320, 280, 257 nm), with a molecular formula C17H18O5 established from EIMS and NMR data. The IR spectrum contained bands at 1740 and 1668 cm–1 consistent with the presence of an acetate and an α,β-unsaturated ketone, respectively. The 13C NMR spectrum (Table 2) showed 17 signals (four methyls, one methylene, four methines and eight quaternary carbons, including three carbonyls), and the chemical shifts and multiplicity observed in the 1H NMR spectrum (Table 1) could be accounted by the furanoeremophilane skeleton with an acetate at C-6 similar to that of decompostin (10). The major difference between radulifolin E (13) and decompostin was the downfield chemical shift for H-1, due to the presence of a ketone at C-2. Therefore, radulifolin E (13) was established as 2-ketodecompostin, previously obtained as a derivative of decompostin via bromination (NBS) and oxidation (AgNO3) [6d]. 1H and 13C NMR assignments (Tables 1 and 2) were confirmed by HMQC and HMBC experiments. The FABMS, 1H and 13C NMR data for compounds 14 and 15 were consistent with the molecular formula C21H28O9. An intense IR band at ca. 3400 cm–1 for both compounds suggested the presence of several hydroxyl groups, and bands at ca. 1660 cm –1 were ascribed to α,β-unsaturated carbonyl groups. 1H and 13C NMR spectra (Tables 1 and 2) indicated the presence of a β-D-glucopyranose fragment and a cacalone aglycon. The anomeric hydrogen of 14 was observed at δ 4.42 and the analysis of the COSY spectrum determined the sequen- Results and discussion Compound (12) was isolated as a yellow solid, and its HREIMS established the molecular formula C15H14O4. UV spectrum showed bands of a conjugated ketone at lmax 336, 277, 245, and 207 nm, and the IR spectrum revealed the presence of hydroxyl group (3583 cm–1), conjugated carbonyl (1664 cm–1) and multiple carbon-carbon bonds (1585, 1463 cm–1). The 1H NMR spectrum (Table 1) showed considerable similarity with that of radulifolin C (7), a compound previously isolated from this species [14], establishing a close structural relationship. The most significant difference was the downfield shift and multiplicity of H-14 (δ 1.85), which appeared as a singlet, in comparison with the chemical shift of the same protons of radulifolin C (δ 1.43), which resonated as a doublet, indicating the presence of a hydroxyl at C-6 in 12, in agreement with the molecular formula. 13C NMR data showed the expected chemical shifts, and the assignments were corroborated by HMQC and HMBC experiments. Therefore this substance was 6-hydroxy-radulifolin C, and named radulifolin D (12). The 3-O-methyl derivative of 12 has been characterized from a chemical analysis of Cacalia hastata L. var. tanakae [15], but it was considered as an artifact due to the lack of optical activity. Radulifolin D (12) is dextrorotatory, while radulifolin C (7) is levorotatory. Considering that the twisting of the A/C rings is in the opposite direction to the pseudo-axial methyl group at C-6 (to 1 O 4 15 OH O OH OH O 12 R1 R2 13 14 R 1 R2 O O OCH3 OCH3 O O O R1 R2 4 CH3 OH Radulifolin A 5 OH CH3 Epi-radulifolin A R2 R1 2 CH3 OH Cacalone 3 OH CH3 Epi-cacalone 1 Cacalol O O HO R1 R 2 OH R1 R2 7 CH3 H Radulifolin C 12 OH CH3 Radulifolin D 6 Radulifolin B OH 8 O-Methyl-1,2dehydrocacalol O OH O O O 9 Adenostin 161 OAc R 10 Ac Decompostin 11 Ang Neoadenostylone 162 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) María Luisa Garduño-Ramírez and Guillermo Delgado Table 1. 1H NMR (500 MHz) Chemical Shift Assignments for Compounds 12-15. H 1 2a 2b 3a 3b 4 6 12 13 14α 14β 15 OCOCH3 1’ 2’ 3’ 4’ 5’ 6’a 6’b -OH a Recorded 12a 13b 14c 15c 8.05 d (8) 6.95 d (8) 6.83 s 2.39 m 1.99 m 1.87 m 4.03 ddd (3,3,3) 2.38 dd (3) 2.0 m 1.79 m 4.03 ddd (3,3,3) 3.37 dq (7, 3) 3.02 qd (7,3) 7.52 q (1) 2.21 d (1) 1.63 s 7.50 q (1) 2.21 (1) 1.60 s 1.21 d (7) 1.22 d (7) 4.42 d (8) 3.01 dd (8, 8) 3.31 dd (9,8) 3.17 dd (9,9) 3.32 m 3.52 m 3.78 m 4.13 br s 4.39 br s 4.44 d (8) 3.07 dd (9,8) 3.37 dd (9,9) 3.29 dd (9,9) 3.31 m 3.65 m 3.84 m 4.18 br s 4.26 br s 4.51 br s 7.43 q (1.5) 2.31 d (1.5) 1.85 s 2.65 s 2.41 d (11) 2.38 dd (11, 5) 2.54 qd (7, 5) 6.43 s 7.53 q (1) 1.97 d (1) 1.31 s 1.08 d (7) 2.25 s in CDCl3 + DMSO. bRecorded in CDCl3. cRecorded in CD3COCD3. J values (in Hz). Table 2. 13C NMR Chemical Shifts Assignments for Compounds 12-15. Position C-1 C-2 C-3 C-4 C-5 C-6 C-7 C-8 C-9 C-10 C-11 C-12 C-13 C-14 C-15 OCOCH3 OCOCH3 C-1’ C-2’ C-3’ C-4’ C-5’ C-6’ a c 12a 126.45 (d) 114.30 (d) 161.03 (s) 124.28 (s) 148.14 (s) 71.216 (s) 142.82 (s) 144.345 (s) 172.99 (s) 123.65 (s) 120.90 (s) 145.08 (d) 8.956 (q) 27.46 (q) 13.44 (q) 13b 14c 15c 129.715 (d) 198.11 (s) 43.26 (t) 38.43 (d) 47.53 (s) 72.63 (d) 137.065 (s) 146.415 (s) 173.45 (s) 156.35 (d) 121.59 (d) 147.76 (s) 8.446 (q) 14.38 (q) 16.904 (q) 21.271 (q) 170.374 (s) 18.36 (t) 20.88 (t) 79.15 (d) 33.88 (d) 158.81 (s) 70.70 (s) 142.34 (s) 146.09 (s) 174.88 (s) 131.02 (s) 121.52 (s) 145.14 (d) 9.01 (q) 27.39 (q) 20.32 (q) 18.44 (t) 20.40 (t) 77.34 (d) 36.00 (d) 161.18 (s) 72.43 (s) 142.53 (s) 146.07 (s) 175.05 (s) 129.94 (s) 121.40 (s) 144.83 (d) 8.86 (q) 28.15 (q) 20.89 (q) 102.71 (d) 74.78 (d) 77.65 (d) 71.64 (d) 77.65 (d) 102.24 (d) 74.91 (d) 77.87 (d) 71.83 (d) 77.40 (d) 62.97 (t) 63.08 (t) Recorded at 125 MHz in CDCl3 + DMSO. b Recorded at 75 MHz in CDCl3. Recorded at 125 MHz in CD3COCD3 at 25 °C. Multiplicity (in parenthesis) deduced by DEPT. New Eremophilanoids from the Roots of Psacalium radulifolium. Hypoglycemic, antihyperglycemic and anti-oxidant evaluations O O O O HO HO OAc O OH O O R1 R2 OH R1 R2 14 CH3 OH Radulifolin F 15 OH CH3 Epi-radulifolin F 13 Radulifolin E O OCH3 O O HO O OAc CH=O 17 Acetylmaturine 16 Maturinone OCH3 OCH3 O O OH O O O CH=O 18 Dimaturine O R1 R2 R1 R2 19 CH3 OH Hydroxycacalolide 20 OH CH3 Epi-hydroxycacalolide tial vecinity for the carbinolic hydrogens of the β-D-glucopyranose (δ 3.78 – δ 3.01), which correlated with the corresponding signals at δ 74.78 (C-2’), 77.65 (C-3’), 71.64 (C-4’), 77.60 (C5’), 62.97 (C-6’) in the HMQC spectrum. The crosspeaks in the HMBC spectrum between H-3 (δ 4.03) and H-2’ (δ 3.01) with C-1’ (δ 102.71) confirmed that the glucopyranose was bound to C-3 of the modifoed eremophilane. NOESY interactions between H-1’ and H-3, and between H-3 and H-4, confirmed the stereochemical assignments. Similar crosspeaks were observed for 15, and therefore, the difference between the two substances was in the aglycon fragment. The stereochemistry at C-6 was deduced by comparing the 1H NMR data of 14 and 15 with those of 2 and 3 [7d]. Specifically, the downfield shift of H-4 in 14 (δ 3.37) with respect to that of H-4 in 15 (δ 3.18) was in agreement with the α- and β- orientation of the hydroxyl groups, respectively [16]. The methanol extract obtained directly from the roots of P. radulifolium was evaluated as hypoglucemic agent in normoglycemic rats, following the standard procedures [17], and the results are shown in Table 3. This residue showed significant decrease of blood glucose concentration (p < 0.05) at 7 h using several doses (30, 100 and 300 mg/kg), without returning to the basal blood glucose level. The effect of the hypoglycemic model drug is also included (glybenclamide, 10 mg/kg). Cacalol (1), The mixture of 2 + 3, O-methyl-1,2dehydrocacalol (8), and decompostin (10) did not display significant hypoglycemic activity at doses of 3.1, 10 and 31 mg/kg. The antihyperglycemic effect of the methanol extract was also tested at the same doses using male diabetic Wistar rats, 163 diabetized via streptozotocin injection, following the standard procedures. The results obtained indicated that this residue did noy display significant antihyperglycemic effect in comparison with the model drug. It has been proposed that the anti-oxidant activity may play a role in the antihyperglycemic and hipoglycemic activities [18], and some reactive oxygen species (superoxide anion, hydrogen peroxide and hydroxyl fee radical) are involved in the physiology of several diseases, including diabetes [19]. Therefore, the antioxidant activity of some isolated compounds (cacalol (1), the mixture of cacalone and epi cacalone (2 + 3), radulifolin C (7), neoadenostylone (11), radulifolin D (12) and radulifolin F (14)) was evaluated via the interaction with the stable free radical DPPH, following the procedures described in the experimental section. The results are included in Table 4, and they showed that only cacalol (1) exhibited significant activity, in agreement with previous reports [20]. It is interesting to note that some pure secondary metabolites did not display hypoglicemic effect, while the activity the methanolic extract of the roots of P. radulifolium was evident. Although the mechanism of action of the eremophilanes, modified eremophilanes or other natural products [21] in changing the blood glucose levels remains unknown, the described results suggest that there is some correlation with the ethnomedical use of the plant as antidiabetic agent. However, this plant, as all the plants used in traditional medicine, should not be used until safety studies are completed. Experimental Section General Experimental Procedures. Melting points are uncorrected. The 1H and 13C NMR spectra were recorded on a Varian Unity Plus-500 instrument, and the chemical shifts are expressed in parts per million (δ) relative to tetramethylsilane. Infrared spectra were recorded with a Nicolet Magna IR TM 750 and Perkin Elmer 283B instruments. MS data were recorded with a JEOL JMS-AX 505 HA mass spectrometer. Electron impact mass spectra were obtained at 70 eV ionization energy. Vacuum chromatography was performed on Merck Kieselgel 60 (0.040-0.863 mm). TLC analyses were performed on TLC plates with Si gel 60 F254 (Merck) or ALUGRAM® SIL G/UV254 silica gel plates. The compounds were detected by their absorbance under UV254 and UV366, or with a charring solution (12 g of ceric ammonium sulfate dihydrate, 22.2 mL of concentrated H2SO 4 and 350 g of ice). Solvents were distilled prior to use. Plant material. The roots of P. radulifolium (HBK) H. Rob. & Brettell were collected in San Luis Potosí in 1995. Plant material was identified by Dr. Robert Bye and M. Sc. Edelmira Linares from the Instituto de Biología de la UNAM. The voucher Bye & Linares 20028 was deposited in the Ethnobotanical Collection, and the voucher Bye & Linares 20149 was deposited in the National Herbarium, of the Instituto de Biología de la UNAM. 164 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) María Luisa Garduño-Ramírez and Guillermo Delgado Table 3. Hypoglycemic Effect (Variation of Percentage Values) of the Methanol Extract of the Roots of P. radulifolium (Oral administration). Time (h) Glybenclamidea 10 mg/kg Doses (extract) 30 mg/kg 100 mg/kg 300 mg/kg 1.5 3 5 7 –15.68 ± 3.66* –31.56 ± 7.75* –35.79 ± 6.74* –35.75 ± 4.33* –0.56 ± 5.87 –3.95 ± 6.52 –12.74 ± 2.97 –21.70 ± 1.77* –3.11 ± 3.91 0.16 ± 7.19 –10.10 ± 5.70 –12.92 ± 8.51 –0.39 ± 4.66 0.03 ± 4.75 –11.80 ± 3.43 –21.66 ± 3.09* a Glibenclamide was used as positive control. *Significative values (p < 0.05). The values represent the mean ± standard deviations from five independent experiments. The negative value (–) indicates a decrease in glucemia. Extraction and isolation. Previously we reported preparation of the n-hexane, CH2Cl2-EtOH (3:2) and MeOH extracts from the roots of this species, and the chemical analysis of the less polar residue [14]. The CH2Cl2-EtOH (3:2) extract (10 g) was loaded onto a column chromatography which was developed under reduced pressure using a gradient of n-hexane-EtOAc as elution system, to afford seven main fractions (named H to N). Fractions H and I were combined (31.2 mg), and this mixture was applied to a preparative TLC which was eluted with n-hexane-EtOAc (20:1), to yield maturinone [6b] (16, 5.6 mg), acetylmaturine [22] (17, 1.7 mg), and triacontanol (9 mg). Column rechromatography over silica gel of fraction J (350 mg) using n-hexane-EtOAc gradient as elution system afforded dimaturine [23] (18, 2.6 mg). Fractions K (15 mg) and L (1.3 g) were combined and subjected to Si-gel column chromatography using n-hexane-EtOAc (4:1) as elution system, affording decompostine [6d] (10, 896.5 mg). Fraction M (3.135 g) was chromatographed over Si gel using n-hexaneEtOAc gradient, to give several fractions. Some of these fractions were purified by column chromatography on Si gel using n-hexane-EtOAc gradient, leading to a mixture hydroxycacalolide [13a] (19) and epi-hydroxycacalolide [13a] (20, 3.3 mg), cacalone (2) and 6-epi-cacalone (3, 20 mg), radulifolin C [14] (7, 2.7 mg), radulifolin D (12, 7.1 mg), radulifolin E (ketodecompostine, 13, 2.2 mg). Column chromatography over Si gel of fraction N (2.56 g) using n-hexane-EtOAc as gradient elution system yielded β-sitosteryl 3-O-β-D-glucopyranoside, 14, 8.9 mg, radulifolin F (3-β-hydroxycacalone-3βO-D-glucopyranoside, 14, 13.2 mg) and epi-radulifolin F (3β-hydroxy-6-epi-cacalone-3β-O-D-glucopyranoside, 15, 26.3 mg) and β-D-glucopyranose. From the polar fractions of the metanolic extract (fractions Ñ to S) were identified O-methyl1,2-dehydrocacalol [6e] (8), cacalol (1), decompostin (10), βsitosterol, stigmasterol, cacalone (2) and 6-epi-cacalone (3), saccharose and β-D-glucopyranose. Radulifolin D (12). 7.1 mg, yellow solid mp 122-124 °C, Rf: 0.226 (7:3 hex-AcOEt), mp 122-124 ºC, [α]D = + 30.0 (c 0.05, MeOH); UV λmax (log ε) 207 (4.50), 245 (4.20), 277 (4.00), 336 (4.00); IR (CHCl3, cm–1): 3583, 3268, 2928, 2854, 1762, 1664, 1585, 1463, 1419, 1354, 1288, 1156, 1113, 996, 996, 918; 1 H and 13 C NMR data, see Tables 1 and 2; EIMS: C15H14O4, 258 [M+] (26), 243 (100), 240 (7), 215 (10), 201 (3), 187 (3), 85 (3), 157 (3), 135 (6), 128 (6), 115 (8), 109 (11), 91 (3), 77 (6), 55 (3), 43 (6). Radulifolin E (ketodecompostin, 13). 4.4 mg, yellow solid mp 222-225 °C (lit. [6d] 220-221 °C), Rf: 0.413 (3:2 hexAcOEt); Mp. 222-225 ºC; UV λ max (log ε): 319.5 (3.95); 280.5 (3.58); 257 (3.71); 240.5 (3.65); 232 (3.66); 224.5 (3.65); 205.5 (3.72).; IR (CHCl3, cm–1): 3037, 2971, 2940, 1747, 1672, 1606, 1531, 1463, 1415, 1372, 1315, 1176, 1050, 1030, 983, 928; 1H and 13C NMR data, see Tables 1 and 2; EIMS: C17H18O5, 302 [M+] (10), 274 (0.5), 260 (64), 242 (100), 227 (15), 214 (9), 199 (14), 191 (24), 163 (5), 161 (5), 137 (20), 123 810), 115 (9), 109 (8), 91 (7), 77 (7), 65 (4), 53 (8), 43 (32), 41 (5). Radulifolin F (3β-hydroxycacalone-3-O-β-D-glucopyranoside, 14). 13.2 mg, yellow oil; Rf: 0.295 (85:15 hexAcOEt); IR (CHCl3, cm–1): 3401, 2936, 1660, 1619, 1603, 1535, 1459, 1445, 1421, 1363, 1162, 1079, 1035, 923, 887; 1H and 13C NMR data, see Tables 1 and 2; FABMS+: C21H28O9, 447 [M+ + Na] (58), 407 (20), 263 (48), 245 (100), 227 (72), 191 (24), 154 (31), 136 (28), 91 (20), 77 (19), 44 (17). Epi-radulifolin F (3β-hydroxy-6-epi-cacalone-3-O-β-D-glucopyranoside, 15). 26.3 mg yellow oil; Rf: 0.295 (85:15 hexAcOEt); IR (KBr, cm–1): 3411, 2930, 1654, 1614, 1535, 1451, 1422, 1369, 1256, 1224, 1201, 1164, 1078, 1038, 936, 888, 814, 623, 595, 532; 1H and 13C NMR data, see Tables 1 and 2; Table 4. Percentage of Inhibition of the Free Radical (DPPH) by Compounds 1-3, 7, 11, 12 and 14. Compound 1 µM Concentrations 10 µM 100 µM 1 2+3 7 11 12 14 15.27 8.91 9.47 6.77 8.8 5.06 47.70 13.60 11.72 10.23 9.09 11.37 73.13 26.98 14.01 25.79 16.76 19.55 165 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) FABMS+: C21H28O9, 447 [M+ + Na] (100), 425 (17), 399 (8), 371 (38), 263 (49), 245 (44), 227 (26), 191 (21), 177 (35), 154 (59), 136 (51), 91 (26), 77 (25), 55 (24), 41 (25), 23 (58). Biological evaluations. The methanol extract used for biological assays was obtained by direct maceration of the dried roots of the plant at room temperature (1 L per each 100 g) by 48 h two times. Male Wistar normoglycaemic rats of 60-65 days old, generally weighing 200-250 g, were used. The animals were housed under standard laboratory conditions and maintained on standard pellet diet and water ad libitum. Rats were placed in single cages with wire-net floors and deprived of food for 18 h before experimentation but allowed free access to tap water throughout. All experiments were carried out using 5 animals per group. Male Wistar rats were made diabetic by an intraperitoneal injection of streptozotocin (60 mg/kg) in citrate buffer, pH 6.3 [24]. Extracts were suspended in 0.05 % of Tween 80 in saline solution. Glibenclamide (10 mg/kg) was used as a hypoglycemic model drug [25]. All extracts were prepared freshly immediately before the experimentation and administered by intragastrical route at 30, 100 and 300 mg/kg. Control rats received the vehicle (0.05 % Tween 80) in the same volume (0.5 mL/100 g) by the same route. Blood samples were collected from caudal vein by means of a little incision in the end of the tail at 0, 1.5, 3, 5, 7 and 9 h after drug administration. Blood glucose concentration was estimated by enzymatic glucose oxidase method using a commercial glucometer (One Touch Basic I, Jonhsons-Johnsons). The percentage variation of glycemia for each group was calculated with respect to initial (0 h) level according to: % variation of glycemia = [(Gt – Gi) / Gi ] × 100, Where Gi was initial glycemia values and Gt was the hypoglycemia value at +1.5, +3, +5, and +7 h, respectively [17b]. Statistical significance was estimated by analysis of variance (ANOVA) followed by Dunnett’s test t. p < 0.05 implies significance. Evaluation of antioxidant activity. The potential antioxidant activity of plant extracts and pure compounds was assessed on the basis of the scavenging activity of the stable 1,1-diphenyl2-picrylhydrazyl (DPPH) free radical [26]. Reaction mixtures containing test samples (dissolved in ethanol, at 1, 10 and 100 µM) and DPPH ethanolic solution (66.66 µM) in ambar vials (4 mL) were stirred for 30 min, and absorbances were measured at 515 nm. Percent of inhibition by sample treatment was determined by comparison with a control group [27]. Acknowledgements We thank Rocío Patiño, Beatriz Quiroz, María Isabel Chávez, Héctor Ríos, Luis Velasco, Javier Pérez-Flores, María Teresa Ramírez-Apan, and Antonio Nieto from the Instituto de Química de la UNAM for technical assistance; and Dr. An- María Luisa Garduño-Ramírez and Guillermo Delgado drés Navarrete, Facultad de Química de la UNAM, for the use of certain research facilities and guidance in some hypoglycemic evaluations assays. Financial support by grants from the DGAPA-UNAM and PROMEP-UAEMor is gratefully acknowledged. References 1. Linares, E.; Bye, R. A. J. Ethnopharmacol. 1987, 19, 153-186 2. Bye, R. A. Econ. Bot. 1986, 40, 103-124. 3. Pérez, R. M.; Ocegueda, G. A.; Muñoz, J. L.; Ávila, J. G.; Morrow, W., W. J. Ethnopharmacol. 1984, 12, 253-262. 4. Sullivan, G. Vet. Hum. Toxicol. 1981, 23, 6-7. 5. Barkley, T. M.; Clark, B. L.; Funston, M. Compositae: Systematics. Proceedings of the International Compositae Conference, Kew, 1994. Hind, D. N. J., Ed. Royal Botanic Gardens, Kew, 1996; Vol. 1., pp. 613-620. 6. (a) Romo, J.; Joseph-Nathan, P. Tetrahedron 1964, 20, 23312337. (b) Correa, J.; Romo, J. Tetrahedron 1966, 22, 685-691. (c) Joseph-Nathan, P.; Morales, J. J.; Romo, J. Tetrahedron 1966, 22, 301-307. (d) Rodríguez-Hahn, L.; Guzmán, A.; Romo, J. Tetrahedron 1968, 24, 477-483. (e) Romo, J.; Rodríguez-Hahn, L.; Manjarrez, A.; Rivera, E.; Bellido, J. Bol. Inst. Quím. Univ. Nac. Autón. Méx. 1968, 20, 19-29. (f) Brown, M.; Thompson, R. H. J. Chem. Soc. 1969, 1184-1186. (g) Ruiz, R. M.; Correa, J.; Maldonado, L. A. Bull. Soc. Chim. Fr. 1969, 3612-3614. (h) Romo, J. Bol. Inst. Quím. Univ. Nac. Autón. Méx. 1969, 21, 9296. (i) Kakisawa, H.; Inouye, Y.; Romo, J. Tetrahedron Lett. 1969, 1929-1932. (j) Samek, Z.; Harmatha, J.; Novotný, L.; Sorm, F. Coll. Czech. Chem. Comm. 1969, 34, 2792-2808. (k) Joseph-Nathan, P.; Negrete, M. C.; González, M. P. Phytochemistry 1970, 9, 1623-1628. 7. (a) Inouye, Y.; Uchida, Y.; Kakisawa, H. Chem. Lett. 1975, 1317-1318. (b) Yuste, F.; Walls, F. Aust. J. Chem. 1976, 29, 2333-2336. (c) Cásares, A.; Maldonado, L. A. Tetrahedron Lett. 1976, 2485-2488. (d) Yuste, F.; Díaz, E.; Walls, F.; Jankowski, K. J. Org. Chem. 1976, 41, 4103-4106. (e) Inouye, Y.; Uchida, Y.; Kakisawa, H. Bull. Chem. Soc. Jpn. 1977, 50, 961-966. (f) Huffman, J. W.; Pandian, R. J. Org. Chem. 1979, 44, 1851-1855. (g) Garofalo, A. W.; Litvak, J.; Wang, L.; Dubenko, L. G.; Cooper, R.; Bierer, D. E. J. Org. Chem. 1999, 64, 3369-3372. 8. Jiménez, M.; Cruz, R.; Valdés, J.; León, J.; Alarcón, G.; Sveshtarova, B. Rev. Latinoam. Quím. 1992, 23, 14-17. 9. Krasovaskaya, N. P.; Kulesh, N. I.; Denisenko, V.; Khim. Prirod. Soed. 1989, 643-645. Chem. Abst. 1990, 112, 95533u. 10. Anaya, A. L.; Hernández-Bautista, B. E.; Torres-Barragán, A.; León-Cantero, J.; Jiménez-Estrada, M. J. Chem. Ecol. 1996, 22, 393-406. 11. Román Ramos, R.; Lara, L. A.; Alarcón Aguilar, F. J.; Flores Sáenz, J. L. Arch. Med. Res. 1992, 23, 105-109. 12. Alarcón, F. J.; Román-Ramos, R.; Jiménez-Estrada, M.; ReyesChilpa, R.; González-Paredes, B.; Flores Sáenz, J. L. J. Ethnopharmacol. 1997, 55, 171-177. 13. (a) Inman, W. D.; Luo, J.; Jolad, S. D.; King, S. R.; Cooper, R. J. Nat. Prod. 1999, 62, 1088-1092. (b) Alarcón Aguilar, F. J.; Jiménez-Estrada, M.; Reyes Chilpa, R.; González Paredes, B.; Contreras Weber, C. C.; Román Ramos, R. J. Ethnopharmacol. 2000, 69, 207-215. 14. Garduño-Ramírez, M. L.; Trejo, A.; Navarro, V.; Bye, E.; Linares, E.; Delgado, G. J. Nat. Prod. 2001, 64, 432-435. 15. Naya, K.; Miyoshi, Y.; Mori, H.; Takai, K.; Nakanishi, M. Chem. Lett. 1976, 73-76. 16. This effect is evident using C6D5N as solvent, as described for the epimers 2 and 3 in reference 7d. New Eremophilanoids from the Roots of Psacalium radulifolium. Hypoglycemic, antihyperglycemic and anti-oxidant evaluations 17. (a) Navarrete, A. Evaluación Farmacológica de Plantas Medicinales. En: Plantas Medicinales de México. Introducción a su Estudio. Editor: E. Estrada. Universidad Autónoma de Chapingo. México. 1996. pp. 255-268. (b) Sharma, S. R.; Dwivedi, S. K.; Swarup, D. J. Ethnopharmacol. 1997, 58, 39-44. (c) Fernando, M. R.; Wickramasinghe, S. M. D. N.; Thabrew, M. I.; Karunanyaka, E. H. J. Ethnopharmacol. 1989, 27, 7-14. (c) Trajanoski, Z.; Brunner, G. A.; Gferer, R. J.; Wach, P.; Pieber, T. R. Diabetes Care 1996, 19, 1412-1415. 18. Becker, M.; Newman, S.; Ismail-Beigi, F. Mol. Cell. Endocrinology 1996, 121, 165-170. 19. (a) Hennani, T.; Parihar, M. S. Ind. J. Phys. Pharm. 1998, 42, 440.452. (b) Anderson, S. M. Drug Develop. Res. 1999, 46, 6779. (c) Sies, H. Oxidative Stress. Introductory Remarks. In: Sies, H., Ed. Academic Press. Orlando, Florida. 1985. pp. 1-8. 20. Lotina-Hennsen, B.; Roque Reséndiz, J. L.; Jiménez, M.; Aguilar, M. Z. Naturforsch. 1991, 46c, 777-780. 166 21. Meckes, M.; Garduño-Ramírez, M. L.; Marquina, S.; Álvarez, L. Rev. Soc. Quím. Méx. 2001, 45, 195-199. 22. Bohlmann, F.; Zdero, Ch.; Grenz, M. Chem. Ber. 1977, 110, 474486. 23. (a) Joseph-Nathan, P.; Negrete M. C.; González, M. P. Phytochemistry 1970, 9, 1623-1628. (b) Bohlmann, F.; Zdero, Ch. Chem. Ber. 1970, 111, 3140-3145. 24. Bwititi, P.; Musabayane, C. T.; Nhachi, C. F. B. J. Ethnopharmacol. 2000, 69, 247-252. 25. Nicki, I.; Nicks, J. L.; Ashcroft, S. J. H. Biochem. J. 1990, 268, 713-718. 26. Cotelle, V.; Vernier, J. L.; Catteau, J. P.; Pommery, J.; Wallet, J. C.; Gaydou, E. M. Free Radical Biol. Med. 1996, 20, 35-43. 27. (a) Waffo, T. P.; Fauconneau, B.; Deffieux, G.; Huguet, F.; Vercauteren, J.; Merillon, J. M. J. Nat. Prod. 1998, 61, 655-657. (b) Rekka, E. A.; Kourounakis, A. P. Res. Comm. Mol. Pathol. Pharmacol. 1996, 92, 361-366. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 167-172 Investigación Isolation and Chemical Transformations of Some Anti-inflammatory Triterpenes from Salvia mexicana L. var. minor Benth. Rosalba Argumedo Delira, Hortensia Parra-Delgado, Ma. Teresa Ramírez Apan, Antonio Nieto Camacho y Mariano Martínez-Vázquez* Instituto de Química, Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México, D. F. Tel: +(52) 56224403; E-mail: [email protected] Recibido el 6 de mayo del 2003; aceptado el 24 de junio del 2003 Dedicated to Dr. Alfonso Romo de Vivar Abstract. The acetone and methanol extracts of aerial parts of Salvia mexicana L. var. minor showed anti-inflammatory and antioxidant properties in the TPA y DPPH models respectively. The chromatography of these extracts led the isolation of β-sitosterol, betulinol, betulinic acid, ursolic acid and arbutin. The presence of these triterpenes is in agreement with previos phytochemical studies of Salvia, however this is the first time that arbutin is isolated from a species of this genus. On the other hand, since none or the isolated compound showed antioxidant properties in the DPPH model, it can be inferred that minor compounds not isolated or synergism effects could account for the antioxidant properties of the extracts. It is known that some pentacyclic triterpene derivatives with an α,β-unsaturated carbonyl in the ring A showed a better nitric oxide synthase inhibition activity that the natural triterpenes. It was decided to synthesize the methyl ester of 2-formyl-3-oxo-urs-28-oic and 2-formil-3-oxo-urs-1en-28-oic acids from ursolic acid and evaluate them using the DPPH and TPA models. The results showed that both compounds have antiinflammatory activity, but only the 2-formyl-3-oxo-ursol-28-oic acid methyl ester was active in DPPH assay, which is in agreement with the proposed mechanism of this test. This is the first chemical study of Salvia mexicana L. var. minor (Benth). Keywords: Triterpenos, Salvia mexicana var. minor, antiinflamatory activity, chemical transformations. Resumen. Los extractos acetónico y metanólico de las partes aéreas de Salvia mexicana L. var. minor, mostraron poseer propiedades antiinflamatorias y antioxidantes en los modelos de TPA y DPPH, respectivamente. La cromatografía de estos extractos permitió el aislamiento de β-sitosterol, betulinol, ácido betulínico, ácido ursólico y arbutina. La presencia de estos triterpenos está de acuerdo con estudios previos de Salvia, sin embargo, es la primera vez que se aísla arbutina de una especie de éste género. Posiblemente las propiedades antiinflamatorias de los extractos se deban a la presencia del β-sitosterol y ácido ursólico, compuestos con probadas actividades antiinflamatorias. Por otro lado, es probable que las propiedades antioxidantes de estos extractos, se deban a la presencia de compuestos minoritarios o a efectos sinérgicos, ya que ninguno de los compuestos aislados fue activo en el modelo de DPPH. Datos recientes en la literatura señalan que algunos derivados de triterpenos pentacíclicos con una cetona α,β-insaturada en el anillo A, presentan una mayor inhibición de la enzima óxido nítrico sintetasa que los triterpenos naturales, por lo que se decidió obtener los ésteres metílicos de los ácidos 2formil-3-oxo-urs-28-oico y 2-formil-3-oxo-urs-1-en-28-oico a partir del ácido ursólico, y evaluarlos en los modelos de DPPH y TPA. Los resultados muestran que ambos tienen propiedades antiinflamatorias, pero solo el éster metílico del ácido 2-formil-3-oxo-ursol-28-oico fue activo en el modelo de DPPH, resultado que está de acuerdo con el mecanismo asociado a esta prueba. Este es el primer estudio químico de la especie Salvia mexicana L. var. minor (Benth). Palabras clave: Triterpenos, Salvia mexicana var. minor, actividad anti-inflamatoria, transformaciones químicas. Introduction Most of the 500 species of Salvia found in Mexico, Central and South America belong to the Subgenus Jungia (formerly Calosphace) [10]. The species Salvia mexicana has been divide in two varieties: S. mexicana var. major Benth. and S. mexicana L. var. minor Benth. Acetone extract of the aerial parts from the former afforded narigenine and a cis-languidulane diterpenoid named salvimexicanolide [10]. In addition, from the chloroform extract of the aerial parts of this species β-sitosterol, betulinic acid and a triterpenic lactone called salviolide were isolated [11]. As a part of our ongoing systematic studies looking for bioactive compounds from Mexican species [12], we report in this paper the chemical study, the free radical scavenging and the anti-inflammatory activities of some extracts and isolates from S. mexicana L. Salvia is an important genus consisting of ca 900 species in the family Lamiaceae (formerly Labiatae). Some species of Salvia have been cultivated worldwide to be used in folk medicine and for culinary purposes [1]. The dried leaves of S. officinalis (sage) L., for example, is well known for their antioxidative properties used in the food processing industry but applicable also to the area of human health [2]. Studies on the chemical constituents of Salvia have been mainly confined to the diterpenoids and the tanshinones [3,4], and several reviews of these components have already been published [5, 6]. In addition, there are several reports on the biological activities of some species of this genus [7-9]. 168 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Rosalba Argumedo Delira et al. R 1 HO 2 R = CH2OH 5 R = COOH HO OH COOR H OH H O HO O HO HO H 3 R=H 6 R = Me H OH 4 H O COOMe COOMe R H O O 7 R=H 8 R= 9 OH H Scheme 1 var. minor. To our knowledge this is the first report on the chemical constituents as well as the free radical scavenging and anti-inflammatory activities of this species. Results and Discussion Flowers and leaves of this species were studied separately. The hexane, acetone and methanol extracts of each limb were obtained. The method of DPPH free radical can be used to evaluate the antioxidant activity of specific compounds or extracts in a short time. It is based on the transformation of the stable free radical 1,1-diphenyl-2-picryl hydrazyl (DPPH) to α,αdiphenyl-β-picryl hydrazine by means of putative antioxidant compounds [13]. On the other hand, the TPA-induced edema test is a screening method to evaluate the ability of test compounds or extracts to prevent an inflammatory reaction in response to the edemogen. The values of the anti-oxidative evaluation, by DPPH method of some extracts of S. mexicana var. minor are shown in Table 1. Those of the anti-inflammatory evaluation assessed by TPA-induced edema in mice are shown in Table 2. According to these results, the acetone and methanol extracts from flowers, as well as the methanol from leaves were active in the DPPH assay. A different pattern is observed in the TPA assay where only the hexane and acetone extracts from the flowers were active. In order to isolate the possibly involved components, all the active extracts were chromatographed. Then, from the flower hexane extract, β-sitosterol (1; 157 mg; 0.034 %) and betulinol (2; 10 mg; 0.002 %) were isolated, while ursolic acid (3; 3.612 g; 0.8 %) and arbutin (4; 563 mg; 0.12 %) were iso- lated from the flower acetone extract. Arbutin (4) was the only compound isolated from the flower and leave methanol extracts, 3.180 g (0.69 %) and 5.103 g (0.96 %) respectively. The presence of 1, 2 and 3 in S. mexicana var. minor are in agreement with previous phytochemical reports of this genus. To our knowledge this is the first time that arbutin is isolated from species of Salvia genus. All the isolated compounds were inactive in DPPH assay, thus indicating that activity of the extracts is due to the minor constituents not isolated or to a synergic effect. These results are in agreement with the assumed mechanism of this reaction, which postulate that the free radical scavenging activity of a compound in DPPH assay is attributed to their hydrogen donating ability [14]. On the other hand, the anti-inflammatory activities of the β-sitosterol (1) and ursolic acid (3) are well documented [15, 16] then the presence of 1 and 3 in this species could account for its anti-inflammatory activity (Table 2). It is known that phorbol esters, such as TPA, induce skin inflammation and a hyperproliferative response with an infiltration of neutrofils [17]. It is also known that TPA stimulates PLA2, and that consequently a release of arachidonic acid and prostaglandins occurs [18]. Although the mechanism by which TPA causes inflammation is not completely clear, it seems to be related in part to the release of eicosanoid mediators. Then inhibitors of cyclooxygenase and lipoxygenase, as ursolic acid, have proven activity in the TPA model [19, 20]. The high output of nitric oxide (NO) produced by inducible nitric oxide synthase (iNOS), which is expressed in activated macrophages, plays an important role in host defense. However, excessive production of NO can also destroy functional normal tissues during acute and chronic inflammation. From a structure-activity study between 80 ursolic and oleanolic derivatives, the 2-cyano-3,12-dioxooleana-1,9-dien-28-oic, is Isolation and Chemical Transformations of Some Anti-inflammatory Triterpenes from Salvia mexicana... Table 1. Free radical scavenging activities of some extracts of S. mexicana L. var. minor. Extracts Concentration (ppm) Reduction of DPPH (%) 10 100 1000 10 100 1000 10 100 1000 7.17 12.64* 84.47* 95.35* 17.74* 92.86* 91.55* 13.55* 90.76* 92.52* 94.69* Acetone (flowers) Methanol (flowers) Methanol (leaves) Nordihydroguaiaretic acid (positive control) The results were analyzed by ANOVA. Statistical comparison were made between control group and the experimental groups using a Dunnet’s test. *p < 0.05. the product with the highest inhibitory activity against production of nitric oxide (NO) induced by interferon γ (IFN-γ) in mouse macrophages. In general, it was found that oleanolic and ursolic derivatives with a 1-en-3-one functionality in ring A have significant inhibitory activity against production of NO. Also it is known that ursolic acid up regulate iNOS and TNF-α expression through NF-κB transactivation in the resting macrophages [21]. Taking this information into account, it was decided to evaluate the free radical scavenging as well as the anti-inflammatory properties of both 8 and 9. Compounds 8 and 9 were synthesized from 3, according to the route illustrated in Scheme 1. The results showed that only 8 was active as free radical scavenger (Table 3). However, both 8 and 9 showed almost the same activity as anti-inflammatory agents as ursolic acid (Table 4). These findings clearly indicate that the free radical scavenger activity of 8 is due to its hydrogen donating ability. On the other hand, in contrast to their inhibitory activity against the production of nitric oxide, the presence of unsaturated moieties in 8 and 9 are not relevant in terms of their antiinflammatory activity, since both of them showed almost the same activity as ursolic acid. Materials and Methods General. The melting points (uncorrected) were determined on a Fisher-Johns apparatus. IR spectra were recorded as KBr pellets or liquid film on a Nicolet spectrophotometer model Magna 750. Mass spectra were recorded at 70 eV on a Jeol JMS-AX505HA mass spectrometer. NMR spectra were measured using Varian-Gemini 200 and Varian VXR-300 (1H, 200 or 300 MHz, 13C, 75 MHz) spectrometers in CDCl3 or DMSO-d6 with TMS as internal standard. Plant material. Aerial parts of S. mexicana L. var. Minor Benth. were obtained from an orchard localized in Xahuen 169 street in San Miguel Tlaixpan (Texcoco, Edo. de Mexico, Mexico) in 2002. A voucher specimen was deposited in the Herbario Nacional (MEXU-1054424). Flowers (456 g) and leaves (531 g) were separately treated. Then, plant material was exhaustively extracted with nhexane, acetone and MeOH, successively. From the flowers, 24.62 g (5.39 %, dry weight) of the hexane extract, 24.47 g (5.36 %, dry weight) of acetone extract and 73.26 g (16.06 %, dry weight) of methanol extract were obtained, while from the leaves, 13.19 g (2.48 %, dry weight), 29.33 g (5.52 %, dry weight) and 179.55 g (33.77 %, dry weight) were obtained respectively. All the extracts were cromatographed using an open column packed with Si-gel (G- Altech, 0.2-0.5 mm, ASTM) in a 1:30 proportion to the extract and eluted with solvent mixtures of increasing polarity starting with hexane and ending with methanol. Chromatography of hexane extract of flowers. From the hexane extract of the flowers a total of 65 fractions of 200 mL each, were collected. Fractions showing similar TLC data were combined, affording eight pools (F1-F8): F3 (fractions 24-26, eluted with hexane-EtOAc, 9:1), F4 (fractions 27-35, eluted with hexane-EtOAc, 8:2). β-sitosterol (1, 157 mg) was isolated from F3 and betulinol (2; 10 mg) from F4. Chromatography of acetone extract of flowers. From the acetone extract of the flowers a total of 265 fractions of 200 mL each were collected. Fractions showing similar TLC data were combined, affording six pools (F1-F6): F3 (fractions 36161, eluted with hexane-EtOAc, 7:3), F4 (fractions 162-237, eluted with hexane-EtOAc, 1:1). Ursolic acid (3; 3.682 g) was isolated from F3 and arbutin (4; 563 mg) from F4. Chromatography of MeOH extract of flowers. From the MeOH extract of the flowers a total of 111 fractions of 200 mL each were collected. Fractions showing similar TLC data were combined, affording nine pools (F1-F9): F5 (fractions 50-77, eluted with EtOAc), F6 (fractions 78-89, eluted with EtOAc-MeOH, 9:1), F7 (fractions 90-93, eluted with EtOAcMeOH, 7:3), F8 (fractions 94-102, eluted with EtOAc-MeOH, 1:1) and F9 (fractions 103-111, eluted with MeOH). Arbutin (4; 3.180 g) was isolated from F5-F8 pools. Chromatography of hexane extract of leaves. When the extract was concentrated, a yellowish solid precipitate (235 mg), which was filtered and chromatographed. A total of 16 fractions of 50 mL each were collected. Fractions showing similar TLC data were combined, affording three pools (F1F3): F1 (fractions 1-4, eluted with hexane), F2 (fractions 5-8, eluted with hexane-EtOAc, 9:1) and F3 (fractions 9-16, eluted with hexane-EtOAc, 8:2). Betulinic acid (5; 84 mg; 36%) was isolated from F3. The remanent extract (13.19 g) afforded a total of 22 fractions of 200 ml each. Fractions showing similar TLC data were combined, affording six pools (F1-F6): F3 (fractions 7-12, eluted with hexane-EtOAc, 7:3), F4 (fractions 170 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Rosalba Argumedo Delira et al. Table 2. Anti-inflammatory activities of some extracts of S. mexicana L. var. minor. Extracts Edema Inhibition (%) (mg, average SE) Control (methanol) Hexane (flowers) Acetone (flowers) Methanol (flowers) Methanol (leaves) Indomethacin (positive control) 15.47 ± 0.32 4.77 ± 0.50 5.60 ± 0.91 13.70 ± 0.31 13.10 ± 1.07 — 69.17* 63.79* 11.42 15.0 1.07 ± 0.03 91.35* All the extracts were tested at 1 mg / ear doses. The results were analyzed by ANOVA. Statistical comparison were made between control group and the experimental groups using a t student test. *p < 0.01 13-14, eluted with hexane-EtOAc, 1:1), F5 (fractions 15-19, eluted with hexane-EtOAc, 3:7) and F6 (fractions 20-22, eluted with EtOAc). β-sitosterol (1; 686 mg) was isolated from F3-F6 pools. Chromatography of acetone extract of leaves. From the acetone extract of the leaves a total of 33 fractions of 200 mL each, were collected. Fractions showing similar TLC data were combined, affording nine pools (F1-F9): F3 (fractions 810, eluted with hexane-EtOAc, 7:3), F4 (fractions 11-15, eluted with hexane-EtOAc, 1:1), F5 (fractions 16-18, eluted with hexane-EtOAc, 3:7), F6 (fractions 19-21, eluted with EtOAc), F7 (fractions 22-25, eluted with EtOAc-MeOH, 7:3), F8 (fractions 26-29, eluted with EtOAc-MeOH, 1:1) and F9 (fractions 30-33, eluted with MeOH). Ursolic acid (3; 4.210 g; 14.35%) was isolated from F3-F5 pools and arbutin (4; 2.103 g; 7.168%) was isolated from F5-F8 pools. Chromatography of MeOH extract of leaves. The MeOH extract was partitioned between n-butanol and CH2Cl2. It afforded the n-butanol extract (46.16 g), which was chromatographed yielding a total of 22 fractions of 200 mL each. Fractions showing similar TLC data were combined, affording nine pools (F1-F9): F6 (fractions 9-13, eluted with EtOAcMeOH, 9:1), F7 (fractions 14-15, eluted with EtOAc-MeOH, 7:3), F8 (fractions 16-18, eluted with EtOAc-MeOH, 1:1) and F9 (fractions 19-22, eluted with MeOH). Arbutin (4; 5.103 g, 4.3 %) was isolated from F6-F9 pools. Ursolic acid methyl ester (6). A solution of ursolic acid (1 g, 2.19 mmol) in a mixture of ether / MeOH (50 mL) was cooled down to 0 °C in an ice-bath. Ethereal diazomethane was added until permanent yellow color was obtained. After 24 h the solvent was removed by distillation under low pressure to give ursol-28-oic acid methyl ester (6; 758 mg; 1.61 mmol; 73.5 % yield). 3-oxo-urs-28-oic acid methyl ester (7). To a solution of 6 (750 mg, 1.6 mmol) in acetone (10 ml) was added an excess of Jones's reagent at 0 °C while stirring. The reaction course was followed by TLC. After 55 min, the excess of Jones's was destroyed by addition of MeOH, and then the reaction mixture was diluted with H 2O (30 mL). Extraction with CH2Cl2 (4 × 10 mL), drying (Na2SO4), filtration and evaporation of the solvent gave a residue, which by crystallization from hexane-EtOAc afforded 7 (321 mg; 0.7 mmol; 43.7 % yield). Mp 182-184°C, IR (KBr) ν max: 2935, 2867, 1726, 1695, 1459, 1380 and 1142 cm–1. EIMS 70eV m/z: 468 (M+, C31H48O3), 453, 419, 407, 262, 249, 203 and 189. 1H NMR 200 MHz CDCl3 δ: 5.27 (1H, m, H-12), 3.61 (3H, s, OMe), 2.60 (1H, d, H-18), 2.25 (2H, m, H-2), 1.08 (3H, s), 1.04 (6H, s), 1.06 (3H, d, J=8 Hz), 0.95 (3H, s), 0.85 (3H, d, J=7Hz), 0.79 (3H, s). 2- Formyl-3-oxo-urs-28-oic acid methyl ester (8). To a solution of 0.7 mmol of 7 in 7 mL of dry pyridine, held under nitrogen, was added 1.5 mL (18.7 mmol) of ethyl formate (distilled from phosphorus pentoxide) followed by 1 mL of a solution of 294 mg (13.3 mmol) of sodium in 6 mL of absolute methyl alcohol. The resulting solution was then kept at room temperature under nitrogen overnight. The reaction was evidenced by the appearance of a deep color and / or the formation of an insoluble precipitate. The mixture was poured into a cold solution of 16 mL of glacial acetic acid in 150 mL of water, and the resulting precipitate was extracted with CH2Cl2. The organic layer was washed with water and then extracted with 3 × 100 mL of 2 % potassium hydroxide solution. The combined basic extract were washed with ether and acidified with 10 mL of glacial acetic acid. Extraction of the aqueous layer with CH2Cl2 in the usual manner, afforded 2formyl-3-oxo-urs-28-oic methyl ester (8; 226 mg; 0.46 mmol; 70 % yield). Reddish viscous liq. IR (CHCl 3) ν max: 2925, 2869, 1725, 1636, 1587, 1455, 1360 and 1147 cm–1. EIMS 70eV m/z: 496 (M+, C32H48O4), 481, 478, 437, 421, 262, 249, 233, 203 and 189. 1H NMR 200 MHz CDCl3 δ: 14.91 (1H, s, OH chelated), 8.57 (1H, s, H-23), 5.29 (1H, m, H-12), 3.61 (3H, s, OMe), 2.32 (1H, d, H-18), 1.25, 1.19, 1.11, 1.09, 0.8 (3H, s, each), 0.93 (3H, d, J=8 Hz), 0.86 (3H, d, J=7Hz). Table 3. Free radical scavenging activities of 8 and 9. Compound 2- Formyl-3-oxours-28-oic acid methyl ester (8) 2- Formyl-3-oxours-1-en-28-oic acid methyl ester (9) Concentration (ppm) Reduction of DPPH (%) 10 100 1000 10 100 1000 12.69* 40.87* 79.43* N. A. N. A. N. A. The results were analyzed by ANOVA. Statistical comparison were made between control group and the experimental groups using a Dunnet’s test. *p < 0.05, N. A. = Not active. Isolation and Chemical Transformations of Some Anti-inflammatory Triterpenes from Salvia mexicana... Table 4. Anti-inflammatory activities of 8 and 9. Compound Control (EtOAc) 2- Formyl-3-oxours-28-oic acid methyl ester (8) 2- Formyl-3-oxours-1-en-28-oic acid methyl ester (9) Edema (mg, average SE) Inhibition (%) 11.80 ± 0.045 3.03 ± 0.86 74.29* 3.03 ± 0.86 74.29* All the compounds were tested at 1 mg / ear doses. The results were analyzed by ANOVA. Statistical comparison were made between control group and the experimental groups using a t student test. *p ≤ 0.01. The reported % of inhibition of ursolic acid is 74.4 % at 1 mg/ear doses [16]. 2- Formyl-3-oxo-urs-1-en-28-oic acid methyl ester (9). PhSeCl (120 mg) was dissolved in 12 mL of CH2Cl2 and cooled to 0 °C and 0.06 g (40 µl) of pyridine was added. After 15 min, 0.2 g of 8 in 3 mL of CH2Cl2 was added and the mixture was stirred for 15 min more. The CH2Cl2 solution was extracted with two 5 mL portions of 10 % HCl and cooled back to 0 °C, at which time 0.1 mL of 30 % H2O2 was added. An additional 0.1 mL of 30 % H2O2 was added after 10 min and again after 20 min. After an additional 10 min, 0.5 mL of H 2O was added and the CH 2Cl 2 layer was separated and washed with 5 mL of saturated NaHCO3. After being dried over Na2SO4, the solution was filtered and the solvent evaporated under vacuum to yield 9 (53 mg; 0.11 mmol; 24 % yield). Reddish viscous liq. IR (CHCl3) ν (cm–1): 2921, 2858, 2721, 1719, 1672, 1604, 1455, 1379, 1224 and 1110. EIMS 70eV m/z: 494 (M+, C32H46O4), 479, 476, 435, 419, 314, 262, 249, 233, 203, 189, 158, 133 and 117. 1H NMR (200 MHz, CDCl3): 10.01 (1H, s, COH), 7.80 (1H, s, H-1), 5.38 (1H, m, H-12), 3.62 (3H, s, OMe), 2.28 (1H, d, H-18), 1.25, 1.18, 1.09, 0.96, 0.87 (3H, s, each), 1.17 (3H, d, J = 7 Hz), 0.87 (3H, d, J = 8Hz). Free radical scavenging Activity. DPPH assay was performed essentially according to the modified method of Cottele. Reaction mixture containing different concentrations of test samples in DMSO and 100 mM DPPH ethanol solution in 96-well microliter plates, were incubated at 37 °C for 30 min, and subsequently the absorbencies were measured at 515 nm in a microplate reader Elx 808. Measurements were performed in triplicate in at least three independent experiments. The % inhibition of each compound was determined by comparison with a DPPH ethanol blank solution [22]. The results were analyzed by ANOVA. Statistical comparisons were made between control group and the experimental groups using Dunnet’s test. Animals. Male CD-1 mice, weighing 20-25 g each were used. Instituto de Fisiología Celular, Universidad Nacional Autónoma de México provided the experimental animals. All animals were held under standard laboratory conditions in the animal 171 house (temperature 27 ± 1 °C). They were fed laboratory diet and water ad libitum. All experiments were carried out using 4-8 animals per group. TPA-induced edema model. Effects of the test substances on TPA-induced ear edema in mice were studied as described by De Young [17] with slight modifications. The substances (1 mg / ear) were applied topically. A solution of TPA (2.5 µg) in EtOH (10 µL) was applied topically to both faces (5 µL each face) of the right ear of the mice, 10 min after the test substances were applied (10 µL each face). The left ear received ethanol (10 µL) first, and 20 µL of the respective solvent subsequently. Four hours later the mice were killed by cervical dislocation. A 7-mm diameter plug was removed from each ear. The swelling was assessed as the difference in weight between right and left ear plugs [19]. Inhibition of edema (EI, %) was calculated by the equation: EI (%) = 100 – [B × 100 / A], with A = edema induced by TPA alone, and B = edema induced by TPA plus sample. Data were expressed as the mean SEM of 4-8 mice. All the extracts and compounds were tested at 1 mg / ear doses. The results were analyzed by ANOVA. Statistical comparisons were made between control group and the experimental groups using a t student test. *p < 0.01. References 1. Lu, Y.; Yeap-Foo, L. Phytochemistry 2002, 59,117-140. 2. Baricevic, D.; Sosa, S.; Della Logia, R.; Tubazo, A.; Simonovska, B.; Krasna, A.; Zumpancic, A. J. Ethnopharmacol. 2001, 75, 125-132. 3. Chang, H. M.; Cheng, K. P.; Choang, T. F.; Chow, H. F.; Chui, K. Y.; Hon, P. M.; Lau Tan, F. W.; Yang, Y.; Zhong, Z. P.; Lee, C. M.; Sham, H. L.; Chan, C. F.; Cui, Y. X. J. Org. Chem. 1990, 55, 3537-3543. 4. Zhang, K. Q.; Bao, Y.; Wu, P.; Rosen, R. T.; HO, C. T. J. Agric. Food Chem. 1990, 38, 1194-1197. 5. Tang, W.; Eisenbrand, G. Chinese Drugs of Plan Origin. Chemistry, Pharmacology, and use in Traditional and Modern Medicine. Springer-Verlag, Berlin, Heidelberg, 1992, pp. 891-902. 6. Rodríguez-Hahn, L.; Esquivel, B.; Cárdenas, J. Recent Advances in Phytochemistry, 1995, 29, 311-332. 7. Takahashi, K.; Ouyang, X.; Komatsu, K.; Nakamura, N.; Hattori, M.; Baba, A.; Azuma, J. Biochem. Pharmacol., 2002, 64, 745750. 8. Cao, C-M.; Xia, Q.; Zhang, X.; Xu, W-H.; Jiang, H-D.; Chen, JZ. Life Sciences, 2003, 72, 2451-2463. 9. Ebringerova, A.; Kardosova, A.; Hromádkova, Z.; Híbalova, V. Fitoterapia, 2003, 74, 52-61. 10. Esquivel, B.; Ramírez-Dávalos, N.; Espinosa-Pérez, G. Heterocycles, 1999, 51, 1647-1651. 11. Collera, O.; Gomora, E.; García Jiménez, F. Rev. Latinoamer. Quím. 1980, 11, 60-62. 12. Estrada-Reyes, R.; Álvarez, C.; López-Rubalcava, C.; Rocha, L.; Heinze, G.; Moreno, J.; Martínez-Vázquez, M. Z. Naturforsch. 2002, 57c, 29-32. 13. Tamil Selvi, A.; Joseph, G. S.; Jayaprakasha, G. K. Food Microbiology 2003, 20, 455-460. 14. Shimada, K. K.; Fujikawa, K. Y.; Nakumara, T. J. Agric. Food Chem. 1992, 40, 945-948. Isolation and Chemical Transformations of Some Anti-inflammatory Triterpenes from Salvia mexicana... 15. Martínez-Vázquez, M.; Ramírez A., T. O.; Lazcano, M. E.; Bye, R. Rev. Soc. Quím. Mex. 1999, 43, 103-105. 16. Recio, M. C.; Giner, R.; Terencio, M. C.; Sanz, M. J.; Ríos, J. L. Planta Med. 1991, 57, A56-57. 17. De Young, L. M.; Kheifets, J. B.; Ballaron, S. L.; Young, J. M. Agents Actions. 1989, 26, 335-341. 18. Fuerstenberg, G.; Richter, H.; Fusening, N. E.; Marks, F. Cancer Lett. 1981, 11, 191-204. 19. Carlson, R. P.; O´Neil-Davis, L.; Chang, J.; Lewis, A. J. Agents Actions. 1985, 17, 197-204. 172 20. Najid, A.; Simon, A.; Cook, J.; Chable-Rabinovitch, H.; Delage, C.; Chulia, A. J.; Rigaud, M. FEBS. 1992, 213-217. 21. You, H. J.; Chul, Y. C.; Ji, Y. K.; Park, S. J. Hahm, K.; Jeong, H. G. FEBS Lett. 2001, 509, 156-160. 22. Cottele, N.; Bernier, J. L.; Catteau, J. P.; Pommery, P.; Wallet, J. C.; Gaydou, E. M. Free Radical Biology & Medicine. 1996, 20, 35-43. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 173-177 Investigación Two New Oleanolic Acid Saponins from the Roots of Viguiera hypargyrea Laura Alvarez,1* Alejandro Zamilpa,2 Silvia Marquina,1 and Manasés González1 Centro de Investigaciones Químicas, Universidad Autónoma del Estado de Morelos, Av. Universidad 1001, Chamilpa, 62210, Cuernavaca, Morelos, México. Tel/Fax: (+52) (01 77) 7329 7997; E-mail: [email protected] 2 Centro de Investigación Biomédica del Sur, Instituto Mexicano del Seguro Social, Argentina No. 1, Centro, 62790 Xochitepec, Morelos, México 1 Recibido el 20 de mayo del 2003; aceptado el 24 de junio del 2003 Dedicated to Professor Alfonso Romo de Vivar Abstract. Two new triterpene saponins whose aglycons are based on the oleanane skeleton (1-2), were isolated from the roots of Viguiera hypargyrea, together with two known triterpene saponins (3 and 4) as well as the triterpenes friedelin, friedelan 3β-ol and oleanolic acid. The structures of the new compounds were established mainly by 2D NMR techniques of their peracetylated derivatives as 3-O-[α-Lrhamnopyranosyl (1 → 3)-β-D-xylopyranosyl (1 → 4)]-β-D-glucopyranosyl-oleanolic acid-28-O-β-D-glucopyranoside and 3-O-[α-Lrhamnopyranosyl (1 → 3)-β-D-xylopyranosyl (1 → 4)]-β-D-glucopyranosyl oleanolic acid respectively. Keywords: Viguiera hypargyrea, Asteraceae, roots, oleanolic acid saponins, bisdesmosides. Resumen. De las raíces de Viguiera hypargyrea se aislaron dos nuevas saponinas triterpénicas (1-2), cuyas agliconas corresponden al esqueleto del oleanano, junto con dos saponinas triterpénicas conocidas (3 y 4), así como los triterpenos friedelina, friedelan-3β-ol y ácido oleanólico. Las estructuras de los compuestos novedosos fueron establecidas principalmente por medio de técnicas de RMN 2D de sus derivados peracetilados como 3-O-[α-L-rhamnopiranosil (1 → 3)-β-D-xilopiranosil (1 → 4)]-β-D-glucopiranosil-ácido oleanólico-28-O-β-D-glucopiranósido y ácido 3-O-[α-L-ramnopiranosil (1 → 3)-β-D-xilopiranosil (1 → 4)]-β-D-glucopiranosil oleanólico respectivamente. Palabras clave: Viguiera hypargyrea, Asteraceae, raíces, saponinas del ácido olanólico, bisdesmósidos. Introduction ranosyl olean-12-en-28-oate (3) and 3-O-[methyl-β-D-glucuronopyranosiduronoate]-28-O-β-D-glucopyranosyl oleanolate (4), which were identified by comparison of their spectroscopic data with those previously described [5, 6]. In this paper, we report the structural determination of the new saponins on the basis of spectroscopic analysis and acidcatalyzed hydrolysis. Compound 1a was obtained as an oil after acetylation of the natural product 1. In the positive-ion FABMS of 1a, quasimolecular ion peaks were observed at m/z 1600 [M + K + H]+, 1584 [M + Na + H]+, and 1561 [M + H]+, and HRFABMS analysis revealed the molecular formula to be C77H108O33. Other significant peaks visible at m / z 1254 [M + K – C14H19O10]+, 1068 [M – C12H17O7 – C11H15O7]+, and 777 [M – C33H45O21]+, indicated the successive loss of one hexosyl, one deoxyhexosyl, one pentosyl and one hexosyl moieties. Another fragment ion at m/z 437 corresponded to the pseudomolecular ion of the aglycon. On acid hydrolysis, 1a liberated oleanolic acid as the genin, and glucose, rhamnose and xylose, which were identified by comparison with authentic samples by co-TLC, IR and NMR. On alkaline hydrolysis, only glucose was detected by co-TLC with an authentic sample, indicating that the glucose was bound to the genin by a glycosidic ester linkage at C-28 [7]. The 1H and 13C NMR spectra of 1a, which are presented in Table 1, showed that most of the signals of the aglycon were in good agreement with literature data for oleanolic acid [8]. Glycosylation shifts were observed Viguiera hypargyrea Blake (Asteraceae) is a perennial herb distributed on Northern Mexico [1]. The roots of this plant are used for gastrointestinal disorders in Mexican traditional medicine and it is commonly known as “plateada” [2]. Diterpenic acids and sesquiterpene lactones have been reported from the leaves [3]. We have recently reported that the n-hexane and ethyl acetate-soluble portions and their principal diterpenic acid components ent-beyer-15-en-19-oic and ent-kaur-16-en19-oic acids showed antispasmodic and antimicrobial effects [4]. Although the methanol-soluble portion did not exhibit apparent antispasmodic and antimicrobial activity at a sample concentration of 25 µg/mL and 10 mg/mL respectively, we had interest in the chemical constituents of this fraction, and here we report the results. Results and discussion Chromatographic separations of the methanol soluble fraction have resulted in the isolation of the known triterpenes friedelin, friedelan-3-β-ol and oleanolic acid, which were identified by direct comparison with authentic samples. Two new triterpene saponins based on the oleanane skeleton (1,2), which were characterized as their peracetate derivatives (1a,2a) were also isolated, together with the known saponins β-D-glucopy- 174 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Laura Alvarez et al. Table 1. 13C (125 MHz) and 1H (500 MHz) NMR Spectral Data for the Aglycon Part of Compounds 1a, 2a and 2b (CDCl3, δ in ppm). 1a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 OMe 38.50 27.69 90.41 38.81 55.56 18.15 31.74 39.33 47.58 36.68 23.41 122.92 142.85 41.02 25.62 22.82 46.81 41.68 45.79 30.58 33.74 32.96 27.69 16.27 15.22 16.92 25.62 175.59 32.96 23.41 2a δH δC 0.82, 1.59 1.07, 1.67 2.97 dd(11.7, 5) 0.64 1.32, 1.46 1.26, 1.61 1.46 1.86 5.31 t(3) 1.1, 1.72 1.28, 1.54 2.81 dd(14, 4) 1.14, 1.62 1.20, 1.32 1.27, 1.36 0.90 0.70 0.86 0.72 1.10 0.89 0.88 2b δC δH δC δH 38.46 27.65 90.44 38.84 55.59 18.13 32.37 39.30 47.65 36.75 23.38 122.64 143.60 40.96 25.69 2.97 46.49 1.60 45.95 30.63 33.80 32.63 27.72 16.25 15.19 17.01 25.85 183.05 33.02 23.54 0.89, 1.56 1.07, 1.67 2.95 dd(13.5, 4.5) 38.6 27.1 90.41 38.80 54.9 18.13 32.5 39.30 47.60 36.74 23.30 122.59 143.61 41.00 25.65 22.95 46.52 41.65 45.93 30.60 33.79 32.59 27.70 16.25 15.20 17.10 25.80 179.80 32.90 23.50 51.82 0.85, 1.60 1.07, 1.66 2.97 dd (13.0, 5) at C-3 and C-28 of the aglycon, indicating that the saccharide units were attached at these positions (i.e., signals at δ 90.44 and 175.56 represented a downfield shift by 10.6 ppm and an upfield shift by 3.6 ppm, respectively, when compared with the analogous data for oleanolic acid). Compound 1a was shown to contain four sugar residues in a HMQC NMR experiment, which revealed the correlations between anomeric carbons in the δ 105-90 range and anomeric proton signals resonating between δ 4.0 and 6.1. Thus, the anomeric 13C signals at δ 103.0, 100.89, 95.98 and 91.58 gave cross-peaks with anomeric protons at δ 4.35 d (J = 8.0 Hz), 4.44 d (J = 8.0 Hz), 5.07 d (J = 1.5 Hz), and 5.58 d (J = 8.0 Hz) respectively. The sugar moieties of 1a were assigned mainly from the 1H1H COSY, HMQC, and HMBC NMR spectra. Evaluation of spin-spin couplings and chemical shifts allowed the identification of one β-xylopyranose unit with the anomeric proton at δ 4.44, one α-rhamnopyranose unit with the anomeric proton at δ 5.07, and two β-glucopyranose units with the anomeric protons resonating at δ 5.58 and δ 4.35 respectively, with the former linked to the carboxylic group of the aglycon through an ester linkage, and the latter being linked to C-3 of the agly- 0.67 1.31, 1.46 1.59, 1.74 dd(13.5, 4.5) 1.49 dd(11, 6.5) 1.86 5.26 t(3.5) 1.64, 1.74 1.60, 1.90 2.80 dd(14, 4) 1.13, 1.60 1.20, 1.33 1.27, 1.36 0.86 0.69 0.87 0.73 1.11 0.89 0.91 0.67 1.31, 1.46 1.55, 1.47 dd (11, 6.5) 1.83 5.25 t(3.5) 1.60, 1.73 1.60, 1.92 2.80 dd (13.5, 5) 1.14, 1.62 1.21, 1.32 1.27, 1.36 0.88 0.69 0.88 0.73 1.11 0.89 0.90 3.61 con [9,10]. The common D-configuration for xylose and glucose and the L-configuration for rhamnose were assumed to be those of the most commonly encountered analogues in the plant kingdom [11]. The sequence of the sugar moieties in 1a was determined from the HMBC and NOESY NMR spectra. In the HMBC spectrum, long-range 13C-1H correlations were observed between the signals at δC 175.56 and δH 5.58, δC 72.65 and δH 5.07, δC 74.87 and δH 4.44, and δC 90.44 and δH 4.35. Accordingly, the glucopyranose unit with the anomeric proton at δ 4.35 was linked to C-3 of the aglycon, and the rhamnose and xylose units were linked to C-3 and C-4 positions of this glucose unit. The other glucose unit (δ 5.58) was linked to C-28 of the aglycon. On the basis of all evidence, the natural product (1) was identified as 3-O-[α-L-rhamnopyranosyl (1 → 3)-β-D-xylopyranosyl (1 → 4)]-β-D-glucopyranosyl-oleanolic acid-28-O-β-D-glucopyranoside. Compound 2a displayed a 1H NMR spectrum very similar to that of compound 1a, with a triplet at δH 5.20 (J = 3.0 Hz) and seven methyl singlets in the high-field region. A significant difference was the absence of the anomeric doublet at δH 5.58. The 13C NMR spectrum contained six carbon less (47 Two New Oleanolic Acid Saponins from the Roots of Viguiera hypargyrea 175 Table 2. 13C (125 MHz) and 1H (500 MHz) NMR Spectral Data for the Sugar Moieties of Compounds 1a, 2a and 2b (CDCl3, δ in ppm; J in Hz). 1a δC C-3-glucose 1 2 3 4 5 6 xylose 1 2 3 4 5) 5 2a δH δC 2b δH δC δH 103.0 73.91 72.65 74.82 72.87 61.70 4.35 d(8) 5.03 dd(9.5, 8) 3.87 t(9.5) 3.79 t(9.5) 3.47 ddd(9.5, 5, 2) 4.01 dd(12, 5) 4.60 dd(12, 2) 103.05 73.92 72.65 74.87 72.86 61.73 4.36 d(8) 5.02 dd(9.5, 8) 3.87 t(9.5) 3.79 t(9.5) 3.47 ddd(9.5, 5, 2) 4.01 dd(12. 5) 4.60 dd(12, 2) 103.05 73.90 72.65 74.85 72.86 61.72 4.36 d(8) 5.02 dd(9.5, 8) 3.87 t(9.5) 3.78 t(9.5) 3.47 ddd(9.5, 5, 2) 4.01 dd(12,5) 4.60 dd(12,2) 100.89 71.33 73.0 68.03 4.44 d(8) 4.96 dd(9.5, 8) 5.14 t(9.5) 5.09 ddd(10.5, 9.5, 5) 100.91 71.36 73.05 68.30 4.44 d(8) 4.96 dd (9.5, 8) 5.14 t(9.5) 5.07 ddd(10.5, 9.5, 5) 100.90 71.35 73.02 68.30 4.44 d(8) 4.96 dd(9.5, 8) 5.14 t(9.5) 5.07 ddd(10.5, 9.5, 62.82 3.28 t(10.5) 4.35 dd(10.5, 5) 62.83 3.28 t(10.5) 4.36 dd(10.5, 5) 62.84 3.28 t (10.5) 4.36 dd (10.5, 5) rhamnose 1 2 3 4 5 6 95.98 70.95 67.85 71.49 66.26 17.13 5.07 d(1.5) 5.01 dd(3.5, 1.5) 5.68 dd(10.5, 3.5) 5.13 t(9.5) 4.53 dd(9.5, 6.5) 1.25 d(6.5) 96.02 70.93 67.87 71.53 66.27 17.13 5.07 d(1.5) 5.01 dd(3.5, 1.5) 5.68 dd(10.5, 3.5) 5.13 t (9.5) 4.53 dd(9.5, 6.5) 1.25 d(6.5) 96.05 70.95 67.87 71.52 66.25 17.13 5.06 d(1.5) 5.01 dd(3.5, 1.5) 5.68 dd (10.5, 3.5) 5.13 t(9.5) 4.53 dd (9.5, 6.5) 1.25 d(6.5) C-28-glucose 1 2 3 4 5 6 91.58 68.28 72.87 69.96 72.46 61.54 5.58 d(8) 5.10 dd(9.5,8) 5.24 t(9.5) 5.17 t(9.5) 3.79 ddd(9.5, 5, 2) 4.04 dd(12, 5) 4.27 dd(12, 2) singlets) than 1a, and the chemical shifts values of the carbons were within the range of 1 ppm of those found for compound 1a, with the exception of C-28, which appeared at δC 183.05 (cf. δC 175.56 for 1a, Tables 1 and 2). These observations indicated the absence of the glucopyranose unit at the carboxyl group. As additional proof, compound 1 was hydrolyzed with KOH giving compound 2 and glucose. Moreover, methylation of 2a with diazomethane afforded the methyl ester derivative (2b). The position of the methyl ester in 2b was determined from the HMBC correlation between the methyl ester proton δH 3.61 (s, -OCH3) and C-28 (δC 179.80). All those data allowed compound 2 to be assigned the structure 3-O-[α-L-rhamnopyranosyl (1 → 3)-β-D-xylopyranosyl (1 → 4)]-β-D-glucopyranosyl oleanolic acid. Although the methanol extract from the roots of Viguiera hypargyrea did not show any antispasmodic and antimicrobial activities [4], the presence of mono and bisdesmoside saponins in this extract is noteworthy, since various triterpene saponins structurally related to those isolated in this work, have shown important biological activities such as inhibitory effects on ethanol absorption [12], as well as hypoglycemic activity [13]. Experimental General Experimental Procedures. Optical rotations were measured on a Perkin-Elmer 241 digital polarimeter at 25 °C. IR spectra were recorded on a Bruker Vector 22 FTIR. All NMR spectra were recorded on a Varian Unity Plus-500 at 500 MHz for 1H NMR, 1H-1H COSY, HMQC, HMBC and 1H-1H TOCSY and 125 MHz for 13C NMR and 13C DEPT in CDCl3. Chemical shifts are reported in ppm relative to TMS. FABMS and HRFABMS were performed using a Hewlett Packard 5985-B and a JEOL-AX 505 HA mass spectrometer, respectively. 176 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Laura Alvarez et al. H OR2 O R2 O R2O O OR2 O O O OR2 O R2O COOR1 OR2 R2 O R1 Glcp Glcp H H Me 1 1a 2 2a 2b R2 H Ac H Ac Ac 18 4 R= O H 3 R=H MeOOC HO HO O O O OH O RO OH OH HO OH Plant Material. The roots of V. hypargyrea were collected near to Durango City, on September 15th of 1997 and identified by Dr. Robert Bye (Instituto de Biología de la UNAM). A Botanical sample was prepared and deposited for reference at the National Herbarium of Mexico (MEXU) with the code number MEXU961417. Extraction and Isolation. 200 g of the methanol extract obtained previously [4] were fractionated by percolation using a gradient of CH2Cl2-MeOH yielding six fractions: Fr. I (9:1, 16.0 g); Fr. II (85:15, 2.4 g); Fr. III (4:1, 4.0 g); Fr. IV (7:3, 7.3 g); Fr. V (1:1, 18.2 g) and Fr. VI (3:7, 22.4 g). Fr. I was purified by column chromatography using mixtures of n-hexane-EtOAc; fractions eluted with 95:5 (n-hexane-EtOAc) afforded 325 mg of friedelin (0.010 %, mp 242-245 °C); from fractions eluted with n-hexane-EtOAc 9:1 crystallized 83 mg of friedelan-3-β-ol (0.0027 %, mp 249-253 °C), fractions eluted with n-hexane-EtOAc (85:15) yielded 54 mg of the mixture of β-sitosterol and stigmasterol. Fr. II was applied to a silica gel column using a gradient system of CH2Cl2-MeOH to yield 1.7 g of oleanolic acid (0.056 %, mp 196-198 °C). Chromatographic analyses of Fr. III showed that this was mainly composed by the saponins glucopyranosyl oleanolate (3), found previously in the ethyl acetate extract [4], and 3-O(methyl-β-D-glucuronopyranosiduronoate)-28-O-β-D-glucopyranosyl oleanolate (4, mp 217-218 °C) isolated previously from V. decurrens [5]. Fr. IV was applied to a silica gel column using EtOAc-MeOH-AcOH-H2O (11:2:2:1) as isocratic elution mixture. Fractions 8-12 afforded 50 mg of 4; fractions 17-26 yielded 12 mg of 1 (mp 129-132 °C) and 9 mg of 2 (mp 276-278 °C), and fractions 31-43 afforded 820 mg of glucose. Fr. V was mainly composed by sucrose, identified by direct comparison with authentic sample. Fr. VI was ground with acetone to yield a mixture of saponins which was acetylated with Ac2O-Py and the residue was chromatographed on silica gel column using mixtures of CH2Cl2-Acetone: Fractions eluted with 9: 1 (CH2Cl2-acetone) yielded 843 mg of 1a (0.028 % of dry plant). Fractions eluted with CH2Cl2-acetone (8:2) afforded sucrose acetylated and 632 mg of 2a (0.021 % of dry plant). Friedelin, friedelan-3-β-ol and oleanolic acid, were identified by direct comparison (IR, TLC) with authentic samples, while compounds 1a, 2a and 3-4 were characterized by means of physicochemical evidence. Compound 1a. Oil, [α] D 25 – 4.4° (c 0.05, MeOH); IR (CHCl3) νmax 2922, 1757, 1452, 1376, 1050 cm–1; 1H NMR (CDCl3, 500 MHz) see Tables 1 and 2; 13C NMR (CDCl3, 125 MHz) see Tables 1 and 2; FABMS m/z 1600 [M + K + H]+ and 1584 [M + Na + H]+, 1254 [M + K – Glc]+, 1068 [M – rham – xyl]+, 785 [M – glc – rham – xyl]+, 437 [C30H46O2]+; HRFABMS m/z 1561.6983 (calcd for C 77H 108O 33, 1561. 6995). Compound 2a. Colorless powder, mp 101-102 °C, αD25 + 6° (c 0.05, CHCl3); IR (CHCl3) νmax 3500-3400, 2922, 1757, 1452, 1376, 1050 cm –1; 1H NMR (CDCl 3, 500 MHz) see Tables 1 and 2 13C NMR (CDCl3, 125 MHz) see Tables 1 and 2; FABMS m/z 1271 [M + K]+, 1255 [M + Na]+, 1233 [M]+, 958 [M – xyl]+, 954 [M – rham]+, 669 [M – rham – xyl]+, 467 [M – rham – xyl – glc]+; HRFABMS m/z 1233.4209 (calcd for C63H92O24, 1233.4212). Compound 2b. Compound 2a (20 mg) was treated with diazomethane in diethyl ether to yield 20 mg of 2b: Oil, [α]D25 + 2.5° (c 0.2, CHCl3), IR (CHCl3) νmax 2958, 1728, 1462, 1377, 1072 cm–1; 1H NMR (CDCl3, 500 MHz) see Tables 1 and 2; 13C NMR (CDCl , 125 MHz) see Tables 1 and 2. 3 Acid hydrolysis of compounds 1 and 2. Saponins 1 (5 mg), and 2 (5 mg) in 0.5 M HCl (dioxane-H2O, 1:1; 5 ml) were refluxed on a water bath at 100 °C for 2 h. After cooling, the nonpolar reaction product was separated by precipitation with ice (3 g) and filtration. The aqueous layer was neutralized with NH4OH and reduced to dryness by lyophilization. The sugars were analyzed by silica gel TLC [EtOAc-MeOH-H2OAcOH (11:2:2:2)] by comparison with standard sugars. Alkaline hydrolysis of compound 1. The saponin (12 mg) in KOH 10 % (4 mL) was heated at 100 °C for 75 min. After acidification with HCl (pH 5), the monodesmoside was extracted with n-BuOH. Comparison with compound 2 demonstrated that both compounds were identical. The aqueous solution contained glucose was identified by TLC comparison with an authentic sample. Two New Oleanolic Acid Saponins from the Roots of Viguiera hypargyrea Acknowledgment We thank Rocío Patiño, María Isabel Chávez, Francisco Javier Pérez, and Luis Velasco (Instituto de Química, UNAM) for assistance. This work was supported in part by CONACYT (Project 3419P-N and grant 96363). References 1. Blake, S. F. Contribution Gray Herbarium Harvard University 1918, 54, 11-16. 2. Martínez, M. Catálogo de Plantas Medicinales de México. Ed. Botas, 1969, 45. 3. Alvarez, L.; Mata, R.; Delgado, G.; Romo de Vivar, A. Phytochemistry 1985, 24, 2973-2976. 4. Zamilpa, A.; Tortoriello, J.; Navarro, V.; Delgado, G.; Alvarez, L. Planta Med. 2002, 68, 281-283. 177 5. Sakai, S.; Katsumata, M.; Satoh, Y.; Nagasao, M.; Miyakoshi, M.; Ida, Y. Phytochemistry 1994, 35, 1319-1321. 6. Marquina, S.; Maldonado, N.; Garduño-Ramírez, M. L.; Aranda, E.; Villarreal, M. L.; Navarro, V.; Bye, R.; Delgado, G.; Alvarez, L. Phytochemistry 2001, 56, 93-97. 7. Shibata, S.; Kitagawa, I.; Fujimoto, H. Tetrahedron Lett. 1965, 3783-3788. 8. Kubota, T.; Hinoh, H. Tetrahedron Lett. 1968, 303-306. 9. Agrawal, P. K.; Jain, D. C.; Gupta, R. K.; Thakur, R. S. Phytochemistry 1985, 24, 2479-2496. 10. Shashi, B.; Sudip, K.; Poddar, G. Phytochemistry 1988, 27, 30573067. 11. Ahmad, V. O.; Basha, A. Spectroscopic Data of Saponins. The triterpenoid glycosides Vol. I-III. CRC Press, Boca Raton, 2000. 12. Yoshikawa, M.; Murakami, T.; Harada, E.; Murakami, N.; Yamahara, J.; Matsuda, H. Chem. Pharm. Bull. 1996, 44, 19151918. 13.Yoshihawa, M.; Murakami, T.; Harada, E.; Murakami, N.; Yamahara, J.; Matsuda, H. Chem. Pharm. Bull. 1996, 44, 19231925. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 178-185 Investigación Análisis isobolográfico de la interacción entre α-sanshool, sesamina, asarinina, fagaramida y piperina sobre la actividad larvicida en Culex quinquefasciatus Say Andrés Navarrete,1,* Alejandro Flores,2 Carmen Sixtos3 y Benito Reyes3 Departamento de Farmacia, Facultad de Química, Universidad Nacional Autónoma de México, Ciudad Universitaria, Coyoacán 04510, México D.F.; Fax: +5622-5329; E-mail: [email protected] 2 Facultad de Estudios Superiores Zaragoza. Universidad Nacional Autónoma de México. Avenida Guelatao 66, Colonia Ejercito de Oriente, Iztapalapa 09230, México D.F., México. 3 Laboratorio de productos Naturales. Área de Química, Universidad Autónoma Chapingo. Texcoco 56230, Estado de México, México 1 Recibido el 29 de abril del 2003; aceptado el 25 de junio del 2003 Dedicado al Dr. Alfonso Romo de Vivar Resumen. Se realizó el análisis isobolográfico de la interacción larvicida entre α-sanshool con sesamina, asarinina, fagaramida y piperina en Culex quinquefasciatus. El α-sanshool fue el compuesto más activo (CL50 = 3.97 ± 0.61 ppm) seguido de fagaramida (CL50 = 7.92 ± 1.22 ppm), piperina (CL50 = 10.02 ± 2.44 ppm), asarinina (CL50 = 69.95 ± 15.00 ppm) y sesamina (CL50 = 277.4 ± 35.75 ppm). La sesamina y la asarinina presentaron una interacción sinérgica con el α-sanshool, en tanto que con fagaramida y con piperina se observó un efecto aditivo. Palabras clave: Análisis isobolográfico, actividad larvicida, α-sanshool, sesamina, asarinina, fagaramida piperina, Culex quinquefasciatus. Abstract. An isobolographic analysis of larvicidal interactions between α-sanshool with sesamin, asarinin, fagaramide and piperine on Culex quinquefasciatus was performed. α-Sanshool was the most active compound (LC50 =3.97 ± 0.61 ppm) followed by fagaramide (LC50 = 7.92 ± 1.22 ppm), piperine (LC50 = 10.02 ± 2.44 ppm), asarinin (LC50 = 69.95 ± 15.00 ppm) and sesamin (LC50 = 277.4 ± 35.75 ppm). Synergistic interaction between a-sanshool and sesamin or asarinin was observed, whereas with fagaramide or piperine produced an additive effect. Keywords: Isobolographic analysis, larvicidal activity, α-sanshool, sesamin, asarinin, fagaramide, piperine, Culex quinquefasciatus. Introducción efectos individuales y que puedan dar lugar a productos de más pronto uso. El α-sanshool 1 (Fig. 1) es una isobutilamida con actividad larvicida aislada de la corteza de Zanthoxylum liebmannianum [3]. El α-sanshool pertenece a un grupo de compuestos de origen natural con propiedades insecticidas importantes, cuyas fuentes principalmente de obtención son las plantas de las familias Piperaceae, Aristolochiaceae y Rutaceae [4-9]. Las isobutilamidas causan la caída rápida y la muerte de insectos voladores [5], sin embargo, la mayoría de ellas son poco estables al medio ambiente y son oxidadas rápidamente por los insectos. Se sabe que las isobutilamidas afectan a los canales de sodio [9] y que bloquean a los canales de calcio [10], pero en realidad, se conoce muy poco de su mecanismo de acción insecticida y menos aún se conoce la forma en la cual causan una variedad de actividades en diversos organismos, incluyéndose a los mamíferos en donde provocan anestesia local y convulsiones [11-15]. En la formulación de insecticidas comerciales es común la adición de otras sustancias para mejorar su efecto. Entre estas sustancias se encuentran aquellas que inhiben a las enzimas responsables de la oxidación microsomal como la sesamina 2 (Fig. 1) [16] o la combinación con otros compuestos que Los mosquitos son transmisores de varias enfermedades entre los que se encuentran la malaria, la filariasis, la encefalitis y el dengue [1]. Los insecticidas sintéticos son los recursos más importantes en la actualidad para el control de los mosquitos, sin embargo, su uso indiscriminado ha tenido un efecto negativo sobre el medio ambiente y muchas especies de mosquitos han creado resistencia a ellos [2]. Estos factores han dado lugar a investigaciones dirigidas a encontrar agentes de control de los mosquitos que sean biodegradables, específicos y en armonía con la ecología. Los productos derivados de las plantas cumplen con algunas de estas características, y han sido utilizados tradicionalmente por comunidades humanas en muchas partes del mundo en contra de las especies de insectos vectores y plagas por sus propiedades larvicidas, reguladoras del crecimiento, repelentes y por sus efectos sobre la ovoposición en insectos [2]. La investigación en los productos naturales para el control de los insectos vectores de enfermedades sigue siendo una actividad prioritaria y aún no agotada. Una estrategia en esta área es la búsqueda de nuevas estructuras bioactivas, otra puede ser la utilización de los compuestos bioactivos en combinaciones estratégicas que mejoren sus Análisis isobolográfico de la interacción entre α-sanshool, sesamina, asarinina,... 179 O O O 3" 2" 7 H 10 9 8 4 5 2 O N 2' 4' 4' 3' O 3' O O O 4' 1 2 3 O 2" O O 3" 1" 6" 4" O 2' 2 3 3 O 5' 1' H H 4 2 1' 6" 5 1 H 6' 3 2' 3' 2' 1 3 H 5' 7 6 1' 5" 6 8 6" 4 2 6' O 5 1 1" O 5" 4" 7 6 8 2" 4" 1" O 12 11 O 3" 1 N 2" 3' O 1' H O 4' 3" 1" 6" 4" 5" 5 2' 2 4 3 1' 1 N 3' 4' 6' 5' 5" 5 4 Fig. 1. Estructura de los compuestos evaluados. α-Sansool 1, sesamina 2, asarinina 3, fagaramida 4 y piperina 5. tienen actividad insecticida, como la asarinina 3, que se sabe incrementa la actividad de las piretrinas [17]. En los últimos años el estudio de la interacción entre dos sustancias activas se ha realizado utilizando el análisis isobolográfico [18], una estrategia utilizada en farmacología que ha dado lugar a la combinación óptima de fármacos de uso clínico [19]. Este método ofrece una evaluación rigurosa de la interacción entre dos sustancias activas, ya que permite definir una simple adición de los efectos individuales (efecto aditivo), una atenuación (efecto subaditivo) o un sinergismo (efecto superaditivo) entre ellas [20]. Mediante la construcción de un isobolograma (Fig. 2), que es una gráfica en coordenadas rectangulares de pares de dosis o concentraciones (z1, z2) de las sustancias respectivas que producen un nivel determinado de efecto (por ejemplo 50 % del efecto) cuando se aplican en forma conjunta. En esta gráfica los interceptos (Z1*,0) y (0, Z2*) corresponden a la concentración del compuesto menos activo (Z1*) y del compuesto más activo (Z2*) que producen individualmente el mismo nivel de efecto (muerte del 50 % de las larvas). La línea que une a estos dos puntos define a la línea de aditividad y todos los puntos sobre esta línea, que tienen las coordenadas (z1, z2), teóricamente representan los pares de dosis aditivas (Z teo) de los constituyentes administrados en forma conjunta que provocan el mismo nivel de efecto que los compuestos individuales. En la Fig. 2 los puntos A y B corresponden a las concentraciones teóricas aditivas (Z teo) para dos proporciones fijas de las dos sustancias activas. Si al realizar el experimento, las concentraciones para provocar el mismo efecto (Z exp) son menores que las teóricas aditivas (Z teo) indicará que existe un efecto superaditivo o sinergista entre ambas sustancias (puntos P y R en la Fig. 2) pero por el contrario, si las concentraciones experimentales (Z exp) son mayores a las concentraciones teóricas aditivas (Z teo) indicará que existe un efecto subaditivo entre ambas sustancias (puntos Q y S en la Fig. 2). En el análisis isobolográfico de la interacción de dos sustancias activas se demuestra estadísticamente si existe diferencia significativa entre las concentraciones teóricas aditivas (Z teo) y las concentraciones experimentales (Z exp) que provocan el mismo nivel de efecto en una proporción determinada de dichas sustancias, si no se encuentran diferencias estadísticamente significativas entre Z teo y Z exp indicará una interacción aditiva entre estas dos sustancias, es decir que el efecto resultante será la suma de los efectos individuales de las dos sustancias activas [20]. Por el contrario, si no hay diferencia entre Z teo y Z exp, esto indicará que el mecanismo de acción por el cual actúan las dos sustancias es similar [20]. En el presente trabajo se realizó el estudio de la interacción del α-sanshool con diferentes proporciones de sesamina, asarinina, fagaramida 4 o piperina 5 (Fig. 1) a través del análisis isobolográfico, con el propósito de conocer de una manera objetiva las combinaciones que permitan mejorar el efecto larvicida del α-sanshool. Un segundo objetivo de este trabajo fue definir, mediante este análisis, qué tipo de interacción se presenta entre estos compuestos bioactivos de origen natural, a fin de proporcionar información dirigida hacia encontrar las combinaciones que permitan el desarrollo de productos larvicidas eficaces en el control de vectores transmisores de enfermedades importantes para el hombre y los animales. Para alcanzar esta meta se determinó el efecto Cuadro 1. Valores de la CL50 ± EEM del efecto larvicida de los compuestos individuales sobre larvas de mosquito común en su cuarto instar. Compuesto CL50 ± EEM (ppm)1 α-Sanshool Sesamina Asarinina Fagaramida Piperina 3.97 ± 0.61 277.4 ± 35.75 69.95 ± 15.00 7.92 ± 1.22 10.02 ± 2.44 Potencia relativa2 1.0 69.8 17.6 1.99 2.51 EEM = error estándar de la media para 4 niveles de concentración con n = 30 en cada nivel y un valor de r > 0.9 del análisis probit. 2 Respecto al valor de α-sanshool: CL 50 del compuesto/CL50 de αsanshool, e indica el número de veces que debe incrementarse la concentración del compuesto para que se presente el mismo nivel del efecto larvicida que el α-sanshool. 1 180 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Andrés Navarrete et al. Cuadro 2. Concentraciones equiefectivas teóricas (Z teo) y experimentales (Z exp) ± EEM de las combinaciones que provocan el 50 % de la muerte de las larvas de C. quinquefasciatusa. Combinación α-Sanshool: sesamina α-Sanshool: asarinina α-Sanshool: fagaramida α-Sanshool: piperina Proporción Zteo ± EEM Zexp ± EEM Fracción totalb 1: 24.45 1: 6.48 1: 0.32 1: 0.11 1: 15.7 1: 6.4 1: 0.13 1: 2.5 1: 0.9 1: 7.3 1: 1.6 14.722 ± 1.52 40.44 ± 4.80 210.96 ± 27.06 250.40 ± 32.16 7.95 ± 1.07 12.95 ± 2.1 62.45 ± 13.3 5.11 ± 0.56 6.07 ± 0.71 4.7 ± 0.61 6.3 ± 1.00 2.19 ± 0.31* 3.512 ± 0.127** 10.074 ± 0.012** 100.15 ± 0.018 1.84 ± 0.13** 4.03 ± 0.29* 10.08 ± 0.03** 4.51 ± 0.47 4.34 ± 0.56 3.89 ± 0.58 4.93 ± 0.6 0.31 0.14 0.05 **0.39 0.22 0.30 0.16 0.88 0.72 0.82 0.78 aLas concentraciones son el total en la combinación expresadas en ppm. bDe acuerdo a la ecuación 1 descrita en la parte experimental, los valores cercanos a 1 indican aditividad y los valores < 1 indican superaditividad o sinergismo [36]. Diferencia significativa con un valor de *p < 0.05 o **p < 0.01 respecto al valor teórico correspondiente [19, 36]. Resultados y discusión Q Z2 * A S B P S R t i 2 EL α-sanshool, la sesamina, la asarinina, la fagaramida y la piperina en forma individual presentaron efecto larvicida sobre Culex quiquefasciatus dependiente de la concentración (p < 0.05). En el Cuadro 1 se presentan los valores de la concentración letal 50 (CL50) de cada una de estas sustancias determinadas por el método probit [22]. El α-sanshool presentó el efecto larvicida más potente, seguido en orden descendente por fagaramida, piperina, asarinina y sesamina (Cuadro 1). Las proporciones y los valores teóricos y experimentales de la concentración equiefectiva de las diferentes combinaciones evaluadas se proporcionan en el Cuadro 2. El isobolograma de la combinación α-sanshool-sesamina en las cuatro diferentes proporciones evaluadas indica que existe una interacción superaditiva entre estas dos sustancias (Fig. 3). La combinación con 10 ppm de sesamina presentó el efecto sinergista mayor, ya que incrementó la actividad del α-sahsnool que pasó de una CL50 = 3.97 ppm (Cuadro 1) a sólo 0.07 ppm en la combinación (Fig. 3). La administración simultánea de α-sanshool + asarinina también indican una interacción superaditiva en las tres combinaciones evaluadas y también la combinación con 10 ppm de asarinina presentó el efecto sinergista mayor (Fig. 4). En contraste, los isobologramas resultantes de la administración simultánea de α-sanshool + fagaramida (Fig. 5) y α-sanshool + piperina (Fig. 6) presentaron una interacción aditiva, a pesar de que los puntos de las concentraciones experimentales que provocan la muerte del 50 % de las larvas de C. quinquefasciatus se encontraron en la zona de superaditividad, las diferencias no fueron lo suficientemente grandes para que existieran diferencias estadísticamente significativas con respecto a las concentraciones teóricas para provocar el mismo efecto como resultado de la adición de los efectos individuales (Cuadro 2). La interacción superaditiva o sinergista de la sesamina con el α-sanshool apoyan la propuesta de que las isobutilamidas actúan de manera similar a las piretrinas; en efecto, se sabe que la sesamina aumenta la actividad insecticida de las piretrinas al inhibir el sistema enzimático oxidante de función mixta [23], en particular el proceso de oxidación dependiente de la isoforma CYP3A [24]. Las propiedades sinérgicas de la asarinina para incrementar la actividad de las piretrinas fue descrito desde 1942 por Haller y sus colaboradores [18]; sin embargo, no se ha descrito cual es su mecanismo de acción por el que se produce el sinergismo. En el análisis isobolográfico (Fig. 4) se observó un comportamiento similar con la Sustancia 2 letal de diferentes concentraciones de los compuestos puros y en diferentes combinaciones sobre larvas de Culex quinquefasciatus en su cuarto instar [21]. 0 Sustancia 1 Z1 * Fig. 2. Isobolograma en donde se muestra la línea de aditividad (línea continua) para las sustancias 1 y 2, determinada por las concentraciones equiefectivas individuales Z1* y Z2*. Las líneas radiales discontinuas representan las combinaciones de dos proporciones fijas de las dos sustancias. La línea radial 0S representa las combinaciones de las sustancias 1 y 2 que guardan una proporción fija 0.5 Z1* : 0.5 Z2*. La línea radial 0Q representa las combinaciones de las sustancias 1 y 2 que guardan otra proporción de Z1* y Z2*. Los puntos A y B representan las cantidades teóricas aditivas (Z teo), los puntos P y R representan las cantidades experimentales (Z exp) con un efecto superaditivo y los puntos Q y S representan las cantidades experimentales (Z exp) con un efecto subaditivo. Análisis isobolográfico de la interacción entre α-sanshool, sesamina, asarinina,... S α-Sanshool (ppm) 4 3 2 h l( 1 ) 0 0 50 100 150 200 250 300 350 Sesamina (ppm) Fig. 3. Isobolograma de la interacción entre α-sanshool y sesamina en las cuatro proporciones de concentraciones evaluadas. Los círculos vacíos sobre la línea oblicua entre los ejes x y y representan los valores teóricos de aditividad, en tanto que los círculos llenos sobre las líneas radiales discontinuas representan los valores experimentales encontrados. Las barras horizontales y verticales indican el error estándar de la media. Los puntos experimentales cayeron muy por debajo de la línea de aditividad, indicando un sinergismo significativo (p < 0.05) en las cuatro proporciones de concentración evaluadas. Parte experimental Material vegetal. Las hojas y la corteza del tronco de Z. liebmannianum fueron colectadas en de 1995 en San Andrés Cacaloapan del municipio de Tehuacán, Puebla. Una muestra de referencia se encuentra depositada en el Herbario de Plantas Útiles Efraim Hernández X de la Universidad Autónoma Chapingo con el registro XOLO19822126. Procedimientos Generales. Los puntos de fusión se determinaron en un aparato Electrothermal Digital IA9100 y no están corregidos. Los espectros de IR se registraron e un espectrómetro Perkin Elmer modelo 599. Los espectros de RMN-1H (300 MHz) y de RMN- 13C (75MHz) se obtuvieron en un 5 4 3 2 S h 1 l( ) sesamina, lo que podría indicar que la asarinina, que es un estereoisómero de la sesamina, actúe también inhibiendo el sistema oxidante de los insectos [23]. Cabe señalar que la sesamina y la asarinina presentaron el efecto larvicida más bajo cuando se evaluaron en forma individual (Cuadro 1). La CL50 (7.92 ± 1.22 ppm) de la fagaramida encontrada en este trabajo, es cercana a la encontrada por Kubo y sus colaboradores, quienes reportan un valor de la CL50 de 15 ppm en Culex pipiens [13]. De acuerdo a la información que proporciona el análisis isobolográfico, el efecto aditivo observado en la interacción entre α-sanshool y fagaramida (Fig. 5), indica que estos dos compuestos actúan por un mecanismo de acción similar [25]. Este efecto aditivo puede deberse a que ambos compuestos pertenecen a la familia de las isobutilamidas, aunque una es alifática y la otra es aromática de cadena corta. La interacción α-sanshool-piperina también resultó en una interacción aditiva (Fig. 6). El análisis isobolográfico permite postular que el efecto larvicida de la piperina se realiza por un mecanismo de acción similar al de las isobutilamidas. La baja actividad antioxidante descrita para la piperina [26] debido más bien a su efecto inhibidor selectivo por las isoformas CYP1A1 y CYP2B1 [27-29] ponen de manifiesto que el α-sanshool no se oxida por estas enzimas o que dichas isoformas del citocromo P450 no existen en las larvas de C. quinquefasciatus. Existe contradicción respecto a un trabajo en el cual, de una manera indirecta, demuestran que la piperina inhibe a la isoforma CYP3A4 en células hepáticas de seres humanos [30], de ser así se esperaría que la piperina presentara un efecto sinergísta similar al de la sesamina, sin embargo, lo que se onservó fue un efecto aditivo. Se requiere de mayor trabajo experimental para definir la importancia de las diferentes isoformas del citocromo P450 en la inhibición de las enzimas oxidantes de las isobutilamidas. En conclusión, el análisis isobolográfico de la interacción del α-sanshool con sesamina y con asarinina, permitió definir una interacción sinérgica entre las mezclas binarias de α-sanshool con estos dos lignanos, atribuida pobablemente al efecto inhibidor del sistema oxidante en las larvas de Culex quinquefasciatus; en tanto que la interacción con fagaramida y con piperina presentó un efecto aditivo, que indica que estos dos insecticidas actúan por un mecanismo de acción similar al del α-sanshool. El incremento importante en la actividad larvicida del α-sanshool por sesamina y por asarinina puede considerarse para el desarrollo de un larvicida con estos productos naturales, ya que supera la actividad larvicida de otros productos naturales individuales y la formación de mezclas binarias que presenten sinergismo puede ser una buena estrategia a considerar para obtener agentes útiles en el control de los mosquitos vectores de enfermedades importantes y que puedan constituirse en productos comercialmente rentables. Por otro lado, la aplicación del análisis isobolográfico en el campo de los biocidas puede representar una estrategia útil para optimizar la actividad de susutancias bioactivas. α-Sanshool (ppm) 5 181 0 0 20 40 60 80 100 Asarinina (ppm) Fig. 4. Isobolograma de la interacción entre α-sanshool y asarinina en las tres proporciones de concentraciones evaluadas. Los puntos experimentales (círculos llenos) cayeron por debajo de la línea de aditividad, indicando un sinergismo significativo (p < 0.05) en las tres proporciones de concentración evaluadas. Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Andrés Navarrete et al. 5 4 4 3 2 -Sanshool (ppm) α h l( 1 ) 0 α-Sanshool (ppm) 5 S α-Sanshool (ppm) 182 3 2 1 0 0 2 4 6 8 10 Fagaramida (ppm) 0 2 4 6 8 10 12 14 Piperina (ppm) Fig. 5. Isobolograma de la interacción entre α-sanshool y fagaramida en las dos proporciones de concentraciones evaluadas. Los valores experimentales (círculos llenos) no fueron estadísticamente diferentes de los valores teóricos (círculos vacíos) representados sobre la línea de aditividad, lo que indica un efecto aditivo entre estas dos sustancias. Fig. 6. Isobolograma de la interacción entre α-sanshool y piperina en las dos proporciones de concentraciones evaluadas. Los valores experimentales (círculos llenos) no fueron estadísticamente diferentes de los valores teóricos (círculos vacíos) representados sobre la línea de aditividad, lo que indica un efecto aditivo entre estas dos sustancias. espectrómetro Varian VXR-3005 en CDCl3, utilizándose TMS como estándar interno. Los espectros de masas se obtuvieron en un espectrómetro Hewlett-Packard modelo 5890 a 70 eV. La rotación óptica se midió en un polarímetro Perkin-Elmer 241 utilizándose cloroformo como disolvente. H-3’) 0.88 (s, 3H, H-4’), 1.75 (d, 3H, J=6Hz, H-12), 1.80 (m, 1H, H-2’), 2.25 (m, 4H, H-4, H-5), 3.10 (dd, 2H, 6.4,12.9 Hz, H-1’), 6.20 (sa, H-N), 6.36-5.37 (m, H-6-H-11), 6.82 (m, H-2, H-3); RMN- 13 C (CDCl 3 , 125 MHz) δ: 165.98 (C-1), 124.22(C-2), 143.38 (C-3), 32.05 (C-4), 26.54 (C-5), 129.62 (C-6), 129.59 (C-7), 125.27 (C-8), 133.46 (C-9), 131.79 (C10), 130.10 (C-11), 18.29 (C-12), 46.88 (C-1’), 28.59 (C-2’), 20.13 (C-3’, C-4’). Extracción e identificación de los compuestos. La corteza seca y molida (2 kg) se extrajo por maceración por períodos de tres días en forma sucesiva con hexano (8 LX3) y con cloruro de metileno (8LX3). Después de eliminar el disolvente a presión reducida se obtuvieron 58 g de extracto de hexano y 141 g de extracto de cloruro de metileno. Una fracción del extracto de cloruro de metileno (31 g) fue separado por cromatografía en columna preparativa (5 d.i. × 80 cm), utilizándose 300 g de gel de sílice (Merck 70-230 mallas). La elución de la columna se realizó con cloruro de metileno y mezclas de cloruro de metileno y acetato de etilo (9:1; 8:2 y 1:1). Se colectaron un total de 80 fracciones de 100 mL cada una. Se reunieron las fracciones 29-46 de la elusión con cloruro de metileno/acetato de etilo (9:1). El total de este conjunto de fracciones (4 g) se separaron en una segunda columna de gel de sílice (Merck 70-230 mallas, 40 g, 2.5 d.i. × 80 cm) utilizándose como mezcla de elusión cloruro de metileno / acetato de etilo (9:2, 20 mL por fracción) de las fracciones 14-65 se obtuvo el α-sanshool 1 en forma de aceite de color amarillo(0.635 g, 0.144 % de rendimiento, considerando la cantidad que se obtendría del total del extracto), el cual fue identificado por comparación de sus espectros de IR, masas, RMN-1H (300 MHz, CDCl3) y RMN-13C (125 MHz, CDCl3) con los datos espectroscópicos descritos previamente para esta amida [4, 31]: IR νmax (CHCl3) cm–1: 3448, 2872, 1674, 1638, 1518, 994, 970; EMIE: m/z (int.rel): 247 (M+, C16H25NO, 29), 204 ([M-C3H7]+, 5), 167(20), 147(10), 141(97), 107(100), 98(12), 91(46), 79(70); RMN-1H (CDCl3, 300 MHz), δ: 0.86 (s, 3H, Para obtener la sesamina 2 y la asarinina 3, 2.9 kg de hojas secas y molidas de Z. liebmannianum se extrajeron por maceración con hexano (7L × 3) por periódos de tres días. Después de eliminar el disolvente se obtuvieron 62 g de extracto. El total del este extracto se separó por cromatografía en columna de gel de sílice (Merck 70-230 mallas, 970 g, 10 d.i. × 120 cm) iniciándose la elusión con hexano y después con mezclas de hexano y acetato de etilo (9:1, 8:2, 1:1) y acetato de etilo. Se colectaron un total de 270 fracciones de 250 mL cada una. De las fracciones 15-67, eluidas con hexano/acetato de etilo (9:1), se obtuvo un sólido cristalino, que después de recristalizarlo de éter isopropílico se obtuvieron 350 mg (0.012 %) de asarinina (p.f 117-118 ºC). De las fracciones 93-103, eluidas también con hexano/acetato de etilo (9:1), cristalizaron 1.3 g (0.044 %) de sesamina (p.f. 121-122 ºC). La identificación de estos compuestos se realizó por comparación de sus espectros de IR, masas y RMN con las descritas previamente [4]. d(+)-Sesamina 2: [α]d20 = +68 (c 0.1, CHCl3); IR ν max (CHCl3) cm–1 :3022, 2881, 1488, 1245, 1041, 935, 810; EMIE: m/z (int.rel): 354 (M+, C20H18O6, 92), 203 (32), 150 (45), 149 (100), 135 (45), 121 (20), 103 (12), 65 (8); RMN-1H (300 MHz, CDCl3) δ: 3.04 (m, 2H, H-1, H-5), 3.85 (dd, 2H, J= 4, 9 Hz, H-4e, H-8e), 4.22 (dd, 2H, J=8, 9Hz, H-4a, H-8a), 4.70 (d, 2H, J=5 Hz, H-2, H-6), 5.90 (s, 4H, OCH2-O), 6.76 (m, 6H, H-2', H-5', H-6'). (+) Asarinina 3: [α]d20 = +120 (c0.1, CHCl3); IR ν max (CHCl3) cm–1:3010, Análisis isobolográfico de la interacción entre α-sanshool, sesamina, asarinina,... 2985, 1500, 1475, 1435, 1250, 1030, 925, 790, 630; EMIE: m/z (int.rel): 354 (M+, C20H18O6, 25), 203,(15), 179 (10), 150 (26), 149 (100), 135 (49), 121(15); RMN- 1H (300 MHz, CDCl3) δ: 3.3 (m, 2H, H-1, H-8e), 4.81 (d, J = 6.43 Hz, 1H, H-2), 3.85 (m, 2H, H-4a, H-8a), 4.08 (d, J = 6.9 Hz, 1H, H4e), 2.85 (m, 1H, H-5), 4.38 (d, J = 7.1 Hz, 1H, H-6), 6.88 (m, 3H, H-2', H-5', H-6'), 5.93 (s, 2H, O-CH2-O), 595 (s, 2H, OCH2-O); ); RMN-13C (CDCl3, 75 MHz) δ: 54.6 (d,C-1), 82.0 (d, C-2), 70.9 (t, C-4), 50.10 (d, C-5), 87.6 (d, C-6), 69.6 (t, C8), 132.2 (s, C-1'), 135.1 (s, C-1''), 106.3 (d, C-2'), 106.5 (d, C-2"), 146.5 (s, C-3'), 147.1 (s, C-3"), 147.6 (s, C-4'), 147.9 (s, C-4"), 108.1 (d, C-5', C-5"), 118.6 (d, C-6'), 119.5 (d, C6"), 100.9 (t, O-CH2-O), 101.0 (t, O-CH2-O). La piperina 5 se obtuvo de la pimienta negra comercial, siguiendo la metodología descrita por Epstein y colaboradores [32]. Brevemente, 60 g de pimienta negra se calentaron a reflujo durante 20 minutos con 120 mL de cloruro de metileno. La mezcla se dejó enfriar a temperatura ambiente (22 ± 2 ºC), se filtró al vacío y el residuo se lavó varias veces con cloruro de metileno. Después de eliminar el disolvente a presión reducida se obtuvo un aceite obscuro al cual se le adicionó éter etílico para inducir la precipitación de la piperina. El filtrado dejó un residuo amarillo en el papel filtro, el cual se lavó varias veces con éter etílico frío y posteriormente se recristalizó de acetona, obteniéndose finalmente 1.36 g de piperina pura. P.f. 130-131 ºC, IR ν max (KBr) cm–1: 1633, 1611, 1583, 1491, 1447, 1252, 1133, 1031, 996, 927; EMIE: m/z (int.rel): 285 (M+, C17H19O3N, 86), 201 (100), 173 (26), 115 (45), 84 (10); RMN-1H (300 MHz, CDCl3) δ: 1.62 (m, 6H, H-3', H-4', H-5'), 3.58 (m, 4H, H-2', H-6'), 5.98 (s, 2H, OCH2-O), 6.43 (d, 1H, J = 15Hz, H-2), 6.74 (m, 2H, H-5", H6"), 6.89 (m, 1H, H-5), 6.98 (s, 1H, H-2"), 7.41 (m, 2H, H-3, H-4). RMN-13C (CDCl3, 75 MHz) δ: 165.36 (C-1), 119.99 (C2), 142.43 (C-3), 125.30 (C-4), 138.15 (C-5), 46.26 (C-2'), 26.09 (C-3'), 24.60 (C-4'), 26.09 (C-5'), 46.26 (C-6'), 130.95 (C-1"), 105.60 (C-2"), 148.12 (C-3"), 148.05 (C-4"), 108.41 (C-5"), 122.43 (C-6"), 101.21 (O-CH2-O). La fagaramida 4 se obtuvo por síntesis siguiendo la técnica descrita por Elliot y colaboradores [33]. Brevemente, Se calentaron a reflujo por 5 h 5.7 mmol de ácido 3-(1,3-benzodioxol-5-il)-2E-propenóico (Aldrich) con 27.3 mmol de cloruro de tionilo en 20 mL de benceno anhidro. Al término de este tiempo se destiló el exceso de cloruro de tionilo. Al cloruro de ácido así obtenido se le agregaron 13 mmol de isobutilamina disuelta en 50 mL de éter etílico anhidro. La mezcla de reacción se dejó con agitación por 20 h a temperatura ambiente. Después de eliminar el éter etílico, la mezcla de reacción se disolvió en acetato de etilo a la cual se le realizaron extracciones sucesivas con ácido clorhídrico al 10% (p/v), bicarbonato de sodio y agua destilada hasta obtener un pH neutro en la fase acuosa. La fase orgánica se secó con sulfato de sodio anhidro y se eliminó a presión reducida; de esta forma se obtuvo la fagaramida, la cual después de recristalizarla de etanol presentó un pf de 115-116 ºC. Sus datos espectroscópicos de RMN coincidieron con los descritos previamente [34]: RMN1H (300 MHz, CDCl ) δ: 0.95 (s, 1H, H-3'), 0.97 (s, 1H, H-4'), 3 183 1.84 (m, 1H, H-2'), 3.22 (dd, J= 6.4, 12.9, 2H, H-1'), 5.66 (sa, 1H, NH), 5.98 (s, 2H, O-CH2-O), 6.26 (m, 1H, H-2), 6.77 (m, 1H, H-5"), 6.80 (m, 1H, H-6"), 7.3 (s, 1H, H-2"), 7.56 (d, J= 15Hz, 1H, H-3); RMN-13C(CDCl3, 75 MHz) δ: 166.05 (C-1), 118.95 (C-2), 140.60 (C-3), 47.11 (C-1'), 28.68 (C-2'), 20.16 (C-3'), 20.16 (C-4'), 129.38 (C-1"), 106.37 (C-2"), 148.24 (C3"), 148.99 (C-4"), 108.52 (C-5"), 123.71 (C-6"), 101.40 (OCH2-O). Determinación de la actividad larvicida. Los huevecillos de Culex quinquefasciatus se colectaron en estanques de agua del "Campo Experimental El Ranchito" de la Universidad Autónoma Chapingo. Los huevecillos se mantuvieron en recipientes con agua y alimento vegetal, la cual se cambió cada 72 h hasta que emergieron los insectos. Los insectos adultos se mantuvieron en una jaula entomológica en condiciones óptimas para que continuaran su ciclo biológico. Como fuente de sangre para la hembras se utilizaron pollos de aproximadamente 15 días de edad, los cuales se introdujeron a la jaula sólo por las noches. Se colocaron recipientes con agua dentro de la jaula entomológica para que las hembras ovopositaran. Diariamente se colectaron los huevecillos y se mantuvieron en recipientes con agua y alimento vegetal. Este procedimiento se siguió hasta la tercera generación, para evitar efectos residuales de la posible exposición a insecticidas. Los huevecillos de la tercera generación se dejaron desarrollar hasta el cuarto instar para realizar los experimentos. Se colocaron 10 larvas de C. quinquefasciatus del cuarto instar en 5 mL de agua, se adicionaron 100 µL de las soluciones de los compuestos de prueba y se aforaron a 10 mL con agua, 24 h después se determinó el número de larvas muertas. En experimentos preliminares se determinó la ventana de actividad biológica de cada uno de los compuestos, ensayándose concentraciones de 1 a 1000 ppm, en espacios logarítmicos. Cada una de las concentraciones de los compuestos de prueba disueltos en acetona se evaluaron por triplicado. Se ajustó el rango de concentraciones de manera tal que al menos existieran 4 niveles de concentración para determinar la Concentración letal 50 (CL 50) por el método probit [22]. Paralelamente se evaluaron lotes controles tratados con 100 µL de acetona [21]. Para el estudio de la interacción, los compuestos se adicionaron en las proporciones definidas en el Cuadro 2 en el mismo volumen de acetona utilizado en la evaluación de los compuestos individuales. Análisis de la interacción. Para caracterizar la interacción entre α-sanshool con sesamina, asarinina, fagaramida y piperina se utilizó un análisis isobolográfico [20]. De acuerdo a este método sólo se consideraron para el análisis las concentraciones equiefectivas (CL50) de cada compuesto y sus combinaciones obtenidas de las curvas concentración-respuesta. Las concentraciones teóricas aditivas (Z teo) se calcularon de las concentraciones equiefectivas (CL50) de los compuestos individuales de acuerdo al método descrito por Tallarida [35]. La comparación de las concentraciones teóricas (Zteo) y experi- 184 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) mentales (Z exp) permite definir la naturaleza de la interacción (subaditividad o superaditividad) o concluir que no hay interacción (aditividad). Los isobologramas se construyeron de acuerdo al procedimiento descrito por Tallarida y colaboradores [19]. Brevemente, las concentraciones equiefectivas (CL50) de cada compuesto se graficó sobre los ejes x y y. En x se graficó el compuesto menos activo y en y el más activo (α-sanshool). Los interceptos con coordenadas (Z1*,0) y (0, Z2*) corresponden a la CL50 del compuesto menos activo y a la CL50 del compuesto más activo, respectivamente. La línea que se forma por la unión de los dos interceptos corresponde a la línea de aditividad. Todos los puntos sobre esta línea con coordenadas (z1, z2), representa el par de concentraciones teóricas aditivas (Z teo) para los compuestos administrados juntos que dan el mismo nivel de efecto que los compuestos administrados individualmente [19]. Las coordenadas de los puntos experimentales de la combinación (Z exp), que caen por debajo de la línea de aditividad indicarán una superaditividad o sinergismo y si caen por arriba de la línea de aditividad indicarán una subaditividad o que uno de ellos está disminuyendo el efecto del otro [21]. Para describir la magnitud de la interacción se calculó el valor de la fracción total (Cuadro 2) de la concentración equiefectiva del compuesto A, del compuesto B y su combinación, de acuerdo a la Ecuación 1 [36]: Fracción total = (Concentración del compuesto A en al combinación/Concentración del compuesto A solo) + (Concentración del compuesto B (1) en la combinación/Concentración del compuesto B solo) El valor de esta fracción total indica la divergencia entre la concentración equiefectiva experimental (Z exp) de la combinación y la concentración teórica aditiva (Z teo). Valores cercanos a 1 indican aditividad; valores menores de 1 implican una interacción sinérgica; y un valor mayor de 1 indica una interacción subaditiva o de atenuación del efecto [36]. La fracción total se calculó con los valores de las concentraciones de los compuestos solos o en la combinación que provocan la muerte del 50 % de las larvas de C. quinquefasciatus (Cuadro 2). Análisis estadístico. Los valores de la CL50 ± EEM de los compuestos individuales o en las combinaciones en las diferentes proporciones se calculó por el método probit [22]. Para determinar que el efecto es dependiente de la concentración, se realizó el análisis de varianza de la regresión lineal entre probits y logaritmo de la dosis [20]. Para distinguir una interacción sinérgica de un efecto aditivo entre los valores experimentales de las concentraciones que provocan el 50 % de la muerte de las larvas en las diferentes combinaciones (Z exp) y las concentraciones aditivas teóricas (Z teo), se utilizó la prueba t de Student, siguiendo el procedimiento para el análisis isobolográfico [19, 36]. Andrés Navarrete et al. Agradecimientos Se agradece a Q. Marisela Gutíerrez, Q. Georgina Duarte, QFB Margarita Guzmán, QFB Rosa Isela Del Villar Morales y QFB Oscar S. Yañes, USAI Facultad de Química UNAM, por el registro de los espectros de IR, Masas y RMN. El presente trabajo fue financiado parcialmente por la Dirección General de Asuntos del Personal Académico de la UNAM a través del Proyecto IN 203902. Referencias 1. Service, M.W. in Pest and Vectors Management in Tropics. Youdeowei, A.; Service, M.W. 1983, 265-280. 2. Sukuman, K.; Petrich, M.J.; Boobar, L.R. J. Am. Mosq. Control Assoc. 1991, 7, 210-237. 3. Reyes, B.; Navarrete, A.; Sixtos, C.; Aguirre, E.; Jiménez, S.; Estrada, E. Rev. Mex. Cienc. Farm. 1991, 21, 30-34. 4. Haral, G. Planta Med. 1984, 80, 366-375. 5. Blade, R.J. in Recent Advances in Chemistry of Insect Control II. The Royal Society of Chemistry. USA 1990, 151-169. 6. Crombie, L. J. Chem. Soc. 1955, 995-999. 7. Jacobson, M. in Naturally Occurring Insecticides. Jacobson, M. y Crosby, D.G. Marcel Decker Inc., New York 1971, 139-176. 8. Miyacado, M.; Nakayama, I.; Yoshioka, H.; Nakatani, N. Agric. Biol. Chem. 1979, 43, 1609-1611. 9. Nakatani, N.; Inatani, R. Agric. Biol. Chem. 1981, 45, 1473-1476. 10. Benner, J.P. Pest. Sci. 1993, 39, 95-102. 11. Elliot, M.; Farnham, A.W.; James, N.F.; Johnson, D.M.; Pulman, D.A. Pestic. Sci. 1987, 18, 191-201. 12. Kubo, I.; Matsumoto, T.; Klocke, J.A.; Kimikawa, T. Experientia 1984, 40, 340-341. 13. Oriowo, M.A. Planta Med. 1982, 44, 54-56. 14. Yasuda, I.; Takeya, K.; Itokawa, H. Chem. Pharm. Bull. 1981, 63, 395-398. 15. Navarrete, A.; Hong, E. Planta Med. 1996, 62, 250-251. 16. Klocke, J.A. in Economic and Medicinal Plant Research, Vol 3, Wagner, H., Hikino, H., Farnworth, N.R., Eds. Academic Press, London, 1989, 103-114. 17. Haller, H.L.; Laforae, F.B.; Sullivan, W.N. J. Am. Chem Soc. 1942, 64, 187-188. 18. Tallarida, R.; Kimmel, H.L.; Holtzman, S.G. Phycopharmacol. 1997, 133, 378-382. 19. Vinik, H.R.; Bradley, E.L.; Kissin, I. Anesth. Analg. 1999, 88, 667-670. 20. Tallarida, R. Drug synergism and dose-effect data analysis. Chapman & Hall / CRC, Florida, 2000. 21. Vahita, R.; Venkatachalam, M.R.; Murugan, K.; Jebanesan, A. Biores. Tech. 2002, 82, 203-204. 22. Infante, S.; Calderón, L.C.; Manual de análisis probit. Colegio de Posgraduados. Centro de Estadística y Cálculo. Chapingo Estado de México. México 1989. 23. Wilkinson, C.F., in Pesticides formulation, Vankenburg, W., Ed., Marcel Dekker, New York, 1976, 1-64. 24. Parker, R.S.; Sontag, T.J.; Swanson, J.E. Biochem. Biophys. Res. Com. 2000, 277, 531-534. 25. Vinik, H.R.; Bradley, E.L.; Kissin, I. Anesth. Analg. 1999, 88, 667-670. 26. Naidu, K.A.; Thippeswamy, N.B. Mol. Cel. Biochem. 2002, 229, 19-23. 27. Liu, H.; Bigler, S.A. Kidney Int. 2002, 62, 868-876. 28. Reen, R.K.; Wiebel, F.J.; Singh, J. J. Ethnopharmacol. 1997, 58, 165-173. 185 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) 29. Reen, R.K.; Roesch, S.F.; Keifer, F.; Wiebel, F.J.; Singh, J. Biochem. Biophys. Res. Com. 1996, 218, 562-569. 30. Bhardwaj, R.K.; Glaeser, H.; Becquemont, L.; Klotz, U.; Gupta, S.K.; Fromm, M.F. J. Pharmacol. Exp. Ther. 2002, 302, 645-650. 31. Yasuda, Yakeya, K.; Itokawa, H. Phytochem. 1982, 21, 12951298. 32. Epstein, W.; Nettz, F.D.; Seidel, L. J. J. Chem Educ. 1993, 7, 598-599. Andrés Navarrete et al. 33. Elliot, M.; Farnham, A.W.; James, N.F.; Johnson, D.M.; Pulman, D.A. Pestic. Sci. 1987, 18, 211-221. 34. Adesina, S.K. J. Nat. Prod. 1986, 49, 715. 35. Tallarida, R. J. Pain 1992, 49, 93-97. 36. Tallarida, R.J.; Stone, D.J.; Raffa, R.B. Life Sci. 1997, 61, PL417-425. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 186-189 Investigación Cytotoxic Evaluation of a Series of Bisalkanoic Anilides and Bisbenzoyl Diamines Luis Chacón-García, M. Elena Rodríguez, and Roberto Martínez* Instituto de Química, Universidad Nacional Autónoma de México, Circuito Exterior, Ciudad Universitaria, Coyoacán 04510, México, D.F. E-mail: [email protected] Dedicated to Profesor Alfonso Romo de Vivar Recibido el 10 de marzo del 2003; aceptado el 26 de junio del 2003 Abstract. A series of bisalkanoic anilides and bisbenzoyl diamines were synthesized with the aim of elucidating the relationship between molecular structure and cytotoxic activity. Twenty-one derivatives were synthesized and tested on three tumoral cell lines. No apparent relationship was observed between electronic effects and cytotoxic activity, but it was found that compounds in which the 4'-phenyl substituent is fluoride or bromide gave the best inhibition of tumoral cell growth. Keywords: Diamides, alkanediamides, cytotoxic activity. Resumen. El objetivo del presente trabajo fue encontrar la relación entre la estructura molecular y la actividad citotóxica de una serie de anilidas de diácidos y diamidas bisbenzoiladas, para lo cual se sintetizaron veintiuno de los compuestos mencionados. Los resultados de la evaluación citotóxica de estos derivados, en tres líneas celulares, no indicaron ninguna relación con respecto a efectos electrónicos de los substituyentes, si bien los derivados 4-bromofenil y 4-fluorofenil son los más activos. Palabras clave: Diamidas, alcano diamidas, actividad citotóxica. Introduction Results and discussion DNA recognizing molecules such as DNA-intercalators and groove binders have been the subject of increasing interest due to the ongoing search for more active antitumoral compounds. DNA-groove binders have been widely studied as anticancer compounds. In addition, they have been studied as anti-HIV agents and have been incorporated as a linker in DNA bisintercalators [1-4]. The most typical DNA-groove binders are the antibiotics Distamycine A (1) and Netropsin (2), which are characterized by polyamide and polyaromatic functional groups along the DNA recognizing chain [5]. The aromatic portion of these compounds is the pyrrolo system; however, recent studies have investigated compounds incorporating thiazolyl (3) or phenyl (4) (Fig. 1) instead of pyrrolyl, and groove binders that contain the benzimidazolyl moiety have been described in earlier reports [6-8]. Recently, we reported a series of N,N’-(diaminophenyl)alkanediamides 5 which differ in the length of the aliphatic portion. These compounds were shown to inhibit the growth of tumoral cell lines, indicating that this topographical factor has an important influence on DNA recognition [9]. However, the cytotoxic activity of the N,N’-(diaminophenyl)alkanediamides was low. The present investigation was undertaken to study the influence of aryl substituents in these compounds and to find compounds of this type with improved cytotoxic activity. To achieve this, we synthesized a series of bisalkanoic anilides and bisbenzoyl diammines (6-27) and their activities as cytotoxic agents were evaluated. The N,N’-diarylalkanediamides (6-20) (Fig. 2) were synthesized by condensation of the respective 4-substitued aniline (2 equiv.) with succinyl, glutaryl or adipoyl chloride (1 equiv.) in acetone while being stirred and cooled in an iced bath. The products were precipitated, filtered, and washed with acetone. Yields varied from 65 to 96 %. Compounds 21-23 and 25-27 were obtained as described for 6-20 but from condensation of the respective benzoyl chloride and ethylenediamine, 1,2-propanediamine, or piperazine as shown in Figure 2. The compounds were obtained in yields of 75 to 95 %. Compound 24 was obtained by reduction of the nitro derivative 23, using Pd/C and hydrazine in ethanol at reflux for 1 h. Recrystallization from methanol afforded the amine derivative. The yields and spectroscopic data of compounds 6-27 are summarized in Table 1. The percentage of inhibition of the growth of the three tumoral cell lines after treatment with each compound at a concentration of 31 µM is given in Table 1. The groups bonded at the 4’ position were selected on the basis of their electron withdrawing or donating properties, and their hydrogen bonding capabilities. The first series of compounds comprises N,N’-diarylalkanediamides with different numbers of methylenes in the aliphatic chain. The first compounds synthesized and probed were 6 to 10 (n = 2). These compounds displayed little activity in the three cell lines. The compound which inhibits cell growth to the greatest extent (57 % in K562) is 6 (R = F), followed by 7 (R = Br) in the same cell line. Cytotoxic Evaluation of a Series of Bisalkanoic Anilides and Bisbenzoyl Diamines 187 Table 1. Physical properties, spectroscopic data and inhibition of the growth of compounds 6-27 at concentration 31 µM. Comp. No. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 R n M.W. Yield (%) m.p. Ref. (a) K562 (b) PC-3 (c) U251 (d) F Br OMe OH H Br F Cl Br I OMe OH H HNCOCH3 CN Br Br NO2 NH2 NO2 Br NO2 2 2 2 2 2 3 4 4 4 4 4 4 4 4 4 — — — — — — — 304 426 328 300 268 318 332 365 454 548 356 328 296 410 346 440 426 358 298 372 452 384 65 69 80 92 83 70 65 75 75 77 89 96 95 80 85 75 80 80 85 75 89 95 243-245 281-282 255-256 273-274 231-232 254-256 230-233 255-256 287-288 [10] [10,11, 12] [10] [13] [10, 12] [10] — [10] [12] — [10, 14] [13, 15] [10, 12] — [14] — — — — — — [16] 57 40 22 0 0 55 96 13 5 0 74 80 24 0 9 83 0 0 0 0 0 0 25 6 13 13 16 0 10 0 0 0 18 34 18 0 13 72 0 8 3 26 0 4 0 0 10 0 0 4 39 0 3 0 17 4 0 0 42 100 5 0 0 0 6 0 233-235 244-245 > 350 272-233 268-270 281-283 253-254 284-285 235-237 270-273 318-320 (a) References of the synthesis for previously reported compounds. (b) Leukemia (c) Prostate (d) CNS. H N O H H N N Me NH2 H2N H N Me N N O O HN O H2N H N N H O HN O N Me N O H2N 23 2 O N S R' O n 3 H ( )n N H N H2N O O N H NH2 5 O 24 O COCl H N H N N H H2N H N R NH2 O Pd / C NH2NH2 4 O 21,25 R'=Me 22,23 R'=H O HN H2N H N N H R R O Me R R R' O NH2 R' HN O 6-20 O NH2 N COCl HN ( )n HN H2N NH2 R HN 1 H ( )n N O R O Me H H N ClCO(CH2)nCOCl H2N O HN O NH2 R N R N O R 26-27 Fig. 1. Examples of compounds containing polyamide and polyaromatic functional groups along the DNA recognizing chain. Fig. 2. Synthesis of compounds 6-27. (R values are reported on Table 1.) To study the influence of the length of the aliphatic chain on cytotoxic activity, we prepared compounds with a four methylene chain (12, 14, 16-18). It should be pointed out that compounds 13 (R = Cl) and 15 (R = I) were included due to the apparent tendency of halogens to present activity. In addition, compounds 19 (R = NHCOCH3) and 20 (R = CN) were included in the study to investigate the effects of the NHCOCH 3 and CN functional groups. In contrast to the almost complete lack of activity shown by the first series (n = 2), compound 12 (R = F) induced almost 100 % inhibition of growth in K562 cell line and the functional groups OMe (16) and OH (17) were found to enhance cytotoxicity. The rest of the compounds showed no activity. 188 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Compounds 21-25 were examined to analyze the importance of the relative position of the amide group and the presence of branching in the aliphatic chain. Surprisingly, compound 21 was the most active in the bromide series, displaying relatively good inhibition in the three cell lines. Given the activity of 21, it is surprising that 22 was inactive. To complete the series of bromide compounds, 11 (n = 3, R = Br) was obtained; it showed greater activity than 7 (R = Br, n = 2) but less than 14 (R = Br, n = 4) in K562 cell line. The inhibition resulted by 21, lead to the resentment that conformation could be implicated in the cytotoxic activity. To test this idea, compounds 26 and 27 were synthesized; however, both of these compounds were inactive. Although these molecules are structurally similar to 21-25, the formers (26 and 27) are not very capable of interacting by hydrogen bonding. This is a very important factor affecting cytotoxicity in DNA groove binders due to the stability of the DNA-ligand complex. Conclusions The data presented here are inconclusive regarding the relationship between electronic factors or hydrogen bonding capability and inhibition of the growth in tumor cell lines. The present results also show no clear link between the presence of halogens or the length of the aliphatic chain and the cytotoxicity of a compound. However, this study did reveal the interesting finding that the compounds which presented cytotoxic activities were primarily those containing fluoride or bromide. Experimental Chemistry General procedure for the preparation of 6-20. Diacyl chloride (0.72 mmol) was added to a solution of 4-R-aniline (1.44 mmol) in 15 mL of acetone at 5 °C. After 2 h stirring, the mixture was filtered and washed with acetone to afford 6-20. 12: 1H NMR (δ, J(Hz)): 1.60 (s, 4H), 2.30 (s, 4H), 7.10 (m, 4H), 7.57 (m, 4H), 9.94 (s, 2H); IR ν (cm–1) 1652, 3305. 15: 1H NMR (δ, J(Hz)): 1.59 (s, 4H), 2.31 (s, 4H), 7.41 (d, J = 8.8, 4H), 7.60 (d, J 8.7, 4H), 9.97 (s, 2H); IR ν (cm–1) 1657, 3292. 19: 1H NMR (δ, J(Hz)) 1.59 (m, 4H), 1.99 (s, 6H), 2.28 (s, 4H), 7.46 (s, 8H), 9.80 (s, 2H), 9.83 (s, 2H); IR ν (cm–1) 1659, 3298. General procedure for the preparation of 21-23 and 25-27. 4bromobenzoyl chloride (1 mmol) was added to a solution of diamine (0.7 mmol) in 15 mL of acetone at 5 °C. After 2 h stirring, water was added and the precipitated filtered and washed with water and acetone to afford 21-23 or 25. 21: 1H NMR (δ, J(Hz)) 1.15 (d, J 6.64, 3H), 3.35 (t, J 9, 2H), 4.23 (m, 1H), 7.64 (d, J 8.6, 4H), 7.76 (d, J 8.5, 2H), 8.34 (d, J 8.2, 1H) 8.63 (t, J 5.6, 1H); IR ν (cm–1) 1637, 3301. Luis Chacón-García et al. 22: 1H NMR (δ, J(Hz)): 1.33 (s, 4H), 7.65 (d, J 8.85, 4H), 7.77 (d, J 8.5, 4H), 8.67 (s, 2H); IR ν (cm–1) 1633, 3287. 23: 1H NMR (δ, J(Hz) 3.47 (d, J=2.7, 4H), 8.07 (d, J = 8.8, 4H), 8.30 (d, J 8.9, 4H), 9.00 (s, 2H); IR ν (cm–1) 1640-3319. 25: 1H NMR (δ, J(Hz)): 1.2 (d, J 6.7, 3H), 3.45 (t, J 6.3, 2H), 4.3 (m, 1H), 8.03 (d, J 8.96, 2H), 8.05 (d, J 9, 2H), 8.63 (d, J 8.14, 1H), 8.92 (t, J 5.6, 1H); IR ν (cm–1) 1661, 3318. 26: 1H NMR (δ, J (Hz)) 3.54 (m, 8H), 7.37 (d, J 8.4, 4H), 7.64 (d, J 8, 4H); IR ν (cm–1). 1635. Preparation of 24. Ethanol (10 ml), Pd/C 5% (0.046 g), Hidrazine (0.818 ml, 25.9 mmol), water (0.93 ml) and 23 (756 mg, 2.59 mmol) were mixed in a bottom flask. The mixture was refluxed for 2h. The resulting solid was dissolved in methanol with heat and filtered at vacuum. Methanol was eliminated up precipitation of a solid that was filtered and crystallized from methanol to afford 24. 1H NMR (δ, J(Hz)): 3.33 (d, J 7.2, 4H), 5.58 (d, J 3.18, 4H), 6.51 (d, J 8.5, 4H), 7.54 (d, J 8.5, 4H), 8.12 (s, 2H); IR ν (cm–1) 1600, 3333, 3437. Cytotoxic Activity Tumoral cell lines were supplied by the National Cancer Institute. The cytotoxicity assays were carried out at 5000 to 7500 cells / mL as reported by Skehan et al. and Monks et al using the sulforhodamine B (SRB) protein assay to estimate cell growth [17, 18]. Compounds were dissolved in DMSO which has not effect on the inhibition has shown by the control. The percentage of inhibition of the growth described for all compounds were obtained from three different experiments. The percentage growth was evaluated spectrophotometrically in a Bio kinetics reader spectrophotometer. Daunomicyne and 5-fluorouracyl were used as references. These compounds under the described conditions gave 100 % of inhibition. Each experiment was made two times by gave triplicate. Acknowledgment. We thank CONACyT (32633-E) and DGAPA-UNAM (IN-211601) for financial support. We also thank M.T. Ramírez Apan for obtaining the biological data, R. Patiño, H. Rios, A. Peña, L. Velasco and J. Pérez for technical assistance. Contribution No. 1765 from Instituto de Química, UNAM. References 1. Timofeeva, O. A.; Ryabinin, V. A.; Sinyakov, A. N.; Zakharova, O. D.; Yamkovoy, V. I.; Tarrago-Litvak, L.; Litvak, S.; Nevinsky, G. A. Mol. Biol. (Moscow) 1997, 31, 359-365. 2. Eliadis, A.; Phillips, D. R.; Reiss, J. A.; Skorobogaty, A. J. Chem. Soc. Chem. Commun 1988, 1049-1052. 3. Martínez, R.; Cogordan, J. A.; Mancera, C.; Díaz, M. L. Fármaco 2000, 55, 631-636. 4. Chacón-García, L.; Martínez, R. Eur. J. Med. Chem. 2002, 37, 261-266. 5. For a review of Lexitropsin see: Lown, J. W. Drug Dev. Res. 1996, 34, 145-183. Cytotoxic Evaluation of a Series of Bisalkanoic Anilides and Bisbenzoyl Diamines 6. Ryabinin, V. A.; Sinyakov, A. N.; Soultrait, V. R.; Caumont, A.; Parissi, V.; Zakharova, O.D.; Vasyutina, E. L.; Yurchenko, E.; Bayandin, R.; Litvak, S.; Tarrago-Litvak, L.; Nevinsky G.A. Eur. J. Med. Chem. 2000, 35, 989-1000. 7. Warner, P. M.; Qi, J.; Meng, B.; Li, G.; Xie, L.; El-Shafey, A.; Jones G. B. Bioorg. Med. Chem. 2002, 12, 1-4. 8. Wemer, D. E.; Dervan, P. B. Curr. Op. Struct. Biol. 1997, 7, 355361. 9. Chacón-García, L.; Martínez, R. Eur. J. Med. Chem. 2001, 36, 731-736. 10. Kubicova, L.; Waisser, K.; Kunes, J.; Kralova, K.; Odlerova, Z.; Slosarek, M.; Janota, J.; Svovoda, Z. Molecules 2000, 5, 714-726. 11. Sanna, A.; Repetto, G. Gazz. Chim. Ital. 1927, 57, 777-780. 12. Richmond, B. C. J. Chem. Soc. 1927, 2923-2929. 189 13. Takahashi, H.; Nobuhara, A.; Kimura, H.; Yakugaku Kenkyu 1965, 36, 149-162; Chem. Abstr., 65 (1966), 2875f. 14. Brisson , J.; Gagné, J.; Brisse, F. Can. J. Chem. 1989, 67, 840849. 15. Michio, N.; Atsushi, T. JAPAN 1970, 70, 37,009; Chem. Abstr., 74 (1970) 64086t. 16. Stebemann, C.; Schnitze, J. J. Am. Chem. Soc. 1943, 2126-2128. 17. Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Nat. Cancer Inst. 1990, 82, 1107-1112. 18. Monks, A.; Scudiero, D.; Skehan, P.; Shoemaker, R.; Paull, K.; Vistica, D.; Hose, C.; Langley, J.; Cronise, P.; Vaigro-Wolff, A.; Gray-Goodrich, M.; Campbell, H.; Mayo, J.; Boyd, M. J. Nat. Cancer Inst. 1991, 83, 757-765. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 190-201 Investigación El análisis conformacional a la luz de la teoría topológica de átomos en moléculas. Contribución de la energía atómica a la energía molecular Fernando Cortés Guzmán,a* Jesús Hernández-Trujillo,b* Gabriel Cuevasa* Instituto de Química. Universidad Nacional Autónoma de México. Circuito Exterior Apdo. Postal 70213. Ciudad Universitaria. C.P.04510 Coyoacán, D.F. México. b Departamento de Fisicoquímica. Facultad de Química. Universidad Nacional Autónoma de México. Circuito Interior. Ciudad Universitaria. C.P.04510 Coyoacán, D.F. México. a Recibido el 4 de mayo del 2003; aceptado el 8 de julio del 2003 Dedicado al Prof. Dr. Alfonso Romo de Vivar en homenaje a su trabajo científico Resumen. La teoría de átomos en moléculas permite definir un átomo en una molécula y provee las herramientas para el cálculo de sus propiedades, entre ellas la energía atómica. Esta energía es utilizada en este trabajo para estudiar el comportamiento tanto de los átomos como de algunos grupos funcionales en las barreras rotacionales del etano y de etanos-1,2-disustituidos, en la preferencia conformacional de ciclohexanos y de 1,3-diheterociclohexanos monosustituidos. Este análisis permitió encontrar que los átomos de carbono en los sistemas de etano son los responsables de las barreras rotacionales y que la preferencia conformacional de los ciclos sustituidos es producto de un balance energético entre el anillo y el sustituyente. Este trabajo muestra una nueva manera de abordar el análisis conformacional, a partir de la contribución de la energía atómica a la energía molecular. Palabras clave: átomos en moléculas, barreras rotacionales, etano, preferencias conformacionales, energía atómica, energía molecular. Abstract. The Topological Theory of Atoms in Molecules allows the description of an atom in a molecule and provides the tools to calculate their properties, including atomic energy. Atomic energy is used in this study to analyze the behavior of atoms and functional groups in the rotational barriers of ethane, and 1,2-disubstituted ethanes and in the conformational preference of cyclohexanes and monosubstituted 1,3-diheterocyclohexanes. The results show that the carbon atoms of the ethane systems are responsible for the rotational barriers and that the conformational preference of the substituted cycles is a product of an energetic balance between the ring and the substituent. This study presents a new method to approach conformational analysis using the contribution of individual atomic energies to the molecular energy. Keywords: Atoms in molecules, rotational barriers, ethane conformational preferences, atomic energy, molecular energy. Introducción Mediante la aplicación del programa proaimv [2], se calcula la contribución de cada átomo a la energía molecular. Debido a que la conformación molecular se refleja en la energía de la molécula, es posible, a través del estudio de la contribución atómica a la energía total, determinar el efecto que tiene cada átomo en el arreglo molecular estudiado y por lo tanto la función que este desempeña. El objetivo de este trabajo es analizar la contribución energética de los átomos en las moléculas a la preferencia conformacional de los derivados del etano, ciclohexanos monosustituidos y 1,3-diheterociclohexanos-2-sustituidos. El etano (1, esquema 1) es la molécula orgánica más sencilla que presenta barreras rotacionales, cuyo valor experimental aceptado es de 2.93 kcal/mol [3]. Esta barrera ha sido explicada de diversas formas: por la presencia de repulsiones estéricas entre átomos o entre enlaces en el confórmero eclipsado, por la presencia de hiperconjugación en el confórmero alternado o por balances entre las componentes de la energia potencial molecular [4]. Cuando se sustituye un hidrógeno de cada carbono por otro grupo como metilo, cloro o flúor, se presentan dos barreras rotacionales, la primera donde los dos sustituyentes eclipsados se encuentran frente a frente y la segunda donde cada grupo esta eclipsado con un átomo de La teoría de átomos en moléculas [1] define un átomo en una molécula como una región del espacio delimitada por una superficie S(r) que exhibe la propiedad de cero flujo, del gradiente de ρ lo que significa que S(r) no es cruzada por ningún vector del gradiente de la densidad electrónica ρ(r). Esta condición se expresa en la ecuación 1, donde n(r) representa al vector unitario perpendicular a la superficie en el punto r. ∇ρ (r) • n(r) = 0 para todo punto r en al superficie S(r) La superficie S(r) se obtiene como solución de la ecuación diferencial 1. Como consecuencia, el espacio tridimensional se divide en regiones disjuntas que se identifican como los átomos de la química, separados por superficies interatómicas S(r). Por lo que la teoría de átomos en moléculas define: 1) la contribución atómica a todas las propiedades moleculares y su aditividad y 2) la transferibilidad de átomos y grupos funcionales de una molécula a otra. La definición rigurosa del átomo en una molécula que la teoría topológica de átomos en moléculas provee, permite estimar propiedades atómicas fundamentales como la energía. El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... H H H H H H H H H H H H H H H 191 las propiedades atómicas, pero además, valida el análisis que a continuación se presenta. La inclusión de la correlación electrónica será objeto de análisis en trabajos posteriores. H H H Etano X4 X3 H6 H5 C1 H 6 H5 X3 H7 H8 H6 X3 X4 H7 H8 X3 H5 H 8 H5 H8 H7 X4 H6 H7 X4 Etanos 1,2-disustituidos 1: X = H, 2: X = CH3; 3: X = Cl; 4: X = F 5: R1 = R2 = H, X = Y = CH2 9-ec: R1 = H, R2 = Cl, X = Y = CH2 6: R1 = R2 = H, X = Y = O 10-ax: R1 = Cl R2 = H, X = Y = O 7: R1 = R2 = H, X = Y = S 10-ec: R1 = H, R2 = Cl, X = Y = O 8: R1 = R2 = H, X = O, Y = S 11-ax: R1 = Cl, R2 = H, X = Y = S 9-ax: R1 = Cl, R2 = H, X = Y = CH2 11-ec: R1 = H, R2 = Cl, X = Y = S Esquema 1 hidrógeno. En estos casos, la barreras se han explicado invocando repulsiones de van der Waals [5]. En los casos en los que el grupo es metilo o cloro, el confórmero más estable es el anti mientras que en caso del flúor, el confórmero más estable es el gauche, este comportamiento se conoce como efecto gauche y se ha explicado por la presencia de dos interacciones hiperconjugativas σC-H → σ*H-F en este confórmero [6]. También es conocido que la mayoría de los sustituyentes en el ciclohexano prefieren adoptar la posición ecuatorial y no la axial. Esta preferencia conformacional ha sido atribuida a que en el confórmero axial existe repulsión entre los hidrógenos de las posiciones 3 y 5 y el sustituyente de la posición 1[7]. Por otro lado el efecto anomérico es la preferencia que muestran los sustituyentes electronegativos por asumir la posición axial cuando sustituye en la posición a respecto a un heteroátomo que forma parte del anillo heterocíclico y no la ecuatorial como sucede en el ciclohexano [8]. Hasta ahora, el origen de este efecto conformacional se considera esencialmente diferente al que opera en el ciclohexano. Todas las moléculas que se estudian aquí se presentan en el Esquema 1 y fueron optimizadas por completo a nivel HF/6-311++G (2d, 2p) con el programa Gaussian 94 [9]. Las funciones de onda obtenidas fueron usadas para determinar la energía atómica empleando el programa PROAIMV. En todos los casos se comparó la suma de las energías atómicas con la energía molecular y se obtuvo en general una diferencia menor a 1 kcal/mol, lo que permite comprobar la aditividad de Resultados A. Etano (1) Las posibles explicaciones que se han dado al aumento de energía del confórmero eclipsado o a la disminución de la misma para el confórmero alternado se han relacionado con la estabilización o desestabilización de átomos dentro de la molécula. En el caso de las explicaciones de origen estérico se espera que los átomos de hidrógeno involucrados se desestabilicen, mientras que en el caso de la participación de interacciones hiperconjugativas se espera la estabilización de los mismos átomos. Hasta el momento no existe una metodología experimental que permita medir la energía de los átomos dentro de una molécula. Por fortuna, empleando métodos computacionales, la teoría de átomos en moléculas provee herramientas para conocer la respuesta de los átomos a los cambios conformacionales. Para esto, se estudio al etano manteniendo fijo el ángulo diedro H-C-C-H y dejando libres el resto de las variables. La optimización parcial se realizo cada cinco grados a nivel HF/6-311++G (2d, 2p). El perfil de energía durante la rotación se muestra en la Fig. 1a, donde se observa una barrera de 3.05 kcal/mol, un valor muy cercano al experimental (2.9 kcal/mol). Como se observa, el confórmero eclipsado es de mayor energía que el alternado. Para cada punto en la curva se genera una función de onda, a partir de la cual se obtiene la energía de cada átomo en cada rotámero. En la Fig. 1b se presentan los perfiles de las energías del átomo de carbono y de hidrógeno durante la rotación y en la Tabla 1 se presentan los valores numéricos. El carbono en el confórmero eclipsado presenta una energía de –37.66808 hartrees mientras que el confórmero alternado es de –37.67199, lo que genera una diferencia de 2.45 kcal/mol por cada carbono. Esto es, el confórmero eclipsado se desestabiliza por 4.9 kcal/mol a causa de los átomos de carbono. Por otro lado, el átomo de hidrógeno en el confórmero eclipsado tiene –0.65271 hartrees de energía mientras que el confórmero alternado se determinan –0.65223 hartrees, lo que genera una diferencia de –0.3 kcal/mol. Por lo tanto, el confórmero eclipsado se estabiliza por –1.8 kcal/mol a causa de los seis átomos de hidrógeno. A diferencia de lo que generalmente se encuentra en los libros de texto sobre química orgánica, aquí Tabla 1. Energía del etano (1) (hartrees) y diferencias energéticas (kcal/mol) calculadas a nivel HF/6-311++G (2d, 2p). átomo C H E (0°) E (60°) DE –37.66808 –0.65271 –37.67199 –0.65223 2.45 –0.3 192 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Fernando Cortés Guzmán et al. b) Energía atómica kcal/mol Energía molecular kcal/mol a) 3.5 3 2.5 2 1.5 1 0.5 0 0 5 10 15 20 25 30 35 40 45 50 55 3 C 2 H 1 0 -1 60 0 5 10 15 20 Angulo ( o) 25 30 35 40 45 50 55 60 Angulo ( o) Fig. 1. a) Barrera rotacional del etano. b) Energía atómica de los átomos de carbono e hidrógeno en el etano durante la rotación. Butano b) Butano 25 20 15 10 5 0 -5 1,2-dicloroetano Angulo ( o) 180 160 140 F H5 20 H7 10 0 180 160 140 120 100 80 60 40 -10 20 170 150 130 110 -8 F 90 Cl 70 CH3 -6 50 -4 C 30 0 -2 Energía atómica 0 30 120 d) c) 10 100 Angulo ( o) Angulo ( o) Energía atómica (grupo) 80 60 40 1,2-difluroetano 20 180 160 140 120 100 80 60 40 20 1,2-difluoroetano 0 12 10 8 6 4 2 0 -2 Energía atómica 1,2-dicloroetano 0 Energía molecular (kcal/mol) a) Angulo ( o) Fig. 2. a) Energía molecular relativa de etanos 1,2-disustituidos. b) Energía relativa (kcal/mol) del átomo de carbono en etanos 1,2-disustituidos. c) Energía relativa (kcal/mol) de X en etanos 1,2-disustituidos. d) Energía relativa (kcal/mol) de los átomos del 1,2-difluoroetano. se puede observar que el carbono se desestabiliza cuando la rotación se realiza del confórmero alternado al eclipsado mientras que el hidrógeno se estabiliza ligeramente en el mismo movimiento. Si se presentara una repulsión entre los átomos de hidrógeno o entre los enlaces C-H se observaría la desestabilización de este átomo. En cambio, el perfil de la energía del átomo de carbono reproduce el perfil del cambio en la energía molecular del etano durante la rotación. La principal contribución a la barrera rotacional se debe a los átomos de carbono, 4.9 kcal/mol, estos son los átomos causantes de la barrera rotacional del etano [1]. B. Etanos disustituidos En el caso de los etanos disustituidos se siguió la misma metodología computacional que para el etano. Se realizaron optimizaciones parciales cada 10 grados manteniendo fijo el ángulo diedro X-C-C-X, donde X puede ser un átomo de cloro, flúor o un grupo metilo, y optimizando el resto de las variables. En la Fig. 2a se muestra el perfil de energía molecu- lar durante la rotación de cero a 180 grados del butano, 1,2dicloroetano y 1,2-difluoroetano. En la Tabla 2 se muestran los valores numéricos. En esta figura se observa que en los tres sistemas la diferencia entre el confórmero gauche y el anti es de 1.19 kcal/mol para butano, 1.77 para el 1,2-dicloroetano y de -0.34 para el 1,2-difluoroetano mientras que los valores experimentales son de 0.89 kcal/mol, 1.1 kcal/mol y -0.5 kcal/mol para estos sistemas respectivamente. Con lo anterior se comprueba que los cálculos realizados reproducen el comportamiento experimental de los tres sistemas. La misma concordancia se observa con los valores de las barreras. Aunque hay que tener en cuenta que el principal objetivo de la química computacional no es solo reproducir los hechos experimentales sino proveer la información que la experimentación no puede proporcionar. En la figura 2b se muestra el perfil de energía del átomo de carbono en los tres sistemas. Los perfiles coinciden con el comportamiento molecular durante la rotación. Para el caso donde los sustituyentes son metilo y cloro se observa que el átomo de carbono es más estable en el confórmero anti mien- El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... 193 Tabla 2. Energía del átomo de carbono de los etanos disustituidos 2 a 4 (Hartress) y diferencias energéticas (kcal/mol) calculadas a nivel HF/6311++G (2d, 2p). X (2) CH3 (3) Cl (4) F E (0°) –37.68222 11.9955 –37.61019 15.0031 –37.34202 22.4285 E (gauche°) –37.69633 3.1413 –37.62993 2.6167 –37.37979 –1.2732 tras que en el caso del flúor el átomo de carbono lo es a los 90 grados. En la figura 2c se muestra la respuesta del sustituyente a la rotación. En todos los casos, los sustituyentes contribuyen a la estabilización de la molécula. En el caso del cloro el cambio es del orden de 2 kcal/mol mientras que con el metilo es de 5 kcal/mol y con el flúor 7 kcal/mol. El hecho de que el confórmero en 70 grados sea el más estable para la molécula del 1,2-difluoroetano, se debe a un balance energético entre el átomo de carbono y el átomo de hidrógeno 7 (Figura 2d, ver Esquema 1 para la numeración), que no es antiperiplanar al átomo de flúor vecinal. En la figura 2d se observa que cuando el ángulo es 70 grados el efecto desestabilizador del hidrógeno 7 es menor. Del análisis anterior se puede concluir que el comportamiento de los etanos disustituidos es producto del cambio energético del átomo de carbono durante de la rotación y no se debe a la repulsión entre los sustituyentes. El efecto gauche es ocasionado por el balance entre la energía del carbono y del hidrógeno que no mantiene una relación de antiperiplanaridad con el flúor vecinal. C. Ciclohexano En el ciclohexano (5), las contribuciones a la energía total son: –37.71790 hartrees del átomo de carbono, –0.66530 para el átomo de hidrógeno que asume la posición axial y –0.66302 para el átomo de hidrógeno que asume la posición ecuatorial. El átomo de la posición axial se estabiliza por 1.43 kcal/mol. En el esquema 2 se muestran las numeraciones correspondientes a este y al resto de los sistemas heterocíclicos analizados adelante. La energía del grupo metileno es de –39.04622 hartrees, por lo que después de la suma de la contribución de cada uno de los metilenos, la energía total es de –234.27732 Hartrees. La diferencia con la energía total determinada por el cálculo es de 0.05 kcal/mol, diferencia ocasionada por el error en la integración numérica y que genera un valor razonable para los fines de análisis que persigue este trabajo. Se conoce que al átomo de deuterio prefiere asumir la posición ecuatorial y no la axial por 6.3 cal/mol [10] lo que prueba la mayor estabilidad del átomo de hidrógeno axial que recibe carga por transferencia electrónica del anillo (0.006 e). Otras consecuencias de esto, son el hecho de que el átomo de hidrógeno axial sufre corrimiento a campo alto en el espectro de resonancia magnética nuclear respecto al ecuatorial y que la constante de acoplamiento a un enlace 1JC-Hax sea menor que la correspondiente constante ecuatorial. La diferencia en E (120°) E (180°) –37.69454 4.2652 –37.62555 5.3633 –37.37631 0.9130 –37.70134 0.0 –37.63410 0.0 –37.37776 0.0 energías de punto cero hace que sea mayor la energía de activación necesaria para romper el enlace σC-D que la C-H, la interacción σC-H → σ*C-H puede ser más accesible en el caso de átomo de hidrógeno que para el de deuterio. D. 1,3-dioxano y sistemas relacionados La introducción de los átomos de oxígeno en el 1,3-dioxano (6, Esquema 1) produce un incremento sustancial de las energías de los átomos de carbono que se unen a átomos de oxígeno. La energía del átomo de carbono del metileno de la posición anomérica, que sufre una doble sustitución, se encuentra a 489.7 kcal/mol por arriba de la energía del carbono del ciclohexano. El carbono del metileno que se encuentra sustituido por un solo átomo de oxígeno (C4,6) es 221.90 kcal/mol superior respecto al del metileno del ciclohexano y finalmente, el carbono del metileno de la posición 5 se situó por debajo del metileno de referencia por 20.43 kcal/mol. De hecho el C5 es el único átomo de carbono que proporciona una energía estabilizante al sistema. Por su parte, los átomos de hidrógeno muestran un comportamiento muy interesante. El átomo de hidrógeno H2ax es más estable que el ecuatorial (∆E = 11.15 kcal/mol), siendo a su vez más estable que el átomo de hidrógeno axial del ciclohexano por apenas 0.71 kcal/mol. El átomo H2ec es menos estable que el átomo de hidrógeno ecuatorial del ciclohexano por 9.01 kcal/mol A este tipo de metileno se le denominará normal por la semejanza que tiene en el orden de estabilidad (Hax más estable que Hec) con el metileno del ciclohexano. Para los metilenos C4,6 el átomo de hidrógeno axial se ubica a 2.4 kcal/mol por arriba del hidrógeno del ciclohexano. Este metileno es normal también por el orden de estabilidad de sus átomos de hidrógeno respecto al ciclohexano. Respecto al ciclohexano H4,6ec se ubica a 6.44 kcal/mol. H4,6ax pierde estabilidad respecto a H2ax por 1.91 kcal/mol y H4,6ec es más estable que H2ec por 2.57 kcal/mol. El metileno de C5 es especialmente interesante pues el átomo de hidrógeno axial es el menos estable de los dos hidrógenos de este carbono (∆E = 1.44 kcal/mol), por lo que es el primer caso inverso con respecto al ciclohexano. En este caso el hidrógeno axial sufre una fuerte pérdida de estabilidad, y respecto a Hax del ciclohexano se ubica a 9.83 kcal/mol. El hidrógeno H5ec es más estable que H5ax por 2.05 kcal/mol, pero menos estable que el hidrógeno Hec del ciclohexano por 6.97 kcal/mol. 194 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Fernando Cortés Guzmán et al. Tabla 3. Energía de las moléculas 5 a 11 (Hartress) y diferencias energéticas (kcal/mol) calculadas a nivel HF/6-311++G (2d,2p). 5 6 7 X C2 Y C4 C5 C6 R1 R2 H4ax H4ec H5ax H5ec H6ax H6ec Contribucion de C Contribucion de H suma C+H suma H axial suma H ecuatorial Energía Total calc. suma anillo Anillo sin susti. ∆E calculo–suma (kcal/mol) –37.71790 –226.30740 –7.96992 –235.60564 –3.99180 3.97812 –234.27740 –234.27732 –75.60167 –36.95073 –75.60168 –37.37647 –37.77724 –37.37647 –0.66643 –0.64866 –0.66338 –0.65276 –0.64963 –0.65192 –0.66338 –0.65276 –149.48091 –5.24892 –154.72983 –2.64282 –2.60610 –305.93341 –305.93318 –397.59313 –37.75316 –397.59312 –37.73416 –37.70975 –37.73420 –0.63371 –0.62279 –64744 –0.64024 –0.64669 –0.65554 –0.64744 –0.64025 –150.93127 –5.13410 –156.06537 –2.57528 –2.55882 –951.25166 –951.25162 –75.56022 –37.33889 –397.65709 –37.73722 –37.73878 –37.36389 –0.64982 –0.63655 –0.64800 –0.64039 –0.64769 –0.65353 –0.66354 –0.65290 –150.17878 –5.19242 –155.37120 –2.60905 –2.58337 –628.58852 –628.58851 0.05 0.14 0.03 0.01 –0.66530 –0.66302 9-ax X C2 Y C4 C5 C6 H1ax Hec R1 R2 H3ax H3ec H4ax H4ec 0.63756 H5ax H5ec H6ax H6ec Contribución de C Contribución de H suma C+H suma H axial suma H ecuatorial Energía Total calc. suma anillo –950.48084 Anillo sin susti. ∆E calculo – suma ∆E conformacional 8 –37.72787 –37.65739 –37.72784 –37.71722 –37.71394 –37.71722 –0.65812 –0.65102 –459.72273 –0.64946 –0.65813 –0.65100 –0.65630 –0.66009 –0.63641 –0.66589 –0.65966 –0.65630 –0.66009 226.26148 –7.22606 –233.49812 –3.29474 –3.93132 –693.21060 –233.48754 –950.49619 –693.21027 0.21 0.95 9-ec –37.73089 –37.66397 –37.73089 –37.71530 –37.71563 –37.71528 –0.65422 –0.65256 –0.65044 –459.71381 –0.65422 –0.65256 –0.66291 –0.65714 10-ax 10-ec 11-ax 11-ec –75.64077 –36.71803 –75.64078 –37.39930 –37.76828 –37.39930 –75.61807 –36.77767 –75.61807 –37.38562 –37.76946 –37.38564 –397.57724 –37.67141 –397.57724 –37.74016 –37.70926 –37.74014 –397.57803 –37.66836 –397.57803 –37.73709 –37.71146 –37.73705 –459.78623 –0.62796 –0.64854 –459.75496 –459.69173 –0.61338 –0.62451 –459.67302 –0.64757 –0.64878 –0.65943 –0.64572 –0.63862 – –0.64571 –0.66246 –0.65946 –0.66295 –0.65712 –226.27196 –7.22602 –233.49798 –3.94716 –3.27886 –693.21211 –233.49798 –0.64886 –0.64612 –0.64878 –0.64757 –149.28491 –4.51564 –153.80055 –1.94521 –2.57043 –764.86943 –305.08210 –0.64570 –0.64761 –0.65943 –0.64572 –149.31836 –4.55215 –153.87051 –2.61310 –1.93905 –764.86159 –305.10665 –0.64702 –0.65264 –0.63861 –0.63756 –150.86097 –4.46539 –155.32636 –1.92425 –2.54114 –1410.17273 –0.64480 –0.65262 –0.64571 –0.63641 –150.85396 –4.48617 –155.34013 –2.56073 –1.92544 –1410.16941 –693.21179 0.19 0.0 –764.86833 0.67 0.0 –764.86161 0.01 4.92 –1410.17257 0.10 0.0 –1410.16921 0.13 2.09 El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... 195 Tabla 4. Energía de los tres confórmeros del dimetoximetano y del ditiometilmetano (Hartrees) y diferencias energéticas (kcal/mol) calculadas a nivel HF/6-311++G (2d, 2p). H8 H6 H 12 H6 X H7 Y H1 1 H1 2 6’-gg C1 X C3 Y C5 H6 H7 H8 H9 H10 H11 H12 H13 Energía Total Energía Total calc. ∆E calculo – suma ∆E conformacional H1 1 H1 0 H9 H1 3 H9 Y H1 0 H7 X H11 H1 0 H6 H8 H1 3 H1 3 H9 H8 Y H1 2 X H7 6'; X = O 7'; X = S 6’-ga 6’-aa 7’-gg 7’-ga 7’-aa –37.34031 –75.60142 –36.94529 –75.60141 –37.34032 –0.64619 –0.64894 –0.65406 –0.65965 –0.65966 –0.64894 –0.64620 –0.65404 –268.04655 –268.04643 0.08 0.0 –37.34316 –75.61180 –36.92899 –75.60582 –37.32604 –0.65525 –0.64552 –0.64590 –0.65988 –0.66903 –0.65467 –0.64323 –0.65367 –268.04313 –268.04296 0.10 2.17 –37.32673 –75.61553 –36.91324 –75.61558 –37.32673 –0.65426 –0.64230 –0.65425 –0.66987 –0.66987 –0.65426 –0.64230 –0.65425 –268.03886 –268.03917 0.19 4.56 –37.71170 –397.58213 –37.74114 –397.58219 –37.71171 –0.62889 –0.62722 –0.63646 –0.62828 –0.62827 –0.62723 –0.62889 –0.63646 –913.37098 –913.37057 0.26 0.0 –37.70986 –397.56662 –37.74700 –397.57915 –37.70655 –0.63679 –0.62839 –0.62853 –0.63024 –0.63493 –0.63700 –0.62744 –0.63627 –913.36904 –913.36877 0.17 1.21 De esta manera, la estabilidad de Hax disminuye en la serie C2 > C4,6 > C5 mientras que Hec varia en la serie C4,6 > C5 > C2. El término efecto Perlin normal se asigna a los grupos metileno en donde la constante de acoplamiento sigue el orden 1J 1 C-Hax < JC-Hec, mientras que el término efecto Perlin inverso se usa para definir el orden relativo inverso: 1JC-Hax > 1JC-Hec [11]. En el 1,3-dioxano, el efecto Perlin es normal para los metilenos C2 y C4,6, mientras que es inverso para el metileno C5, hecho que es concordante con el orden de estabilidad relativa determinado para los diferentes átomos de hidrógeno del dioxano. El origen de esta estabilidad está en que los metilenos en C2 y C4,6 del dioxano, sufren la participación de la interacción estereoelectrónica nO → σ*C-Hax, mientras que el efecto Perlin inverso en el metileno C5 se debe a la participación de la interacción homoanomérica nO → *σ*C-Hec como ha descrito Anderson et al. [12]. Es importante destacar el hecho que el efecto de la sustitución del grupo metileno por oxígeno, independientemente de su origen, disminuye paulatinamente con la distancia y que la transferencia electrónica asociada con la interacción hiperconjugativa produce la estabilización del átomo que la recibe. Se sabe que el 1,3-dioxano (6) es más estable que el 1,4dioxano debido a la participación de dos interacciones estereoelectrónicas de tipo n O → σ* C-O pues ambos átomos de oxígeno mantienen un par de electrones no compartido antipe- –37.70586 –397.56666 –37.74700 –397.56666 –37.70582 –0.63635 –0.62712 –0.63576 –0.63756 –0.63755 –0.63575 –0.62713 –0.63636 –913.36592 –913.36558 0.21 3.18 riplanar a un enlace C-O. [13] En un sistema anular como en el 1,3-dioxano, es imposible tener el arreglo conformacional en el que los enlaces C-O se mantengan en conformación anti por lo que no es posible evaluar el efecto que tiene el cambio relativo en la orientación de los pares electrónicos no compartidos. Para evaluar el efecto que tiene la interacción estereoelectrónica nO → σ*C-H, se estudiaron las contribuciones a la energía molecular de los diferentes átomos de los tres confórmeros del dimetoximetano. En la Tabla 4 se incluyen las energías de los confórmeros 6'-g,g, 6'-g,a y 6'-a,a, mismas que guardan una razonable similitud con las previamente descritas [14]. El confórmero g,g es el más estable de los tres. Se acepta que dos interacciones estereoelectrónicas de tipo nO *σ*C-O participan en su estabilización y es el confórmero que se considera análogo al 1,3-dioxano, aunque los grupos metilo no mantienen la misma relación que guardan los metilenos de las posiciones 4 y 6 en este último. En el confórmero a,g, solo puede existir una interacción estereoelectrónica de tipo nO* → σ*C-O, razón por la que este confórmero se encuentra 2.15 kcal/mol por arriba del anterior. Finalmente, el menos estable de todos (a 4.82 kcal/mol respecto al más estable) es el confórmero a,a, en donde no existe posibilidad alguna de interacción nO → σ*C-O. En el confórmero g,g el carbono C3 (anomérico) es el menos estable de los tres átomos de carbono que forman la molécula (∆E = 247.88 kcal/mol), evidenciando el efecto de la 196 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Hax X 5 4 3 1 1 6 X Hec 2 Hec 5 6 Hec Hec Hec Y 2 Hax 3 Hax 4 Hax Esquema 2. Numeración de los derivados del ciclohexano y heterociclohexanos. sustitución por átomos electronegativos. Este átomo de carbono tiene una estabilidad similar a la del carbono anomérico del 1,3-dioxano (∆E = 3.41 kcal/mol), y por lo tanto ambos presentan desestabilización con respecto al carbono del ciclohexano (∆E = 484.82 kcal/mol). Los carbonos 1 y 5 muestran también valores similares al de los metilenos C4,6 del 1,3dioxano, aun cuando la comparación debe hacerse con cuidado, pues un átomo de hidrógeno sustituye al átomo de carbono anular y su contribución es ligeramente desestabilizante con respecto a estos (∆E = 22.69 kcal/mol), y desestabilizante también con respecto al metileno de referencia en el ciclohexano. Por su parte, los átomos de oxígeno muestran una estabilidad similar a la que presentan los oxígenos en el 1,3-dioxano. Estos confórmeros prueban que los efectos de estabilización de los átomos de oxígeno y de desestabilización de los átomos de carbono, principalmente el anomérico, no son sólo producto de la conectividad, sino de la conformación. El átomo de carbono del metileno anomérico pierde estabilidad, al incrementar su energía en 20.11 kcal/mol, pero los átomos de oxígeno se estabilizan por 232.30 kcal/mol. Los carbonos de los grupos metilo pierden estabilidad también (∆E = 8.52 kcal/mol). Por lo tanto, se puede concluir que el efecto anomérico ocasiona una notable estabilización de los heteroátomos a costa de la desestabilización del átomo central del segmento anomérico, en este caso del átomo de carbono y de los átomos de hidrógeno unidos a él. Con respecto a los átomos de hidrógeno del metileno anomérico en el confórmero g,g, cada uno es antiperiplanar a un enlace O-CH3 y a un par electrónico no compartido, a diferencia del metileno anomérico del 1,3-dioxano en el que un átomo de hidrógeno es antiperiplanar a dos pares electrónicos no compartidos y el otro hidrógeno es antiperiplanar a dos enlaces C-H, pues los grupos metilo se mantienen lo más distantes posible en el sistema abierto. Desde el punto de vista energético, este par de átomos no está diferenciado. En el confórmero a,a, cada uno de los átomos de hidrógeno del metileno en cuestión es antiperiplanar a dos pares electrónicos no compartidos y susceptible de estabilizarse a través de dos interacciones nO → σ*C-H, por lo que el cálculo predice apropiadamente, además de una energía similar para ambos átomos de hidrógeno, una mayor estabilidad (∆E = 6.41 kcal respecto a los hidrógenos anoméricos del confórmero g,g). En el confórmero a,g, uno de los átomos de hidrógeno del metileno anomérico es susceptible de estabilizarse a través de dos interacciones nO → σ*C-H, mientras que el otro sólo puede estabilizarse a través de una interacción de este tipo. Estos Fernando Cortés Guzmán et al. átomos de hidrógeno son diferenciados por el cálculo, siendo aquel que cuenta con dos interacciones nO → σ*C-H, más estable respecto al que solo tiene una (∆E = 5.73 kcal/mol). Los átomos de hidrógeno de los tres confórmeros pueden clasificarse en tres tipos: — Aquellos que son antiperiplanares a pares electrónicos localizados (que no toman parte en interacciones estereoelectrónicas de tipo nO → σ*C-O); por lo tanto son antiperiplanares a buenos donadores y se estabilizan. — Aquellos que mantienen una relación de antiperiplanaridad con enlaces C-O con baja capacidad donadora y por lo tanto no ganan estabilidad. — Aquellos átomos de hidrógeno que son antiperiplanares a heteroátomos que se encuentran participando en interacciones de tipo nO → σ*C-O. En el confórmero g,a los átomos de hidrógeno con los números 6,7 y 8 (ver estructuras de la tabla 4 para la numeración) pertenecen al grupo metoxi donador (nO) mientras que los hidrógenos 11,12 y 13 pertenecen al grupo metoxi aceptor (σ* C-O ). Los átomos de hidrógeno H 6 y H 11 son antiperiplanares a pares electrónicos no compartidos que no participan en interacciones estereoelectrónicas y se ven estabilizados, al igual que H13. El hidrógeno H7 es antiperiplanar al par electrónico no compartido que participa en la interacción nO → σ*O-C, por lo que no se ve estabilizado en igual magnitud que su análogos anteriores. Finalmente los hidrógenos H8 y H12 al ser antiperiplanares a enlaces C-O se encuentran desestabilizados. En el confórmero g,g, los dos grupos metilo son equivalentes por simetría, sin embargo, existen los tres tipos de átomos de hidrógeno descritos anteriormente. Los átomos de hidrógeno H8 y H13 se estabilizan fuertemente debido a que son antiperiplanares a un par de electrones localizado. Los átomos de hidrógeno H6 y H12 pierden estabilidad debido a que se ubican en forma antiperiplanar a enlaces C-O, y finalmente los átomos de hidrógeno H7 y H11 se estabilizan poco, debido a que son antiperiplanares a enlaces que participan en la interacción nO → σ*O-C. En el tercer confórmero, los átomos de hidrógeno 8, 13, 6 y 11 se encuentran dispuestos en forma antiperiplanar a pares electrónicos no compartidos, por lo que se estabilizan respecto a los hidrógenos H7 y H12, los que al mantenerse en forma antiperiplanar a enlaces C-O, se encuentran por arriba en términos de energía. E. 1,3-ditiano y sistemas relacionados En el 1,3-ditiano (7, esquema 1) se observa la estabilización del carbono anomérico con respecto al carbono del ciclohexano (∆E = 22.13 kcal/mol), lo que contrasta con el 1,3-dioxano analizado previamente. Los átomos de carbono C4,6, que también ganan estabilidad, se ubican 10 kcal/mol por debajo del carbono de referencia pero, a diferencia del ciclohexano, el El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... carbono 5 es el menos estable de todos los átomos de carbono (∆E = 5.11 kcal/mol). La diferencia entre la energía de los átomos de hidrógeno del metileno C2 es de 6.85 kcal/mol, la de los metilenos C4,6 es de 4.52 kcal/mol y la de C5 es de 5.55 kcal/mol. La energía relativa de los átomos de hidrógeno unidos a C2 y C4,6 indica que el átomo de hidrógeno axial es más estable que el correspondiente ecuatorial, mientras que para C5 se invierte el orden de estabilidad. Es bien sabido que todos los metilenos en el 1,3-ditiano presentan un efecto Perlin de tipo inverso [15]. Sin embargo, el orden relativo de estabilidad de los átomos de hidrógeno de los diferentes metilenos es análogo al del dioxano (6), que muestra, como ya se dijo, efectos normales. Esto hace pensar que el efecto que da origen al efecto Perlin no solo ocasiona la estabilización del átomo de hidrógeno axial, sino que, cuando el carbono asociado al metileno gana estabilidad como en el 1,3-ditiano, en lugar de perderla como en el 1,3-dioxano, se observará un efecto Perlin inverso, pero de forma contraria, si el átomo de carbono se desestabiliza, entonces se observará un efecto Perlin normal. En el caso en que la estabilidad de los átomos de hidrógeno se invierte y el carbono se mantenga estabilizado, entonces se observará un efecto Perlin inverso (metileno 5 del 1,3-dioxano). Nuevamente, el problema del efecto anomérico en el segmento S-C-S no puede ser abordado con sólo estudiar al 1,3ditiano, ya que es necesario tener un modelo en el que la interacción nS* → σ*C-S se pueda modificar. Para ello, se abordó el estudio de los tres confórmeros del ditiometilmetano (7’, Tabla 4). Como se puede apreciar en la Tabla 4, y en analogía al dimetoximetano, el confórmero g,g es más estable que el g,a por apenas 1.21 kcal/mol y 3.18 kcal/mol más estable que el a,a, poniendo de manifiesto la menor capacidad donadora del átomo de azufre [16]. Como se puede apreciar en el confórmero g,g, análogo al 1,3-ditiano (7), el carbono anomérico se estabilizar con respecto al carbono del ciclohexano (∆E = 14.58 kcal/mol), pero es menos estable que en el caso del metileno del 1,3-ditiano (∆E = 7.54 kcal/mol). Nuevamente, el efecto de la sustitución es estabilizante para el carbono anomérico del análogo de 7 y no como en el caso del análogo de 6 para el que es desestabilizante. Los átomos de azufre pierden estabilidad con respecto a los átomos correspondientes en el 1,3-ditiano (∆E = 6.90 kcal/mol), comportamiento compartido por los átomos de carbono de los grupos metilo, aunque este comportamiento debe considerarse con cuidado dada la sustitución de un átomo de hidrógeno por un átomo de carbono. Al parecer, la pérdida de la estabilidad en el átomo de azufre ocasiona la estabilización de los átomos vecinos. En el confórmero aa, no existe la posibilidad de que se produzcan interacciones estereoelectrónicas siendo el orbital aceptor el σ*C-S. En este sistema, los átomos de azufre pierden estabilidad respecto al confórmero g,g (∆E = 9.71 kcal/mol), pero el carbono anomérico se estabiliza. Los átomos de carbono de los grupos metilo también sufren estabilización. En el confórmero g,a, un segmento mantiene la interacción estereoelectrónica nS → σ*C-S pero el otro no. La energía de los átomos involucrados da evidencia de esto. El átomo de 197 azufre donador se estabiliza respecto al confórmro a,a, pero es menos estable que el mismo átomo en el confórmero g,g. Sorprendentemente, el átomo de azufre que sólo es aceptor en el segmento, es isoenergético con respecto al confórmero a,a, al igual que el carbono anomérico, pero el átomo de azufre que es donador se estabiliza por 7.84 kcal/mol, aproximándose energéticamente al carbono que es donador. El comportamiento de los átomos de hidrógeno de esta molécula es idéntico al de los átomos de hidrógeno del dimetoximetano, tomando sólo en consideración la menor capacidad donadora del átomo de azufre, y la estabilidad inversa de los átomos de carbono de los metilos. En el caso del dimetoximetano la estabilidad ganada por los átomos de oxígeno está ligada a la desestabilización que sufre de carbono anomérico, y ambos factores se incrementan en ausencia de la interacción estereoelectrónica nO → σ*C-O. En el caso del sistema con azufre, el comportamiento es contrario. El carbono anomérico se estabiliza en la medida en que se desestabiliza el heteroátomo involucrado en el segmento. Es claro que el mecanismo de estabilización no puede ser el mismo, aunque tienen en común su dependencia con la estereoquímica, ya que el comportamiento de la sustitución en el segmento es diferente. Eso puede deberse a que el átomo de carbono y el de azufre tienen una electronegatividad similar a diferencia de la relación entre átomo de oxígeno-carbono. F. 1,3-oxatiano Surge ahora la pregunta con respecto al comportamiento que experimentaría una molécula en donde los heteroátomos O y S se enfrentasen en un mismo segmento anomérico, pues si operan los mismos mecanismos de estabilización que en las moléculas ya analizadas, se esperaría la desestabilización del átomo de oxígeno con estabilización del carbono anomérico por participación de nO → σ*C-O, y la desestabilización del carbono anomérico con estabilización del átomo de azufre por participación del mecanismo nS → σ*C-S. Posiblemente este hecho contradictorio sea la explicación del porqué la síntesis de 1,3-oxatianos es complicada y los rendimientos son bajos cuando se le prepara por condensación de un hidroxitiol y un compuesto carbonílico o al hecho de que el 2-metoxi-1,3-ditiano no se pueda preparar por condensación entre ortoformiato de metilo y el 3-hidroxipropanotiol, además de que es responsable de que los dos átomos de hidrógeno del metileno anomérico del 1,3-oxatiano sean sincrónicos en el espectro de 1H RMN [17]. El compuesto 8 combina al átomo de oxígeno y el de azufre en un mismo segmento. El átomo de oxígeno pierde la estabilidad que gana en el dioxano 6 (∆E = 26.01 kcal/mol), mientras que el átomo de azufre gana estabilidad (∆E = 40.14 kcal/mol). La estabilidad del carbono anomérico es intermedia con respecto a la de los carbonos anoméricos de 6 y 7, la cual es menor a la del ciclohexano (∆E = 237.84 kcal/mol). Este comportamiento sería el esperado a partir de la observación hecha con anterioridad, y pone en evidencia la participación de ambos mecanismos en la estabilización de 8. 198 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Por su parte, el metileno C4 (α respecto al átomo de azufre) es más estable que el carbono metilénico del ciclohexano, y de energía muy similar a la del C5 (∆E = 12.12 y 13.10 kcal/mol respectivamente). Finalmente, el metileno C6 (α al átomo de oxígeno) muestra una mayor desestabilización respecto al carbono de referencia (∆E = 222.14 kcal/mol). Desde el punto de vista experimental, se ha descrito un efecto Perlin compensado para C2 y C4, inverso para C5 y normal para C6, lo que aunado a lo descrito con anterioridad permite establecer relación entre el efecto Perlin y la estabilidad del átomo de carbono que constituye el metileno. Así, el metileno C6 muestra Hax más estable que Hec (∆E = 6.68 kcal/mol) y al átomo de carbono que lo soporta más estable respecto al del metileno del ciclohexano. Por otra parte, el metileno C 4 muestra Hax 4.78 kcal/mol menos energía que Hec, pero el carbono es más estable que el átomo de carbono de referencia. Un hecho similar sucede en C5, en donde el carbono metilénico se encuentra más estabilizado con respecto a la referencia, pero Hax es menos estable que Hec (∆E = 3.66 kcal/mol), lo que produce un efecto compensado. Finalmente, el carbono anomérico de la posición 2 en la molécula 8 es menos estable que el carbono metilénico del ciclohexano y Hax es más estable que Hec (∆E = 8.32 kcal/mol). En estas condiciones, una mayor desestabilización del C2 ocasionaría un efecto Perlin normal, pero esta no es suficiente y se produce un efecto Perlin compensado. G. Sistemas sustituidos La energía total del confórmero axial del clorociclohexano axial (9-ax) al nivel de teoría empleado en este trabajo es de –693.21060 hartrees, mientras que la del confórmero ecuatorial (9-ec) es de –693.21210 hartrees, lo que lleva a una diferencia de 0.95 kcal/mol a favor del confórmero ecuatorial, de acuerdo con la observación experimental. (Signo negativo en la tabla 3). (∆E = 0.80 kcal/mol [18]. La introducción del átomo de cloro ocasiona la desestabilización del átomo de carbono que lo soporta (C2) siendo más afectado por la sustitución del confórmero axial que el ecuatorial, con una diferencia energética de 4.13 kcal/mol. Los carbonos C3 y C5 ganan estabilidad en el confórmero ecuatorial (∆E = 1.91, 1.06 kcal/mol), mientras que los carbonos C4,6 son más estables en el axial por 1.2 kcal/mol. Nuevamente, el efecto del sustituyente disminuye al aumentar la distancia. Por su parte, el átomo de cloro es considerablemente más estable en el confórmero axial que en el ecuatorial por 5.6 kcal/mol, hecho que contrasta con la idea general de que en el ciclohexano la preferencia es por la posición ecuatorial. Los hidrógenos de las posiciones 1,3,5 son más estables en ambos confórmeros que sus correspondientes hidrógenos ecuatoriales. Para el confórmero 9-ax: ∆EC1,3 = 4.47, ∆EC5 = 3.90; para 9-ec: ∆EC1,3 = 1.04, ∆EC5 = 1.88 kcal/mol (la numeración no sigue la regla establecida, ver el Esquema 1). Sólo en el confórmero 9ax para el metileno C4,6 el orden de estabilidad se invierte, siendo más estable el hidrógeno ecuatorial con ∆E = 2.38 kcal/mol. ∆E para C4,6 en el confórmero 5-ec es de 3.60 kcal/mol. Fernando Cortés Guzmán et al. Un argumento para justificar la preferencia del sustituyente por la posición ecuatorial en el ciclohexano, es el de la repulsión estérica que puede sufrir con los átomos de hidrógeno syn-diaxiales de las posiciones 4,6. El análisis de la deslocalización electrónica permite establecer que los átomos involucrados no presentan una superficie interatómica, por lo que no hay una interacción enlazante entre ellos [19]. Posiblemente un efecto electrónico de naturaleza hiperconjugativa podría ser el origen de ésto. El hecho es que el átomo de hidrógeno de la posición axial del metileno C3 en el confórmero 9-ax se ubica 5.4 kcal/mol con respecto al hidrógeno en C5-eq. La saturación de la participación del enlace C2-Hax en la interacción σC-H → σ*C-Cl y no con σ*C3-Hax puede ser el origen de esto. Con respecto al ciclohexano, los átomos de carbono C2, C4,6 y C5 se desestabilizan en el confórmero 9-ax por 37.97, 0.43 y 2.48 kcal/mol respectivamente, mientras que C1,3 se estabiliza por 6.24 kcal/mol. En el confórmero ecuatorial todos los átomos de carbono se desestabilizan con las diferencias: 33.84 para C2, 1.63 para C4,6 y 1.42 para C5 y 8.15 para C1,3 kcal/mol respectivamente. De manera contraria a lo que experimenta el átomo de cloro, el átomo de hidrógeno de la posición C2 es más estable en el confórmero ecuatorial que en el axial con ∆E = 0.61 kcal/mol. Los átomos de hidrógeno de las posiciones 2ax, 3ec, 4ax y 4ec son más estables en el confórmero axial por 2.45, 1.85, 2.15 y 0.13 kcal/mol respectivamente, mientras que el de la posición 2-ec es más estable en el confórmero ecuatorial por 0.98 kcal/mol. En el ciclohexano, los seis átomos de carbono contribuyen a la estabilidad del sistema con –226.30740 Hartress, energía que está 28.82 kcal/mol por debajo de la contribución de los seis átomos de carbono del isómero 9-ax y 22.24 kcal/mol con respecto a 9-ec. El anillo completo en el derivado 9-ax contribuye con –233.49798 Hartrees (∆E = 6.55 kcal/mol), lo que permite establecer que la preferencia conformacional observada proviene del balance entre la estabilización lograda por el átomo de cloro, que prefiere asumir la posición axial, y el anillo de ciclohexano, que es más estable cuando sufre la sustitución ecuatorial. En este caso, la contribución del anillo domina y la preferencia observada es la ecuatorial. El cálculo a nivel HF/6-311++G(2d,2p) establece que el confórmero 10-ax es más estable que su análogo 10-ec por prácticamente 5 kcal/mol. Como se ha visto en este análisis, la introducción de uno o dos átomos electronegativos ocasiona un aumento en la energía del átomo de carbono que los soporta. La introducción del átomo de cloro en la posición anomérica del dioxano ocasiona una pérdida adicional de estabilidad del átomo. Con respecto al 1,3-dioxano (6) el átomo C2 del confórmero axial (10-ax) se desestabiliza por 146.02 kcal/mol, mientras que el mismo átomo en el confórmero 10ec se desestabiliza por 108.60 kcal/mol, poniendo de manifiesto que la desestabilización es aditiva con respecto a la incorporación de átomos electronegativos, y que el efecto El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... desestabilizante es mayor para el confórmero con el átomo de cloro axial que para el ecuatorial (∆E = 37.41 kcal/mol). Por su parte, la diferencia energética entre los átomos de cloro es de 19.62 kcal/mol, contribuyendo más a la estabilidad molecular el átomo de cloro axial que el ecuatorial. El átomo de hidrógeno de la posición anomérica es más estable cuando adopta la conformación axial que la ecuatorial (∆E = 12.91 kcal/mol). El átomo de cloro en el 1,3-dioxano axial es 39.85 kcal/mol más estable que el átomo de cloro axial en el ciclohexano, mientras que para el átomo de cloro ecuatorial la diferencia es también a favor del dioxano por 25.82 kcal/mol. Los átomos de oxígeno juegan un papel fundamental en este caso. La diferencia energética entre los átomos de oxígeno de 10-ax y 10-ec es de 14.24 kcal/mol en favor del confórmero axial. De esta forma, los tres átomos electronegativos que sustituyen en la posición anomérica aportan mayor estabilidad al confórmero axial que al ecuatorial, a pesar de ocasionar una fuerte desestabilización en el carbono anomérico. Con respecto al 1,3-dioxano, cada átomo de oxígeno del confórmero 10-ax se estabiliza por 24.54 kcal/mol con la introducción de un átomo de cloro, mientras que en 10-ec la estabilización es tan solo de 10.29 kcal/mol. El efecto estabilizante de la sustitución por un átomo electronegativo también afecta a los átomos de carbono C4,6 y lo hace con más intensidad en el confórmero 10-ax (∆E = 14.33 kcal/mol) que en el 6-eq (∆E = 5.74 kcal/mol), pero esta sustitución produce desestabilización con respecto al ciclohexano. Esta sustitución es desestabilizante para el carbono del metileno C5 con respecto a 10, por 2.56 kcal/mol para 10ax y 4.88 kcal/mol para 10-ec. Los átomos de hidrógeno de las posiciones 4,6-ax son más estables que los correspondientes ecuatoriales en el confórmero 10-ec, (∆E = 8.60 kcal/mol) pero el orden se invierte en el confórmero 10-ax, en forma similar a como sucede en 9ax. (∆E = 0.76 kcal/mol). H4ax en el confórmero 10-ec es 7.44 kcal/mol más estable que en 10-ax, lo que implica que está operando un mecanismo similar al que opera en el clorociclohexano. En el metileno C5 del confórmero 10-ax, Hax es más estable que Hec, siguiendo el comportamiento normal de un metileno en ciclohexano, pero el orden de estabilidad se invierte en 10-ec (∆E = 1.72 y 1.20 kcal/mol respectivamente) en donde la interacción homoanomérica no compite con la interacción nO → σ*C-Cl. La contribución a la estabilidad total de los átomos de carbono e hidrógeno del anillo del 1,3-dioxano en el confórmero axial es de –153.80055 Hartrees, mientras que la del confórmero ecuatorial es de –153.87051 (∆E = 43.90 kcal /mol). Cuando se incorporan los átomos de oxígeno, se produce la estabilización del confórmero ecuatorial, pero la diferencia disminuye sustancialmente: 6-ax: –305.08210; 6-ec: –305.10665; ∆E = 15.41 kcal/mol. El átomo de cloro, más estabilizante en el confórmero axial que en el ecuatorial como se indicó (∆E = 19.62 kcal/mol), invierte la preferencia conformacional observada. 199 Los cálculos predicen mayor estabilidad del confórmero 11-ax con respecto al 11-ec, por 2.09 kcal/mol. Este valor es consistente con el esperado pues se predice menor al del 2 cloro-1,3-dioxano [20]. La introducción del átomo de cloro en la posición anomérica del 1,3-ditiano ocasiona severos cambios con respecto a los sucedidos en el análogo con oxígeno. Por ejemplo, los átomos de azufre son más estables en el confórmero ecuatorial, pero el metino anomérico es más estable en el axial. Mientras que en los sistemas relacionados con el dioxano los heteroátomos de 10-ax y de 10-ec se estabilizan con respecto a 6, en 11-ax y 11-ec son menos estables con respecto al 1,3-ditiano (7). Además, el átomo de cloro, que se estabiliza en el dioxano con respecto al ciclohexano, se desestabiliza en el ditiano. Sin embargo, nuevamente el átomo de cloro es más estable en el confórmero axial de que en el ecuatorial (∆E = 11.74 kca/mol). Esta diferencia es de 19.76 en 10, mientras que a diferencia de 10, el átomo de azufre en 11 es apenas más estable en el confórmero ecuatorial que en el axial (∆E = 0.5 kcal/mol). La diferencia energética entre los carbonos anoméricos es de 1.95 kcal/mol a favor del confórmero axial a diferencia del comportamiento del mismo metileno en 10. La diferencia energética de la contribución de todos los átomos de carbono exceptuando los de cloro (contribución del anillo), es de 9.63 kcal/mol a favor del confórmero ecuatorial, pero la estabilidad ganada por el átomo de cloro decide la tendencia conformacional. Este hecho también sucede en el dioxano, pero no en el ciclohexano, en donde la estabilidad ganada por el átomo de cloro no es suficiente para contrarrestar la estabilidad del anillo a ser sustituido en la posición ecuatorial. El átomo de cloro tiene la tendencia natural a ocupar la posición axial debido a que se estabiliza, independientemente del tipo de anillo, ocasionando desestabilización del carbono que sustituye. Este hecho obliga a redefinir el término Efecto Anomérico, ya que, al menos para el caso del cloro, la tendencia de este átomo en el ciclohexano también es por la posición axial. Es importante destacar que este análisis le da un origen común al efecto anomérico y a la preferencia conformacional del ciclohexano. El anillo es más estable cuando se le sustituye en posición ecuatorial, pero no el heteroátomo, ya que su estabilización se produce a expensas de la desestabilización anular. El efecto anomérico se observa cuando la estabilización del sustituyente tiene la capacidad de contrarrestar la desestabilización del heterociclo. En este sentido, el efecto anomérico es estabilizante, pues cuando opera el anillo no pierde sustancialmente la estabilidad al ocasionar la estabilización del heteroátomo. En los sistemas que incorporan átomos de oxígeno, todos los heteroátomos sustituyentes en la posición anomérica se estabilizan, mientras que el carbono anomérico sufre una fuerte desestabilización (la energía del C2 en el confórmero 6ax es la más elevada de todas la determinadas en este estudio).En el derivado del 1,3-ditiano se produce la estabilización del carbono anomérico, ocasionando la desestabilización de los sustituyentes heteroatómicos vecinos (después del C2 del 1,3-ditiano, el C2 de 11-ax es el carbono anomérico más 200 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) estable del sistema), pero esta desestabilización afecta más a 11-ec que a 11-ax. Conclusiones La teoría topológica de átomos en moléculas permite descomponer la energía molecular total en sus componentes atómicas gracias a su definición de un átomo en una molécula, de manera tal que es posible explicar en forma satisfactoria observaciones experimentales como la preferencia conformacional y el efecto Perlin. La barrera rotacional en el etano se produce por un cambio en la energía del átomo de carbono. Esta energía es mayor en el confórmero eclipsado que en el alternado, mientras que el hidrógeno presenta el comportamiento inverso, es más estable en el confórmero eclipsado que en el alternado. Esto contrasta con la idea de que la barrera se deba a repulsiones de los átomos de hidrógeno. Cuando se estudia a los etanos 1,2-disustituidos se encuentra que el causante de la barrera es también el átomo de carbono, ya que su energía presenta el mismo comportamiento que el perfil de energía de la molécula, mientras que los sustituyentes siempre tienen una contribución estabilizante. Este también es el origen del efecto gauche en el 1,2difluoroetano. El efecto anomérico en segmentos que presentan átomos de oxígeno, se caracteriza por la desestabilización del átomo de oxígeno con estabilización del carbono anomérico por participación de la interacción estereoelectrónica nO → σ*C-O, mientras que por una interacción de naturaleza opuesta, pero también de tipo estereoelectrónica, se espera desestabilización del carbono anomérico con estabilización del átomo de azufre. Ambos mecanismos muestran una notable dependencia de la estreoquímica molecular. Los átomos electronegativos tienen la tendencia a estabilizarse más cuando ocupan la posición axial y no la ecuatorial, no siendo necesaria la presencia de átomos electronegativos en la red anular para conseguir este efecto. Por esta razón, se debe revisar la definición del efecto anomérico, pues el átomo de cloro muestra preferencia por la posición axial (sustituyente electronegativo) independientemente si el anillo es heterocíclico o carbocíclico. La estabilización del sustituyente se produce a costa de la estabilidad del anillo, que, si presenta un mecanismo de estabilización interna, no perderá demasiada estabilidad (efecto anomérico) y entonces se observará el confórmero axial, mientras que cuando no existe este mecanismo, la desestabilización del anillo dominará a la estabilización del sustituyente y se observará la preferencia por la posición ecuatorial. Fernando Cortés Guzmán et al. Referencias 1. Bader, R. F. W. Atoms in Molecules - A Quantum Theory; Oxford University Press 1990. 2. Biegler-Koing, F. W.; Bader, R. F. W.; Tang, T. H. J. Comput. Chem. 1982, 3, 317-328 3. Hirota, E.; Saito, S.; Endo, Y. J. Chem. Phys. 1979, 71, 11831187. 4. Pitzer, R. M. Acc. Chem. Res. 1983, 16, 207-210. Goodman, L.; Pophristic, V.; Weinhold, F. Acc. Chem. Res. 1999, 32, 983-993. Isaacs, N. S. Physical Organic Chemistry; Longman Scientific 1987. Murcko, M. A.; Castejon, H.; Wiberg, K. B. J. Am Chem. Soc. 1988, 110, 8029-8038. Reed, E. A.; Weinhold, F. Israel J. Chem. 1991, 31, 277-285. Bader, R. F. W.; Cheeseman, J. R.; Laidig, K. E.; Wiberg, K. B.; Breneman, C. J. Am. Chem. Soc. 1990, 112, 6530-6536. 5. Lowe, J. P. Prog. Phys. Org. Chem. 1968, 6, 1-80. Compton, D. A. C.; Montero, S.; Murphy, W. F. J. Phys. Chem. 1980, 84, 3587-3591. Stidham, H. D.; During, J. R. Spectrochim. Acta 1986, 42A, 105-111. Eliel, E. L.; Allinger, N. L.; Angyal, S. J.; Morrison, G. Conformational Analysis; Wiley: New York 1967. Mizushima, S. Structure of molecules and internal rotation; Academic 1954. Murcko, M. A.; Castejon, H.; Wiberg, K. B. J. Chem. Phys. 1996, 100, 16162-16168. 6. Wolfe, S. Acc. Chem. Res. 1972, 5, 102. Juaristi, E.; Cuevas, G. The anomeric effect, CRC Press. 1994. 7. Bushweller, C. H. In conformational behavoir of six member rings, Juaristi, E. Ed. VCH, 1995. Wiberg, K. B.; Hammer, J. D.; Castejon, H.; Bailey, W. F.; DeLeon, E. L.; Jarret, R. M. J. Org. Chem. 1999, 64, 2085-2095. Wiberg, K. B.; Castejon, H.; Bailey, W. F.; Ochterski, J. J. Org. Chem. 2000, 65, 1181-1187. Wiberg, K. B. J. Org. Chem. 1999, 64, 6387-6393. 8. Kirby, A.J. The Anomeric Effect and Related Stereoelectronic Effects at Oxygen. Springer Verlag: Berlin, 1983. Szarek, W.A.; Horton, D., Eds. Anomeric Effect: Origin and Consequences; ACS Symposium Series No. 87; American Chemical Society: Washington, D.C. 1979. Juaristi, E.; Cuevas, G. The Anomeric Effect. CRC Press: Boca Raton, FL. 1994. Juaristi, E., Cuevas, G. Tetrahedron, 1992, 48, 5019-5087. Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry, Pergamon: New York, 1983. Thatcher, G.R.R., (Ed), The Anomeric Effect and Associated Stereoelectronic Effects, ACS Symposium Series 539; American Chemical Society: Washington, 1993. 9. Gaussian 98 (Revision A.x), M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. HeadGordon, E. S. Replogle and J. A. Pople, Gaussian, Inc., Pittsburgh PA, 1998. 10. Anet, FAL.; Kopelevich, M. J. Am. Chem. Soc. 1986, 108, 13551356. Anet, F.A.L., O’Leary, D.J. Tetrahedron Lett. 1989, 10591062. 11. Perlin, A.S.; Casu, B. Tetrahedron Lett. 1969, 2921-2924, Cuevas, G.; Juaristi, E. Vela, A. J. Phys. Chem. A. 1999, 103, 932-937. Juaristi, E.; Cuevas, G. Vela, A. J. Am. Chem. Soc. 1994, 116, 5796-5804. Juaristi, E.; Cuevas, G.; Vela, A. J. Mol. Struct. (Theochem) 1997, 418, 231-241. Juaristi, E.; Cuevas, G. Tetrahedron Lett. 1992, 33, 1847-1850. El análisis conformacional a la luz de la teoría topológica de átomos en moléculas... 12. Anderson, J.E.; Bloodworth, A.J.; Cai, J.; Davies, A.G.; Tallant, N.A. J. Chem. Soc., Chem. Commun. 1992, 1689-1691. Anderson, J.E.; Bloodworth, A.J.; Cai, J.; Davies, A.G.; Schiesser, C.H. J. Chem. Soc., Perkin Trans. 2, 1993, 1927-1929. 13. Dávalos, J.Z.; Flores, H.; Jiménez, P.; Notario, R.; Roux, M.V.; Juaristi, E.; Hosmane, R.S.; Liebman, J.F. J. Org. Chem. 1999, 64, 9328-9336 14. Westiuk, N.H.; Laidig, K.E.; Ma. J. Application of Quantum Theory of Atoms in Molecules to study of Anomeric Effect in Dimethoxymethane. En The Anomeric Effect and Associated Stereoelectronic Effects. Thatcher G.R.J. (Ed). ACS Symposium Series No. 539. Washington, D.C. 1992. p. 176. 15. Cuevas, G.; Juaristi, E. J. Am. Chem. Soc. 2002, 124, 1308813096. 201 16. Salzner, U.; Schleyer, P.v.R. J. Am. Chem. Soc. 1993, 115, 10231-10236. Salzner, U.; Schleyer, P.v.R. J. Org. Chem. 1994, 59, 2138-2155. 17. Juaristi E.; Cuevas G.; Flores-Vela A. Tetrahedron Lett. 1992, 33, 6927-6930 18. Eliel, E.L.; Wilen, S.H. Stereochemistry of Organic Compounds, Wiley: New York, 1994, 19. Wiberg, K. B.; Castejon, H.; Bailey, W. F.; Ochterski, J. J. Org. Chem. 2000, 65, 1181-1187 20. Se ha descrito que el efecto anomérico que involucra átomos de la tercera fila de la tabla periódica debe ser menor a los que generan los átomos de la segunda. Schleyer, P.v.R.; Gemís, J.E.; Spitznagel, G.W. J.Am. Chem. Soc. 1985, 107, 6393-6394. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 202-206 Investigación Synthesis and Properties of 2-diazo-1-[2-(thiophen-2-ylmethoxy)phenyl]-ethanone. Intramolecular Cyclization Through a Carbenoid Intermediate Erick Cuevas Yañez,1 Abraham Arceo de la Peña,1 Joseph M. Muchowski2 y Raymundo Cruz Almanza1* Instituto de Química, UNAM. Circuito Exterior, Ciudad Universitaria, Coyoacán, 04510, México, D.F. Tel. (52) (55) 5622-4408; Fax: (52) (55) 5616-2217. E-mail: [email protected] 2 Roche Biosciences Department of Chemistry, Roche Palo Alto, 3431 Hillview Avenue, Palo Alto, CA 94304-1320, USA. 1 Recibido el 24 de junio del 2003; aceptado el 10 de julio del 2003 Dedicated to Dr. Alfonso Romo de Vivar Romo Resumen. Se describe la síntesis de la 2-diazo-1-[2-(2-tiofenilmetoxi)-fenil]-etanona (1) la cual involucra la reacción de sustitución nucleofílica aromática del (2-tiofenil) metóxido de sodio con 2-flurorobenzaldehído, la oxidación en condiciones suaves del grupo aldehído al ácido carboxílico y posteriormente la transformación a α-diazocetona a través de un anhídrido carbónico-carboxílico como intermediario de reacción. El tratamiento del compuesto 1 con cantidades catalíticas de acetato de rodio (II) da la tiofenilmetil benzofuranona 10 por medio de una transposición sigmatrópica [2,3] del iluro de oxonio 9 derivado del carbenoide de rodio que se postula como intermediario de reacción. Palabras clave: α-diazocetona, ciclización, carbenoide de rodio. Abstract. The synthesis of 2-diazo-1-[2-(thiophen-2-ylmethoxy)phenyl]-ethanone (1) is described. The procedure involves an aromatic nucleophilic substitution between the sodium 2-thiophenyl methoxide and 2-fluorobenzaldehyde, the subsequent mild-condition oxidation to the corresponding carboxylic acid, and final transformation through carbonic-carboxylic anhydride intermediate to the a-diazoketone. Treatment of compound 1 with catalytic amounts of rhodium (II) acetate gives the thienylmethyl benzofuranone 10 which comes from the [2,3] sigmatropic rearrangement of the oxonium ylide 9 derived from the rhodium carbenoid that is postulated as reaction intermediate key. Key words: α-diazoketones, cyclization, rhodium carbenoid. Introduction structure a thiophene ring and a diazo group, in such a way they could react to each other to form 5 or 6 membered rings when a carbenoid from a rhodium or copper salt were generated. For this purpose we believed molecule 1 was appropriate for the study and we proceeded to prepare it. Initially, we reacted the sodium salt of 2-hydroxymethylthiophene 2 with 2-fluorobenzaldehyde 3 to obtain the alkyl aryl ether 4 in 73 % yield through a nucleophilic aromatic substitution. Previously, the substitution on 2-fluorobenzaldehyde was reported as a process that readily occurs [4]. The 2hydroxymethylthiophene used in this part was prepared by The diazo group chemistry has taken a new interest as a consequence of the development of catalytic methods with transition metals that convert diazoketones in valuable tools in organic synthesis with several applications in homologations, X-H bond insertions, ylide formations, and reactions with alkenes and aromatic compounds [1]. Our interest has especially been focused in aryl and heteroaryl diazoketone synthesis since there are few examples that show the reactivity of this kind of compounds. Recently Doyle and coworkers [2] took advantage of the furan reactivity to prepare a macrocycle by the intramolecular insertion of the rhodium carbenoid from a w-furanyl diazoketone. On the other hand, Capretta and coworkers [3] reported that the treatment of diverse diazoalkanoyl thiophenes with catalytic rhodium (II) acetate gave the intramolecular cyclization compounds as major reaction products, and they isolated 5 and 6 mermbered-fused rings to thiophene. O S O 1 Fig. 1. O O Results and discussion OH H S Taking the preceding reactions and as a part of a project that involves a study of the ω-diazoalkanoylthiophene reactivity, we were interested in synthesize a molecule which had in their N2 2 Scheme 1 3 F NaH H DMF O 4 S Synthesis and properties of 2-Diazo-1-[2-(thiphen-2-ylmethoxy)-phenyl]... O O AgNO3, NaOH H O OH THF-H 2O S S O 5 4 Scheme 2 O O OH O 5 N-methylmorpholine S O ClCO2Et,ether 0°C, 15 min O O OEt S 6 CH2N2/ether O O N2 S 1 Scheme 3 Table 1. Crystal data for compound 1. Crystal data Empirical formula Formula weight Crystal color Crystal system Space group a, Å b, Å c, Å Volume, Å3 Z Density (calcd.), g / cm3 Absorption coefficient, (mm–1) F(000) Crystal Size (mm) C13H10N2O2S 258.29 Colorless needle Orthorrombic Pbca 5.5049(2) α = 90° 17.914(2) β = 90° 24.750(1) γ = 90° 2440.8(3) 8 1.406 2.327 1072 0.56 × 0.12 × 0.07 Data collection Temperature, K Radiation, l (Å) θ min, max,° Index ranges Reflections Collected Independent reflections Observed reflects[I > 2.0 σ (1)] 292(2) 1.54178 1.50,56.50 –5 [h [5,-19 [k[19,-26[1[26 3015 1508 (Rint = 0.0830) R1 = 0.0723, wR2 = 0.2159 Refinement Data-to-parameter ratio R, wR2 G.O.F. Largest diff. peak,hole,e Å–3 1508 / 0 / 164 R1 = 0.0723, wR2 = 0.2159 0.996 0.264,-0.288 203 sodium borohydride reduction of thiophene-2-carboxaldehyde, which is commercially available [5]. The next step in the synthetic route consisted on the aldehyde group oxidation in molecule 4 to the corresponding carboxylic acid. For this purpose, aldehyde 4 was treated with silver oxide, which was prepared in situ from the reaction of silver nitrate and sodium hydroxide. The oxidation reaction was performed at room temperature during 48 h in THF. After work up, benzoic acid 5 was obtained in 38 % yield. Finally, the thiophen-2-ylmethoxy benzoic acid 5 became diazoketone 1 by the reaction with N-methylmorpholine and ethyl chloroformate in ether at 0 °C during 15 min, and subsequent addition of an excess of ethereal diazomethane (10:1) to the carbonic-carboxylic anhydride 6 which is postulated like the acylating agent. This technique has been successful in the synthesis of diverse pirrolyl diazoketones [6], and in this case gave quantitative yield. Compounds 1, 2, 4 and 5 were characterized by the conventional spectroscopic techniques, and especially the diazoketone 1 was a crystalline solid which was studied by X-ray diffraction. The compound structure is represented in Fig. 2 and some important crystallographic data are in Table 1. The X-ray study revealed some interesting data about the compound structure. For example, we observed that the angle formed between the nitrogen atoms of the diazo group is 177°, which indicates a strong tendency toward the lineal structure given by a resonant structure that favors two consecutive double bonds, one between C2 and N1 and the other between N1 and N2. Additionally, the bond distance between C1 and C2 is shorter than other distances with sp2 atoms, such as the C1-C3 bond distance. This can be explained by the resonance effects that should exist between the carbonyl group and the diazo group. In Table 2 some relevant angle and bond distances from the X-Ray study are presented for molecule 1. In the Fig. 3 we also could appreciate the packing cell representation for compound 1 that shows a great separation among the ring substituents. This suggests a strong repulsion between the ortho substituents which is probably due to the steric effect, and is reflected in the dihedral angle (C1-C3-C4O2) with 8 ° instead of 0 °. Therefore, the substituents are not coplanar. It is important to mention that there are not reported examples where a diazoketone has been studied and characterized by X-ray diffraction. Thus, the present work contributes with new data about the space distribution that presents this functional group. Once prepared, diazoketone 1 reacted with catalytic amounts of rhodium (II) acetate. At the moment, when we carried out the reaction, we considered two possible reaction routes: the first consisted on the C-H insertion on the benzylic position of the carbenoid species 7 that it is formed from the diazoketone 1 and the catalyst, which would give thienylchromanone 8 as the major product; while in the second route we intended the formation of an oxonium ylide (9) after a sigmatropic rearrangement [2, 3] would afford benzofuranone 10. In this respect, we were not able to predict the reaction course 204 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) Erick Cuevas Yañez et al. Table 2. Selected bond distances (Å) and angles (°) for 1. Bond Distance (Å) S1-C13 O1-C1 O2-C9 N1-C2 C1-C3 C3-C4 C3-C8 C4-C5 S1-C10 O2-C4 N1-N2 C1-C2 C9-C10 C11-C12 1.689(8) 1.221(9) 1.440(8) 1.316(9) 1.483(9) 1.409(9) 1.407(10) 1.370(9) 1.707(9) 1.366(9) 1.117(8) 1.433(12) 1.487(10) 1.419(11) Bond Angle (°) C13-S1-C10 N2-N1-C2 O1-C1-C3 N1-C2-C1 C8-C3-C1 O2-C4-C5 C5-C4-C3 C11-C10-C9 C9-C10-S1 C4-O2-C9 O1-C1-C2 C2-C1-C3 C8-C3-C4 C4-C3-C1 O2-C4-C3 O2-C9-C10 C11-C10-S1 92.0(5) 177.0(10) 121.6(8) 115.0(8) 116.4(6) 124.0(7) 120.6(7) 126.1(8) 122.3(6) 117.4(5) 121.0(6) 117.3(7) 117.9(6) 125.7(7) 115.4(6) 108.2 111.5(6) Fig. 2. ORTEP representation for compound 1. because there are not overwhelming evidences about the reaction mechanism. The McKervey group [7-9] found that varying the catalyst they could modify the course of this kind of reactions to obtain both chromanones and benzofuranones. However, the five-membered ring formation and subsequent Stevens rearrangement is more common in these processes [10]. Treatment of diazoketone 1 with rhodium (II) acetate in dichloromethane at room temperature under inert atmosphere gave a reaction product whose physical and spectroscopic constants did not correspond to those expected for chromanone 8 as have been reported previously [11, 12]. In this way, spectroscopic data indicated that the compound obtained was thienylmetylbenzofuranone 10. Additionally, Gefflaut and Périe [13] informed a similar cyclization process when the diazoketone derived from 2-benzyloxybenzoic acid reacted with catalytic rhodium (II) acetate. Therefore, in this case the carbenoid 7 derived from the diazoketone 1 reacted in an intramolecular way with the phenoxy oxygen to generate a five-membered ring and also an oxonium ylide which is kinetically favored and is rearranged later to the benzofuranone 10. In conclusion, this work presents the first synthesis of the 2-diazo-1-[2-(thiophen-2-ylmethoxy)-phenyl]-ethanone 1 and its intramolecular cyclization to the benzofuranone 10, which expands the possibilities to carry out reactivity studies with transition metals and to synthesize some derivatives from this molecule. Experimental section The starting materials were purchased from Aldrich Chemical Co. and were used without further purification. Solvents were distilled before use, ether and tetrahydrofuran (THF) were dried over sodium using benzophenone as indicator. Diazomethane was prepared from N-methyl-N-nitroso-p-toluenesulfon- amide (Diazald ®) using a minimum amount of water and ethanol as co-solvent, and dried over KOH pellets before use. Silica gel (230-400 mesh) and neutral alumina were purchased from Merck. Silica plates of 0.20 mm thickness were used for thin layer chromatography. Melting points were determined with a Fisher-Johns melting point apparatus and they are uncorrected. 1H and 13C NMR spectra were recorded using a Varian Gemini 200, chemical shifts (d) are given in ppm relative to TMS as internal standard (0.00). For analytical purposes, mass spectra were recorded on a JEOL JMS-5X 10217 in the EI mode, 70 eV, 200 °C via direct inlet probe. Only molecular and parent ions (m/z) are reported. IR spectra were recorded on a Nicolet Magna 55-X FT instrument. For X-Ray diffraction studies, crystals of compound 1 were obtained by slow evaporation of a dilute ethanol solution, and reflections were acquired with a Nicolet P3 / F diffractometer. Three standard reflections every 97 reflections were used to monitor crystal stability. The structure was solved by direct methods, missing atoms were found by difference-Fourier synthesis, and refined on F2 by a full-matrix least-squares procedure using an isotropic displacement parameters using SHELX-97. Synthesis and properties of 2-Diazo-1-[2-(thiphen-2-ylmethoxy)-phenyl]... 205 washed with water (150 mL), dried over Na2SO4 and the solvent was removed in vacuo. Purification by column chromatography (SiO2, hexane / AcOEt 9:1) yield a colorless oil (2.67 g, 73 %). IR (CHCl3, cm–1). 3105, 1687. 1H NMR (CDCl3, 200 MHz) δ 5.35 (s, 2H), 7.07 (m, 3H), 7.35 (dd, 2H), 7.55(m, 1H), 7.85 (m, 1H), 10.5 (s, 1H). MS [EI+] m/z (%): 218 [M]+ (10), 97 [M-C7H5O2]+ (100). 2-(Thiophen-2-ylmethoxy)benzoic acid (5). Silver nitrate (8.4 g, 49.5 mmol) was added to a solution of sodium hydroxide (2.97 g, 74.2 mmol) in water (75 mL). The mixture was added to a solution of compound 4 (2.67 g, 12.3 mmol) in THF (14 mL) and the resulting mixture was stirred for 48 h at room temperature. The mixture was filtered and acidified to pH = 4, and the product was extracted with ethyl acetate (3 × 100 mL). The organic phase was dried over Na2SO4 and the solvent was removed in vacuo. The product was purified by crystallization (1.1 g, 38 %). m.p. 100 °C. IR (CHCl3, cm–1). 3511, 1664. 1H NMR (CDCl3, 200 MHz) δ 5.46 (s, 2H), 7.05 (m, 3H), 7.29 (dd, 1H), 7.45 (m, 1H), 7.6 (m, 1H), 8.22 (dd, 1H). MS [E I+] m/z (%): 234 [M]+ (7), 97 [M-C7H5O3]+ (100). Fig. 3. Representation of the packing cell of compound 1. 2-Hydroxymethylthiophene (2). To a suspension of sodium borohydride (0.339 g, 8.92 mmol) in absolute ethanol (9 mL) a solution of 2-thiophenecarboxaldehyde (2 g, 17.85 mmol) in ethanol (25 mL) was added maintaining the temperature below 25 °C. The resulting mixture was then heated at 50 °C during 1 h. The solvent was removed in vacuo, and water (50 mL) was added, the solution acidified with diluted HCl (10 %) to pH = 5. The product was extracted with ether (3 × 50 mL), the organic phase was dried over Na2SO4 and the solvent was removed in vacuo to yield a colorless oil (1.93 g, 95 %), which was used without additional purification. IR (CHCl3, cm–1). 3434, 2930, 1668. 1H NMR (CDCl3, 200 MHz) δ 2.25 (s, 1H), 4.79 (s, 1H), 6.98 (m, 2H), 7.33 (m, 1H). MS [EI+] m/z (%): 114 [M]+ (100), 113 [M - H]+ (40). 2-(Thiophen-2-ylmethoxy)benzaldehyde (4). To a suspension of sodium hydride (0.97 g, 20.3 mmol) in DMF (17 mL) a solution of 2-hydroxymethylthiophene 2 (1.93 g, 16.9 mmol) in DMF (17 mL) was added at 0 °C, the resulting mixture was stirred under a nitrogen atmosphere at room temperature for 15 min. The mixture was cooled at 0 °C and 2-fluorobenzaldehyde (2.2 g, 17.8 mmol) was added. The resulting mixture was stirred for additional 15 minutes at room temperature. The reaction was quenched by addition of water (100 mL) and diluted HCl (10 %) to pH = 5. The aqueous phase was extracted with ether (3 × 75 mL), the organic phase was 2-Diazo-1-[2-(thiophen-2-ylmethoxy)-phenyl]-ethanone (1). An ice-cold solution of the acid 5 (0.28 g, 1.2 mmol) in freshly distilled ether (2 mL) was treated successively with ethyl chloroformate (0.14g, 1.3 mmol) and N-methylmorpholine (0.12 g, 1.2 mmol), the mixture was stirred under nitrogen atmosphere for 15 min at 0 °C, then an ether solution of diazomethane (12 mmol) from N-methyl-N-nitroso-4-toluenesulfonamide was added at 0 °C. A vigorous evolution of nitrogen occurred, and the mixture was allowed to warm to room temperature overnight. The solvent was removed in vacuo and the product was purified by column chromatography (SiO2, hexane / AcOEt 9:1) to yield a yellow crystalline product (100 %).m. p 70 °C. IR (CHCl3, cm–1). 3139, 2104, 1762. 1H NMR (CDCl3, 200 MHz) δ 5.27 (s, 1H), 5.46 (s, 2H), 7.05 (m, 3H), 7.29 (dd, 1H), 7.45(m, 1H), 7.6 (m, 1H), 8.22 (dd, 1H). MS [EI+] m/z (%): 258 [M]+ (5), 120 [C7H4O2]+ (100). 2-(Thiophen-2-ylmethyl)benzofuran-3-one (10). A solution of the diazopropanone 1 (0.3096 g, 1.2 mmol) in dry CH2Cl2 O O O N2 S S O 7 Rh2(OAc)4 -Rh2(OAc)4 H -N2 1 O Rh Rh2(OAc)4 O S 8 -N2 O O 7 Scheme 4 O Rh O -Rh2(OAc)4 +O S 9 S O 10 S 206 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) (10 mL) was stirred with Rh2(OAc)4 (2 mg) under nitrogen atmosphere at room temperature. After 2 h, the mixture was evaporated in vacuo and purified by column chromatography (SiO2, hexane / AcOEt 9:1) to yield a white solid (0.118 g, 43 %). m.p. 45 °C. IR (CHCl3, cm–1). 2923, 1716, 1612. 1H NMR (CDCl3, 200 MHz) δ 3.30 (dd, 1H), 3.62 (dd, 1H), 4.78 (dd, 1H), 6.95-7.22 (m, 4H), 7.60-7.80 (m, 3H). MS [EI+] m/z (%): 230 [M]+ (15), 121 [C7H5O2]+ (100), 97 [C5H5S]+ (75). Acknowledgments Financial support from CONACyT (no. 27997E) is gratefully acknowledged. The authors would like to thank Rocío Patiño, Angeles Peña, Javier Pérez and Rubén A. Toscano for their technical support. Erick Cuevas Yañez et al. References 1. Doyle, M. P.; McKervey, M.A.; Ye, T.; Modern Synthetic Methods using Diazocompounds: from Cyclopropanes to Ylides, John Wiley & Sons: New York, 1998. 2. Doyle, M.P., Chapman, B.J.; Hu, W.; Peterson, C.S.; McKervey, M.A.; Garcia, C.F.; Org. Lett. 1999,1, 1327-1329. 3. a)Frampton, C.S.; Pole, D.L.; Yong, K.; Capretta, A. Tetrahedron Lett. 1997, 38, 5081-5084. b) Yong, K.; Salim, M.; Capretta, A. J. Org. Chem. 1998, 63, 9828-9833. 4. Yeager, G.W.; Schissel, D.N. Synthesis 1995, 28-30. 5. Gronowitz, S.; Liljefors, S. Chem. Scr., 1978-79, 13, 39-45. 6. Jefford, C.W.; Kubota, T.; Zaslona, A. Helv. Chim. Acta 1986, 69, 2048-2061. 7. Pierson, N.; Fernández-García, C.; McKervey, M. A.; Tetrahedron Lett. 1997, 38, 4705-4708. 8. Ye, T.; Fernandez-García, C.; McKervey, M. A.; J. Chem. Soc. Perkin Trans. 1, 1995, 1373-1379. 9. Ye, T.; McKervey, M.A. J. Chem. Soc. Chem. Commun. 1992, 823-824. 10. Hodgson, D. M.; Pierard, F. Y. T. M.; Stupple, P.A. Chem. Soc. Rev. 2001, 30, 50-61. 11. De, M.; Majundar, D. P.; Kundu, N. G. J. Indian Chem. Soc. 1999, 76, 665-674. 12. Corvaisier, A. Bull. Soc. Chim. Fr. 1962, 528-535. 13. Gefflaut, T.; Périe, J. Synth. Commun. 1994, 24, 29-33. Revista de la Sociedad Química de México, Vol. 47, Núm. 2 (2003) 207-209 Investigación Estudio fitoquímico de Salvia uruapana† René Manjarréz, Bernardo A. Frontana-Uribe y Jorge Cárdenas* Instituto de Química, Universidad Nacional Autónoma de México, Circuito Exterior Ciudad Universitaria, Coyoacán 04510, México D.F. Tel. +52 (55) 5622-4413; Fax. +52 (55) 5616-2217; E-mail: [email protected] Recibido el 20 de junio del 2003; aceptado el 14 de julio del 2003 En homenaje a los 50 años de vida académica del Dr. Alfonso Romo de Vivar Resumen. De las partes aéreas de Salvia uruapana se aislaron dos diterpenos, salviafaricina y tonalensina ambos previamente descritos en Salvia tonalensis, así como 7-O-luteolina diglucósido peracetilado. La mezcla de los ácidos oleanólico y ursólico se aisló de las fracciones de baja polaridad. Palabras clave: Salvia uruapana; Labiatae; Diterpenos; neo-clerodano; 5-10-seco-neo-clerodano; flavona; triterpenos. Abstract: From the aerial parts of Salvia uruapana were isolated two diterpenic compounds previously isolated from Salvia tonalensis: salvifaricin and tonalensin, as well as the peracetylated 7-O-luteolin diglucoside. The mixture of oleanolic and ursolic acids was also isolated from the low polarity fractions. Keywords: Salvia uruapana; Labiatae; Diterpenos; neo-clerodane; 5-10-seco-neo-clerodane; flavone; triterpenoids. Introducción Discusión de resultados El género Salvia, miembro de la familia Labiatae, consta de aproximadamente 900 especies en el mundo. En México existen más de 300 salvias que se encuentran predominantemente en bosques de pinos-abeto y encino por encima de los 1000 m de altura, lo que hace de México uno de los países con mayor diversidad botánica en este género [1]. De los estudios fitoquímicos de especies americanas del género Salvia se han obtenido diterpenos con esqueleto de tipo clerodano, abietano y pimarano, además de los esqueletos modificados como riacofano y tilifolano, que se propone provienen de precursores clerodánicos. Se han obtenido otros compuestos como flavonoides, ácidos triterpénicos y β-sitosterol por mencionar algunos otros. Esta gran riqueza y diversidad fitoquímica, aunada a la abundancia de especies de este género en nuestro país, ha alentado la búsqueda de compuestos con propiedades biológicas interesantes en plantas de este género. Algunas especies vegetales pertenecientes a este género se han utilizado con fines medicinales por sus propiedades antitumorales, bactericidas, bacteriostáticas, carminativas, entre otras [2]. Continuando con estudios fitoquímicos realizados en plantas del género Salvia endémicas de México [3, 4], en este trabajo se presentan los resultados del estudio de Salvia uruapana (Fern.) Labiatae. Esta planta crece en la zona central de México, encontrándose abundantemente en los estados de Michoacán, Jalisco y Colima. Es una hierba con flor azul intenso, de talla moderada (40-60 cm) que crece en zonas altas (> 1300 msnm) y húmedas. El producto blanco cristalino mostró en el espectro de IR las señales características de un furano monosubstituido (1505 y 875 cm –1) y de una γ-lactona α,β-insaturada (1755, 1670 cm –1 ). Con la espectroscopia de RMN y experimentos bidimensionales RMN 1H-RMN 13C, se logró la determinación de la estructura 1 para este compuesto. Este ha sido reportado previamente de la partes aéreas de Salvia farinacea asignándole el nombre de salvifaricina [9]. Se observaron claramente las señales reportadas como típicas para este compuesto en el espectro de RMN 1H, como son las generadas por la sustitución β del anillo de furano, el sistema de metilo secundario entre C-8 y C-17, el sistema A-B del metileno de la γ-lactona α,β-insaturada y el protón cetálico de C-20. En el primer reporte de 1 se describe parcialmente la espectroscopía de RMN y es en un reporte reciente [5] donde se confirman las O 16 O 15 12 O 20 1 9 10 O O 17 8 O 4 O 18 19 O O 1 AcO AcO AcO O AcO OAc No 1768 del Instituto de Química, UNAM. O O OAc 3 †Contribución OAc OAc O O AcO 2 O OAc O Fig. 1. Compuestos aislados de Salvia uruapana. 208 Rev. Soc. Quím. Méx. Vol. 47, Núm. 2 (2003) asignaciones y se completan los datos espectroscópicos faltantes, mismos que concuerdan correctamente con los obtenidos para el compuesto 1. De las aguas madres de donde se obtuvo el producto anterior, se logró aislar por cristalización otro producto que mostró tener un espectro de RMN 1H y RMN 13C complicado. Al variar la temperatura, este último mostró cambios importantes por lo que se sospechó la presencia de confórmeros. Mediante el estudio de cristalografía de Rayos-X de un monocristal (Fig. 2), se logró obtener la estructura 2 para el compuesto aislado. Este producto ha sido reportado previamente del estudio fitoquímico de Salvia tonalensis y se denominó tonalensina [10]. Los datos reportados en el estudio cristalográfico y los obtenidos en este estudio concuerdan, así como las señales reportadas para los espectros de RMN 1H y RMN 13C [6]. La salvifaricina podría derivarse de la tonalensina vía la fusión de los anillos A y B mediante un reacción electrocíclica del compuesto 2 permitida térmicamente [7]. Con el fin de verificar esta hipótesis, la tonalensina se sometió a reflujo durante 48 h en diferentes disolventes monitoreando la reacción por ccf. En tolueno (p.eb. 110 °C) y 1,1,2,2-tetracloroetano (p.eb. 142 °C) no hubo reacción; en decalina (mezcla de isómeros p.eb. 183 °C) se observó descomposición del producto, pero ninguno de los productos observados en la placa correspondió a la salvifaricina. Esto demostró que el simple calentamiento del producto no induce la reacción y se requiere la presencia de otro tipo de catalizadores, tal vez del tipo enzimáticos, que se podrían encontrar en la planta. La acetilación de la fracción de 20 % metanol en acetato de etilo permitió obtener un compuesto peracetilado. Este compuesto mostró claramente en RMN 1 H el patrón de señales para dos sistemas aromáticos, uno de ellos tetra-substituido y el otro tri-substituido, además del protón singulete característico de las flavonas. Diez grupos hidroxilo fueron identificados mediante las correspondientes señales de los acetatos (δ 1.99-2.40). Las glucosas fueron identificadas mediante las dos señales en RMN 13C de los carbonos CH2 base de oxígeno y los dos dobletes en RMN 1H típicos del protón anomérico del sistema disacárido. Los sistemas obser- Fig. 2. Estructura obtenida por rayos X de la tonalensina (2). René Manjarréz et al. vados son característicos de la luteolina 7-O-diglucósido, compuesto frecuentemente encontrado en los productos naturales [8]. El compuesto mostró un pico molecular en 1030 uma que confirmó a la luteolina con un disacárido de dos glucosas. S. uruapana pertenece a la sección Angulatae del subgénero Calosphace y S. tonalensina y pertenecen a la sección Polystachyae del mismo subgénero [1]. El hecho de haber encontrado los mismos productos reportados en dos secciones diferentes podría indicar una relación botánica muy cercana de ambas plantas. Parte experimental La Salvia uruapana (Fern.) fue obtenida por Bernardo A. Frontana Uribe y Dagoberto Alavés en el camino de Teretán hacia Zirimícuaro, Edo. de Michoacán, México en noviembre de 1999. El especimen fue identificado por la Biol. Irene Díaz del Instituto de Biología UNAM y se depositaron dos ejemplares en el herbario del mismo Instituto con el registro MEXU 967718 y 967719. Los puntos de fusión no están corregidos y fueron determinados en un equipo Fisher-Johns. Los espectros de IR se obtuvieron con un espectrofotómetro Nicolet Magna 750. Las espectrometrías de masas de baja resolución se obtuvieron con la técnica de impacto electrónico a 70 eV en un equipo Jeol JMS-AX 505. Las espectrometrías de alto peso molecular se obtuvieron con la técnica FAB+ con un equipo Jeol JMS-SX 102A. Los experimentos de 1H RMN (300 MHz) y 13C RMN (75 MHz) se obtuvieron con un equipo Varian Unity 300 y con TMS como estándar interno empleando como disolvente deuterocloroformo. Las cromatografías de placa fina (CCF) se realizaron en hojas de aluminio precubiertas con sílica gel (Macherey-Nagel Alugram Sil G / UV254). Las cromatografías Flash se realizaron empleando sílica gel (Merck 60 0.0300.075 mm) y las cromatografías al vacío empleando sílica gel para cromatografía en placa fina (Merck 60). Extracción y aislamiento 2.98 kg de planta seca se sometieron a maceración con acetona, se evaporó el disolvente obteniendo 21.5 g de residuo acetónico. Se cromatografió en columna al vacío empacada con sílice en una proporción de 1 a 10 con respecto al peso del residuo obtenido. La elución se efectuó iniciando con hexano, y mezclas hexano-acetato de etilo de polaridad creciente, acetato de etilo y finalmente con mezclas acetato de etilometanol de polaridad creciente hasta un 20 % de metanol. La mezcla de los ácidos triterpénicos ursólico y oleanólicos se separó de las fracciones con polaridad 20-40 % de AcOEt. De la fracción obtenida con 50 % de AcOEt en hexano y por recristalización por par de disolventes con CH2Cl2 y éter (1:1), se purificó la salvifaricina (1). De las aguas madres de esta misma fracción y con recristalización por par de disolventes con CH2Cl2 y éter (1:1) se aisló la tonalensina (2). De la frac- Estudio fitoquímico de Salvia uruapana 209 ción obtenida con 25 % de hexano en AcOEt, mas los restos de la fracción anterior, se obtuvo la mezcla de los ácidos ursólico y oleanólico con un p.f. 225-228 °C. Finalmente, mediante la acetilación con anhídrido acético y piridina de la fracción de 20 % metanol en acetato de etilo, se obtuvo el derivado peracetilado de la luteolina 7-O-diglucósido (3). 619 (2), 550 (2), 522 (2), 413 (5), 371 (31), 331 (42), 289 (12), 169 (86) 154 (55), 136 (55), 127 (28), 109 (109), 84 (19), 77 (18), 69 (19), 55 (23), 43 (100). Salvifaricina (1): p.f. 214-216 °C (lit. p.f. 214-215°C [9]), IR (CHCl 3) ν max cm –1: 3050, 2964, 2943, 2904, 1755, 1670, 1579, 1505, 1465, 1238, 1164,1052, 1008, 980, 875. EM m/z (abundancia relativa) M+ 340 (100), 282 (22), 259 (26), 244 (23), 217 (44), 189 (22), 163 (100), 135 (72), 95 (97), 81 (61), 77 (25), 55 (26), 39 (16). Los autores agradecen a las siguientes personas por su ayuda en la obtención de los datos espectroscópicos: Javier Pérez, Nieves Zavala, Alejandrina Acosta, Rocío Patiño y Alfredo Toscano; a Irene Díaz por la clasificación del especimen vegetal y a Carmen Márquez por su ayuda en la separación por cromatografía preparativa. Este trabajo fue parcialmente financiado con el proyecto CONACyT J34873-E. Tonalensina (2): p.f. 192-194 °C (lit. p.f. 191-193 °C [10]), IR (CHCl3) νmax cm–1: 1750, 1635, 1502, 1465, 1350, 1313, 1148, 1022, 946, 875. EM m/z (abundancia relativa) M+ 340 (73), 322 (13), 294 (42), 279 (22), 265 (24), 217 (59), 201 (54), 185 (50), 171 (45), 141 (42), 128 (54), 115 (53), 94 (100), 81 (78), 77 (53), 65 (33). Luteolina 7-O-[β-D-Glucopiranosyl-D-Glucopiranósido] (3): p.f. 121-124 °C; IR (CHCl3) νmax cm–1: 3100, 2940, 2877, 1756, 1645, 1617, 1428, 1370, 1119, 1072, 1039. 1H RMN CDCl3 δ J (Hz): 6.58 (s, 1H, H-3), 6.71 (d, 1H, J = 2.6, H-6), 7.0 (d, 1H, J = 2.6, H-8), 7.77 (d, 1H, J = 2.2, H-2’), 7.36 (d, 1H, J = 8, H-5’), 7.71 (d, 1H, J = 8, 2.2, H-6’), 5.33-4.9 (m, 6H), 4.73 (d, 1H, J = 2), 4.33-3.66 (m, 7H), 2.43-1.99 (10s, 30H); 13C RMN CDCl3 δ 167.2 (C-2), 102.3 (C-3), 176.1 (C4), 160.5 (C-5), 100.8 (C-6), 167.9 (C-7), 98.3 (C-8), 158.3 (C9), 108.9 (C-10), 129.8 (C-1’), 121.5 (C-2’), 144.7 (C-3’), 150.6 (C-4’), 124.2 (C-5’), 124.5 (C-6’), 20.37 q, 20.57 q, 21.09 q, 29.17 q, 29.66 q, 61.5 t, 61.88 t, 68.01 d, 68.15 d, 71.05 d, 71.9 d, 72.0 d, 72.7 d, 74.05 d, 109.6 d, 112.6 d, 160.3 s, 169.3 s, 169.4 s, 169.6 s, 169.8 s, 170.3 s, 170.5 s. EM FAB+, m/z (abundancia relativa): [M++1] 1031 (28), 989 (12), Agradecimientos Referencias 1. Epling, C. Repert. Spec. Nov. Regni Veg. 1939, 110, 1-383. 2. Esquivel, B.; Calderón, J.S.; Sánchez, A.A., Ramamoorthy, T.P.; Flores, E.A., Domínguez, R.M.; Rev. Latinoamer. Quím. 1996, 24, 44-64. 3. Rodríguez-Hahn, L.; Esquivel, B.; Cárdenas, J. En: Secondary metabolites from Mexican plants: Chemistry and Biological, Properties, Rodríguez-Hahn, L. Ed., Signpost, New Deli, India, 1996, p. 19. 4. Rodríguez-Hahn, L.; Esquivel, B.; Cárdenas, J. En: Phytochemistry of Medicinal Plants, Arnason, J.T. Ed. Plenum Press, New York EUA, 1995, Cap. 12, p. 311. 5. Rodríguez, B. Mag. Reson. Chem. 2001, 39, 150-154. 6. Ortega, A.; Maldonado, E.; Díaz, E.; Reynolds W. F. Spectrochim. Acta A 1998, 54, 659-670. 7. March, J. Advanced Organic Chemistry, 4a Ed. John Wiley and Sons, USA, 1992, p.1110 8. Imperato,F.; Nazzaro, R. Phytochemistry 1996, 41, 337-338. 9. Rodríguez, B.; Pascual, C.; Savona G. Phytochemistry 1984, 23, 1193-194. 10. Toscano, R. A.; Maldonado, E.; Ortega, A. J. Chem. Crystal. 1996, 26, 239-242.