FACULTAT DE FARMÀCIA DEPARTAMENT DE FARMACOLOGIA

FACULTAT DE FARMÀCIA

DEPARTAMENT DE FARMACOLOGIA

Nuevas isoquinoleínas bioactivas

inspiradas en la naturaleza

New bioactive isoquinolines inspired by nature

TESIS DOCTORAL

presentada por:

Laura Moreno Gálvez

Valencia, Septiembre 2013



D. Diego M. Cortes Martínez, catedrático del Departamento de Farmacología

de la Universitat de València y Dña. Nuria Cabedo Escrig, Colaborador Científico

Adjunto del Centro de Ecología Química Agrícola,

CERTIFICAN:

Que el trabajo presentado por la Lda. Laura Moreno Gálvez, titulado:

“Nuevas isoquinoleínas bioactivas inspiradas en la naturaleza", ha sido realizado en

el Departamento de Farmacología de la Universitat de València, bajo nuestra

dirección y asesoramiento.

Concluido el trabajo experimental y bibliográfico, autorizamos la presentación

de esta Tesis Doctoral para que sea juzgada por el tribunal correspondiente.

Valencia, 17 de Septiembre de 2013

Dr. Diego Cortes Dra. Nuria Cabedo



La presente Tesis Doctoral ha sido financiada con los siguientes

proyectos y ayudas:

a) Proyecto públicos

Proyecto concedido por el Ministerio de Educación y

Ciencia (SAF2007-63142).

“Síntesis de Isoquinoleínas Dopaminérgicas”

Proyecto concedido por el Ministerio de Ciencia y

Tecnología (SAF2011-23777).

“Estudio de los mecanismos moleculares y celulares en la

disfunción endotelial asociada a enfermedades con

inflamación sistémica que podrían inducir desórdenes

cardiovasculares”

b) Proyectos financiados por empresas

Proyecto concedido por Fundación Medina, 2010.

“Aislamiento de acetogeninas”

Proyecto concedido por Laboratorios Servier, 1997-2012.

“Semisíntesis de estiril-lactonas antitumorales”

Proyecto concedido por Valentia Biopharma 2011-2013.

“Síntesis de moléculas bioactivas”

c) Becas

“Beca colaboración” con cargo al proyecto Semisíntesis de

Estiril-Lactonas Antitumorales, dirigido por el profesor

Diego M. Cortes Martínez, y financiado por Laboratorios

Servier (2010-2012).

“Beca investigación” con cargo al proyecto Semisíntesis

de Estiril-Lactonas Antitumorales, dirigido por el profesor

Diego M. Cortes Martínez, y financiado por Laboratorios

Servier (2012-2013).

Seguramente, lo más difícil de toda esta Tesis Doctoral va a ser

agradecer a toda la gente que lo merece sin ocupar otro tomo entero.

En primer lugar, quiero expresar mi profundo agradecimiento a mi

director Diego, cuya confianza en mi desde el primer día ha significado

más de lo que pueda parecer. Gracias por formarme como la

investigadora que soy ahora y por preocuparte siempre tanto por todos

nosotros. Gracias por darme la oportunidad de realizar esta Tesis

Doctoral. A mi directora Nuria, gracias por su amabilidad, por los buenos

ratos pasados trabajando juntas y por toda la ayuda prestada. Gracias por

ofrecerme, además de mil enseñanzas, tu amistad incondicional. Para

vosotros dos, no hay gracias en el mundo suficientes. A Loles y a Maria

Jesús por su disposición a ayudar. A Almu por ofrecer su colaboración

desinteresada. Al profesor Xavier Franck de la Universidad de Rouen,

por acogerme en esas estancias en Francia. A Mari Carmen y Arturo, de

Valentia Biopharma, por confiar en mí y hacer el trabajo tan gratificante.

A todos los profesores del Departamento de Farmacología.

A todo el personal de secretaría, muy especialmente a Ángel, cuyo

impecable trabajo en RMN ha sido imprescindible para la realización de

este trabajo. A mis compañeros, los de antes, y los de ahora. A

Noureddine, a Sònia y a Paloma porque me enseñaron todo lo que sé

sobre el trabajo en el laboratorio. Especialmente a mis compañeros Javi y

Abraham, por todas esas horas compartidas en el labo, por convertirlas en

horas de risas, de anécdotas y de vivencias. A Isabel, Elisa, Marta, Rosa,

María y en general a todos los compañeros del fondo, por su amistad. A

los compañeros del otro labo, especialmente a Carmen, Rita, Julia, Cris y

Fermín, por esas charlas en el pasillo que muchas veces te alegran el día.

En el plano personal, tengo la fortuna de tener a muchas personas a

las que agradecer. A mis amigos de Canet, por interesarse por mi

investigación y apoyarme cuando ha sido necesario. Por los años que

llevamos juntos y los que nos quedan. Muy especialmente a Edu, por

darme siempre un motivo por el que sacar fuerzas. A Paloma y Ainhoa,

por su entusiasmo sobre mi trabajo y su apoyo. A Maria Jesús y Lluís,

por animarme en los días duros y celebrar los buenos con unas cervecitas

en el Fata. A mis amigos de la universidad: Pablo, Lorena, Sara, Adrián,

Chelo, Marina, Bea y Eva por aportar su granito de arena, cada uno a su

forma, para esta Tesis.

A mi familia, especialmente a mis padres, por la libertad a la hora

de elegir que siempre me han dado, por nunca habernos presionado a

ninguno. A mis hermanos: Ángela, Miguel, Ana, Jose y Elena, por

siempre interesarse por mi trabajo y creer en su hermana pequeña. A mis

sobrinos, Ángela y Miguel. A los que ya no están.

Especialmente, y por encima de todas las cosas, a Alejandro, por

haberme sostenido cuando me fallaban las fuerzas. Por casi haberme

obligado a continuar en ocasiones. Sin él, simplemente no habría llegado

hasta aquí.

A todos los que no he nombrado pero me han ayudado durante

estos años. A todos, mis más profundas GRACIAS.

Abreviaturas utilizadas

AMPc: Adenosin monfosfato cíclico

Asp: Ácido Aspártico

ATP: Adenosin trifosfato

BP: Benzopiranos prenilados

CI50: Concentración Inhibitoria 50

COMT: Catecol-O-metil transferasa

CoQ: Coenzima Q

CRM: Cadena Respiratoria Mitocondrial

CYP51: Lanosterol 14α-demetilasa

DAT: Transportador de Dopamina

DFT: Teoría Funcional de la Densidad

IQ: Isoquinoleína

LHON: Neuropatía Óptica Hereditaria de Leber

MAO: Monoamina Oxidasa

MD: Dinámica Molecular

MSA: Metabolitos Secundarios Activos

NADH: Nicotinamida adenina dinucleótico

Phe: Fenilalanina

QM: Mecánica Cuántica

QTAIM: Teoría Cuántica de Atomos en Moléculas

RD: Receptor Dopaminérgico

REA: Relaciones Estructura-Actividad

RMN: Resonancia Magnética Nuclear

Ser: Serina

SNC: Sistema Nervioso Central

THIQ: Tetrahidroisoquinoleínas

ÍNDICE Pág.

INTRODUCCIÓN 1

1. Introducción y objetivos 3

2. Alcaloides isoquinoleínicos: generalidades 7

3. Biosíntesis de isoquinoleínas 10

4. Tipos de isoquinoleínas 12

5. Síntesis de isoquinoleínas 17

5.1. Síntesis de pirrolo[2,1-a]isoquinoleínas 17

5.2. Síntesis de benzo[a]quinolicidinas 20

5.3. Síntesis de alcaloides fenantrénicos 24

6. Actividades biológicas de isoquinoleínas 25

6.1. Actividad antimicrobiana 25

6.2. Actividad inhibidora de la CRM 27

6.3. Actividad dopaminérgica 32

6.3.1. Receptores dopaminérgicos 35

REFERENCIAS BIBLIOGRÁFICAS 37

RESUMEN-DISCUSIÓN RESULTADOS 53

Resumen Capítulo I 57

Resumen Capítulo II 58

Resumen Capítulo III 61

Resumen Capítulo IV 64

ENGLISH VERSION 73

Outline Chapter I 77

Outline Chapter II 78

Outline Chapter III 81

Outline Capítulo IV 84

ANEXO I: Artículos científicos 93

Capítulo 1 95

Artículo 1 97

“Las Annonáceas: fuente de inspiración para la obtención de

nuevos medicamentos” (En: Annonáceas, plantas antiguas,

estudios recientes. Universidad de Ciencias y Artes de

Chiapas, México. 2012)

Capítulo 2 115

Artículo 2 117

“Synthesis of new antimicrobial pyrrolo [2,1-a]isoquinolin-

3-ones” (En: Bioorganic & Medicinal Chemistry, 2012, 20,

6589)

Capítulo 3 127

Artículo 3 129

“Synthesis of pirido[2,1-a]isoquinolin-4-ones and

oxazino[2,3-a]isoquinolin-4-ones: New inhibitors of

mitocondrial respiratory chain” (En: European Journal of

Medicinal Chemistry 2013, 69, 69)

Capítulo 4 137

Artículo 4 139

“3,4-Dihydroxy- and 3,4-methylenedioxy phenanthrene-type

alkaloids with high selectivity for D2 dopamine receptor”

(En: Bioorganic and Medicinal Chemistry Letters, 2013, 23,

4824)

Artículo 5 143

“Tetrahydroisoquinolines acting as dopaminergic ligands. A

molecular modeling study using MD simulations and QM

calculations” (En: Journal of Molecular Modelling, 2012, 18,

419)

Artículo 6 157

“2-(3chloro-4-hydroxyphenyl)ethylamine acting as ligand of

the D2 dopamine receptor. Molecular modelling, synthesis

and D2 receptor affinity” (Enviado a European Journal of

Medicinal Chemistry)

ANEXO II: Espectros RMN 193

Introducción y objetivos

Introducción

3

1. Introducción y objetivos

La naturaleza sigue siendo una fuente de inspiración

inagotable en el diseño de nuevos medicamentos. Numerosos

productos naturales y sus análogos estructurales forman parte del

arsenal terapéutico, mientras que otros sin propiedades farmacológicas

reconocidas, son utilizados por la industria farmacéutica como

productos de partida en la síntesis de nuevos medicamentos.

La Farmacoquímica Natural estudia los principios activos

obtenidos a partir de materias primas de origen natural, ya sean

vegetales, animales o microorganismos. Estos principios activos son

sustancias, también denominadas “metabolitos secundarios activos”

(MSA), que la naturaleza biosintetiza y que pueden llegar a utilizarse

directamente como fármacos, o constituir la fuente de inspiración en

la preparación de nuevos medicamentos de síntesis. La

Farmacoquímica Natural aborda, por tanto, el estudio de estos MSA

desde diferentes puntos de vista:

1) Origen biológico

2) Origen biogenético

3) Propiedades físico-químicas

4) Determinación estructural

5) Propiedades farmacológicas

6) Aplicaciones terapéuticas

Introducción

4

Nuestro grupo de investigación ha trabajado desde hace

años en el campo de la Farmacoquímica Natural. Dentro de la misma,

el estudio de las plantas de la familia de las Annonáceas y de sus

interesantes propiedades terapéuticas ha constituido uno de los temas

centrales de nuestra investigación.

Figura 1. Ejemplos de MSA presentes en Annonáceas

Las plantas pertenecientes a la familia Annonáceas son

conocidas desde hace tiempo tanto por su interés económico

(proporcionan frutos comestibles y aceites esenciales) como por su

utilización en agroquímica y medicina tradicional (plaguicidas y

antiparasitarias) [1]. Así pues, el aislamiento, caracterización y estudio

de sus MSA responsables de estas actividades ha suscitado un gran

interés para numerosos grupos de investigación. Entre estos MSA,

cabe destacar la importancia de las acetogeninas y de los alcaloides

isoquinoleínicos (IQ), ya que se trata de dos de los tipos de MSA más

abundantes y ampliamente distribuidos en las Annonáceas (Figura 1).

Introducción

5

a) Acetogeninas. Se trata de moléculas presentes

exclusivamente en especies de la familia Annonácea y que presentan

interesantes propiedades biológicas. Entre ellas destacan las

propiedades citotóxicas, antitumorales, antiparasitarias e insecticidas

[2]. Estructuralmente se caracterizan por presentar una larga cadena de

32 o 34 átomos de carbono con varias funciones oxigenadas y una

lactona terminal. Su actividad citotóxica se debe fundamentalmente a

la inhibición selectiva, de rango nanomolar, que ejercen sobre el

complejo I de la Cadena Respiratoria Mitocondrial (CRM) [3].

b) Estiril-lactonas. Estos compuestos se caracterizan

por poseer un esqueleto básico de 13 átomos de carbono con una

estructura formada por un grupo estiril o pseudoestiril unido a un

anillo lactona. Las estiril-lactonas se distribuyen especialmente en

especies del género Goniothalamus, ampliamente utilizadas en

medicina popular, y han sido descritas como citotóxicas [4]. En los

ensayos in vitro, en líneas celulares tumorales humanas realizados

sobre estos compuestos, destaca la actividad antitumoral de

crassalactona B, que fue capaz de inhibir el crecimiento de la línea

celular T de leucemia (Jurkat) con una CI50 de 0.45 µM [5].

c) Benzopiranos prenilados. Los benzopiranos

prenilados (BP) son otro grupo de MSA presentes en especies de la

familia de las Annonáceas, concretamente en las del género

Polyalthia. Debido a la similitud estructural de los BP con el

Introducción

6

ubicromanol (derivado del coencima Q, transportador de electrones de

la CRM) se han realizado sobre polycerasoidin, polycerasoidol y

polyalthidin, BP de Polyalthia cerasoides, ensayos de actividad

inhibidora del complejo I de la CRM, encontrando que polyalthidin

era el inhibidor más potente con una CI50 de 4.4 µM [6]. Esta

actividad inhibidora de la cadena transportadora de electrones de los

BP, podría explicar las propiedades citotóxicas y antitumorales que

presentan estas especies vegetales.

d) Alcaloides Isoquinoleínicos. Los alcaloides IQ

pueden encontrarse en especies de un gran número de familias

botánicas dentro del orden de las Magnoliales, entre las que se

encuentra la familia de las Annonáceas. Abarcan un amplio grupo de

estructuras, incluidas las 1-benciltetrahidroisoquinoleínas (1-bencil-

THIQ), aporfinas y protoberberinas. Cabe destacar que los alcaloides

IQ se caracterizan por ser capaces de ejercer numerosas y variadas

actividades biológicas.

En la presente Tesis Doctoral hemos realizado un estudio

químico-farmacológico de las IQ sintetizadas, en base a los estudios

de afinidad por los receptores dopaminérgicos que se habían realizado

previamente sobre diversas IQ aisladas en la familia de las

Annonáceas [7]. De esta manera, nos fijamos tres objetivos:

Introducción

7

Desarrollo de métodos de síntesis total de

diferentes alcaloides isoquinoleínicos:

a) pirrolo[2,1-a]isoquinolin-3-onas,

b) pirido y oxazino[2,3-a]isoquinolin-4-onas,

c) alcaloides aporfínicos y fenantrénicos

Estudio de sus actividades biológicas: Citotóxica,

incluyendo antimicrobiana e inhibición de CRM, y

dopaminérgica.

Relaciones Estructura-Actividad (REA).

2. Alcaloides Isoquinoleínicos: Generalidades

Dentro del grupo de los alcaloides, las IQ representan una

clase de compuestos naturales y sintéticos que han recibido una

atención considerable debido a la importancia de sus actividades

biológicas, entre las que se encuentran la inhibición de la proliferación

celular y propiedades antitumoral, antiespasmódica, hipnoanalgésica y

antifúngica. Un gran número de alcaloides IQ han sido aislados a

partir de fuentes naturales y se ha analizado su capacidad de inhibir el

transporte de neurotransmisores y su afinidad por los sitios de unión a

receptores [7]. Actúan tanto sobre los receptores de membrana

dopaminérgicos y adrenérgicos, como sobre los canales de calcio

voltaje-dependientes e incluso sobre segundos mensajeros

intracelulares como el AMPc.

Introducción

8

En la actualidad, el gran avance en la química de síntesis y

la disponibilidad de nuevas y sofisticadas técnicas de determinación

estructural han permitido sintetizar un gran número de alcaloides

diferentes. Además, la utilización de métodos computacionales y de

modelización molecular ha reducido notoriamente la gran cantidad de

caminos y ensayos que se deben realizar para obtener resultados

satisfactorios. Mediante estudios de modelización molecular se han

podido determinar posibles farmacóforos, es decir, la parte de la

molécula responsable de la actividad farmacológica.

El aislamiento, caracterización y desarrollo de nuevos

métodos de síntesis de moléculas con esqueleto isoquinoleínico

(aporfinas, protoberberinas, cularinas, 1-bencil-THIQs y bisbencil-

THIQs), así como la evaluación de la afinidad por los receptores

dopaminérgicos (RD), es una de las líneas de investigación más

relevantes seguidas por nuestro grupo de investigación [8-16].

Resultados previos sugieren que las 1-bencil-THIQs naturales y

sintéticas tienen afinidad por los RD (tipo-D1 y tipo-D2) y, en algunos

casos, pueden inhibir el proceso de recaptación de la dopamina.

En trabajos anteriores de nuestro grupo, describimos la

síntesis enantioselectiva de pares de 1-bencil-THIQs dopaminérgicas

(1S) y (1R), utilizando (S) y (R)-fenilglicinol como auxiliar quiral. Los

enantiómeros (1S) resultaron ser de 5 a 15 veces más efectivos frente a

los RD tipo-D1 y tipo-D2 que los enantiómeros (1R) (Figura 2) [11].

Introducción

9

Figura 2. 1-bencil-THIQ (1S) y (1R)

Posteriormente, se describió la preparación en “one pot”

de una secuencia de THIQs 1-ciclohexilmetil-7,8-dioxigenadas

(Figura 3), sustituidas y no sustituidas en el anillo ciclohexil mediante

la aplicación de una transposición de photo-Fries, seguida por una

ciclación-reducción en tándem. De hecho, conseguimos por primera

vez una hidrogenación regioselectiva del anillo bencílico en el sistema

THIQ. Todas las IQs 1-ciclohexilmetil sintetizadas fueron capaces de

desplazar a los radioligandos específicos de los RD tipo-D2 de sus

lugares de unión en cuerpo estriado de rata, mientras que los derivados

N-metilados mostraron también afinidad por los RD tipo-D1 [12].

Figura 3. THIQ 1-ciclohexilmetil-7,8-dioxigenada

A través de estos resultados, preparamos nuevas IQs 1-

sustituídas. Mediante estudios de modelización molecular se pudo

confirmar la presencia de dos farmacóforos: la sustitución alquílica

Introducción

10

del nitrógeno IQ, así como la del anillo aromático, preferentemente

por un OH y un Cl [13, 14].

3. Biosíntesis de isoquinoleínas

Los alcaloides son sustancias orgánicas nitrogenadas de

carácter básico y de origen natural, generalmente biosintetizadas a

partir de aminoácidos, y de estructura química muy variada y

compleja. Los alcaloides con esqueleto IQ constituyen uno de los

grupos más abundantes en la naturaleza, se encuentran principalmente

distribuidos en el reino vegetal y abarcan una amplia variedad de tipos

estructurales. En particular, las 1-bencil-THIQ ocupan un lugar

privilegiado dentro de este grupo, puesto que a partir de ellas se

biosintetizan multitud de análogos estructurales, tales como

protoberberinas, aporfinas, bisbencil-THIQ, cularinas, ftalil-IQ,

morfínicos, etc. La primera 1-bencil-THIQ de la ruta biogenética es el

alcaloide norcoclaurina, cuya biosíntesis tiene lugar mediante una

reacción de tipo Pictet-Spengler, a través de la condensación de una

molécula de dopamina con otra de p-hidroxifenilacetaldehído,

formadas ambas a partir de L-tirosina. (Figura 4).

Introducción

11

NH

HO

HOH

OH

COOH

NH2HO

HO

O

HNH2HO

HO

NCH3H

HO

HO

O

NH

H

O

O

NHH3CO

H

OCH3

OH

OH

NCH3H3CO

HO

H3CO OH

NH2HO

NH2HO

HO COOH COOH

NH2HO

(S)-norcoclaurina

morfina

tirosina

dopamina p-OH-fenilacetaldehído

generación del anillo "tetrahidro-isoquinoleína"

(THIQ)

condensacióntipo "Pictet-Spengler"

aporfina

(S)-crasifolina

tiramina L-dopa p-OH-fenilpirúvico

cularina

A B

C

Figura 4. Biosíntesis de isoquinoleínas

Introducción

12

4. Tipos de isoquinoleínas

En la naturaleza se han aislado numerosos alcaloides IQ

con esqueletos diversos [17]. Algunos de los más abundantes, y cuyo

estudio abordamos en la presente Tesis Doctoral, son los siguientes:

a) IQ derivadas biogenéticamente de 1-bencil-

THIQ: Las 1-bencil-THIQ son precursoras de un gran número de

alcaloides con diferentes esqueletos (Figura 4). Muchos de estos

derivados poseen potentes propiedades farmacológicas, como por

ejemplo los analgésicos morfina y codeína (esqueleto morfínico), los

antimicrobianos sanguinarina y berberina (esqueleto

protoberberínico), el curarizante tubocurarina (bisbencil-IQ), el

relajante de la musculatura lisa papaverina (1-bencil-IQ) o el

antitusivo noscapina (ftalil-IQ) (Figuras 4 y 5).

- Protoberberinas: Las protoberberinas son alcaloides IQ

de origen natural presentes, entre otras, en especies de las

familias Annonáceas, Berberidáceas y Ranunculáceas, como

Hydrastis canadensis (sello de oro), Coptis chinensis (coptis o

hebra de oro), Berberis aquifolium (uva de Oregón), B.

vulgaris (agracejo) y B. aristata (árbol de cúrcuma). Este tipo

de alcaloides presentan un esqueleto tetracíclico originado a

partir de un sistema 1-bencil-THIQ. Los alcaloides

tetrahidroprotoberberínicos (THPB) presentan actividades

Introducción

13

farmacológicas muy interesantes y variadas. En los últimos

años se ha observado que pueden actuar tanto sobre receptores

dopaminérgicos, como inhibiendo la acetilcolinesterasa y

butirilcolinesterasa, además de tener actividad bactericida y ser

incluso útiles para tratar la diabetes [18, 19].

Figura 5. Estructuras de IQ derivadas de 1-bencil-THIQ

- Aporfinas: Los alcaloides aporfínicos constituyen

uno de los grupos más amplios dentro de las isoquinoleínas

[20]. Se encuentran distribuidos en el reino vegetal, como por

ejemplo en las familias de las Annonáceas, Lauráceas,

Monimiáceas, Hernandiáceas y Ranunculáceas. Muchos de

ellos poseen propiedades farmacológicas, como antioxidantes,

Introducción

14

antiagregantes plaquetarios, antineoplásicos o anti

parkinsonianos, entre otros [21]. Algunas aporfinas tienen

aplicación terapéutica, como es el caso de la apomorfina

(Figura 5), alcaloide semisintético obtenido a partir de la

morfina, capaz de actuar como agonista de los receptores

dopaminérgicos D2, lo que implica estimulación locomotora

con aplicaciones en el tratamiento de la enfermedad de

Parkinson. Otro ejemplo sería la bulbocapnina, que actúa como

antagonista dopaminérgico, produciendo reducción de la

actividad motora e inducción de catalepsia. El grupo de las

aporfinas abarca diversas subestructuras tanto naturales como

sintéticas: oxoaporfinas, proaporfinas y alcaloides

fenantrénicos, entre otros.

- Cularinas: La (+)-cularina fue aislada por Manske

en 1938 a partir de especies pertenecientes a la familia

Fumariáceas. Su estructura fue determinada en 1950.

Numerosos alcaloides cularínicos han sido aislados de especies

de Ceratocapnos, Corydalis, Dicentra y Sarcocapnos [22-25].

Como ya se ha mencionado en la biosíntesis de IQs, se

considera que la 1-bencil-THIQ 7,8-dioxigenada crassifolina

es la precursora en la biosíntesis de los alcaloides cularínicos,

que contienen un sistema oxepina (Figura 4).

Introducción

15

b) Pirrolo[2,1-a]isoquinoleínas. Estructuralmente se

caracterizan por poseer un anillo nitrogenado de cinco miembros

fusionado con el núcleo IQ (Figura 6). Dentro de las IQ, las

pirroloisoquinoleínas han recibido considerable atención dadas las

interesantes actividades biológicas que son capaces de ejercer. Entre

ellas, cabe destacar que han sido descritas como antidepresivas,

agonistas muscarínicos, antiagregantes plaquetarios o antineoplásicas

[26]. Curiosamente, las pirrolo[2,1-a]isoquinoleínas fueron

sintetizadas mucho antes de que fuesen aisladas como productos

naturales [26]. La actividad antineoplásica que algunas de ellas son

capaces de desempeñar despertó un gran interés por las mismas. De

este grupo cabe destacar los productos naturales crispina A y crispina

B, con propiedades antitumorales [27], y trollina, descrita como

antimicrobiana y citotóxica [28].

Figura 6. Estructuras de pirrolo[2,1-a]isoquinoleínas

Además, también se ha encontrado en la naturaleza este

esqueleto integrado en otras estructuras policíclicas, como es el caso

de las lamelarinas aisladas de invertebrados marinos [29].

Introducción

16

c) Benzo[a]quinolicidinas. El esqueleto benzo[a]

quinolicidina puede encontrarse en multitud de compuestos

bioactivos, como es el caso de los alcaloides bis-IQ de la Ipecacuana:

emetina y alcaloides relacionados (Figura 7). Estos alcaloides son

interesantes desde un punto de vista farmacológico dado que poseen

actividades biológicas como antiamebianas [30], antitumorales [31] y

se utilizan en clínica como eméticos. En trabajos recientes, se ha

puesto de manifiesto que el sistema tricíclico que presentan estas

moléculas resulta esencial para su efecto citotóxico [32].

Figura 7. Estructura de benzo[a]quinolicidinas

Las benzo[a]quinolicidinas también forman parte de

sistemas más grandes, como por ejemplo en las protoberberinas, IQ

anteriormente mencionadas (Figura 5).

Introducción

17

5. Síntesis de Isoquinoleínas

Como se ha comentado anteriormente, las IQ se

encuentran ampliamente distribuidas en la naturaleza. Utilizando la

misma como fuente de inspiración, numerosos grupos han centrado

sus esfuerzos en preparar análogos sintéticos que permitan optimizar

las actividades biológicas mostradas por los productos naturales.

5.1. Síntesis de pirrolo[2,1-a]isoquinoleínas

La atractiva estructura y las interesantes propiedades

biológicas de las pirrolo[2,1-a]isoquinoleínas han atraído la atención

de un gran número de investigadores, que han descrito diversas

secuencias sintéticas.

Una de las síntesis más representativas descritas en la

literatura implica la ciclación del anillo de cinco miembros a través de

una reacción intramolecular de ciclación oxidativa entre una IQ y el

triple enlace de un grupo propargilo en posición 1, promovida por

acetato de plata (Esquema 1) [33].

Introducción

18

Esquema 1. Síntesis de pirrolo[2,1-a]isoquinoleínas a partir de

dihidroisoquinoleínas

Otras estrategias conocidas proceden principalmente vía

fenetil succinimidas, en las cuales un ión N-acil iminio experimenta

diferentes tipos de ciclación intramolecular para crear el enlace C10a-

C10b. En este caso, las diferentes aproximaciones sintéticas abarcan:

- Adición carbofílica del reactivo de organolitio

sobre una succinimida y reacción de ciclación en tándem del ión N-

acil iminio para obtener el esqueleto isoquinoleínico (Esquema 2)

[34].

Introducción

19

Esquema 2. Síntesis de pirrolo[2,1-a]isoquinoleínas mediante adición

carbofílica

- Aplicación de la ciclación tipo Parham a N-fenetil

imidas halogenadas, en las cuales el halógeno se intercambia por el

metal y se produce una ciclación intramolecular, obteniendo las

correspondientes enamidas (Esquema 3) [34].

Esquema 3. Síntesis de pirrolo[2,1-a]isoquinoleínas utilizando la ciclación

tipo Parham

Introducción

20

- La ciclación inducida por la rección de Pummerer

que implica la generación de un ión iminio a partir de un intermedio

O-imido sulfóxido, para obtener el sistema azabicíclico (Esquema 4)

[35].

Esquema 4. Síntesis de pirrolo[2,1-a]isoquinoleínas utilizando la ciclación

de Pummerer

5.2. Síntesis de benzo[a]quinolicidinas

Diferentes aproximaciones a la síntesis de

benzo[a]quinolicidinas han sido descritas en la literatura. La mayoría

de ellas implican el cierre del anillo isoquinoleínico a través de la

formación del enlace C11a-C11b por ciclodeshidratación de Bischler-

Napieralski o por ciclación catalizada por paladio a través de una

Introducción

21

reacción tipo Heck de un intermedio enaminona, sustituida en

posición 6 (Esquema 5) [36].

Esquema 5. Síntesis de benzo[a]quinolicidinas por ciclación catalizada por

paladio

Otro método descrito en la literatura consiste en la

formación del enlace C1-C11b por la ciclación de Mannich, entre una

dihidroisoquinoleína y una β-dicetona (Esquema 6) [37]. Roy y cols.

describieron un procedimiento efectivo para la síntesis de

benzo[a]quinolicin-4-tionas a través de la reacción de 3,4-

dihidroisoquinoleínas con diferentes β-oxoditioésteres (Esquema 7)

[38]. Recientemente, Padwa y cols. han obtenido esta estructura

Introducción

22

mediante una cascada de cicloadición dipolar-adición conjugada

(Esquema 8) [39].

Esquema 6. Síntesis de benzo[a]quinolicidinas a través de la ciclación de

Mannich

Esquema 7. Síntesis de benzo[a]quinolicidinas por reacción de

dihidroisoquinoleínas con β-oxoditioésteres

Introducción

23

Esquema 8. Síntesis de benzo[a]quinolicidinas por cascada de cicloadición

dipolar-adición conjugada

Pemberton y cols. describieron una síntesis de

benzo[a]quinolicidinas haciendo reaccionar dihidroisoquinoleínas con

derivados de los ácidos de Meldrum o 1,2-dioxin-4-onas como fuente

de acilcetenos (Esquema 9) [40].

Esquema 9. Síntesis de Pemberton

Introducción

24

5.3. Síntesis de alcaloides fenantrénicos

De manera similar a las pirrolo[2,1-a]isoquinoleínas y las

benzo[a]quinolicidinas, los alcalodes fenantrénicos también han sido

estudiados por numerosos grupos de investigación, debido a su

particular estructura y a sus variadas actividades biológicas.

Consecuentemente, varios métodos de síntesis total para estos

compuestos han sido descritos.

Puesto que en la naturaleza los alcaloides fenantrénicos

son productos de degradación de las aporfinas, encontramos en la

literatura diversas síntesis a partir de las mismas, utilizando

generalmente la degradación de Hofmann (Esquema 10) [41, 42].

Esquema 10. Síntesis de alcaloides fenantrénicos a partir de aporfinas

Introducción

25

Asimismo, también se han puesto a punto métodos de

síntesis de estos compuestos sin pasar por las aporfinas, utilizando una

ciclación fotoquímica (Esquema 11) [43, 44].

Esquema 11. Síntesis de alcaloides fenantrénicos por ciclación fotoquímica

6. Actividades biológicas de isoquinoleínas

6.1. Actividad antimicrobiana

Entre las variadas actividades biológicas que presentan las

IQ se encuentran la actividad antimicrobiana (frente a bacterias y

hongos), antimalárica y citotóxica, entre otras.

Introducción

26

Se ha observado que diversas isoquinoleín-5,8-dionas

(quinonas), aisladas de esponjas marinas del género Xestospongia y de

su molusco nudibranquio asociado Jorunna funebris, poseen actividad

antimicrobiana frente a bacterias Gram-positivas (Bacillus subtilis y

Staphyloccoccus aureus) y frente a hongos (Cladosporium

cucumerinum) (Figura 8). Estos compuestos también han demostrado

ser insecticidas (Spodoptera littoralis) [45-47]. La interesante

actividad antibacteriana y antitumoral mostrada por las quinonas, ha

llevado a la síntesis de una serie de derivados isoquinoleín-5,8-dionas

que mostraron ser activos frente a B. subtilis [48].

Figura 8. Isoquinoleínas antimicrobianas

En estudios previos se ha generado la hipótesis, mediante

ensayos in silico de binding, que algunas THIQs poseen afinidad por

el lugar de unión al sustrato del enzima fúngico lanosterol 14α-

demetilasa (CYP51). Diversos derivados pirazino[2,1-

a]isoquinoleínicos han mostrado tener una potente actividad

antifúngica in vitro frente a varias especies de hongos patógenos

(Candida albicans, Cryptococcus neoformans, Aspergillus fumigatus

y Trichophyton rubrum) [49, 50].

Introducción

27

Como se ha comentado anteriormente, también las

pirrolo[2,1-a]isoquinoleínas han sido descritas como antimicrobianas,

destacando entre ellas la trollina [28].

6.2. Actividad inhibidora de la CRM

La enzima NADH:ubiquinona oxidorreductasa, también

conocida como complejo I mitocondrial, es el complejo proteico más

grande de la membrana mitocondrial interna implicado en la

generación de energía celular. Su estructura cristalina ha sido descrita

recientemente [51]. Está considerado el principal punto de entrada de

electrones en la cadena respiratoria mitocondrial (CRM) y es el

encargado de catalizar la transferencia de éstos desde el nicotinamida

adenina dinucleótido reducido (NADH) de la matriz mitocondrial

hasta un aceptor, la ubiquinona (CoQ), de la membrana mitocondrial

interna (Figura 9).

Su estudio ha suscitado un gran interés debido a que un

amplio espectro de patologías, tanto hereditarias como adquiridas, ha

sido asociado a defectos funcionales de la enzima. Tal es el caso de

diversas enfermedades neurodegenerativas como Parkinson [52],

neuropatía óptica hereditaria de Leber (LHON) [53] o Alzheimer [54].

Introducción

28

Figura 9. Transporte electrónico a través de la cadena respiratoria y

fosforilación oxidativa

Asimismo, está claramente establecido su papel en el

desarrollo de otros desórdenes tales como síndrome de Leigh [55],

cardiopatía isquémica [56], diabetes mellitus [57], esquizofrenia [58]

y alteraciones nutricionales [59]. De igual modo, el propio proceso de

envejecimiento celular ha sido asociado principalmente a un

progresivo deterioro del funcionamiento de este complejo enzimático

[60]. Por todo ello, el complejo I mitocondrial es objeto de una intensa

investigación que abarca tanto los aspectos funcionales como los

estructurales y mecanísticos.

Del papel crucial que desempeña el complejo I como paso

limitante del mantenimiento de la función bioenergética celular, se

desprende la gran importancia que la inhibición de dicho complejo

puede acarrear sobre el metabolismo y la supervivencia celular. El

COMPLEJO I

NADH

Q

Cit c

COMPLEJO III COMPLEJO IV

ATPasa

½ O2 H2O

Espacio

intermembrana

Matriz

mitocondrial

H+ = Gradiente electroquímico [H+]

COMPLEJO II

ADP + Pi

ATP

Introducción

29

interés por un mejor conocimiento tanto de la estructura como de la

función de esta enzima, con el fin de entender los mecanismos

moleculares de las patologías en las cuales se encuentra implicado, ha

impulsado la búsqueda de nuevos inhibidores específicos. Estos

podrían convertirse en potentes herramientas para diseccionar el

mecanismo funcional del complejo I [61]. Por otra parte, este

complejo ha adquirido gran relevancia gracias a un creciente interés

industrial y comercial como posible diana de insecticidas, acaricidas y

fungicidas [62]. Pero, sin duda alguna, el aspecto de mayor relevancia

es el de la posible aplicación de inhibidores del complejo I como

antitumorales, dado que se ha descrito que algunos de estos

compuestos inhiben la proliferación de líneas celulares tumorales

humanas mediante la inducción de apoptosis [63,64].

Figura 10. IQ inhibidoras de la CRM

Durante los últimos años se ha conseguido caracterizar un

elevado número de inhibidores de esta enzima, hecho que continúa

actualmente en aumento. En este sentido, nuestro grupo de

Introducción

30

investigación ha descrito varias IQ naturales y sintéticas como

efectivos inhibidores de la CRM [65, 66] (Figura 10).

6.3. Actividad dopaminérgica

La dopamina o 3,4-dihidroxifeniletilamina, es una de las

catecolaminas naturales biosintetizada en las terminaciones neuronales

dopaminérgicas, localizadas fundamentalmente en el cuerpo estriado y

en el sistema límbico, almacenándose en las vesículas sinápticas. Se

biosintetiza a partir del aminoácido esencial L-tirosina, mediante una

cascada de reacciones enzimáticas. Este neurotransmisor

catecolaminérgico se libera a la hendidura sináptica debido a

variaciones en los potenciales de acción. En el espacio sináptico, la

dopamina ejerce su función uniéndose a los receptores postsinápticos,

los cuales transfieren las señales entre las células.

La biosíntesis de dopamina y su liberación en las neuronas

puede ser modulada por dos mecanismos adicionales que afectan la

neurotransmisión dopaminérgica:

a) La estimulación de los autoreceptores presinápticos

tipo-D2 por los agonistas dopaminérgicos [67-70].

b) El mecanismo de recaptación desde el espacio

extraneuronal a las neuronas presinápticas por el transportador de

membrana de dopamina (DAT) [71, 72].

Introducción

31

No hay que olvidar también la inactivación metabólica

producida por la acción combinada de las enzimas MAO y COMT

[73] (Figura 11).

Figura 11. Terminación dopaminérgica

La dopamina es un importante neurotransmisor del

sistema nervioso central (SNC) implicada en la regulación de un gran

número de actividades fisiológicas incluyendo el control de los

movimientos, estados emocionales, funciones cognitivas, secreciones

neuroendocrinas y desórdenes de sustancias abusivas. Estas

actividades están asociadas a cuatro vías dopaminérgicas

principalmente: vía mesocortical, vía mesolímbica, vía nigroestriada y

vía túberoinfundibular, de manera que una degeneración parcial de

Introducción

32

estas vías se manifiesta en el desarrollo de diversos tipos de

desórdenes psicóticos y neurológicos, como el Parkinson.

6.3.1. Receptores dopaminérgicos

En función de su localización celular y de sus

características farmacológicas, se pueden considerar varios aspectos

relacionados con los RD. En primer lugar, son receptores de

membrana, estructuralmente constituidos por siete dominios

transmembrana hidrofóbicos acoplados a una proteína G. Fueron

inicialmente clasificados en dos subtipos, D1 y D2, basados en sus

propiedades farmacológicas y en su diferente activación de la

adenilciclasa [74]. Por otra parte, la aplicación de métodos de

clonación usando técnicas de biología molecular, ha permitido

reagruparlos en dos subfamilias, tipo-D1 (D1 y D5) y tipo-D2 (D2, D3 y

D4) con secuencias de aminoácidos y algunas propiedades

farmacológicas comunes.

La activación de los receptores tipo-D1, estimula la

adenilciclasa a través de la proteína Gs y conduce a un aumento del

AMPc intracelular en la neurona postsináptica, mientras que la

activación de los receptores tipo-D2 inhibe la adenilciclasa a través de

una proteína Gi o Go, disminuyendo en consecuencia el contenido de

AMPc intracelular [74-77].

Introducción

33

Los receptores tipo-D1 presentan un 80% de similitud y

propiedades fisiológicas similares, son receptores postsinápticos y se

localizan en casi todas las áreas cerebrales con inervación

dopaminérgica. Los receptores D5 se diferencian de los D1 en que son

menos abundantes y están menos distribuídos, se localizan en el

córtex frontal, cuerpo estriado, hipocampo e hipotálamo [78].

Los receptores tipo-D1 poseen alta afinidad por ligandos

benzacepínicos, SCH 23390 y SKF 83566 (Figura 12) y una

moderada afinidad por agonistas clásicos de dopamina (apomorfina) y

por agonistas selectivos como SKF 38393.

Figura 12. Ligandos específicos de receptores D1

En cuanto a los receptores tipo-D2, se sabe que D2 y D3

muestran un 75% de similitud, mientras que entre los receptores D2 y

D4 sólo hay un 53% de similitud. Se encuentran mayoritariamente

distribuidos en los sistemas nigroestriado y mesolímbico y se

localizan tanto a nivel presináptico (autoreceptores, regulan la

liberación de dopamina), como postsináptico. Las butiferonas, como

haloperidol, y las benzamidas sustituídas, como sulpirida y racloprida

(antagonistas D2) (Figura 13), poseen alta afinidad por dichos

Introducción

34

receptores, al igual que las fenotiazinas y los tioxantenos. Sin

embargo, existen algunas diferencias entre las afinidades de estos tres

receptores (D2, D3 y D4) con algunos ligandos específicos. La

benzamida racloprida, por ejemplo, presenta alta afinidad por los

receptores D2 y D3 y baja por los receptores D4. Dado que los

receptores D3 y D4 se expresan fundamentalmente en las regiones

corticales y límbicas, implicadas en el control del conocimiento y de

las emociones, son objeto de estudio de nuevas generaciones de

fármacos para muchos de los trastornos neurológicos y psiquiátricos

con baja incidencia de efectos colaterales extrapiramidales [75, 78].

Figura 13. Antagonistas de receptores D2

Farmacológicamente, los antagonistas dopaminérgicos son

agentes que se utilizan para el tratamiento clínico de la esquizofrenia,

manía, delirios y la enfermedad de Huntington, mientras que los

agonistas dopaminérgicos se utilizan en desórdenes neuroendocrinos y

en la enfermedad de Parkinson [71]. Muchos de los inhibidores del

transportador de dopamina que inactivan el proceso de recaptación,

actúan como antidepresivos puesto que aumentan la concentración de

Introducción

35

dopamina en la hendidura sináptica y por tanto activan la

neurotransmisión dopaminérgica.

Figura 14. Técnica de fijación de radioligandos

Los estudios de afinidad por los receptores tipo-D1 y tipo-

D2 se llevan a cabo mediante ensayos in vitro utilizando técnicas de

fijación con radioligandos específicos denominadas técnicas de

binding (Figura 14), en cuerpo estriado de rata. Son experimentos de

competición donde se evalúa la capacidad de los compuestos

ensayados para desplazar los radioligandos, [3H]-SCH 23390 (ligando

selectivo de receptores tipo-D1) y [3H]-racloprida (ligando selectivo de

receptores tipo-D2) de sus lugares de unión al receptor [79].

Referencias bibliográficas

Referencias

39


[1] Barbalho, S. M.; Alvares Goulart, R.; Vasques, F. M.; Soares, M. S.;

Bueno, P. C. S.; Landgraf Guiguer, E.; Cressoni Araujo, A. y Groppo,

M.

Annona sp: Plants with Multiple Applications as Alternative Medicine

- A Review.

Curr. Bioact. Compd. 2012, 8, 277.

[2] Liaw, C. C.; Wu, T. Y.; Chang, F. R. y Wu, Y. C.

Historic Perspectives on Annonaceous Acetogenins from the

Chemical Bench to Preclinical Trials.

Planta Med. 2010, 76, 1390.

[3] Bermejo, A.; Figadère, B.; Zafra-Polo, M. C.; Barrachina, I.;

Estornell, E. y Cortes, D.

Acetogenins from annonaceae. Recent progress in isolation, synthesis,

and mechanisms of action.

Nat. Prod. Rep. 2005, 22, 269.

[4] Fatima, A.; Modolo, L. V.; Conegero, L. S.; Pilli, R. A.; Ferreira, C.

V.; Kohn, L. K. y Carvalho, J. E.

Styryl lactones and their derivatives: biological activities, mechanisms

of action and potential leads for drug design.

Curr. Med. Chem. 2006, 13, 3371.

[5] Popsavin, V.; Benedeković, G.; Popsavin, M.; Kojić, V. y

Bogdanović, G.

Divergent synthesis of cytotoxic styryl lactones isolated from

Polyalthia crassa. The first total synthesis of crassalactone B.

Tetrahedron Lett. 2010, 51, 3426.

Referencias

40

[6] Zafra-Polo, M. C.; González, M. C.; Tormo, J. R.; Estornell, E. y

Cortes, D.

Polyalthidin: New Prenylated Benzopyran Inhibitor of the Mammalian

Mitochondrial Respiratory Chain.

J. Nat. Prod. 1996, 59, 913.

[7] Cabedo, N.; Berenguer, I.; Figadère, B. y Cortes, D.

An overview on benzylisoquinoline derivatives with dopaminergic

and serotonergic activities.

Curr. Med. Chem. 2009, 16, 2441.

[8] Protais, P.; Arbaoui, J.; Bakkali, E-H.; Bermejo, A. y Cortes, D.

Effects of various isoquinoline alkaloids on in vitro 3H-Dopamine

uptake by rat striatal synaptosomes.

J. Nat. Prod. 1995, 58, 1475.

[9] Cabedo, N.; Protais, P.; Cassels, B. K. y Cortes, D.

Synthesis and Dopamine Receptor Selectivity of the Benzyl-

tetrahydroisoquinoline, (R)-(+)-nor-Roefractine

J. Nat. Prod. 1998, 61, 709.

[10] Andreu, I.; Cortes, D.; Protais, P.; Cassels, B. K.; Chagraoui, A. y

Cabedo, N.

Preparation of dopaminergic N-alkyl-benzyltetrahydroisoquinolines

using a «One-Pot» procedure in acid medium.

Bioorg. Med. Chem. 2000, 8, 889.

[11] Cabedo, N.; Andreu, I.; Ramírez de Arellano, M. C.; Chagraoui, A.;

Serrano, A.; Bermejo, A.; Protais, P. y Cortes, D.

Enantioselective syntheses of dopaminergic (R)- and (S)-benzyltetra-

hydroisoquinolines.

J. Med. Chem. 2001, 44, 1794.

Referencias

41

[12] Andreu, I.; Cabedo, N.; Torres, G.; Chagraoui, A.; Ramírez de

Arellano, M. C.; Gil, S.; Bermejo, A.; Valpuesta, M.; Protais, P. y

Cortes, D.

Syntheses of dopaminergic 1-cyclohexylmethyl-7,8-dioxygenated

tetra-hydroisoquinolines by selective heterogenous tandem

hydrogenation.

Tetrahedron 2002, 58, 10173.

[13] Suvire, F. D.; Andreu, I.; Bermejo, A.; Zamora, M. A.; Cortes, D. y

Enriz, R. D.

Conformational study of N-alkyl-benzyltetrahydroisoquinoline

alkaloids.

J. Mol. Struc. (Teochem) 2003, 666-667, 109.

[14] Suvire, F. D.; Cabedo, N.; Chagraoui, A.; Zamora, M. A.; Cortes, D. y

Enriz, R. D.

Molecular recognition and binding mechanism of N-alkyl-

benzyltetrahydroisoquinolines to the D1 dopamine receptor. A

computational approach.

J. Mol. Struc. (Teochem) 2003, 666-667, 455.

[15] Berenguer, I.; El Aouad, N.; Andújar, S.; Romero, V.; Suvire, F.;

Freret, T.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Boulouard, M.;

Cabedo, N. y Cortes, D.

Tetrahydroisoquinolines as dopaminergic ligands: 1-Butyl-7-chloro-6-

hydroxy-tetrahydroisoquinoline, a new compound with

antidepressant-like activity in mice.

Bioorg. Med. Chem. 2009, 17, 4968.

[16] El Aouad, N.; Berenguer, I.; Romero, V.; Marin, P.; Serrano, A.;

Andújar, S.; Suvire, F.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.;

Cabedo, N. y Cortes, D.

Structure-activity relationship of dopaminergic halogenated 1-benzyl-

tetrahydroisoquinoline derivatives.

Eur. J. Med. Chem. 2009, 44, 4616.

Referencias

42

[17] Bentley, K. W.

β-Phenylethylamines and the isoquinoline alkaloids.

Nat. Prod. Rep. 2006, 23, 444.

[18] Xiao, H.; Peng, J.; Liang, Y.; Yang, J.; Bai, X.; Hao, X. Y.; Yang, F.

M. y Sun, Q. Y.

Acetylcholinesterase inhibitors from Corydalis yanhusuo.

Nat. Prod. Res. 2011, 25, 1418.

[19] Slobodníková, L.; KoSt’álová, D.; Labudová, D.; Kotulová, D. y

Kettmann, V.

Antimicrobial activity of Mahonia aquifolium. Crude extract and its

major isolated alkaloids.

Phytother. Res. 2004, 18, 674.

[20] Guinaudeau, H.; Leboeuf, M. y Cavé, A.

Aporphinoid alkaloids. V.

J. Nat. Prod. 1994, 57, 1033.

[21] Zhang, A.; Zhang, Y.; Branfman, A. R.; Baldessarini, R. J.; y

Neumeyer, J. L.

Advances in Development of Dopaminergic Aporphinoids.

J. Med. Chem. 2007, 50, 171.

[22] Guinaudeau, H. y Allais, P.

New isoquinoline alkaloids from Corydalis claviculata.

Heterocycles 1984, 22, 107.

[23] Blanco, O.; Castedo, L.; Cortes, D. y Villaverde, M. C.

Alkaloids from Sarcocapnos saetabensis.

Phytochemistry 1991, 30, 2071.

[24] Valpuesta, M.; Posadas, N.; Ruis, I.; Silva, M. V.; Gómez, A. I.;

Suau, R.; Pérez, B.; Pliego, F. y Cabezudo, B.

Alkaloids from Ceratocapnos heterocarpa plants and in vitro cultures.

Phytochemistry 1995, 38, 113.

Referencias

43

[25] Campello, M. J.; Castedo, L.; Saa, J. M.; Suau, R. y Vidal, M. C.

Cancentrine-type cularine alkaloids: sarcocapnine and

oxosarcocapnine.


[26] Passler, U. y Knolker, H. J.

The Pyrrolo[2,1-a]isoquinoline Alkaloids.

The Alkaloids: Chemistry and Biology; Ed. Elsevier: San Diego, 2012;

70, 79–151.

[27] Zhang, Q.; Tu, G.; Zhao, Y. y Cheng, T.

Novel bioactive isoquinoline alkaloids from Carduus crispus.

Tetrahedron 2002, 58, 6795.

[28] Wang, R. F.; Yang, X. W.; Ma, C. M.; Cai, S. Q.; Li, J. N. y Shoyama,

Y.

A bioactive alkaloid from the flowers of Trollius chinensis.

Heterocycles 2004, 63, 1443.

[29] Andersen, R. J.; Faulkner, D. J.; He, C. H.; Van Duyne, G. D. y

Clardy, J.

Metabolites of the marine prosobranch mollusk Lamellaria sp.

J. Am. Chem. Soc. 1985, 107, 5492.

[30] Bansal, D.; Sehgal, R.; Cawla, Y.; Mahajan, R. C. y Malla, N.

In vitro activity of antiamoebic drugs against clinical isolates of

Entamoeba histolytica and Entamoeba dispar.

Ann. Clin. Microb. Antimicrob. 2004, 3, 27.

[31] Zhou, Y. D.; Kim, Y. P.; Mohammed, K. A.; Jones, D. K.;

Muhammad, I.; Dunbar, D. C. y Nagle, D. G.

Terpenoid Tetrahydroisoquinoline Alkaloids Emetine, Klugine, and

Isocephaeline Inhibit the Activation of Hypoxia-Inducible Factor-1 in

Breast Tumor Cells.

J. Nat. Prod. 2005, 68, 847.

Referencias

44

[32] Zaarur, N.; Gabai, V. L.; Porco, J. A.; Claderwood, S. y Sherman,

M. Y.

Targeting heat shock response to sensitize cancer cells to proteasome

and Hsp90 inhibitors.

Cancer Res. 2006, 66, 1783.

[33] Knölker, H. J. y Agarwal, S.

Total synthesis of the antitumor active pyrrolo[2,1-a]isoquinoline

alkaloid (±)-crispine A.


[34] Collado, M. I.; Manteca, I.; Sotomayor, N.; Villa, M. J. y Lete, E.

Metalation vs. Nucleophilic Addition in the Reactions of N-

Phenethylimides with Organolithium Reagents. Ready Access to

Isoquinoline Derivatives via N-Acyliminium Ions and Parham-Type

Cyclizations.

J. Org. Chem. 1997, 62, 2080.

[35] Padwa, A.; Henning, R.; Kappe, C. O. y Reger, T. S.

A Triple Cascade Sequence as a Strategy for the Construction of the

Erythrinane Skeleton.

J. Org. Chem. 1998, 63, 1144.

[36] Kirschbaum, S. y Waldeman, H.

Three-Step Access to the Tricyclic Benzo[a]quinolizine Ring System.

J. Org. Chem. 1998, 63, 4936.

[37] Strandtmann, M.; Cohen, M. P. y Shravel, J.

A new general synthesis of benzo[a]quinolizines,

dibenzo[a,f]quinolizines, and related compounds.

J. Org. Chem. 1966, 31, 797.

Referencias

45

[38] Roy, A.; Sukumar, N.; Hiriyakkanavar, I. y Hiriyakkanavar, J.

An Expedient Route to 2,3-Substituted and Fused

Benzo[a]quinolizine-4-thione Framework via Ring Annulation with β-

Oxodithioesters.

Org. Lett. 2001, 3, 229.

[39] Stearman, C. J.; Wilson, M. y Padwa, A.

Conjugate addition-dipolar cycloaddition cascade for the synthesis of

benzo[a]quinolizine and indolo[a]quinolizine scaffolds: application to

the total synthesis of (+/-)-yohimbenone.

J. Org. Chem. 2009, 74, 3491.

[40] Pemberton, N.; Jakobsson, L. y Almqvist, F.

Synthesis of multi ring-fused 2-pyridones via an acyl-ketene imine

cyclocondensation.

Org. Lett. 2006, 8, 935.

[41] Shamma, M. y Hwang, D. Y.

The synthesis of (±)-talphenine, thaliglucine and thaliglucinone.


[42] Milian, L.; Estellés, E.; Abarca, B.; Ballesteros, R.; Sanz, M. J. y

Blázquez, M.A.

Reactive Oxygen Species (ROS) Generation Inhibited by Aporphine

and Phenanthrene Alkaloids Semi-Synthesized from Natural Boldine.

Chem. Pharm. Bull. 2004, 52, 696.

[43] Estévez, J. C.; Villaverde, M. C.; Estévez, J. A.; Seijas, J. A. y

Castedo, L.

New total synthesis of phenanthrene alkaloids.

Can. J. Chem. 1990, 68, 964.

[44] Kini, S. V. y Ramana, M. M. V.

A new convenient synthesis of phenanthrene alkaloids from 1-

arylmethyl-1,2,3,4-tetrahydroisoquinolines.


Referencias

46

[45] Edrada, R. A.; Proksch, P.; Wray, V.; Christ, R.; Witte, L. y Van

Soest, R. W. M.

Bioactive isoquinoline quinone from an undescribed philippine marine

sponge of the genus Xestospongia.

J. Nat. Prod. 1996, 59, 973.

[46] Mckee, T. C. y Irelan, C. M.

Cytotoxic and antimicrobial alkaloids from the fijian sponge

Xestospongia caycedoi.

J. Nat. Prod. 1987, 50, 754.

[47] Kuwabara, N.; Hayashi, H.; Hiramatsu, N.; Choshi, T.; Kumemura,

T.; Nobuhiro, J. y Hibino, S.

Syntheses of the antibiotic alkaloids renierone, mimocin, renierol,

renierol acetate, renierol propionate, and 7-methoxy-1,6-

dimethylisoquinoline-5,8-dione.


[48] Rao, K. V. y Beach, J. W.

Streptonigrin and related compounds. 5. Synthesis and evaluation of

some isoquinoline analogues.

J. Med. Chem. 1991, 34, 1871.

[49] Tang, H.; Zheng, C.; Lv, J.; Wu, J.; Li, Y.; Yang, H.; Fu, B.; Li, C.;

Zhou, Y. y Zhu, J.

Synthesis and antifungal activities in vitro of novel pyrazino[2,1-a]

isoquinoline derivatives.

Bioorg. Med. Chem. Lett. 2010, 20, 979.

[50] Tang, H.; Zheng, C.; Zhu, J.; Fu, B.; Zhou, Y. y Lv, J.

Design and synthesis of novel pyrazino[2,1-a]isoquinoline derivatives

with potent antifungal activity.

Arch. Pharm. Chem. Life Sci. 2010, 343, 360.

Referencias

47

[51] Baradaran, R.; Berrisford, J. M.; Minhas, G. S. y Sazanov, L. A.

Crystal structure of the entire respiratory complex I .

Nature 2013, 494, 443.

[52] Keeney, P.; Xie, J.; Capaldi, R.A. y Bennett, J. P.

Parkinson’s disease brain mitochondrial complex I has oxidatively

damaged subunits and is functionally impaired and misassembled.

J. Neurosci. 2006, 26, 5256.

[53] Qi, X.; Lewin A. S.; Hauswirth, W. W. y Guy, J.

Suppression of complex I gene expression induces optic neuropathy.

Ann. Neurol. 2003, 53, 198.

[54] Kim, S. H.; Vlkolinsky, R.; Cairns, N.; Fountoulakis, M. y Lubec, G.

The reduction of NADH ubiquinone oxidoreductase 24- and 75-kDa

subunits in brains of patients with Down syndrome and Alzheimer’s

disease.

Life Sci. 2001, 68, 2741.

[55] Scacco, S.; Petruzzella, V.; Budde, S.; Vergari, R.; Tamborra, R.

Panelli, D.; Van den Heuvel, L.P.; Smeitink, J. A. y Papa, S.

Pathological mutations of the human NDUFS4 gene of the 18-kDa

(AQDQ) subunit of complex I affect the expression of the protein and

the assembly and function of the complex.

J. Biol. Chem. 2003, 278, 44161.

[56] Bénit, P.; Beugnot, R.; Chretien, D.; Giurgea, I.; De Lonlay-Debeney,

P.; Issartel, J. P.; Corral-Debrinsky, M.; Kerscher, S.; Rustin, P.;

Rötig, A. y Munnich, A.

Mutant NDUFV2 subunit of mitochondrial complex I causes early

onset hypertrophic cardiomyopathy and encephalopathy.

Human mutation 2003, 21, 582.

Referencias

48

[57] Ohkubo, E.; Aida, K.; Chen, J.; Hayashi, J. I.; Isobe, K.; Tawata, M. y

Onaya, T.

A patient with type 2 diabetes mellitus associated with mutations in

calcium sensing receptor gene and mitochondrial DNA.

Biochem. Biophys. Res. Commun. 2000, 278, 808.

[58] Ben-Shachar, D.

Mitochondrial dysfunction in schizophrenia: a possible linkage to

dopamine.

J. Neurochem. 2002, 83, 1241.

[59] Briet, F.; Twomey, C. y Jeejeebhoy, K. N.

Effect of feeding malnourished patients for 1 mo on mitochondrial

complex I activity and nutritional assessment measurements.

Am. J. Clin. Nutr. 2004, 79, 787.

[60] Merlo-Pich, M.; Raule, N.; Catani, L.; Fagioli, M. E.; Faenza, I.;

Cocco, L. y Lenaz, G.

Increased transcription of mitochondrial genes of complex I in human

platelets during ageing.

FEBS Lett. 2004, 558, 19.

[61] Estornell, E.

Mitochondrial complex I: new insights from inhibitor assays.

Protoplasma 2000, 213, 11.

[62] Jewess, P. J. y Devonshire, A. L.

Kinetic microplate-based assays for inhibitors of mitochondrial

NADH:ubiquinone oxidoreductase (complex I) and

succinate:cytochrome c oxidoreductase.

Anal. Biochem. 1999, 272, 56.

Referencias

49

[63] Li, N.; Ragheb, K.; Lawler, G.; Sturgis, J.; Rajwa, B.; Melendez, J. A.

y Robinson, J. P.

Mitochondrial complex I inhibitor rotenone induces apoptosis through

enhancing mitochondrial reactive oxygen species production.

J. Biol. Chem. 2003, 10, 8516.

[64] Tormo, J. R.; Gallardo, T.; González, M. C.; Bermejo, A.; Cabedo, N.;

Andreu, I. y Estornell, E.

Annonaceous acetogenins as inhibitors of complex I.

Curr. Top. Phytochem. 1999, 2, 69.

[65] Granell, S.; Andreu, I.; Marti, D.; Cavé, A.; Aragon, R.; Estornell, E.;

Cortes, D. y Zafra-Polo, M. C.

Bisbenzyltetrahydroisoquinolines, a new class of inhibitors of the

mitochondrial respiratory chain complex I.

Planta Med. 2004, 70, 266.

[66] Andreu, I.; Cabedo, N.; Tormo, J. R.; Bermejo, A.; Mello, R. y Cortes,

D.

Synthesis of N-diisopropyl phosphoryl benzyltetrahydroisoquinoline,

a new class of mitochondrial complexes I and III inhibitors.

Bioor. Med. Chem. Lett. 2000, 10, 1491.

[67] Carlsson, A.

Dopamine receptor agonists: intrinsic activity vs state of receptor.

J. Neural. Transm. 1983, 57, 309.

[68] Clarck, D.; Hjorth, S. y Carlsson, A.

Dopamine receptor agonists: mechanisms underlying autoreceptor

selectivity. I. Review of the evidence.

J. Neural. Transm. 1985, 62, 151.

[69] Clarck, D.; Hjorth, S. y Carlsson, A.

Dopamine receptor agonists: mechanisms underlying autoreceptor

selectivity. II. Theoretical considerations.

J. Neural. Transm. 1985, 62, 171.

Referencias

50

[70] Tamminga, C. A.

Partial dopamine agonists and the treatment of psychosis.

Curr. Neuropharmacol. 2005, 3, 3.

[71] Marsden, C. A.

Dopamine: the rewarding years.

Br. J. Pharmacol. 2006, 147, S136.

[72] Gainetdinov, R. R. y Caron, M. G.

Monoamine transporters: from genes to behaviour.

Annu. Rev. Pharmacol. Toxicol. 2003, 43, 261.

[73] Zhang, A.; Neumeyer, J. L. y Baldessarini, R. J.

Recent progress in development of dopamine receptor subtype-

selective agents: potential therapeutics for neurological and

psychiatric disorders.

Chem. Rev. 2007, 107, 274.

[74] Kebabian, J. W. y Calne, D. B.

Multiple receptors for dopamine.

Nature 1979, 277, 93.

[75] Missale, C.; Nash, S. R.; Robinson, S. W.; Jaber, M. y Caron, M. G.

Dopamine receptors: from structure to function.

Physiol. Rev. 1998, 78, 189.

[76] Strange, P. G.

Dopamine receptors.

Tocris Cookson 1997, 1.

[77] Sunahara, R. K.; Guan, H. C.; O´Dowd, B. F.; Seemann, P.; Laurier,

L. G.; Ng, G.; George, S. R.; Torchia, J.; Van Yol, H. H. M. y Niznik,

H. B.

Cloning for a gene for human dopamine D5 receptor with higher

affinity for dopamine than D1.

Nature 1991, 350, 614.

Referencias

51

[78] Levant, B.

The D3 dopamine receptor: neurobiology and potential clinical

relevance.

Pharmacol. Rev. 1997, 49, 231.

[79] Protais, P.; Cortes, D.; Pons, J. L.; López, S.; Villaverde, M. C. y

Castedo, L.

Displacement activity of some natural cularine alkaloids at striatal

[3H]-SCH 23390 and [

3H]-raclopride binding sites.

Experientia 1992, 48, 27.

Resumen-Discusión de los resultados

Resumen-Discusión Resultados

55

En la presente Tesis Doctoral hemos realizado la síntesis

y evaluación biológica de isoquinoleínas (IQ) con esqueletos diversos,

basándonos en los resultados previos obtenidos con este tipo de

alcaloides aislados en la naturaleza. En este sentido, las IQ son uno de

los metabolitos secundarios activos (MSA) más abundantes y de

mayor interés biológico de los aislados en especies de una de las

familias botánicas más estudiadas por nuestro grupo de investigación:

las Annonáceas.

En el presente trabajo hemos estudiado diversas rutas

sintéticas que nos han conducido a la preparación de varias series de

isoquinoleínas, que han sido evaluadas frente a tres dianas biológicas:

a) Pirrolo[2,1-a]isoquinoleínas antimicrobianas

b) Pirido y oxazino[2,1-a]isoquinoleínas

inhibidoras de la cadena respiratoria mitocondrial

c) Alcaloides aporfínicos y fenantrénicos con

afinidad por los receptores dopaminérgicos

Los artículos a los que ha dado lugar la presente Tesis

Doctoral se pueden dividir en cuatro capítulos:


56

Capítulo I: Las Annonáceas: fuente de inspiración para la

obtención de nuevos medicamentos

Capítulo II: Pirrolo[2,1-a]isoquinoleínas antimicrobianas

Capítulo III: Pirido y oxazino[2,1-a]isoquinoleínas

inhibidoras de la cadena respiratoria mitocondrial

Capítulo IV: Alcaloides aporfínicos y fenantrénicos

dopaminérgicos. Modelización molecular de IQs 1-sustituidas y

dopamina.


57

CAPÍTULO I

Artículo 1: “Las Annonáceas: fuente de inspiración

para la obtención de nuevos medicamentos” (En: Annonáceas, plantas

antiguas, estudios recientes. Universidad de Ciencias y Artes de Chiapas, México.

2012)

La Farmacoquímica Natural estudia los principios activos

obtenidos a partir de materias primas de origen natural, ya sean

vegetales, animales o microorganismos. Nuestro grupo de

investigación ha trabajado desde hace años en el campo de la

Farmacoquímica Natural, prestando una especial atención al estudio

de especies pertenecientes a la familia de las Annonáceas.

Esta familia botánica presenta un gran potencial

terapéutico, ya que posee numerosos MSA con potentes actividades

farmacológicas. Las Annonáceas son conocidas tanto por su interés

económico (frutos comestibles, condimentos alimentarios, aceites

esenciales), como por su utilización en medicina popular y en

agroquímica (como agentes antiparasitarios y plaguicidas).

Los MSA más abundantes y de mayor interés en esta

familia botánica son las acetogeninas y los alcaloides

isoquinoleínicos, pero también cabe destacar la presencia de

benzopiranos y estiril-lactonas entre otros. La síntesis de alcaloides IQ


58

con similitud estructural a los encontrados en la naturaleza, y

especialmente en Annonáceas, es el principal objetivo de la presente

Tesis Doctoral.

CAPÍTULO II

Artículo 2: “Synthesis of new antimicrobial pyrrolo

[2,1-a]isoquinolin-3-ones” (En: Bioorganic & Medicinal Chemistry, 2012, 20,

6589)

Las pirrolo[2,1-a]isoquinoleínas son MSA poco frecuentes

en la naturaleza, presentan una original estructura y han mostrado

actividad antimicrobiana. La búsqueda de nuevos antibióticos y

antifúngicos que puedan sustituir a fármacos poco eficaces, a causa de

las resistencias generadas por los microorganismos, fue lo que nos

impulsó a realizar la síntesis de nuevas pirrolo[2,1-a] isoquinoleínas

con actividad antimicrobiana.

Figura 1. Mecanismo de ciclación de pirrolo[2,1-a]isoquinoleínas

Existen diversos métodos de síntesis de

pirroloisoquinoleínas. Nosotros hemos desarrollado un nuevo

procedimiento basado en una doble ciclación intramolecular a partir


59

de β-fenilacetamidas (Figura 1). Mediante este método, las

pirroloisoquinoleínas fueron obtenidas con buenos rendimientos.

Utilizando este procedimiento, realizamos la síntesis de

tres series de pirrolo[2,1-a] isoquinoleínas (Figura 2). Entre ellas, se

sintetizó el compuesto natural (±)-trollina (3b).

Figura 2. Pirrolo[2,1-a]isoquinoleínas antimicrobianas

Los trece compuestos sintetizados fueron evaluados como

bactericidas y antifúngicos, mediante la técnica de difusión en agar.

En lo que respecta a los test bactericidas, nuestros compuestos fueron

ensayados frente a diversas bacterias patógenas humanas: B. cereus, S.

Aureus, E. faecalis, S. typhii, E. coli y E. carotovora.


60

En general, pudimos observar que la introducción de un

grupo lipofílico en posición 8 y/o 9 resultó adecuada para la actividad

antimicrobiana, en comparación con los compuestos con hidroxilos

libres en estas posiciones, como en el caso de (±)-trollina (3b). El

compuesto que presentó una actividad más relevante fue 2a, que posee

un átomo de cloro en posición 8 y un grupo metoxilo en posición 9.

Este compuesto demostró actividad bactericida frente a todas las cepas

ensayadas. Además, mostró el mayor halo de inhibición frente a S.

aureus y E. carotovora de entre todas las pirroloisoquinoleínas

sintetizadas. En el caso de ésta última, la potencia bactericida de 2a

fue equiparable a la del compuesto de referencia (tetraciclina).

La actividad fungicida fue comprobada frente a algunas

cepas de hongos fitopatógenos: A. parasiticus, T.viridae, F.

culmorum, G. candidum y P. citrophthora. Dentro de la serie 1, las

pirroloisoquinoleínas más activas fueron 1a y su análogo fluorado 1e.

Estos compuestos mostraron una potencia similar entre ellos, lo que

sugiere que el grupo O-bencílico en posición 8 de las

pirroloisoquinoleínas contribuye positivamente a sus propiedades

antifúngicas, independientemente de que contenga o no un grupo

halógeno. El compuesto 2a, el más activo en los ensayos bactericidas,

demostró un efecto antifúngico moderado.


61

En conclusión, cabe destacar la actividad bactericida de la

pirroloisoquinoleína 8-cloro-9-metoxi, 2a, y la fungicida de las

pirroloisoquinoleínas 8-bencílicas, 1a y 1e.

CAPÍTULO III

Artículo 3: “Synthesis of pirido[2,1-a]isoquinolin-4-

ones and oxazino[2,3-a]isoquinolin-4-ones: New inhibitors of

mitocondrial respiratory chain” (En: European Journal of Medicinal

Chemistry 2013, 69, 69)

La cadena respiratoria mitocondrial (CRM) de transporte

electrónico comprende cuatro complejos enzimáticos (del I al IV), el

coenzima Q (ubiquinona) y el citocromo C. Este sistema juega un

papel esencial en la síntesis del ATP, el metabolismo de las especies

reactivas del oxígeno y la apoptosis. Por ello, sus inhibidores

presentan importantes perspectivas para convertirse en nuevos agentes

antitumorales.

Los alcaloides benzo[a]quinolicidínicos poseen

interesantes propiedades biológicas, así como una similitud estructural

con algunos inhibidores de la CRM conocidos. En este trabajo

decidimos realizar la síntesis de moléculas con dicho esqueleto. Entre

los métodos descritos en la literatura, elegimos el de Pemberton,

puesto que nos permitía explorar la reactividad de las iminas y

aprovecharnos de la diversidad estructural que proporcionaba, dado


62

que con él se obtienen pirido[2,1-a]isoquinoleínas y oxazino[2,1-

a]isoquinoleínas.

Figura 3. Pirido y oxazino[2,1-a]isoquinoleínas inhibidoras de la CRM

Las diferentes iminas se sintetizaron mediante la ciclación

de Bischler-Napieralski y se hicieron reaccionar con diferentes

dioxinonas. El producto mayoritario obtenido de esta reacción

depende del pH del medio: en condiciones neutras, los productos

mayoritariamente aislados fueron las pirido[2,1-a]isoquinoleínas,

mientras que al añadir una base, se formaron preferentemente las

oxazino[2,1-a]isoquinoleínas. Siguiendo este procedimiento se


63

sintetizaron tres series de compuestos, a partir de tres iminas

diferentes, dando un total de 11 compuestos a ensayar como posibles

inhibidores de la CRM.

En general, las oxazino[2,1-a]isoquinoleínas mostraron

mayor potencia que sus respectivos análogos piridínicos. Por ejemplo:

el compuesto 1b (Figura 3) mostró una CI50 de 8.33 M mientras que

su análogo oxigenado 1c presentó una capacidad de inhibición cinco

veces superior. Una tendencia similar se observa en todos los

productos ensayados. En cuanto al sustituyente en posición 2, un

grupo fenilo mantuvo la actividad (1e) y un anillo furánico

proporcionó el compuesto más activo, 1g. La desprotección de los

metoxilos del anillo A (1h) también mantuvo la actividad. En el

mismo sentido, cuando se introdujo un anillo aromático en posición 1

(Series 2 y 3), las pirido[2,1-a]isoquinoleínas (2b, 3b) inhibieron la

CRM con potencias moderadas, mientras que sus respectivos análogos

oxazino[2,1-a]isoquinoleínicos (2c, 3c) mostraron una mayor

actividad. La presencia de un átomo de bromo sobre el anillo

aromático no afectó significativamente a su actividad.

Comparado con el compuesto de referencia, rotenona (un

potente inhibidor del complejo I, altamente tóxico), tanto las pirido

como las oxazino[2,1-a]isoquinoleínas mostraron actividad inhibitoria

de la CRM con potencias moderadas. Dado que la mayoría de

inhibidores de la CRM con potenciales aplicaciones en terapéutica


64

actúan en el rango alto nanomolar o bajo micromolar, estos

compuestos, y especialmente las oxazino[2,1-a]isoquinoleínas, pueden

ser considerados efectivos inhibidores de la CRM.

CAPÍTULO IV

Artículo 4: “3,4-Dihydroxy- and 3,4-methylenedioxy

phenanthrene-type alkaloids with high selectivity for D2 dopamine

receptor” (En: Bioorganic and Medicinal Chemistry Letters, 2013, 23, 4824)

La neurotransmisión mediada por dopamina juega un

papel muy importante en diversos desórdenes psiquiátricos y

neurológicos. Modular la vía dopaminérgica a través de los receptores

de la dopamina es un medio potencial para tratar enfermedades como

la esquizofrenia y el Parkinson.

Dado que existen dos importantes familias de receptores

dopaminérgicos (RD) (tipo-D1 y tipo-D2), ha surgido el interés en

encontrar ligandos que a la vez tengan afinidad por estos receptores y

dispongan de una elevada selectividad, dado que de esta manera se

pueden evitar los efectos adversos asociados a la unión al receptor no

deseado. Desde un punto de vista terapéutico, los receptores tipo-D2

han adquirido una especial relevancia puesto que los antagonistas D2

son los fármacos más empleados en el tratamiento de la esquizofrenia,


65

depresión y trastorno bipolar, y los agonistas en la enfermedad de

Parkinson.

Figura 4. Aporfinas y fenantrenos sintetizados

Los alcaloides fenantrénicos son buenos candidatos para

unirse a los receptores dopaminérgicos debido a su similitud

estructural a la dopamina y la disposición libre de su cadena lateral

aminoetílica. La secuencia general para la preparación de estos

compuestos se basó en la síntesis de las fenilacetamidas, seguida de la

ciclación de Bischler-Napieralski. De esta forma se obtuvieron las 1-

bencil-THIQ, que mediante una reacción de arilación directa

catalizada por paladio permitieron obtener las aporfinas

correspondientes. La N-metilación de éstas últimas dio lugar a las

sales de amonio cuaternarias, que fueron sometidas a la degradación

de Hofmann, obteniéndose así los alcaloides fenantrénicos.


66

Todos los compuestos sintetizados (Figura 4), excepto los

cuaternarios, fueron capaces de desplazar los radioligandos

específicos para cada receptor (D1: 3H-SCH 23390 y D2:

3H-

racloprida) a concentraciones micro o nanomolares.

Las aporfinas no cuaternarias (3b, 3d, 3e) mostraron una

afinidad similar por los receptores D1. La afinidad por los receptores

D2 fue muy superior, especialmente en las moléculas con OH libres o

grupos metilendioxi (3d, 3e). Un comportamiento similar fue

observado en los fenantrenos no cuaternarios.

En conclusión, tanto las aporfinas como los fenantrenos,

poseen afinidades por los receptores D2 de rango nanomolar. Cabe

destacar que los alcaloides fenantrénicos, poseen una mayor

selectividad, como puede observarse en la Figura 5, dado que su

afinidad por los receptores D1 es menor que en el caso de las

aporfinas.

Figura 5. Curvas de desplazamiento de radioligandos para 3d y 4b


67

La gran selectividad de la que disponen estos compuestos,

especialmente los alcaloides fenantrénicos, les puede permitir evitar

los efectos adversos asociados a la estimulación de los receptores D1 y

por tanto ser aplicados en terapéutica con mayor eficacia.

Artículo 5: “Tetrahydroisoquinolines acting as

dopaminergic ligands. A molecular modeling study using MD

simulations and QM calculations” (En: Journal of Molecular Modelling,

2012, 18, 419)

La utilización de técnicas computacionales y de

modelización molecular ha reducido los numerosos ensayos que se

deben de realizar para obtener un nuevo fármaco. Mediante estudios

de modelización molecular se han podido determinar las partes de una

molécula responsables de la actividad farmacológica.

En el presente trabajo, se realizó un estudio de

modelización molecular sobre 16 THIQ 1-sustituidas como ligandos

dopaminérgicos. Se calcularon las energías de unión de las IQ al

interaccionar con los receptores dopaminérgicos D2.

Nuestro estudio de modelización molecular se realizó en

dos pasos. En primer lugar, se llevaron a cabo simulaciones de las

interacciones entre los compuestos ensayados y los receptores


68

dopaminérgicos D2. En un segundo paso, sistemas modelo reducidos

fueron optimizados usando cálculos de mecánica cuántica.

Comparando los resultandos obtenidos sobre los diferentes

complejos ligando-receptor, nuestras simulaciones nos proporcionaron

interesantes conclusiones generales. Destacó la importancia del Asp

86 negativamente cargado para la unión de los ligandos ensayados. El

grupo carboxilo terminal puede considerarse como punto de anclaje

para ligandos que poseen un grupo amino protonado (Figura 6).

NH

Cl

HO

9

Figura 6. Unión de la 1-bencil-THIQ 9 al receptor dopaminérgico D2

Los datos de afinidad muestran que los grupos hidroxilo

son de vital importancia para estabilizar la unión a los receptores D2 a

través de restos de serina. No obstante, los residuos de Ser 141 y 144

de los receptores D2 pueden no ser igualmente importantes para la


69

afinidad. Mutaciones individuales de los dos residuos en la hélice

transmembrana 5 produjeron efectos dispares en la unión al receptor.

Estos resultados demostraron que la Ser 141 es otro punto importante

para la unión de estas IQ al RD.

Artículo 6: “2-(3chloro-4-hydroxyphenyl)ethylamine

acting as ligand of the D2 dopamine receptor. Molecular

modelling, synthesis and D2 receptor affinity”

En este trabajo, el estudio de modelización molecular se

realizó en tres pasos. En el primero, se evaluaron las interacciones

entre los compuestos y el RD D2 mediante simulaciones de dinámica

molecular. En el segundo, se hizo uso de sistemas modelo reducidos.

Por último, los complejos más representativos fueron analizados a

través de un estudio QTAIM, utilizando cálculos DFT.

Figura 7. Dopamina y análogos clorados

De las energías de unión obtenidas en las simulaciones se

puede observar que la sustitución de un OH en posición meta respecto

a la cadena aminoetílica por un átomo de cloro (Figura 7, compuesto

3), aumenta la afinidad por el RD D2 en comparación con la dopamina


70

(1). Sin embargo, la sustitución del OH en posición para por el cloro

(2) ejerce un efecto contrario.

Figura 8. Energías de interacción para compuesto 3

Como ya se había sugerido en el Artículo 5, la existencia

de un resto carboxilo en el lugar de unión es muy importante para la

interacción ligando-receptor, puesto que la fuerte interacción de los

ligandos con el Asp 114 se mantuvo para todos los complejos

evaluados. Respecto a la Val 115 (Figura 8), este aminoácido forma

diversas interacciones con el anillo aromático de la dopamina. En

cuanto a la serina, la Ser197 aporta la mayor contribución a las

energías de interacción de los complejos con los compuestos que

poseen un OH en meta (1 y 2), mientras que la Ser 193 es más

importante para los compuestos que tienen un cloro en esta posición

(3). Por otro lado, el análisis de descomposición de residuos también


71

revela una destacable contribución de la Phe 389, siendo más

importante en el caso de los compuestos con OH en para (1 y 3).

English version

English version

75

In the present Thesis, we have carried out the synthesis

and the biological evaluation of several isoquinolines (IQ) with

diverse structures based on previous natural occurring IQ. These

alkaloids are one of the most abundant and relevant metabolites

isolated from the Annonaceae family.

In the present work we have studied several synthetic

routes that led us to different types of isoquinolines, and each type of

synthesized isoquinolines has been tested on different biological

activities:

a) Antimicrobial pyrrolo[2,1-a]isoquinolines

b) Pyrido and oxazino[2,1-a]isoquinolines as

mitocondrial respiratory chain inhibitors

c) Dopaminergic aporphines and phenantrene

alkaloids

This Thesis has generated six original articles which have

been classified in four chapters:

Chapter I: Annonaceae as an inspiration source for new

medicinal products generation

Chapter II: Antimicrobial pyrrolo[2,1-a]isoquinolines

Chapter III: Pyrido[2,1-a]isoquinolines and oxazino

English version

76

[2,1-a]isoquinolines as inhibitors of mitochondrial

respiratory chain

Chapter IV: Dopaminergic aporphine and phenanthrene

alkaloids. Molecular modeling of 1-substituted isoquinolines.

English version

77

CHAPTER I

Article 1: “Las annonáceas: fuente de inspiración para

la obtención de nuevos medicamentos” (In: Anonáceas, plantas antiguas,

estudios recientes. Universidad de Ciencias y Artes de Chiapas, México. 2012)

Natural Pharmaco-chemistry studies the active principles

obtained from natural sources regardless their plant, animal or

microbial origin. Our research group has been working in the Natural

Pharmaco-chemistry field for many years, paying special attention to

the Annonaceae family.

This botanic family presents an enormous therapeutic

potential, since plants belonging to this family possess many active

secondary metabolites with potent pharmacologic activities.

Annonaceae family is known either for its economical interest (edible

fruits, alimentary condiments, essential oils) or its uses in popular

medicine and agro-chemistry (anti-parasitic, pesticides).

The most remarkable secondary metabolites in this family

are acetogenins and isoquinoline alkaloids, but the presence of

benzopyranes and stiril-lactones is also noteworthy. The synthesis of

new isoquinoline alkaloids with similar structure to those found in

nature, and more specifically in Annonaceae family, is the main aim

of this Thesis.

English version

78

CHAPTER II

Article 2: “Synthesis of new antimicrobial pyrrolo[2,1-

a]isoquinolin-3-ones” (In: Bioorganic & Medicinal Chemistry, 2012, 20, 6589)

Pyrroloisoquinolines present an original structure and they

have been reported as antimicrobial products. Inasmuch, the discovery

of new antimicrobial agents has turned into a clear need in the last few

years due to the development of resistance to classic drugs as a

consequence of their extensive use.

Figure 1. Mechanism of cyclization of pyrrolo[2,1-a]isoquinolines

Although several methods have been employed for the

synthesis of this structure, we have applied a new methodology based

on double intramolecular cyclization, conducted by Bischler-

Napieralski cyclodehydration from an ester phenylethylamide. It

involves the subsequent reduction of the imine intermediate (Figure

1).

Using this method we performed the synthesis of three

series of compounds with pyrroloisoquinoline structure, obtaining a

English version

79

total of 13 different compounds. The natural product (±)-trolline (3b)

was synthesized, among other compounds.

Figure 2. Antimicrobial pyrrolo[2,1-a]isoquinolines

All the synthesized pyrrolo[2,1-a]isoquinolines were

assayed in vitro for their ability to inhibit bacterial and fungal growth.

In the antimicrobial assays, our compounds were tested against several

human pathogenic bacteria as well as economically important

phytopathogenic bacteria and/or fungi. The bacterial agents were: B.

cereus, S. aureus, E. faecalis, S. typhii, E. coli and E. carotovora.

In general, a lipophilic group at the 8- and/or 9-position

seemed to provide moderate activity compared with that displayed by

a free hydroxyl group such as 1b and (±)-trolline (Figure 2, 3b). The

most noteworthy compound was 2a, which possessed both a chlorine

English version

80

atom and a methoxyl group at the 8- and 9-position, respectively. This

compound exerted bactericidal activity against all the tested strains.

Furthermore, 2a showed the highest inhibition zone against S. aureus

and E. carotovora among all the tested pyrroloisoquinolines. For the

latter microorganism, 2a showed a potency within the range displayed

by the reference compound (tetracycline).

Fungicide activity was tested against some

phytopathogen fungi strains: A. parasiticus, T. viridae, F. culmorum,

G. candidum and P. citrophthora. Compounds 1a, 2a, 1c and 1e

inhibited fungal growth in vitro. In series 1, the most active

compounds were 1a and its fluorinated analog 1e, which showed

similar potency. Consequently, it seems that the benzylic moiety

located at the 8-position on the pyrroloisoquinoline structure

contributed positively to its antifungal properties regardless the

presence of a halogen atom. By contrast, compound 2a, which was the

most potent bactericidal agent against human pathogens, only exerted

a moderated antifungal activity.

Therefore, 8-chloro-9-methoxy compound, 2a, was the

most relevant compound in bactericidal assays and 8-benzyl

pyrroloisoquinolines 1a and 1e in the fungicidal assays.

English version

81

CHAPTER III

Article 3: “Synthesis of pyrido[2,1-a]isoquinolin-4-ones

and oxazino[2,3-a]isoquinolin-4-ones: New inhibitors of

mitocondrial respiratory chain” (in: European Journal of Medicinal

Chemistry 2013, 69, 69)

Mitochondrial electron transport chain, comprises the

enzymatic respiratory complexes (I-IV), the coenzyme Q (ubiquinone)

and cytochrome c. Mitochondria play essential roles in the ATP

synthesis, ROS metabolism and apoptosis. It is important that cells

preserve an appropriate level of intracellular ROS to keep redox

balance and signaling cellular proliferation. Some anticancer agents

have the ability to inhibit mitocondrial electron transport and/or

increase superoxide radical generation in tumor cells, resulting in cell

apoptosis through ATP levels reduction and/or cell-damaging.

Benzo[a]quinolizine is an important heterocyclic

framework that can be found in numerous bioactive compounds.

Benzo[a]quinolizines are structurally related to some mitocondrial

respiratory chain inhibitors (mycotoxins, such as circumdatins) and in

this study we have carried out the synthesis of these compunds.

Among the different depicted approaches, we have applied the

Pemberton method that uses dioxinones as acyl-ketenes precursors.

Therefore, taking advantage of the chemical diversity generated by

English version

82

this procedure, different pyrido[2,1-a]isoquinolines and oxazino[2,1-

a]isoquinolines were obtained.

Figure 3. Pyrido and oxazino[2,1-a]isoquinolines

Several imines were prepared using Bischler-Napieralski

ciclyzation and then reacted with dioxinones. The major product

obtained relied on the pH of the medium. Under neutral conditions,

the formation of pyridones was clearly favored, and under basic

conditions oxazinones were preferentially formed. Following this

procedure, we synthesized three series of compounds, starting from

English version

83

three different imines, to get a total of 11 compounds. Then, their

potential effect on respiratory chain inhibition was evaluated.

In general, oxazino[2,3-a]isoquinolin-4-ones seemed to

display greater activity than their respective pyrido[2,1-a]isoquinolin-

4-ones analogues. For instance, while compound 1c (Figure 3) IC50

was 1.36 µM, its pyridone analogue (1b) showed decreased NADH

oxidase activity (IC50=8.33 µM). By contrast, the size and nature of

the substituent at 2-position of the oxazine did not seem to affect

NADH oxidase activity. In this regard, not only the existence of a

phenyl group (1e) kept the activity but the presence of a furan ring at

C2 position provided the most potent compound 1g. Moreover, the

deprotection of the methoxy groups in the A ring to obtain a catechol

(1h), maintained the activity. On the other hand, when the aromatic

ring was placed on C1 position (series 2 and 3), the pyrido and

oxazino[2,1-a]isoquinolin-4-ones were able to inhibit respiratory

chain without any noticeable influence of the halogen group.

Compared with the highly toxic compound, rotenone (a

high-affinity inhibitor of complex I), the pyridone and oxazinone

derivatives showed a moderated inhibitory activity. Given that most of

the respiratory chain inhibitors with potential therapeutic interest act

in the high nanomolar and low micromolar ranges, these compounds

and especially the oxazinone derivatives, can be considered effective

inhibitors of the whole respiratory chain.

English version

84

CHAPTER IV

Article 4: “3,4-Dihydroxy- and 3,4-methylenedioxy

phenanthrene-type alkaloids with high selectivity for D2 dopamine

receptor” (In: Bioorganic and Medicinal Chemistry Letters, 2013, 23, 4824)

Dopamine-mediated neurotransmission plays an important

role in several psychiatric and neurological disorders. Researchers

have focused on various approaches towards modulating

dopaminergic activity via the dopamine receptors (DR) as a potential

means for treating schizophrenia and Parkinson’s diseases.

There are two important families of DR: D1-like and D2-

like. Nowadays there is a big interest in finding ligands which not only

posses affinity for these receptors but also display good selectivity in

order to avoid adverse effects associated to binding to the undesired

receptor. From a therapeutic point of view, drugs acting at D2-like DR

have become very relevant since most used antipsychotics in

schizophrenia or bipolar disease treatment are D2 antagonists, and they

are also involved in dopamine’s release.

English version

85

Figura 4. Aporphine and phenanthrene alkaloids synthesized

Phenanthrene alkaloids are good candidates to bind to DR

with high affinities due to their structural similarity to dopamine and

the flexible disposition of their aminoethyl chain. The general

synthetic plan for obtaining these compounds consisted in preparing

the appropriate β-phenyl acetamide to be then cyclized by Bischler-

Napieralski cyclodehydration and followed by imine reduction. This

sequence led us to obtain the expected 1-(2’-bromobenzyl)-

tetrahydroisoquinolines, which were then subjected to direct arylation

to generate aporphines. Aporphines were then N-methylated to obtain

ammonium salts which, under basic Hofmann’s degradation

conditions, gave the corresponding phenanthrene derivatives (Figure

4).

English version

86

All the tested compounds, except those with a quaternary

amonium, were able to displace both 3H-SCH 23390 and

3H-

raclopride from their specific binding sites in rat striatum at

micromolar or nanomolar concentrations.

The non-quaternary aporphine derivatives (3b, 3d, 3e)

displayed similar potency against D1 receptor. Interestingly, the

affinity was substantially increased towards D2 receptor, with Ki

values reaching the nanomolar range, when the oxygenated functions

were in the catechol or the methylenedioxy form (3d, 3e). A similar

behavior was observed in the non-quaternary phenanthrene alkaloids.

In conclusion, both aporphines and phenanthrenes possess

similar affinities towards D2 DR, but phenantrene alkaloids displayed

higher selectivity than aporphines’ since they showed lower affinity

for D1 DR (Figure 5).

English version

87

Figura 5. Selectivity of compounds 3d and 4b at the D2-like dopaminergic

receptors ([3H]-raclopride binding)

The therapeutic potential of these compounds is therefore

noticeable due to the pharmacological relevance of D2 ligands as

potential antipsychotic or anti-Parkinson's drugs. Inasmuch, the high

selectivity of the phenanthere alkaloids synthesized has may be

revealed a new therapeutic class of drugs since they lack of D1-

associated adverse affects.

Article 5: “Tetrahydroisoquinolines acting as

dopaminergic ligands. A molecular modeling study using MD

simulations and QM calculations” (In: Journal of Molecular Modelling,

2012, 18, 419)

The use of computational techniques and molecular

modeling, have reduced the number of assays necessary to obtain new

bioactive molecules. Using molecular modelling, it has become

possible to indentify the parts of a molecule that are implied in its

pharmacological effects.

In the present work, a molecular modeling study on 16 1-

substituted-THIQ as dopaminergic ligands was performed. Binding

energies of BTHIQs interacting with the human dopamine D2 receptor

were calculated.

English version

88

Theoretical calculations were carried out in two steps. In a

first step, MD simulations of the molecular interactions between

compounds 1–16 and D2 DR we performed. In the second step,

reduced model systems were optimized using quantum mechanics

calculations.

The comparisons between the results obtained for the

different complexes, led to interesting general conclusions. Consistent

with previous experimental and theoretical data, the simulations

revealed the importance of the negatively charged Asp 86 for ligand

binding. In this regard, the terminal carboxyl group seems to function

as an anchoring point for ligands with protonated amino groups

(Figure 6).

NH

Cl

HO

9

Figure 6. Compound 9 binding to D2 RD

English version

89

Affinity data indicate that the hydroxyl groups of

dopaminergic ligands are key groups for binding stabilization and

further suggest that serine residues 141 and 144 of the D2 receptor

may not be equally important for binding affinity. Individual mutation

of serines 141 and 144 in TM5 to alanine produced asymmetrical

effects on dopamine receptor binding. These results indicate that Ser

141 might be an important group for isoquinoline binding to DR.

Article 6: “2-(3chloro-4-hydroxyphenyl)ethylamine


modelling, synthesis and D2 receptor affinity”

This study was carried out in three steps. First, a molecular

dynamic simulations to evaluate the molecular interactions between

the compounds and the D2 DR were performed. Second, the reduced

model systems were optimized. And third, the most representative

complexes obtained in the previous steps were further analyzed from a

QTAIM study using DFT calculations.

Figure 7. Dopamine and chlorinated analogues

English version

90

From the relative binding energies obtained in the MD

simulations, it can be observed that the replacement of the OH group

in meta respect to the aminoethyl chain by a chlorine atom (Figure 7,

Compound 3) increases the affinity towards D2 DR when compared

with DA (1). In contrast, the replacement of the OH group in para

position (2) resulted in the opposite effect.

Figure 8. Interaction energies for compound 3

As it was previously suggested in Article 5, the strong

interaction with the carboxylic group of an aspartate is very important

for ligand’s binding since the interaction with Asp 114 was

maintained for all the tested complexes (Figure 8). On the other hand,

Val 115 can form several interactions with the DA aromatic ring. In

regard to other serine residues of the binding site, Ser 197 is the

residue that exerted the largest contribution to the interaction energy

English version

91

for complexes with 1 and 2. In addition, Ser 193 showed the most

important contribution in the complex with compound 3. Finally, the

residue decomposition analysis exhibited a significant contribution of

Phe 389. The contribution of this residue was greater for DA/D2 DR

or compound 3/D2 DR complexes than for compound 2/D2 DR

complex.

Anexo I: Artículos científicos

Capítulo I:

Las Annonáceas: fuente de inspiración

para la obtención de nuevos medicamentos

Artículo 1: “Las Annonáceas: fuente de inspiración para

la obtención de nuevos medicamentos” (En: Annonáceas,

plantas antiguas, estudios recientes. Universidad de

Ciencias y Artes de Chiapas, México. 2012)

Las Annonáceas: Fuente de inspiración para la

obtención de nuevos medicamentos

Nuria Cabedo#, Laura Moreno, Sonia López, Paloma Marín, Javier Párraga y Diego Cortes

Departamento de Farmacología, Laboratorio de Farmacoquímica, Facultad de Farmacia,

Universidad de Valencia, 46100 Burjassot, Valencia, España – www.farmacoquimicavalencia.es

La Farmacoquímica Natural estudia los principios activos obtenidos a partir de

materias primas de origen natural, ya sean vegetales, animales o microorganismos. Estos

principios activos son moléculas, también denominadas “metabolitos secundarios activos”

(MSA), que la naturaleza biosintetiza y que pueden llegar a utilizarse directamente como

fármacos, o constituir la fuente de inspiración en la preparación de nuevos medicamentos

de síntesis [1]. La Farmacoquímica Natural aborda por tanto el estudio de estos MSA

desde diferentes puntos de vista: 1) origen biológico (vegetal, animal y microorganismo);

2) origen biogenético; 3) propiedades fisico-químicas (solubilidad, polaridad, métodos de

extracción y purificación); 4) determinación estructural (Resonancia Magnética Nuclear,

espectrometría de masas y rayos X); 5) propiedades farmacológicas (actividad, mecanismo

de acción, REA); y 6) aplicaciones terapéuticas (enfermedades para las cuales estos

compuestos podrán ser utilizados) [1].

Un ejemplo representativo de fármaco de origen natural lo constituye la penicilina G.

Este MSA se aísla del hongo Penicillium chysogenum y se biosintetiza a partir de la

condensación de aminoácidos. Su estructura, presenta un grupo ácido carboxílico que

permite la extracción de la penicilina G en forma de sal, y se caracteriza por poseer como

farmacóforo un anillo β-lactama, además de contar con agrupamientos fundamentales

como son el anillo tiazolidina y un enlace amídico. Su capacidad de inhibir la síntesis de la

pared bacteriana, permiten su uso como antibiótico en diversas enfermedades infecciosas,

tales como ciertas amigdalitis, meningitis o enfermedades venéreas.

Otro ejemplo de fármaco de origen natural sería la morfina, obtenida de las cápsulas

de adormidera (Papaver somniferum). Su alta afinidad por los receptores opioides dota a

esta molécula de una potente actividad analgésica. Ambos MSA y sus análogos

estructurales, son alguno de los muchos ejemplos de fármacos de origen natural

ampliamente utilizados hoy día en terapéutica (Figura 1).

# Centro de Ecología Química Agrícola-Instituto Agroforestal del Mediterraneo, UPV, Campus de Vera, Edificio 6C,

46022 Valencia, España

2

Figura 1.

Estos productos naturales o MSA, y muchos otros, ocupan un papel fundamental en

la industria farmacéutica, ya que en la actualidad la mitad de los medicamentos prescritos

tienen un origen natural, directamente o mediante transformaciones químicas [2].

Concretamente, en el área del cáncer y enfermedades infecciosas se encuentran más del

60% y 75%, respectivamente, de fármacos de origen natural. Hoy en día, se está dando

especial relevancia a los microorganismos de origen marino, éstos ocupan una parte

extensa de la naturaleza todavía por explorar (> 70% de la superficie del planeta) dotada

de gran biodiversidad. Aunque pocos metabolitos procedentes del mar se encuentran

actualmente en el mercado o en ensayos clínicos, los organismos marinos representan la

principal fuente de nuevos productos bioactivos con potencial farmacéutico, y con una

diversidad inusual en sus estructuras químicas. La investigación de metabolitos de origen

marino se inició en los años 1950 por Werner Bergman y John Faulkner, y alcanzó mayor

interés en las últimas décadas. Cabe destacar de los estudios realizados por el Instituto

Nacional del Cáncer (EEUU), que alrededor de un 1.8% de los extractos de especies

marinas presentan algún tipo de actividad antitumoral, mientras que en organismos

terrestres sólo un 0.4% de los extractos presentan actividad. En el mar habitan numerosos

organismos invertebrados de cuerpo blando y/o sésiles capaces de sintetizar u obtener (a

partir de otros microorganismos) metabolitos secundarios tóxicos para defenderse de los

depredadores. Hasta el momento, se han encontrado más de 15000 productos naturales de

origen marino, incluyendo numerosos compuestos con potencial farmacéutico. Algunos de

estos compuestos ya se encuentran en el mercado, por ejemplo: el potente analgésico del

SNC Prialt® (ziconotida, un derivado sintético del péptido ω-conotoxina aislado del

molusco Conus magus) y el agente antineoplásico Yondelis® (trabectedina o ET-743, una

ecteinascidina aislada de un tunicado caribeño, Ecteinascidia turbinata); otros compuestos

antineoplásicos se encuentran en fase clínica, por ejemplo dentro del grupo de los

ciclodepsipéptidos destacan: Aplidin® (obtenido inicialmente mediante semisíntesis a

N

SHN H

COOHO

O

ácido

carboxílicoenlace amídico

β-lactama

anillo

tiazolidina

OHO

N

CH3

H

OH

penicilina G morfina

3

partir de didemnina A, un metabolito aislado a partir del tunicado Aplidium albicans),

Kahalalida F (aislado de un molusco marino sin caparazón Elysia rufescens y del alga

Bryopsis pennata que forma parte de la dieta del molusco) (Figura 2) y las dolastatínas

(pentapéptidos lineales que se obtienen del molusco Dolabella auricularia); en el grupo de

los alcaloides tipo pirroloquinolina encontramos las isobatzellinas A-D (aislados de

esponjas pertenecientes a distintos géneros, destacando Zyzzya y Batzella sp.).

N

OHN

NH O

HN

NH

H2N

O

O

HN

O

NH

O

O

HN

OOHN

OHNHN

O

O

HN

O

O

Val

Pro

OrnVal

Val

Phe

Figura 2. Kahalalida F

Algunos metabolitos de origen marino con propiedades antitumorales, son además

inhibidores de la cadena respiratoria mitocondrial (CRM), y por lo tanto de la síntesis de

ATP, pudiendo ser éste un posible mecanismo de su acción citotóxica. Un ejemplo son los

diterpenos aislados de la esponja Dysidea sp., puupehenona y haterumadienona, que

mostraron unas CI50 de 1.28 µM y 4.71 µM, respectivamente (Figura 3) [3].

O

O

OH

H

H

puupehenona heterumadienona

O

H

H

O

Figura 3.

El complejo I de la CRM es el mayor y más complejo de los sistemas enzimáticos

implicados en los procesos bioenergéticos de la membrana mitocondrial interna. Su

función es la oxidación del NADH formado en la matriz mitocondrial por las vías

catabólicas oxidativas, canalizando los equivalentes de reducción hacia el espacio

intermembrana. Los electrones siguen su vía hacia la reducción del oxígeno molecular a

4

través de los complejos III y IV, incrementando así el gradiente de protones en el espacio

intermembrana. Finalmente, el gradiente electroquímico generado es utilizado para la

generación de ATP por la ATPasa mitocondrial, proceso conocido como fosforilación

oxidativa.

En el presente trabajo, abordamos la obtención y el estudio de diversos tipos de

MSA aislados en especies de la familia Annonáceas. Las plantas pertenecientes a esta

familia se caracterizan por su gran contenido en diferentes esqueletos químicos con

múltiples propiedades farmacológicas. Los MSA descritos serán aquellos de carácter

“neutro” como las acetogeninas, estiril-lactonas, benzopiranos prenilados, y algunas

isoquinoleínas (amídicas); así como los MSA de carácter “iónico” de los que forman parte

las azafluorenonas y aquellas isoquinoleínas que contienen un nitrógeno secundario o

terciario, capaz de protonarse en condiciones ácidas.

1. Acetogeninas

Las plantas pertenecientes a la familia Annonáceas son conocidas desde hace tiempo

tanto por su interés económico (proporcionan frutos comestibles y aceites esenciales)

como por su utilización en medicina tradicional y popular (pesticidas y antiparasitarias).

Los MSA en gran parte responsables de estas propiedades, son las acetogeninas (ACG),

moléculas presentes exclusivamente en especies de la familia Annonáceas y que se

caracterizan tanto por su original estructura química, como por las actividades biológicas y

farmacológicas que son capaces de desarrollar. Entre ellas destacan las propiedades

NADH

Q

Cit c

COMPLEJO III COMPLEJO IV

ATPasa ADP + Pi ATP

½ O2 H2O

Complejo I = NADH:Ubiquinona oxidoreductasa Complejo II = Succinato:ubiquinona oxidoreductasa Complejo III = Ubiquinol:Citocromo c oxidoreductasa Complejo IV = Citocromo c oxidasa

Espacio intermembrana

Matriz mitocondrial COMPLEJO I

COMPLEJO II

5

citotóxicas, antitumorales, antiparasitarias e insecticidas [4]. La actividad citotóxica de las

ACG se debe fundamentalmente a la inhibición selectiva, de rango nanomolar, que ejercen

sobre el complejo I de la CRM [5]. Estructuralmente se caracterizan por presentar una

cadena de 32 o 34 átomos de C con varias funciones oxigenadas y una γ-lactona terminal.

Su origen biogenético es similar al de los ácidos grasos. En los últimos años, se ha llevado

a cabo una serie de estudios que han permitido postular la forma en que las ACG

interaccionan con el complejo I. Para conocer mejor que parte de la molécula interacciona

con dicho complejo, la estructura de las ACG se ha dividido en cuatro zonas: a) dominio

lactónico terminal, b) dominio espaciador (cadena hidrocarbonada entre la lactona y los

anillos THF), c) dominio polioxigenado (sistema THF), y d) dominio hidrofóbico terminal

(Figura 4).

threo/trans/threo/trans/erythro

O

15O O

O

24

34

37

OH OH

1

2

OH

28

dominio lactónico terminal

dominio espaciador dominio polioxigenado dominio hidrofóbico terminal

Figura 4. Estructura de squamocin

Nuestro grupo de investigación ha centrado su interés en el aislamiento, evaluación

de las propiedades biológicas y transformación química de las ACG aisladas en diversas

especies tropicales de la familia Annonáceas. Algunos de los datos recientes sobre REA de

las ACG, corresponden al efecto producido por la sustitución de la γ-lactona terminal en

squamocin (una potente ACG bis-THF, con una CI50 de 0.9 nM) [6], así como la

influencia de la acetilación de los hidroxilos presentes en el dominio polioxigenado de

guanaconetins (un grupo homogéneo de ACG bis-THF aisladas en semillas de Annona

spraguei) [7,8]. Los resultados obtenidos, apoyaron la idea de la importancia del dominio

lactónico terminal para la actividad inhibidora del complejo I, ya que su sustitución por

otros grupos, dando lugar a derivados γ-cetoamidas y γ-cetoácidos, disminuye

drásticamente esta actividad; si bien la γ-lactona puede ser sustituída por algún bioisóstero

sin que la actividad se vea notoriamente afectada. En cuanto al dominio polioxigenado, los

datos obtenidos en las ACG guanaconetins nos muestran que los grupos acetilo,

especialmente en las monoacetiladas, modulan la actividad tanto frente a líneas celulares

tumorales como en la inhibición de los complejos de la CRM, destacando la actividad en

6

ensayos in vitro de guanaconetin-3, con una CI50 de 2.4 nM y 17 µM para el complejo I de

la CRM y frente a la línea celular tumoral HepG2, respectivamente (Figura 5).

O

3015O O

O

24

34

37

35

OHOH OCOCH3

Figura 5. Estructura de guanaconetin-3

2. Estiril-lactonas

Por otra parte, en el curso de nuestra investigación sobre MSA citotóxicos presentes

en especies de la familia Annonáceas, aislamos compuesto con esqueleto estiril-lactonas,

descritos en la literatura por su actividad citotóxica y antitumoral frente a varias líneas

celulares tumorales humanas. Estos compuestos se caracterizan por poseer un esqueleto

básico de 13 átomos de carbono con una estructura formada por un grupo estiril o

pseudoestiril unido a un anillo γ- o δ-lactona. Biogenéticamente tienen un origen mixto:

derivan de la vía del ácido shikímico y del acetil-CoA, para dar lugar a una molécula de

ácido cinámico (C6-C3) y dos unidades de acetato/malonato (C4), respectivamente. Las

estiril-lactonas se distribuyen especialmente en especies del género Goniothalamus,

ampliamente utilizadas en medicina popular. Recientemente, se han aislado nuevas estiril-

lactonas citotóxicas, las crassalactonas A-D, a partir de un extracto de hojas y ramas de

Polyalthia crassa. En los ensayos in vitro, en líneas celulares tumorales humanas

realizados sobre estos compuestos, destacó la actividad antitumoral de crassalactona B,

que fue capaz de inhibir el crecimiento de la línea celular T de leucemia (Jurkat) con una

CI50 de 0.45 µM [9].

En nuestro grupo, aislamos en el extracto metanólico de las cortezas del tronco de

Goniothalamus arvensis, (+)-altholactona, como producto mayoritario, además de las

furanopironas (+)-isoaltholactona, (+)-O-acetilaltholactona y (-)-etharvensin, así como

otras estiril-lactonas con diferentes esqueletos (Figura 6) [10-13]. Altholactona, 5-

acetoxigoniothalamin óxido y O-acetilaltholactona, fueron evaluados a nivel de su

capacidad para inhibir la CRM en mitocondrias de corazón de ternera. Los resultados

mostraron que estas estiril-lactonas naturales eran potentes inhibidores de la cadena

transportadora de electrones [14].

7

O

O

OH

H

OH

6

7

5

3

2

(+)-altholactona

O

O

OOH

(+)-isoaltholactona (-)-etharvensin

O

O

OHO

H

H

EtO

Figura 6. Furanopironas de Goniothalamus arvensis

Debido al interés biológico de las estiril-lactonas y con la finalidad de establecer

criterios de REA, se sintetizaron una serie de nuevos análogos de altholactona (derivados

1-9), así como del producto natural goniothalesdiol (derivados 10-11) (Figura 7). Estudios

in vitro de la actividad citotóxica de las estiril-lactonas naturales y sintéticas, frente a

líneas celulares tumorales de leucemia L-1210, mostraron que altholactona y algunos de

sus derivados (compuestos 4-6), eran capaces de inhibir la proliferación celular, actuando

a nivel de la fase G2+M del ciclo celular (estudios realizados en colaboración con la

empresa farmacéutica Laboratoires Servier) [15]. Además, todos los derivados mostraron

capacidad para inhibir el complejo I de la CRM, destacando O-acetilaltholactona con una

CI50 de 4.7 µM [16].

O

O

OH

H

OAc

O

HOOH

H3COH

5 6

1

4

goniothalesdiol

O

H3COH2COOCH2OCH3

H3COH

10

O

O

OOH

RO

OR

1: R= H2: R= COCH33: R+R= CH2

R1

O

O

OH

H

OH

R

O

O

OOCH2OCH3

8

O

O

OOCH2OCH39

OCH3

O

H3COH2COOCH2OCH3

H3COH

11

OCH3

R2

R2

4: R1= NO2; R2= H5: R1= H; R2= NO2

6: R= Cl7: R= OCH3

O O

O

Figura7. Estiril-lactonas semisíntéticas (1-11)

8

3. Benzopiranos prenilados

Los benzopiranos prenilados, son otro grupo de MSA presentes en especies de la

familia Annonáceas, concretamente en las del género Polyalthia. Estos compuestos

bioactivos pueden encontrarse además en plantas superiores de otras familias y en

numerosos organismos marinos (corales, esponjas marinas, tunicados, etc).

Biogenéticamente, son metabolitos secundarios de origen mixto, ya que proceden del

ácido shikímico y del acetil-CoA. El género Polyalthia incluye más de 150 especies

siendo todas ellas típicas de zonas tropicales de África, Asia y Oceanía. Nuestro estudio se

llevó a cabo sobre el extracto metanólico de la corteza del tronco de dos especies: P.

cerasoides y P. sclerophylla. En dichas especies se aislaron nuevos benzopiranos

prenilados: polycerasoidin, polycerasoidol, (6E,10E)-isopolycerasoidol, polycerasoidin

metil éster y polyalthidin (Figura 8) [17,18]. Debido a la similitud estructural de los

benzopiranos prenilados con el ubicromanol (derivado del coencima Q, complejo de la

CRM) se realizaron sobre polycerasoidin, polycerasoidol y polyalthidin ensayos de

actividad inhibidora del complejo I de la CRM, pudiendo concluir que polyalthidin era el

inhibidor más potente de los tres benzopiranos ensayados, con una CI50 de 4.4 µM frente a

11 µM y 37 µM para policerasoidin y polycerasoidol, respectivamente [19]. Esta actividad

inhibidora de la cadena transportadora de electrones de los benzopiranos prenilados,

podría explicar las propiedades citotóxicas y antitumorales que presentan estas especies

vegetales.

O

CH3

RO

CH3 CH3

CH3

COOH

O

CH3

HO

CH3 CH3

COOH

CH3 O

CH3

H3CO

CH3 CH3

CH3

COOCH3

O

CH3

H3CO

CH3

CH3

COOH

R= OCH3; Polycerasoidin

R= H; Polycerasoidol

Polycerasoidin metilester(6E,10E)-Isopolycerasoidol

Polyalthidin

Figura 8. Benzopiranos prenilados de especies de Polyalthia

9

4. Isoquinoleínas

Los alcaloides isoquinoleínicos abarcan un amplio grupo de estructuras incluidas las

benciltetrahidroisoquinoleínas (BTHIQ), aporfinas (AP) y protoberberinas (PB), que

pueden encontrarse en especies de un gran número de familias botánicas dentro del orden

de las Magnoliales, entre las que se encuentra la familia de las Annonáceas.

Biogenéticamente, al igual que otros esqueletos isoquinoleínicos, las AP y PB se forman a

partir de los alcaloides BTHIQs que a su vez proceden de la tirosina, junto con los

correspondientes productos de descarboxilación. La dopamina (obtenida a partir de la

tirosina) se condensa con el p-hidroxifenilacetaldehido para dar lugar a la primera BTHIQ,

norcoclaurina. Estos metabolitos destacan por presentar afinidad por los receptores de la

dopamina, que son receptores de membrana acoplados a una proteína G. La aplicación de

métodos de clonación usando técnicas de biología molecular ha permitido clasificarlos en

dos subfamilias, D1-like (D1 y D5) y D2-like (D2, D3 y D4), con secuencias de aminoácidos

y propiedades farmacológicas comunes. Los receptores D1-like poseen alta afinidad por

ligandos benzacepínicos, SCH 23390 y SKF 83566 (Figura 9) y una moderada afinidad

por agonistas clásicos de dopamina (apomorfina) y por agonistas selectivos como SKF

38393.

NCH3

HO

HO

NCH3

Cl

HO

SCH23390SKF83566

Figura 9. Ligandos específicos de receptores D1

Los receptores D2-like se localizan tanto a nivel presináptico (autorreceptores,

regulan la liberación de dopamina) como postsináptico. Las butiferonas (haloperidol) y las

benzamidas sustituídas (sulpirida y racloprida, antagonistas D2) (Figura 10), poseen alta

afinidad por dichos receptores, al igual que las fenotiazinas y los tioxantenos. Dado que

los receptores D3 y D4 se expresan fundamentalmente en las regiones corticales y límbicas,

implicadas en el control del conocimiento y de las emociones, son el objetivo del estudio

de una nueva generación de fármacos que permitan paliar algunos trastornos neurológicos

y psiquiátricos con baja incidencia de efectos colaterales extrapiramidales.

10

Cl

Cl

OH

OCH3

NH

O

N

raclopridaF

N

OH

Cl

haloperidol

Figura 10. Antagonistas de receptores D2

Farmacológicamente, los antagonistas dopaminérgicos son agentes que pueden ser

utilizados en el tratamiento clínico de la esquizofrenia, manía, delirios y la enfermedad de

Huntington, mientras que los agonistas dopaminérgicos se utilizan en desórdenes

neuroendocrinos y en la enfermedad de Parkinson. Muchos de los inhibidores del

transportador de dopamina inactivan el proceso de recaptación, actuando como

antidepresivos puesto que aumentan la concentración de dopamina en la hendidura

sináptica y por tanto activan la neurotransmisión dopaminérgica. Actualmente, los

inhibidores del transportador de dopamina han sido prescritos para el tratamiento del

déficit de atención en los casos de hiperactividad (ADHDs) y propuestos como fármacos

potenciales en los tratamientos de adición a la cocaína. Entre ellos destacan compuestos

naturales y sintéticos, con esqueletos variados como tropánicos (tipo cocaína), núcleos

isoquinoleínicos (tipo nomifensina), estructuras tricíclicas (tipo amineptina) o estructuras

aril-piperacínicas (tipo GBR-12783).

Los estudios de afinidad por los receptores D1-like y D2-like se llevan a cabo

mediante ensayos in vitro utilizando técnicas de fijación con radioligandos específicos,

denominadas técnicas de binding, en cuerpo estriado de rata. Son experimentos de

competición donde se evalúa la capacidad de los compuestos ensayados para desplazar los

radioligandos 3H-SCH 23390 (ligando selectivo de receptores D1-like) y 3H-racloprida

(ligando selectivo de receptores D2-like) de sus lugares de unión al receptor. En el caso de

la recaptación de dopamina, los estudios in vitro de inhibición se realizan en los

sinaptosomas de cuerpo estriado de rata, utilizando 3H-dopamina como sustrato de los

transportadores dopaminérgicos.

Estudios previos realizados por nuestro grupo de investigación, determinaron la

importancia de algunos grupos funcionales en el esqueleto de las isoquinoleínas para

mejorar su afinidad por los receptores dopaminérgicos.

11

4.1. Bencilisoquinoleínas

Se han llevado a cabo numerosos estudios de REA sobre BTHIQ, lo que ha

permitido conocer algunas características estructurales que modulan la afinidad de dichas

moléculas por los receptores dopaminérgicos: a) la presencia de un átomo de cloro en

posición 7 y de un hidroxilo en posición 6, b) la configuración (S) del centro

estereogénico, y c) la naturaleza del sustituyente en posición 1; habiendo podido

determinarse, en las series 1-butil-, 1-fenil- y 1-bencil-THIQ, que la afinidad por los

receptores D2 es mayor cuando en posición 1 existe un sustituyente butil o bencil, mientras

que la afinidad por D1 aumenta cuando en dicha posición el sustituyente es un fenil,

probablemente debido a la flexibilidad del grupo butil o bencil para adoptar una

configuración espacial determinada que favorezca la unión a D2.

Además, una de las moléculas sintetizadas de tipo 1-butil-BTHIQ y con OH en C-6 y

Cl en C-7, mostró una elevada afinidad y selectividad por los receptores D2 (CI50 66 nM),

así como una actividad antidepresiva, deducida mediante ensayos de comportamiento

realizados en experimentos in vivo con ratones (Figura 11) [20-22].

1-butil-THIQ

NRHO

1

6

NRHO

NRHO

1-bencil-THIQ1-fenil-THIQ

Cl Cl Cl

Figura 11. Isoquinoleínas 1-sustituidas dopaminérgicas

Por otra parte y dentro de este tipo de estructuras, sintetizamos una serie de 1-

benzoil-3,4-dihidroisoqinoleínas (1-benzoil-DHIQ), mediante oxidación controlada del

carbono bencílico de la imina precursora (resultados patentados con Laboratoires Servier)

[23,24]. Estos derivados α-cetoimina (con nitrógeno no alcalino y por tanto sin capacidad

para fijarse a los receptores dopaminérgicos), fueron capaces de inhibir el crecimiento de

líneas celulares tumorales de leucemia (L1210) e interferir en la fase G1 del ciclo celular

(Figura 12).

12

N

O

N

O

OCH36-butiloxi-1-benzoil-DHIQ6-benciloxi-1-benzoil-DHIQ

O O6

1

Figura 12. Benzoildihidroisoquinoleínas citotóxicas

4.2. Aporfinas

La aporfinas (AP) han mostrado interesantes propiedades farmacológicas:

antitumoral, anticonvulsivante, antipalúdica y antibacteriana, entre otras. Una de las más

utilizadas en terapéutica es la (R)-apomorfina, se trata de una AP semisintética, agonista

de los receptores dopaminérgicos y utilizada clínicamente en la enfermedad del Parkinson.

Algunas de las características químicas fundamentales de las AP, necesarias para aumentar

la afinidad por estos receptores son: a) la presencia de un grupo OH en posición 11, b) la

configuración (R) en posición 6a, c) el tamaño del sustituyente sobre el átomo de

nitrógeno, y d) la presencia de halógenos en posición 3 (Figura 13) [25]. A nivel de la

recaptación de dopamina, se considera importante la presencia de un grupo metilendioxi

sobre el motivo β-feniletilamina de las AP, como en el caso de anonaina. Esta AP se ha

encontrado frecuentemente en especies de Annonáceas, habiendo sido aislada por nosotros

en Xylopia papuana [26].

NH

O

OH

annonaína

NCH3

H

apomorfina

HO

HO

Figura 13. Aporfinas dopaminérgicas

13

4.3. Protoberberinas

Dentro de los alcaloides protoberberínicos (PB), la berberina ha sido uno de los

compuestos más estudiados por sus propiedades bactericida, fungicida y antitumoral. A

partir de especies del género Guatteria: G. ouregou y G. schomburgkiana, aislamos PB

con alta afinidad por los receptores dopaminérgicos D2: (R)-corypalmine, (R)-

tetrahydropalmatine, (R)-10-demetildiscretine y principalmente (R)-coreximine [27]. Las

características más importantes en su estructura son: a) la posición de los grupos metoxi e

hidroxi (meta y para respectivamente, al motivo β-feniletilamina), y b) la configuración

(R) en posición 13a. Recientemente, otros grupos se han interesado en la obtención de este

tipo de isoquinoleínas, con objeto de avanzar en la terapia del Parkinson y del Alzheimer.

Así, las PB monooxigenadas en posición meta al motivo β-feniletilamina, parecen mostrar

elevada afinidad por los receptores dopaminérgicos (Figura 14) [28].

N

HO

N

H3CO

N

H3CO

HOH

OCH3

OH

(R)-coreximine

2

3

13a1

Figura 14. Protoberberinas naturales y sintéticas

5. Azafluorenonas

Las azafluorenonas, son un grupo de alcaloides que se han asilado únicamente en

especies de la familia Annonáceas. Macondina, darienina y ursulina, fueron algunas de las

azafluorenonas que encontramos en las cortezas del tronco de Oxandra major [29].

Recientemente se han aislado la 5,8-dihidroxi-6-metoxi-oniquina y la 5-hidroxi-6-metoxi-

oniquina (esta última isómero de macondina), a partir de Mitrephora diversifolia (Figura

15) [30]. Estas moléculas presentaron propiedades antimaláricas, sin ser tóxicas frente a

células embrionarias de riñón, con un actividad frente a P. falciparum de rango

micromolar [30]. Además, algunos de estos compuestos mostraron actividad

antibacteriana, siendo capaces de dañar la estructura del DNA [31].

14

NO

H3CO

HO

7

8

1

4

macondina

5

NO

OH

OCH3

6

5-OH-6-MeO-onychine

Figura 15. Azafluorenonas naturales

6. ββββ-Carbolinas-pirimidinas: “annomontina”

Las Annonáceas por tanto pueden considerarse una familia de gran diversidad y

riqueza en MSA, muchos de los cuales presentan propiedades farmacológicas de gran

interés como ha podido observarse. Finalmente, cabe destacar, como fruto de las

investigaciones realizadas entre las Universidades de Chiapas en México y de Valencia en

España, el aislamiento (a partir de las raíces de Annona purpurea), determinación

estructural y actividad ansiolítica de “annomontina”, una molécula con esqueleto β-

carbolina-pirimidina, muy poco frecuente en Annonáceas, sin duda con interesantes

perspectivas de futuro desarrollo farmacológico (Figura 16) [32].

HB

HA

HD

HC NH

N

N

NH2N

HB

HA

HA

HB

sistema ABorto piridínico

sistema ABorto pirimidínico

anillo aromáticoorto-disustituido: sistema ABCD

H

H

H

H NH

N

N

NH2N

H

H

H

H

4

3

1a

4a

4'

2' 6'

8a

5

8

6

7

5a

Correlación heteronuclear 1H-13C - ( 3J)

( ) HMBC de annomontina

Correlación homonuclear 1H-1H

( ) COSY de annomontina

Figura 16. Determinación estructural por RMN de annomontina

15


[1] Cortes, D. Farmacoquímica Natural. Editorial Moliner, Valencia 2007.

[2] Silverman, R. B. The organic chemistry of drug design and drug action, J. Hayhurst Editor,

Elsevier Academic Press, Londres 2004.

[3] Ciavatta, M. L., López-Gresa, M. P., Gavagnin, M.; Romero, V., Melck, D., Manzo, E.,

Guo, Y. W.; Van Soest, R. y Cimino, G. Studies on Puupehenone-metabolites of a Dysidea

sp.: structure and biological activity. Tetrahedron 2007, 67, 1380-1384.

[4] Londerhausen, M., Leicht, W., Lieb, F., Moeschler, H. y Weiss, H. Molecular mode of

action of annonins. Pestic. Sci. 1991, 33, 427-438.

[5] Bermejo, A., Figadère, B., Zafra-Polo, M. C., Barrachina, I., Estornell, E. y Cortes, D.

Acetogenins from Annonaceae: recent progress in isolation, synthesis and mechanisms of

action. Nat. Prod. Rep., 2005, 22, 269-303.

[6] Duval, R. A., Lewin, G., Peris, E., Chahboune, N., Garofano, A., Dröse, S., Cortes, D.,

Brandt, U. y Hocquemiller, R. Heterocyclic analogues of squamocin as inhibitors of

mitochondrial complex I. On the role of the terminal lactone of annonaceous acetogenins

Biochemistry 2006, 45, 2721-2728.

[7] Barrachina, I., Royo, I., Baldón. H. A., Chahboune, N., Suvire, F., DePedro, N., Zafra-Polo,

M. C., Bermejo, A., El Aouad, N., Cabedo, N., Saez, J., Tormo, J. R., Enriz, R. D. y Cortes,

D. New antitumoral acetogenin ‘Guanacone type’ derivatives: Isolation and bioactivity.

Molecular dynamics simulation. Bioorg. Med. Chem. 2007, 15, 4369-4381.

[8] Chahboune, N., Barrachina, I., Royo, I., Romero, V., Saez, J., Tormo, J. R., DePedro, N.,

Estornell., E., Zafra-Polo, M. C., Peláez, D. y Cortes, D. Guanaconetins, New Antitumoral

Acetogenins, Mitochondrial Complex I and Tumor Cell Growth Inhibitors diacetyl-

guanacone. Bioorg. Med. Chem. 2006, 14, 1089-1094.

[9] Popsavin, V., Benedeković, G., Popsavin, M., Kojić, V. y Bogdanović, G. Divergent

synthesis of cytotoxic styryl lactones isolated from Polyalthia crassa. The first total

synthesis of crassalactone B. Tetrahedron Letters 2010, 51, 3426-3429.

[10] Bermejo, A., Lora, M. J., Blazquez, M. A., Rao, K. S., Cortes, D. y Zafra-Polo, M. C. (+)-

Goniotharvensin, a novel styryl-lactone from the stem bark of Goniothalamus arvensis. Nat.

Prod. Lett. 1995, 7, 117-122.

[11] Bermejo, A., Blázquez, M. A., Serrano, A., Zafra-Polo, M. C. y Cortes, D. Preparation of 7-

Alkoxylated Furanopyrones: Semisynthesis of (−)-Etharvensin, a New Styryl-Lactone from

Goniothalamus arvensis. J. Nat. Prod. 1997, 60, 1338-1340.

[12] Bermejo, A., Blázquez, M. A., Rao, K. S. y Cortes, D. Styryl-pyrones from Goniothalamus

arvensis. Phytochemistry 1998, 47, 1375-1380.

[13] Bermejo, A., Blázquez, M. A., Rao, K. S. y Cortes, D. Styryl-lactones from Goniothalamus

species-a review. Phytochem. Anal. 1999, 10, 127-131.

16

[14] Peris, E., Estornell, E., Cabedo, N., Cortes, D. y Bermejo, A. 3-Acetylaltholactone and

related styryl-lactones, mitochondrial respiratory chain inhibitors. Phytochemistry 2000, 54,

311-315.

[15] Bermejo, A., Léonce, S., Cabedo, N., Andreu, I., Caignard, D. H., Atassi, G. y Cortes, D.

Semisynthesis and cytotoxicity of styryl-lactone derivatives. J. Nat. Prod. 1999, 62, 1106-

1109.

[16] Peris, E., Cavé, A., Estornell, E., Zafra-Polo, M. C., Figadère, B., Cortes, D. y Bermejo, A.

Semisynthesis of new tetrahydrofuranic alkyl ester and furano-pyrone derivatives as

inhibitors of the mitochondrial complex I. Tetrahedron 2002, 58, 1335-1342.

[17] González, M. C., Serrano, A., Zafra-Polo, M. C., Cortes, D. y Rao, K. S. Polycerasoidin and

polycerasoidol, two new prenylated benzopyran derivatives from Polyalthia cerasoides J.

Nat. Prod. 1995, 58, 1278-1284.

[18] González, M. C., Sentandreu, M. A., Rao, K. S., Zafra-Polo, M. C. y Cortes, D. Prenylated

benzopyran derivatives from two Polyalthia species. Phytochemistry 1996, 43, 1361-1364.

[19] Zafra-Polo, M. C., González, M. C., Tormo, J. R., Estornell, E. y Cortes, D. Polyalthidin:

New Prenylated Benzopyran Inhibitor of the Mammalian Mitochondrial Respiratory Chain.

J. Nat. Prod. 1996, 59, 913-916.

[20] Cabedo, N., Andreu, I., Ramírez De Arellano, M. C., Chagraoui, A., Serrano, A. Bermejo,

A., Protais, P. y Cortes, D. Enantioselective Syntheses of dopaminergic (R)- and (S)-

benzyltetrahydroisoquinolines. J. Med. Chem. 2001, 61, 1794-1801.

[21] Berenguer, I., El Aouad, N., Andujar, S., Romero, V., Suvire, F., Freret, T., Bermejo, A.,

Ivorra, M. D., Enriz, R. D., Boulouard, M., Cabedo N. y Cortes, D. Tetrahydroisoquinolines

as dopaminergic ligands: 1-Butyl-7-chloro-6-hydroxy-tetrahydroisoquinoline, a new

compound with antidepressant-like activity in mice Bioorg. Med. Chem. 2009, 17, 4968-

4980.

[22] El Aouad, N, Berenguer, I., Romero, V., Marín, P., Serrano, A., Andujar, S., Suvire, F.,

Bermejo, A., Ivorra, M. D., Enriz, R. D., Cabedo N. y Cortes, D. Structure–activity

relationship of dopaminergic halogenated 1-benzyl-tetrahydroisoquinoline derivatives. Eur.

J. Med. Chem. 2009, 44, 4616-4621.

[23] Andreu, I., Cabedo, N., Atassi, G., Pierre, A., Caignard, D. H., Renard, P., Cortes, D. y

Bermejo, A. An efficient method for the preparation of antitumoral α-keto-imines

benzyldihydroisoquinolines by selective benzylic oxidation with C/Pd in acetonitrile.

Tetrahedron Letters 2002, 43, 757-759.

[24] Bermejo, A., Andreu, I., Suvire, F., Léonce, S., Caignard, D. H., Renard, P., Pierré, A.,

Enriz, R. D., Cortes, D. y Cabedo, N. Syntheses and Antitumor Targeting G1 Phase of the

Cell Cycle of Benzoyldihydroisoquinoline and Related 1-Substituted Isoquinolines. J. Med.

Chem. 2002, 45, 5058-5068.

17

[25] Cabedo, N., Berenguer, I., Figadère, B. y Cortes, D. An overview on benzylisoquinoline

derivatives with dopaminergic and serotonergic activities. Curr. Med. Chem. 2009, 16,

2441-2467.

[26] Protais, P., Arbaoui, J., Bakkali, E. H., Bermejo, A. y Cortes, D. Effects of Various

Isoquinoline Alkaloids on In Vitro 3H-Dopamine Uptake by Rat Striatal Synaptosomes. J.

Nat. Prod. 1995, 58, 1475-1484.

[27] Cortes, D., Arbaoui, J. y Protais, P. High affinity and selectivity of some

tetrahydroprotoberberine alkaloids for rat striatal 3H-raclopride binding sites. Nat. Prod.

Letters 1993, 3, 233-238.

[28] Schulze, M., Müller, F. K. U., Mason, J. M., Görls, H., Lehmann, J. y Enzensperger,

Ch. Dibenzazecine scaffold rebuilding-Is the flexibility always essential for high dopamine

receptor affinities? Bioorg. Med. Chem. 2009, 17, 6898-6907.

[29] Arango, G., Cortes, D., Cassels, B. K., Cavé, A. y Mérienne, C. Azafluorenones from

Oxandra cf. major and biogenetic considerations. Phytochemistry 1987, 26, 2093-2098.

[30] Mueller, D., Davis, R. A., Duffy, S., Avery, V. M., Camp, D. y Quinn, R. J. Antimalarial

Activity of Azafluorenone Alkaloids from the Australian Tree Mitrephora diversifolia. J.

Nat. Prod. 2009, 72, 1538-1540.

[31] Kraus, G. A. y Kempema, A. Synthesis of Azafluorenone Antimicrobial Agents. J. Nat.

Prod. 2010, 73, 1967–1968.

[32] Rejón-Orantes, J. C., González-Esquinca, A. R., Pérez de la Mora. M., Roldan, R. G. y

Cortes D. Annomontine, an alkaloid isolated from Annona purpurea has anxiolytic-like

effects in the elevated plus-maze. Planta Medica 2010 in press.

.

Capítulo II:

Pirrolo[2,1-a]isoquinoleínas

antimicrobianas

Artículo 2: “Synthesis of new antimicrobial pyrrolo

[2,1-a]isoquinolin-3-ones” (En: Bioorganic & Medicinal

Chemistry, 2012, 20, 6589)

Bioorganic & Medicinal Chemistry 20 (2012) 6589–6597

Contents lists available at SciVerse ScienceDirect

Bioorganic & Medicinal Chemistry

journal homepage: www.elsevier .com/locate /bmc

Synthesis of new antimicrobial pyrrolo[2,1-a]isoquinolin-3-ones

Laura Moreno a, Javier Párraga a, Abraham Galán a, Nuria Cabedo b, Jaime Primo b, Diego Cortes a,⇑a Departamento de Farmacología, Facultad de Farmacia, Universidad de Valencia, 46100 Burjassot, Valencia, Spainb Centro de Ecología Química Agrícola-Instituto Agroforestal Mediterraneo, UPV, Campus de Vera, Edificio 6C, 46022 Valencia, Spain

a r t i c l e i n f o

Article history:Received 21 April 2012Revised 13 September 2012Accepted 15 September 2012Available online 24 September 2012

Keywords:Pyrrolo[2,1-a]isoquinolin-3-onesBischler-Napieralski cyclodehydrationBactericideFungicide

0968-0896/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.bmc.2012.09.033

⇑ Corresponding author.E-mail address: [email protected] (D. Cortes).

a b s t r a c t

The attractive structure of the pyrroloisoquinoline moiety, together with its potential antimicrobial activ-ity, encouraged us to prepare six 8-substituted and seven 8,9-disubstituted-1,2,3,5,6,10b-hexahydropyr-rolo[2,1-a]isoquinolin-3-ones in a few steps with good yields. We applied a convenient methodology viadouble intramolecular cyclization conducted by a Bischler-Napieralski cyclodehydration-imine reductionsequence, which is widely employed in the synthesis of isoquinoline alkaloids. Therefore, we synthesizedthree series of these pyrrolo[2,1-a]isoquinolin-3-ones characterized by the substituent at the 8-positionor 8,9-positions of the aromatic ring: (a) different side chains are attached to an 8-OH group (series 1); (b)a chlorine atom is attached to the 8-position (series 2); and (c) 8- and 9-carbons are bearing an identicalgroup (series 3). The compounds bearing a benzylic moiety at the 8-position, for example, 8-benzyloxy-pyrrolo[2,1-a]isoquinolin-3-one (1a) and 8-(4-fluorobenzyloxy)-pyrrolo[2,1-a]isoquinolin-3-one (1e), aswell as, a 8-chloro-9-methoxy moiety including the 8-chloro-9-methoxy-pyrrolo[2,1-a]isoquinolin-3-one (2a), provided the most fungicide and bactericide agents, respectively.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Isoquinoline alkaloids are a large family of natural productswith a variety of powerful biological activities,1 including inhibi-tion of cellular proliferation.2 Within the isoquinoline family, pyr-roloisoquinoline alkaloids have been paid considerable attention inrecent years because they display interesting biological activitiessuch as antidepressant,3 muscarinic agonist,4 antiplatelet5 andantitumor activity.6 In 1968, the first natural pyrroloisoquinolinealkaloids were isolated from the peyote cactus and were identifiedas peyoglutam and mescalotam.7 Since then, the pyrrolo[2,1-a]iso-quinoline ring system is the main core of a wide variety of biolog-ically active alkaloids, including antitumor crispine A, isolatedfrom Carduus crispus,8 tetracyclic compounds such as antiglycemicjamtine, isolated from the climbing shrub Cocculus hirsutus,9 anderythrina type alkaloids with curare-like neuromuscular blockingactivities.10,11 In 1984, a tetracyclic framework bearing a pyrrolo-isoquinoline lactam was found in nuevamine12 which displayedanti-inflammatory, antimicrobial, anti-leukemic and antitumorproperties.13 In 2004, tricyclic lactam trolline was isolated fromTrollius chinensis. This alkaloid exhibited in vitro significant anti-bacterial activity against some strains of Klebsiella pneumoniae,Pseudomonas aeruginosa, Haemophilus influenza, Staphylococcusaureus, Streptococcus pneumonia, S. pyogenes, and moderate antivi-ral activity against influenza viruses A and B.14

ll rights reserved.

Actually, pyrrolo[2,1-a]isoquinolines were synthesized long be-fore they were isolated as natural products, and were incorporatedinto larger ring systems; for example, in the lamellarins skeleton.15

Nevertheless, given their attractive biological activities, the syn-thesis of new compounds bearing this structural framework hasgreatly increased in recent years.16 One representative syntheticstrategy involves the annulation of the pyrrole ring by intramolec-ular reaction of a 1,3-difunctionalized three-carbon building blockwith a 3,4-dihydroisoquinoline.17–19 Other strategies proceed prin-cipally via phenethyl succinimides, in which a N-acyliminium ionundergoes different types of intramolecular aromatic p-cyclizationreactions to provide the C10a–C10b bond formation. In this case,some broadly used approaches involve: (1) tandem carbophilicaddition of the organolithium reagent-N-acyliminium ion cycliza-tion sequence to obtain the isoquinolone skeleton;20 (2) applica-tion of Parham-type cyclization to halogenated imides, giving thecorresponding enamides;21,22 and (3) the Pummerer/Mannich in-duced cyclization cascade, which involves a thionium-N-acylimin-ium ion cyclization to give the azabicyclic ring system.23

Our research group is focusing since a long time on isolatingand synthesizing isoquinoline-containing compounds to obtain avariety of chemically diverse structures with dopaminergic activ-ity.24–28 However, the discovery of new antimicrobial agents inthe last few years has become a need since some microorganismsdevelop resistance to classic drugs due to their extensive use. Asmentioned above, previous works have shown the antimicrobialproperties of pyrroloisoquinolines, including the nuevamine-typeand trolline-type alkaloids.13,14 The attractive structure of the

http://dx.doi.org/10.1016/j.bmc.2012.09.033

mailto:[email protected]

http://dx.doi.org/10.1016/j.bmc.2012.09.033

http://www.sciencedirect.com/science/journal/09680896

http://www.elsevier.com/locate/bmc

6590 L. Moreno et al. / Bioorg. Med. Chem. 20 (2012) 6589–6597

pyrroloisoquinoline moiety, together with its potential antimicro-bial activity, encouraged us to synthesize trolline analogs contain-ing a lactam-ring pharmacophore similar to antibiotic b-lactams.Although several methods have been portrayed for the synthesisof this framework, we applied a typical methodology that wewidely employed in the course of our research into isoquinolinesynthesis. It is based on double intramolecular cyclization, con-ducted by Bischler-Napieralski cyclodehydration from an esterb-phenylethylamide and involves the subsequent reduction ofthe imine intermediate. Therefore, we prepared six 8-substitutedand seven 8,9-disubstituted-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-ones, including the known alkaloid (±)-trolline, ina few steps with good yields for the purpose of exploring their anti-microbial activities. Specifically, we followed the same approach tosynthesize three series of these pyrrolo[2,1-a]isoquinolinonescharacterized by the substituent at the 8-position of the aromaticring. In these series: (a) different side chains are attached to an8-OH group (series 1); (b) a chlorine atom is attached to the 8-po-sition (series 2); and (c) 8- and 9-carbons are bearing an identicalgroup (series 3). The biological results of these thirteen compoundshave enabled us to draw conclusions on these groups’ influence onantimicrobial activity.

2. Results and discussion

2.1. Chemistry

The general synthetic plan for these compounds focused onpreparing the appropriate b-phenyl acetamides (1–3) by standardmethods25,26 to be cyclized by Bischler-Napieralski cyclodehydra-tion. Firstly, these were prepared by starting with benzaldehydederivatives: 3-hydroxybenzaldehyde (series 1, Scheme 1) and3-chloro-4-methoxy-benzaldehyde (series 2, Scheme 2); and fromthe 3,4-dimethoxy-b-phenyl-ethylamine (series 3, Scheme 3).Therefore, by commencing with these benzaldehydes, and by asuccessive nitromethane-reduction sequence, b-phenylethylamineintermediates were obtained and then condensed with the ethylsuccinyl chloride under Shotten-Bauman conditions to give theb-(3-benzyloxy-phenyl)acetamide (1), b-(3-chloro-4-methoxy-phenyl)acetamide (2) and b-(3,4-dimethoxy-phenyl)-acetamide(3), as outlined in Schemes 1–3, respectively. These amides 1–3were treated with a POCl3 reagent to lead us to the expecteddihydroisoquinolines by Bischler-Napieralski cyclization. The ob-tained imine was reduced with NaBH4 to give the tetrahydroiso-quinoline skeleton and, simultaneously, the formed nucleophilic

R=HR=Bn

(a)3-OH-benzaldehyde3-OBn-benzaldehyde

H

ORO BnO

(b)

RO

N

1a: R= Bn1b: R= H

O

(g)

1 2

3

567

8

9

10

(98 %)

(72 %)(43 %)

3-benzyβ-nitrost

A B

C10b

BnO

NH

OO

Scheme 1. Synthesis of pyrroloisoquinolin-3-ones 1a,1b (series 1). Reagents and conditireflux, overnight; (c) LiAlH4, THF/Et2O, N2, reflux, 2 h; (d) Ethyl succinyl chloride, NaOH 5%concd HCl–EtOH 1:1, reflux, 3 h.

amine quickly attacked the carbonyl ester leading to C ringclosure by a second intramolecular cyclization to give 8-benzyl-oxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1a),8-chloro-9-methoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]iso-quinolin-3-one (2a) and 8,9-dimethoxy-1,2,3,5,6,10b-hexahydro-pyrrolo[2,1-a]isoquinolin-3-one (3a) for series 1–3, respectively.In order to obtain different substitutions at the 8 and 9-positionsto explore their influence on antimicrobial activity, the benzylicand the methoxy groups of 1a (series 1) and 2a (series 2) weredeprotected under acid medium to give 8-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1b), and by BBr3

to obtain 8-chloro-9-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (2b) (Schemes 1 and 2). In addition, thosesubstituents that seemed to enhance the antimicrobial activitywere placed in both 8 and 9-positions on the aromatic A-ring(series 3), also by previous deprotection of 3a by BBr3 to obtain(±)-trolline (3b) (Scheme 3).

Secondly, in series 1, one carbamate and three O-alkylatedderivatives were prepared from 8-hydroxy-pyrroloisoquinolone1b (Scheme 4) to give: 8-ethylcarbamate- (1c), 8-(1-piperidineth-oxy)- (1d), 8-(4-fluorobenzyloxy)- (1e) and 8-phenylacetamide-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-ones (1f). Inseries 2, one carbamate derivative was prepared from 8-chloro-9-hydroxy-pyrroloisoquinolone 2b to obtain 8-chloro-9-ethylcarba-mate-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (2c)(Scheme 2). In series 3, one 8,9-bis(4-fluorobenzyloxy)- (3c) andone 8,9-bis(phenylacetamide)-1,2,3,5,6,10b-hexahydropyrrol-o[2,1-a]isoquinolin-3-ones (3d), were prepared from (±)-trolline(3b) (Scheme 3). All the target compounds were synthesized usinga short approach with good yields. In addition, and to our knowl-edge, compounds 1a–1f, 2a–2c, 3c and 3d are reported for the firsttime. Their structures were confirmed by NMR spectral data andMS spectrometry.

2.2. Antimicrobial activity

All the synthesized pyrrolo[2,1-a]isoquinolines were assayedin vitro for their ability to inhibit bacterial and fungal growth. Inthe antimicrobial assays, our compounds were tested against sev-eral human pathogenic and economically important phytopatho-genic bacteria and/or fungi. The bacterial agents were distributedover Gram(+) and Gram(�) bacteria: B. cereus, S. aureus, andE. faecalis as Gram(+) and S. typhii, E. coli and E. carotovora asGram(�) (Table 1). The inhibition zones exhibited by compounds1a–1f, 2a–2c and 3a–3c are summarized in Tables 1 and 2 for

NO2 BnO

NH2

(c)

1

BnO

NHO

EtO

3-benzyloxy-phenyl-ethylamine

O

(d)

BnO

N

OEtO

(87 %)

(77 %)

(97 %)loxy-yrene

Et

(e)(f)

ons: (a) Benzyl chloride, K2CO3, EtOH, reflux, 6 h; (b) Nitromethane, NH4OAc, AcOH,, CH2Cl2, rt, overnight; (e) POCl3, CH2Cl2, N2, reflux, 6 h; (f) NaBH4; MeOH, rt, 2 h; (g)

(b)

(54 %)3

MeO

NHO

EtO O

MeOMeO

N OMeO

(c)

HO

N OHO

(e)

O

N OO

O

N OO

3a

3b

NH

O

O

HN

3d

(d)

F

F

(93 %)

(63 %)

3c(80 %)

MeO

NH2

3,4-dimethoxy-phenyl-ethylamine (88 %)

MeO(a)

Scheme 3. Synthesis of pyrroloisoquinolin-3-ones 3a–3d (series 3). Reagents and conditions: (a) Ethyl succinyl chloride, NaOH 5%, CH2Cl2, rt, overnight; (b) POCl3, CH2Cl2, N2,reflux, 3 h; and NaBH4; MeOH, rt, 2 h; (c) BBr3, CH2Cl2, rt, 2 h; (d) p-fluorobenzyl chloride, K2CO3, EtOH, reflux, overnight; (e) 2-bromo-N-phenylacetamide, K2CO3, EtOH,reflux, 6 h.

H

OCl Cl NO2 Cl

NH2(a) (b)

2

Cl

NHO

EtO O

(c)

Cl

N

2a

O

(90 %)

(53 %)(36 %)

(88 %)

H3CO H3CO3-chloro-4-methoxy-benzaldehyde

3-chloro-4-methoxy-β-nitrostyrene

β-(3-chloro-4-methoxyphenyl)ethylamine

H3CO

H3CO

H3CO

Cl

N

OEtO

Cl

NH

OEtO

(d)

H3CO H3CO(e)

Cl

N

2b

O

(91 %)

HO

Cl

N

2c

O

(85 %)

O(f) (g)

ONH

Scheme 2. Synthesis of pyrroloisoquinolin-3-one 2a–2c (series 2). Reagents and conditions: (a) Nitromethane, NH4OAc, AcOH, reflux, 6 h; (b) LiAlH4, THF/Et2O, N2, reflux, 2 h;(c) Ethyl succinyl chloride, NaOH 5%, CH2Cl2, rt, overnight; (d) POCl3, CH3CN, N2, reflux, 4 h; (e) NaBH4; MeOH, rt, 2 h; (f) BBr3, CH2Cl2, rt, 2 h; (g) Ethyl isocyanate, acetone,reflux, 3 h.

L. Moreno et al. / Bioorg. Med. Chem. 20 (2012) 6589–6597 6591

bactericidal and antifungal activity, respectively. In general, alipophilic group at the 8- and/or 9-position seemed to providemoderate activity if compared with a free hydroxyl group such

as 1b and (±)-trolline (3b). The introduction of the O-benzyl group(1a) was adequate for activity, and a halogen atom (1e) over thebenzyl moiety or even two p-fluorobenzyloxy groups (3c), did

HO

N O

1b

(a)

O

N O

1c

HN

O

O

N O

1d

N

O

(b)

O

N O

1e

O

N O

1f

NH

OF

(70 %) (82 %)

(80 %) (80 %)

(c) (d)

Scheme 4. Synthesis of pyrroloisoquinolin-3-ones 1c–1f. Reagents and conditions: (a) Ethyl isocyanate, acetone, reflux, 3 h; (b) 2-bromo-1-(piperidin-1-yl)ethanone, K2CO3,EtOH, reflux, 6 h; (c) p-fluorobenzyl chloride, K2CO3, EtOH, reflux, overnight; (d) 2-bromo-N-phenylacetamide, K2CO3, EtOH, reflux, 6 h.

Table 1

Strains Bactericidal activityInhibition zone (mm) 24 h (means ± SE)a

1ab 1bb 2ab 1cb 1eb 1fb 3bb 3cb Tetracyclineb

B. cereus 8.66 ± 0.20A 7.50 ± 0.35B 7.33 ± 0.41BC 6.50 ± 0.35C 7.50 ± 0.61B 8.83 ± 0.20A 6.66 ± 0.41BC 7.33 ± 0.41BC 24.33 ± 0.41D

S. aureus 0 ± 0 0 ± 0 7.83 ± 0.20A 0 ± 0 6.33 ± 0.20B 0 ± 0 0 ± 0 0 ± 0 27.0 ± 0.71C

E. faecalis 6.83 ± 0.20AB 0 ± 0 6.83 ± 0.20AB 8.50 ± 0.35C 0 ± 0 7.33 ± 0.41B 0 ± 0 6.33 ± 0.20A 26.0 ± 0.71D

S. typhii 6.0 ± 0A 0 ± 0 7.33 ± 0.73B 7.50 ± 0.35B 0 ± 0 7.33 ± 0.20B 0 ± 0 5.83 ± 0.20A 24.33 ± 0.41C

E. coli 405 6.33 ± 0.20A 0 ± 0 6.83 ± 0.20AB 6.83 ± 0.54AB 6.16 ± 0.20A 7.50 ± 0.35B 0 ± 0 6.83 ± 0.20AB 25.66 ± 0.41C

E. carotovora 7.26 ± 0.18AB 0 ± 0 10.33 ± 0.41C 8.83 ± 0.20D 8.0 ± 0.61B 8.0 ± 0.35BD 0 ± 0 7 ± 0A 13.33 ± 0.41E

E. coli 100 0 ± 0 0 ± 0 6.16 ± 0.20A 6.33 ± 0.41A 0 ± 0 0 ± 0 0 ± 0 0 ± 0 25.66 ± 0.41B

a Each value represents the average and the standard error of three independent experiments. Within each line, the mean values labeled with the same superscript (A–E) donot present statistically significant differences (P >0.05).

b Dose: 0.2 mg/disk. Compounds 1d, 2b, 2c and 3d did not show bactericidal activity. Compound 3a showed bactericidal activity at 0.3 mg/disk for S. aureus, E. faecalis,E. coli 405 and E. carotovora.

Table 2

Strains Antifungal activityInhibition zone (mm) 72 h (means ± SE)a

1ab 2ab 1cb 1eb Benomyl

A. parasiticus 6.83 ± 0.20A 7.50 ± 0.35B 0 ± 0 7.83 ± 0.20B 29.33 ± 0.41C,c1

T. viride 6.0 ± 0A 0 ± 0 0 ± 0 8.10 ± 0.25B 29.33 ± 0.81C,c1

F. culmorum 12.0 ± 0.71A 7.50 ± 0.35B 0 ± 0 10.50 ± 0.35C 27.66 ± 1.08D,c2

G. candidum 6.16 ± 0.20A 6.33 ± 0.41A 0 ± 0 7.0 ± 0.35A 0 ± 0P. citrophthora 11.33 ± 0.41A 6.83 ± 0.20B 9.66 ± 0.41C 10.66 ± 0.41A 44.33 ± 0.41D,c3

a Each value represents the average and the standard error of three independent experiments. Within each line, the mean values labeled with the same superscript (A–D)do not present statistically significant differences (P >0.05).

b Dose: 0.2 mg/disk.c1 Dose: 20 lg/disk.c2 Dose: 0.2 mg/disk.c3 Dose: 30 lg/disk. Compounds 1b, 1d, 1f did not show antifungal activity.


not significantly improve it. Moreover, a carbamate bond at the 8-position (1c) also provided activity against most of the testedstrains but not at 9-position (2c) on the aromatic A-ring. A similareffect was observed for 1f and 3d with one and two aromaticamide side chains, respectively. However, saturated amide 1d at8-position did not display any bactericidal activity. The most note-worthy compound was 2a, which possessed both a chlorine atomwhich draws electron density away from the p system and an elec-tron donating methoxyl group at the 8- and 9-position, respec-tively. This compound displayed bactericidal activity against allthe tested strains. However, the presence of two methoxyl groups(3a) at 8 and 9-positions was detrimental to the activity of thecompound and doses of 0.3 mg/disk were required to get a moder-ate bactericidal effect. Furthermore, 2a showed the highest inhibi-

tion zone against S. aureus and E. carotovora among all the testedpyrroloisoquinolines. For this latter microorganism, 2a possesseda potency that almost fell in the same range as the reference com-pound (tetracycline), suggesting that it could be a potential alter-native bactericidal agent to this ubiquitous plant pathogen witha wide host range. Thus, the presence in 2a of a chlorine atomand a methoxyl group at the 8-position and the 9-position, respec-tively, appeared to enhance activity.

Fungicide activity was tested against some phytopathogen fun-gi strains: A. parasiticus, T. viridae, F. culmorum, G. candidum andP. citrophthora (Table 2). Compounds 1a, 2a, 1c and 1e inhibitedfungal growth in vitro. Compounds 1d and 1f, in which thesubstitution at the 8-position was an amide side chain, did not dis-play growth inhibition of the selected fungi at 0.2 mg/disk, unlike


the bactericidal test. In series 1, the most active compounds were1a and their fluorinated analog 1e, which showed similar potency.Consequently, it seems that the benzylic moiety located at the 8-position on the pyrroloisoquinoline structure contributed posi-tively to its antifungal properties, even if there was, or was not, ahalogen atom. Compound 2a, which was seen to be the most po-tent bactericidal agent among the tested pyrroloisoquinolines,had a moderate antifungal effect. Its derivatives 2b and 2c didnot display any antifungal activity even at 0.4 mg/disk. Surpris-ingly, other disubstituted series 3 analogous (3a–3d) also showedno fungicidal effect at the highest dose tested (0.4 mg/disk).

In conclusion, we prepared new eleven 8-substituted pyrrol-o[2,1-a]isoquinolinones together with the known 3a and trolline(3b) via a double cyclization unleashed by Bischler-Napieralskicyclodehydration and an imine reduction sequence in a few stepswith good yields. The SAR studies reveal that the benzylic moietyat the 8-position, as in compounds 1a (O-benzyl group) and 1e(O-p-fluoro-benzyl group), and the 8-chloro-9-methoxy substitu-tion as in 2a, provide the most fungicide and bactericide agents,respectively.

3. Material and methods

3.1. General instrumentation

Melting points were taken on a Cambridge microscope instru-ment coupled with a Reichert-Jung. EIMS was recorded in a VGAuto Spec Fisons spectrometer instrument (Fisons, Manchester,United Kingdom). 1H NMR and 13C NMR spectra were recordedwith CDCl3 as a solvent in a Bruker AC-300, AC-400 or AC-500.Multiplicities of 13C NMR resonances were assigned by DEPTexperiments. COSY, HSQC and HMBC correlations were recordedat 400 and 500 MHz (Bruker AC-400 or AC-500). The assignmentsof all compounds were made by COSY, DEPT, HSQC and HMBC.All the reactions were monitored by analytical TLC with silica gel60 F254 (Merck 5554). Residues were purified by silica gel 60(40–63 lm, Merck 9385) column chromatography. Solvents andreagents were purchased from commercial sources. Quoted yieldsare of purified material.

3.2. General procedure for the synthesis of amides (1–3)

3.2.1. 3-Benzyloxy-benzaldehydeA mixture of 3-hydroxybenzaldehyde (3 g, 24.59 mmol), benzyl

chloride (4.1 mL, 35.79 mmol) and anhydrous K2CO3 (2.4 g,17.39 mmol) in absolute EtOH (30 mL) was refluxed for 6 h. Thenthe reaction mixture was concentrated to dryness, redissolved in10 mL of CH2Cl2 and washed with 5% aqueous NaOH (3 � 10 mL).The organic layer was dried with anhydrous Na2SO4 and evapo-rated to dryness. The residue was purified by silica gel columnchromatography (hexane/EtOAc, 8:2) to afford 5.1 g of 3-benzyl-oxy-benzaldehyde (98%) as a white solid. Mp: 54–56 �C; 1H NMR(300 MHz, CDCl3): d = 9.85 (s, 1H, CHO), 7.32 (m, 9H, H-2, H-4,H-5, H-6, Ph), 5.11 (s, 2H, OCH2Ph); 13C NMR (75 MHz, CDCl3) d192.5 (CHO), 159.7 (C-3), 138.2 (C-1), 136.7 (C-10), 130.5 (CH-5),129.1 (CH-30, CH-50), 128.6 (CH-40), 127.9 (CH-20, CH-60), 124.1(CH-6), 122.6 (CH-4), 113.6 (CH-2), 70.6 (OCH2Ph); ESMS m/z (%):213 (100) [M+1]+.

3.2.2. 3-Benxyloxy-b-nitrostyreneA mixture of 3-benzyloxy-benzaldehyde (1 g, 4.71 mmol), nitro-

methane (0.7 mL, 12.89 mmol) and NH4OAc (0.8 g, 10.41 mmol) inAcOH (12.5 mL) was refluxed overnight. After cooling, the mixturewas diluted with water and extracted with CH2Cl2 (3 � 10 mL). Theorganic solution was washed with brine (2 � 10 mL) and water(2 � 10 mL), dried with anhydrous Na2SO4 and evaporated to dry-

ness to obtain the 3-benzyloxy-b-nitrostyrene (1.2 g, 97%) as yel-low needles, which was used in the following step with nofurther purification. Mp: 80–83 �C; 1H NMR (300 MHz, CDCl3):d = 7.88 (d, J = 13.8 Hz, 1H, H-b), 7.48 (d, J = 13.8 Hz, 1H, H-a),7.40 (m, 7H, H-2, H-5, Ph), 7.01 (m, 2H, H-4, H-6), 5.11 (s, 2H,OCH2Ph); 13C NMR (75 MHz, CDCl3) d = 159.6 (C-3), 139.4 (CH-b),137.8 (C-1), 136.7 (C-10), 130.9 (CH-a), 129.1-127.9 (6C, CH-5,CH20-CH-60), 122.4 (CH-6), 119.2 (CH-4), 115.5 (CH-2), 70.6(OCH2Ph); ESMS m/z (%): 256 (100) [M+1]+.

3.2.3. 3-Chloro-4-methoxy-b-nitrostyrene3-Chloro-4-methoxy-benzaldehyde (1.0 g, 5.87 mmol) was sub-

mitted to the same conditions depicted above to obtain the 3-chloro-4-methoxy-b-nitrostyrene (1.1 g, 88%) as yellow needles,which was used in the following step with no further purifica-tion.26 Mp: 143–145 �C; 1H NMR (400 MHz, CDCl3): d = 7.91 (d,J = 13.7 Hz, 1H, H-b), 7.59 (d, J = 2.2 Hz, 1H, H-2), 7.52(d, J = 13.7 Hz, 1H, H-a), 7.44 (dd, J = 8.6, 2.2 Hz, 1H, H-6), 6.97(d, J = 8.6 Hz, 1H, H-5), 3.90 (s, 3H, OCH3-4); 13C NMR (100 MHz,CDCl3) d = 158.4 (C-4), 138.4 (CH-b), 136.4 (CH-a), 130.9 (CH-2),130.1 (CH-6), 124.2 (C-1), 123.7 (C-3), 112.7 (CH-5), 56.8 (OCH3);MS (EI) m/z (%): 213.5 (55) [M]+, 185 (100).

3.2.4. b-(3-Benzyloxy-phenyl)ethylamineA mixture of 3-benzyloxy-b-nitrostyrene (600 mg, 2.35 mmol)

in 10 mL of anhydrous THF was added dropwise to a well-stirredsuspension of LiAlH4 (0.3 g, 8.7 mmol) in 13 mL of anhydrousEt2O under nitrogen atmosphere, and was refluxed for 2 h. Thenthe reaction mixture was cooled and the excess of reagent de-stroyed by a dropwise addition of H2O. After a partial evaporationof the filtered portion, the aqueous solution was extracted withCH2Cl2 (3 � 10 mL). The organic layers were washed with brine,dried over Na2SO4 and the solvent was removed to give the b-(3-benzyloxy-phenyl)ethylamine (465 mg, 87%) as a yellow oil. 1HNMR (300 MHz, CDCl3): d = 7.20 (m, 6H, H-5, Ph), 6.83 (m, 3H, H-2, H-4, H-6), 5.03 (s, 2H, OCH2Ph), 2.88 (t, J = 13.5 Hz, 2H, CH2-b),2.65 (t, J = 13.5 Hz, 2H, CH2-a); 13C NMR (75 MHz, CDCl3):d = 159.3 (C-3), 140.1 (C-10), 137.4 (C-1), 130.0 (CH-5), 129.9 (CH-30, CH-50), 128.3 (CH-40), 127.9 (CH-20, CH-60), 121.9 (CH-6), 116.0(CH-2), 112.8 (CH-4), 70.3 (OCH2Ph), 43.7 (CH2-b), 40.3 (CH2-a);ESMS m/z (%): 228 (100) [M+1]+.

3.2.5. b-(3-Chloro-4-methoxyphenyl)ethylamine3-Chloro-4-methoxy-b-nitrostyrene (1.0 g, 4.70 mmol) was

submitted to the same conditions depicted above to obtain the2-(3-chloro-4-methoxyphenyl)ethylamine (905 mg, 90%) as a yel-low oil. The compound was used in further reaction without puri-fication.26 1H NMR (400 MHz, CDCl3): d = 7.30 (d, J = 2.2 Hz, 1H,H-2), 7.20 (dd, J = 8.5, 2.2 Hz, 1H, H-6), 6.80 (d, J = 8.5 Hz, 1H,H-5), 3.90 (s, 3H, OCH3-4), 3.10 (m, 2H, H-b), 2.80 (m, 2H, H-a);13C NMR (100 MHz, CDCl3): d = 153.8 (C-4), 133.3 (C-1), 130.8(CH-2), 128.4 (CH-6), 122.6 (C-3), 112.5 (CH-5), 56.5 (OCH3), 43.8(CH2-a), 39.1 (CH2-b); MS (EI) m/z (%): 185 (45) [M]+.

3.2.6. Ethyl [b-(3-benzyloxy)phenethylamino]-oxobutanoate (1)An amount of 0.47 mL of ethyl succinyl chloride (3.31 mmol)

was added dropwise at 0 �C to a solution of b-(3-benzyloxy-phenyl)ethylamine (446 mg, 2.12 mmol) in CH2Cl2 and 5% aqueousNaOH (4 mL). The reaction was stirred at room temperature over-night to be then extracted with CH2Cl2 (3 � 10 mL). The combina-tion of the organic phases was washed with brine (2 � 10 mL) andH2O (2 � 10 mL), dried over Na2SO4 and evaporated to dryness. Theresidue was purified by silica gel column chromatography (hexane/EtOAc 6:4) to afford 540 mg of amide 1 (77%) as a yellow powder.Mp: 68–71 �C; 1H NMR (500 MHz, CDCl3): d = 7.22 (m, 6H, H-5, Ph),6.76 (m, 3H, H-2, H-4, H-6), 5.09 (s, 2H, OCH2Ph), 4.01 (q, J = 7.9 Hz,


2H, CO2CH2CH3), 3.40 (q, J = 6.3 Hz, 2H, CH2-a), 2.81 (t, J = 6.3 Hz,2H, CH2-b), 2.65 (t, J = 6.5 Hz, 2H, CH2CO), 2.37 (t, J = 6.5 Hz,CH2CONH), 1.10 (t, J = 7.9 Hz, 3H, CO2CH2CH3); 13C NMR(125 MHz, CDCl3): d = 172.9 (CO2CH2CH3), 171.3 (CONH), 159.0(C-3), 140.5 (C-10), 136.9 (C-1), 129.6 (CH-5), 128.5 (CH-30, CH-50), 127.9 (CH-40), 127.4 (CH-20, CH-60), 121.3 (CH-6), 115.3 (CH-2), 112.8 (CH-4), 69.8 (OCH2Ph), 60.6 (CO2CH2CH3), 40.5 (CH2-a),35.6 (CH2-b), 31.0 (CH2CONH), 29.5 (CH2CO), 14.1 (CO2CH2CH3);ESMS m/z (%): 356 (100) [M+1]+.

3.2.7. Ethyl [b-(3-chloro-4-methoxy)phenethylamino]-oxobut-anoate (2)

b-(3-Chloro-4-methoxyphenyl)ethylamine (1 g, 5.39 mmol)was submitted to the same conditions depicted above. The residuewas purified by silica gel column chromatography (hexane/EtOAc5:5) to afford 900 mg of amide 2 (53%) as a yellow powder. Mp:109–112 �C; 1H NMR (500 MHz, CDCl3): d = 7.17 (d, J = 2.1 Hz, 1H,H-2), 7.04 (dd, J = 8.3, 2.1 Hz, 1H, H-6), 6.84 (d, J = 8.3 Hz, 1H, H-5), 4.11 (q, J = 7.9 Hz, 2H, CO2CH2CH3), 3.85 (s, 3H, OCH3–4), 3.43(q, J = 6.3 Hz, 2H, CH2-a), 2.70 (t, J = 6.3, 2H, CH2-b), 2.60 (t,J = 6.5 Hz, 2H, CH2CO), 2.41 (t, J = 6.5 Hz, CH2CONH), 1.22 (t,J = 7.9 Hz, 3H, CO2CH2CH3); 13C NMR (125 MHz, CDCl3): d = 173.1(CO2CH2CH3), 171.8 (CONH), 153.5 (C-4), 131.9 (C-1), 130.3 (CH-2), 127.8 (CH-6), 122.2 (CH-3), 112.1 (CH-5), 60.7 (CO2CH2CH3),56.1 (OCH3), 40.6 (CH2-a), 34.3 (CH2-b), 30.9 (CH2CONH), 29.5(CH2CO), 14.0 (CO2CH2CH3); ESMS m/z (%): 314.5 (100) [M+1]+.

3.2.8. Ethyl [b-(3,4-dimethoxy)phenethylamino]-oxobutanoate(3)

b-(3,4-Dimethoxyphenyl)ethylamine (1 g, 5.52 mmol) was sub-mitted to the same conditions depicted above. The residue waspurified by silica gel column chromatography (hexane/EtOAc 5:5)to afford 1.5 g of amide 3 (88%) as a white powder. Mp: 48–51 �C. 1H NMR (500 MHz, CDCl3): d = 6.81 (d, J = 2.1 Hz, 1H, H-5),6.70 (m, 2H, H-2, H-6), 4.11 (q, J = 7.9 Hz, 2H, CO2CH2CH3), 3.85(s, 3H, OCH3-4), 3.84 (s, 3H, OCH3-3), 3.43 (q, J = 6.3 Hz, 2H,CH2-a), 2.70 (t, J = 6.3, 2H, CH2-b), 2.60 (t, J = 6.5 Hz, 2H, CH2CO),2.41 (t, J = 6.5 Hz, CH2CONH), 1.22 (t, J = 7.9 Hz, 3H, CO2CH2CH3);13C NMR (125 MHz, CDCl3): d = 173.4 (CO2CH2CH3), 171.9 (CONH),149.4 (C-3), 148.0 (C-4), 131.9 (C-1), 121.1 (CH-6), 112.4 (CH-5),111.8 (CH-2), 61.0 (CO2CH2CH3), 56.2 (2� OCH3), 41.2 (CH2-a),35.6 (CH2-b), 31.4 (CH2CONH), 29.8 (CH2CO), 14.5 (CO2CH2CH3);ESMS m/z (%): 310 (100) [M+1]+.

3.3. General procedure for the synthesis of 1,2,3,5,6,10b-pyrrolo[2,1-a]isoquinolin-3-ones (1a–3a)

3.3.1. 8-Benzyloxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1a)

A solution of ethyl 4-[b-(3-benzyloxyphenyl)ethylamino]-4-oxobutanoate (1) (300 mg, 0.84 mmol) in dry CH2Cl2 (30 mL) wastreated with POCl3 (0.39 mL, 4.20 mmol) and was refluxed for 6 hin a nitrogen atmosphere. The reaction mixture was diluted withH2O (10 mL) and extracted with CH2Cl2 (3 � 10 mL). The organicsolution was dried over Na2SO4 and evaporated to dryness. Theresidue was dissolved in MeOH (25 mL) and treated with NaBH4

(400 mg, 10.57 mmol) at room temperature. The reaction mixturewas stirred for 2 h. Afterward, H2O (5 mL) was added and the or-ganic solvent was removed under reduced pressure. The aqueousmixture was made basic and extracted with CH2Cl2 (3 � 10 mL),dried over Na2SO4 and concentrated. The residue was purified bysilica gel column chromatography (toluene/EtOAc/MeOH/Et3N,6:3:1:0.1) to obtain 105 mg of the 8-benzyloxy-pyrrolo[2,1-a]iso-quinolin-3-one 1a (43%) as a yellow oil. 1H NMR (500 MHz, CDCl3):d = 7.36 (m, 5H, Ph), 7.02 (d, J = 8.5 Hz, 1H, H-10), 6.87 (dd, J = 8.5,2.5 Hz, 1H, H-9), 6.75 (d, J = 2.5 Hz, 1H, H-7), 5.04 (s, 2H, OCH2Ph),

4.71 (t, J = 8 Hz, 1H, H-10b), 4.23 (ddd, J = 12.4, 5.8, 2.7 Hz, 1H, H-5a), 3.05 (m, 1H, H-5b), 2.91 (m, 1H, H-6a), 2.71 (m, 1H, H-6b),2.61 (m, 1H, H-1a), 2.54 (m, 1H, H-2a), 2.45 (m, 1H, H-2b), 1.82(m, 1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 173.2 (NCO), 157.4(C-8), 136.7 (C-10), 134.8 (C-6a), 130.0 (C-10a), 128.5 (CH-30, CH-50), 127.9 (CH-40), 127.3 (CH-20, CH-60), 125.8 (CH-10), 114.6 (CH-7), 113.8 (CH-9), 69.9 (OCH2Ph), 56.3 (CH-10b), 36.8 (CH2-5),31.6 (CH2-2), 28.7 (CH2-6), 27.5 (CH2-1); ESMS m/z (%): 293 (100)[M]+.

3.3.2. 8-Chloro-9-methoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (2a)

Ethyl b-(3-chloro-4-methoxy-phenethylamino)-4-oxobutano-ate (2) (100 mg, 0.31 mmol) was submitted to the same conditionsdepicted above. The residue was purified by silica gel column chro-matography (toluene/EtOAc/MeOH/Et3N, 6:3:1:0.1) to obtain28 mg of 8-chloro-9-methoxy-pyrrolo[2,1-a]isoquinolin-3-one 2a(36%) as a yellow oil. 1H NMR (500 MHz, CDCl3): d = 7.15 (s, 1H,H-7), 6.62 (s, 1H, H-10), 4.72 (t, J = 8.0 Hz, 1H, H-10b), 4.28 (ddd,J = 12.9, 6.2, 2.7 Hz, 1H, H-5a), 3.86 (s, 3H, OCH3), 2.99 (m, 1H,H-5b), 2.84 (m, 1H, H-6a), 2.68 (m, 1H, H-6b), 2.66 (m, 1H,H-1a), 2.56 (m, 1H, H-2a), 2.48 (m, 1H, H-2b), 1.85 (m,1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 173.0 (NCO), 153.6(C-9), 136.9 (C-10a), 130.5 (CH-7), 127.8 (C-6a), 121.1 (C-8),108.3 (CH-10), 56.2 (CH-10b), 56.1 (OCH3), 36.9 (CH2-5), 31.6(CH2-2), 27.5 (CH2-6), 27.4 (CH2-1); ESMS m/z (%): 251 (100) [M]+.

3.3.3. 8,9-Dimethoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (3a)

Ethyl b-(3,4-dimethoxy-phenethylamino)-4-oxobutanoate (3)(1.2 g, 3.88 mmol) was submitted to the same conditions depictedabove. The residue was purified by silica gel column chromatogra-phy (CH2Cl2/MeOH/NN4OH, 95:5:0.1) to obtain 520 mg of 8, 9-dimethoxy-pyrrolo[2,1-a]isoquinolin-3-one 3a (54%) as a greenoil. 1H NMR (500 MHz, CDCl3): d = 6.59 (s, 1H, H-7), 6.54 (s, 1H,H-10), 4.70 (t, J = 8.0 Hz, 1H, H-10b), 4.26 (ddd, J = 12.9, 6.2,2.7 Hz, 1H, H-5a), 3.83 (s, 3H, OCH3-9), 3.82 (s, 3H, OCH3-8), 2.99(m, 1H, H-5b), 2.85 (m, 1H, H-6a), 2.65 (m, 1H, H-6b), 2.60 (m,1H, H-1a), 2.55 (m, 1H, H-2a), 2.44 (m, 1H, H-2b), 1.80 (m, 1H,H-1b); 13C NMR (125 MHz, CDCl3): d = 173.1 (NCO), 148.0 (C-8),147.8 (C-9), 129.2 (C-10a), 125.4 (C-6a), 111.6 (CH-7), 107.6 (CH-10), 56.5 (CH-10b), 55.9 (OCH3-8), 55.8 (OCH3-9), 36.9 (CH2-5),31.6 (CH2-2), 27.9 (CH2-6), 27.6 (CH2-1); ESMS m/z (%): 247 (100)[M]+.

3.4. General procedure for the synthesis of 1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-ones (1b–1f)

3.4.1. 8-Hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1b)

8-benzyloxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1a) (200 mg, 0.68 mmol) was refluxed for 3 h in a mixtureof equal volumes of ethanol and concentrated HCl (50 mL). Thereaction mixture was evaporated to dryness and the residue puri-fied by silica gel column cromatography (CH2Cl2/MeOH 94:6) togive 101 mg of 8-hydroxy-pyrrolo[2,1-a]isoquinolin-3-one 1b(72%) as a white oil. 1H NMR (500 MHz, CDCl3): d = 6.76 (d,J = 8.3 Hz, 1H, H-10), 6.53 (dd, J = 8.3, 2.0 Hz, 1H, H-9), 6.42 (d,J = 2.0 Hz, 1H, H-7), 4.56 (t, J = 8 Hz, 1H, H-10b), 3.90 (m, 1H, H-5a), 2.91 (m, 1H, H-5b), 2.67 (m, 1H, H-6a), 2.56 (m, 1H, H-6b),2.45 (m, 1H, H-1a), 2.37 (m, 1H, H-2a), 2.22 (m, 1H, H-2b), 1.65(m, 1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 174.0 (NCO), 155.3(C-8), 134.2 (C-6a), 128.0 (C-10a), 125.4 (CH-10), 114.7 (CH-7),113.9 (CH-9), 56.5 (CH-10b), 36.9 (CH2-5), 31.3 (CH2-2), 28.1(CH2-6), 27.1 (CH2-1); ESMS m/z (%): 203 (100) [M]+.


3.4.2. 8-Ethylcarbamate-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1c)

A solution of 8-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]-isoquinolin-3-one (1b) (35 mg, 0.17 mmol) in dry acetone (10 mL)was treated with ethyl isocyanate (0.34 mmol, 0.03 mL). Afterrefluxing for 3 h, the reaction mixture was concentrated to dryness,redissolved in 10 mL of CH2Cl2 and washed with H2O (3 � 10 mL).The organic layer was dried with anhydrous Na2SO4, filtered andevaporated under reduced pressure. The residue was purifiedthrough a silica gel column (CH2Cl2/MeOH 97:3) to afford 8-ethylc-arbamate-pyrrolo[2,1-a]isoquinolin-3-one 1c (34 mg, 70%) as ayellow oil. 1H NMR (500 MHz, CDCl3): d = 7.09 (d, J = 8.3 Hz, 1H,H-10), 7.00 (dd, J = 8.3, 2.0 Hz, 1H, H-9), 6.91 (d, J = 2.0 Hz, 1H, H-7), 4.71 (t, J = 8.0 Hz, 1H, H-10b), 4.26 (m, 1H, H-5a), 3.30 (m, 2H,CH3CH2NHCO), 3.07 (m, 1H, H-5b), 2.92 (m, 1H, H-6a), 2.77 (m,1H, H-6b), 2.63 (m, 1H, H-1a), 2.53 (m, 1H, H-2a), 2.45 (m, 1H,H-2b), 1.86 (m, 1H, H-1b), 1.20 (m, 3H, CH3CH2NHCO); 13C NMR(125 MHz, CDCl3): d = 173.2 (NCO), 154.4 (C-8), 149.6 (NHCO),134.8 (C-6a), 134.4 (C-10a), 128.7 (CH-10), 125.1 (CH-7), 120.3(CH-9), 56.5 (CH-10b), 36.8 (CH2-5), 36.7 (CH3CH2NHCO), 31.3(CH2-2), 28.5 (CH2-6), 27.5 (CH2-1), 15.0 (CH3CH2NHCO); ESMSm/z (%): 297 (100) [M+Na]+.

3.4.3. 8-(1-Piperidinethoxy)-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1d)

A mixture of 8-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1b) (30 mg, 0.14 mmol), 2-bromo-1-(piperi-din-1-yl)ethanone (19 mg, 0.14 mmol) and anhydrous K2CO3

(19 mg, 0.14 mmol) in absolute EtOH (10 mL) was refluxed for6 h. Afterward, the reaction mixture was concentrated to dryness,redissolved in 10 mL of CH2Cl2 and washed with 5% aqueous NaOH(3 � 10 mL). The organic layer was dried with anhydrous Na2SO4,filtered and evaporated to dryness. The residue was purified by sil-ica gel column chromatography (toluene/EtOAc/MeOH/Et3N,6:3:1:0.1) to afford 40 mg of 8-(1-piperidinethoxy)-pyrrolo[2,1-a]isoquinolin-3-one 1d (82%) as a green oil. 1H NMR (500 MHz,CDCl3): d = 7.02 (d, J = 8.5 Hz, 1H, H-10), 6.84 (dd, J = 8.5, 2.6 Hz,1H, H-9), 6.70 (d, J = 2.6 Hz, 1H, H-7), 4.71 (t, J = 6.7 Hz, 1H, H-10b), 4.65 (s, 2H, OCH2CO), 4.23 (m, 1H, H-5a), 3.55 (t, J = 5.3 Hz,2H, CH2N), 3.46 (t, J = 5.3 Hz, 2H, CH2N), 3.04 (m, 1H, H-5b), 2.89(m, 1H, H-6a), 2.73 (m, 1H, H-6b), 2.56 (m, 1H, H-1a), 2.52 (m,1H, H-2a), 2.44 (m, 1H, H-2b), 1.82 (m, 1H, H-1b), 1.64–1.54 (m,6H, (CH2)3N); 13C NMR (125 MHz, CDCl3): d = 173.2 (NCO-3),166.0 (NCOCH2O), 156.7 (C-8), 135.0 (C-6a), 130.6 (C-10a), 125.9(CH-10), 114.5 (CH-7), 113.6 (CH-9), 67.5 (OCH2CO), 56.4 (CH-10b), 46.3 and 43.2 (2 � CH2N), 36.9 (CH2-5), 31.7 (CH2-2), 28.6(CH2-6), 27.5 (CH2-1), 26.4 and 25.4 (2 � CH2CH2N), 24.3(CH2(CH2)2N); ESMS m/z (%): 328 (100) [M]+.

3.4.4. 8-(4-Fluorobenzyloxy)-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1e)

A mixture of 8-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1b) (20 mg, 0.10 mmol), p-fluorobenzyl chlo-ride (0.01 mL) and anhydrous K2CO3 (10 mg) in absolute ethanol(10 mL) was refluxed overnight. Afterward, the reaction mixturewas concentrated to dryness, redissolved in 10 mL of CH2Cl2 andwashed with 5% aqueous NaOH (3 � 10 mL). The organic layerwas dried with anhydrous Na2SO4, filtered and evaporated to dry-ness under reduced pressure. The residue was purified by silica gelcolumn chromatography (CH2Cl2/MeOH 97:3) to afford 25 mg of 8-(4-fluorobenzyloxy)-pyrrolo[2,1-a]isoquinolin-3-one 1e (80%) as ayellow oil. 1H NMR (500 MHz, CDCl3): d = 7.38 (m, 2H, PhF), 7.07(m, 3H, PhF, CH-10), 6.86 (dd, J = 8.5, 2.5 Hz, 1H, H-9), 6.74 (d,J = 2.5 Hz, 1H, H-7), 5.00 (s, 2H, OCH2Ph), 4.73 (t, J = 7.8 Hz, 1H,H-10b), 4.25 (m, 1H, H-5a), 3.04 (m, 1H, H-5b), 2.91 (m, 1H, H-6a), 2.75 (m, 1H, H-6b), 2.63 (m, 1H, H-1a), 2.56 (m, 1H, H-2a),

2.46 (m, 1H, H-2b), 1.82 (m, 1H, H-1b); 13C NMR (125 MHz, CDCl3):d = 173.4 (NCO), 162.5 (C-40, 1JCF = 246 Hz), 157.3 (C-8), 135.0 (C-10), 132.5 (C-6a), 130.2 (C-10a), 129.3 (2CH, CH-20, CH-60), 125.9(CH-10), 115.5 (2CH, CH-30, CH-50, 2JCF = 21.7 Hz), 114.5 (CH-7),113.9 (CH-9), 69.3 (OCH2Ph), 56.4 (CH-10b), 36.9 (CH2-5), 31.7(CH2-2), 28.7 (CH2-6), 27.6 (CH2-1); ESMS m/z (%): 311 (100) [M]+.

3.4.5. 8-Phenylacetamide-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1f)

A mixture of 8-hydroxy-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]isoquinolin-3-one (1b) (30 mg, 0.14 mmol), 2-bromo-N-phenyl-acetamide (29 mg, 0.14 mmol) and anhydrous K2CO3 (19 mg,0.14 mmol) in EtOH (10 mL) was refluxed for 6 h. Then the reactionmixture was concentrated to dryness, redissolved in 10 mL ofCH2Cl2 and washed with 5% aqueous NaOH (3 � 10 mL). The organ-ic layer was dried with anhydrous Na2SO4 and evaporated to dry-ness. The residue was purified by silica gel columnchromatography (toluene/EtOAc/MeOH/Et3N, 6:3:1:0.1) to afford40 mg of 8-phenylacetamide-pyrrolo[2,1-a]isoquinolin-3-one 1f(80%) as a yellow oil. 1H NMR (500 MHz, CDCl3): d = 7.58 (d, 2H,J = 7.6 Hz, H-20, H-60), 7.36 (t, 2H, J = 7.5 Hz, H-30, H-50), 7.16 (t,1H, J = 7.5 Hz, H-40), 7.10 (d, J = 8.5 Hz, 1H, H-10), 6.90 (dd, J = 8.5,2.5 Hz, 1H, H-9), 6.78 (d, J = 2.5 Hz, 1H, H-7), 4.74 (t, J = 7.9 Hz,1H, H-10b), 4.60 (s, 2H, OCH2CO), 4.28 (ddd, J = 12.9, 6.2, 2.7 Hz,1H, H-5a), 3.06 (m, 1H, H-5b), 2.93 (m, 1H, H-6a), 2.87 (m, 1H,H-6b), 2.65 (m, 1H, H-1a), 2.57 (m, 1H, H-2a), 2.47 (m, 1H, H-2b), 1.84 (m, 1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 173.2(PhNCO), 166.0 (NCO-3), 155.7 (C-8), 136.7 (C-10), 135.6 (C-6a),131.8 (C-10a), 129.1 (2CH, CH-20, CH-60), 126.3 (CH-10), 124.9(CH-40), 120.1 (2CH, CH-30, CH-50), 114.9 (CH-7), 113.7 (CH-9),67.7 (OCH2CO), 56.3 (CH-10b), 36.8 (CH2-5), 31.7 (CH2-2), 28.7(CH2-6), 27.6 (CH2-1); ESMS m/z (%): 336 (100) [M]+.

3.5. General procedure for the synthesis of 1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-ones (2b and 3b)

3.5.1. 8-Chloro-9-hydroxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-one (2b)

8-Chloro-9-methoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]iso-quinolin-3-one (2a, 0.23 mmol, 60 mg) in dry CH2Cl2 was stirred at�78 �C. Then, 0.10 mL of BBr3 were added under nitrogen atmo-sphere and the resulting mixture was stirred for 2 h at room tem-perature. The reaction mixture was evaporated to dryness andpurified by silica gel column chromatography (CH2Cl2/MeOH90:10) to afford 50 mg of 8-chloro-9-hydroxy-1,2,3,5,6,10b-hexa-hydropyrrolo[2,1-a]isoquinolin-3-one 2b (91%) as a colorless oil.1H NMR (500 MHz, CDCl3): d = 7.18 (s, 1H, H-7), 6.97 (s, 1H, H-10), 4.48 (t, J = 8.0 Hz, 1H, H-10b), 4.30 (ddd, J = 12.9, 6.2, 2.7 Hz,1H, H-5a), 2.87 (m, 1H, H-5b), 2.67 (m, 1H, H-6a), 2.51 (m, 1H,H-6b), 2.42 (m, 1H, H-1a), 2.35 (m, 1H, H-2a), 2.31 (m, 1H, H-2b), 1.62 (m, 1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 172.5(NCO), 149.3 (C-9), 138.1 (C-10a), 131.3 (CH-7), 130.3 (C-6a),119.9 (C-8), 113.6 (CH-10), 56.2 (CH-10b), 37.0 (CH2-5), 31.7(CH2-2), 27.5 (CH2-6), 27.4 (CH2-1); ESMS m/z (%): 237.5 (100)[M]+.

3.5.2. (±)-Trolline (3b)8,9-Dimethoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquino-

lin-3-one (3a, 0.80 mmol, 200 mg) in dry CH2Cl2 was stirred at�78 �C. Then, 0.25 mL of BBr3 were added under nitrogen atmo-sphere and the resulting mixture was stirred for 2 h at room tem-perature. The reaction mixture was evaporated to dryness andpurified by silica gel column chromatography (CH2Cl2/MeOH85:15) to afford 165 mg of 8,9-dihydroxy-1,2,3,5,6,10b-hexahydro-pyrrolo[2,1-a]isoquinolin-3-one 3b (93%) as a white powder. Mp:245–247 �C; 1H NMR (500 MHz, CDCl3): d = 6.99 (s, 1H, H-7),


6.98 (s, 1H, H-10), 4.54 (t, J = 8.0 Hz, 1H, H-10b), 4.32 (ddd, J = 12.9,6.2, 2.7 Hz, 1H, H-5a), 2.89 (m, 1H, H-5b), 2.74 (m, 1H, H-6a), 2.48(m, 1H, H-6b), 2.43 (m, 1H, H-2a), 2.35 (m, 1H, H-2b), 2.33 (m, 1H,H-1a) 1.66 (m, 1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 172.6(NCO), 146.3 (C-8), 146.1 (C-9), 129.3 (C-10a), 124.7 (C-6a), 118.7(CH-10), 116.5 (CH-7), 56.4 (CH-10b), 37.4 (CH2-5), 31.9 (CH2-2),28.1 (CH2-6), 28.0 (CH2-1); ESMS m/z (%): 219 (100) [M]+.

3.6. General procedure for the synthesis of 9-substituted-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinolin-3-ones (2c, 3cand 3d)

3.6.1. 8-Chloro-9-ethylcarbamate-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]isoquinoline-3-one (2c)

A solution of 8-chloro-9-hydroxy-1,2,3,5,6,10b-hexahydropyr-rolo[2,1-a]isoquinolin-3-one 2b (20 mg, 0.084 mmol) in dry ace-tone (10 mL) was treated with ethyl isocyanate (0.17 mmol,0.01 mL). After refluxing for 3 h, the reaction mixture was concen-trated to dryness, redissolved in 10 mL of CH2Cl2 and washed withH2O (3 � 10 mL). The organic layer was dried with anhydrousNa2SO4, filtered and evaporated under reduced pressure. The resi-due was purified through a silica gel column (CH2Cl2/MeOH 95:5)to afford 8-chloro-9-ethylcarbamate-1,2,3,5,6,10b-hexahydropyr-rolo[2,1-a]isoquinolin-3-one 2c. (85%) as a yellow oil. 1H NMR(500 MHz, CDCl3): d = 7.10 (s, 1H, H-7), 6.78 (s, 1H, H-10), 4.68 (t,J = 8.0 Hz, 1H, H-10b), 4.42 (ddd, J = 12.9, 6.2, 2.7 Hz, 1H, H-5a),3.72 (m, 2H, CH3CH2NHCO), 3.01 (m, 1H, H-5b), 2.83 (m, 1H, H-6a), 2.63 (m, 1H, H-6b), 2.61 (m, 1H, H-1a), 2.55 (m, 1H, H-2a),2.46 (m, 1H, H-2b), 1.80 (m, 1H, H-1b), 1.20 (m, 3H, CH3CH2NHCO);13C NMR (125 MHz, CDCl3): d = 173.3 (NCO), 158.5 (C-9), 150.5(NHCO), 137.4 (C-10a), 130.6 (CH-7), 126.0 (C-6a), 119.3 (C-8),112.5 (CH-10), 56.5 (CH-10b), 37.4 (CH3CH2NHCO), 37.0 (CH2-5),31.7 (CH2-2), 27.4 (CH2-6), 27.3 (CH2-1), 15.4 (CH3CH2NHCO);ESMS m/z (%): 308.5 (100) [M]+.

3.6.2. 8,9-Bis(4-fluorobenzyloxy)-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]isoquinolin-3-one (3c)

A mixture of 8, 9-dihydroxy-1,2,3,5,6,10b-hexahydropyrrol-o[2,1-a]isoquinolin-3-one (3b) (50 mg, 0.22 mmol), p-fluorobenzylchloride (0.04 mL) and anhydrous K2CO3 (40 mg) in absolute etha-nol (20 mL) was refluxed overnight. Afterward, the reaction mix-ture was concentrated to dryness, redissolved in 10 mL of CH2Cl2

and washed with 5% aqueous NaOH (3 � 10 mL). The organic layerwas dried with anhydrous Na2SO4, filtered and evaporated to dry-ness under reduced pressure. The residue was purified by silica gelcolumn chromatography (CH2Cl2/MeOH 97:3) to afford 70 mg of 8-(4-fluorobenzyloxy)-pyrrolo[2,1-a]isoquinolin-3-one 1e (80%) as ayellow oil. 1H NMR (500 MHz, CDCl3): d = 7.38 (m, 4H, PhF), 7.04(m, 4H, PhF, CH-10), 6.69 (s, 1H, H-7), 6.64 (s, 1H, H-10), 5.05 (s,4H, OCH2Ph), 4.66 (t, J = 7.8 Hz, 1H, H-10b), 4.26 (m, 1H, H-5a),2.99 (m, 1H, H-5b), 2.84 (m, 1H, H-6a), 2.65 (m, 1H, H-6b), 2.56(m, 1H, H-1a), 2.51 (m, 1H, H-2a), 2.42 (m, 1H, H-2b), 1.76 (m,1H, H-1b); 13C NMR (125 MHz, CDCl3): d = 173.1 (NCO), 163.4 (C-40, 1JCF = 246 Hz), 161.4 (C-400, 1JCF = 246 Hz), 147.9 (C-8), 147.7 (C-9), 132.8 (C-10, C-100), 130.4 (C-10a), 129.2 (4CH, CH-30, CH-50,CH-300, CH-500), 126.9 (C-6a), 115.4 (4CH, CH-20, CH-60, CH-200, CH-600), 115.2 (CH-7), 112.2 (CH-10), 71.2 (OCH2Ph), 70.7 (OCH2Ph),56.4 (CH-10b), 36.9 (CH2-5), 31.7 (CH2-2), 28.0 (CH2-6), 27.5(CH2-1); ESMS m/z (%): 435 (100) [M]+.

3.6.3. 8,9-Bis(phenylacetamide)-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]isoquinolin-3-one (3d)

A mixture of 8, 9-dihydroxy-1,2,3,5,6,10b-hexahydropyrrol-o[2,1-a]isoquinolin-3-one (3b) (30 mg, 0.13 mmol), 2-bromo-N-phenylacetamide (61 mg, 0.26 mmol) and anhydrous K2CO3

(30 mg, 0.22 mmol) in EtOH (10 mL) was refluxed for 6 h. Then

the reaction mixture was concentrated to dryness, redissolved in10 mL of CH2Cl2 and washed with 5% aqueous NaOH (3 � 10 mL).The organic layer was dried with anhydrous Na2SO4 and evapo-rated to dryness. The residue was purified by silica gel columnchromatography (CH2Cl2/MeOH, 98:2) to afford 40 mg of 8,9-bis(phenylacetamide)-1,2,3,5,6,10a-hexahydropyrrolo[2,1-a]iso-quinolin-3-one 3d (63%) as a brown oil. 1H NMR (500 MHz, CDCl3):d = 7.52 (m, 4H, H-20, H-60, H-200, H-600), 7.26 (m, 4H, H-30, H-50, H-300, H-500), 7.10 (m, 2H, H-40, H-400), 6.75 (s, 1H, H-7), 6.72 (s, 1H,H-10), 4.68 (s, 2H,OCH2CO), 4.67 (t, J = 7.9 Hz, 1H, H-10b), 4.25(ddd, J = 12.9, 6.2, 2.7 Hz, 1H, H-5a), 2.98 (m, 1H, H-5b), 2.82 (m,1H, H-6a), 2.66 (m, 1H, H-6b), 2.62 (m, 1H, H-2a), 2.53 (m, 1H,H-2b), 2.43 (m, 1H, H-1a), 1.74 (m, 1H, H-1b); 13C NMR(125 MHz, CDCl3): d = 173.0 (NCO-3), 165.9 (PhNCO), 146.9 (C-9),146.2 (C-8), 136.6 (C-10), 132.4 (C-10a), 128.9 (C-30, C-50, C-300,C-500), 124.9 (CH-40, CH-400), 128.7 (C-6a), 120.0 (4CH, CH-20, CH-60, CH-200, CH-600), 115.7 (CH-7), 111.9 (CH-10), 69.6 (OCH2CO),56.2 (CH-10b), 36.7 (CH2-5), 31.6 (CH2-6), 27.9 (CH2-2), 27.4(CH2-1); ESMS m/z (%): 485 (100) [M]+.

3.7. Pharmacological assays

3.7.1. Target microorganismsFungicidal activity was measured against 5 phytopathogens:

Aspergillus parasiticus (CECT 2681), Trichoderma Viride (CECT2423), Fusarium culmorum (CCM 172), Phytophthora citrophthora(CECT 2353) and Geotrichum candidum (CCM 245). Seven differentbacterial strains were used to determine bactericidal activity:Bacillus cereus (CECT 148), Staphylococcus aureus (CECT 86), Entero-coccus faecalis (CECT 481), Salmonella typhi (CECT 409), Escherichiacoli (CECT 405), Escherichia coli (CECT 100) and Erwinia carotovora(CECT 225). The strains were provided by the Colección Españolade Cultivos Tipo (CECT) or by the Colección de la Cátedra de Micro-biología (CMM) of the Biotechnology Department (UniversidadPolitécnica de Valencia).

3.7.2. Antifungal and antibacterial activitiesThese assays were determined in triplicate by the paper disk-

agar diffusion assay according to Cole.29 The doses used in the as-says were 10, 15 or 20 lg/mm2 (0.2, 0.3 or 0.4 mg/disk). Fungalstrains were seeded in Petri dishes containing PDA culture mediumand were incubated for 7 days at 28 �C. Then, a Tween 80 solution(0.05%) in sterile distilled water was used to obtain a suspensioncontaining �106 conidia/mL. Next 1 mL of this conidia suspensionwas added to 15 mL of PDA in a Petri dish. After solidification, fourWhatman disks (No. 113, 0.5 cm diameter) impregnated with thetested products, at appropriate doses, were added to these Petridishes. The PDA plates containing disks impregnated with onlythe solvent used to dissolve the tested compounds were used asnegative controls, and the disks with benomyl (methyl-1-[butylc-arbamoyl]-2-benzimidazolecarbamate) (Sigma), at different con-centrations according to the fungal species assayed, were used aspositive controls. Fungicidal activity was determined by measuringthe inhibition zone developed around the paper disk, indicating azone of no growth.

In the bactericidal tests, 24-h cultures of each bacterium, main-tained in inclined tubes on solid culture medium, were reactivatedwith a Nutrient Broth (Difco) and were incubated for 24 h at 28 or37 �C according to the bacterium. Then, 1 mL of this suspensionwas inoculated in a Petri plate, and 15 mL of the culture mediumPlate Count Agar (Difco) was added. When the medium was com-pletely solidified, five paper disks loaded with the tested productswere placed in the dish. These plates were incubated for 24 h in thedark at 28 or 37 �C, according to the bacterium. The plate CountAgar plates containing disks impregnated with only the solventused to dissolve the tested compounds were used as negative


controls, and a positive control with tetracycline chlorhydrate(0.2 mg/disk) was performed to appraise the level of activities. Bac-tericidal activity was determined by measuring the halo developedaround the paper disk.

3.8. Statistical analysis

Analysis of variance (ANOVA) was performed for the fungicidaland bactericidal data (Tables 1 and 2), and the least significant dif-ference (LSD) test was used to compare means (Statgraphics centu-rion XVI version).

References and notes

1. Bentley, K. W. Nat. Prod. Rep. 2005, 22, 249.2. Juszczak, R. J.; Russell, J. H. J. Biol. Chem. 1989, 264, 810.3. Elwan, N. M.; Abdelhadi, H. A.; Abdallah, T. A.; Hassaneen, H. M. Tetrahedron

1996, 52, 3451.4. Loesel, W.; Roos O.; Schonorrenberg, G. US Patent 4694085 1987; Chem. Abstr.

1988, 108, 204513.5. Kuo, R.-Y.; Wu, C.-C.; Chang, F.-R.; Yeh, J.-L.; Chen, I.-J.; Wu, Y.-C. Bioorg. Med.

Chem. Lett. 2003, 13, 821.6. Anderson, W. K.; Heider, A. R.; Raju, N.; Yucht, J. A. J. Med. Chem. 1988, 31, 2097.7. Kapadia, G. J.; Fales, H. M. Chem. Commun. 1968, 1688.8. Zhang, Q.; Tu, G.; Zhao, Y.; Cheng, T. Tetrahedron 2002, 58, 6795.9. Ahmad, V. U.; Rahman, A.-U.; Rasheed, T.; Rehman, H.-U. Heterocycles 1987, 26,

1251.10. Tanaka, H.; Tanaka, T.; Etoh, H.; Goto, S.; Terada, Y. Heterocycles 1999, 51, 2759.11. Zhang, F.; Simpkins, N. S.; Blake, A. J. Org. Biomol. Chem. 2009, 7, 1963–1979.12. Valencia, E.; Freyer, A. J.; Shamma, M. Tetrahedron Lett. 1984, 25, 599.

13. Moreau, A.; Couture, A.; Deniau, E.; Grandclaudon, P.; Lebrun, S. Tetrahedron2004, 60, 6169.

14. Wang, R. F.; Yang, X. W.; Ma, C. M.; Cai, S. Q.; Li, J. N.; Shoyama, Y. Heterocycles2004, 63, 1443.

15. Andersen, R. J.; Faulkner, D. J.; He, C. H.; Van Duyne, G. D.; Clardy, J. J. Am. Chem.Soc. 1985, 107, 5492.

16. Pässler, U.; Knölker, H.-J. The Pyrrolo[2,1-a]isoquinoline Alkaloids In TheAlkaloids: Chemistry and Biology; Knölker, H.-J., Ed.; Elsevier: San Diego, 2012;Vol. 70, pp 79–151.

17. Shono, T.; Hamaguchi, H.; Sasaki, M.; Fujita, S.; Nagami, K. J. Org. Chem. 1983,48, 1621.

18. Knölker, H. J.; Agarwal, S. Tetrahedron Lett. 2005, 46, 1173.19. Wu, T. R.; Chong, J. M. J. Am. Chem. Soc. 2006, 128, 9646.20. Hiemstra, H.; Speckamp, W. N. Additions to N-Acyliminium Ions In

Comprehensive Organic Chemistry; Trost, B. M., Fleming, I., Eds.; PergamonPress: Oxford, 1991; Vol. 2, pp 1047–1082.

21. Parham, W. E.; Bradsher, C. K. Acc. Chem. Res. 1982, 15, 300.22. Collado, M. I.; Manteca, I.; Sotomayor, N.; Villa, M. J.; Lete, E. J. Org. Chem. 1997,

62, 2080.23. Padwa, A.; Henning, R.; Kappe, C. O.; Reger, T. S. J. Org. Chem. 1998, 63, 1144.24. Bermejo, A.; Protais, P.; Blázquez, M. A.; Rao, K. S.; Zafra-Polo, M. C.; Cortes, D.

Nat. Prod. Lett. 1995, 6, 57.25. Berenguer, I.; El Aouad, N.; Andujar, S.; Romero, V.; Suvire, F.; Freret, T.;

Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Boulouard, M.; Cabedo, N.; Cortes, D.Bioorg. Med. Chem. 2009, 17, 4968.

26. El Aouad, N.; Berenguer, I.; Romero, V.; Marín, P.; Serrano, A.; Andujar, S.;Suvire, F.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Cabedo, N.; Cortes, D. Eur. J.Med. Chem. 2009, 44, 4616.

27. Cabedo, N.; Berenguer, I.; Figadère, B.; Cortes, D. Curr. Med. Chem. 2009, 16,2441.

28. Andujar, S.; Suvire, F.; Berenguer, I.; Cabedo, N.; Marín, P.; Moreno, L.; Ivorra,M. D.; Cortes, D.; Enriz, R. D. J. Mol. Model. 2012, 18, 419.

29. Cole, M. D. Biochem. Syst. Ecol. 1994, 22, 837.

Capítulo III:

Pirido[2,1-a]isoquinoleínas y oxazino[2,1-

a] isoquinoleínas inhibidoras de la cadena

respiratoria mitocondrial

Artículo 3: “Synthesis of pirido[2,1-a]isoquinolin-4-

ones and oxazino[2,3-a]isoquinolin-4-ones: New

inhibitors of mitocondrial respiratory chain” (En:

European Journal of Medicinal Chemistry 2013, 69, 69)

lable at ScienceDirect

European Journal of Medicinal Chemistry 69 (2013) 69e76

Contents lists avai

European Journal of Medicinal Chemistry

journal homepage: http: / /www.elsevier .com/locate/ejmech

Original article

Synthesis of pyrido[2,1-a]isoquinolin-4-ones and oxazino[2,3-a]isoquinolin-4-ones: New inhibitors of mitochondrial respiratory chain

Laura Moreno a, Nuria Cabedo b,*, Agathe Boulangé c, Javier Párraga a, Abraham Galán a,Stéphane Leleu c, María-Jesús Sanz d,e, Diego Cortes a, Xavier Franck c,**

aDepartamento de Farmacología, Facultad de Farmacia, Universidad de Valencia, Burjassot, 46100 Valencia, SpainbCentro de Ecología Química Agrícola-Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia, UPV, Campus de Vera s/n, Edificio 6C, 46022Valencia, SpaincNormandie Univ, COBRA, UMR 6014 et FR 3038, CNRS, Univ Rouen, INSA Rouen, 1 rue Tesnières, 76821 Mont-Saint-Aignan Cedex, FrancedDepartamento de Farmacología, Facultad de Medicina, Universidad de Valencia, 46013 Valencia, Spaine Institute of Health Research-INCLIVA, University Clinic Hospital of Valencia, Valencia, Spain


Article history:Received 13 May 2013Received in revised form9 August 2013Accepted 10 August 2013Available online 19 August 2013

Keywords:Benzo[a]quinolizinesBischlereNapieralski cyclodehydrationAcyl-ketene imine cyclocondensationRespiratory chain inhibitionCytotoxicity

* Corresponding author. Tel.: þ34 963 87 90 58.** Corresponding author. UMR-CNRS 6014 COBRA,Tesnière, 76131 Mont Saint Aignan Cedex, France. Tel

E-mail addresses: [email protected], [email protected] (X. Franck).

0223-5234/$ e see front matter � 2013 Elsevier Mashttp://dx.doi.org/10.1016/j.ejmech.2013.08.013

a b s t r a c t

Benzo[a]quinolizine is an important heterocyclic framework that can be found in numerous bioactivecompounds. The general scheme for the synthesis of these compounds was based on the preparation ofthe appropriate dihydroisoquinolines by BischlereNapieralski cyclization with good yields, followed bythe Pemberton method to form the oxazinones or pyridones derivatives via acyl-ketene imine cyclo-condensation. All the synthesized compounds were assayed in vitro for their ability to inhibit mito-chondrial respiratory chain. Most of the tested compounds were able to inhibit the integrated electrontransfer chain, measured as NADH oxidation, which includes complexes I, III and IV, in the low micro-molar range. Oxazino[2,3-a]isoquinolin-4-ones displayed greater activity than their pyrido[2,1-a]iso-quinolin-4-ones analogs. Indeed, the presence of a furan ring in C2 position of oxazino[2,3-a]isoquinolin-4-ones provided the compound (1g) with the most potent biological activity. Therefore, these com-pounds and especially the oxazinone derivatives are in the tendency of the new less toxic antitumoragents that target mitochondrial electron transport chain in a middle range potency.

� 2013 Elsevier Masson SAS. All rights reserved.

1. Introduction

Isoquinoline alkaloids are a large family of natural products witha variety of powerful biological activities [1e3], including the in-hibition of cellular proliferation [4]. Our research group has for along time been focused on the isolation and synthesis ofisoquinoline-containing compounds and a variety of chemicallydiverse structures with dopaminergic activity were obtained [5e9].In addition, we have also been interested in the search of newcompounds with antitumor properties via inhibition of mito-chondrial respiratory chain, such as Annonaceous acetogenins [10e12]. Mitochondrial electron transport chain, comprises the enzy-matic respiratory complexes (IeIV), the coenzyme Q (ubiquinone)

Université de Rouen, 1 rue.: þ33 (0) 2 35 52 24 [email protected] (N. Cabedo),

son SAS. All rights reserved.

and cytochrome c. In the electron transport chain a transfer ofelectrons occurs during the oxidative phosphorylation and a protongradient is established as the energy source for the ATP generation.In addition, when an electron escapes, it may react with molecularoxygen to form the superoxide radical (O2

��) which can be con-verted to hydrogen peroxide (H2O2) and other reactive oxygenspecies (ROS). Therefore, mitochondria play essential roles in theATP synthesis, ROS metabolism and apoptosis. It is important thatcells preserve an apposite level of intracellular ROS to keep redoxbalance and signaling cellular proliferation [13]. An increase ofintracellular ROS leads to cellular damage, including lipid peroxi-dation, oxidative DNA modifications, protein oxidation, enzymeinactivation and finally, the cell death. Furthermore, DNA damageand mutations in the enzymatic complexes are related to the mostcommon human neurodegenerative diseases, including Parkinson’sdisease [14] and aging [15]. Damage to the mitochondria has beenalso detected in cancer cells that increase the glycolytic activity(Warburg effect) to produce the ATP required for cellular functions[16]. Some anticancer agents have the ability to inhibit mitochon-drial electron transport and/or increase superoxide radical

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name

Delta:1_surname

Delta:1_given name




http://crossmark.dyndns.org/dialog/?doi=10.1016/j.ejmech.2013.08.013&domain=pdf

www.sciencedirect.com/science/journal/02235234

http://www.elsevier.com/locate/ejmech

http://dx.doi.org/10.1016/j.ejmech.2013.08.013



Fig. 1. Structure of benzo[a]quinolizine framework and examples of natural productscontaining it.

L. Moreno et al. / European Journal of Medicinal Chemistry 69 (2013) 69e7670

generation in tumor cells, causing apoptosis by reducing ATP levelsand/or cell-damaging. Indeed, the research on respiratory chaininhibitors has been focused on pharmacological [17] and agro-chemical targets, providing compounds with high potential forantitumor therapy [18] and pest control [19]. In addition, mito-chondrial respiratory chain inhibitors have been also extremelyuseful to establish the functional mechanisms and crystal struc-tures of complex I [20].

Benzo[a]quinolizine is an important heterocyclic frameworkthat can be found in numerous bioactive compounds (Fig. 1) [1]such as berberine [21] and emetine [22]. Furthermore, manybiologically-relevant 2-pyridones like the yohimbane-type alkaloidsempervilam [23] and the antineoplasic agent camptothecin arenatural products that contain multi ring-fused systems (Fig. 2).Recently, synthetic benzo[a]quinolizine derivatives have showneffect in reversing tumor cell resistance with IC50 values in themicromolar range [24].

The biological properties and chemical features of these mole-cules have attracted the chemists’ interest leading to the descrip-tion of a large number of synthetic routes [25]. Therefore, both theattractive structure of the benzo[a]quinolizine moiety togetherwith their potential activity as inhibitors of cellular proliferation,encouraged us to prepare pyrido[2,1-a]isoquinolin-4-ones and theoxygenated analogs oxazino[2,3-a]isoquinolin-4-ones in order toexplore their properties as mitochondrial respiratory chaininhibitors.

Several approaches for the synthesis of benzo[a]quinolizineshave been described in the literature [26], most of which involvedB-ring closure by formation of de C11aeC11b bond either by Bis-chlereNapieralski cyclodehydration or by related palladium-catalyzed cyclization [27,28]. Other reported methods include theformation of C1eC11b bond via Mannich cyclization of a dihy-droisoquinolinium ion [29] or B-ring closure by the formation ofthe C7eC7a bond from the appropriate N-substituted 2-arylpiperidine or pyridine derivatives [30]. Roy et al. [31] alsoshowed an effective procedure for the synthesis of benzo[a]qui-nolizin-4-thiones through the reaction of 3,4-dihydroisoquinolineswith different b-oxodithioesters. Furthermore, a conjugateaddition-dipolar cycloaddition cascade has been described for thesynthesis of benzo[a]quinolizines [32]. Among the different

Fig. 2. Examples of natural 2-pyridones and structures of pyrido[2

depicted approaches we are now interested in the study of thereactivity of imines since it can lead to original and biologicallyunexplored frameworks. Indeed, the microwave-assisted acyl-ketene imine cyclocondensation has been a useful strategy togenerate oxazinones. In this regard, Presset et al. [33] reported theuse of cyclic 2-diazo-1,3-diketones as acyl-ketene source by Wolffrearrangement, and Pemberton et al. [34] started from eitherMeldrum’s acid derivatives or 1,3-dioxine-4-ones to react withdihydroisoquinolines. Therefore, we have applied the Pembertonmethod using dioxinones as acyl-ketenes precursors takingadvantage of the chemical diversity generated by this procedure.The structureeactivity relationship (SAR) of the synthesizedproducts was further evaluated.

2. Results and discussion

2.1. Chemistry

The general synthetic plan to obtain these compounds wasbased on the preparation of the appropriate b-phenylacetamides(1e3) by standard methods [6,35] and further cyclization by Bis-chlereNapieralski cyclodehydration (Scheme 1). First, 3,4-dimethoxyphenethylamine was subjected to Ac2O/Pyr or Schot-teneBauman conditions to give the appropriate amides (1e3).Then, amides 1e3 were treated with POCl3 to lead to the expecteddihydroisoquinolines 1ae3a by BischlereNapieralski cyclizationwith good yields.

Once the different imines were prepared following Pemberton’sprocedure [34], imine 1a was reacted with the commerciallyavailable 2,2,6-trimethyl-4H-1,3-dioxin-4-one (4a) under neutralor basic conditions, either using microwave irradiation or conven-tional heating. Similar to Pemberton et al. [34], the formation ofoxazinone 1c was favored by basic conditions and it was almostexclusively formed by the use of conventional heating. Neutralconditions clearly favored the formation of pyridone 1b undereither conventional or microwave activation (Table 1, Scheme 2).

Once the reaction conditions were set up [38], this methodologywas applied to the synthesis of oxazinones (1ce3c, 1e, 1g) andpyridones (1be3b, 1d, 1f) from different 1-substituted dihy-droisoquinolines (1ae3a) and dioxinones (4ae4c). When 1-benzyldihydroisoquinolines (2a, 3a) were used, yields were lowerthan those for 1-methyldihydroisoquinoline (1a) (Schemes 3 and4). In addition, in order to explore their biological activity, depro-tection of the methoxy groups of compound 1b was performed.Catechol 1h was then obtained in good yield using BBr3. To ourknowledge, compounds 1c, 1fe1h, 2b, 2c, 3b and 3c are here re-ported for the first time and their structures were confirmed byNMR spectral data and mass spectrometry.

2.2. Respiratory chain inhibition (SAR)

All the synthesized compoundswere assayed in vitro against theNADH oxidase activity of beef-heart submitochondrial particles, asa model of mammalian respiratory chain. Most of the tested

,1-a]isoquinolin-4-ones and oxazino[2,3-a]isoquinolin-4-ones.

Scheme 1. Synthesis of dihydroisoquinolines 1ae3a. Reagents and Conditions: (a) Ac2O, Pyr, rt, 3 h; (b) POCl3, CH3CN, N2, reflux, 1 h; (c) 2-bromophenylacetyl chloride or 2-phenylacetyl chloride, CH2Cl2, 5% aq NaOH, rt, 16 h.

Table 1Reaction of dihydroisoquinoline 1awith 2,2,6-trimethyl-4H-1,3-dioxin-4-one underdifferent reaction conditions.

Solvent Heating Pyridone (1b) Oxazinone (1c)

Toluene MW 70% -TEA þ Toluene MW Complex mixtureToluene Conventional 78% 1%TEA þ Toluene Conventional 2% 68%

L. Moreno et al. / European Journal of Medicinal Chemistry 69 (2013) 69e76 71

compounds were able to inhibit the whole respiratory chain in themicromolar range. Indeed, this enzymatic assay detected a decay ofthe NADH oxidase activity in the integrated electron transferchain which involves all three energy-conserving enzymatic com-plexes: NADH:ubiquinone oxidoreductase (complex I), ubiq-uinol:cytochrome c reductase (complex III) and cytochrome coxidase (complex IV). NADH is the physiological electron donor ofcomplex I that catalyzes the reduction of lipid soluble endogenouscoenzyme Q or ubiquinone (CoQ10 in beef heart). In fact, in theintegrated electron transfer chain, electrons are first carried from

Table 2Inhibitory potency of compounds 1be3c against the NADH oxidase activity of mammali

Compound IC50 (mM) � SEa

1b 8.23 � 0.98*

1c 1.36 � 0.50*1d >301e 1.32 � 0.03*1f 12.50 � 1.66*1g 0.90 � 0.03*1h 3.44 � 0.05*2b 4.24 � 0.03*2c 1.39 � 0.20*3b 4.69 � 0.11*3c 1.53 � 0.10*Rotenone 5.10 � 0.90* (nM)

*p < 0.05.a Data are presented as the mean � SD of thee independent determinations for each c

complex I or from complex II (succinate:cytochrome c reductase) tocomplex III by ubiquinone, and then from complex III to complex IVby the peripheral membrane protein cytochrome c. Thus, theinhibitory action of the active compounds might be affecting one ormore of the electron transfer chain enzymatic complexes [39].

Compared with the highly toxic rotenone (a high-affinity in-hibitor of complex I), the pyridone (1b, 1f, 1h, 2b and 3b) andoxazinone (1c, 1e, 1g, 2c and 3c) cytotoxic derivatives showed amoderate inhibitor activity that was in the same order of magni-tude that the structural-related mycotoxins such as circumdatins(circumdatin E and H, IC50 ¼ 2.5 and 1.5 mM, respectively) andflavacol (IC50 ¼ 13 mM) [40]. Given that most of the respiratorychain inhibitors with potential therapeutic interest act in the highnanomolar and low micromolar ranges to avoid an extremetoxicity, as it happens with Annonaceus acetogenins (IC50 values inthe low nanomolar range), these compounds and especially theoxazinone derivatives can be consider effective inhibitors of thewhole respiratory chain [41] (Table 2).

In general, oxazino[2,3-a]isoquinolin-4-ones seemed to displaygreater activity than their respective pyrido[2,1-a]isoquinolin-4-

an respiratory chain.

ompound.

Scheme 2. Mechanism of reaction.

Scheme 3. Synthesis of pyrido[2,1-a]isoquinolin-4-ones and oxazino [2,3-a]isoquinolin-4-ones 1be1g. Reagents and Conditions: (a) 2,2,6-trimethyl-4H-1,3-dioxin-4-one 4a,toluene, N2, MW, 150 �C, 5 min; (b) 2,2,6-trimethyl-4H-1,3-dioxin-4-one 4a, Et3N, toluene, N2, reflux, 3 h; (c) 2,2-dimethyl-6-phenyl-4H-1,3-dioxin-4-one 4b, toluene, N2, MW,110 �C, 5 min; (d) 2,2-dimethyl-6-phenyl-4H-1,3-dioxin-4-one 4b, Et3N, toluene, N2, reflux, 3h; (e) 6-(furan-2-yl)-2,2-dimethyl-4H-1,3-dioxin-4-one 4c, toluene, N2, MW, 110 �C,5 min; (f) 6-(furan-2-yl)-2,2-dimethyl-4H-1,3-dioxin-4-one 4c, Et3N, toluene, N2, reflux, 3 h; (g) BBr3, CH2Cl2, N2, rt, 2 h.


Scheme 4. Synthesis of pyrido[2,1-a]isoquinolin-4-ones and oxazino[2,3-a]iso-quinolin-4-ones 2be3c. Reagents and Conditions: (a) 2,2,6-trimethyl-4H-1,3-dioxin-4-one 4a, toluene, N2, reflux, 3 h; (b) 2,2,6-trimethyl-4H-1,3-dioxin-4-one 4a, Et3N,toluene, N2, reflux, 3 h.


ones analogs. For instance, while the IC50 of the compound 1c was1.36 mM, its pyridone analog (1b) showed decreased NADH oxidaseactivity (IC50 ¼ 8.33 mM). In addition, when the methyl group in C2position of the pyridone was replaced by an aromatic group, theactivity decreased or almost disappeared (1d and 1f). By contrast,the size and nature of the substituent in 2-position of the oxazinedid not seem to affect NADH oxidase activity. In this regard, not onlythe existence of a phenyl group (1e) kept the activity but thepresence of a furan ring in C2 position provided the most potentcompound 1g. Interestingly, the deprotection of the methoxygroups in A ring to obtain a catechol (1h) maintained the activity.

On the other hand, when the aromatic ring was placed on C1position (series 2 and 3), the pyrido[2,1-a]isoquinolin-4-ones wereable to inhibit respiratory chain at moderated potencies (com-pounds 2b and 3b with IC50 of 4.24 mM and 4.69 mM respectively)without any noticeable influence of the halogen group. Howeverand in agreement with the previous observations, the oxazine an-alogs displayed greater NADH oxidase inhibition than theirrespective pyridones (compounds 2c and 3c with IC50 of 1.39 mMand 1.53 respectively). Likewise, the presence of a bromine atom inthe aromatic ring did not seem to affect the activity. Therefore,pyridone and mainly oxazine derivatives can interfering with themitochondrial respiratory chain by direct inhibition of one or moreenzyme complexes and/or, similar to many potent inhibitors thatmimic the quinoid head of ubiquinone, they can function as falseelectron acceptors (carriers) by extracting electrons from in-termediates in the respiratory chain in competition with theirnatural substrates [42]. We could hypothesize that the differentelectron density of oxazine and pyridone nucleus might establishtheir different electron acceptor properties whereas substituentson C1 or C2 positions favors the passage through the hydrophobicenvironment of the ubiquinone catalytic binding site.

3. Conclusions

In conclusion, we have synthesized via Pemberton methodeleven benzo[a]quinolizines including pyrido[2,1-a]isoquinolin-4-ones (1b, 1d, 1f, 1h, 2b and 3b) and oxazino[2,3-a]isoquinolin-4-ones (1c, 1e, 1g, 2c and 3c). Most of these compounds were foundto be moderate inhibitors of the electron transport chain in thesame range that known mycotoxins. Oxazino[2,3-a]isoquinolin-4-ones displayed greater activity than their pyrido[2,1-a]iso-quinolin-4-ones analogs. Indeed, the presence of a furan ring in C2

position of oxazino[2,3-a]isoquinolin-4-ones provided the com-pound (1g) with the most potent biological activity. Therefore,these compounds and especially the oxazinone derivatives are inthe tendency of the new less toxic antitumor agents that targetmitochondrial electron transport chain in a middle range potency.

4. Experimental section

4.1. Chemistry

Melting points were taken on a Cambridge microscope instru-ment coupled with a Reichert-Jung. EIMS was recorded in a VGAuto Spec Fisons spectrometer instrument (Fisons, Manchester,United Kingdom). Microwave reactions were performed using aCEM Focused Microwave Synthesis System apparatus, ModelDiscover. The device is a continuous focused microwave powerdelivery system with operator selectable power output from 0 to300 W. All the reactions were performed in special 10 mL glassvessels under argon atmosphere. Reaction mixture temperatureswere measured during microwave heating with an IR surfacesensor located at the base of the Discover. The temperature wasfixed at 150 �C or 110 �C and maintained for 5 min (i.e., hold time:time in which the system maintains the control parameters) andwas usually reached within 1 min (i.e., run time: maximum runtime for the method in situations where the control point is notreached). 1H NMR and 13C NMR spectra were recorded with CDCl3as a solvent in a Bruker AC-300, AC-400 or AC-500. Multiplicities of13C NMR resonances were assigned by DEPT experiments. COSY,HSQC and HMBC correlations were recorded at 400 and 500 MHz(Bruker AC-400 or AC-500). The assignments of all compoundswere made by COSY, DEPT, HSQC and HMBC. All the reactions weremonitored by analytical TLC with silica gel 60 F254 (Merck 5554).Residues were purified by silica gel 60 (40e63 mm, Merck 9385)column chromatography. Solvents and reagents were purchasedfrom commercial sources. Quoted yields are of purified material.

4.1.1. General procedure for the synthesis of dihydroisoquinolines(1ae3a)4.1.1.1. 6,7-Dimethoxy-1-methyl-3,4-dihydroisoquinoline (1a).A solution of 3,4-dimethoxyphenethylamine (1 g, 5.52 mmol) indry Ac2O (10 mL) and pyridine (0.5 mL) was stirred under nitrogenatmosphere at room temperature for 3 h. Then, the reactionmixture was diluted with water (15 mL) and extracted with CH2Cl2.The combined organic phases were dried over Na2SO4 and evapo-rated to dryness to give 1.02 g (81%) of 1 as awhite solid, which wasused in the next step with no further purification. Next, a solutionof N-(3,4-dimethoxyphenethyl)acetamide 1 (500 mg, 2.24 mmol)in dry CH3CN (20 mL) was treated with POCl3 (0.37 mL, 4.1 mmol)and refluxed for 1 h in a nitrogen atmosphere. The reaction mixturewas concentrated to dryness, re-dissolved in water and basifiedwith NH4OH until pH ¼ 9. The mixture was extracted with CH2Cl2.The organic solution was dried over Na2SO4 and evaporated todryness to give 312 mg of 1a (67%) which was used in the next stepwith no further purification. Characterization data for 1 and 1awere in agreement with published data [35].

4.1.1.2. 6,7-Dimethoxy-1-(2-bromobenzyl)-3,4-dihydroisoquinoline(2a). 2-Bromophenylacetyl chloride (1.2 mL, 8.25 mmol) was addeddropwise at 0 �C to a solution of 3,4-dimethoxyphenetylamine (2 g,5.50 mmol) in 15 mL of CH2Cl2 and 5% aqueous NaOH (2.5 mL). Thereaction was stirred at room temperature overnight and thenextracted with CH2Cl2 (3 � 10 mL). The combination of the organicphaseswaswashedwithbrine (2�10mL) andH2O (2�10mL), driedover Na2SO4 and evaporated to dryness. The residue was purified bysilica gel column chromatography (hexane-EtOAc, 5:5) to afford 1.8 g


of amide 2 as a white powder (4.73 mmol, 86%). Next, a solution of 2(100 mg, 0.27 mmol) in dry CH3CN (5 mL) was treated with POCl3(0.07 mL, 0.75 mmol) and refluxed for 1 h in a nitrogen atmosphereunder the same conditions described above to obtain 1a. The residuewas used in the next step with no further purification. Characteriza-tion data for 2 and 2awere in agreement with published data [36].

4.1.1.3. 6,7-Dimethoxy-1-benzyl-3,4-dihydroisoquinoline (3a).The compound was 3 prepared according to the proceduredescribed above to obtain 2a. Firstly, 2-phenylacetyl chloride(2.18 mL, 16.5 mmol), 3,4-dimethoxyphenetylamine (2 g, 11 mmol)and NaOH 5% (2.5mL). The residuewas purified by silica gel columnchromatography (hexane/EtOAc 5:5) to afford 1.9 g of amide 3 as awhite powder (6.27 mmol, 57%). Next, a solution of 3 (200 mg,0.65 mmol) in dry CH3CN (15 mL) was treated with POCl3 (0.12 mL,1.35mmol) and refluxed for 1 h in a nitrogen atmosphere under thesame conditions described above to obtain 1a. The residue wasused in the next step with no further purification. Characterizationdata for 3 and 3a were in agreement with published data [37].

4.1.2. General procedure for the synthesis of pyrido[2,1-a]isoquinolin-4-ones (1b, 1d, 1f)4.1.2.1. 9,10-Dimethoxy-2-methyl-6,7-dihydro-4H-pyrido[2,1-a]iso-quinolin-4-one (1b). 2,2,6-Trimethyl-4H-1,3-dioxin-4-one (0.065mL,0.48 mmol) was added to a stirred solution of imine 1a (50 mg,0.24 mmol) dissolved in dry toluene (0.5 mL). The mixture wasirradiated at 150 �C, 300 W, for 5 min and then cooled to roomtemperature. The resulting solution was diluted with CH2Cl2 andwashed with saturated aq. NaHCO3, water and brine. The combinedorganic extracts were dried with Na2SO4, filtered and concentrated.Purification by column chromatography (CH2Cl2eMeOH, 98:2) gave1b (42 mg, 70%) as a white powder. Characterization data were inagreement with published data [33,31].

4.1.2.2. 9,10-Dimethoxy-2-phenyl-6,7-dihydro-4H-pyrido[2,1-a]iso-quinolin-4-one (1d). This compound was prepared following thesame procedure described above for the synthesis of 1b, and using1a (0.24 mmol) and 2,2-dimethyl-6-phenyl-4H-1,3-dioxin-4-one(89 mg, 0.46 mmol). The mixture was irradiated at 110 �C, 300 W,for 5 min and then cooled to room temperature. Purification bycolumn chromatography (CH2Cl2-AcOEt, 60:40) gave 1d (35 mg,60%) as a yellow powder. Characterization in agreement withpublished data [33,31].

4.1.2.3. 2-(Furan-2-yl)-9,10-dimethoxy-6,7-dihydro-4H-pyrido[2,1-a]isoquinolin-4-one (1f). This compound was prepared followingthe same procedure described above for the synthesis of 1b, andusing 1a (0.24 mmol) and 6-(furan-2-yl)-2,2-dimethyl-4H-1,3-dioxin-4-one (93 mg, 0.48 mmol). The mixture was irradiated at110 �C, 300 W, for 5 min and then cooled to room temperature.Purification by column chromatography (CH2Cl2-AcOEt, 70:30)gave 1f (23 mg, 31%) as a yellow oil. IR (cm�1): 2932, 1648, 1566,1504, 1459, 1341; 1H NMR (500 MHz, CDCl3): d ¼ 7.57 (d, J ¼ 1.3 Hz,1H, H-30), 7.23 (s, 1H, H-11), 6.87 (d, J ¼ 3.3 Hz, 1H, H-50), 6.85 (d,J ¼ 1.7 Hz, 1H, H-1), 6.79 (d, J ¼ 1.7 Hz, 1H, H-3), 6.75 (s, 1H, H-8),6.54 (dd, J ¼ 3.5, 1.8 Hz, 1H, H-40), 4.27 (m, 2H, CH2-6), 3.98 (OCH3-9), 3.94 (OCH3-10), 2.92 (m, 2H, CH2-7); 13C NMR (125MHz, CDCl3):d ¼ 162.7 (CO), 151.1 (C-9), 151.0 (C-10), 148.5 (C-2), 144.0 (CH-30),143.5 (C-11b), 139.6 (C-10), 129.1 (C-7a), 121.5 (C-11a), 112.1 (CH-40),110.4 (CH-8), 109.9 (CH-3), 109.4 (CH-50), 108.2 (CH-11), 98.0 (CH-1), 56.3 (OCH3-9), 56.0 (OCH3-10), 39.1 (CH2-6), 27.6 (CH2-7); ESMSm/z (%) 324 [M þ H]þ (100).

4.1.2.4. 9,10-Dihydroxy-2-methyl-6,7-dihydro-4H-pyrido[2,1-a]iso-quinolin-4-one (1h). A solution of 1b (75 mg, 0.33 mmol) in dry

CH2Cl2 (5 mL) was cooled to �78 �C. Then, BBr3 (0.05 mL,0.50 mmol) was added dropwise under nitrogen atmosphere. After15 min, the reaction mixture was warmed up to room temperatureand stirred for 2 h. The reaction was terminated by the addition ofMeOH (1 mL) dropwise and the mixture was further stirred foranother 30 min. The solvent was concentrated to dryness. Theresiduewas purified by silica gel column chromatography (CH2Cl2eMeOH, 90:10) to afford 70 mg of 1h (87%) as a brown solid. Mp:281e283 �C; IR (cm�1): 3328, 2971, 2734, 2589, 1646, 1509, 1298;1H NMR (500 MHz, CDCl3): d ¼ 7.59 (s, 1H, H-11), 6.98 (s, 1H, H-8),6.46 (d, J ¼ 1.1 Hz, 1H, H-3), 6.37 (d, J ¼ 1.1 Hz, 1H, H-1), 4.19 (m, 2H,CH2-6), 2.66 (m, 2H, CH2-7), 2.03 (s, 3H, CH3); 13C NMR (125 MHz,CDCl3): d ¼ 165.2 (CO), 153.3 (C-9), 151.9 (C-10), 148.6 (C-2), 145.8(C-11b), 130.5 (C-7a), 120.9 (C-11a), 117.6 (CH-1), 117.4 (CH-8), 115.6(CH-11), 106.6 (CH-3), 41.9 (CH2-6), 29.5 (CH2-7), 23.3 (CH3); ESMSm/z (%) 234 (100) [M]þ

4.1.3. General procedure for the synthesis of oxazino[2,3-a]isoquinolin-4-ones (1c, 1e, 1g)4.1.3.1. 9,10-Dimethoxy-2,11b-dimethyl-7,11b-dihydro-4H,6H- [1,3]oxazino[2,3-a]isoquinolin-4-one (1c).2,2,6-Trimethyl-4H-1,3-dioxin-4-one (0.06 mL, 0.44 mmol) wasadded to a stirred solution of imine 1a (46 mg, 0.22 mmol) dis-solved in dry toluene (5 mL). Et3N (0.06 mL, 0.44 mmol) was addedand the heated mixture was refluxed for 3 h and then cooled toroom temperature. The resulting solution was diluted with CH2Cl2and washed with saturated aq. NaHCO3, water and brine. Thecombined organic extracts were dried with Na2SO4, filtered andconcentrated. Purification by column chromatography (CH2Cl2-AcOEt, 60:40) gave 1c (43 mg, 68%). IR (cm�1): 2932, 1662, 1515,1409; 1H NMR (500 MHz, CDCl3): d ¼ 6.99 (s, 1H, H-8), 6.96 (s, 1H,H-11), 5.26 (s, 1H, H-3), 4.62 (ddd, J ¼ 12.7, 5.1, 2.7 Hz; 1H, H-6a),3.91 (s, 3H, OCH3-9), 3.66 (s, 3H, OCH3-10), 3.01 (m, 1H, H-6b), 2.91(m, 1H, H-7a), 2.66 (m, 1H, H-7b), 1.97 (CH3-2), 1.81 (CH3-11b); 13CNMR (125 MHz, CDCl3): d ¼ 163.4 (CO), 162.1 (C-2), 149.2 (C-10),148.8 (C-9), 128.5 (C-11a), 126.9 (C-7a), 110.7 (CH-8), 108.8 (CH-11),98.4 (CH-3), 90.3 (C-11b), 56.1 (OCH3-10), 55.8 (OCH3-9), 35.9 (CH2-6), 27.6 (CH2-7), 23.1 (CH3-11b), 19.8 (CH3-2); ESMS m/z (%) 290[M þ H]þ (100)

4.1.3.2. 9,10-Dimethoxy-11b-methyl-2-phenyl-7,11b-dihydro-4H,6H-[1,3]oxazino[2,3-a]isoquinolin-4-one (1e). This compound was pre-pared following the same procedure described above for the syn-thesis of 1c, and using 1a (75 mg, 0.36 mmol) and 2,2-dimethyl-6-phenyl-4H-1,3-dioxin-4-one (178 mg, 0.88 mmol), Et3N (0.12 mL,0.88 mmol), and dry toluene (3 mL). Purification by column chro-matography (CH2Cl2-AcOEt, 90:10) gave 1e (40 mg, 50%). Charac-terization data were in agreement with published data [33].

4.1.3.3. 2-(Furan-2-yl)-9,10-dimethoxy-11b-methyl-7,11b-dihydro-4H,6H- [1,3]oxazino[2,3-a]isoquinolin-4-one (1g). This compoundwas prepared following the same procedure described above forthe synthesis of 1c, and using 1a (50 mg, 0.24 mmol) and 6-(furan-2-yl)-2,2-dimethyl-4H-1,3-dioxin-4-one (126 mg, 0.66 mmol), Et3N(0.06 mL, 0.33 mmol), and dry toluene (2 mL). Purification by col-umn chromatography (CHCl2-AcOEt, 90:10) gave 1g (20mg, 26%) asawhite powder. Mp: 123e125 �C; IR (cm�1): 2943,1653, 1515,1414,1364; 1H NMR (500 MHz, CDCl3): d ¼ 7.53 (d, J ¼ 1.1 Hz, 1H, H-30),7.02 (s, 1H, H-11), 6.80 (d, J¼ 3.3 Hz,1H, H-50), 6.62 (s, 1H, H-8), 6.50(dd, J ¼ 3.4, 1.8 Hz, 1H, H-40), 5.89 (s, 1H, H-3), 4.67 (ddd, J ¼ 13.0,5.3, 2.8 Hz, 1H, H-6a), 3.94 (OCH3-10), 3.87 (OCH3-9), 3.09 (m, 1H,H-6b), 2.97 (m, 1H, H-7a), 2.70 (dt, J ¼ 13.5, 3.3 Hz, 1H, H-7b), 1.89(CH3-11b); 13C NMR (125 MHz, CDCl3): d ¼ 162.2 (CO), 152.3 (C-2),149.4 (C-9), 148.1 (C-10), 147.1 (C-10), 145.1 (CH-30), 128.4 (C-11a),127.1 (C-7a), 111.8 (CH-40, CH-50), 110.7 (CH-8), 108.9 (CH-11), 94.9


(CH-3), 90.7 (C-11b), 56.1 (OCH3-10), 55.9 (OCH3-9), 36.2 (CH2-6),27.6 (CH2-7), 23.0 (CH3-11b); ESMS m/z (%) 341 [M]þ (100).

4.1.4. General procedure for the synthesis of pyrido[2,1-a]isoquinolin-4-ones 2be3b4.1.4.1. 1-(2-Bromophenyl)-9,10-dimethoxy-2-methyl-6,7-dihydro-4H-pyrido[2,1-a]isoquinolin-4-one (2b). A residue of the imine 2a(from 100 mg, 0.27 mmol of amide 2) was dissolved in dry toluene(5 mL), and treated with 2,2,6-trimethyl-4H-1,3-dioxin-4-one(0.06 mL, 0.44 mmol). The reaction mixture was refluxed under ni-trogen atmosphere for 3 h and then cooled to room temperature. Theresulting solution was diluted with CH2Cl2 and washed with satu-rated aq. NaHCO3, water and brine. The combined organic extractswere dried with Na2SO4, filtered and concentrated. Purification bycolumn chromatography (CH2Cl2-AcOEt, 10:90) gave 2b (40 mg, 33%from amide 2) as a white solid. Mp: 216e218 �C. IR: 2961, 1639,1491,1466, 1450, 1212; 1H NMR (500 MHz, CDCl3): d ¼ 7.65 (dd, 1H,J¼ 8.02, 1.12 Hz, CH-20), 7.31 (td,1H, J¼ 7.49, 1.19 Hz, CH-60), 7.20 (td,1H, J¼ 7.77,1.71 Hz, CH-40), 7.14 (dd,1H, J¼ 7.55,1.65 Hz, CH-40), 6.66(s, 1H, H-8), 6.52 (s, 1H, H-3), 6.43 (s, 1H, H-11), 4.45 (m, 1H, H-6a),3.97 (m, 1H, H-6b), 3.86 (OCH3-9), 3.23 (OCH3-10), 2.85 (m, 2H, CH2-7), 1.88 (s, 3H, CH3); 13C NMR (125MHz, CDCl3): d¼ 161.7 (CO), 150.2(C-9),149.5 (C-10),146.5 (C-2),140.4 (C-1),140.1 (C-10), 133.2 (CH-30),132.7 (CH-50), 131.6 (C-11b),129.2 (CH-40), 128.2 (CH-60), 126.5 (C-20),121.4 (C-11a), 118.4 (C-7a), 117.7 (CH-3), 112.4 (CH-11), 109.6 (CH-8),55.7 (OCH3-9), 55.3 (OCH3-10), 40.3 (CH2-6), 28.6 (CH2-7), 21.0(CH3); ESMS m/z (%) 425 [M þ H]þ (100).

4.1.4.2. 9,10-Dimethoxy-2-methyl-1-phenyl-6,7-dihydro-4H-pyrido[2,1-a] isoquinolin-4-one (3b). This compound was preparedfollowing the same procedure described above for the synthesis of2b, and using a residue of the imine 3a (from 200 mg, 0.65 mmol ofamide 3) and treated with 2,2,6-trimethyl-4H-1,3-dioxin-4-one(0.18 mL, 1.30 mmol) in dry toluene (10 mL). Purification by col-umn chromatography (CH2Cl2eMeOH, 95:5) gave 3b (75 mg, 33%from amide 3). IR (cm�1): 2933, 1690, 1517, 1373, 1262, 1212; 1HNMR (500 MHz, CDCl3): d ¼ 7.45 (m, 2H, CH-30, CH-50), 7.30 (m, 1H,CH-40), 7.29 (m, 2H, CH-20, CH-60), 6.96 (s, 1H, H-8), 6.89 (s, 1H, H-11), 6.16 (s, 1H, H-3), 4.63 (m, 1H, H-6a), 3.98 (m, 1H, H-6b), 3.83(OCH3-9), 3.25 (OCH3-10), 2.92 (m, 2H, CH2-7), 1.93 (s, 3H, CH3); 13CNMR (125 MHz, CDCl3): d ¼ 161.0 (CO), 148.9 (C-9), 148.1 (C-2),146.5 (C-10), 140.2 (C-1), 135.6 (C-11b), 132.5 (C-10), 128.9 (CH-20,CH-60), 128.7 (C-7a), 128.6 (CH-30, CH-50), 127.9 (CH-40), 123.5 (C-11a), 118.7 (CH-3), 111.4 (CH-8), 109.7 (CH-11), 55.8 (OCH3-9), 55.4(OCH3-10), 41.3 (CH2-6), 27.2 (CH2-7), 19.0 (CH3); ESMS m/z (%) 347(100) [M]þ

4.1.5. General procedure for the synthesis of oxazino[2,3-a]isoquinolin-4-ones (2ce3c)4.1.5.1. 11B-(2-Bromobenzyl)-9,10-dimethoxy-2-methyl-7,11b-dihy-dro-4H,6H- [1,3]oxazino[2,3-a]isoquinolin-4-one (2c). A residue ofthe imine 2a (from 100 mg, 0.27 mmol of amide 2) in dry toluene(5 mL) was treated with 2,2,6-trimethyl-4H-1,3-dioxin-4-one(0.06 mL, 0.44 mmol) and Et3N (0.06 mL, 0.44 mmol). The reac-tion mixture was refluxed under nitrogen atmosphere for 3 h andthen cooled to room temperature. The resulting solution wasdiluted with CH2Cl2 and washed with saturated aq. NaHCO3, waterand brine. The combined organic extracts were dried with Na2SO4,filtered and concentrated. Purification by column chromatography(Toluene-AcOEt, 70:30) gave 2c (15 mg, 12% from amide 2). IR(cm�1): 2938, 1664, 1515, 1460, 1264, 1212; 1H NMR (500 MHz,CDCl3): d ¼ 7.44 (m, 1H, C-20), 7.16 (m, 1H, CH-50), 7.07 (m, 1H, CH-40), 6.94 (m, 1H, CH-60), 6.56 (s, 1H, H-11), 6.55 (s, 1H, H-8), 5.29 (s,1H, H-3), 4.33 (m, 1H, H-6a), 3.84 (OCH3-9), 3.80 (m, 1H, CH2Ph-a),3.66 (OCH3-10), 3.65 (m, 1H, CH2Ph-b), 3.18 (m, 1H, H-6b), 2.81 (m,

1H, H-7a), 2.50 (m, 1H, H-7b), 2.05 (s, 3H, CH3); 13C NMR (125 MHz,CDCl3): d ¼ 162.7 (CO), 161.9 (C-2), 149.3 (C-9), 147.3 (C-10), 134.6(C-10), 132.8 (CH-30), 132.5 (CH-60), 128.6 (CH-40), 127.5 (C-20), 126.9(CH-50),126.8 (C-7a),126.4 (C-11a),110.3 (CH-8),109.6 (CH-11), 98.8(CH-3), 91.7 (C-11b), 55.8 (OCH3-9), 55.8 (OCH3-10), 41.8 (CH2Ph),36.7 (CH2-6), 27.1 (CH2-7), 19.8 (CH3); ESMS m/z (%) 444 (100)[Mþ1]þ

4.1.5.2. 11B-Benzyl-9,10-dimethoxy-2-methyl-7,11b-dihydro-4H,6H-[1,3]oxazino[2,3-a]isoquinolin-4-one (3c). This compound was pre-pared following the same procedure described above for the syn-thesis of 2c, and using a residue of the imine 3a (from 200 mg,0.65 mmol of amide 3) in dry toluene (10 mL), and treated with2,2,6-trimethyl-4H-1,3-dioxin-4-one (0.18mL,1.30mmol) and Et3N(0.18 mL). The reaction mixture was refluxed under nitrogen at-mosphere for 2.5 h and then cooled to room temperature. Purifi-cation by column chromatography (CH2Cl2eMeOH, 95:5) gave 3c(80 mg, 33% from amide 3). IR (cm�1): 2933, 1698, 1515, 1373, 1324,1212; 1H NMR (500 MHz, CDCl3): d ¼ 7.18 (m, 3H, CH-30, CH-40, CH-50), 7.83 (m, 2H, CH-20, CH-60), 6.70 (s, 1H, H-11), 6.54 (s, 1H, H-8),5.36 (s, 1H, H-3), 4.16 (m, 1H, H-6a), 3.86 (OCH3-9), 3.75 (OCH3-10),3.53 (m, 2H, CH2Ph), 2.87 (m,1H, H-6b), 2.72 (m,1H, H-7a), 2.35 (m,1H, H-7b), 2.04 (s, 3H, CH3); 13C NMR (125 MHz, CDCl3): d ¼ 163.1(CO), 162.2 (C-2), 149.2 (C-9), 147.4 (C-10), 134.6 (C-10), 130.9 (CH-20,CH-60), 128.0 (CH-30, CH-50), 127.8 (C-7a), 127.0 (CH-40), 126.3 (C-11a), 110.4 (CH-11), 109.7 (CH-8), 99.1 (CH-3), 91.8 (C-11b), 55.9(OCH3-9), 55.8 (OCH3-10), 42.7 (CH2Ph), 37.2 (CH2-6), 27.2 (CH2-7),19.8 (CH3); ESMS m/z (%) 366 (100) [M þ H]þ.

4.2. Biological assays

NADH and other biochemical reagents were purchased fromSigmaeAldrich Chemical Co. Stock solutions (30 mM in absoluteEtOH) of target compounds were prepared and kept in the darkat�20 �C. Appropriate dilutions (10e30mM)weremade before theexperiments. Inverted submitochondrial particles (SMP) from beefheart were obtained following Fato’s method [39] by extensiveultrasonic disruption of frozen-thawed mitochondria to produceopen membrane fragments where permeability barriers to sub-strates were lost, and they were stored at �70 �C at 28 mg/mL(protein measured by the Bradford method). The beef heart SMPwere transferred to glass test tubes, diluted to 0.5 mg/mL in250 mM sucrose and 10 mM TriseHCl buffer, pH 7.4, and treatedwith 300 mM NADH to activate complex I before starting the ex-periments. Aliquots of the stocks solutions (1 mL) were addedsuccessively to 500 mL of the SMP preparations with 5 min of in-cubation on ice after each addition (ethanol never exceeded 2% ofthe total volume). After incubation, aliquots of the treated SMP(25 mL) were diluted to 6 mg/mL in 50 mM potassium phosphatebuffer (pH 7.4) and 1 mM EDTA, in a cuvette at 22 �C, always in thepresence of 75 mM NADH. Immediately, NADH oxidase activity wasmeasured as the aerobic oxidation of NADH. Reaction rates werecalculated for each inhibitor (at increasing concentrations) from thelinear decrease of NADH concentration (l¼ 340 nm, 3¼ 6.22 mM�1

cm�1) measured in an end-window photomultiplier spectropho-tometer ATI-Unicam UV4-500. The inhibitory concentration (IC50)was taken as the final compound concentration in the assay cuvettethat yielded 50% inhibition of the NADH oxidase activity. Data fromindividual titrations were used to assess the means and standarddeviations of three independent assays for each compound.

Acknowledgments

This study was supported by grants SAF2011-23777, SpanishMinistry of Economy and Competitiveness, RIER RD08/0075/0016,


Carlos III Health Institute, Spanish Ministry of Health and the Eu-ropean Regional Development Fund (FEDER). A. Galán was therecipient of a fellowship from FPU program of Spanish Ministry ofEducation, Culture and Sport.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejmech.2013.08.013.

References

[1] K.W. Bentley, b-Phenylethylamines and the isoquinoline alkaloids, Nat. Prod.Rep. 22 (2005) 249e268.

[2] S.S. Kulkarni, S. Singh, J.R. Shah, W.-K. Low, T.T. Talele, Synthesis and SARoptimization of quinazolin-4(3H)-ones as poly(ADP-ribose)polymerase-1 in-hibitors, Eur. J. Med. Chem. 50 (2012) 264e273.

[3] O.P.J. van Lindena, C. Farenc, W.H. Zoutmanc, L. Hameetmanc, M. Wijtmans,R. Leurs, C.P. Tensenc, G. Siegal, I.J.P. de Escha, Fragment based lead discoveryof small molecule inhibitors for the EPHA4 receptor tyrosine kinase, Eur. J.Med. Chem. 47 (2012) 493e500.

[4] R.J. Juszczak, J.H. Russell, Inhibition of cytotoxic T lymphocyte-mediated lysisand cellular proliferation by isoquinoline sulfonamide protein kinase in-hibitors. Evidence for the involvement of protein kinase C in lymphocytefunction, J. Biol. Chem. 264 (1989) 810e815.

[5] A. Bermejo, P. Protais, M.A. Blázquez, K.S. Rao, M.C. Zafra-Polo, D. Cortes,Dopaminergic isoquinoline alkaloids from roots of Xylopia papuana, Nat. Prod.Lett. 6 (1995) 57e62.

[6] I. Berenguer, N. El Aouad, S. Andujar, V. Romero, F. Suvire, T. Freret,A. Bermejo, M.D. Ivorra, R.D. Enriz, M. Boulouard, N. Cabedo, D. Cortes, Tet-rahydroisoquinolines as dopaminergic ligands: 1-butyl-7-chloro-6-hydroxy-tetrahydroisoquinoline, a new compound with antidepressant-like activity inmice, Bioorg. Med. Chem. 17 (2009) 4968e4975.

[7] N. El Aouad, I. Berenguer, V. Romero, P. Marín, A. Serrano, S. Andujar, F. Suvire,A. Bermejo, M.D. Ivorra, R.D. Enriz, N. Cabedo, D. Cortes, Structure-activityrelationship of dopaminergic halogenated 1-benzyl-tetrahydroisoquinolinederivatives, Eur. J. Med. Chem. 44 (2009) 4616e4621.

[8] N. Cabedo, I. Berenguer, B. Figadère, D. Cortes, An overview on benzyliso-quinoline derivatives with dopaminergic and serotonergic activities, Curr.Med. Chem. 16 (2009) 2441e2467.

[9] S. Andujar, F. Suvire, I. Berenguer, N. Cabedo, P. Marín, L. Moreno, M.D. Ivorra,D. Cortes, R.D. Enriz, Tetrahydroisoquinolines acting as dopaminergic ligands.A molecular modeling study using MD simulations and QM calculations,J. Mol. Model. 18 (2012) 419e431.

[10] A. Bermejo, B. Figadere, M.C. Zafra-Polo, I. Barrachina, E. Estornell, D. Cortes,Acetogenins from annonaceae. Recent progress in isolation, synthesis, andmechanisms of action, Nat. Prod. Rep. 22 (2005) 269e303.

[11] S. Granell, I. Andreu, D. Marti, A. Cave, R. Aragon, E. Estornell, D. Cortes,M.C. Zafra-Polo, Bisbenzyltetrahydroisoquinolines, a new class of inhibitorsof the mitochondrial respiratory chain complex I, Planta Med. 70 (2004)266e268.

[12] I. Andreu, N. Cabedo, J.R. Tormo, A. Bermejo, R. Mello, D. Cortes, Synthesis ofN-diisopropyl phosphoryl benzyltetrahydroisoquinoline, a new class ofmitochondrial complexes I and III inhibitors, Bioorg. Med. Chem. Lett. 10(2000) 1491e1494.

[13] D. Trachootham, W. Lu, M.A. Ogasawara, N. Rivera-del Valle, P. Huang, Redoxregulation of cell survival, Antioxid. Redox Signal. 10 (2008) 1343e1374.

[14] T.M. Dawson, V.L. Dawson, Molecular pathways of neurodegeneration inParkinson’s disease, Science 302 (2003) 819e822.

[15] R.S. Balaban, S. Nemoto, T. Finkel, Mitochondria, oxidants, and aging, Cell 120(2005) 483e495.

[16] H. Pelicano, L. Feng, Y. Zhou, J.S. Carew, E.O. Hileman, W. Plunkett,M.J. Keating, P. Huang, Inhibition of mitochondrial respiration. A novelstrategy to enhance drug-induced apoptosis in human leukemia cells by areactive oxygen species-mediated mechanism, J. Biol. Chem. 278 (2003)37832e37839.

[17] A. Szewczyk, L. Wojtczak, Mitochondria as a pharmacological target, Phar-macol. Rev. 54 (2002) 101e127.

[18] V. Gogvadze, S. Orrenius, B. Zhivotovsky, Mitochondria in cancer cells, what isso special about them? Trends Cell Biol. 18 (2008) 165e173.

[19] F. Schuler, J.E. Casida, The insecticide target in the PSST subunit of complex I,Pest. Manag. Sci. 57 (2001) 932e940.

[20] R. Baradaran, J.M. Berrisford, G.S. Minhas, L.A. Sazanov, Crystal structure of theentire respiratory complex I, Nature 494 (2013) 443e448.

[21] G. Memetzidis, J. Stambach, L. Jung, C. Schott, C. Heitz, J.C. Stocklet, Structure-affinity relationships of berbines or 5,6,13,13a-tetrahydro-8H-dibenzo[a,g]quinolizines at a-adrenoceptors, Eur. J. Med. Chem. 26 (1991) 605e611.

[22] C. Szantay, L. Toke, P. Kolonits, Synthesis of protoemetine. A new total syn-thesis of emetine, J. Org. Chem. 31 (1966) 1447e1451.

[23] N. Kogure, C. Nishiya, M. Kitajima, H. Takayama, Six new indole alkaloids fromGelsemium sempervirens Ait. f. Tetrahedron Lett. 46 (2005) 5857e5861.

[24] Y. Kanintronkul, R. Worayuthakarn, N. Thasana, P. Winayanuwattikun,K. Pattanapanyasat, R. Surarit, S. Ruchirawat, J. Svasti, Overcoming multidrugresistance in human lung cancer with novel benzo[a]quinolizin-4-ones,Anticancer Res. 31 (2011) 921e927.

[25] W. Du, Towards new anticancer drugs: a decade of advances in synthesis ofcamptothecins and related alkaloids, Tetrahedron 59 (2003) 8649e8687.

[26] S.U. Saraf, The chemistry of the benzo(a)- and benzo(c)quinolizinium ions,Heterocycles 16 (1981) 803e841.

[27] S. Kirschbaum, H. Waldeman, Construction of the tricyclic benzoquinolizinering system by combination of a tandem Mannich-Michael reaction with aHeck reaction, Tetrahedron Lett. 38 (1997) 2829e2832.

[28] S. Kirschbaum, H. Waldeman, Three-step access to the tricyclic benzo[a]qui-nolizine ring system, J. Org. Chem. 63 (1998) 4936e4946.

[29] M. Strandtmann, M.P. Cohen, J. Shravel, A new general synthesis of benzo[a]quinolizines, dibenzo[a, f]quinolizines, and related compounds, J. Org. Chem.31 (1966) 797e802.

[30] M. Rubiralta, A. Diez, A. Balet, J. Bosch, New synthesis of benzo[a]quinolizidin-2-ones via protected 2-aryl-4-piperidones, Tetrahedron 43 (1987) 3021e3030.

[31] A. Roy, S. Nandi, H. Ila, H. Junjappa, An expedient route to 2,3-substituted andfused benzo[a]quinolizine-4-thione framework via ring annulation with b-oxodithioesters, J. Org. Lett. 3 (2001) 229e232.

[32] C.J. Stearman, M. Wilson, A. Padwa, Conjugate addition-dipolar cycloadditioncascade for the synthesis of benzo[a]quinolizine and indolo[a]quinolizinescaffolds: application to the total synthesis of (�)-yohimbenone, J. Org. Chem.74 (2009) 3491e3499.

[33] M. Presset, Y. Coquerel, J. Rodriguez, Microwave-assisted domino and multi-component reactions with cyclic acylketenes: expeditious syntheses of oxa-zinones and oxazindiones, Org. Lett. 11 (2009) 5706e5709.

[34] N. Pemberton, L. Jakobsson, F. Almqvist, Synthesis of multi ring-fused 2-pyridones via an acyl-ketene imine cyclocondensation, Org. Lett. 8 (2006)935e938.

[35] M.R. Pitts, J.R. Harrison, C.J. Moody, Indium metal as a reducing agent inorganic synthesis, J. Chem. Soc. Perkin Trans. 1 (2001) 955e977.

[36] M. Lafrance, N. Blaquière, K. Fagnou, Direct intramolecular arylation ofunactivated arenes: application to the synthesis of aporphine alkaloids, Chem.Commun. 24 (2004) 2874e2875.

[37] M. Ariza, A. Díaz, R. Suau, M. Valpuesta, Synthesis of new dopamine D1antagonist SCH 23390 analogues by the stereoselective Stevens rearrange-ment, Eur. J. Org. Chem. (2011) 6507e6518.

[38] P.A. Peixoto, A. Boulangé, S. Leleu, X. Franck, Versatile synthesis of acylfur-anones by reaction of acylketenes with a-hydroxy ketones: application to theone-step multicomponent synthesis of cadiolide B and its analogues, Eur. J.Org. Chem. (2013) 3316e3327.

[39] R. Fato, E. Estornell, S. Bernando, F. Palloti, G. Parenti-Castelli, G. Lenaz,Steady-state kinetics of the reduction of coenzyme Q analogs by complex I(NADH:ubiquinone oxidoreductase) in bovine heart mitochondria and sub-mitochondrial particles, Biochemistry 35 (1996) 2705e2716.

[40] M.P. López-Gresa, M.C. González, J. Primo, P. Moya, V. Romero, E. Estornell,Circumdatin H, a new inhibitor of mitochondrial NADH oxidase from Asper-gillus ochraceous, J. Antibiot. 58 (2005) 416e419.

[41] A. Bermejo, J.R. Tormo, N. Cabedo, E. Estornell, B. Figadère, D. Cortes, Enan-tiospecific semisynthesis of (þ)-almuheptolide-A, a novel natural heptolideinhibitor of the mammalian mitochondrial respiratory chain, J. Med. Chem. 41(1998) 5158e5166.

[42] M. Degli Esposti, A. Ngo, A. Ghelli, B. Benelli, V. Carelli, H. McLennan,A.W. Linnane, The interaction of Q analogs, particularly hydroxydecyl ben-zoquinone (idebenone), with the respiratory complexes of heart mitochon-dria, Arch. Biochem. Biophys. 330 (1996) 395e400.



http://refhub.elsevier.com/S0223-5234(13)00519-9/sref1











































































































































































Capítulo IV:

Alcaloides aporfínicos y fenantrénicos

dopaminérgicos. Modelización molecular

de IQs 1-sustituidas y dopamina

Artículo 4: “3,4-Dihydroxy- and 3,4-methylenedioxy

phenanthrene-type alkaloids with high selectivity for D2

dopamine receptor” (En: Bioorganic and Medicinal

Chemistry Letters, 2013, 23, 4824)

Artículo 5: “Tetrahydroisoquinolines acting as

dopaminergic ligands. A molecular modeling study using

MD simulations and QM calculations” (En: Journal of

Molecular Modelling, 2012, 18, 419)

Artículo 6: “2-(3chloro-4-hydroxyphenyl)ethylamine


modelling, synthesis and D2 receptor affinity” (Enviado a

European Journal of Medicinal Chemistry)

Bioorganic & Medicinal Chemistry Letters 23 (2013) 4824–4827

Contents lists available at SciVerse ScienceDirect

Bioorganic & Medicinal Chemistry Letters

journal homepage: www.elsevier .com/ locate/bmcl

3,4-Dihydroxy- and 3,4-methylenedioxy- phenanthrene-typealkaloids with high selectivity for D2 dopamine receptor

0960-894X/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.bmcl.2013.06.078

⇑ Corresponding author. Tel.: +34 963544975.E-mail address: [email protected] (D. Cortes).URL: http://www.farmacoquimicavalencia.es (D. Cortes).

Laura Moreno a, Nuria Cabedo b, María Dolores Ivorra a, María-Jesús Sanz c,d, Arturo López Castel e,M. Carmen Álvarez e, Diego Cortes a,⇑a Departamento de Farmacología, Laboratorio de Farmacoquímica, Facultad de Farmacia, Universidad de Valencia, Avda. Vicente Andrés Estellés s/n, Burjassot, 46100 Valencia, Spainb Centro de Ecología Química Agrícola-Instituto Agroforestal Mediterráneo, UPV, Campus de Vera, Edificio 6C, 46022 Valencia, Spainc Departamento de Farmacología, Facultad de Medicina, Universidad de Valencia, 46013 Valencia, Spaind Institute of Health Research-INCLIVA, University Clinic Hospital of Valencia, Valencia, Spaine Valentia BioPharma, Parque Científico de la Universidad de Valencia, Catedrático Jose Beltrán 2, Paterna, 46980 Valencia, Spain


Article history:Received 30 May 2013Revised 24 June 2013Accepted 27 June 2013Available online 4 July 2013

Keywords:AporphinesPhenanthrene alkaloidsDopamine receptorsStructure–activity relationships

a b s t r a c t

Dopamine-mediated neurotransmission plays an important role in relevant psychiatric and neurologicaldisorders. Nowadays, there is an enormous interest in the development of new drugs acting at the dopa-mine receptors (DR) as potential new targets for the treatment of schizophrenia or Parkinson’s disease.Previous studies have revealed that isoquinoline compounds such as tetrahydroisoquinolines (THIQs)can behave as selective D2 dopaminergic alkaloids. In the present study we have synthesized five apor-phine compounds and five phenanthrene alkaloids and evaluated their potential dopaminergic activity.Binding studies on rat striatal membranes were used to evaluate their affinity and selectivity towardsD1 and D2 DR. Phenanthrene type alkaloids, in particular the 3,4-dihydroxy- and 3,4-methylenedioxyderivatives, displayed high selectivity towards D2 DR. Therefore, they are potential candidates to be usedin the treatment of schizophrenia (antagonists) or Parkinson’s disease (agonists) due to their scarce D1

DR-associated side effects.� 2013 Elsevier Ltd. All rights reserved.

Dopamine-mediated neurotransmission plays an important rolein several psychiatric and neurological disorders which affect sev-eral million people worldwide. Researchers have focused on vari-ous approaches towards modulating dopaminergic activity viathe dopamine receptors (DR) as a potential means of treatingschizophrenia and Parkinson’s diseases. The consequences of anactivation or blockade of DR are wide-ranging, and a perturbationof dopamine neurotransmission may result in profound neurolog-ical, psychiatric, or physiological signs and symptoms. For thesereasons, much research has focused on the discovery of noveldopaminergic ligands as potential drug candidates.1 From a thera-peutic point of view, drugs acting at D2-like DR have become veryrelevant since most used antipsychotics in schizophrenia or bipolardisease treatment are D2 antagonists and they are also involved indopamine’s release.2

Isoquinoline alkaloids are a large family of natural productswith a variety of powerful biological activities,3 including seroto-ninergic and dopaminergic.4 Tetrahydroisoquinolines (THIQs), themost numerous naturally occurring alkaloids, include 1-benzyl-THIQs and aporphines, both of which share structural similarities

with dopamine and can interact with DR.5,6 Aporphine alkaloidsconstitute one of the largest groups of isoquinolines,7 and manyof them display pharmacological activities such as antioxidant,antiplatelet, antitumor, anticonvulsant, antineoplastic, cytotoxic,and antiparkinsonian.8 In addition, aporphines’ Hofmann degrada-tion products, phenanthrene alkaloids, have also attractedresearchers’ attention since they can exert varied and powerfulbiological activities including a1-adrenoceptor antagonism,9 inhi-bition of acetylcholinesterase10 and intestinal glucose uptake11,antioxidant activity12,13 or impairment of leukocyte–endothelialcells interactions.14 Moreover, since they have been reported toproduce effects associated with DR stimulation,15 interaction withDR is therefore expected but barely explored.

Our group has long been interested in finding new and potentdopaminergic ligands. Previous observations reported by us havedemonstrated that some natural and synthetic 1-benzyl-THIQs alka-loids can bind to DR.16–20 In the present study, we have carried outthe total synthesis of phenanthrene alkaloids, and take advantageof the diversity generated over the synthetic route of compoundswhich might display dopaminergic activity. Phenanthrene alkaloidsare good candidates to bind to DR with high affinities due totheir structural similarity to dopamine. Furthermore, as opposed to1-benzyl-tetrahydroisoquinolines (BTHIQ) or aporphines wherethe amino function is held into a ring system, the flexible disposition

http://crossmark.dyndns.org/dialog/?doi=10.1016/j.bmcl.2013.06.078&domain=pdf

http://dx.doi.org/10.1016/j.bmcl.2013.06.078


http://www.farmacoquimicavalencia.es

http://dx.doi.org/10.1016/j.bmcl.2013.06.078

http://www.sciencedirect.com/science/journal/0960894X

http://www.elsevier.com/locate/bmcl

H3CO

NH2

3,4-dimethoxy-phenyl-ethylamine

H3CO

H3CO

NHH3CO O

Br

H3CO

NHH3CO

Br

H3CO

NH3CO

Br

Boc

(a) (b)

1(77 %)

(c)

(d)(e)

2b(50%)

2c(97%)

H3CO

NH3CO

Br2a

(77%)

H3CO

NH3CO Boc

3a(84%)

1

2

6

8

11

Scheme 1. Synthesis of aporphine 3a. Reagents and conditions: (a) 2-bromophenylacetyl chloride, CH2Cl2, 5% aq NaOH, rt, 16 h; (b) POCl3, CH3CN, N2, reflux, 1 h; (c) NaBH4,CH3OH, N2, rt, 2 h; (d) di-tert-butyl dicarbonate, CH2Cl2, rt, 2 h; (e) 2-diphenylphosphino-20-(N,N-dimethylamino)biphenyl, Pd(OAc)2, K2CO3, DMA, 130 �C, 4 h.

Scheme 2. Synthesis of phenathrenes 4a–4c. Reagents and conditions: (a) LiAlH4, THF, reflux, 4 h; (b) CH3I, CH3OH, Et2O, rt, 3 h; (c) KOH 3 N, CH3OH, 45 �C, 6 h; (d) BBr3,CH2Cl2, N2, rt, 2 h; (e) CH3I, CH3OH, Et2O, rt, 1 h.

HO

NHO

(100%)3d

N

O

O

N+

NN+ O

O

O

O

O

O

(76%)

(100%)(84%)(99%)

3e

3f4d4e

H3CO

NH3CO

3b

(a) (b)

(c)

(d)(e)

Scheme 3. Synthesis of phenanthrenes 4d, 4e. Reagents and conditions: (a) BBr3,CH2Cl2, N2, rt, 2 h; (b) CH2Cl2, CsF, DMF, reflux, 2 h; (c) CH3I, CH3OH, Et2O, rt, 3 h; (d)KOH 3 N, CH3OH, 45 �C, 6 h; (e) CH3I, CH3OH, Et2O, rt, 1 h.

L. Moreno et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4824–4827 4825

of the aminoethyl chain in phenanthrene alkaloids might reach therequired conformation for an optimal DR interaction.

The general synthetic plan for these compounds consisted inpreparing the appropriate b-phenyl acetamide 1 under Shotten–Baumann conditions to be then cyclized by Bischler–Napieralskicyclodehydration followed by imine reduction.18,19 This sequenceled us to obtain expected 1-(20-bromobenzyl)-tetrahydroisoquino-line 2b. The BTHIQ derivatives with a bromine atom placed overthe phenyl ring (after protection of amino function with a BOCgroup) were then subjected to direct arylation to generate apor-phine 3a (Scheme 1).21 After reduction of the carbamate protectinggroup, aporphine 3b was N-methylated to obtain ammonium salt3c which, under basic Hofmann’s degradation conditions, gavethe corresponding phenanthrene derivative 4a (Scheme 2).22

Thereafter, these phenanthrene alkaloids were again N-methylatedin order to further explore the influence of a quaternary ammo-nium group over dopaminergic activity (4c). The deprotection ofdimethoxy groups from aporphine and phenanthene compoundswas performed using BBr3 to afford catechols 3d and 4b. Methyl-enedioxy derivative 3e was formed by the reaction of 3d withCsF and dichloromethane (Scheme 3).17–19 Using this sequence,several known alkaloids were obtained, aporphines (±)-nuciferine(3b), (±)-roemerine (3e), (±)-roemrefidine (3f), and phenanthrenesatherosperminine (4a) and stephenanthine (4d).8

All the aporphines and phenanthrenes were assayed in vitro fortheir ability to displace selective D1 and D2�DR ligands from theirrespective specific binding sites in the striatal membranes

(Table 1).17–19 Many of these compounds were able to displaceboth 3H-SCH 23390 and 3H-raclopride from their specific bindingsites in rat striatum at micromolar or nanomolar concentrations.

In general, all the tested compounds followed the same ten-dency: (a) The nonquaternary aporphine compounds displayedsimilar potency against D1 receptor (in the low micromolar range)regardless whether the oxygenated functions over the A-ring wereprotected or not. However, the affinity was substantially increased

Table 1Affinity values (Ki, pKi) and selectivity (ratio Ki�D1/Ki�D2) determined in binding experiments to D1 and D2�DR of compounds 3b, 3d, 3e, and 4a, 4b, 4d

Compounds Specific ligand D1 [3H]-SCH 23390 Specific ligand D2 [3H]-raclopride Ki D1/D2

Ki (lM) pKi Ki (lM) pKi

Aporphines3b 0.630 ± 0.149 6.231 ± 0.123f 0.209 ± 0.086 6.761 ± 0.1953d 0.401 ± 0.044 6.401 ± 0.047g 0.045 ± 0.019 7.413 ± 0.174a,d 8.93e 0.377 ± 0.148 6.489 ± 0.167h 0.084 ± 0.012 7.080 ± 0.061b,d 4.4Phenanthrenes4a 2.944 ± 0.304 5.535 ± 0.04 0.323 ± 0.026 6.494 ± 0.037b 9.14b 7.699 ± 0.330 5.208 ± 0.212 0.049 ± 0.018 7.414 ± 0.243c,f 1574d 6.347 ± 0.415 5.383 ± 0.274 0.066 ± 0.009 7.189 ± 0.063c,e 96

Data were displayed as mean ± SEM for 3–5 experiments.ANOVA, post Newmann–Keuls multiple comparison tests.

a p < 0.05.b p < 0.01.c p < 0.001, versus D1-like DR.d p < 0.05 versus 3b.e p < 0.05 versus 4a.f p < 0.01 versus 4a.g p < 0.001 versus 4b.h p < 0.001 versus 4d.

Figure 1. Selectivity of compounds 3d and 4b at the D2-like dopaminergic receptors ([3H]-raclopride binding). Data were displayed as mean ± SEM for 3–5 experiments.

4826 L. Moreno et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4824–4827

towards D2 receptor, with Ki values reaching the nanomolar range,when the oxygenated functions were in the catechol or the meth-ylenedioxy form. The highest affinity of compounds with free phe-nol groups has been previously described in severalisoquinolines.17–20 (b) A similar behavior was observed in the non-quaternary phenanthrene alkaloids. Both, the methylenedioxy andthe catechol groups improved the affinity towards D2�DR (showingpotent affinities within the nanomolar range). In contrast, theirinteraction with the D1�DR was clearly undermined. (c) Regardingthe quaternary ammonium compounds, both aporphines (3c, 3f)and phenanthrenes (4c, 4e) were inactive, probably due to thenitrogen atom in the quaternary form hindering its role as anchor-ing group with Asp86 which is essential for DA binding to D2�DR.23

When both series were compared, they displayed similar D2

affinities, but the conformationally restricted aporphines boundbetter to D1 receptor than phenanthrenes (Table 1, Fig. 1). There-fore, it seems that the disposition of all carbon and hydrogen atomslying in one plane (the phenanthrene ring) was slightly unfavor-able towards D1�DR interaction but not to D2�DR. Indeed, phenan-threnes showed increased selectivity towards D2�DR (compound4b D1/D2 ratio: 157) which may be of relevance from a pharmaco-logical point of view. As previously mentioned, drugs acting atD2-like DR are of special therapeutic potential since while the

antagonists are used in the treatment of schizophrenia (antipsy-chotics), the agonists are employed in the treatment of Parkinson’sdisease.24,25 Of note, other studies have even revealed the potentialantidepressant effect of D2 agonist.3,18 Therefore, the therapeuticpotential of these compounds is noticeable given their selectivityon D2-related activities and their minimal D1-associated collateraleffects.

In summary, we have carried out the total synthesis of fiveaporphine compounds and five phenanthrene alkaloids and evalu-ated their potential dopaminergic activity towards D1 and D2 DR.Our results have demonstrated that phenanthrene alkaloids, inparticular the 3,4-dihydroxy- and 3,4-methylenedioxy derivatives,display high selectivity towards D2�DR. Taking into account thatD1�DR interaction may lead to undesirable side effects, includingdyskinesia and cardiovascular disturbances, the D2 selective com-pounds synthesized may constitute new and useful candidatesfor the treatment of schizophrenia (antagonists) or Parkinson’s dis-ease (agonists).

Acknowledgments

This study was supported by grants SAF2011-23777, SpanishMinistry of Economy and Competitiveness, RIER RD08/0075/

L. Moreno et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4824–4827 4827

0016, Carlos III Health Institute, Spanish Ministry of Health and theEuropean Regional Development Fund (FEDER).

References and notes

1. Oloff, S.; Mailman, R. B.; Tropsha, A. J. Med. Chem. 2005, 48, 7322.2. Beaulieu, J. M.; Gainetdinov, R. R. Pharmacol Rev. 2011, 63, 182.3. Bentley, K. W. Nat. Prod. Rep. 2006, 23, 444.4. Cabedo, N.; Berenguer, I.; Figadère, B.; Cortes, D. Curr. Med. Chem. 2009, 16, 2441.5. Na, Y.; Neumeyer, J. L.; Baldessarini, R. J.; Xuechu, Z. Chem. Rev. 2007, 107, 274.6. Sondergaard, K.; Kristensen, J. L.; Palner, M.; Gillings, N.; Knudsne, G. M.; Roth,

B. L.; Begtrup, M. Org. Biomol. Chem. 2005, 3, 4077.7. Zhang, A.; Zhang, Y.; Branfman, A. R.; Baldessarini, R. J.; Neumeyer, J. L. J. Med.

Chem. 2007, 50, 171.8. Guinaudeau, H.; Lebœuf, M.; Cavé, A. J. Nat. Prod. 1994, 57, 1033.9. Guh, J. H.; Hsieh, C. H.; Teng, C. H. Eur. J. Pharmacol. 1999, 374, 503.

10. Chiou, C. M.; Kang, J. J.; Lee, S. S. J. Nat. Prod. 1998, 61, 46.11. Lin, C. J.; Chen, C. H.; Liu, F. W.; Kang, J. J.; Chen, C. K.; Lee, S. L.; Lee, S. S. Life Sci.

2006, 79, 144.12. Teng, C. M.; Hsiao, G.; Ko, F.; Lin, D. T.; Lee, S. S. Eur. J. Pharmacol. 1996, 303, 129.13. Milian, L.; Estellés, E.; Abarca, B.; Ballesteros, R.; Sanz, M. J.; Blázquez, M. A.

Chem. Pharm. Bull. 2004, 52, 696.

14. Estellés, R.; López-Martín, J.; Milian, L.; O’Connor, J. E.; Martínez-Losa, M.;Cerdá-Nicolás, M.; Anam, E. M.; Ivorra, M. D.; Issekuts, A. C.; Cortijo, J.;Morcillo, J. E.; Blázquez, M. A.; Sanz, M. J. Br. J. Pharmacol. 2004, 142, 229.

15. Bhattacharya, S. K.; Bose, R.; Ghosh, P.; Tripathi, V. J.; Ray, A. B.; Dasgupta, B.Psychopharmacology 1978, 59, 29.

16. Bermejo, A.; Protais, P.; Blázquez, M. A.; Rao, K. S.; Zafra-Polo, M. C.; Cortes, D.Nat. Prod. Lett. 1995, 6, 57.

17. Cabedo, N.; Andreu, I.; Ramírez, M. C.; Chagraoui, A.; Serrano, A.; Bermejo, A.;Protais, P.; Cortes, D. J. Med. Chem. 2001, 44, 1794.

18. Berenguer, I.; El Aouad, N.; Andujar, S.; Romero, V.; Suvire, F.; Freret, T.;Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Boulouard, M.; Cabedo, N.; Cortes, D.Bioorg. Med. Chem. 2009, 17, 4968.

19. El Aouad, N.; Berenguer, I.; Romero, V.; Marín, P.; Serrano, A.; Andujar, S.;Suvire, F.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Cabedo, N.; Cortes, D. Eur. J.Med. Chem. 2009, 44, 4616.

20. Andujar, S.; Suvire, F.; Berenguer, I.; Cabedo, N.; Marín, P.; Moreno, L.; Ivorra,M. D.; Cortes, D.; Enriz, R. D. J. Mol. Model. 2012, 18, 419.

21. Lafrance, M.; Blaquiere, N.; Fagnou, K. Eur. J. Org. Chem. 2007, 811.22. Shamma, M.; Hwang, D. Y. Tetrahedron 1974, 30, 2279.23. Andujar, S.; Tosso, R. D.; Suvire, F.; Angelina, E.; Peruchena, N.; Cabedo, N.;

Cortes, D.; Enriz, R. D. J. Chem. Inf. Model. 2012, 52, 99.24. Luthra, P. M.; Kumar, J. B. J. Med. Chem. 2012, 14, 1556.25. Poewe, W. Neurology 2009, 72, S65.

http://refhub.elsevier.com/S0960-894X(13)00811-1/h0005








































ORIGINAL PAPER

Tetrahydroisoquinolines acting as dopaminergic ligands.A molecular modeling study using MD simulationsand QM calculations

Sebastián Andujar & Fernando Suvire & Inmaculada Berenguer & Nuria Cabedo &

Paloma Marín & Laura Moreno & María Dolores Ivorra & Diego Cortes &

Ricardo D. Enriz

Received: 15 December 2010 /Accepted: 22 March 2011 /Published online: 27 April 2011# Springer-Verlag 2011

Abstract A molecular modeling study on 16 1-benzyltetrahydroisoquinolines (BTHIQs) acting as dopaminergicligands was carried out. By combining molecular dynamicssimulations with ab initio and density functional theory(DFT) calculations, a simple and generally applicableprocedure to evaluate the binding energies of BTHIQsinteracting with the human dopamine D2 receptor (D2 DR)is reported here, providing a clear picture of the bindinginteractions of BTHIQs from both structural and energeticviewpoints. Molecular aspects of the binding interactionsbetween BTHIQs and the D2 DR are discussed in detail. Asignificant correlation between binding energies obtainedfrom DFT calculations and experimental pKi values wasobtained, predicting the potential dopaminergic effect ofnon-synthesized BTHIQs.

Keywords 1-Benzyl-THIQ . Halogenated-1-benzyl-THIQ .

D2-dopamine receptor . Structure-activity relationship .

Molecular dynamic simulation . Ab initio and DFTcalculation

Introduction

The dopamine D2 receptor (D2 DR) has been implicated inthe mechanism of drugs used in the treatment of disorderssuch as schizophrenia and Parkinson’s disease. For thesereasons, a great deal of research has focused on thediscovery of novel dopaminergic ligands as potential drugcandidates [1]. DR can be classified into two pharmaco-logical families (D1 and D2-like) that are encoded by atleast five genes. Which receptor(s) needs to be activated toobtain therapeutic effects in Parkinson’s disease has beenthe subject of controversy [2]. The D2-like DR show highaffinities for drugs (antagonists) used in the treatment ofschizophrenia (antipsychotics) and those (agonists) utilizedto treat the Parkinson’s disease [3].

Tetrahydroisoquinolines (THIQs)—the most numerousnaturally occurring alkaloids—include 1-benzyl-THIQs andaporphines, both of which have structural similarities todopamine and can interact with DR [4]. Previous results inour group suggested that some natural and synthetic 1-benzyl-THIQs alkaloids were able to bind to DR [5–7]. Inthis way, we described the enantioselective syntheses of pairsof dopaminergic (1S)- and (1R)-benzyl-THIQs using (R)- and(S)-phenylglycinol as the chiral source, and we observedthat, in these series of 1-benzyl-THIQs, (1S)-enantiomerswere 5–15 times more effective at D1-like and D2-likedopamine receptors than (1R)-enantiomers [8] (Fig. 1). Onthe other hand, we described the preparation in a ‘one-pot’sequence of 1-cyclohexylmethyl 7,8-dioxygenated-THIQ,substituted and unsubstituted in the C ring by application ofthe photo–Fries transposition, followed by a tandem reduction-cyclization and further reduction. Indeed, we accomplished forthe first time a regioselective hydrogenation of the benzyl ringin the THIQ system. All 1-cyclohexylmethyl THIQs studied in

S. Andujar : F. Suvire :R. D. Enriz (*)Departamento de Química, Universidad Nacional de San Luis,San Luis, Argentinae-mail: [email protected]

S. Andujar :R. D. EnrizIMIBIO-SL (CONICET),Chacabuco 915,5700, San Luis, Argentina

I. Berenguer :N. Cabedo : P. Marín : L. Moreno :M. Dolores Ivorra :D. CortesDepartamento de Farmacología, Facultad de Farmacia,Universidad de Valencia,46100 Burjassot Valencia, Spain

J Mol Model (2012) 18:419–431DOI 10.1007/s00894-011-1061-0

this work were able to displace the D2-like DR radioligandfrom its specific binding sites in rat striatal membranes, whilethe N-methylated derivatives also showed affinity for the D1-like DR. Recently, we reported the influence of thesubstitution at the 1-position over a 7-chloro-6-hydroxy-THIQ core [9, 10]. In previous works [8, 11–15], wedetermined the role of certain structural requirements forimproving the affinity for D1 and D2-like DR. Thus, we areable to postulate that the presence of a hydroxyl (OH) and ahalogen group (Cl) in the THIQ A-ring could lead toobtaining molecules that can bind selectively to one of thetwo groups of receptors mentioned above [13, 14]. Preservingthe chlorine and hydroxyl (or methoxyl) groups at the C-6and C-7 positions of the THIQ A-ring, respectively, with asecondary (NH) or a tertiary (NMe) amine, we explored theimpact of inclusion of aliphatic and aromatic groups such asbutyl-, phenyl-, benzyl-[8] as well as halogenated-1-benzylbenzyl moieties at 1-position [9] to determine theirinfluence over dopaminergic activity. Thus, we have recentlyreported five series of 1-substituted-THIQs: 1-butyl-THIQs(compounds 1–3 in Fig. 2), 1-phenyl-THIQs (4–6), 1-benzyl-THIQs (7–9), 2′-bromobenzyl-THIQs (11–13) and 2′,4′-dichlorobenzyl-THIQs (14–16). Compound 10, which wasobtained from a fortuitous synthesis, was also included insuch a report [8]. During a Bischler-Napieralski cyclization,we observed the same fact reported by Doi et al. in 1997 [16]when preparing 1-butyl-THIQs. The need to add P2O5 (andPOCl3 at a molar ratio of 1:1) to the cyclodehydrationreaction, because of the difficulty of cyclizing the amidewhen there is a chlorine in the structure (originally at the C-6position of the A-ring), causes an aberrant cyclization, by

means of formation of a nitrilium intermediate, which givestwo positional isomers, clearly identified after the reductionstep: 6-chloro-7-hydroxy-1-butyl-THIQ (compound 2), and 6-hydroxy,7-chloro-1-butyl-THIQ (compound 10: unexpectedcyclization product), in a 1:2 ratio.

All these compounds were assayed in vitro for theirability to displace selective radioligands of D1 and D2 DRfrom their respective specific binding sites in rat striatalmembranes, and were tested for their ability to inhibit invitro 3[H]-dopamine uptake in rat striatal synaptosomes.Many of these compounds were able to displace both 3[H]-SCH 23390 and 3[H]-raclopride at nano or micromolar (nMor μM) concentration from their specific binding sites in ratstriatum, but all compounds had only low or no effect on3[H]-dopamine uptake [8, 9]. The replacement at the C-1position of THIQs, is an important factor modulating theselectivity at DR. Compounds 1, 3, 10 and 11 (Table 1)show a greater affinity towards D2 receptors when a butylor a benzyl moiety, respectively, is located in that position.The different activities and selectivity obtained for thesecompounds can be explained by the different spatialorientations adopted by the varied hydrophobic portionslocated at C-1, which could give different molecularinteractions with the D1 and D2 receptors. Since someBTHIQs have shown a great affinity for the D2 DR,considerable interest has developed in delineating theportions of the BTHIQ molecular structure responsible forits dopaminergic properties and interactions with the D2DR. The process of drug design could be considerablyimproved if receptors and their mode of interaction withligands were known in precise molecular detail. Such

Fig. 1 a–d Structural featuresof tetrahydroisoquinoline(THIQ) compounds. a (1S)-1-Benzyl-6,7-methylenedioxy-1,2,3,4-THIQ. b (1R)-1-Benzyl-6,7-methylenedioxy-1,2,3,4-THIQ. c (1S)-N-Propyl-1-benzyl-6,7-dihydroxy-1,2,3,4-THIQ. d (1R)-N-Propyl-1-benzyl-6,7-dihydroxy-1,2,3,4-THIQ (previously reportedin [8])

420 J Mol Model (2012) 18:419–431

information could then be used to design more definedstructures in which the pharmacophoric groups are orientedin the appropriate spatial arrangement for optimal receptorinteraction.

In the present work, we report a molecular modeling studyperformed on 16 BTHIQs acting as dopaminergic ligands.Combined molecular dynamics (MD) simulations and quan-tum mechanics (semiempirical, ab initio and DFT) calcu-lations were employed in our study to evaluate the molecular

interactions between the BTHIQs and the D2 DR. Anexcellent correlation between binding energies obtained fromDFT calculations and experimental pKi was obtained.

Materials and methods

Theoretical calculations were carried out in two steps. In afirst step, we performed MD simulations of the molecular

Fig. 2 Structural features of the16 BTHIQs reported here,showing the different torsionalangles

Compound Relative binding energy (BE) (kcal/mol) Specific-D2 ligand[3H]-raclopride

EU(RHF/6-31G(d))

Δ EU(RHF/6-31G(d))

EU(B3LYP/6-31G(d,p))

Δ EU(B3LYP/6-31G(d,p))

pKi

1 -98.91 22.16 -114.29 24.58 6.108±0.165 [9]

2 -80.00 41.07 -93.62 40.25 5.424±0.026 [9]

3 -111.92 9.15 -135.43 3.44 7.117±0.151 [9]

4 -75.46 45.61 -94.89 43.98 5.212±0.124 [9]

5 -83.02 38.05 -103.00 35.87 5.670±0.406 [9]

6 -90.48 30.59 -105.35 33.52 5.950±0.198 [9]

7 -99.94 21.13 -118.48 20.39 6.014±0.049 [9]

8 -84.29 36.78 -99.93 38.94 5.816±0.181 [9]

9 -116.35 4.72 -137.83 1.04 7.178±0.091 [9]

10 -114.03 7.04 -138.43 0.44 7.220±0.139 [9]

11 -100.47 20.6 -128.07 10.8 6.630±0.092 [10]

12 -81.59 39.48 -113.31 25.56 5.896±0.099 [10]

13 -121.07 0 -138.87 0 7.391±0.139 [10]

14 -74.61 46.46 -101.31 37.56 5.507±0.105 [10]

15 -77.31 43.76 -101.10 37.77 5.230±0.096 [10]

16 -102.79 18.28 -131.14 7.73 6.996±0.105 [10]

Table 1 Relative bindingenergies obtained for the differ-ent complexes. Previouslyreported experimental pKi dataare shown in the last column

J Mol Model (2012) 18:419–431 421

interactions between compounds 1–16 and D2 DR. In thesecond step, reduced model systems were optimized usingquantum mechanics calculations. Semiempirical (AM1)combined with ab initio [RHF/6-31G(d)] and B3LYP [6-31G(d,p)] calculations were employed in these optimizations.

Molecular dynamics simulations

It must be pointed out that the principal goal of the MDsimulations performed here was not to obtain a new D2 DRby homology. Our aim in this study was less ambitious; wewished to obtain a reasonable indication of the relationshipbetween the structures of compounds 5–7 and theirpotential affinities for the binding pocket of D2 DR. Thus,for this purpose, we considered it more appropriate to use apreviously reported and extensively tested model for D2DR [17]. In fact, there are many molecular modelingstudies in the literature reporting D2 DRs obtained byhomology, all of them structurally closely related [18–20].Thus, in the present study, we used the D2DR modelpreviously reported in reference [17]. The ligand topologieswere built using the mktop program [21]. For this purpose,we used the previously optimized geometry at RHF/6-31G(d) level of theory of the global minimum of each ligand. Inthe present study, we used an approach where manualdocking was guided by information from site-directedmutagenesis and short docking simulations, with both thereceptor and the ligand free to move. Structurally similarparts of the ligands were oriented in similar positions in thereceptor model, which was described by Mansour et al. [22]and Lan et al. [23]. Thus, receptor–ligand complexes wereprepared in order to obtain the input files forMD runs. Severaldocking positions were considered and the strongest receptorinteractions were examined in detail.

The MD simulations and analysis were performed usingthe GROMACS 3.2.1 simulation package [24, 25] with theOPLS-AA force field [26–30] and the rigid SPC watermodel [31, 32] in a cubic box with periodic boundaryconditions. Receptor–ligand complexes were embedded ina box containing the SPC water model that extended to atleast 1 nm between the receptor and the edge of the box,resulting in a box of 7.17 nm in side length. The totalnumber of water molecules was 11,330 for the differentsimulations. Three Na+ ions were then added to the systemsby replacing water in random positions, thus making thewhole system neutral. The time step for simulations was0.001 ps for a complete simulation time of 5 ns. For long-range interactions, the particle-mesh Ewald (PME) [33–35]method was used with a 1 nm cut-off and a Fourier spacingof 0.12 nm. The MD protocol consisted of severalpreparatory steps: energy minimization using the conjugategradient model [36, 37] density stabilization (NPT con-ditions), and finally production of the MD simulation

trajectory. All production simulations were performedunder NVT conditions at 310 K, using Berendsen’scoupling algorithm [38] for keeping the temperatureconstant. The compressibility was 4.8×10−5 bar−1. Allcoordinates are saved every 5 ps. The SETTLE [24]algorithm was used to keep water molecules rigid. TheLINCS [39] algorithm was also used to constrain all C-αatom positions for the receptor in order to avoid unfoldingproblems. The simulations were analyzed using the analysistools provided in the Gromacs package.

Histidine in the active site is a potential problem becausethe state of His (neutral or protonated) is a controversialtopic. We were particularly interested in performingsimulations under physiological conditions (pH≈7). Previ-ous reports have indicated that, under physiological con-ditions (pH≈7), histidine located in a hydrophobicenvironment (hydrophobic pocket without water molecules)is in neutral form [40]. In addition, previous simulationsperformed for D3DR by Micheli et al. [20] also consideredthe histidine residue to be neutral. Thus, on the basis ofthese results, we considered His in neutral form in ourcalculations. This amino acid was calculated as follow:protons were added using the program pdb2gmx, in theGROMACS suite of programs, for optimization of thehydrogen bond network. His protons were placed bydefault; these selections were done automatically (Hiswas in neutral form). This is based on an optimalhydrogen bonding conformation. Hydrogen bonds aredefined based on simple geometric criteria, specified bythe maximum hydrogen–donor–acceptor angle and donor–acceptor distance.

It should be noted that the compounds reported herepossess one chiral center, and are therefore enantiomericwith the possibility of two isomers (1-S and 1-R). However,we did not perform an enantiomeric resolution for previouslyreported biological assays; thus, only one isomer of eachcompound was evaluated in our MD simulations. To choosethe isomeric forms of each compound, we considered on theone hand previously reported results [15] and, on the other,preliminary and specially performed exploratory simulationsdetermining the spatially preferred form of each compound(results not shown). Our previous experimental results onstructurally related compounds suggested that the S formwould be the preferred isomer for these compounds [15].The preliminary and exploratory MD simulations are inagreement with these experimental data, indicating that thespatial ordering adopted by 1-S forms gives adequateorientation of the molecules to interact in the active site ofthe dopamine D2 receptor.

The equilibrium state of the complexes was observedfrom the onset of simulation until 5 ns. The temperaturewas stabilized at 310±4 K for all complexes. The potentialenergy stabilized in a short time period (around 0.5 ns), and

422 J Mol Model (2012) 18:419–431

the values obtained suggested that the system was wellequilibrated.

Considering the 5 ns of MD simulation, and from the timeprofiles, it was concluded that some properties of the ligand–receptor complexes reached stable average values at around0.5 ns, whereas others take longer time periods. For thisreason, and to ensure full equilibration, only the last 4.5 nswere taken into account for the analysis. After discarding thefirst 0.5 ns of the trajectory, we followed the changes in spatialordering of the ligand–receptors complexes.

Quantum mechanics calculations

The binding pocket of the D2 L–R (ligand-receptor) wasdefined according to Teeter et al. [41] and Neve et al. [42]. Inour reduced model system, only 13 amino acids wereincluded in molecular simulations. The size of the molecularsystem simulated and the complexity of the structures underinvestigation restricted the choice of the quantum mechanicalmethod to be used. Consequently, the semiempirical AM1method was selected combined with ab initio calculations(RHF/6-31G(d)). The torsional angles of the ligands and theflexible side-chains of the amino acids as well as the bondangles and bond lengths of the moieties involved in thepotential intermolecular interactions were optimized at thesemiempirical level. Next, the torsional angles of the ligandsand the flexible side-chains of the amino acids as well as thepotential intermolecular interactions were optimized at RHF/6-31G(d) and DFT [B3LYP/6-31G(d,p)] levels of theory. Incontrast, the torsional angles of backbones as well as thebond angles and bond lengths of non-interacting residueswere kept frozen during the calculations.

The binding energy of the complexes was calculated, withthe approximation neglecting the superimposition of error dueto the difference between the total energies of the complexwith the sum of the total energies of the components:

BEQM¼EL=D2DR � ED2DR þ ELð Þ ð1Þ

where BEQM is the binding energy, EL/D2DR is the complexenergy, ED2DR the energy of the reduced receptor model(binding pocket) and EL the energy of the ligand.

All the quantum mechanical calculations reported herewere carried out using the Gaussian 03 program [43].

Spatial views shown in Figs. 3, 9 and 10 were constructedusing the UCSF Chimera program [44] as the graphicinterface.

Results and discussion

Our molecular modeling study was carried out in two steps.First, we performed MD simulations of the molecular

interactions between the compounds shown in Fig. 2 withthe human D2 DR (Fig. 3). In the second step, reducedmodel systems (shown as a circle in Fig. 3) were optimizedusing quantum mechanic calculations. Semiempirical(AM1) combined with ab initio [RHF/6-31G(d)] and DFT[B3LYP/6-31G(d,p)] calculations were employed for theseoptimizations.

Molecular dynamic simulations

Comparing the results obtained for the different complexesled to interesting general conclusions. Consistent withprevious experimental [22] and theoretical [45] results,our simulations indicate the importance of the negativelycharged aspartate 86 for binding of these ligands. A highlyconserved aspartic acid (Asp 86) in trans-membrane helix 3(TM3) is important for the binding of both agonists andantagonists to the D2 receptor, [22, 46, 47], and its terminalcarboxyl group may function as an anchoring point forligands with a protonated amino group [23, 41, 42, 47]. Inthe present study, all the compounds simulated were dockedinto the receptor with the protonated amino group near Asp86. After 5 ns of MD simulations, the ligands had movedslightly but in a different form compared with the initialposition. However, the strong interaction with Asp 86 wasmaintained for all complexes (see Fig. 4), supporting thesuggestion that Asp 86 functions as an anchoring point forligands with a protonated amino group.

Pharmacological data [22, 48] indicate that the hydroxylgroups of dopaminergic ligands are of primary importancein stabilizing binding, suggesting that the serine residues(141 and 144) of the D2 receptor may not be equallyimportant for binding affinity. Individual mutation ofserines 141 and 144 in TM5 to alanine produced asymmet-rical effects on dopamine receptor binding. These resultsindicated that Ser 141 might be differentially important fordopamine binding. In addition, site-directed mutagenesisstudies have indicated that a cluster of serine residues inTM5 (Ser 141, Ser 144) and in TM4 (Ser 122 and Ser 118)is important for agonist binding and receptor activation [45,47–49]. It was suggested that the serine cluster anddopamine form a hydrogen-bonding network. Such ahydrogen-bonding network was reproduced by the MDsimulation of these complexes (Fig. 5). In these complexes,the strongest contributor to the network was Ser 141, whichis consistent with the experimental observation that a Ser 141Ala mutated receptor completely lost dopamine-inducedactivation [22]. The 7-hydroxyl group of compound 3displayed another significant hydrogen bond interaction withSer 122; however, this interaction is weaker with respect tothe hydrogen bond with Ser 141.

Figure 5 shows that compounds 3 and 9 display stronghydrogen bond interactions with Ser141 during the entire

J Mol Model (2012) 18:419–431 423

simulation period. Similar results were obtained for com-pounds 6, 10, 13 and 16. However, for the rest of theBTHIQs evaluated here, such interactions were slightlyweaker. It should be noted that in compounds 3, 6, 9, 10, 13and 16, the hydroxyl group on the ring-A is acting as aproton-donor; whereas the oxygen atom of the OH group ofSer141 is the proton-acceptor counterpart. In contrast, inthe case of compounds 1, 2, 4, 5, 7, 8, 11, 12, 14 and 15,the OH group of Ser 141 is the proton-donor and themethoxyl group on the ring-A is the acceptor counterpart.MD simulations predict that these interactions are weaker incomparison to those observed for hydroxyl ligands on thering-A.

Aromatic side chains are bulky, have low barriers forrotation, and are ideal for adjusting to the changingconformation of the hydrophobic moiety of the ligand. In

0 1000 2000 3000 4000 50000.2

0.3

0.4

0.5

0.6

0.7

0.8

9 3 6

dist

ance

(nm

)

time (ps)

Fig. 4 Bond lengths obtained for the salt bridge between Asp 86 andthe protonated amine group in compounds 3, 6 and 9

Fig. 3 Spatial view obtained forthe dopamine D2 receptor (D2DR) model. The plot was per-formed using the UCSFChimera program [44] programas a graphic interface. Confor-mations used as starting geome-tries for the molecular dynamics(MD) simulations of compounds3 (cyan), 6 (green) and 9(yellow) are shown. The bindingpocket optimized from quantummechanics calculations isdenoted with a circle. Thenumbers of the amino acidsincluded correspond to reference[17] and not to those given inthe crystal data

0 1000 2000 3000 4000 5000

0

1N

umbe

r of

hyd

roge

n bo

nds

time (ps)

time (ps)

a

0 1000 2000 3000 4000 5000

0

1

Num

ber

of h

ydro

gen

bond

s

b

Fig. 5 Hydrogen bonds obtained for compounds 9 (a) and 3 (b).These interactions are between Ser 141 and the catecholic hydroxyls

424 J Mol Model (2012) 18:419–431

the dopamine D2 receptor, the binding site proved to bealigned with aromatic side chains, and such residues canadjust to the different shapes and flexibility of the ligands inthe binding site. Thus, Phe 82, Val 83 and Val 87(TM3);Phe 145 (TM5); and Trp 182, Phe 185, Phe 186 and His189 (TM6) form a mostly hydrophobic pocket for ligands(Fig. 3).

It is interesting to note that the only structural differencesbetween compounds 3, 6, 9, 13 and 16 are the differentsubstituents at C-1. Whereas compound 6 has a relativelyrigidly held phenyl ring, the corresponding butyl, benzyland halogenated-benzyl substituents on compounds 3, 9, 13and 16, respectively, are free to rotate, allowing betteraccommodation of these hydrophobic moieties to interactwith the cluster of aromatic and non polar residues. Theseresults might be better appreciated by observing thedifferent conformational behaviors obtained for the torsionalangles of their respective hydrophobic portions during thesimulations (Figs. 5–8). The conformational behaviorsobserved for the torsional angles θ1 and θ2 of compound 9are shown in Fig. 4. Whereas θ1 is maintained relativelyfixed at about 250° during the simulation (Fig. 6a), thetorsional angle θ2 displayed a high molecular flexibility,adopting conformations from 20° to 300° (Fig. 6b). Closelyrelated results were obtained for the torsional angles θ1 andθ2 of compounds 13 and 16. The hydrophobic portion ofcompound 3, the butyl moiety, also displayed a highmolecular flexibility. Figure 6 gives the conformationalbehaviors of torsional angles f1–f3 of compound 3. Thetorsional angle f1 adopts a relatively rigid planar form closeto 170° (Fig. 7a) but the other two torsional angles f2 and f3displayed a high molecular flexibility (Fig. 7b and c,respectively). Very similar results were obtained for thebutyl portion of compound 10. In contrast, the conforma-tional behavior obtained for the phenyl ring of compounds4–6 displayed a very restricted molecular flexibility, keepinga spatial ordering almost perpendicular with respect to therest of the molecule during the entire simulation (Fig. 8). Thedifferent affinities previously reported for compounds 6 and9 suggest that the orientation of the substituent at C-1 maybe a more important factor in the different effects on receptoraffinity for the two ligands. This argument also applies to 3,10, 13 and 16, where the orientations of the butyl andhalogenated-benzyl substituents are more favorable forhydrophobic interactions. Thus, the different affinities andselectivities obtained for these compounds might beexplained, at least in part, by the different spatial orientationsadopted by the varied hydrophobic portions located at C-1,which give different molecular interactions with the D2receptor. These aspects are discussed in detail in terms ofquantum mechanics calculations in the next section.

In the next step of our study, we evaluated thebinding energies (BE) obtained for the different com-

plexes. From the binding energies obtained in our MDsimulations, one can distinguish a very good binderfrom a very weak binder (−441,217.42 kJ mol−1 forcompound 9 vs −441,015.35 kJ mol−1 for compound 4)but cannot distinguish ligands with similar binding affinities(−441,217.42 kJ mol−1 for compound 9 vs −441,004.26 kJmol−1 for compound 3 and −441,015.35 kJ mol−1 forcompound 4 vs −440,155.71 kJ mol−1 for compound 1among other examples). This is not an unexpected result; canwe realistically expect to make accurate and reliablepredictions with what are decidedly crude representationsof the molecular interactions involved in the bindingprocess? Any model that neglects or only poorlyapproximates the terms that are playing determinantroles, such as, e.g., lone pair directionality in hydrogenbonds, explicit п-stacking polarization effects, hydrogenbonding networks, induced fit, and conformational entropy,

0 1000 2000 3000 4000 5000

dihe

dral

ang

le (

degr

ee)

time (ps)

0 1000 2000 3000 4000 5000

time (ps)

θ1

θ2

a

dihe

dral

ang

le (

degr

ee)

b

020406080

100120140160180200220240260280300320340360

020406080

100120140160180200220240260280300320340360

Fig. 6 Evolution of the angles θ1 (a) and θ2 (b) of compound 9 withtime during the simulation

J Mol Model (2012) 18:419–431 425

among others, cannot reasonably be expected to distinguishbetween compounds possessing relatively similar binding

energies. There are several works supporting this concept inthe literature [50, 51].

At this stage of our work, we considered the trendpredicted for the MD simulations as certainly significantbut, on the other hand, we might be reluctant to assign it aquantitative significance, because of the approximationsinvolved in this mode of approach. It should be noted thatwe are dealing with relatively weak interactions andtherefore MD simulations might underestimate such inter-actions. Thus, in the next step, we optimized reduced modelsystems using combined semiempirical, ab initio and DFTcalculations.

Quantum mechanics calculations

AM1 calculations combined with RHF/6-31G(d) and DFT[B3LYP/6-31G(d,p)] optimizations were performed byconsidering all receptor amino acids that could interactafter initial positioning of the ligands against Asp 86 andSer 141 residues. The binding pocket designed in this way(Fig. 3) provided data that matched experimental resultspreviously reported from binding assays [8, 9].

Figure 9a shows ligand 13 interactions with the D2 DRoptimized using quantum mechanical calculations. The saltbridge between the protonated amino group and thecarboxyl group of Asp 86, as well as the hydrogen bondbetween the 7-hydroxyl group with Ser 141 can be seen inthis figure. From Fig. 9a it is clear that a strong salt bridgeexists in this compound between the protonated aminogroups and the carboxyl group of Asp 86 (calculateddistance of 3.47Å). The hydrogen bond between 13 and Ser141 is a bifurcated interaction in which the oxygen atom ofthe hydroxyl group and the oxygen of carbonyl group ofSer 141 are the proton-acceptors, giving interatomic

0 1000 2000 3000 4000 50000

20406080

100120140160180200220240260280300320340360

dihe

dral

ang

le (

degr

ee)

dihe

dral

ang

le (

degr

ee)

dihe

dral

ang

le (

degr

ee)

time (ps)

0 1000 2000 3000 4000 5000time (ps)

0 1000 2000 3000 4000 5000time (ps)

φ1

φ2

φ3

a

b

c

020406080

100120140160180200220240260280300320340360

020406080

100120140160180200220240260280300320340360

Fig. 7 Evolution of the angles f1 (a), f2 (b) and f3 (c) of compound 3with time during the simulation

0 1000 2000 3000 4000 50000

20406080

100120140160180200220240260280300320340360

ángu

lo d

iedr

o (g

rado

s)

tiempo (ps)

χ

Fig. 8 Evolution of the angle χ of compound 6 with time during thesimulation

426 J Mol Model (2012) 18:419–431

distances of 2.28Å and 2.40Å, respectively. Figure 9bshows ligand 11 interaction with the D2 DR. In this case,the 7-methoxyl group acts as proton-acceptor while thehydroxyl group of Ser 141 is the proton-donor, displayingan interatomic distance of 2.32Å.

Table 1 gives the BE calculated for the differentcomplexes using RHF/6-31 G(d) and B3LYP/ 6-31 G(d,p)

calculations. All compounds possessing 7-methoxyl groupsdisplayed higher BE with respect to the 7- hydroxylhomologues (cf. 1 with 3; 4 with 6; 7 with 9; 11 with 13,and 14 with 16). Previously, we reported that a 7-hydroxylgroup acting as a proton-donor gives a stronger hydrogenbond than those derivatives possessing a 7-methoxyl group[16]. The present results are in agreement with previously

Fig. 9 Interactions ofcompound 13 (a) and 11 (b)with the binding pocket of D2DR. Spatial view of two inter-actions: salt bridge (Asp 86 withprotonated amino group) to theright and hydrogen bondbetween meta-hydroxyl groupwith Ser 141to the left

Fig. 10 Interactions of compound 9 (a), 13 (b), 16 (c), 3 (d), 6 (e) and 10 (f) with the binding pocket D2 DR. Different spatial views show thehydrophobic interactions at the hydrophobic zone

J Mol Model (2012) 18:419–431 427

reported calculations for isolated and solvated molecules, aswell as with previously reported experimental bindingaffinities [8, 9] (see Table 1).

Figure 10a shows ligand 9 interactions with the bindingpocket. In this case, a different spatial view with respect toFig. 9 is shown in order to better appreciate the hydropho-bic interactions. From this figure, we can observe that thebenzyl group of 9 adopts an adequate conformation tointeract with Phe 186, Phe 82 and His 189. A similar spatialordering was obtained for compounds 13 and 16 (Fig. 10band c, respectively). For compounds 13 and 16, the halogensubstituent confers a higher polarizability on the benzylgroup, allowing a stronger hydrophobic interaction. Inshould be noted that compound 13 displayed the highestof pKi value in this series. These hydrophobic interactionscould explain, at least in part, the strongest affinity obtainedfor this compound. The butyl group of 3 displays a spatialordering closely related to that of the benzyl group of 9, 13and 16, also giving closely related hydrophobic interactionswith the same hydrophobic residues (Fig. 10d). In contrast,the phenyl group of 6 displayed a different spatial ordering,giving adequate distance to interact only with Phe 186(Fig. 10e). Interestingly, the bonding energies obtained forthese complexes are: 13/D2 DR < 10/D2 DR < 9/D2 DR <3/D2 DR < 16/D2 DR, which are in complete agreementwith their respective pKi values obtained from our previousexperimental results (see Table 1). Compound 10 adopts adifferent spatial ordering at the binding site; thus, the butylportion of this compound interacts with three aromaticresidues: Trp 182, Phe 82 and Phe 186 (Fig. 10f).Compounds 4–6 possess a phenyl ring perpendicular tothe rest of the ligand from the ring containing theprotonated nitrogen [52]. These compounds docked in theD2 receptor model have few interactions in the bindingpocket because their 1-phenyl substituents extend towardthe extracellular surface of the receptor, parallel to the helixaxes. These results are in agreement with those previouslyreported [23]. Thus, it appears that the shape and flexibility ofthe side chain at the C-1 position affects the receptor subtypeselectivity of ligands to an extent that depends on thegeometry, flexibility and stacking potential of ligand sub-stituents. Lan et al. [23], reported that the D1 selective ligandSCH23390 contains a phenyl ring perpendicular to the restof the molecule and the membrane plane, and parallel to thehelix axes, which could explain its selectivity. Our results arein agreement with those results. Compounds type 4–6 in thisseries displayed a conformational behavior closely related tothat reported for SCH23390.

Regarding the general structure of BTHIQs reportedhere, it is reasonable to think that the presence of a chlorineatom at C6, and consequently halogen bonding interactions,could be operative for the ligand–receptor complex forma-tion. Thus, this chlorine possibly could be interacting

through either a positive sigma hole with a negative sitein its vicinity or through its negative lateral ring ofelectrostatic potential with a positive site in the vicinity. Acomprehensive study on electrostatically driven non-covalent interactions has been reported recently by Politzeret al. [53]. In this latter article, the possibility that halogenand other σ-hole interactions can be competitive withhydrogen bonding has been clearly established. Unfortunately,from the limited information obtained from our relatively low-level theory calculations, it is not possible to properlydetermine if the halogen bonding interactions could take placehere. It is clear that further, more accurate calculations, as wellas quantum atoms in molecules (QAIM) [54, 55] analysis arenecessary for a detailed description of these interactions.Such calculations are now in progress in our laboratory andwill be reported later in a separate paper.

0 10 20 30 40 505

6

7

8

14

415

12

2

5

86

17

11

16

310

9

pKi

pKi

ΔE (kcal/mol)

0 10 20 30 40 50

ΔE (kcal/mol)

R=0.9557 R2=0.9134

a

13

5

6

7

8

b

R=0.9726 R2=0.9460

13

109

316

11

1

76

8512

2

154

14

a

b

Fig. 11 Correlations obtained between the experimental pKi valuesversus the binding energies (BE) calculated from a ab initio [RHF/631G(d)] calculations, and b DFT (B3LYP/6-31G(d,p)) computations

428 J Mol Model (2012) 18:419–431

Figure 11a gives a graphical representation of thecalculated BEs obtained from RHF/6-31G(d) calculationsversus experimental pKi values, obtained in binding studiesin rat striatum [8, 9]. This figure has a correlation coefficientR2=0.9134. Theisresult is very satisfactory when oneconsiders the type of approximations used. Figure 11b showsthe same correlations but in this case using BEs obtainedfrom DFT [B3LYP/6-31G(d,p)] calculations. In this draft,the correlation coefficient is R2=0.9460 indicating that DFTcalculations give a better correlation with the experimentaldata. Although both linear correlations are good enough topredict the biological activity of BTHIQs, it is clear that DFTcalculations give a significantly better correlation withrespect to RHF computations. This is particularly evidentfrom the high squares of correlation coefficients, r2 obtainedusing RHF and DFT calculations (0.9134 and 0.9460,respectively). From our results, it is clear that the predictedfirst-principles structure of the primary binding pocket of D2DR leads to correct predictions of the critical residues forbinding THIQs, and gives relative binding affinities thatcorrelate fairly well with those obtained in experimentsperformed in native tissue. This good correlation providesadditional validation for the predicted structure and function.

It should be noted that the AM1 method it is notadequate to describe the hydrogen bonds. In addition, theab initio and DFT calculations performed here probably donot properly consider the dispersion interactions. Fortu-nately, in this case it appears that such limitations are notsevere enough to prevent us obtaining our objectives. Suchan assumption appears to be reasonable, considering thesignificant correlation obtained between the experimentaldata and the theoretical calculations performed. However,we cannot exclude that a kind of error-cancellation couldhave taken place in this case. Thus, it must be pointed outthat the approaches used in this study could be operativeonly for THIQs and structurally related compounds. Toextend these approaches to other compounds possessingdifferent structures would require additional validation andmore accurate calculations.

Conclusions

A molecular modeling study on 16 BTHIQs acting asdopaminergic ligands was carried out. By combining MDsimulations with ab initio and DFT calculations, a simpleand generally applicable procedure to evaluate the bindingenergies of BTHIQs interacting with the D2 DR is reportedhere, providing a clear picture of the binding interactions ofBTHIQs from both structural and energetic viewpoints.Thus, our results give interesting information that may behelpful in obtaining a better understanding of the molecularinteractions between BTHIQs and the D2 DR.

A significant correlation between binding energiesobtained from DFT calculations and experimental pKivalues was obtained. These results could predict thepotential dopaminergic effect of non-synthesized BTHIQswith an acceptable degree of accuracy. Such informationcould be essential in determining a priori the putativeactivity of new BTHIQ derivatives. It is prudent to remark thatthe excellent correlation obtained here between experimentaldata and the theoretical calculations performed here could belimited to BTHIQs and structurally related compounds.However, we believe our results may be helpful in thestructural identification and understanding of the mini-mum structural requirements for these molecules, and canprovide a guide to the design of BTHIQs with this biologicalactivity.

Acknowledgments Grants from Universidad Nacional de San Luis(UNSL) partially supported this work. This research was alsosupported by the Spanish “Ministerio de Educación y Ciencia” grantSAF 2007–63142. S.A.A. thanks a postdoctoral fellowship ofCONICET-Argentina. R.D.E. is a member of the Consejo Nacionalde Investigaciones Científicas y Técnicas (CONICET-Argentina) staff.

References

1. Oloff S, Mailman RB, Tropsha A (2005) Application of validatedQSAR models of D1 dopaminergic antagonist for databasemining. J Med Chem 48:7322–7332

2. Sit SY, Xie K, Jacutin-Porte S, Boy KM, Seanz J, Taber MT,Gulwadi AG, Korpinen CD, Burris KD, Molski TF, Ryan E, XuC, Verdoorn T, Johnson G, Nichols DE, Mailman RB (2004)Synthesis and SAR exploration of dinapsoline analogues. BioorgMed Chem 12:715–734

3. Strange PG (1997) Dopamine receptors. Tocris Reviews 15.Tocris Cookson, Bristol

4. Zhang A, Neumeyer JL, Baldessarini RJ (2007) Recent progressin development of dopamine receptor subtype-selectiveagents:potential therapeutics for neurological and psychiatric disorders.Chem Rev 107:274–302

5. Protais P, Arbaoui J, Bakkali EH, Bermejo A, Cortes D (1995)Effects of various isoquinoline alkaloids on in vitro 3H-dopamineuptake. J Nat Prod 58:1475–1484

6. Bermejo A, Protais P, Blázquez MA, Rao KS, Zafra-Polo MC,Cortes D (1995) Dopaminergic isoquinoline alkaloids from rotosof Xylopia papuana. Nat Prod Res 6:57–62

7. Cabedo N, Protais P, Cassels BK, Cortes D (1998) Synthesis anddopamine receptor selectivity of the benzyltetrahydroisoquinoline,(R)-(+)-nor-roefractine. J Nat Prod 61:709–712

8. Cabedo N, Andreu I, Ramírez de ArellanoMC, Chagraoui A, SerranoA, Bermejo A, Protais P, Cortes D (2001) Enantioselective synthesesof dopaminergic (R)- and (S)-benzyltetrahydroisoquinolines. J MedChem 44:1794–1801

9. Berenguer I, El Aouad N, Andujar S, Romero V, Suvire F, FreretT, Bermejo A, Ivorra MD, Enriz RD, Boulouard M, Cabedo N,Cortes D (2009) Tetrahydroisoquinolines as dopaminergicligands: 1-butyl-7-chloro-6-hydroxy-tetrahydroisoquinoline, anew compound with antidepressant-like activity in mice. BioorgMed Chem 17:4968–4980

10. El Aouad N, Berenguer I, Romero V, Marín P, Serrano A, AndujarS, Suvire F, Bermejo A, Ivorra MD, Enriz RD, Cabedo N, Cortes

J Mol Model (2012) 18:419–431 429

D (2009) Structure–activity relationship of dopaminergic haloge-nated 1-benzyl-tetrahydroisoquinoline derivatives. Eur J MedChem 44:4616–4621

11. Andreu I, Cabedo N, Torres G, Chagraoui A, Ramirez de ArellanoMC, Gil S, Bermejo A, Valpuesta M, Portais P, Cortes D (2002)Syntheses of dopaminergic 1-cyclohexylmethyl-78-dioxygenatedtetrahydroisoquinolines by selective heterogeneous tandem hydro-genation. Tetrahedron 58:10173–10179

12. Bermejo A, Andreu I, Suvire F, Léonce S, Caignard DH, RenardP, Pierré A, Enriz RD, Cortes D, Cabedo N (2002) Syntheses andantitumor targeting G1 phase of the cell cycle of benzoyldihy-droisoquinoline. J Med Chem 45:5058–5068

13. Suvire FD, Andreu I, Bermejo A, Zamora MA, Cortes D, EnrizRD (2003) Conformational study of N-alkyl-benzyltetrahydroiso-quinolines alkaloid. J Mol Struct THEOCHEM 666–667:109–116

14. Suvire FD, Cabedo N, Chagraoui A, Zamora MA, Cortes D, EnrizRD (2003) Molecular recognition and binding mechanism of N-alkyl-benzyltetrahydro-isoquinolines to the D1 dopamine receptor.A computational approach. J Mol Struc THEOCHEM 666–667:455–467

15. Andreu I, Cortes D, Protais P, Cassels BK, Chagraoui A, CabedoN (2000) Preparation of dopaminergic N-Alkyl-benzyltetrahydroi-soquinolines using a ‘One-Pot’ procedure in acid medium. BioorgMed Chem 8:889–895

16. Doi S, Shirai N, Sato Y (1997) Abnormal products in theBischler-Napieralski isoquinoline synthesis. J Chem Soc PerkinTrans 1:2217–2221

17. Andujar SA, Migliore de Angel BM, Charris JE, Israel A, Suárez-Roca H, López SE, Garrido MR, Cabrera EV, Visual G, Rosales C,Suvire FD, Enriz RD, Angel-Guío JE (2008) Synthesis dopaminergicprofile and molecular dynamics calculations of N-aralkyl substituted2-aminoindans. Bioorg Med Chem 16:3233–3244

18. Kalani MYS, Vaidehi N, Hall SE, Trabanino RJ, Freddolino PL,Kalani MA, Floriano WB, Wai Tak Kam V, Goddard WA III(2005) The predicted 3D structure of the human D2 dopaminereceptor and the binding site and binding affinities for agonistsand antagonists. Proc Natl Acad Sci USA 101:3815–3820

19. Becker OM, Marantz Y, Shacham S, Inbal B, Heifetz A, Kalid O,Bar-Haim S, Warshaviak D, Fichman M, Noiman SG (2004)Protein-coupled receptors: in silico drug discovery in 3D. ProcNatl Acad Sci USA 101:11304–11309

20. Micheli F, Bonanomi G, Blaney FE, Braggio S, Capelli AM,Checchia A, Curcuruto O, Damiani F, Di Fabio R, Donati D,Gentile G, Gribble A, Hamprecht D, Tedesco G, Terreni S, Tarsi L,Lightfoot A, Stemp G,MacDonald G, Smith A, Pecoraro M, PetroneM, Perini O, Piner J, Rossi T, Worby A, Pilla M, Valerio E, GriffanteC, Mugnaini M, Wood M, Scott C, Andreoli M, Lacroix L, SchwarzA, Gozzi A, Bifone A, Ashby CR Jr, Hagan JJ, Heidbreder C (2007)124-Triazol-3-yl-thiopropyl-tetrahydrobenzazepines: a series ofpotent and selective dopamine D3 receptor antagonists. J MedChem 50:5076–5089

21. Ribeiro AA, Horta BAC, de Alencastro RB (2008) MKTOP: aprogram for automatic construction of molecular topologies. JBraz Chem Soc 19:1433–1435

22. Manzour A, Meng F, Meador-Woodruff JH, Taylor LP, Civelli O,Akil H (1992) Site-directed mutagenesis of the human dopamineD2 receptor. Eur J Pharm Mol Pharmacol Sec 227:205–214

23. Lan H, DuRand CJ, Teeter MM, Neve KA (2006) Structuraldeterminants of pharmacological specificity between D1 and D2

dopamine receptors. Mol Pharmacol 69:185–19424. Berendsen HJC, Van der Spoel D, Van Drunen R (1995)

GROMACS: a message-passing parallel molecular dynamicsimplementations. Comput Phys Commun 91:43–56

25. Lindahl E, Hess B, van der Spoel D (2001) GROMACS 3.0: apackage for molecular simulations and trajectory analysis. J MolModel 7:306–317

26. van Buuren AR, Marrink SJ, Berendsen HJC (1993) A moleculardynamics study of the decane/water interface. J Phys Chem36:9206–9212

27. Mark AE, van Helden SP, Smith PE, Janssen LHM, vanGunsteren WF (1994) Convergence properties of free energycalculations. A-cyclodextrin complexes as a case study. J AmChem Soc 116:6293–6302

28. Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, KleinML (1983) Comparison of simple potential functions forsimulating liquid water. J Chem Phys 79:926–935

29. van Buuren AR, Berendsen HJC (1993) Molecular dynamicssimulation of the stability of a 22 residue alpha-helix in water and30% trifluoroethanol. Biopolymers 33:1159–1166

30. Liu H, Muller-Plathe F, van Gunsteren WF (1995) A force fieldfor liquid dimethyl sulfoxide and physical properties of liquiddimethyl sulfoxide calculated using molecular dynamics simulation.J Am Chem Soc 117:4363–4366

31. Miyamoto S, Kollman PA (1992) SETTLE—an analytical versionof the SHAKE and RATTLE algorithm for rigid water models. JComput Chem 13:952–962

32. Berendsen HJC, Postma HJC, Van Gunsteren WF, Hermans WF(1981) Interaction models for water in relation to proteinhydration. In: Pullman B (ed) Intermolecular forces. Reidel,Dordrecht, pp 331–342

33. Darden T, York D, Pedersen L (1993) Particle mesh Ewald—an N.log(n) method for Ewald sums in large systems. J Chem Phys98:10089–10092

34. Essmann U, Perera L, Berkowitz ML, Darden T, Lee H, PedersenLG (1995) A smooth particle mesh Ewald method. J Chem Phys103:8577–8593

35. Luty B, Tironi IG, van Gunsteren WF (1995) Lattice–summethods for calculating electrostatic interactions in molecularsimulations. J Chem Phys 103:3014–3021

36. Zimmerman K (1991) All purpose molecular mechanics simulatorand energy minimizer. J Comput Chem 12:310–319

37. Ferguson DM (1995) Parameterization and evaluation of a flexiblewater model. J Comput Chem 16:501–511

38. Berendsen HJC, Postma JPM, DiNola A, Haak JR (1984)Molecular dynamics with coupling to an external bath. J ChemPhys 81:3684–3690

39. Hess B, Bekker H, Berendsen HJC, Fraaije JG (1997) E.M.LINCS: a linear constraint solver for molecular simulations. JComput Chem 18:1463–1472

40. Kampmann T, Mueller DS, Mark AE, Young PR, Kobe B (2006)The role of histidine residues in low-pH-mediated viral membranefusion. Structure 14:1481–1487

41. Teeter MM, Froimowitz MF, Stec B, Durand CJ (1994)Homology modeling of the dopamine D2 receptor and its testingby docking of agonists and tricyclic antagonists. J Med Chem37:2874–2888

42. Neve KA, Cumbay MG, Thompson KR, Yang R, Buck DC, WattsVJ, Durand CJ, Teeter MM (2001) Modeling and mutationalanalysis of a putative sodium-binding pocket on the dopamine D2

receptor. Mol Pharmacol 60:373–38143. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,

Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN, BurantJC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B,Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, HadaM, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M,Nakajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, KnoxJE, Hratchian HP, Cross JB, Adamo C, Jaramillo J, Gomperts R,Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C,Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P,Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, StrainMC, Farkas O, Malick DK, Rabuck AD, Raghavachari K,Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cioslowski

430 J Mol Model (2012) 18:419–431

J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I,Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY,Nanayakkara A, Challacombe M, Gill PMW, Johnson B, ChenW, Wong MW, Gonzalez C, Pople JA (2003) Gaussian 03,Revision B.05. Gaussian Inc, Pittsburgh

44. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM,Meng EC, Ferrin TE (2004) UCSF Chimera—a visualization systemfor exploratory research and analysis. J Comput Chem 25:1605–1612

45. Hjerde E, Dahl SG, Sylte I (2005) Atypical and typicalantipsychotic drug interactionswith the dopamine D2 receptor.Eur J Med Chem 40:185–194

46. Cho W, Taylor LP, Mansour A, Akil A (1995) Hydrophobicresidues of the D2 dopamine receptor are important for bindingand signal transduction. J Neurochem 65:2105–2115

47. Cox BA, Henningsen RA, Spanoyannis A, Neve RL, Neve KA(1992) Contributions of conserved serine residues to the interactionsof ligands with dopamine D2 receptors. J Neurochem 59:627–635

48. Wiens BL, Nelson CS, Neve KA (1998) Contribution of serineresidues to constitutive and agonist-induced signaling via theD2Sdopamine receptor: evidence for multiple agonist-specificactive conformations. Mol Pharmacol 54:435–444

49. Wilcox RE, Huang WH, Brusniak MYK, Wilcox DM, PearlmanRS, Teeter MM, Durand CJ, Wiens BL, Neve KA (2000)CoMFA-based prediction of agonist affinities at recombinant wildtype versus serine to alanine point mutated D2 dopaminereceptors. J Med Chem 43:3005–3019

50. Page CS, Bates PA (2008) Can MM-PBSA calculations Predictthe specifiicties of protein kinase inhibitors? J Comput Chem27:1990–2007

51. Mertz KM (2010) Limits of free energy computation for protein–ligand interactions. J Chem Theor Comput 6:1769–1776

52. Breneman CM, Wiberg KB (1990) Determining atom-centeredmonopoles from molecular electrostatic potentials. The need forhigh sampling density in formamide conformational analysis. JComput Chem 11:361–373

53. Politzer P, Murray JS, Clark T (2010) Halogen bonding: anelectrostatically-driven highly directional noncovalent interaction.Phys Chem Chem Phys 12:7748–7757

54. Bader RFW (1990) Atoms in molecules. A quantum theory.Oxford University Press, Oxford

55. Popelier PLA (1999) Atoms in molecules. An introduction.Pearson, Harlow

J Mol Model (2012) 18:419–431 431

2-(3chloro-4-hydroxyphenyl)ethylamine

acting as ligand of the D2 dopamine receptor. Molecular modelling,

synthesis and D2 receptor affinity.

Emilio Angelina a, Laura Moreno

b, Francisco Garibotto

,a.c, Javier Párraga

b , Nelida Peruchena

d,

Nuria Cabedoe, Margarita Villeco

f Diego Cortes

b, Sebastian Andujar

a,c* and Ricardo D. Enriz

a,c,*

a Instituto Multidisciplinario de Investigaciones Biologicas (IMIBIO-SL -CONICET), Chacabuco 915, 5700 San Luis,

Argentina.

b Departamento de Farmacología, Laboratorio de Fármacoquímica, Facultad de Farmacia, Universidad de Valencia,

46100 Burjassot, Valencia, España.

cFacultad de Química, Bioquímica y Farmacia, Universidad Nacional de San Luis, Chacabuco 915, 5700 San Luis,

Argentina.

d Laboratorio de Estructura Molecular y Propiedades, Área de Química Física, Departamento de Química, Facultad de

Ciencias Exactas y Naturales y Agrimensura, Universidad Nacional del Nordeste, Avda. Libertad 5460, (3400)

Corrientes, Argentina.

e Centro de Ecología Química Agrícola-Instituto Agroforestal Mediterráneo, Universidad Politécnica de Valencia,

Campus de Vera, Edificio 6C, 46022 Valencia, España,

f Instituto de Química Orgánica, Facultad de Bioquímica Química y Farmacia, Universidad Nacional de Tucumán,

Ayacucho 471, S. M. de Tucumán, T4000INI. Argentina

__________________________________________________________________________

* To whom correspondence should be addressed.

Phone: (54) 26664423789; E-mail: [email protected]; [email protected]


Abstract

A molecular modeling study of 2-(3chloro-4-hydroxyphenyl)ethylamine and structurally related

compounds was carried out. By combining molecular dynamics simulations with semiempirical

(PM6), ab initio and density functional theory (DFT) calculations, a simple and generally applicable

procedure to evaluate the binding energies of these ligands interacting with the D2 Dopamine

Receptors (D2-DR) is reported here, providing a clear picture of the binding interactions of these

compounds from both structural and energetic viewpoints.

A reduced model for the binding pocket was used. This approach allows us to perform more

accurate quantum mechanical calculations as well as to obtain a detailed electronic analysis using

Quantum Theory of Atoms in Molecules (QTAIM) technique. Thus, molecular aspects of the

binding interactions between ligands and D2-DR are discussed in detail. A good correlation between

the relative binding energies obtained from theoretical calculations and experimental IC50 values

was obtained, predicting that -(3chloro-4-hydroxyphenyl) ethylamine posses a significant affinity

by the D2-DR. Thus, our theoretical predictions were experimentally corroborated when we

synthesize and test 2-(3chloro-4-hydroxyphenyl) ethylamine which displayed a similar affinity by

the D2-DR to that reported for DA.

Introduction

In the past the potential clinical usefulness of centrally acting dopamine (DA) receptor agonists has

stimulated intense research on new dopaminergic agents. Thus, in general, different efforts in

medicinal chemistry and neuropharmacology have yield substantial numbers of compounds with

activity and selectivity at each of the major DA (dopamine) receptors [1-6]. However, the design

and development of DA receptor-selective ligands remains largely empirical, quite conservative in

following molecular precedents, and somewhat unpredictable and not ready for routine applications

of computer-aided drug design techniques. The computational limitations and/or errors can arise in

many ways depending on the methodology: some possibilities include errors in the structure of the

host, choice of ionization state, structure of the complex, inadequate sampling of internal degrees of

freedom and the so called “ligand-receptor stereo-electronic problem”, which has received

relatively little attention in this case, at least in comparison with the other aspects.

In order to obtain new dopaminergic agonists, The DA molecule (compound 1 in figure 1) has been

modified mainly on the amino group and on the ethylamine chain, but only seldom on the catechol

moiety [7-15]. Claudi et al reported the synthesis and binding affinity for D1 and D2 subtypes of

DA receptors of 2-(4fluoro-3-hydroxyphenyl)ethylamine (compound 7, figure 1)[14]. This

compound showed about two-fold lower affinity than DA for both binding sites. Previously

Cardinelli et all reported low activity for other fluorine derivatives of DA including 2-(3fluoro-4-

hydroxyphenyl)ethylamine (compound 6) [13]. In fact relatively few examples can be found in the

literature on the replacement of catechol hydroxyl groups by chlorine on DA derivatives. Studies on

aminotetralins showed that 2-amino-6-chloro-7-hydroxytetralins (compound 8) are weakly effective

in the binding assays [16]. In addition Claudi et al reported that 2-(4-Chloro-3-

hydroxyphenyl)ethylamine (compound 2) is not able to discriminate between the two subtypes of

DA receptors (D1 and D2), and has seven-fold lower affinity than DA for both sites.

It must be pointed out that although 3-Chloro-4-hydroxy-β-phenylethylamine has been previously

reported by Chung et al [17] and Nakazawa et al [18]; however, according to our bibliographic

information, the dopaminergic effect of 2-(3chloro-4-hydroxyphenyl) ethylamine (compound 3) has

not been analysed yet. One possible explanation for this is that due to the low activity obtained for

the analogues of this compound, compound 3 has been directly discarded as a potential ligand of

interest for the DA D2 receptor.

It is interesting to note that in the series of benzazepines the 2,3,4,5-tetrahydro-7,8-dihidro-3-

methyl-1-phenyl-1H-3-benzazepine (compound 9) had only micromolar affinity at D1 and D2

receptors [19]. However 2,3,4,5-tetrahydro-7-chloro-8-hydroxy-3-methyl-1-phenyl-1H-3-

benzazepine (compound 10), obtained by replacing the 7-OH of 9 with a chlorine, has nanomolar

affinity possessing a high selectivity by D1 receptors. In the same way our own studies performed

on a series of benzyl-tetrahydroisoquinolines (BTHIQs) indicated that the presence of a chlorine

group at C7 is very important for the adrenergic effect of such compounds (compounds 11-14) [20-

22]. In fact BTHIQs possessing a chlorine group at C7 displayed the strongest affinity for both D1

and D2 DA receptors in this series [21]. It should be noted that the inductive effect of chlorine could

influence the acidity of the phenolic group and produce different interactions at the binding pocket

altering the affinity for D2-DR receptor. One question which arises is if compound 3, which has not

been previously studied, effectively have low affinity for D2-DR and in that case try to find an

explanation at the sub-molecular level for such behaviour. A question which might arise is why to

study compound 3 and why now? There are different reasons to analyze this molecule with the

computational approaches now available. On one hand it is possible to exploit prior information

obtained for dopamine respect to its biologically relevant conformation [23]. On the other hand is

also now possible to study with some accuracy the electronic effects introduced to the ligand when

replacing an OH by Cl and the effects of such change when the ligand is interacting with its

biological receptor.

Recently we reported a comprehensive conformational study of DA interacting with the D2-DR D2,

using a combination of MD simulations, semiempirical and DFT calculations [23]. In addition, a

detailed electronic analysis using Quantum Theory of Atoms in Molecules (QTAIM) [24-26]

technique was also carried out. By using this approach it is possible to evaluate with some details

the conformational and electronic behaviours of DA and its halogenated derivatives.

Halogens, especially the lighter fluorine and chlorine, are widely used substituents in medicinal

chemistry. Until recently, they were merely perceived as hydrophobic moieties and Lewis bases in

accordance with their electronegativities. Much in contrast to this perception, compounds

containing chlorine can also form directed close contacts of the type R−X···Y−R′, where the

halogen X acts as a Lewis acid and Y can be any electron donor moiety. This interaction, referred to

as “halogen bonding” since 1978 [27] is driven by the σ-hole, a positively charged region on the

hind side of X along the R−X bond axis that is caused by an anisotropy of electron density on the

halogen. [28,29]. The intriguing formal similarity between halogen bonding, R–X---B, and

hydrogen bonding, R–H---B, has been noted and discussed in detail by Legon; [30,31]. However, he

also does point out that the hydrogen bond is more likely to be nonlinear. In turn a halogen bond is a

highly directional, electrostatically-driven non-covalent interaction between a region of positive

electrostatic potential on the outer side of the halogen X in a molecule R–X and a negative site B,

such as a lone pair of a Lewis base or the p-electrons of an unsaturated system. In a very elegant

paper Politzer et al have introduced the “σ-hole concept” allowing to explain the empirically-

observed characteristics of halogen bonding: its marked directionality, its dependence upon the

polarizability and electronegativity of the halogen atom, and the role of the electron-withdrawing

power of the R portion of any molecule [32].

In a recent review on different types of protein−ligand interactions relevant to medicinal chemistry,

[33] the authors conclude that halogen bonds are a useful addition to the arsenal of favorable

interactions in molecular recognition and can lead to significant affinity gains in some cases. There

are a number of experimental studies where the effect of halogen substitution on binding affinity

has been systematically evaluated. [34-36]. The strengths of halogen bonds can be evaluated

theoretically through quantum chemical model calculations. It is evident that halogen bonding is

best described theoretically using high-level quantum chemical methods such as coupled cluster

[37] (CCSD-(T)) and perturbation theory [38,39] (MP2) calculations; yet using these methods

greatly limits the size of the model systems amenable to computational studies. Thus, much larger

systems (like those studied here) can be treated using QM/MM calculations [40, 41], semiempirical

studies [42] or using reduced model systems [23].

It is well known that the binding of DA to the receptor is substantially affected by multiple

serine/alanine mutations. The multiple mutations including a Ser 193 susbstitution produce the

greatest effect [43,44]. It has been previously reported that the effects of the multiple mutations

were not additive, with the single serine mutation having relatively larger effects, which is

indicative of a very precise network of hydrogen bonds between the TMV (transmembrane

spanning region) serine residues and the catechol hydroxyls of the DA molecule [45]. A further

interesting finding was that addition of a 4-hydroxyl group (p-tyramine) to beta-phenylethylamine

(compound 4) does not enhance affinity, but addition of a 3-hydroxyl group (m-tyramine)

(compound 5) is favourable. When the 3-and 4- hydroxyls are present (the case of dopamine), the

affinity is enhanced over the effects of the individual hydroxyl groups. These results are consistent

with a productive and specific interaction of the two hydroxyl groups of DA with a network of

serine residues. Our previous results [23] suggested that the region near these serine residues may

be rather mobile which is consistent with a model of receptor function where the binding of DA

locks the receptor into a relatively rigid conformation with precise interactions between ligand and

receptor.

On the other hand it is well-known that the non covalent interactions generally are weaker than the

covalent ones; such interactions are more difficult to describe properly. However, recent advances

in computational calculation of the electron charge density make possible the proper description of

the three-dimensional network of bonding and non bonding interactions in biological systems [23,

46-48] in the context of the quantum theory of atoms in molecules (QTAIM) [49]. Starting with

strong and moderated hydrogen bonds or halogen bonds, moving on to weaker polar interactions

and ending with stacking and T-shape interactions between aromatic rings, all of them can be

evaluated by QTAIM analysis [50]. In fact, nowadays it is well know that the stacking amino acids

aromatic rings in proteins is evidently much more important than it has been previously believed

and, indeed, can form one of the dominant stabilizing contributions [26]. Our previous results show

that the interactions of the catechol OH groups of the ligand, in the different conformations of the

DA/D2D-R complex, determine the decrease or increase of the electron density on the aromatic ring

of dopamine. In turn, the electronic population of the aromatic ring of dopamine defines its

orientation within the binding site and the type of interactions that are established with the aromatic

rings of the receptor. Needless to say that a description non quantum mechanical of the problem

would overlook these electronic effects that are crucial to understand the binding modes of the

ligand within the receptor binding site. Taking advantage of structural information available from

different experimental and theoretical studies of DA at the D2-DR) [23,43-44], we reported here a

comparative study among the complexes of compounds 1-5 with the D2D-R. It is clear that an

exhaustive analysis of the different stabilizing and destabilizing interactions for ligands possessing

different substituents at the 3- and 4- position of the catechol moiety would be of paramount

importance to determine the intricacies of this network of serine residues. Thus, we chose to study

compounds 1-5 in the molecular modelling study. It should be noted that these compounds were

carefully selected in function of their substituents at the catecholic portion. The selected compounds

were: compounds 4 and 5 possessing only one hydroxyl group at para and meta position,

respectively and compounds 2 and 3 possessing chlorine and hydroxyl groups at para and meta

positions, respectively. We also include in our comparative study DA (compound 1), possessing the

catecholic ring. Thus, the comparison between the conformational and electronic behavior of 1 with

structurally related compounds and the degree of similarity between the different complexes will

provide an indication of the physics of binding. The challenge is to understand the structural and

energetic differences between these complexes so as to successfully design new molecules that

adopt the desired shape with little or no energetic penalty.

First we performed MD simulations for the different complexes formed between compounds 1-5

with the D2-DR. The next step was to construct a reduced model for the binding site of this

receptor. The third step was to simulate the molecular interactions between the different ligands and

the D2D-R using a QTAIM analysis; being the principal goal of such calculations try to obtain a

detailed description of the molecular interactions which stabilize and destabilize the different

complexes. Finally in order to obtain an experimental corroboration we synthesized compound 3

and we tested its affinity for the D2-DR; a good correlation between our theoretical and

experimental results was obtained

Results and Discussion

Molecular Modelling

Our study was carried out in four steps. In the first one we performed docking simulations using the

Autodock 4.0 program in order to localize the different ligands in the binding pocket. The obtained

complexes were used as starting structures for the MD simulations. Next we performed molecular

dynamics simulations evaluating the molecular interactions between compounds 1-5 with the D2D-

R (figures 2-5). In the third step, reduced model systems were optimized using quantum mechanics

calculations. Semiempirical (PM6) calculations were employed in these optimizations. The most

representative complexes obtained in the previous steps were further analysed from a QTAIM study

using DFT calculations.

Molecular dynamic simulations

Comparing the results obtained for the different complexes, interesting general conclusions might

be obtained. Consistent with previous experimental [43-45] and theoretical [22,23] results, our

simulations indicate the importance of the negatively charged aspartate 114 for the binding of these

ligands. It is well-known that a highly conserved aspartic acid (Asp 114) in trans-membrane helix 3

(TM3) is important for the binding of both agonists and antagonists to the D2D-R [23, 51,52]. In

the present study, all the compounds simulated were docked into the receptor with the protonated

amino group near Asp 114. After 10 ns of MD simulations, the ligands had moved slightly but in a

different form compared with the initial position. However, the strong interaction with Asp114 was

maintained for all the complexes (see figures 2-4 and 1S and 2S in supporting information) which

support that Asp 114 is the anchoring point for ligands with a protonated amino group.

In the next step of our study we evaluated the relative free energy (ΔΔG Gbind) obtained for

the different complexes (table 1). From the relative binding energies obtained in our MD

simulations, one can appreciate that replacing the OH in meta by a chlorine atom increases the

affinity for the D2D-R in comparison to compound 1, indicated by the lowest value of the relative

free energy obtained for compound 3. In contrast replacing the OH in para position displays the

opposite effect, ie decreases the affinity for the receptor (5.57 Kcal/mol above the value obtained

for 3). This was a surprising result because we did not expect that compound 3 presents a relative

binding energy as low indicating good affinity for the D2 receptor. The results obtained for

compounds 4 and 5 were as expected since both compounds showed a lower affinity for the D2DR

than compound 1. Compound 4 displayed lower affinity than compound 5 which is also in

agreement with the experimental data. To try to better understand these results we compare the

results obtained for 1/D2D-R complex with those complexes obtained for compounds 2-5.

Figures 2 and 3 show the structures obtained for the complexes of compounds 3 and 2,

respectively; the complex obtained for 1/D2-DR was superimposed in both figures for comparison.

The superposition of the complexes obtained for compounds 1 and 3 (see Figure 2) shows that both

structures are very similar in terms of the orientation of the ligand in the receptor binding site. In

both complexes the ethylammonium side chain is located in a staggereded conformation (torsional

angle 1=171 ° and 177 °, respectively) and therefore the aromatic ring of compound 3 overlaps

almost perfectly with the catecholic ring of 1, with the difference that this ring is rotated 180° with

respect to the DA ring. In contrast overlapping the complex obtained for compound 2 with DA (see

Figure 3) shows a marked deviation from the p-Cl position in the binding site, with respect to DA.

Unlike compounds 1 and 3, the side chain of ethylammonium of compound 2 is located in a biased

conformation with a twist angle of 97°. It should be noted that in Figures 2 and 3 in both

chlorinated isomers, the chlorine atom is located in an hydrophobic environment composed by

residues Ile184, Phe389, Val190 and bonds C--H de His393; in addition, the chlorine atom of

compound 3 establishes other interactions with residues Ser193 and Ser194.

In the case of 3/D2-DR complex, rotation of its aromatic ring at an angle of 180° with

respect to compoud 1 allows anchorage of the chlorine atom into the hydrophobic region of the

binding site, keeping the rest of the structure in same position as 1. Whereas in compound 2 the

chlorine atom replaces the OH at para position and therefore it is necessary a change of the torsional

angle of the side chain of ethylammonium in order to allow the chlorine atom might be anchored in

the hydrophobic pocket. This behavior determines a different spatial orientation to those observed

for compounds 1 and 3 and therefore a different pattern of interactions too.

Observing Figures 2 and 3 one can see that in complexes of compounds 1 and 3 the side-

chain of Ser197 interacts with Thr119 in TM3. In contrast, in 2/D2D-R complex this interaction

breaks and the side chain of Ser197 is associated with the backbone of Ser193, causing distortion of

the backbone interactions of TM5 and therefore the change in the structure of the transmembrane

domain (see figure 2S (supporting information). It is worth noting that the change that has this

domain in the complex 2/D2-DR reflects the importance of preserving the flexibility of the

backbone of TM5 during the simulation, as suggested by other authors [53,54].

Possible role of the aromatic residues of the binding site

Figures 1S and 2S give a spatial view of the same complexes shown in figures 2 and 3 but in

this case showing the spatial distribution of the aromatic residues of the binding site. Figure 1S

shows that the side chains of Phe389, Phe390 and Phe198 overlap almost perfectly in both

complexes 1/D2D-R and 3/ D2D-R. These three residues are linked together by interactions type C-

H...π (some of these interactions are shown in the molecular graphs later) forming a conserved

structure that has been observed also in D2D-R complexes with alkaloids possessing a

tetrahidroprotoberberine structure [47]. This conserved structure forms specific interactions with the

ligand, contributing, together with the principal interaction Asp114 to its anchoring in the proper

orientation within the binding site. Moreover, in the complex 2/D2-DR the side chains of

phenylalanine triad show a marked deviation in the position of atoms relative to 1/D2DR (see

Figure 2S). This alteration in the phenylalanine triad structure is a consequence of the distortion

undergoes by TM5 in the complex, which produces a displacement of the Phe198 residue so that it

can no longer be associated with Phe190. Since phenylalanine triad produces specific interactions

with the ligand altering its structure, such interactions necessarily affect the mode of binding of the

ligand binding site, this being another factor that helps to explain the change in orientation of

compound 2 within the binding site with respect to compounds 1 and 3.

Figure 4 gives the structures obtained for the complexes of compounds 4 and 5. As shown in

this figure, both ligands (compounds 4 and 5) act as proton donor against Ser197. In addition the

OH group of compound 5 provides other interactions with Cys118 and Thr119. In contrast, the OH

group of compound 4 provides no significant additional interaction other than the hydrogen bond

with Ser197. Note that in the latter complex, the hydrophobic side chain of Ile184 is interposed

between the OH group of the ligand and the side chains of Ser193 and Ser194, avoiding the

interactions with these serine residues. Thus, the OH group of compound 5 is anchored in a region

of the binding site where it could form a greater number of interactions which might explain that

this isomer is associated with higher affinity to the D2D-R than compound 4.

On the basis of these results, one might think that the compound 3 would have a D2 receptor

affinity at least similar to that of 1. However, we must bear in mind that these results have been

obtained by MD simulations and therefore some caution must be taken with them. Any model that

neglects or else poorly approximates the terms that are playing determinant roles like for example

lone pair directionality in hydrogen bonds, explicit п-stacking polarization effects, hydrogen

bonding networks, induced fit, conformational entropy among others cannot reasonably be expected

to distinguish between compounds possessing relatively similar binding energies. There are in the

literature several works supporting this concept [55-56].

At this stage of our work we consider the trend predicted for the MD simulations as certainly

significant but, on the other hand they might be reluctant to give a quantitative significance, because

of the approximations involved in this mode of approach. It should be noted that we are dealing

with relatively weak interactions and therefore MD simulations might underestimate such

interactions. Thus, in the next step of our study we optimized reduced model systems using

combined semiempirical and DFT calculations.

Constructing the reduced models for the binding site

The use of model systems to calculate and simulate molecular interactions (MI) is necessary

since the ligands interacting at the active site of D2/DR are a molecular system too large for

accurate QM MO calculations. Moreover a model system representing the D2D-R binding pocket

may be desirable in order to evaluate the ability of the ligands to interact with the active site. By

using a model, one avoids dealing with complexities due to the rest of the D2D-R. Thus a better

understanding of the inherent electronic properties of the ligands reflected in the MI may be gained.

When choosing a model system, the ability to reproduce electronic and hydrophobic properties of

the ligands is crucial. Thus, to acquire a more-detailed insight into the mechanisms driving the

bindings of ligands to the active site of D2D-R, the structure-affinity relationship was analysed. The

information obtained from these calculations is very important for quantitative analyses and is

highly useful to the understanding of the binding mechanism.

Figure 5 shows the ligand-residue interaction spectra calculated by the free energy decomposition,

which suggests that the interaction spectra compounds 1-5 with D2D-R are closelly related and

reflects their similar binding modes. As shown in Figure 5, the residues Asp114 and Val115 are

those with the greatest contribution to the interaction energy, this is true not only for compounds 1-

5, but for other compounds with dopaminergic activity previously reported [47]. Since the salt

bridge with Asp114 is the strongest interaction established in the binding site is considered a

guideline interaction of the ligand binding site of D2D-R. With respect to Val115, this aminoacid

forms several interactions type C-H...π with the aromatic ring of DA. These interactions are not

shown in graphs for easy viewing molecular interactions of meta-OH/Cl and para-OH/Cl. On the

other hand among the serine residues of the binding site, Ser197 is the residue that has the largest

contribution to the interaction energy in complexes of compounds 1 and 2, being higher in the first,

while Ser193 presents the most important contribution in the complex of compound 3. Finally, the

residue decomposition analysis also shows a significant contribution of one of the residues of the

phenylalanine triad, Phe389, its contribution being greater in 1/D2D-R and 3/D2D-R than in

2/D2D-R.

From these results we considered prudent to include in the reduced model not just those aminoacids

involved in the most relevant molecular interactions displayed in the different spectra, but also to

include all the residues involved in stabilizing and destabilizing interactions displaying more than 1

Kcal/mol. Thus, residues Asp114, Val115, Met116, Cys118, Thr119, Ile184, Phe189, Val190,

Val191, Tyr192, Ser193, Ser194, Ile195, Val196, Ser197, Phe198, Trp386, Phe389, Phe390, His393,

Tyr408, Thr412 and Tyr416 were included in the reduced model for the binding pocket of D2D-R

and therefore a final number of 23 aminocids were included in our model. A spatial view of this

reduced model is shown in figure 6.

Quantum Mechanics Calculations and topological analysis of the electron density

The starting geometries for each complex were obtained from the coordinates of the energy

minimum during the simulation time. PM6 optimizations were performed by considering all

receptor amino acids that could interact after initial positioning of the ligands against Asp114

residue. Next DFT (PBE/6-31G(d,p) single point calculations were carried out for each complex

optmized with PM6 computations.

Evaluating the molecular interactions for the different complexes

The topological analysis of the electron density constitutes a powerful tool to investigate the

electronic properties of the molecular system and allows a deep examination of the molecular

interactions. This methodology has been successfully applied in the study of the properties of a

variety of conventional and unconventional HBs, aromatic HBs as well as π−π stacking [24-26].

From the QTAIM it is possible to determine in an unequivocal way the different strong and weak

interactions between two atoms observing the existence of Bond Critical Points (BCPs) and their

respective bond paths. It should be noted that this detailed analysis is not possible from the

evaluation of the geometrical parameters (bond distances and angles).

Figures 7, 8 y 9 show the molecular graph of electron density obtained for compounds 1, 3

and 2 in the D2R binding site, respectively. These molecular graphs allow to visualize critical points

and bond paths that connect the atoms of the ligand with the receptor residues. The molecular graph

obtained for compounds 4 and 5 are shown in figures 3S and 4S as supporting information; a brief

discussion for each figure is given in the same section and such results are in full agreement with

those obtained for compounds 1-3.

Catechol interactions

The molecular graph of Figure 7 shows the most important interactions of compound 1 with

the residues of the binding site. In particular, it is observed that OH in the meta position (m-OH)

establishes a strong hydrogen bond (HB) type O-H....S with Cys118 (b = 0.0202 au), other weaker

HBs type C-H....O with Phe390 (b = 0.0023 au), Ser197 (b = 0.0065 au) and Val115 (b =

0.0032 au) and two contacts type O....O with Thr119 (b = 0.0061 au) and Ser197 (b = 0.0212

au). While in position para the OH acts as a proton donor against Ser197 (b = 0.0212 au), and as

proton acceptor versus C-H bonds of Phe390 (b = 0.0034 au). Furthermore, like m-OH, p-OH also

establishes a contact type O ... O with Ser193 backbone (b = 0.0027 au).

It is important highlight that exist three Odonor...Oacceptor interactions in the 1/D2DR complex.

Even more interesting, the oxygen in meta position (m-O) join up to the hydroxylic oxygen of

Ser197 and Thr119; the latter two are connected through a strong hydrogen bond (OH .. O). Thus,

the three oxygen atoms are connected directly or indirectly by bond paths to give a topological ring

(see figure 7).

Interactions of the chlorine atom

As seen in Figure 8, the chlorine atom of compound 3 (in meta position) establishes six

interactions with the binding site of D2D-R. Three interactions type C-H.... Cl with residues Ile184,

Val190 and His393 (Σ b = 0.0149 au), two interactions type O.... Cl with Ser193 (b = 0.0128 au)

and Ser194 (b = 0.0013 au) and a strong intramolecular HB interaction type (O-)H...Cl with the

OH in para position (b = 0.0240 au). These two Cl .. O interactions could explain the favorable

binding behavior observed for compound 3. In other words, the behavior of compound 3 in the

binding site mimics the behavior of dopamine which establishes several O....O interactions with

carbonyl and hydroxyl groups of their biological receptor.

Moreover, in the complex of compound 2, the chlorine atom is still more "buried" in the

hydrophobic region (see Figure 9) forming only HBs interactions type C-H...Cl (a total of six

interactions with residues Ile184, Phe189, Val190 and His393 (Σ b = 0.0326 au)). Note that in this

complex the chlorine atom is not forming an intramolecular hydrogen bond with m-OH as in the

case of compound 3.

In short and as was discussed previously, in 1-D2DR complex there are several O…O interactions

between the catecholic and receptor carbonyl/hydroxyl groups. This kind of contacts have been

already described in previous reports (23) and in such sense we believe they could be involved in

the ligand/receptor recognition process. It must be remarked that when, the meta OH of dopamine is

replaced by a chlorine atom, some of these O…O contacts are replaced by Cl...O interactions that

show a similar behavior than the first ones.

Effects of chlorine substitution on the interactions of the catecholic OH

Figure 10 gives the values of b summation (Σ b) corresponding to the interactions

between different groups of compounds 1-5 in the D2R binding site. In this figure we only show the

values of Σ b corresponding to the interactions of the substituents in the meta (m-OH/Cl) and para

position (p-OH/Cl). One can see in this figure that when replacing the meta position of 1 by a

chlorine atom, it can not be anchored to the binding site with the same force as the m-OH of 1.

However, the introduction of Cl at meta position improves the interactions of p-OH with respect to

the same group in 1. Thus, both groups together, m-Cl and p-Cl, achieve to be anchored to the

binding site with almost the same strength as the corresponding catecholic hydroxyls of 1, as shown

in Figure 5a.

Figure 2 shows that the position of p-OH in the binding site is essentially the same in both

complexes, 1/D2D-R and 3/D2D-R; where one would expect to find the same pattern of

interactions. However, the presence of the chlorine atom in the meta position modifies the binding

site in order to strengthen interaction of p-OH.

The molecular graph of Figure 8 shows the most important interactions obtained for

compound 3 in the D2R binding site. It is observed that p-OH acts as proton donor with the

hydroxyl oxygen of Ser193 ( b = 0.0200 au) and simultaneously forms HBs type C-H....O ─ with

Phe390 and Ser197 (Σ b = 0.0180 au). Moreover, in the complex 1/D2D-R, Ser193 is literally

"trapped" between the backbone of TM5 and the ring of Phe189, forming strong interactions Type

O-H.... with the aromatic ring of Phe189 that prevent the association with 1 (see Figure 2).

Something similar happens in compound 2/D2D-R complex (see Figures 3 and 9).

There are two changes which produce m-Cl on the binding site, which allows the interaction

of p-OH with Ser193:

(i) As shown previously, m-Cl forms a strong interaction type O...Cl with Ser193 that would

explain per se the conformational change experienced by this residue to pass from

1/D2D-R to 3/D2D-R which enables training of hydrogen bond with p-OH.

(ii) On the other hand, it was seen that m-Cl also forms the same type of interaction O....Cl with

Ser194, which contribute to the conformational change of Ser193 indirectly through the

following mechanism (see Figure 2) : (i) When the DA/D2D-R complex pass to the

3/D2D-R complex, the side chain of Ser194 is oriented towards m-Cl as a result of the

O....Cl interaction, (ii) since the OH of Ser194 is associated with the backbone of Val190

forming an HB type O-H....O=C, the conformational change of Ser194 is followed by a

sliding of the backbones of Phe189 and Val190 in a counterclockwise, attempting to

preserve the O-H....O=C interaction. (iii) It should be noted that due to the sliding of the

backbone, the side chain of Phe389 is not properly oriented to interact with Ser193; in

fact this latter may interacts with p-OH, and m-Cl.

Returning to Figure 10, it can be seen that when we replace the hydroxyl atom of para

position of 1 by a chlorine, the halogen provides stronger interactions than the corresponding OH

group. However, the introduction of chlorine significantly weakens the interactions of m-OH in

comparison to the same group in 1/D2D-R complex, so that both groups together bind more weakly

to the binding site that the corresponding catecholic OH groups of 1.

As discussed above, in the complex 2/D2D-R the chlorine atom is more "buried" in the

hydrophobic pocket formed by residues Ile48, Phe154 and Val190 that in the 3/D2D-R complex.

This determines that the m-OH is poorly positioned to interact with the serines of the

binding site. As shown in Figure 3, in the 2/D2D-R complex, the hydrogen atom of the m-OH is

oriented towards Ser197, like p-OH in the 1/d2D-R complex. In fact, Figure 9 shows a critical point

of the corresponding link and link roads connecting the hydrogen atom of m-OH with the oxygen

atom of the hydroxyl group of Ser197. However, it should be noted that this interaction is very

weak, with a value of b of only 0.0002 au.

Observing figure 10 it is clear that the OH group of compound 5 is more strongly anchored

than the OH group of compound 4 in the binding site. Moreover, as seen previously the OH group

of compound 5 produces similar interactions to those observed for the m-OH group of compound 1;

similarly the pattern of interactions of compound 4 is closely related to those of the p-OH of 1.

However, although the pattern of interactions of the isolated OH groups is similar to those of the

corresponding OH groups of compound 1 it can be seen from Figure 10 that these OH groups are

more strongly associated when they are isolated than when they are simultaneously present in DA.

In other words, the OH groups of compounds 4 and 5 are more firmly anchored to the binding site

in comparison to the OH groups of 1. This finding is contrary to the assumption that the

simultaneous presence of both hydroxyls improves interactions they establish separately. Moreover,

it is also interesting to compare the interactions observed in the isolated OH groups with respect to

the OH groups in the same position in the chlorinated analogues. For example, introduction of a

chlorine atom in the para position of compound 5 to give compound 2 decreases drastically the

strength of the interactions of m-OH (see Figure 10). Conversely, introduction of a chlorine atom in

the meta position of compound 4 to give compound 3 slightly improves the interactions of OH in

the para position. As discussed previously, in both chlorinated compounds the chlorine atom is

anchored in the hydrophobic pocket formed by residues Ile184, Phe389 and Val190. The anchor in

the hydrophobic pocket affects more the orientation of the ligand in the 2/D2DR complex than in

the 3/D2DR complex, and therefore the interactions of the OH groups in the first complex is more

affected than the second one. Figures 3 and 4 show very well that the m-OH of compound 2 is in a

totally different environment with respect to the equivalent OHs of compounds 1 and 5.

Interactions with the aromatic rings of the binding site

Molecular graphs given in Figures 7, 8 and 9 show the interactions of the ligands with the

aromatic residues of the binding site. In the three complexes studied, Phe389 residue forms two C-

type interactions C-H... with the side chain of ethylammonium and the aromatic ring of the ligand.

Consistent with the results of the decomposition of interaction energy by residue, these interactions

are much stronger in the 1/D2D-R and 3/D2D-E complexes (Σ b = 0.0177 au and Σ b = 0.0167 au

respectively) than in the 2/D2D-R complex (Σ b = 0.0022 au). Moreover, another 3F aromatic

residues, Phe390, form HB interactions type C─H ....O with the OH groups of compounds 1 and 3,

but not with the m-OH of compound 2. In other words, compound 2 provides either weaker

interactions with the 3F residues or directly do not produce any, which is related to the distortion

observed in 3F for the compound 2/D2D-R complex.

Experimental corroboration for the theoretical calculations

Our theoretical results indicated that compounds 3 could possess a similar affinity for the

D2D-R to that previously reported for 1. Thus, we considered very interesting to synthesize and test

the binding affinity of this compound in order to corroborate our theoretical results.

Synthesis of 3-chloro-4-hydroxy-β-phenylethylamine (3)

3-Chloro-4-hydroxy-β-phenylethylamine (3) has been previously obtained by the Chung’s

method [17], which consists in the oxidative chlorination of phenols with HCl and m-chloro

perbenzoic acid (m-CPBA) in N,N-dimethyl formamide, and also using an enzymatic procedure

[18]. In the course of our work on the synthesis of 1-substituted-1,2,3,4-tetrahydroisoquinolines

[20], the availability of the N-(4-(benzyloxy)-3-chlorophenethyl)benzamide (1) allowed us to obtain

the 3-chloro-4-hydroxy-β-phenylethylamine (3) by hydrolysis of the benzamide bond. Initially and

unsuccessfully, various basic and acid conditions excessively strong for the hydrolysis of the

benzamide (1) were assayed. Finally, we used a hydrolysis reaction in milder conditions [57], a

mixture of 2.5 N HCl/HOAc (4:1) at reflux for 42 h, which afforded the 3-chlorinated dopamine

analogue (3) in good yield (scheme 1)

BindingAffinity of compound 3

In order to minimize the experimental errors due to the methodology employed we used a

very similar experimental protocol to that reported in reference [58] (see experimental section). Our

experimental measurements indicated that compounds 3 possess a pKi value of 6.83 M. It should

be noted that this compound displayed a significant affinity by the D2D-R, which is in a complete

agreement with our theoretical predictions. These experimental results are an additional support to

our predictive molecular modelling study. Even when there are numerical differences between the

experimental and theoretical values it is also true that the thoretical calculations predicted that the

affinity for the D2D-R of compounds 3 could be closely related to those displayed by DA (pKi =

6.25M ) [58]. It is clear that such a prediction was totally corroborated with our experimental

results.

Conclusions

It is noteworthy that a molecule relatively simple as compound 3 has not been tested before as a

possible ligand of the D2D-R. One possible reason is that in that moment when the structurally

related compounds were estudied, this compound was discarded without an exhaustive previous

analysis. It is also fair to note that at the time in which the analogues of compound 3 were reported

was not possible to perform theoretical calculations with a relatively detailed electronic analysis as

has been done in this work.

In this paper we performed a molecular modeling study of 2-(3chloro-4-hydroxyphenyl)ethylamine

and analogues. By combining MD simulations with semiempirical and DFT calculations, a simple

and generally applicable procedure to evaluate the binding energies of ligands interacting with the

D2D-R has been reported here which provides a relativelly clear picture for the binding interactions

of ligands from both structural and energetic points of view. Thus, our results give interesting

information which may be helpful in obtaining a better understanding of the molecular interactions

between these ligands and the D2D-R.

Considering that the level of confidence in the design of new compounds based on molecular

modelling has increased, it is important to emphasize that any modeled binding mode remains a

model until it is experimentally validated. This has been the main reason to synthesize and evaluate

the affinity of compound 3 for the D2-DR. Thus, our theoretical and experimental results contribute

to the understanding of the non covalent interactions in the context of the ligand - receptor binding

event in a two-way manner, by providing a detailed topological description of the interaction

network of the ligand in the receptor binding pocket and by showing the convenience of going

beyond the concept of pair-wise interactions in order to “see” the electronic effects within the

intricate biological environment. Undoubtedly the results presented here show the importance to

perform these studies as comprehensive as possible when hydrogen bonds are involved and even

more if there are halogen atoms involved in these interactions.

It is interesting to note that the multiplicity of interactions present in a single receptor-ligand

complex is a compromise of attractive and repulsive interactions that is very difficult to analyse in a

separate way. By focusing on observed interactions, one neglects a large part of the thermodynamic

cycle represented by a binding free energy: solvation processes, long-range interactions,

conformational changes. It is clear that a complex is not characterized by a single structure but by

an ensemble of structures. Changes in the degrees of freedom of both partners (ligand and receptor)

during the binding event have a large impact on binding free energy. Considering all these factors it

is clear that our calculations are still far from considering the above variables. In any case we can

only say is that our study was conducted as deep as the size of our biological system has allowed.

So some caution is required to not extend these results far beyond the type of molecular interactions

analysed here nor to another molecular systems in particular to those which are structurally different

to those studied here. However, we believe our results may be helpful in the structural identification

and understanding of the minimum structural requirements for these molecules and can provide a

guide in the design of new ligands for the D2 receptor of dopamine.

MATERIAL AND METHODS

Chromatographic and Spectroscopic Analysis

The reaction was monitored by analytical TLC with silicagel 60 F254 (Merck 5554). The residue was

purified through silica gel C-18 (SPE, Alltech, 100 mg/1.5 mL) column chromatography. Isolation

and purification was carried out by a Waters HPLC system with a 600 pump and both a 2996

Photodiode Array Detector (PDA) and ELSD 2420 Detector (Milford, MA). 1H, NMR spectra was

recorded on a Bruker AV 300 MHz instrument (Rheinstetten, Germany). Chemical shifts (δ) are

reported in ppm for a solution of the compound in CDCl3 and the coupling constants (J) values are

given in Hz. High resolution ESIMS (electrospray) data were carried out on a Micromass Q-TOF

MicroTM

coupled with a HPLC Waters Alliance 2695 (Milford, MA). The instrument was calibrated

by using a PEG mixture from 200 to 1000 MW (resolution specification 5000 FWHM, deviation <5

ppm RMS in the presence of a known lock mass). The HCl salts of the synthesized compounds

were prepared from the corresponding base with 5% HCl in MeOH. N-(4-benzyloxy)-3-

chlorophenethyl)benzamide was prepared by standard methods from 3-chloro-4-

methoxybenzaldehyde and benzoyl chloride [20].

Synthesis of 3-chloro-4-hydroxy-β-phenylethylamine (3)

A solution of the N-(4-benzyloxy)-3-chlorophenethyl)benzamide (1, 20 mg, 0.055 mmol) in a

mixture of 2.5 N HCl/ HOAc (4:1 mL) was heated for 42 h at 100 ºC. Then, the solvent was

removed under reduced pressure and the residue partitioned between H2O/EtOAc. The phases were

separated and the acid aqueous layer evaporated under reduced pressure to give a residue which was

purified by column chromatography C-18 (SPE, Alltech, 100 mg/1.5 mL) and eluted in a gradient

from 100 % H2O to 100 % MeOH. The first fraction eluted with H2O/MeOH (9:1) was evaporated

and purified by semi-preparative HPLC using a Tracer Excel 120 ODS-B C18 column, 5 µm (25.0

x 1 cm), and MeOH/H2O in 1% HOAc (20:80) as mobile phase with a flow of 2 mL/min. The 3-

chloro-4-hydroxy-β-fenilethylamine (7 mg, 0.041 mmol, 75% yield) was isolated with a retention

time (Rt) of 14.5 min as a white solid. 1

H NMR* (300 MHz, CDCl3) δ 7.20 (s, 1H, H-2), 6.9 (s, 1H,

H-6), 6.8 (s, 1H, H-5), 3.2 (m, 2H, CH2-α), 2.8 (m, 2H, CH2-β); ESMS m/z (%) 172 [M+1]+

(39),

155 (100).

Binding experiments.

These experiments were performed on striatal membranes. Each striatum was homogenized in 2 mL

ice-cold Tris-HCl buffer (50 mM, pH = 7.4 at 22 ºC) with a Polytron (4s, maximal scale) and

immediately diluted with Tris buffer. The homogenate was centrifuged either twice ([3H] SCH

23390 binding experiments) on four times ([3H] raclopride binding experiments) at 20000g for 10

min at 4 ºC with resuspension in the same volume of Tris buffer between centrifugations. For [3H]

SCH 23390 binding experiments, the final pellet was resuspended in Tris buffer containing 5 mM

MgSO4, 0.5 mM EDTA and 0.02% ascorbic acid (Tris-Mg buffer), and the suspension was briefly

sonicated and diluted to a protein concentration of 1mg/mL. A 100μL aliquot of freshly prepared

membrane suspension (100 μg of striatal protein) was incubated for 1h at 25 ºC with 100μL Tris

buffer containing [3H] SCH 23390 (0.25 nM final concentration) and 800μL of Tris-Mg buffer

containing the required drugs. Non-specific binding was determined in the presence of 30 μM

SK&F 38393 and represented around 2-3% of total binding. For [3H] raclopride binding

experiments, the final pellet was resuspended in Tris buffer containing 120 mM NaCl, 5 mM KCl, 1

mM CaCl2, 1 mM MgCl2 and 0.1% ascorbic acid (Tris-ions buffer), and the suspension was treated

as described above. A 200 μL aliquot of freshly prepared membrane suspension (200 μg of striatal

protein) was incubated for 1 h at 25 °C with 200 μL of Tris buffer containing [3H] raclopride (0.5

nM, final concentration) and 400 μL of Tris-ions buffer containing the drug under investigation.

Non-specific binding was determined in the presence of 50 μM apomorphine and represented

around 5-7% of the total binding. In both cases, incubations were stopped by addition of 3 mL of

ice-cold buffer (Tris-Mg buffer or Tris-ions buffer, as appropriate) followed by rapid filtration

through Whatman GF/B filters. Tubes were rinsed with 3 mL of ice-cold buffer, and filters were

washed with 3 x 3 mL ice-cold buffer. After the filters had been dried, radioactivity was counted in

4 mL BCS scintillation liquid at an efficiency of 45%. Filter blanks corresponded to approximately

0.5% of total binding and were not modified by drugs.

Molecular Modelling

A 3D model of the human D2 DR was used for the MD simulations. This model is based on

the homology model from the crystallized D3 DR, β2 adrenoceptor and Α2α adenosine receptor as

templates 49 [47,54]

Docking simulations and determination of binding affinity

The AutoDock4 program simulations were performed using rotatable bonds in the ligands and

rigid/rotatable side chains in amino acids of the fifth transmembrane domain in the receptor due to

its implication in the formation of binding pocket [59,60].

For that, Kollman united atom partial charges for all protein atoms, solvent parameters and non-

merge polar hydrogen’s were assigned; and the possible rotatable bonds, torsions and atomic partial

charges (Gasteiger) of the ligands were assigned using the AutoDock Tools 1.5.2 [60].

The ligands were then docked inside a cubic GRID box (70 x 70 x 70 A , grid points separated by

0.375A ) centered at the midpoint between the Asp114 and Ser194 alfa carbons (both residues

conserved at the D2DR putative binding site). This docking simulation was achieved under the

hybrid Lamarckian genetic algorithm, which had an initial population of 100 randomly placed

individuals and a maximum number of energy evaluations set at 1x107. The resulting docked

orientations within a root mean square deviation (RMSD) of 0.5 A were clustered together. The

lowest energy cluster returned by AutoDock4 for each compound was used for further affinity and

conformational binding analysis. All of the other parameters were maintained at their default

settings [60]. The visualizations were performed using the VMD 1.8.6 program [61].

Molecular dynamics simulations of complexes

The complex geometries from docking were soaked in boxes of explicit water using the

TIP3P model [62] and subjected to MD simulation. All MD simulations were performed with the

Amber software package40 using periodic boundary conditions and cubic simulation cells. The

particle mesh Ewald method (PME) [63,64] was applied using a grid spacing of 1.2 A˚ , a spline

interpolation order of 4 and a real space direct sum cutoff of 10A˚. The SHAKE algorithm was

applied53 allowing for an integration time step of 2 fs. MD simulations were carried out at 300 K

target temperature and extended to 10 ns overall simulation time. The NPT ensemble was employed

using Berendsen coupling to a baro/thermostat (target pressure 1 atm, relaxation time 0.1 ps). Post

MD analysis was carried out with program PTRAJ[65].

MM-GBSA free energy decomposition

In order to determine the residues of the D2 DR active site involved in the interactions, the

residues proposed by Andujar et al. [17] and Soriano-Ursua et al. were first identified.[54] Then

MM-GBSA free energy decomposition using the mm_pbsa program in AMBER[66] was employed

to corroborate the amino acids interacting with the ligands. This calculation can decompose the

interaction energies of each residue considering molecular mechanics and solvation energies.[67-

71] Each ligand–residue pair includes four energy terms: van der Waals contribution (Evdw),

electrostatic contribution (Eele), polar desolvation term (GGB) and nonpolar desolvation term

(GSA), which are summarized in the following equation:

For MM-GBSA methodology, snapshots were taken at 10 ps time intervals from the

corresponding last 1000 ps MD trajectories and the explicit water molecules were removed from the

snapshots.

Quantum mechanics calculations and topological study of the electron charge density

distribution

23 amino acids were included in this reduced model based on the generated data. Five

molecular complexes, 1/D2DR, 2/D2DR, 3/D2DR, 4/D2DR, and 5/D2DR, obtained for our “reduced

model system”, were selected due to their reprensentative chemical feautures for the calculation of

the charge density. Single point calculations were performed with Gaussian 03[72] and employing a

hybrid PBE functional and 6-31G(d) as basis set. This type of calculation has been recently used in

studies on the topology of ρ(r) because it ensures a reasonable compromise between the wave

function quality required to obtain reliable values of the derivatives of ρ(r) and the computer power

available, due to the extension of the system in study. The topological properties of a scalar field

such as ρ(r) are summarized in terms of their critical points, i.e., the points rc where Δρ(r) = 0.

Critical points are classified according to their type (ω,σ) by stating their rank, ω, and signature, σ.

The rank is equal to the number of nonzero eigenvalues of the Hessian matrix of ρ(r) at rc, while the

signature is the algebraic sum of the signs of the eigenvalues of this matrix. Critical points of (3, -1)

and (3, +1) type describe saddle points, while the (3, -3) is a maximum and (3, +3) is a minimum in

the field. Among these critical points, the (3,-1) or bond critical points are the most relevant ones

since they are found between any two atoms linked by a chemical bond. The determination of all

the bond critical points and the corresponding bond paths connecting these point with bonded

nuclei, were performed with the AIMAll software [73].

Acknowledgments

Grants from Universidad Nacional de San Luis (UNSL), partially supported this work. R. D.

Enriz; F. Garibotto and S.A.Andujar are members of the Consejo Nacional de Investigaciones

Científicas y Técnicas (CONICET-Argentina) staff.

REFERENCES

[1] . Bettinetti, S. ber, H. H bner, P. Gmeiner. Parallel synthesis and biological screening of

dopamine receptor ligands taking advantage of a click chemistry based BAL linker, J. Comb.

Chem., 7 (2005) 309−316.

[2] J.W. Kebabian, F.I. Tarazi, N.S. Kula, R.J. Baldessarini. Compounds selective for dopamine

receptor subtypes, Drug Discovery Today, 2 (1997) 333−340.

[3] R.J. Baldessarini, Dopamine receptors and clinical medicine. In The Dopamine Receptors;

Neve, K. A., Neve, R. ., Eds.; Humana Press: Totowa, NJ, 1997; pp 457−498.

[4] R.B. Mailman, D.E. Nichols, A. Tropsha, Molecular drug design for dopamine receptors. In

The Dopamine Receptors; K. A. Neve, R.L. Neve, Eds.; Humana Press: Totowa, NJ, 1997; pp

105−133.

[5] J.P. Kelleher, F. Centorrino, M.J. Albert, R.J. Baldessarini, Advances in atypical

antipsychotics for the treatment of schizophrenia: New formulations and new agents. CNS Drugs 16

(2002) 249−261.

[6] M. Rowley, L.J. Bristow, P.H. Hutson, Current and novel approaches to the drug treatment

of schizophrenia, J. Med. Chem. 44 (2001) 477−501.

[7] G. Gallagher, P.G. Lavanchy, J.W. Wilson, J.P. Hieble, R.M. Marinis, 4-[2-(Di-n-

propylamino)ethyl]-2(3H)-indolone: A Prejunctional Dopamine Receptor Agonist, J. Med. Chem.

28 (1985) 1533-1536.

[8] K.L. Kirk, C.R. Creveling, The Chemistry and Biology of Ring- Fluorinated Biogenic

Amines. Med. Res. Reu. 4 (1984) 189-220.

[9] R.M. DeMarinis, G. Gallagher, R.F. Hall, R.G. Franz, C. Webster, W.F. Huffmann, M.S.

Schwartz, C. Kaiser, S.T. Ross, J.W. Wilson, P. Hieble, Syntheses and in Vitro Evaluation of 4-(2-

Aminoethyl)-2(3H)-indolones and Related Compounds as Peripheral Prejunctional Dopamine

Receptor Agonists, J. Med. Chem. 29 (1986) 939-947.

[10] J.R. McCarthy, J. McCowan, M.B. Zimmerman, M.A. Wenger, L.W. Emmert, Synthesis and

Renal Vasodilator Activity of 2-Chlo-rodopamine and N-Substituted Derivatives, J. Med. Chem. 29

(1986) 1586-1590.

[11] J. Weinstock, D.E. Gaitanopouloe, O.D. Stringer, R.G. Franz, J.P. Hieble, L.B. Kinter, W.A.

Maan , K.E. Flaim, G. Gessner, Synthesis and Evaluation of Non-catechol D-1 and D-2 Dopamine

Receptor Agonists: Benzimidazol-2-one. Benzoxazol-2-one, and the Highly Potent Benzothiazol-2-

one 7-Ethylamines, J. Med. Chem. 30 (1987) 1166-1176.

[12] L. Nedelec, C. Dumont, C. Obrlander, D. Frechet, J. Laurent, J.R. Bohier, Synthesis and

Evaluation of Dopaminergic Activity of Diphenylethylamines. Eur. J. Med. Chem. 13 (1978) 553-

563.

[13] M. Cardellini, G.M. Cingolani, F. Claudi, V. Perlini, W. Balduini, F. Cattabeni, Synthesis and

Dopamine Receptors Binding Affinity of 2-(3-Fluoro-4-hydroxyphenyl)ethylamine and its N-Alkyl

Derivatives. Zl Farmaco, Ed. Sci. 43 (1988) 49-59.

[14] F. Claudi, M. Cardelhi, G.M. Ciolani, A. Piergentili, G. Peruzzi, W. Balduini, SynWi,

Dopamine Receptor Affinities of 2-(4-Chloro-3-hydroxyphenyl)ethylamine and N-Substituted

Derivatives. J. Med. Chem. 33 (1990) 2408-2412.

[15] F. Claudi, G. Giorgioni, A. Di Stefano, M.P. Abbracchio, A.M. Paoletti, W. Balduinio,

Synthesis and Pharmacological Characterization of 2-(4-Chloro-3-hydroxyphenyl)ethylamine and

N,N-Dialkyl Derivatives as Dopamine Receptor Ligands, J. Med. Chem. 35 (1992) 4408-4414.

[16] J. Weinstock, D.E. Gaitanopoulos, O. Hye-Ja, F.R. Pfeiffer, C.B. Karaeh, J.W. Venslavsky,

H.M. Sarau, K.E. Flaim, J.P. Hieble, C. Kaiser, Synthesis and Dopaminergic Activity of Some

Halogenated Mono- and Dihydroxylated 2-Aminotetralins, J. Med. Chem. 29 (1986) 1613-1627.

[17] K.H. Chung, H.J. Kim, H.R. Kim, E.K. Ryu, Oxidative chlorination of phenols with

hydrogen chloride/m-chloroperbenzoic acid/N,N-Dimethyl-formamide system, Synth. Commun. 20

(1990) 2991-2997.

[18] H. Nakazawa, K. Sano, K. Matsuda, K. Mitsugi, Enzymic preparation of aromatic

ethylamines from aromatic L-amino acids. Biosci. Biotech. Biochem. 57 (1993) 1210-1211.

[19] K.M. O`Boyle, J.L. Waddington, Structural Determinants of Selective Affinity for Brain D-1

Receptors within a Series of 1-Phenyl- 1H-3-benzazepines Analogues of SK&F 38393 and SCH

23390, Eur. J. Pharmacol. 115 (1985) 291-295.

[20] I. Berenguer, N. El Aouad, S. Andujar, V. Romero, F. Suvire, T. Freret, A. Bermejo, M.D.

Ivorra, R.D. Enriz, M. Boulouard, N. Cabedo, D. Cortes, Tetrahydroisoquinolines as dopaminergic

ligands: 1-Butyl-7-chloro-6-hydroxytetrahydroisoquinoline, a new compound with antidepressant-

like activity in mice, Bioorganic & Medicinal Chemistry, 17 (2009) 4968–4980.

[21] N. El Aouad, I. Berenguer, V. Romero, P. Marín, A. Serrano, S. Andujar, F. Suvire, A.

Bermejo, M.D. Ivorra, R.D. Enriz, N. Cabedo, D. Cortes, Structure–activity relationship of

dopaminergic halogenated1-benzyl-tetrahydroisoquinoline derivatives, European Journal of

Medicinal Chemistry, 44 (2009) 4616–4621.

[22] Andujar, S; Suvire, F; Berenguer, I; Cabedo, N.; Marín, P; Moreno, L.; Ivorra, M.D.; Cortes,

D.; Enriz, R.D. Tetrahydroisoquinolines acting as dopaminergic ligands. A molecular modeling

study using MD simulations and QM calculations. J Mol Model. 18 (2012) 419–431

[23] S.A. Andujar, R.D. Tosso, F.D. Suvire, E. Angelina, N. Peruchena, N. Cabedo, D. Cortes,

R.D. Enriz, Searching the "biologically relevant" conformation of dopamine: a computational

approach. J Chem Inf Model. 2012, 52(1), 99-112.

[24] U. Koch, P.L. Popelier, A Characterization of C-H…O Hydrogen Bonds on the Basis of the

Charge Density, J. Phys. Chem, 99 (1995) 9747-9754.

[25] C.F. Matta, N. Castillo, R.J. Boyd, Extended Weak Bonding Interactions in DNA: π-

Stacking (Base−Base), Base−Backbone, and Backbone−Backbone Interactions, J. Phys. Chem. B,

110 (2005) 563-578.

[26] R.A. Mosquera, M.J.G. Moa, L. Estévez, M. Mandado, A.M. Graña, An Electron Density-

Based Approach to the Origin of Stacking Interactions, In Quantum Biochemistry; Matta, C. F., Ed.;

WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2010.

[27] J.M. Dumas, H. Peurichard, M. Gomel, CX4•••base interactions as models of weak charge-

transfer interactions: comparison withstrong charge-transfer and hydrogen-bond interactions. J.

Chem. Res., 2 (1978) 54−55.

[28] T. Clark, M. Hennemann, J.S. Murray, P. Politzer, Halogen bonding: the sigma-hole. J. Mol.

Model., 13(2) (2007), 291−296.

[29] T. Clark, σ-Holes. Wiley Interdiscip. Rev.: Comput. Mol. Sci. [Online early access] DOI:

10.1002/wcms.1113. Published Online:Aug 8, 2012.

[30] A. Legon, The halogen bond: an interim perspective. Phys. Chem. Chem. Phys., 12 (2010)

7736-7747.

[31] A.C. Legon, Angew. Chem., Int. Ed., 38 (1999) 2686–2714.

[32] P. Politzer, J.S. Murray, T. Clark, Halogen bonding: an electrostatically-driven highly

directional noncovalent interaction. Phys. Chem. Chem. Phys., 12 (2010) 7748–7757.

[33] C. Bissantz, B. Kuhn, M. Stahl, A medicinal chemist’s guide to molecular interactions. J.

Med. Chem., 53(14) (2010) 5061−5084.

[34] L.A. Hardegger, B. Kuhn, B. Spinnler, L. Anselm, R. Ecabert, M. Stihle, B. Gsell, R.

Thoma, J. Diez, J. Benz, J.M. Plancher, G. Hartmann, D.W. Banner, W Haap, F. Diederich,

Systematic investigation of halogen bonding in protein−ligand interactions. Angew. Chem., Int. Ed.,

50(1) (2011) 314−318.

[35] L.A. Hardegger, B. Kuhn, B. Spinnler, L. Anselm, R. Ecabert, M. Stihle, B. Gsell, R.

Thoma, J. Diez, J. Benz, J.M. Plancher, G. Hartmann, Y. Isshiki, K. Morikami, N. Shimma, W.

Haap, D.W. Banner, F. Diederich, Halogen bonding at the active sites of human cathepsin L and

MEK1 kinase: efficient interactions in different environments. Chem Med Chem, 6(11) (2011)

2048−2054.

[36] R. Wilcken, X. Liu, M.O. Zimmermann, T.J. Rutherford, A.R. Fersht, A.C. Joerger, F.M.

Boeckler, Halogen-enriched fragment libraries as leads for drug rescue of mutant p53. J. Am.

Chem. Soc. 134(15) (2012) 6810−6818.

[37] (30) H.G. K mmel, Biography of the Coupled Cluster Method; World Scientific Publishing:

Singapore, 2002.

[38] (31) C. Møller, M.S. Plesset, Note on an approximation treatment for many-electron

systems. Phys. Rev., 46(7) (1934) 618.

[39] (32) J.A. Pople, J.S. Binkley, R. Seeger, Theoretical models incorporating electron

correlation. Int. J. Quantum Chem., (1976), 1−19.

[40] Y. Lu, Y. Wang, W. Zhu, Nonbonding interactions of organic halogens in biological systems:

implications for drug discovery and biomolecular design. Phys. Chem. Chem. Phys. 12(18) (2010)

4543−4551.

[41] Y. Lu, T. Shi, Y. Wang, H. Yang, X. Yan, X. Luo, H. Jiang, W. Zhu, Halogen bonding a novel

interaction for rational drug design? J. Med. Chem. 52(9) (2009) 2854−2862.

[42] (33) P. Dobe , J. ez , J. Fanfrlík, M. Otyepka, P. Hobza, Semiempirical quantum

mechanical method PM6-DH2X describes the geometry and energetics of CK2-inhibitor complexes

involving halogen bonds well, while the empirical potential fails. J. Phys. Chem. B, 115(26) (2011)

8581−8589.

[43] C. Coley, R. Woodward, A.M. Johansson, P.G. Strange, L.H. Naylor, Effect of Multiple

Serine/Alanine Mutations in the Transmembrane Spanning Region V of the D2 Dopamine Receptor

on Ligand Binding. J Neurochem., 74(1) (2000) 358-66.

[44] R Woodward; C Coley; S Daniell; Louise H Naylor; Philip G Strange, Investigation of the

role of conserved serine residues in the long form of the rat D2 dopamine receptor using site-

directed mutagenesis. Journal of neurochemistry, 66(1) (1996) 394-402.

[45] B.A. Cox, R.A. Henningsen, A. Spanoyannis, R.L. Neve, K.A. Neve, Contributions of

conserved serine residues to the interactions of ligands with dopamine D2 receptors. J. Neurochem.

1992, 59, 627-635.

[46] R.D. Tosso, S.A. Andujar, L. Gutierrez, E. Angelina, R. Rodríguez, M. Nogueras, H.

Baldoni, F.D. Suvire, J. Cobo, R.D. Enriz, Molecular Modeling Study of Dihydrofolate Reductase

Inhibitors. Molecular Dynamics Simulations, Quantum Mechanical Calculations, and Experimental

Corroboration. J. Chem. Inf. Model., (2013) DOI:10.1021/ci400178h.

[47] J. Párraga, N. Cabedo, S. Andujar, L. Piqueras, L. Moreno, A. Galán, E. Angelina, R.D.

Enriz, M.D. Ivorra, M.J. Sanz, D. Cortes, 2,3,9- and 2,3,11-Trisubstituted

Tetrahydroprotoberberines as D2 Dopaminergic Ligands, European Journal of Medicinal Chemistry

(2013), doi: 10.1016/j.ejmech.2013.07.036.

[48] E. Angelina, S.A. Andujar, R.D. Tosso, R. Enriz, N. Peruchena, Non-covalent interactions in

receptor-ligand complexes. A study based on the electron charge density" which you submitted to

the Journal of Physical Organic Chemistry (accepted for publication)

[49] R. Bader, Atoms in Molecules: A Quantum Theory, Oxford University Press., New York

1990

[50] S.J. Grabowski, QTAIM characteristics of halogen bond and related interactions. J. Phys.

Chem. A 116 (2012) 1838-1846.

[51] M.M. Teeter, M.F. Froimowitz, B. Stec, C.J. Durand, Homology Modeling of the Dopamine

D2 Receptor and Its Testing by Docking of Agonists and Tricyclic Antagonists. J. Med. Chem., 37

(1994) 2874-2888.

[52] A. Mansour, F. Meng, J.H. Meador-Woodruff, L.P. Taylor, O. Civelli, H. Akil, Site-directed

mutagenesis of the human dopamine D2 receptor. Eur J Pharmacol., 227 (1992) 205–214.

[53] V. Katritch, K.A. Reynolds, V. Cherezov, M.A. Hanson, C.B. Roth, M. Yeager, R. Abagyan,

Analysis of full and partial agonists binding to beta2-adrenergic receptor suggests a role of

transmembrane helix V in agonist-specific conformational changes. J Mol Recognit, 22 (2009) 307-

318.

[54] M.A. Soriano-Ursúa, J.O. Ocampo-López, K. Ocampo-Mendoza, J.G. Trujillo-Ferrara, J.

Correa-Basurto, Theoretical study of 3-D molecular similarity and ligand binding modes of

orthologous human and rat D2 dopamine receptors. Comput Biol Med, 41 (2011) 537-445.

[55] C.S. Page, P.A. Bates, Can MM-PBSA calculations Predict the specificities of protein kinase

inhibitors? J. Comp. Chem., 27 (2008) 1990–2007.

[56] K.M. Mertz, Limits of free energy computation for protein-ligan interactions. J. Chem.

Theory Comput., 6 (2010) 1769-1776.

[57] M. Alías, M.P. López, C. Cativiela, An efficient and stereodivergent synthesis of threo- and

erythro-b-methylphenylalanine. Resolution of each racemic pair by semipreparative HPLC.

Tetrahedron, 60 (2004) 885–891.

[58] L. Kozell, K.A. Neve, Constitutive Activity of a Chimeric D2/D1 Dopamine Receptor,

Molecular Pharmacology, 52 (1997) 1137–1149.

[59] M.Y. Kalani, N. Vaidehi, S.E. Hall, R.J. Trabanino, P.L. Freddolino, M.A. Kalani, W.B.

Floriano, V.W. Kam, W.A. Goddard 3rd, , The predicted 3D structure of the human D2 dopamine

receptor and the binding site and binding affinities for agonists and antagonists, Proc. Natl. Acad.

Sci. USA 101 (2004) 3815–3820.

[60] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, E. Hart, R.K. Belew, A.J. Olson,

Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy

function, J. Comp. Chem. 19 (1998) 1639–1662.

[61] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol. Graph. 14

(1996) 33–38.

[62] W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. W. Impey and M. L. Klein, J. Chem.

Phys., 79 (1983) 926–935.

[63] T. Darden, D. York and L. Pedersen, J. Chem. Phys., 98 (1993) 10089–10092.

[64] U. Essmann, L. Perera, M. L. Berkowitz, T. Darden, H. Lee and L. G. Pedersen, J. Chem.

Phys., 103 (1995) 8577–8593.

[65] D. A. Case, T. E. Cheatham, T. Darden, H. Gohlke, R. Luo, K. M. Merz, Jr., A. Onufriev, C.

Simmerling, B. Wang and R. J. Woods, J. Comput. Chem., 26 (2005) 1668–1688.

[66] D.A. Case, T.A. Darden, T.E. Cheatham III, C.L. Simmerling, J. Wang, R.E. Duke, R. Luo,

R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. Seabra, J. Swails, A.W.

Goetz, I. Kolossváry, K.F. Wong, F. Paesani, J. Vanicek, R.M. Wolf, J. Liu, X. Wu, S.R. Brozell, T.

Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews,

M.G. Seetin, R. Salomon-Ferrer, C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, P.A.

Kollman, AMBER12, University of California, San Francisco, 2012.

[67] T. Hou, N. Li, Y. Li, W. Wang, Characterization of domain-peptide interaction interface:

prediction of SH3 domain-mediated protein-protein interaction network in yeast by generic

structure-based models. J. Proteome Res., 11 (2012) 2982-95.

[68] H. Gohlke, C. Kiel, D.A. Case, Insights into protein-protein binding by binding free energy

calculation and free energy decomposition for the Ras-Raf and Ras-RalGDS complexes. J. Mol.

Biol., 330 (2003) 891-913.

[69] T. Hou, W. Zhang, D.A. Case, W. Wang, Characterization of domain-peptide interaction

interface: a case study on the amphiphysin-1 SH3 domain. J. Mol. Biol., 376 (2008) 1201-14.

[70] T. Hou, Z. Xu, W. Zhang, W.A. McLaughlin, D.A. Case, Y. Xu, W. Wang, Characterization

of domain-peptide interaction interface: a generic structure-based model to decipher the binding

specificity of SH3 domains. Mol. Cell Proteomics., 8 (2009) 639-649.

[71] T. Hou, Y. Li, W. Wang, Prediction of peptides binding to the PKA RIIalpha subunit using a

hierarchical strategy. Bioinformatics, 27 (2011) 1814-21.

[72] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J.

A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.

Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M.

Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,

M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.

Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.

Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A.

D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,

J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.

Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.

Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, and

J. A. Pople, Gaussian, Inc., Wallingford CT,Gaussian 03, revision B.05; Gaussian Inc.: Pittsburgh,

PA, 2003

[73] T. Keith, AIMA11, 12.11.09; Gristmill Software: Overland Park, KS, USA, 2012.

Figures Captions

Figure 1: Structure of dopamine and derivatives.

Figure 2: Spatial view of the complex obtained for compound 3 (cyan). The complex obtained for

DA is overimposed in gray for comparison.

Figure 3: Spatial view of the complex obtained for compound 2 (orange). The complex obtained for

DA is overimposed in gray for comparison.

Figure 4: Overimposed spatial view of the complexes obtained for compounds 4 (cyan) and 5

(brown).

Figure 5: Histograms of interaction energies partitioned for D2D-R amino acids when complexed

with compound 1 (a), compound 2 (b), compound 3 (c), compound 4 (d) and compound 5

(e). The x-axis denotes the residue number of D2D-R, and the y-axis shows the

interaction energy between the compound and the specific residue. Negative and positive

values represent favourable or unfavourable binding, respectively.

Figure 6: Spatial view of compound 3 (green)/D2DR interaction. Magnification of the receptor

active site at the right. The names of the residues involved in the main interactions are

written in the figure.

Figure 7: Molecular graph of compound 1 interaction with the binding site. Large spheres represent

attractors or nuclear critical points (3, -3) attributed to the atomic nuclei. The connecting

nuclei lines are bond paths and small spheres on them are bond critical points (3, -1).

Figure 8: Molecular graph of compound 3 interacting with the binding site.

Figure 9: Molecular graph of compound 2 interacting with the binding site.

Figure 10: Σb values obtained for the interactions of m-OH/Cl, p-OH/Cl and phenylethylamonium

of the three complexes analysed. Magnification of the m-OH/Cl and p-OH/Cl interactions

in figure b.

Scheme 1: Shyntesis of compound 3

Figure 1S: Similar spatial view to that shown in figure 2 but in this case displaying the spatial

distribution of the aromatic residues of the binding site.

Figure 2S: Similar spatial view to that shown in figure 3 but in this case displaying the spatial

distribution of the aromatic residues of the binding site.

Figure 3S: Molecular graph of compound 4 interacting with the binding site

Figure 4S: Molecular graph of compound 5 interacting with the binding site

Scheme 1

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 8

Figure 9

Figure 10

Anexo II: Espectros de RMN

respresentativos

Espectros

193

ESPECTROS RELATIVOS AL ARTÍCULO 2

Espectros en CDCl3 de

1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 1a

1a

Espectros

194


1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 1c

Espectros

195

Espectros en CDCl3 de HSQC y COSY 45 de 1c

Espectros

196


1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 3a

Espectros

197

Espectros en CDCl3 de HSQC y COSY 45 de 3a

Espectros

198


1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 3c

Espectros

199


E

Espectros en CDCl3 de 1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 1b

Espectros

200

Espectros en CDCl3 de DEPT y HSQC de 1b

Espectros

201


1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 1g

Espectros

202

Espectros en CDCl3 de COSY 45 y HMBC de 1g

Espectros

203


1H-RMN (500 MHz) y


Espectros

204

Espectros en CDCl3 de HSQC y HMBC de 2b

Espectros

205



1H-RMN (500 MHz) y


Espectros

206

Espectros en CDCl3 de DEPT y HMBC (125 MHz) de 3b

Espectros

207

Espectros en CDCl3 +DMSO de 1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 3d

Espectros

208

Espectros en CDCl3 de 1H-RMN (500 MHz) y

13C-RMN (125 MHz) de 4d

Espectros

209

Espectros en CDCl3 de COSY 45 y HSQC de 4d

Documents

FACULTAT DE FARMÀCIA DEPARTAMENT DE FARMACOLOGIA