Departamento de Farmacia y Tecnología Farmacéutica Facultad de Farmacia
DE NAVARRA
TESIS DOCTORAL “Nanosystems for the oral administration of alkyl-‐ lysophospholipid anti-‐tumor agents: development of lipid nanotransporters and preclinical studies”
Trabajo presentado por Ander Estella Hermoso de Mendoza para obtener el Grado de Doctor
Fdo. Ander Estella Hermoso de Mendoza Pamplona, 2011
Servicio de Publicaciones de la Universidad de Navarra ISBN 84-8081-097-1
Dña. Mª José Blanco Prieto, Profesora Investigadora de la Universidad de Navarra, informa que: El presente trabajo: “Nanosystems for the oral administration of alkyl-‐ lysophospholipid
anti-‐tumor
agents:
development
of
lipid
nanotransporters and preclinical studies” presentado por D. Ander Estella Hermoso de Mendoza para optar al grado de Doctor, ha sido realizado bajo su dirección en el Departamento de Farmacia y Tecnología Farmacéutica de la Facultad de Farmacia de la Universidad de Navarra y, una vez revisado, no encuentra objeciones para que sea presentado a su lectura y defensa. Y para que así conste, firma el presente informe. Fdo. Dra. María José Blanco Prieto Pamplona, 2011
Las investigaciones realizadas en el presente trabajo se han llevado a cabo dentro de los proyectos del Ministerio de Ciencia e Innovación (SAF2007-‐61261, SAF2010-‐15547, PCT-‐ 090100-‐2007-‐27) del Gobierno de España, de la Fundación Caja Navarra, Ibercaja, del Departamento de Salud del Gobierno de Navarra (beca Ortiz de Landázuri, ref: 63/09) y de la Línea Especial “Nanotecnologías y liberación controlada de fármacos” de la Universidad de Navarra. Así mismo, agradezco al Departamento de Educación del Gobierno Vasco (BFI06.37), al Departamento de Educación del Gobierno de Navarra y a la Asociación de Amigos de la Universidad de Navarra las becas predoctorales concedidas para la consecución de este trabajo.
A mis padres, Javier y Mª Jose
Muchas gracias… A la Universidad de Navarra y al Departamento de Farmacia y Tecnología Farmacéutica, por permitirme realizar esta tesis doctoral. Al Gobierno Vasco (Dpto. Educación), al Ministerio de Ciencia e Innovación, al Gobierno de Navarra, a Caja Navarra y a la Asociación de Amigos de la Universidad por la ayuda económica aportada durante estos años. A la Dra. María Blanco, por guiarme todos estos años, por confiar en mi persona y por toda la ayuda que me has ofrecido estos años. Gracias porque además de directora, has sido amiga y me has dado infinidad de consejos. Creo que no podía haber caido en mejores manos para hacer esta tesis. Muchas gracias! A Faustino Mollinedo, por colaborar activamente en el diseño de experimentos y ayudarnos a interpretar los resultados; y a su laboratorio en el Centro de Investigación del Cáncer de Salamanca, por ayudarme durante todas las estancias que he realizado allí, especialmente Janis de la Iglesia-‐Vicente, porque sin su ayuda no habría podido realizar varios experimentos de este trabajo. Muchas gracias, chati. Al Dr. Miguel Ángel Campanero (Campi), por toda la ayuda prestada durante todos estos años, que ha sido mucha, y por todos los buenísimos momentos que hemos pasado en el departamento y fuera de él. La cromatografía no será lo mismo sin tu ayuda… Al resto de doctores, profesores y personal del departamento: Dña. Marije Renedo, Dña. Pilar Ygartua, Dña. Carmen Dios, Dña. María del Mar Goñi, D. Felix Recarte, Dña. Pilar Guillén (por recordarme cuántas pastas o croissants me comía al cabo de la mañana los días que había cumpleaños… un besazo, Pili!), D. Juan Luis Martín, Dña. Paula Oteiza, D. Fernando Martínez, Dña. Socorro Espuelas, Dña. Maribel Calvo, Dña. Conchita Tros, Dña. María Jesús Garrido, D. Iñaki Fdez. De Troconiz y D. Juan Manuel Irache. Dentro de este grupo tan amplio, me gustaría dar las gracias especialmente a la Dra. Pilar Ygartua, por ser una gran persona, porque me ha escuchado y ha sabido darme consejos durante toda mi andadura, especialmente en este último tramo, porque personas como ella quedan pocas… Muchas gracias por darme tu apoyo, Pilar!
Muchas gracias… A mis compañeros de grupo de trabajo, pilares sobre los que me he podido apoyar durante mis momentos de flaqueo, que bien saben que han sido muchos: A Edurne Imbuluzqueta, la mujer que más nombres ha tenido en éste nuestro departamento, Imbuluz, Aduren, Imbulurguete, Edrine Imbuluzquetz,… gracias de verdad por estar siempre ahí, en los buenos y malos momentos, por todos tus consejos, por los mimos que me has dado y por enseñarme que siempre se puede sacar algo bueno de las cosas. Mila esker! A David González, por tu compañerismo sin límite, por dar todo por los demás sin esperar nada a cambio, por esos momentos de confidencias en el animalario y largas charlas que hemos tenido en el masas y por todo el apoyo que me has dado. Un abrazo, bordini! A Fabio Rocha, por ser auténtico (“shi”), buen amigo, buen compañero, por los ánimos que me has dado en los momentos bajos y por los buenos momentos que hemos compartido. Mucho ánimo, que no te queda nada! A Eduardo Ansorena, que aunque ya no está en el departamento (ni en el país) ha sido una persona muy importante para mí, y ha formado parte junto con David y conmigo de los míticos “Bordini”, gracias Edu por las risas que nos hemos echado juntos en tantas ocasiones (que han sido muchas y en muchos lugares del mundo!) y por los ánimos incondicionales que me has dado en los momentos complicadetes. A Elisa Garbayo, gracias por poner siempre ese punto de alegría en todo lo que hacemos, por esa risa contagiosa que tantas veces ha conseguido animarme, por todas las aventuras que hemos tenido en estos años, porque aunque nos dejaste una temporadita, es como si jamás te hubieras ido. Te he echado mucho de menos, Elisica! A Hugo Lana, porque te considero una parte fundamental en nuestro grupo y en mi tesis, porque sin tu ayuda esta tesis se habría quedado a la mitad; gracias por los infinitos ratos que hemos pasado en el animalario, machacando órganos y procesando muestras, gracias por estar siempre ahí, por sacar tiempo para tomarte un redbull y escucharme, por ser tan descerebrado como yo y seguirme el juego en todas las “gansadas” que nos hemos pegado. GRACIAS, Lana!!
Muchas gracias… A los que recientemente habéis llegado: Bea Lasa, “Blasa”, que continuará mis pasos y por su aportación sanguínea a la ciencia, Teresa Simón, “Maritere”, por su templanza, Esther Tamayo, por su inagotable capacidad de trabajo y organización de reuniones, y a las últimas incorporaciones al clan, Cristina (…) Tabar e Izaskun Imbuluzqueta (Imbuluz Senior!), mis competidoras por las meriendas de Martín y demás historias. A Noelia Ruz, porque empecé contigo toda mi andadura en la investigación, gracias por enseñarme todo lo que debía saber para sobrevivir en un laboratorio; gracias también por todos los buenos ratos que hemos pasado (casas rurales, cenas,…), por ser siempre tan buenísima persona y por tus consejos. A Onintza Sayar, muchísimas gracias por todo tu apoyo en los 8 años que he pasado en este departamento, porque SIEMPRE has estado cuando lo he necesitado. Me has ayudado a entender muchas cosas de la vida. Gracias por enseñarme que el tiempo lo cura todo, por los consejos (que han sido muchos) que me has dado y por los que todavía me vas a dar! Mila esker, Onin, mi paso por el departamento no habría sido el mismo sin tu ayuda. A Daniel Moreno, Victor López y Cristina Ederra, sois únicos, lo sabéis. Aunque ya no estéis en este departamento, sabéis que formáis parte de él y de mi tesis. Gracias por ayudarme en mis quebraderos mentales y no mentales, por vuestros sabios y “honorables” consejos, por dar ese toque de humor en los momentos más apropiados. Os quiero, chicos! A Raquel Martins, mi “Rachel”, lo más parecido a una hermana que he tenido, comenzamos juntos este camino, sufrimos juntos, “sobrevivimos” juntos y lo vamos a terminar juntos… Muchas gracias por tu amistad, tu comprensión, tu compañerismo, por todos los ratos que hemos disfrutado… Let’s dance now! A Cristina Aranda, Koldo Urbiola y Lorena de Pablo, voy a echar de menos trabajar con vosotros, pero sobre todo las risas, las lecturas matinales de horóscopo, los cafés, en fin, todas aquellas cosas que nos hacen inseparables en el tiempo… como el pensar que una tesis pueden defenderla dos personas… (Lorena pitrus dixit).
Muchas gracias… A Patricia Calleja, Sara Zalba, Judit Huarte (la mac-‐nífica) y Patricia Ojer, mis cuatro “jinetas” del apocalipsis departamental… mil gracias por ser tan buenas amigas, por todos los momentos de complicidad que hemos tenido y por todas las muestras de cariño que me habéis dado estos años. Gracias por todo vuestro apoyo, sabéis que me habéis ayudado mucho! A la mac-‐zona, Luisa Ruiz, Maite Agüeros e Irene Esparza, nunca escribir en el departamento había sido tan divertido, gracias por vuestro apoyo y consejos, por ayudarme sin esperar nada a cambio en todo momento, especialmente en los últimos días de mi tesis y por todos los buenos ratos. A Arianna Madrid, María Matoses, Zinnia Parra, Nieves Vélez, Marta Rayo, Guiomar Perez, Verónica Madrid, Sheyla Rehecho, Hesham Salman, Nacho Melgar, Izaskun Goñi, Amaya Lasarte… y muchos más que os podéis dar por agradecidos! To Dr. Véronique Préat, and all my colleagues during my stay in the “Unité de Pharmacie Galenique” of the Université Catholique de Louvain in Brussels. Thanks for making me feel like home. A Javi Alonso, por aguantarme todos estos años, gracias por tener siempre respuesta a todas mis preguntas, por tenderme tu mano en los momentos duros, que bien sabemos han sido muchos. Quiero que sepas que, a pesar de todo lo que pase, buena parte de esta tesis te pertenece. Gracias de corazón. Y por último, pero con diferencia los más importantes para mí, muchísimas gracias a Javier y Mª Jose, mis padres, los que siempre habéis estado y estaréis. Sé que, como yo, habéis pasado buenos y malos momentos a lo largo de esta tesis, pero tal y como me habéis enseñado siempre, las cosas si se hacen despacio y con cabeza acaban saliendo bien… y creo que este trabajo es muestra de ello. En serio, sin vuestro apoyo esta tesis no habría tenido sentido y por eso, os agradezco la fe que habéis tenido en mi, aunque en ocasiones os haya parecido que iba sin rumbo… Gracias, de verdad.
Y a todos los que habéis aportado luz a mi camino, ¡¡gracias mil!!
"Lo que con mucho trabajo se adquiere, más se ama." (Aristóteles)
Index
INDEX ................................................................................................................................................. i ABBREVIATIONS / ABREVIATURAS .............................................................................. iii INTRODUCTION ......................................................................................................................... 1 “Lipid nanomedicines for anticancer drug therapy” HYPOTHESIS AND OBJECTIVES / HIPOTESIS Y OBJETIVOS ........................... 25 CHAPTER 1 ................................................................................................................................ 35 “Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice” CHAPTER 2 ................................................................................................................................. 57 “Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization” CHAPTER 3 ................................................................................................................................. 81 “Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against mantle cell lymphoma” CHAPTER 4 ............................................................................................................... 113 “In vitro and in vivo efficacy of edelfosine-‐loaded lipid nanoparticles against glioma” GENERAL DISCUSSION .......................................................................................... 135 DISCUSION GENERAL ............................................................................................. 175 GENERAL CONCLUSIONS ...................................................................................... 217 CONCLUSIONES GENERALES ............................................................................... 221 FUTURE PERSPECTIVES / PERSPECTIVAS FUTURAS .................................. 225
i
Index
ANNEXES .................................................................................................................... 229 ANNEX I .......................................................................................................................... 231 “Lipid nanoparticles in biomedicine” ANNEX II ......................................................................................................................... 257 “Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems” ANNEX III ....................................................................................................................... 267 “In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts” ANNEX IV ........................................................................................................................ 279 “Lipid raft-‐targeted therapy in multiple myeloma” ANNEX V ..................................................................................................................... 291 Commentary: “The Rafts of the Medusa: cholesterol targeting in cancer therapy”
ii
Abbreviations / Abreviaturas
AFM
Atomic force microscopy
Microscopía de fuerza atómica
AIDS
Acquired immune deficiency syndrome Síndrome de inmunodeficiencia adquirida
ALP
Alkyl lysophospholipid
Alquil lisofosfolípido
ApoE
Apolipoprotein E
Apolipoproteína E
AUC
Area under the curve
Área bajo la curva
BBB
Blood brain barrier
Barrera hematoencefálica
BSA
Bovine serum albumin
Albúmina sérica bovina
bw
Body weight
Peso
CHOL
Cholesterol
Colesterol
CL
Clearance
Aclaramiento
CLL
Chronic lymphocytic leukemia
Leucemia linfocítica crónica
Cmax
Peak concentration
Concentración máxima
Cmin
Minimum concentration
Concentración mínima
CSF
Cerebrospinal fluid
Fluido cerebroespinal
D
Dose
Dosis
DCP
Dicetyl phosphate
Dicetil fosfato
DDAB
Didecyldimethylammonium bromide
Bromuro de didecildimetilamonio
DOPE
Dioleoylphosphatidyl ethanolamine
Dioleoilfosfatidil etanolamina
DOTAP
Dioleoyl trimethylammonium propane
Dioleoil trimetilamonio propano
DSC
Differential scanning calorimetry
Calorimetría diferencial de barrido
EDTA
Ethylenediaminetetraacetic acid
Ácido etilendiaminotetraacético
ELL-‐12
Liposomal edelfosine
Edelfosina liposomal
EPR
Enhanced permeability and retention
Incremento de la permeabilidad y retención
ESI
Electrospray ionization
Ionización por electrospray
ET-‐18-‐OCH3 Edelfosine
Edelfosina
FBS
Fetal bovine serum
Suero fetal bovino
FCA
Freud's complete adjuvant
Adyuvante completo de Freud
FCS
Fetal calf serum
Suero fetal bovino
FITC
Fluorescein isothiocyanate
Isotiocianato de fluoresceína
GRAS
Generally recognised as safe
Comunmente reconocidos como seguros
HBV
Hepatitis B virus
Virus de hepatitis B
HIV
Human immunodeficiency virus
Virus de inmunodeficiencia humana
iii
Abbreviations / Abreviaturas
HPLC-‐MS
High performance liquid chromatography mass spectrometry
Cromatografía líquida de alta resolución acoplada a espectrometría de masas
HPTLC
High performance thin layer chromatography
Cromatografía en capa fina de alta resolución
IARC
International Agency for Research on Cancer
Agencia Internacional de Investigación del Cáncer
IC50
Inhibitory concentration 50
Concentración inhibitoria 50
IFN
Interferon
Interferón
IgG
Immunoglobulin G
Inmunoglobulina G
k
Distribution constant rate
Constante de distribución
LD
Laser diffractometry
Difractometría láser
LDC
Lipid-‐drug conjugates
Conjugados lípido-‐fármaco
LDL
Low density lipoproteins
Lipoproteinas de baja densidad
LHRH
Luteinizing-‐hormone-‐ releasing hormone
Hormona liberadora de hormona luteinizante
LN
Lipid nanoparticles
Nanopartículas lipídicas
LOD
Limit of detection
Límite de detección
LOQ
Limit of quantitation
Límite de cuantificación
MCD
Methyl cyclodextrin
Metil ciclodextrina
MCL
Mantle cell lymphoma
Linfoma de manto
MDR
Multidrug resistant
Multirresistente a fármacos
MDT
Mean dissolution time
Tiempo medio de disolución
MM
Multiple myeloma
Mieloma múltiple
MPS
Mononuclear phagocytic system
Sistema fagocítico mononuclear
MRM
Multiple reaction monitoring
Monitorización de reacciones múltiples
MRT
Mean residence time
Tiempo medio de residencia
MTT
3-‐(4,5-‐Dimethylthiazol-‐2-‐yl)-‐2,5-‐ diphenyltetrazolium bromide
Bromuro de 3-‐(4,5-‐dimetiltiazol-‐2-‐il)-‐2,5-‐ difeniltetrazolio
NLC
Nanostructured lipid carriers
Transportadores lipídicos nanoestructurados
O/W
Oil-‐in-‐water simple emulsion
Emulsión simple orgánico-‐en-‐acuoso
P-‐gp
P-‐glycoprotein
P-‐glicoproteína
PAF
Platelet activating factor
Factor activador de plaquetas
PBL
Peripheral blood lymphocytes
Linfocitos de sangre periférica
PBS
Phosphate-‐buffered saline
Tampón fosfato salino
iv
Abbreviations / Abreviaturas
PC
Phosphatidyl choline
Fosfatidilcolina
PCS
Photon correlation spectroscopy
Espectroscopía de correlación fotónica
PDI
Polydispersity index
Indice de polidispersión
PEG
Polyethylene glycol
Polietilen glicol
PLA
Polylactic acid
Ácido poliláctico
PLGA
Polylactic co-‐glycolic acid
Ácido poliláctico co-‐glicólico
PMS
N-‐methyl-‐dibenzopyrazine methylsulphate
Metilsulfato de N-‐metil-‐dibenzopirazina
PP
Peyer's patches
Placas de Peyer
QC
Quality control
Control de calidad
QD
Quantum dot
Quantum dot
RES
Reticuloendothelial system
Sistema reticuloendotelial
RSD
Relative standard deviation
Desviación estándar relativa
S/N
Signal-‐to-‐noise ratio
Ratio señal/ruido
SA
Stearyl amine
Estearil amina
SCID
Severe combined immunodefficiency
Inmunodeficiencia severa combinada
SIM
Selected ionization monitoring
Monitorización de ionización seleccionada
SLN
Solid lipid nanoparticles
Nanopartículas lipídicas sólidas
t½Ka
Absorption half life
Semivida de absorción
t½α
Distribution half life
Semivida de distribución
t½β
Elimination half life
Semivida de eliminación
TLC
Thin layer chromatography
Cromatografía en capa fina
Tmax
Time to peak concentration
Tiempo requerido para alcanzar la concetración máxima
TRAIL
Tumor necrosis factor-‐related apoptosis-‐inducing ligand
Ligando inductor de apoptosis asociado a factor de necrosis tumoral
UHPLC-‐ MS/MS
Ultra high performance liquid cromatography tandem mass spectrometry
Cromatografía líquida ultrarrápida acoplada a espectrometría de masas en tandem
VEGF
Vascular endothelial growth factor
Factor de crecimiento endotelial vascular
Vss
Volume of distribution
Volumen de distribución
W/O
Water-‐in-‐oil simple emulsion
Emulsión simple acuoso-‐en-‐orgánico
W/O/W
Water-‐in-‐oil-‐in water multiple emulsion
Emulsión múltiple acuoso-‐en-‐orgánico-‐en-‐ acuoso
WHO
World Health Organization
Organización Mundial de la Salud
v
Introduction. Lipid nanomedicines for anticancer drug therapy
Introduction
Lipid nanomedicines for anticancer drug therapy Ander Estella-‐Hermoso de Mendoza1, Miguel A. Campanero2, Faustino Mollinedo3, María J. Blanco-‐Prieto1 1Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia,
University of Navarra, E-‐31008, Spain 2Servicio de Farmacología Clínica, Clínica Universitaria, E-‐31080 Pamplona, Spain 3Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain Journal of Biomedical Nanotechnology, 2009 (5) p. 323-‐343
1
Introduction. Lipid nanomedicines for anticancer drug therapy
Copyright © 2009 American Scientific Publishers All rights reserved Printed in the United States of America
Journal of Biomedical Nanotechnology Vol. 5, 1–21, 2009
Lipid Nanomedicines for Anticancer Drug Therapy Ander Estella-Hermoso de Mendoza1 , Miguel A. Campanero2 , Faustino Mollinedo3 , and María J. Blanco-Prieto1!∗ 1
Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia, University of Navarra, E-31008, Spain 2 Servicio de Farmacología Clínica, Clínica Universitaria, E-31080 Pamplona, Spain 3 Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del Cáncer, CSIC-Universidad de Salamanca, Campus Miguel de Unamuno, E-37007 Salamanca, Spain
REVIEW
More than half of all people diagnosed with cancer receive chemotherapy. Unfortunately, most chemotherapy drugs cannot tell the difference between a cancer cell and a healthy cell. In this sense, some other drawbacks often encountered with antineoplastic compounds, such as poor stability and specificity and a high occurrence of drug-resistant tumor cells may be overcome to some degree by incorporating them into drug delivery systems. Solid Lipid Nanoparticles (SLN) have arisen considerable interest in recent years. These are particles of submicron size made from a lipid matrix that is solid at room and body temperature. Moreover, the biodegradable and biocompatible nature of SLN makes them less toxic than other nanoparticulate systems. SLN are capable of encapsulating hydrophobic and hydrophilic drugs, and they also provide protection against chemical, photochemical or oxidative degradation of drugs, as well as the possibility of a sustained release of the incorporated drugs. Along with these last issues, the feasibility of scaling up for large scale production and the low cost of lipids as compared to biodegradable polymers or phospholipids have favoured their use as potential drug delivery systems. This review focuses on the recent advances in SLN as carriers for chemotherapeutic agent delivery
Keywords: Solid Lipid Nanoparticles, SLN, Nanomedicines, Cancer, Antineoplastic.
CONTENTS 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Lipid Based Nanosystems . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Solid Lipid Nanoparticles . . . . . . . . . . . . . . . . . . . . . 2.2. Nanostructured Lipid Carriers (NLC) . . . . . . . . . . . . . . 2.3. Lipid Drug Conjugates (LDC) . . . . . . . . . . . . . . . . . . 2.4. Surface Modified Lipid Based Nanosystems: Active Targeting . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Preparation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. High Pressure Homogenization (Hot and Cold) . . . . . . . 3.2. Microemulsion Technique . . . . . . . . . . . . . . . . . . . . . 3.3. Emulsion Formation Solvent-Evaporation or -Diffusion Method . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Water-in-Oil-in-Water (w/o/w) Double Emulsion Method . . . . . . . . . . . . . . . . . . . . . 3.5. Emulsification Dispersion Followed by Ultrasonication . . 3.6. Hot Homogenization by High Shear Homogenization and/or Ultrasonication . . . . . . . . . . . . . . . . . . . . . . . 3.7. Solvent Injection Method . . . . . . . . . . . . . . . . . . . . . 4. Characterization of SLN . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Particle Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Surface Charge—Zeta Potential . . . . . . . . . . . . . . . . . 4.3. Atomic Force Microscopy . . . . . . . . . . . . . . . . . . . . . 4.4. Crystallinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ∗
Author to whom correspondence should be addressed.
J. Biomed. Nanotechnol. 2009, Vol. 5, No. 4
1 3 3 6 6 6 7 7 7 7 8 8 8 8 8 8 9 9 9
4.5. Stability . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Lyophilization . . . . . . . . . . . . . . . . . . . . . 5. Efficacy and Toxicity Studies of SLN Loaded with Cytotoxic Drugs . . . . . . . . . . . . . . . . . . . . . . . 6. Pharmacokinetics and Tissue Biodistribution of SLN 6.1. General Aspects . . . . . . . . . . . . . . . . . . . . 6.2. Routes of Administration . . . . . . . . . . . . . . 6.3. Pharmacokinetics . . . . . . . . . . . . . . . . . . . 6.4. Tissue Distribution . . . . . . . . . . . . . . . . . . 7. Concluding Remarks . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . Reference and Notes . . . . . . . . . . . . . . . . . . . .
....... .......
9 10
. . . . . . . . .
10 12 12 13 15 16 18 18 18
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
1. INTRODUCTION Cytotoxic drugs remain the mainstay of cancer chemotherapy.1 The effective use of cancer chemotherapy requires a thorough understanding of the principles of neoplastic cell growth kinetics, basic pharmacologic mechanisms of drug action, pharmacokinetic and pharmacodynamic variability, and mechanisms of drug resistance. Recent scientific advances in the field of molecular oncology have led to the identification of large numbers of potential targets for novel anticancer therapies. Hence, from a pharmacological point of view, these drugs present
1550-7033/2009/5/001/021
doi:10.1166/jbn.2009.1042
1
3
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
different mechanisms of action which make them appropriate for one type of cancer or another. Some anticancer agents induce their cytotoxic effects during specific phases of the cell cycle, whereas others show non-specific cytotoxic activity. Antimetabolites, such as fluorouracil (5-FU) and methotrexate, are more active against S-phase cells, whereas the vinca alkaloids and taxanes are relatively more M-phase specific. These differences in
Estella-Hermoso de Mendoza et al.et al. Mendoza
the mechanism of action may have clinically important consequences for cancer chemotherapy. For example, cellcycle–nonspecific agents, such as the alkylating agents and platinum derivatives, generally have linear dose-response curves (e.g., increasing the dose increases cytotoxicity). In contrast, cell-cycle–specific agents will often plateau in their concentration-dependent effects because only a subset of proliferating cells remain fully sensitive to
REVIEW
Ander Estella-Hermoso de Mendoza was got his Bachelor of Science in Pharmacy (2005) and his Master of Science in Pharmaceutical Technology (2007) at the University of Navarra. He is currently a Ph.D. student at the Department of Pharmacy and Pharmaceutical Technology in the same University. His research studies are focused on the development, in vitro and in vivo characterization of polymer and lipid based drug delivery systems containing alkyl lysophospholipid antineoplastic drugs.
Miguel A. Campanero studied Pharmacy at the University of Navarra and received his Ph.D. at the same University in 1998. He is associate professor of Pharmacology at the Faculty of Medicine and Head of the Pharmacokinetics Laboratory from the Clinical Investigation Unit of the University of Navarra. His current research field is preclinical and clinical evaluation of drug delivery systems, covering aspects of bioanalytical evaluation and pharmacokinetic/pharmacodynamic characterization of drug distribution profiles.
Faustino Mollinedo studied Chemistry at the Complutense University of Madrid (Spain), where he obtained his Ph.D. in 1982. After postdocs in Dartmouth Medical School (Hanover, NH, USA), and the New York University Medical Center (NY, USA) (1982–1985), he became a group leader in 1986 in the Center of Biological Research (CIB, CSIC) in Madrid (Spain). In 1994 he moved as group leader and Main Investigator to the Institute of Biology and Molecular Genetics (University of Valladolid-CSIC9 in Valladolid (Spain) (1994–2000), and then to the Center for Cancer Research (University of Salamanca-CSIC) of Salamanca (Spain) (2000-present), becoming Full Professor in 2002. His current work is mainly focused on phospholipid ethers, selective killing of tumor cells, apoptosis and cancer, lipid rafts in apoptosis, and neutrophil differentiation, activation and apoptosis. María J. Blanco-Prieto received her Pharmacy Degree from the University of Santiago de Compostela (Spain) in 1991, followed by a Ph.D. in Pharmaceutical Sciences (1996) from the University of Paris-Sud (France). From 1996 to 1999 she completed a post-doctoral training at the Swiss Federal Institute of Technology (ETH), Zürich, (Switzerland) and then joined the University of Navarra as research scientist and associate professor. Since January 2006, she is the director of the line of research “Nanotechnologies and Drug Delivery Systems” at the Department of Pharmacy and Pharmaceutical technology, University of Navarra. Her research lay in the field of biomaterials and advanced drug carrier systems including the design and the development of polymer and lipid based micro- and nanoscale carriers, their biological evaluation in in vitro cell cultures (toxicity, mechanism of action, intracellular drug release) and also their pharmacokinetic and dynamic impact in vivo (using relevant animal models of the diseases). Research carried out in this field is mainly applied to cancer, infectious, cardiovascular and neurodegenerative diseases. 2
J. Biomed. Nanotechnol. 5, 1–21, 2009
4
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
Mendoza et al. de Mendoza et al. Estella-Hermoso
Table I.
Classification of antineoplastic agents vectorized in SLN.
Group Alkaloids Alkylating agents Alkyl Lysophopholipids Anthracyclines Antimetabolites Hormonal agents Podophyllotoxin derivatives Taxanes Others
Examples Mitotic inhibitor: Vinorelbine Topoisomerase I inhibitor: Camptothecin Nitrogen mustard: Chlorambucil Edelfosine Doxorubicin, Idarubicin Folic acid analog: Methotrexate Pyrimidine analog: 5–Fluorouracil Estrogen receptor antagonist: Tamoxifen Etoposide Paclitaxel, Docetaxel All trans retinoic acid Beta–elemene Cholesteryl butirate Cisplatin Topoisomerase II inhibitor: Mitoxantrone
J. Biomed. Nanotechnol. 5, 1–21, 2009
route of administration for all these cytotoxic drugs is the intravenous bolus or infusion, classically in the form of free drug solutions, and even if they present a very long history of use and new drug regimes have been developed in order to improve their clinical accomplishment, it is still frequent to encounter treatment failures.1! 15 This review focuses on the recent advances in SLN as carriers for the delivery of chemotherapeutic agents listed in Table I.
2. LIPID BASED NANOSYSTEMS SLN, NLC and LDC have been designed to overcome the main problems of membrane stability and drug leakage associated with liposomes and conventional emulsions.16 Table II summarizes the antineoplastic drug-loaded lipid formulations developed so far. 2.1. Solid Lipid Nanoparticles SLN are a relatively new class of drug carriers. It was in the mid-1980s when Speiser developed the first micro and nanoparticles (named nanopellets), made of solid lipids for oral delivery.17 Since then, lipids have turned out to be attractive materials for drug delivery system preparation. SLN are characterized as being colloidal particles made of lipids that remain in solid state at room and body temperature, which are generally recognised as safe (GRAS) or have a regulatory accepted status. They present sizes from 50 to 500 nm, depending on the method and materials employed for the manufacturing process. They were introduced by Gasco in the early 1990s, produced by the dilution of a warm microemulsion,18 and by Müller et al.,19 who produced them by a high pressure homogenization method. Advantages that present SLN are the use of biodegradable physiological lipids, the bioavailability improvement of poorly water-soluble molecules, enhanced drug penetration into the skin after topical administration and protection of chemically labile agents from degradation in the gut.20–23 Different stabilizing agents or surfactants confer various interesting properties on the systems. For instance, the use of steric stabilizers such as polysorbates or poloxamers hinders the anchoring of the lipase/colipase complexes in the gastrointestinal tract and consequently, the degradation of the SLN.24 Besides, the use of Tween 80 or sodium dodecylsulfate can make SLN cross the blood brain barrier and deliver drugs into the brain.25–28 Some limitations to these SLN are the likely expulsion of the encapsulated drug during storage and relatively low drug loading. This expulsion phenomenon can occur along the time-dependent restructuring process that happens during storage, which alters the crystallization pattern of the lipid in which the drug is embedded. The step from an imperfect crystalline lattice of the lipid to a more perfect lipid crystalline structure leads to drug expulsion.29–31
REVIEW
drug-induced cytotoxicity. These cell-cycle-specific agents tend to be schedule dependent, because the only way to increase the total cell kill is by extending the duration of exposure, not by increasing the dose.2 In this sense, lipid based particulate systems offer great promise as drug carriers to improve the therapeutic effectiveness and safety profile of these conventional forms of chemotherapeutic agents, by providing a sustained release of the drug over time.3–7 All these DNA-affecting drugs treat cancer principally by being toxic to cells that grow and divide very rapidly. Unfortunately, most of these drugs also affect non cancerous cells, leading to undesirable adverse effects. However, new synthetic ether-linked analogues of phosphatidylcholine and lysophosphatidylcholine, collectively named as antitumour lipids (ATLs), have been developed.8–10 They were initially synthesized in the late 1960s, but have attracted greater interest since the finding that the prototype molecule of the group, the ether lipid 1-O-octadecyl-2-O-methyl-rac-glycero3-phosphocholine (ET-18-OCH3 , edelfosine), selectively accumulates in tumor tissue11 and induces a selective apoptotic response in tumour cells, sparing normal cells. Unlike most chemotherapeutic agents currently used, edelfosine does not interact with DNA, but acts at the cell membrane, and therefore its effects seem to be independent of the proliferative state of target cells.12 Recent progress has unveiled the molecular mechanism underlying the apoptotic action of edelfosine, involving membrane rafts and Fas/CD95 death receptor,13 and has led to the proposal of a two-step model for the selective action of edelfosine on cancer cells, namely: (a) edelfosine uptake into the tumour cell, but not into normal cells; (b) intracellular activation of Fas/CD95 through translocation and capping into membrane rafts. Edelfosine constitutes the first antitumour drug acting through the intracellular activation of the Fas/CD95 death receptor.9 However, drawbacks such as poor oral bioavailability and hemolytic toxicity have been described for this drug.11! 14 The main
3
5
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy Table II.
Overview of antineoplastic agents vectorized in lipid nanoparticles, composition and preparation methods.
Drug
Lipid matrix
Surfactants/ligands
5-fluorouracil
Triolein, cholesterol
All trans retinoic acid
Tricaprin
Beta-elemene Camptothecin
Monostearin, Precirol Precirol Compritol Precirol, squalene Stearic acid
Hydrogenated soybean phosphatidylcholine, DSPE, Tween 80, ferritin Egg phosphatidyl choline, Tween 80, DSPE-PEG Pluronic F68, Myverol, Pluronic F68 Myverol, Pluronic F68 Myverol, Pluronic F68 Soybean lecithin, Pluronic F68
Cisplatin
Monostearin
Tween 80, Pluronic F68
Chlorambucil
Stearic acid
Cetylpalmitate
Pluronic F68, sodium glycocholate Tyloxapol, sodium glycocholate Pluronic F68, sodium glycocholate, PEG2000 -DSPE Pluronic F68, Tween 60, DDAB Epikuron 200, sodium taurocholate, butanol Tween 80, Span 85, DOTAP
Precirol
Tween 80, DOTAP
Docetaxel
Trimyristin
Doxorubicin
Stearic acid
Egg phosphatidyl choline, galactosylated DOPE Epikuron 200, sodium taurocholate Epikuron 200, sodium taurocholate, PEG2000 Tween 80
Stearic acid, oleic acid
REVIEW
Estella-Hermoso de Mendoza et al.et al. Mendoza
Stearic acid Stearic acid Cholesteryl butirate DNA
Stearic acid Edelfosine
Compritol, stearic acid
Monostearin
Etoposide
Stearic acid
Hydrogenated soybean phosphatidylcholine, sodium tauroglycocholate Hydrogenated soybean phosphatidylcholine, sodium tauroglycocholate Hydrogenated soybean phosphatidylcholine, sodium tauroglycocholate Epikuron 200, sodium taurocholate Soybean lecithin, sodium taurodeoxycholate Soybean lecithin
Monostearin
Soybean lecithin
Tristearin
Soybean lecithin
Compritol
Soybean lecithin
Compritol
Lecithin, polyoxyl-40-stearate
Distearin
Tripalmitin
Idarubicin
Stearic acid
Methotrexate
Stearic acid
Mitoxantrone
Preparation method
% EE
Reference
Solvent injection method
36.8
39
High pressure homogenization Hot homogenization Hot homogenization Hot homogenization Hot homogenization High pressure homogenization Emulsification dispersion– ultrasonication method Hot homogenization
78.7
57
97–99 n.d.
99.6
78 35 35 35 58
82.3
66
>95
67
Hot homogenization
67
Hot homogenization
67
Hot homogenization
67
Microemulsion technique
n.d.
6
High pressure homogenization Emulsion formation solvent evaporation method Hot homogenization
>90
63–65
n.d.
74
>90
68
99
69, 72
Microemulsion technique Microemulsion technique Emulsion formation solvent evaporation method High pressure homogenization
4, 70 85
30
>95
59
High pressure homogenization
59
High pressure homogenization
59, 60
Microemulsion technique Hot homogenization Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsification dispersion– ultrasonication method
98
71
51.3
16
55–75
75 75 75 75
87.3
5
continued
4
J. Biomed. Nanotechnol. 5, 1–21, 2009
6
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
Estella-Hermoso Mendoza et al. de Mendoza et al. Table II.
Continued Lipid matrix
Paclitaxel
Tristearin
Pluronic F68
Stearic acid
Pluronic F68
Compritol
Pluronic F68
Monostearin
Pluronic F68, PEG
Monostearin
Pluronic F68, Folic acid
Trimyristin
Egg phosphatidyl choline, PEG2000 -DSPE Epikuron 200, cholesteryl hemisuccinate, sodium taurocholate, butanol Epikuron 200, sodium taurocholate Lecithin, oleic acid, Pluronic F68
Tripalmitin
Tamoxifen
Palmitic acid
Vinorelbine bitartrate
Monostearin
Surfactants/ligands
Preparation method
% EE
Reference
Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method Emulsion formation solvent diffusion method High pressure homogenizaton Microemulsion technique
36.2
48
59–77
48
Microemulsion technique High pressure homogenization
48 48 48 72–89
61
n.d.
95
28–31
3
80
62
Abbreviations: DOPE—Dioleoylphosphatidyl ethanolamine, DDAB—Dimethyl distearyl ammonium bromide, DSPE—Distearoylphosphatidyl ethanolamine, DOTAP— dioleoyl trimethylammonium propane, PEG2000 —Polyethylene glycol 2000, n.d.—Not determined.
Therefore, the incorporation capability of drugs depends not only on physical-chemical properties of the drug but also on the miscibility, solubility of the drug in the lipid melt, type of matrix material and the degree of crystallinity and polymorphic form.23! 32 Drug molecules that
are not accommodated within the crystalline lattice may adsorb onto nanoparticle surface. There are different models described in the literature to explain the mechanisms of drug incorporation (Fig. 1(A)). The major work has been performed by the research group of Mehnert and
REVIEW
Drug
(A)
Drug-enriched core
Drug-enriched shell
Homogeneous matrix or solid solution
(B)
Fig. 1.
(A) Drug incorporation models for the three types of SLN and (B) basic structure of NLC.
J. Biomed. Nanotechnol. 5, 1–21, 2009
5
7
Introduction. Lipid nanomedicines for anticancer drug therapy
REVIEW
Lipid Nanomedicines for Anticancer Drug Therapy
co-workers.21! 33 The drug-enriched core is formed when cooling of the hot o/w emulsion leads to precipitation of the drug first. This happens preferentially in lipid solutions with drug dissolved at its saturation solubility in the lipid at production temperature. During cooling, super saturation and subsequent drug precipitation are obtained. The drug-enriched shell model can be explained as a function of the solubility of the drug in the water-surfactant mixture at increased temperature during the production process. The drug partially leaves the lipid particle and dissolves in the aqueous phase during hot homogenization. Cooling of the oil/water nanoemulsion reduces the drug solubility in the aqueous phase, and the drug tries to distribute into the lipid particles leading to enrichment in the particle shell if the particle core has already started to solidify. In the case of a solid solution, the drug is molecularly dispersed uniformly in the particle matrix. Drug release takes place by diffusion from the solid lipid matrix and additionally by further lipid nanoparticle degradation.34 2.2. Nanostructured Lipid Carriers (NLC) The aforementioned problems of drug loading with SLN can be minimized by the recently developed NLC (Fig. 1(B)) by the creation of a lipid particle matrix as imperfect as possible in order to accommodate the active molecules. To achieve this aim, a blend of solid lipid and liquid lipid is used to produce nanoparticles that remain solid at temperatures up to 40 ! C. NLC present considerable crystal disorder, translated into a higher drug loading and less drug expulsion during storage.35 Nevertheless, it has been observed that the controlled release properties of the carrier may be compromised due to the decrease in diffusion length of the lipid matrix.36 However, this drawback can be modulated by altering the proportions of solid and liquid lipids.23 Upon administration of SLN and NLC via the parenteral route of administration, improved biodistribution profile, targeting, and enhanced cytotoxicity against multidrug resistant cancer cells have been observed.35! 37 2.3. Lipid Drug Conjugates (LDC) Due to their lipophilic nature, SLN and NLC present a great advantage for highly efficient incorporation of hydrophobic drugs, but with only a moderate loading capacity for hydrophilic compounds.34 Therefore, in order to solve this problem LDC were developed. In this case, the hydrophilic drug is transformed into a more hydrophobic molecule by conjugation with a lipid compound. Additionally, LDC present advantages for oral and parenteral administration. After oral administration, the bioavailability will be enhanced in comparison to the administration of micronized conjugate powder. After intravenous (i.v.) administration, the LDC nanoparticles offer improved possibilities for drug targeting and distribution through 6
Estella-Hermoso de Mendoza et al.et al. Mendoza
the organism according to the partition coefficient of the drug.38 LDC can be prepared by salt formation or alternatively by covalent linkage. Most of the lipid conjugates melt around 50–100 ! C. These conjugates are melted and dispersed in a hot stabilizing agent solution. Further processing is performed equal to SLN and NLC. The obtained emulsion system is cooled, and the conjugate recrystallises and forms LDC nanoparticles.38 This LDC suspension could to some extent be considered as a nanosuspension of a pro-drug. 2.4. Surface Modified Lipid Based Nanosystems: Active Targeting The in vivo fate of SLN can be customized by a surface modification procedure. This goal can be easily accomplished by the use of ligands that specifically bind to surface receptors on target sites.26! 39 In the case of cancer, there are many receptors reported in the literature that seem to be overexpressed in neoplastic cells. Some examples are folic acid or ferritin receptors (overexpressed in carcinomas),39–42 peptide receptors (growth factors, gastrin related peptides, vasoactive intestinal peptide)43–45 and receptors for LDL (in breast cancer, colon cancer).46! 47 The attachment of suitable ligands on to the surface of lipid nanoparticles results in increased selectivity for tumor tissue.48–52 But the antigen that will allow molecular targeting must be expressed exclusively on cancer cells, be an integral part of an essential cellular function of the cancer cells, and should not easily mutate as the cancer cells proliferate.53 Khalid et al.52 observed the increase in the biological half-life while providing substantial accumulation at the tumour site after the administration of poly(ethylene glycol)-coated SLN to subcutaneously implanted C26 colon adenocarcinoma bearing mice. Stella et al.54 exploited folic acid as a targeting agent to actively target tumor cells, especially in ovarian cancer, while other authors showed that the folic acid-coated lipid nanoparticles containing paclitaxel could enhance the cellular uptake of SLN and the cellular cytotoxicity of the drug by the improved endocytosis mediated by folate receptor in A549 lung cancer cells.48 Stevens et al.51 demonstrated that the treatment of mice bearing lung carcinoma M109 tumours with folate receptor-targeted SLN containing the same drug resulted in significantly greater tumour growth inhibition and animal survival compared to treatment with nontargeted SLN or free paclitaxel formulated in Cremophor EL. Similarly, Jain et al.39 administered 5-fluorouracil (5-FU), SLN and ferritin-coated SLNs to MDA-MB-468 tumour bearing mice. The administration of ferritin-coated SLN formulation also resulted in effective reduction in tumour growth as compared with free 5-FU and plain SLN. Intact antibodies or their fragments have also been used as highly specific targeting agents. Beduneau et al.55 conjugated a whole monoclonal antibody to LNC directed J. Biomed. Nanotechnol. 5, 1–21, 2009
8
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
against transferrin receptors, enhancing the uptake of these particles by tissues overexpressing such receptor. Recently, delivery of actives to different tissues using nanoparticulate drug carriers in combination with novel targeting principles of “differential protein adsorption (PathFinder Technology)” has been studied in depth.28! 38! 56 This technology identifies the naturally occurring mechanism for localization of material in different parts of the body via adsorbed blood proteins. The production of particles with appropriate surface properties leads in vivo to preferential adsorption of the targeting protein and subsequently to site-specific i.v. delivery.38
3. PREPARATION METHODS
Table III. Main components employed in the formulation of SLN in the field of cancer. Lipid matrix
Caprylic/capryc triglyceride (Mygliol 812) Cetylpalmitate Cholesterol Glyceryl trilaurate (Dynasan 112) Glyceryl trimyristate (Dynasan 114) Glyceryl tripalmitate (Dynasan 116) Glyceryl tristearate (Dynasan 118) Glyceryl monostearate (Imwitor 900) Glyceryl behenate (Compritol 888 ATO) Oleic acid Precirol ATO 5 Solid paraffin Stearic acid
Surfactants
Phosphatidyl choline (Epikuron) Lecithins (egg, soy) Pluronic F68 Polysorbate 80 (Tween 80) Solutol HS15
Co-surfactants
Sodium glycocholate Sodium taurocholate Sodium tauroglycocholate Sodium taurodeoxycholate Cholesteryl hemisuccinate
Cryoprotectants
Trehalose Sucrose Sorbitol Maltose Glucose Mannose
Molecules on surface
Polyethylene glicol Folic acid
J. Biomed. Nanotechnol. 5, 1–21, 2009
3.1. High Pressure Homogenization (Hot and Cold) Patented by Müller and Lucks,19 the high pressure homogenization is a method used to achieve particles with narrow distribution in size. In hot high pressure homogenization, drugs are usually dispersed or dissolved in a molten solid lipid, normally 5–10 ! C above the melting point of the lipid. Next, the drug-lipid melt is dispersed in a surfactant solution kept at the same temperature by high-speed stirring, yielding an aqueous o/w primary emulsion. This pre-emulsion is passed through a high pressure homogenizer, at the same temperature as the previous step. Classic homogenization parameters for this method are from one to three homogenization cycles of 500 bar. The more cycles that are performed, the greater the reduction in size and polidispersity index obtained.22! 34 In the case of the cold homogenization process, the drug is dissolved in the melted lipid followed by a fast cooling by means of dry ice or liquid nitrogen, which leads to the formation of a solid solution. This solution is then micronized with the aid of a mortar or a ball mill, and those solid lipid microparticles are then dispersed and emulsified in a cool aqueous phase containing surfactants. This method overcomes most formulation problems encountered with thermolabile drugs. All-trans retinoic acid,57 camptothecin,58 etoposide,59! 60 paclitaxel,61 vinorelbine bitartrate62 or DNA63–65 are some examples of drugs incorporated into lipid nanosystems by high pressure homogenization.
REVIEW
SLN can be obtained in many ways, from the methods in which organic solvents are employed to more modern methods incorporating the use of high pressure homogenizers. The great variety of materials (solid lipids, liquid lipids, surfactants) and homogenizing techniques can lead us to the achievement of the multiple lipid systems described previously, each one presenting its own physicalchemical profile. Table III summarizes the most widely used excipients in the formulation of antineoplastic drug loaded SLN.
Lipid Nanomedicines for Anticancer Drug Therapy
3.2. Microemulsion Technique The method developed by Gasco et al.18 has been adapted or modified by many authors according to their needs.3! 16! 35! 39! 48! 52! 57! 60! 61! 66–71 Briefly, this method can be explained as the preparation of a warm microemulsion by stirring, containing typically around 10% molten solid lipid, 15% surfactant and up to 10% cosurfactant. This warm microemulsion is then dispersed under stirring in excess cold water (typical ratio 1:50) using a specially developed thermostated syringe. The excess water is removed either by ultra-filtration or by lyophilization in order to increase the particle concentration. Drugs encapsulated by means of this technique are cholesteryl butirate,6 tamoxifen,3 paclitaxel,6 idarubicin71 and doxorubicin.4! 69! 70! 72 3.3. Emulsion Formation Solvent-Evaporation or -Diffusion Method In the emulsion formation solvent-evaporation method,5! 30 the lipid is completely dissolved in a water-immiscible organic solvent (e.g., chloroform) which is then emulsified in an aqueous surfactant solution before evaporation of the solvent under reduced pressure. Once the solvent evaporates, the lipid precipitates forming SLN. An important 7
9
Introduction. Lipid nanomedicines for anticancer drug therapy
REVIEW
Lipid Nanomedicines for Anticancer Drug Therapy
advantage of this method is the avoidance of heat during the preparation, which makes it suitable for the incorporation of highly thermolabile drugs. Drawbacks might arise due to solvent residues in the final suspension. These suspensions are generally quite dilute, because the lipid presents a limited solubility in the organic material. In the end, lipid concentrations in the final SLN dispersion range around 0.1 g/l, and therefore the particle concentration has to be increased by means of ultra-filtration or evaporation. In the solvent-diffusion technique, partially watermiscible solvents (e.g., benzyl alcohol, ethyl formate) are used.73 Initially, they are both saturated with water to make a certain initial thermodynamic balance of both liquids. Later, the lipid is dissolved in the water-saturated solvent and subsequently emulsified with solvent-saturated aqueous surfactant solution at a raised temperature. After the addition of excess water (typical ratio: 1:5–1:10), the SLN precipitate due to diffusion of the organic solvent from the emulsion droplets to the continuous aqueous phase. Similar to the production of SLN via microemulsions, the particle suspension needs to be concentrated by means of ultra-filtration or lyophilization. Mean particle sizes around 100 nm with very narrow size distributions can be achieved by both solvent evaporation methods. Edelfosine 30 and DNA74 are two examples of drugs that have been encapsulated via solvent evaporation method, and paclitaxel 48 or methotrexate,75 examples of drugs incorporated in SLN via solvent diffusion method. 3.4. Water-in-Oil-in-Water (w/o/w) Double Emulsion Method This method consists of the solubilization of the drug in the internal phase of a double emulsion, along with a surfactant capable of preventing the drug from leaching to the external phase during solvent evaporation. In this case, the lipid melt is emulsified with an aqueous drug and surfactant solution at a high temperature by high shear homogenization (e.g., Ultra Turrax). The obtained warm w/o nanoemulsion is the dispersed in the aqueous external phase of the w/o/w emulsion containing a stabilizer under mechanical stirring at 2–3 ! C. The resultant SLN suspension is then concentrated and purified by diafiltration.30! 76! 77 3.5. Emulsification Dispersion Followed by Ultrasonication This simple method is based on the dissolution of the drug, lipid and emulsifier in a common solvent, which is then evaporated under reduced pressure in order to obtain a solvent free drug dispersed lipid phase. This phase is then homogenized with a hot aqueous surfactant solution using a homogenizer, and then ultrasonicated to obtain the nanoemulsion. The SLN are formed upon cooling 8
Mendoza Estella-Hermoso de Mendoza et al.et al.
at room temperature. Cisplatin66 and mitoxantrone5 have been encapsulated following this method. 3.6. Hot Homogenization by High Shear Homogenization and/or Ultrasonication These techniques do not require the involvement of organic solvents or large amounts of surfactants. Molten lipid is added and dispersed in an aqueous solution with the help of high shear homogenization or ultrasonication. The obtained emulsion is left to cool down at room temperature and may be concentrated by ultrafiltration. Even if these techniques are relatively easy and require tools that are readily available in most laboratories, they present some disadvantages such as low dispersion quality, which is often affected by the presence of a small percentage of microparticles. Metal contamination is another major problem presented by ultrasonication. Many antineoplastic drugs have been successfully incorporated into lipid nanocarriers by this method: beta-elemene,78 camptothecin,35 chlorambucil,67 docetaxel68 and methotrexate.16 3.7. Solvent Injection Method This technique is based on the solvent diffusion method. Here, SLN are prepared by rapidly injecting a solid lipid solution in water miscible solvents into water. SLN precipitate due to diffusion of the organic solvent from the emulsion droplets to the continuous aqueous phase. The most widely used solvents in this method are ethanol, acetone, isopropanol and methanol. An example of drug successfully vectorized by this method is 5-fluorouracil.39
4. CHARACTERIZATION OF SLN The physical-chemical characterization of nanoparticles is essential for the development of any formulation, and so for SLN formulations that will be further tested both in vitro and in vivo. 4.1. Particle Size Particle size plays a critical role not only in their clearance by the reticuloendothelial system (RES), but also in the gastrointestinal uptake and in the prevention of embolisms after intravenous administration. For this reason, the precise determination of the particle size is a must. Sizes below 5 "m are required in order not to cause embolisms due to the blocking of the thin capillaries after parenteral administration of lipid nanoparticles.79 On the other hand, sizes no larger than 300 nm are advisable for the intestinal transport to the thoracic duct.80 The most established method for particle size measurement is photon correlation spectroscopy (PCS).81 Nanoparticles are usually polydisperse in nature and polydispersity index (P. I.) gives a J. Biomed. Nanotechnol. 5, 1–21, 2009
10
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
measure of size distribution of the nanoparticle population. Therefore, the lower the P. I. value, the more monodispersed a SLN suspension is. For acceptable polydisperse systems some authors affirm that the P. I. should be less than 0.5,26! 82 however, most researchers accept P. I. values below 0.3 as optimum values.30! 83–86 This technique can sometimes lead to erroneous results when there is presence of different particle sizes which may be accidental microbial contamination, particle agglomerates or dust. To avoid these untrue data it is always wise to corroborate the results with other appropriate methods such as electron microscopy or atomic force microscopy (AFM).
(A)
(C)
4.2. Surface Charge—Zeta Potential
4.3. Atomic Force Microscopy AFM is an important tool for studying either particulate or biological samples due to its ability to image surfaces under liquids.88 Currently, this is the only technique available that directly provides structural, mechanical, functional and topographical information about surfaces with nanometer-to angstrom-scale resolution. Figure 2 shows an example of AFM images taken from edelfosine loaded lipid nanoparticles.30 4.4. Crystallinity Differential scanning calorimetry (DSC) and X-ray diffractometry (XRD) are two of the main tools employed to determine the crystallinity and polymorphic behaviour of the components of the SLN. DSC gives information on the melting and crystallization behaviour of all solid and liquid constituents of the particles, while XRD is used to identify specific crystalline compounds, both mineral and organic, based on their crystal structure. In XRD, a powder made up of small crystals is irradiated with a monochromatic beam of X-rays with the assumption that the orientation of the crystallite is random. The beam is diffracted at angles J. Biomed. Nanotechnol. 5, 1–21, 2009
REVIEW
The zeta potential is the overall charge a particle acquires in a specific medium. The significance of zeta potential is that its value can be related to the stability of colloidal dispersions. The zeta potential indicates the degree of repulsion between close and similarly charged particles in dispersion. For particles that are small enough, a high zeta potential will confer stability, i.e., the solution or dispersion will resist aggregation. When the potential is low, attraction exceeds repulsion and the dispersion will break and flocculate. So, colloids with high zeta potential (negative or positive) are electrically stabilized while colloids with low zeta potentials tend to coagulate or flocculate. In general, zeta potentials superior to −60 mV are essential for excellent electrostatic stabilization. Values of −30 mV were enough for good stabilization, hence, indicating good physical stability.87
(B)
Fig. 2. Atomic force microscopy images of freeze-dried Compritol® lipid nanoparticles: multi-particles (A); zoom-in of the selected area of A (B); 3D morphological image (C).
determined by the spacing of the planes in the crystals and the type and arrangement of the atoms. A scanning detector records the diffracted beams as a pattern. The intensity and position of the diffractions are unique to each type of crystalline material. This pattern can be identified, even in complex mixtures, by comparison to reference spectra using a computerized database. XRD investigations have been most important in the elucidation of the manner of arrangement of lipid molecules, their multiplemelting phenomena, phase behaviour and the characterization and identification of the structure of lipid and drug molecules.29! 30 The determination of the cristallinity of the components of the formulation is very important as the incorporated drug may undergo a polymorphic transition. Lipid matrices can also change their polymorphic behaviour leading to a possible undesirable drug expulsion during storage.31 4.5. Stability Unlike polymeric nanoparticles,89 SLN do not tend to swell in the presence of water.75 However, the type of lipid matrices and surfactants, and its concentration may critically affect the stability of formulations. Chemical instability can be assessed by means of decomposition of lipids. In a study of stability of different lipid matrices, Radomska-Soukharev et al.90 concluded that the decomposition of formulations with mono-, di- and 9
11
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
triglycerides was around 10% and that the decomposition increased as the concentration of surfactants increased. The formation of a gel is one of the typical physical instabilities of SLN leading to a change in surface area. This gelation takes place by the SLN building up a network and lipid bridges between them.87 Lipid formulations with Imwitor 900 or Dynasan 118 (tristearin) can be named as examples that form gels at temperatures of 40 ! C.90 More recently, Paliwal et al.75 studied the stability of methotrexate-loaded SLN. These authors concluded that 60 days after they were prepared SLN maintained their size at 4 ! C, while this size increased when these particles were kept at 25 ! C.
REVIEW
4.6. Lyophilization Obtaining a dry product in order to ensure stability of SLN is very important. Lyophilization is one of the most used techniques to produce dry powders from nanoparticulate suspensions.91–95 However, parameters employed in this procedure depend on the lipid used for the SLN formulation. Zimmerman et al. optimized the temperature and time parameters for the lipid matrix Imwitor 900 (glyceryl monostearate).79 Firstly, the aqueous SLN-dispersion was diluted with a solution of a cryoprotectant (trehalose or fructose) and frozen at −70 ! C right after production in a deep-freeze. After thermal treatment of 2 hours at −22 ! C and a posterior cooling down to −40 ! C for another 2 hours, formulations underwent a primary drying process at 1.03 mbar (−30 ! C 7 h, −10 ! C 2 h, 20 ! C 12 h). Finally, a secondary drying process for 3 hours at 30 ! C and 0.001 mbar was performed to obtain the lyophilized product. Consequently, the lyophilization process subjects the lipid nanosystems to freezing and drying, two important stresses that can affect their physical stability if suitable stabilizers or cryoprotectants, namely sugars, are not employed. In fact, when SLN are lyophilized without cryoprotectants, the final product commonly results in the aggregation of particles, which acquire a rubbery aspect. The most generally employed cryoprotectants in the formulation of SLN are trehalose, sucrose, sorbitol, maltose, glucose and mannose, among others. In a study of freeze-drying of SLN, Schwarz and Mehnert96 concluded that trehalose was the most effective cryoprotectant in preventing particle growth. Following this line, Del Pozo-Rodríguez et al.74 studied the short- and longterm stability of lyophilized DNA-containing PrecirolSLN cryoprotected with glucose and trehalose. Samples were frozen at −80 ! C. After 24 hours, frozen samples were lyophilized at −55 ! C and 0.2 mbar for 48 hours. 5% glucose-cryprotected SLN rendered a powdered product that became rubber after 48 h, while the use of other glucose concentrations led to rubbery samples from the beginning. In the case of trehalose, a concentration of 10% 10
Estella-Hermoso de Mendoza et al.et al. Mendoza
manifested the best results, yielding a powdered product that remained over the time. This same study showed that the lyophilization process of the SLN meant an increase in the particle size of about 100 nm, whatever the trehalose concentration used. However, this size hardly varied along the first 9 months at room temperature.
5. EFFICACY AND TOXICITY STUDIES OF SLN LOADED WITH CYTOTOXIC DRUGS The local or systemic chemotherapeutic response of a drug may be improved by incorporation of anticancer agents into a delivery system that can facilitate sustained drug release for a sufficient period of time. Therefore, efficacy and therapeutic value of antitumour drug could be substantially diminished by severe systemic toxicity. This is consistent with an in vivo study with Ehrlich ascite carcinoma (EAC) bearing mice performed by Ruckmany et al.16 They administered methotrexateloaded SLN made of stearic acid intraperitoneally for 9 days. After day 9, mean survival time for the mice treated with SLN was nearly 30% higher than that of the mice treated with methotrexate solution. Reddy et al.59 also increased by 5–8 days the mean survival time of mice bearing Dalton’s lymphoma after etoposide-loaded SLN were administered intraperitoneally. SLN were made of glycerol monostearate, glycerol distearate and tripalmitin. Among the three nanoparticle formulations, tripalmitin lipid nanoparticles showed significantly higher apoptosis and increase in survival time of tumour bearing mice. Xu et al.68 observed that docetaxel-loaded galactosylated SLN showed the potential of increasing the efficacy in the treatment of hepatocellular carcinomas expressing the asialoglycoprotein receptor. Furthermore, a SLN-treated group of hepatoma bearing mice demonstrated the most dramatic efficacy of all the tested treatments, and all animals treated showed complete tumour regression with no significant toxicity after a multiple dose administration of docetaxelloaded SLN. In another study, mitoxantrone-loaded SLN also showed efficacy against lymph node metastases. The treatment with SLN gave a mean size of lymph node three times smaller than that of the mitoxantrone solution.5 Table IV shows the most representative in vitro studies performed with antineoplastic agents loaded in SLN. To date, there is no SLN product for cancer therapy on the market. In order to develop a SLN formulation for its administration in clinical practice it is crucial to establish the biocompatibility of the components with blood components and other tissues. For this reason, various in vitro and in vivo studies have been performed to study the toxicity and biocompatibility of lipid nanoparticles. Experiments with the interaction of SLN with human granulocytes revealed that SLN had much lower uptake by the RES, resulting in extended circulation time in blood. The study also showed the nearly 10-fold J. Biomed. Nanotechnol. 5, 1–21, 2009
12
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
Estella-Hermoso Mendoza et al. de Mendoza et al. Table IV.
In vitro studies performed with different tumoural cell lines and corresponding IC50 (half maximal inhibitory concentration).
Drug
Lipid matrix
Molecule on surface
Cell line
5-fluorouracil Cholesteryl butirate Docetaxel Doxorubicin
HSPC Epikuron 200 Trimyristin Stearic acid
Ferritin — — —
Paclitaxel
Tristearin Compritol Monostearin Monostearin Trimyristin
— — PEG Folic acid PEG2000 -DSPE
Tamoxifen
Palmitic acid
—
MDA-MB-648 HT-29 BEL7402 HT-29 U-373 Y-79 HL-60 MCF-7 A549 A549 A549 A549 OVCAR-3 MCF-7 MCF-7
IC50 (!M) free solution
IC50 (!M) SLN
IC50 (!M) targeted SLN
5.02 ≥ 600 — 0.13 2.15 >0.51 0.07 0.02 2.13 2.13 2.13 2.13 ≈ 20 ≈ 0"5 ≈ 100
3.56 300 — 0.08 0.99 0.16 0.001 0.001 0.84 1.33 0.63 0.63 — — ≈ 100
1.28 — — — — — — — — — 1.74 0.19 ≈ 20 0.5—1 —
Reference 39 6 68 69, 72
48 48 48 48 61 3
Abbreviations: DSPE—Distearoylphosphatidyl ethanolamine, HSPC—Hydrogenated soya phosphatidylcholine, PEG2000—Polyethylene glycol 2000, DSPE— Distearoylphosphatidyl ethanolamine, HSPC—Hydrogenated soya phosphatidylcholine, PEG2000—Polyethylene glycol 2000.
J. Biomed. Nanotechnol. 5, 1–21, 2009
an in vitro degradation study.99 In contrast, a lower dose of Compritol SLN (2.5%, w/w) was well tolerated. The binding of SLN formulations to different cells, including red blod cells, is important because it may affect not only blood clotting or embolic effects but also may produce a change in the pharmacokinetic behaviour of the drug. SLN have diverse affinities to erythrocytes depending on the surfactant used. Olbrich et al.24 showed by flow cytometry that SLN formulations (made of Compritol as matrix material and Tween 80 and Poloxamer 188 as surfactants) exhibit a binding to erythrocytes below 10.0%. In contrast, when Span 85 was employed, blood cell affinity of SLN was increased, which translated into a significant aggregation of red blood cells (75.3%). Surfactants are also extremely important components of SLN and their biocompatibility and cytotoxicity must be investigated. Good tolerability depends in the first line on the surfactant used and secondarily on the lipid. To formulate parental SLN, surfactant of GRAS status must be employed, e.g., lecithin, Tween 80, Poloxamer 188, Span 85 or sodium glycocholate. Many authors have studied the toxicity of different stabilizing agents on cells and their capability to reverse multidrug resistance (MDR), suggesting that MDR reversal by polyoxyethylene surfactants involves alterations of the lipids physical state of the plasma membrane in resistant cells as one of the possible mechanism(s).101# 102 Concerning Solutol® HS15, Woodcock et al.102 concluded that 66% of the R100 cells (MDR-derivative of human leukaemia cell line) were lysed when incubated for only 1 h with this polyethoxylated solubilising agent at concentrations equivalent to 1:100 (w/v) and 100% were lysed by concentrations of 1:10 (w/v). This surfactant is greatly employed in the formulation of LNC. Hence, Allard et al.103 investigated the cytotoxicity of a tamoxifen derivative loaded in LNC prepared with
REVIEW
lower cytotoxicity of glyceride SLN compared to that of PLGA nanoparticles.97 As a consequence, the authors concluded that SLN can be used as drug carriers because of their prolonged circulation time and high toxicological acceptance. In an attempt to study the toxicity of lipids employed, SLN made from lipids consisting of stearic acid or dimethyl-dioctadecylammonium bromide were assayed in vitro using murine macrophages in the presence of different concentration and different size of nanoparticles.98 It was concluded that the nanoparticles were cytotoxic at the concentration of 0.01%, independently of the size of the SLN. However, SLN made from lipids consisting of triglycerides, paraffin or cetylpalmitate were found to be harmless at the same concentration. These data agree with the results obtained by Weyhers et al.,99 who in an in vivo study provided a better understanding of the toxicity created by SLN in vivo when two types of SLN, one made from Compritol (GRAS) and the other made from cetyl palmitate (which is less physiologic), were i.v. administered six times within a period of 20 days. The doses that were administered in this study were extremely high, up to 1 g of solid lipid per kg of body weight (10% w/w). Translated into human administration, this could be compared to bolus injection of 75 g of solid particles intravenously.100 The hepatic and splenic tissues were analysed histopathologically. The results showed that the toxicity was dependent not only on the lipid matrix but also on the dose administered. Thus, for the less physiological lipid cetyl palmitate containing SLN no pathological results were obtained, whereas Compritol containing formulations led to a partially reversible accumulation of the lipid in liver and spleen and subsequently to pathological alterations. This accumulation was attributed to the slow degradation of the Compritol matrix compared to the faster degradation of cetyl palmitate that could be observed in
11
13
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso de Mendoza et al.et al. Mendoza
REVIEW
Lipid Nanomedicines for Anticancer Drug Therapy
Solutol® HS15 using rat 9 L gliosarcoma cells. Tamoxifen derivative-loaded LNC demonstrated cytotoxic activity on 9 L cells 150-fold higher than blank LNC, while cell survival curves were not significantly different between the derivative in solution and loaded in LNC, indicating that the activity of the drug was totally recovered in vitro after encapsulation. The same authors performed an in vivo study using the same tumour cell line-bearing rats, observing that animals treated with a single administration of drug-loaded LNC at an equivalent dose of 2.5 mg/kg presented tumours three times smaller than those treated by blank LNC, indicating that the cytostatic activity of the tamoxifen derivative was preserved in vivo. An in vitro antiproliferative activity of tamoxifen-loaded SLN was also assayed on MCF-7 cell line (human breast cancer cells),3 showing that tamoxifen incorporated in SLN maintained the same antitumour activity as when administered as a free drug. Many in vitro studies performed with doxorubicin loaded SLN have clearly shown the efficacy of the systems.69! 72 In all tested tumour cell lines, doxorubicinloaded SLN inhibited cell growth more strongly than the free solution. Authors could associate the enhanced cytotoxicity with an increased drug accumulation in tumour cells.72 Cytotoxicity of paclitaxel-loaded SLN has also been widely assayed.6! 48! 51! 61! 69! 104 Different authors have discovered that the cellular uptakes of different types of SLN are time-dependent, concentration-dependent and that the intracellular accumulation of drugs may be further optimized through receptor-mediated endocytosis of the drug-loaded SLN, as previously mentioned in this review. These authors determined that the order of cellular uptake ability for the tested SLN was tristearin SLN > monostearin SLN > stearic acid SLN > Compritol ATO888 SLN. Concerning cytotoxicity, IC50 values for paclitaxel demonstrated that monostearin SLN were the less cytotoxic.48 In a another study, Lee et al.61 found no significant differences in cytotoxicity between paclitaxelloaded SLN and the paclitaxel containing Cremophor EL-based formulation, presenting an IC50 value of approximately 20 "M against OVCAR-3 cell line. Compared with MCF-7 cells, the paclitaxel containing Cremophor EL-based formulation had a slightly lower IC50 value than the paclitaxel-loaded SLN, even if this difference was not statistically significant. You et al.62 incorporated vinorelbine bitartrate into SLN and tested the cytotoxicity of the system in a MCF-7 breast cancer cell line. Compared to the free vinorelbine bitartrate, the cellular cytotoxicity of the drug loaded in SLN increased approximately 20% after 24 h in the same incubation conditions, while blank SLN presented a cellular viability of above 80%. 12
6. PHARMACOKINETICS AND TISSUE BIODISTRIBUTION OF SLN 6.1. General Aspects Anticancer drugs are a diverse class of compounds that present specific problems when used in cancer treatment. Poor specificity, high toxicity and susceptibility to induce drug resistance are the main drawbacks that have been observed after their use in humans. Conventionally administered cytotoxic agents extensively and indiscriminately bind to body tissues in a highly unpredictable manner. Only a small fraction of the drugs reaches the tumour site. This may both reduce the therapeutic efficacy and increase systemic drug toxicity. Moreover, even though cytotoxic drugs ideally should only kill cancer cells, in reality they do not differenciate between healthy and cancerous cells. Tumour response is determined by the in vivo drug half-life, the cellular residence time of the drug and the nature of observed cellular responses. Generally, tumours are associated with a defective, leaky vascular architecture as a result of the poorly regulated nature of tumour angiogenesis. Moreover, the interstitial fluid within a tumour is usually inadequately drained by a poorly formed lymphatic system. As a result, submicron-sized particulate matter such as SLN can preferentially extravasate into the tumour and be retained there, maximizing the amount of drug that can reach the targeted tumour sites.7 This phenomenon is known as the enhanced permeability and retention (EPR) effect and is widely exploited in the drug delivery system community. Chemical and physical properties of the nanoparticles, including size, composition, and surface, are also important factors that determine their pharmacokinetics and biodistribution. In general particles, as with all intravenously injected and colloidal particulates, are opsonized and cleared from the circulation by the reticuloendothelial system or mononuclear phagocyte system. To facilitate drug targeting, in tumour tissue for example, a reticuloendothelial system avoidance (stealth) facility may be incorporated as described well for classical polymeric nanoparticels in the past. SLN may be used to target drugs at particular organs without surface modification because of their small size (50–200 nm). Moreover, solid lipid nanoparticles have drawn increasing interest from every branch of medicine owing to their ability to overcome certain biological barriers, resulting in increased therapeutic efficacy of the encapsulated drug. SLN increase the tumour accumulation104 and allow brain delivery of anticancer drugs not capable of crossing the blood brain barrier (BBB).4 Different sub-classes of cytotoxic agents and their derivatives, including anthracyclines (e.g., doxorubicin, idarubicin), taxanes (e.g., paclitaxel), camptothecins (camptothecan, 7-ethyl-10-hydroxy-20(S)-camptothecin (SN38)), etoposide, flurodooxyuridine (FudR) and retinoic acid and other J. Biomed. Nanotechnol. 5, 1–21, 2009
14
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
investigational compounds (e.g., cholesteryl butyrate) have been encapsulated, and their disposition evaluated in preclinical studies on different species. 6.2. Routes of Administration
6.2.1. Oral Administration of SLN Oral delivery is the easiest and the most convenient route for drug administration, but many drugs are not administered orally because they are not absorbed or are degraded in the gastrointestinal tract. SLN have been proposed to overcome these problems. Absorption of SLN by the gastrointestinal tract has been studied extensively for the last decade. The factors controlling intestinal absorption of particulate systems are now known. Size, nature of the lipid material, inner structure of the vehicle and the employed surfactants have been determined as critical factors influencing particle uptake.98! 106 Reduction in the particle size is a key factor for improving the peroral performance of SLN of poorly soluble drugs. In SLN formulations, the particle size range has been reduced to less than 400 nm, resulting in an increase in surface area and saturation solubility, and making it possible to reach high concentrations of SLN in the gastrointestinal tract.106 Due to their small particle size, SLNs may exhibit bioadhesion properties to the gastrointestinal tract wall or enter the intervillar spaces thus increasing their residence time in the gastrointestinal tract. This increase in adhesion results in enhanced bioavailability. J. Biomed. Nanotechnol. 5, 1–21, 2009
The mechanism of oral absorption for SLN still needs to be revealed. Different mechanisms for the particle absorption from the gastrointestinal tract have been described (Fig. 3). Transport through endothelial tight junctions, endocytosis by the epithelial membranes and M-cell mediated transport have been proposed as possible mechanisms. The main SLN uptake pathway in rats is the M-cells transport overlaying the Peyer’s patches.107 The particle size is a critical determinant of the absorption of nanoparticles administered orally. Larger particles (>300 nm) may be retained for longer periods in Peyer’s patches, while smaller particles are transported to the thoracic duct. Moreover, the composition of SLN can also play a role in relation to the uptake pathway. A wide variety of solid lipids, such as highly purified triglycerides, complex glyceride mixtures, phospholipids or even waxes are employed to prepare SLN. Surfactants are used to stabilize the obtained drug delivery formulation. It is known that liquid lipid vehicles constituted of some unsaturated fatty acids and triglycerides can favour intestinal lymphatic absorption of orally administered drugs. Briefly, fatty acids with chain lengths of 14 or greater are more highly lymphatically transported, whereas shorter chain fatty acids are primarily absorbed via the portal blood. As such, triglycerides composed of long chain fatty acids more effectively support lymphatic drug transport than their medium and short chain counterparts.108 Transport of such drugs or carriers through the intestinal lymphatics, via the thoracic lymph duct to the systemic circulation joining at the junction of the jugular and left subclavian vein, avoids presystemic hepatic metabolism, and as a result, enhances the amount of orally administered drugs reaching into systemic circulation. Several experiments have been performed to ensure the SLN lymphatic uptake after duodenal administration to rats. Transmission electron microscopy has been used to observe the presence of SLN in lymph and blood.107! 109 In some recent research, SLN suspensions were administered by oral gavages to rats with and without thoracic lymph duct cannulation. Approximately 77.9% of absorbed SLN was transported into systematic circulation via lymph, whereas the remaining 22.1% of SLN was absorbed by non-lymphoid pathway.110 The intestinal lymphatic transport of SLN offers some potential advantages. Firstly, direct lymphatic transport bypasses the first pass metabolism in the enterocyte and potentially reduces the access to efflux proteins such as p-glycoprotein, enabling the oral route to be used for the administration of some drugs limited to parenteral route administration. Besides, SLN drug formulations afford the possibility of targeting drugs to the lymph for the treatment of lymphatic cancers. Finally, the lymphatic system acts as a prolonged-release system, affording a sufficiently high level of administered drug even after 24 h.75! 108 In previous research, idarubicin-SLN administered by the duodenal route gave
REVIEW
SLN-anticancer systems are generally injected either intravenously, intraperitoneally or administered orally. The route of administration is an important parameter in the delivery of SLN-encapsulated anticancer agents to tumours. Some hypotheses highlight the ability of SLN to enhance the oral bioavailability of actives by either physical-chemical or biochemical mechanisms. Physical-chemical mechanisms are based on enhancing the solubility of the drug. On the other hand, the biochemical mechanisms base their success in the inhibition of the efflux transporters. Different routes of administration may result in varying effects on the SLN biodistribution pattern. Reddy et al.105 studied the biodistribution and tumour uptake of etoposide-SLN after s.c., i.v. or intraperitoneal administration to tumour bearing BALB/c mice. The obtained results showed that s.c. injection reduced the distribution of etoposide-SLN to all the tissues studied, whereas after intraperitoneal injection, the distribution of etoposide-SLN to tissues was higher. The intravenous injection also resulted in lower concentrations of etoposide nanoparticles in the organs of RES. The influence of the administration route on the bioavailability of SLN must be explained before studying the SLN biodistribution profile and tumour uptake.
Lipid Nanomedicines for Anticancer Drug Therapy
13
15
Introduction. Lipid nanomedicines for anticancer drug therapy
Lipid Nanomedicines for Anticancer Drug Therapy
Estella-Hermoso deMendoza Mendozaetetal. al.
SLN
Adhesion
REVIEW
Anchoring of lipases
Drug release + Lipid degradation products
Degradation drug release
ABSORPTION OF DRUG AND LIPIDS
Micelle, drug solubilized + Bile salts
Mixed micelle
Gut wall Blood
Gut Fig. 3.
Mechanisms of absorption promoting effect of lipids formulated as a lipid nanoparticle.
higher idarubicin levels (around 40%) than those after i.v. administration of the same formulation at 24 h after administration.71 The surfactants contribute to an increase in the permeability of the intestinal membrane or improved the affinity between lipid particles and the intestinal membrane. In previous research,111 the absorption was greatly improved by increasing the amount of Pluronic F68 and Tween 80 on SLN nanoparticles. SLN show a certain physical stability in the gut lumen and lymph ensures a partial protection of the incorporated 14
Mucus
drugs.107 The physical stability of SLN reduces the local adverse effect of some drugs on the gastrointestinal tract, and protects biodegradable molecules, such as peptides and proteins, when orally administered. The encapsulation of anticancer drugs in lipid nanocapsules has also led to an improvement of oral exposure compared to the free drug. The inhibition of the P-glycoprotein by the surfactants contained in the lipid nanocapsules could explain the improvement of oral paclitaxel exposure when it was administered in lipid nanocapsules to male Sprague rats.112 J. Biomed. Nanotechnol. 5, 1–21, 2009
16
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
6.2.2. Parenteral Administration of SLN
6.2.3. Topical Administration of SLN SLN are considered to be the latest generation of colloidal drug carriers for topical use. Compared with other vehicles, such as creams, tinctures and emulsions, and other delivery systems like liposomes and microparticles, they combine such advantages as local tolerance, good adhesive properties, the protection of the encapsulated drugs against chemical decomposition, and the possibility of modulating the drug release.20 The bioavailability of drugs in nanoparticulate carriers may be increased by the close contact between the small particle size of nanoparticles and stratum corneum. However, the limited voids in the lipid bilayers of stratum corneum make the diffusion of intact particles into the skin impossible. SLN have a more favourable occlusive effect than other pharmaceutical preparations and hence the particle permeation through the skin is increased. SLN form adhesive films of densely packed spheres on the surface of the skin, increasing skin hydration.113 J. Biomed. Nanotechnol. 5, 1–21, 2009
The enhanced skin permeation of the SLN is probably due to the additives included in the formulation. Surfactants, which can loosen or fluidize the lipid bilayers of the stratum corneum, may act as permeation enhancers.20 The biodistribution data after subcutaneous injection (s.c.) injection of etoposide-SLN to tumour bearing BALB/c mice showed that the distribution of encapsulated drug was slow from the injection site. The increase in the concentration of etoposide in tissues at 6 h and 24 h postinjection indicates the etoposide-SLN residence at the injection site for a longer time.105 However, in most studies the evaluation of the observed pharmacological effect is the only procedure to assess the ability of a delivery system in transdermal applications. In vitro measurements can be useful to assert SLN occlusive effects.22 6.2.4. Intratumour Administration of SLN Few studies have been performed to study the effectiveness of SLN after direct intratumour administration. Mitoxantrone is example of drug that after s.c. injection induces more serious local toxicity than intravenous injection or intraperitoneal injection (i.p.). After multiple s.c administration of mitoxantrone loaded SLN every four days, for 28 days, it could be concluded that mitoxantrone-SLN had a strong affinity to the tumour tissue, contributing to the therapeutic effect of the antitumour agent. After mitoxantrone solution s.c. administration, no such affinity to the tumour surface was observed.5
REVIEW
SLN can be used for all parenteral applications suitable for polymeric nanoparticles.22 As mentioned before, the particle size of intravenously administered drug must be below 5 !m to avoid blocking of fine capillaries leading to embolism. Because of their minimum size below 400 nm, SLN formulations can be used for systemic body distribution with a minimized risk of blood clotting and aggregation. Using SLN as carriers, a three to five-fold enhancement of i.v. plasma peaks of the encapsulated drug as well as a prolonged residence time are generally observed after SLN administration. Uptake of the nanoparticles in cells of the RES or mononuclear phagocyte system differed according to the size and composition of the particles. Uptake of SLN can be avoided by PEGylation leading to long circulating particles, so-called stealth SLN. Surprisingly, stealth SLN showed similar circulating time and pharmacokinetic behaviour in comparison to unmodified nanoparticles.4 The similar low surface hydrophobicity of both types of particles could explain the results obtained when stealth and non-stealth doxorubicin SLN were evaluated in vivo. Minimizing opsonin protein binding is the key point for developing a long circulation nanoparticle formulation. The nature of the lipid material and the surfactant employed in the SLN preparation might play an important role to avoid adsorption of blood proteins that mediates liver uptake.23 The incorporated drug is gradually released on erosion (e.g., degradation by enzymes) or by diffusion from the particles. The rate of release may be controlled by the nature of the lipid material, particle size, and choice of surfactant and also by inner structure of SLN34 as discussed above.
Lipid Nanomedicines for Anticancer Drug Therapy
6.3. Pharmacokinetics Administration of anticancer drugs incorporated in SLN leads to different pharmacokinetic profiles to those obtained after free drug administration. When the drugs are encapsulated in SLN, the drugs are protected from chemical degradation in gastrointestinal tract, bypass the effect of extrusion membrane proteins, reduce the drug exposure to hepatic metabolizing enzyme activity and avoid the renal clearance due to increased size. Reduced liver metabolism and renal clearance of drugs encapsulated in SLN often result in prolonged blood circulation, with an increased chance of accumulation in the target tissue. Table V summarizes the key pharmacokinetic parameters in some studies of anticancer agents delivered by SLN. All non-stealth SLN formulations led to significantly higher levels of drug remaining in the systemic circulation for longer periods of time. Using SLN as carriers, mostly a three to twenty-fold enhancement of drug plasma peaks is observed when SLN are administered by i.v. route. The terminal half-lives of the encapsulated drugs are also significantly increased when administered via SLN. Plasma concentration-time profile for SLN shows a bi-exponential curve with high AUC, a lower rate of clearance, and a smaller volume of distribution in comparison to the free drug.107" 109" 110" 114 The biphasic behaviour can probably 15
17
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso de Mendoza et al.et al. Mendoza
Lipid Nanomedicines for Anticancer Drug Therapy Table V.
Pharmacokinetic parameters of anticancer drugs delivered by lipid nanoformulations. AUC (!g l−1 h)
Animal
Drug
Solution
Non-stealth
Stealth
Solution
Non-stealth
Stealth
References
Mice Rat Rabbit Rat Mice Rat Rat Mice Mice Rat Mice Mice
Camptothecin Doxorubicin Doxorubicin Doxorubicin FuDR Idarubicin (iv) Idarubicin (or) Paclitaxel (or) Paclitaxel (iv) Tamoxifen (iv) Metothrexate (iv) Docetaxel (iv)
66a 5"88 2"86 5"02 0"04 6"51 1"88 0"382 0"624 0"004 1"185 7200
324a 102"8 7"39 48"87 0"138 27"53 40"45 1"05 1"15 0"021 1"987 N.A.
N.A. N.A. 15.6–25.99b 67.27 N.A. N.A. N.A. N.A. N.A. N.A. N.A. 29100
N.A. 2"54 0"80 3"72 2"10 0"79 1"16 N.A. 1"36 2"68 8"2 0"3
0"36 3"36 1"19 4"11 19"0 23"51 7"22 4"5 10"06 9"58 14"46 N.A.
N.A. N.A. 1.20–1.83b 3.52 N.A. N.A. N.A. N.A. N.A. N.A. N.A. 1.4
58 109 4 4 119 71 71 112 104 120 16 52
a
REVIEW
Terminal half-life (h)
Unit of AUC is h ng g−1 . b Stealth SLN with three different levels of stealth agent (PEG2000-stearic acid). N.A. — Not applicable.
be explained by the slow distribution of SLN to organs and tissues. It is interesting to note that sometimes two peaks were observed in the concentration-time profile of encapsulated drug after SLN administration, whichever the administration route was. The first peak of encapsulated drug in blood took place immediately after SLN intravenous administration and for 1–2 h after oral SLN administration which indicated that SLN were absorbed quickly from the gastrointestinal tract into systemic circulation. SLN concentrations then began to decrease, which may be related to the uptake of SLN by the macrophages,115# 116 the distribution of SLN among particular organs and the release of encapsulated drug. The second peak occurred at about 6–8 h, and the maximum concentrations were lower than that of the first drug peak. The second peak of the encapsulated drug may be caused by the re-distribution of SLN released from those particular organs into systemic circulation.110 The interaction of SLN with the major circulatory protein, serum albumin, has been investigated in the present decade. Albumin adsorption on the particle surface formed a capping layer of 17 nm. The SLNs are protected by this layer against flattening on surfaces. At physiological albumin concentrations (35–50 g/l), the increased size was not important enough to explain blood cell aggregation.117 After oral administration of SLN to rats, the bioavailability of the encapsulated drug increased considerably. The AUC values obtained after administration of idarubicin-SLN was higher (around 21-fold) than that of idarubicin free solution.71 The administration route affects the bioavailability of SLN encapsulated drugs strongly. The AUC of SLN of idarubicin after administration into the duodenum was higher than that obtained after SLN i.v. administration.71 The difference between the two administration routes for the most part lies in the predominant transmucosal transport of SLN to the lymph; lymphatic transport predominantly contributes to SLN absorption, which could affect the drug bioavailability. 16
Coating SLN with PEG2000 further improved their abilities to evade the RES clearance.58# 71# 109 The AUC values significantly increased after SLN were coated with the stealth agents (Table V). However, the increase in circulation time was generally not as substantial. The opposite effect was observed when anticancer agents were formulated in lipid nanocapsules. Non-PEGylated docetaxel lipid nanocapsules were rapidly cleared from the systemic circulation, whereas the increase of PEG content from 6 to 10% gave a three-fold elevation of docetaxel blood levels at the same time after intravenous administration on female BALB/c mice inoculated with C26 colon adenocarcinoma cells.52 6.4. Tissue Distribution Like most drugs, anticancer drugs only exercise their therapeutic effect when they diffusses into the tumour. Therefore it is very important to know where exactly an anticancer drug has been distributed. Unfortunately, in vivo efficacy studies of SLN loaded anticancer drugs have so far been rare. Up until now, the study of the biodistribution profile of anticancer drug-loaded SLN in tumour-bearing animals has been considered an adequate tool to predict the therapeutic effect of the loaded drug. The tissue distribution profiles of SLN encapsulated drug differ significantly from the observed profile after administration of the free anticancer drug. With the exception of the brain, concentrations of drugs encapsulated are lower after SLN administration than with the obtained after drug free administration, especially for the liver, spleen, and kidneys, heart and lungs. SLN are widely distributed in muscles, bone, and fatty tissues.58# 108# 114 The reason for this different pattern of body distribution is not so far completely understood. The slower diffusion of SLN through biological barriers and the slower drug release from the SLN are the main reasons that could explain this. J. Biomed. Nanotechnol. 5, 1–21, 2009
18
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
J. Biomed. Nanotechnol. 5, 1–21, 2009
Stealth SLN increases the brain concentration of the encapsulated drug. Zara et al.71 evaluated the effect of the stealth agent concentration in the brain uptake of doxorubicin SLN in rabbits. They observed an increase in the brain concentration of doxorubicin on increasing the stealth agent. The amount of drug present in the rabbit brain ranged from 27.5 ng/g for nonstealth SLN to 242.0 ng/g for stealth SLN preparation loaded with 0.45% PEG after 30 min. The presence of surfactant agent in the SLN improves its brain uptake. The brain uptake of Poloxamer 188 stabilized stearic acid camptothecin-loaded SLN was studied after both oral and i.v. administration in mice. The aforementioned particles showed a significant brain uptake, and the maximum concentration (Cmax ) increased by 180% when compared to the Cmax of the drug solution.114 Surface charged SLN were also proposed in order to achieve brain targeting. Positively (surface charge: +5 mV) and negatively (surface charge: −47 mV) charged tripalmitin SLN were loaded with labelled etoposide and their brain uptake was compared to that of free labeled drug in mice after i.v. administration. Positively charged SLN showed a higher brain accumulation compared with both negatively charged SLN and free labeled drug. The etoposide concentration, achieved using positively charged SLN, was 10- and 14-fold higher with respect to the negatively charged SLN at 1 and 4 hours after administration, respectively.60 Unfortunately, the strategy of using surface charged SLN to cross the brain blood barrier has shown some important drawbacks such as the toxic effects at brain microvasculature endothelium of some cationic nanoparticles, which limit its usefulness.27
REVIEW
In general, the major limiting factor for the systemic use of nanoparticles is their rapid clearance from the blood circulation by the RES. The physicochemical properties of SLN are responsible for its wide distribution profile. SLN, particularly those in the range of 120–200 nm, are not taken up readily by the cells of the RES, and thus bypass liver and spleen filtration. Moreover, controlled release of the incorporated drug can be achieved for up to several weeks.26 The reduced volume of distribution of SLN is probably one of the reasons for the low amount of free drug present in tissues after SLN administration and could be related to slower distribution through biological barriers and/or the slow release from the SLN in the targeted organs.4 In particular, lower concentrations of free drug were observed in the heart after SLN than after free drug administration. Therefore, SLN might be useful to reduce the toxicity and to increase the clinical efficacy of anticancer drugs. Higher drug concentrations were detected in the brain in all of the studies when the drugs were encapsulated and delivered in SLN. Moreover, the low intrinsic toxicity and biodegradability of lipids used in SLN is a valuable feature in brain tumour treatment. In the late 1990s, SLN were proposed for brain drug targetting application independently by two research groups.109! 114 They developed scientific research to study the pharmacokinetics of two anticancer agents, namely camptothecin and doxorubicin, after the administration of free drug solution and drugloaded SLN to rats, observing drug accumulation into the brain after both oral and i.v. administration when loaded into SLN. Following SLN i.v. administration, the maximum concentration (Cmax ) increased by 180% when compared to the Cmax of the drug solution. The area under the curve (AUC)/dose and the MRT of SLN were 10.4- and 4-fold higher, respectively.58 In another independent study, stearic acid labelled SLN was found in rat CSF 20 min after i.v. administration. 30 min after administration of stealth SLN, the same doxorubicin concentration (around 10 "g/g of tissue) was found in the brain, heart, liver, lungs and spleen. However, after 24 hours brain doxorubicin concentrations remained higher than those observed in heart, liver, lungs and spleen. In recent years, various mechanisms for nanoparticle mediated drug uptake by the brain have been outlined. These include the enhancement of drug carrier retention in the brain blood capillaries by adsorption on to the capillary walls, resulting in a high concentration gradient across the blood brain barrier; the opening of tight junctions due to the presence of nanoparticles; the transcytosis of nanoparticles through the endothelium; solubilization of endothelial cell membrane lipids and membrane fluidization due to surfactant effects of polysorbates present in the drug carrier, or inhibition of efflux system of the blood brain barrier, especially P-glycoprotein.25 However, the mechanism for SLN uptake is not completely understood.
Lipid Nanomedicines for Anticancer Drug Therapy
6.4.1. Tumour Accumulation Several improvements in tumour drug accumulation have been observed when anticancer drugs are administered loaded to SLN. Elevated tumour etoposide concentrations were found when SLN were i.v. administered, compared to free drug to Dalton’s lymphoma tumour-bearing mice (67% increase 1 h post-injection, 30% increase 24 h postinjection).105 The accumulation of docetaxel in tumours was 1.7 to 2.4 times higher compared with the free drug at 6 h after intravenous injection to female nude mice bearing hepatoma.68 AUC values of DNA alkylating agent chlorambucil loaded SLN in tumour were significantly (p < 0.01) higher than the obtained after i.v. administration of the free drug to C57 BL/6 male mice subcutaneously inoculated with colon-38 tumour fragments from donor mice.67 The high accumulation of the anticancer drugs loaded in SLN could be attributed to following reasons; first, encapsulation of anticancer drugs into SLN resulted in higher accumulation in both the liver and the tumour. SLN can increase drug concentration in tumour via the previously explained EPR effect. Nanoparticles with a diameter below 200 nm can significantly accumulate in tumours 17
19
Introduction. Lipid nanomedicines for anticancer drug therapy
REVIEW
Lipid Nanomedicines for Anticancer Drug Therapy
by this mechanism.118 Second, the internalization of SLN was further enhanced by the target of the SLN to specific carcinoma cells by modifying the nanoparticles with a ligand that can bind to specific receptor. As it has previously been mentioned in this review, the requirements of the antigen that will allow molecular targetting are the exclusive expression in cancer cells, to be an integral part of an essential cellular function of the cancer cells and not to easily mutate as the cancer cells proliferate.53 Docetaxel incorporated in galactosylated SLN showed a tumour increased accumulation in the treatment of hepatocellular carcinomas expressing the asialoglycoprotein receptor compared with unmodified SLN.68 On the other hand, a folate receptor-targeted SLN system significantly improved in vivo tumour growth inhibition and tumour bearing animal survival compared to non-targetted SLN.51 These findings will likely lead to the development of more targeted-SLN formulations designed for anticancer use. Significant improvements in tumour drug accumulation were observed when the etoposide formulations were injected intraperitoneally or subcutaneously rather than when intravenously administered.105 The slower and progressive penetration of the nanoparticles (and hence the loaded drug) from the peritoneum or the s.c. injection site into the tumour may result in more favourable patterns of drug distribution. The route of administration of a SLN formulation will be a key aspect to consider when designing animal or clinical studies using SLN for anticancer drug delivery. In another study, a 5-fold increase in AUCtumor compared to the free drug was observed after i.v. administration of PEGylated docetaxel-lipid nanocapsules to female BALB/c mice inoculated with C26 colon adenocarcinoma cells.52 Finally, a study was carried out with local injection of mitoxantrone SLN against breast cancer and its lymph node metastases.5 The drug concentration using SLN was much higher in local lymph nodes, and the drug concentration in other tissue was lower than that of mitoxantrone solution.
7. CONCLUDING REMARKS SLN-based anticancer drug delivery vehicles have been shown to be more efficient than conventional drug solutions, not only because the drug efficacy is improved while its toxicity is reduced, but also, because better pharmacokinetics and drug biodistribution profile were observed when the antitumour agent was encapsulated in these systems. Moreover, SLN have emerged as an effective alternative to liposomes, not only because of their better stability profile but also because they are easily scalable; since they are based on low cost biocompatible lipid excipients, their commercialisation will be easier compared to liposomes or polymeric nanoparticles. 18
Estella-Hermoso de Mendoza et al.et al. Mendoza
A certain number of lipid nanoparticulate systems have shown abilities to treat cancers that are usually refractory to cytotoxic drug treatment. Even though there is no SLNbased product for the treatment of cancer on the market so far, it can be expected that new modified types of SLN, such as NLC, LDC, stealth SLN, targeted-SLN, and SLN loaded with drug combinations, will be perfected and utilized to further improve the efficacies and side effect profiles of anticancer drugs. Consequently, the introduction of SLN to chemotherapy offers great promise for the management of cancerous diseases. Acknowledgments: We gratefully acknowledge support from the Spanish Ministry of Science and Innovation (SAF2007-61261, SAF2008-02251, PCT090100-2007-27) and Caja Navarra Foundation. Ander Estella-Hermoso de Mendoza is supported by the research grant from the Department of Education of the Basque Government (BF106.37).
References and Notes 1. R. B. Ewesuedo and M. J. Ratain, Oncologic therapeutics, edited by E. E. Vokes and H. M. Golomb (eds.). Springer, New York (2003), pp. 19–66. 2. C. H. Takimoto and E. Calvo, Cancer management: A multidisciplinary approach, 10th edn., edited by R. Pazdur, L. R. Coia, W. J. Hoskins, and L. D. Wagman (eds.). CMPMedica (2007). 3. G. Fontana, L. Maniscalco, D. Schillaci, G. Cavallaro, and G. Giammona, Solid lipid nanoparticles containing tamoxifen characterization and in vitro antitumoral activity. Drug Deliv. 12, 6 (2005). 4. A. Fundaro, R. Cavalli, A. Bargoni, D. Vighetto, G. P. Zara, and M. R. Gasco, Non-stealth and stealth solid lipid nanoparticles (SLN) carrying doxorubicin: pharmacokinetics and tissue distribution after i.v. administration to rats. Pharmacol. Res. 42, 4 (2000). 5. B. Lu, S. B. Xiong, H. Yang, X. D. Yin, and R. B. Chao, Solid lipid nanoparticles of mitoxantrone for local injection against breast cancer and its lymph node metastases. Eur. J. Pharm. Sci. 28, 1 (2006). 6. L. Serpe, M. G. Catalano, R. Cavalli, E. Ugazio, O. Bosco, R. Canaparo, E. Muntoni, R. Frairia, M. R. Gasco, M. Eandi, and G. P. Zara, Cytotoxicity of anticancer drugs incorporated in solid lipid nanoparticles on HT-29 colorectal cancer cell line. Eur. J. Pharm. Biopharm. 58, 3 (2004). 7. H. L. Wong, R. Bendayan, A. M. Rauth, Y. Li, and X. Y. Wu, Chemotherapy with anticancer drugs encapsulated in solid lipid nanoparticles. Adv. Drug Deliv. Rev. (2007). 8. W. J. Houlihan, M. Lohmeyer, P. Workman, and S. H. Cheon, Phospholipid antitumor agents. Med. Res. Rev. 15, 3 (1995). 9. F. Mollinedo, J. L. Fernandez-Luna, C. Gajate, B. Martin-Martin, A. Benito, R. Martinez-Dalmau, and M. Modolell, Selective induction of apoptosis in cancer cells by the ether lipid ET-18-OCH3 (Edelfosine): Molecular structure requirements, cellular uptake, and protection by Bcl-2 and Bcl-X(L). Cancer Res. 57, 7 (1997). 10. W. E. Berdel and S. Okamoto, Ether lipids in cancer chemotherapy. Keio J. Med. 39, 2 (1990). 11. A. Estella-Hermoso de Mendoza, M. A. Campanero, J. de la Iglesia-Vicente, C. Gajate, F. Mollinedo, and M. J. Blanco-Príeto, Anti-tumour alkyl ether lipid edelfosine: Tissue distribution and pharmacokinetic behaviour in healthy and tumour-bearing immune suppressed mice. Clin. Cancer Res. (2009), In press.
J. Biomed. Nanotechnol. 5, 1–21, 2009
20
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
J. Biomed. Nanotechnol. 5, 1–21, 2009
34. M. Muchow, P. Maincent, and R. H. Muller, Lipid nanoparticles with a solid matrix (SLN, NLC, LDC) for oral drug delivery. Drug Dev. Ind. Pharm. 34, 12 (2008). 35. Z. R. Huang, S. C. Hua, Y. L. Yang, and J. Y. Fang, Development and evaluation of lipid nanoparticles for camptothecin delivery: A comparison of solid lipid nanoparticles, nanostructured lipid carriers, and lipid emulsion. Acta Pharmacol. Sin. 29, 9 (2008). 36. V. Jenning, K. Mader, and S. H. Gohla, Solid lipid nanoparticles (SLN) based on binary mixtures of liquid and solid lipids: a (1)HNMR study. Int. J. Pharm. 205, 1 (2000). 37. H. L. Wong, A. M. Rauth, R. Bendayan, J. L. Manias, M. Ramaswamy, Z. Liu, S. Z. Erhan, and X. Y. Wu, A new polymer-lipid hybrid nanoparticle system increases cytotoxicity of doxorubicin against multidrug-resistant human breast cancer cells. Pharm. Res. (2006). 38. R. H. Muller and C. M. Keck, Challenges and solutions for the delivery of biotech drugs–a review of drug nanocrystal technology and lipid nanoparticles. J. Biotechnol. 113, 1 (2004). 39. S. K. Jain, A. Chaurasiya, Y. Gupta, A. Jain, P. Dagur, B. Joshi, and V. M. Katoch, Development and characterization of 5-FU bearing ferritin appended solid lipid nanoparticles for tumour targeting. J. Microencapsul. 25, 5 (2008). 40. S. Markert, S. Lassmann, B. Gabriel, M. Klar, M. Werner, G. Gitsch, F. Kratz, and A. Hasenburg, Alpha-folate receptor expression in epithelial ovarian carcinoma and non-neoplastic ovarian tissue. Anticancer Res. 28, 6A (2008). 41. C. O. Evans, C. Yao, D. Laborde, and N. M. Oyesiku, Folate receptor expression in pituitary adenomas cellular and molecular analysis. Vitam. Horm. 79 (2008). 42. Y. Lu, E. Sega, and P. S. Low, Folate receptor-targeted immunotherapy: Induction of humoral and cellular immunity against haptendecorated cancer cells. Int. J. Cancer. 116, 5 (2005). 43. C. L. Denholt, P. R. Hansen, N. Pedersen, H. S. Poulsen, N. Gillings, and A. Kjaer, Identification of novel peptide ligands for the cancer-specific receptor mutation EFGRvIII using a mixturebased synthetic combinatorial library. Biopolymers 91, 3 (2009). 44. D. B. Cornelio, L. Meurer, R. Roesler, and G. Schwartsmann, Gastrin-releasing peptide receptor expression in cervical cancer. Oncology 73, 5 (2007). 45. K. Zhang, M. R. Aruva, N. Shanthly, C. A. Cardi, C. A. Patel, S. Rattan, G. Cesarone, E. Wickstrom, and M. L. Thakur, Vasoactive intestinal peptide (VIP) and pituitary adenylate cyclase activating peptide (PACAP) receptor specific peptide analogues for PET imaging of breast cancer: In vitro/in vivo evaluation. Regul. Pept. 144, 1 (2007). 46. M. G. Caruso, C. Messa, A. Orlando, B. D’Attoma, and M. Notarnicola, Early induction of LDL receptor gene expression by genistein in DLD-1 colon cancer cell line. Fitoterapia 79, 7 (2008). 47. C. Peterson, S. Vitols, M. Rudling, H. Blomgren, F. Edsmyr, and L. Skoog, Hypocholesterolemia in cancer patients may be caused by elevated LDL receptor activities in malignant cells. Med. Oncol. Tumor Pharmacother. 2, 3 (1985). 48. H. Yuan, J. Miao, Y. Z. Du, J. You, F. Q. Hu, and S. Zeng, Cellular uptake of solid lipid nanoparticles and cytotoxicity of encapsulated paclitaxel in A549 cancer cells. Int. J. Pharm. 348, 1 (2008). 49. X. G. Zhang, J. Miao, Y. Q. Dai, Y. Z. Du, H. Yuan, and F. Q. Hu, Reversal activity of nanostructured lipid carriers loading cytotoxic drug in multi-drug resistant cancer cells. Int. J. Pharm. 361, 1 (2008). 50. S. M. Stephenson, P. S. Low, and R. J. Lee, Folate receptormediated targeting of liposomal drugs to cancer cells. Methods Enzymol. 387 (2004). 51. P. J. Stevens, M. Sekido, and R. J. Lee, A folate receptor-targeted lipid nanoparticle formulation for a lipophilic paclitaxel prodrug. Pharm. Res. 21, 12 (2004).
REVIEW
12. F. Mollinedo, C. Gajate, S. Martin-Santamaria, and F. Gago, ET-18OCH3 (edelfosine): A selective antitumour lipid targeting apoptosis through intracellular activation of Fas/CD95 death receptor. Curr. Med. Chem. 11, 24 (2004). 13. C. Gajate and F. Mollinedo, The antitumor ether lipid ET-18OCH(3) induces apoptosis through translocation and capping of Fas/CD95 into membrane rafts in human leukemic cells. Blood. 98, 13 (2001). 14. J. Kotting, N. W. Marschner, W. Neumuller, C. Unger, and H. Eibl, Hexadecylphosphocholine and octadecyl-methyl-glycero3-phosphocholine: A comparison of hemolytic activity, serum binding and tissue distribution. Prog. Exp. Tumor Res. 34 (1992). 15. S. London, Intermittent oxaliplatin dosing prolongs time to treatment failure. Oncol. News Int. (2008). 16. K. Ruckmani, M. Sivakumar, and P. A. Ganeshkumar, Methotrexate loaded solid lipid nanoparticles (SLN) for effective treatment of carcinoma. J. Nanosci. Nanotechnol. 6, 9 (2006). 17. P. Speiser, Lipidnanopellets als tragersystem fur arzneimittel zur peroralen anwendung. Eur. Patent Appl. EP 0,167,825 (1986). 18. M. R. Gasco, Method for producing solid lipid microspheres having a narrow distribution. No. 5,250,236, (1993). 19. R. H. Müller and J. S. Lucks, Medication vehicles made of solid lipid particles (solid lipid nanospheres—SLN). No. 0605497, (1993). 20. J. Y. Fang, C. L. Fang, C. H. Liu, and Y. H. Su, Lipid nanoparticles as vehicles for topical psoralen delivery: Solid lipid nanoparticles (SLN) versus nanostructured lipid carriers (NLC). Eur. J. Pharm. Biopharm. 70, 2 (2008). 21. W. Mehnert and K. Mader, Solid lipid nanoparticles: production, characterization and applications. Adv. Drug Deliv. Rev. 47, 2 (2001). 22. R. H. Muller, K. Mader, and S. Gohla, Solid lipid nanoparticles (SLN) for controlled drug delivery—A review of the state of the art. Eur. J. Pharm. Biopharm. 50, 1 (2000). 23. S. S. Shidhaye, R. Vaidya, S. Sutar, A. Patwardhan, and V. J. Kadam, Solid lipid nanoparticles and nanostructured lipid carriers– innovative generations of solid lipid carriers. Curr. Drug Deliv. 5, 4 (2008). 24. C. Olbrich and R. H. Muller, Enzymatic degradation of SLN-effect of surfactant and surfactant mixtures. Int. J. Pharm. 180, 1 (1999). 25. Y. Chen, G. Dalwadi, and H. A. Benson, Drug delivery across the blood-brain barrier. Curr. Drug Deliv. 1, 4 (2004). 26. I. P. Kaur, R. Bhandari, S. Bhandari, and V. Kakkar, Potential of solid lipid nanoparticles in brain targeting. J. Control. Rel. (2008). 27. P. R. Lockman, J. M. Koziara, R. J. Mumper, and D. D. Allen, Nanoparticle surface charges alter blood-brain barrier integrity and permeability. J. Drug Target. 12, 9 (2004). 28. R. H. Muller and C. M. Keck, Drug delivery to the brain–realization by novel drug carriers. J. Nanosci. Nanotechnol. 4, 5 (2004). 29. H. Bunjes, F. Steiniger, and W. Richter, Visualizing the structure of triglyceride nanoparticles in different crystal modifications. Langmuir. 23, 7 (2007). 30. A. Estella-Hermoso de Mendoza, M. Rayo, F. Mollinedo, and M. J. Blanco-Prieto, Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: Development and in vitro characterization. Eur. J. Pharm. Biopharm. 68, 2 (2008). 31. E. B. Souto, W. Mehnert, and R. H. Muller, Polymorphic behaviour of Compritol888 ATO as bulk lipid and as SLN and NLC. J. Microencapsul. 23, 4 (2006). 32. K. Manjunath, J. S. Reddy, and V. Venkateswarlu, Solid lipid nanoparticles as drug delivery systems. Meth. Find. Exp. Clin. Pharmacol. 27, 2 (2005). 33. A. Zur Muhlen, C. Schwarz, and W. Mehnert, Solid lipid nanoparticles (SLN) for controlled drug delivery–drug release and release mechanism. Eur. J. Pharm. Biopharm. 45, 2 (1998).
Lipid Nanomedicines for Anticancer Drug Therapy
19
21
Introduction. Lipid nanomedicines for anticancer drug therapy
REVIEW
Lipid Nanomedicines for Anticancer Drug Therapy 52. M. N. Khalid, P. Simard, D. Hoarau, A. Dragomir, and J. C. Leroux, Long circulating poly(ethylene glycol)-decorated lipid nanocapsules deliver docetaxel to solid tumors. Pharm. Res. 23, 4 (2006). 53. R. Abou-Jawde, T. Choueiri, C. Alemany, and T. Mekhail, An overview of targeted treatments in cancer. Clin. Ther. 25, 8 (2003). 54. B. Stella, S. Arpicco, M. T. Peracchia, D. Desmaele, J. Hoebeke, M. Renoir, J. D’Angelo, L. Cattel, and P. Couvreur, Design of folic acid-conjugated nanoparticles for drug targeting. J. Pharm. Sci. 89, 11 (2000). 55. A. Beduneau, P. Saulnier, F. Hindre, A. Clavreul, J. C. Leroux, and J. P. Benoit, Design of targeted lipid nanocapsules by conjugation of whole antibodies and antibody Fab’ fragments. Biomaterials 28, 33 (2007). 56. Y. Gupta, A. Jain, and S. K. Jain, Transferrin-conjugated solid lipid nanoparticles for enhanced delivery of quinine dihydrochloride to the brain. J. Pharm. Pharmacol. 59, 7 (2007). 57. S. J. Lim and C. K. Kim, Formulation parameters determining the physicochemical characteristics of solid lipid nanoparticles loaded with all-trans retinoic acid. Int. J. Pharm. 243, 1 (2002). 58. S. C. Yang, L. F. Lu, Y. Cai, J. B. Zhu, B. W. Liang, and C. Z. Yang, Body distribution in mice of intravenously injected camptothecin solid lipid nanoparticles and targeting effect on brain. J. Control. Rel. 59, 3 (1999). 59. L. H. Reddy, J. S. Adhikari, B. S. Dwarakanath, R. K. Sharma, and R. R. Murthy, Tumoricidal effects of etoposide incorporated into solid lipid nanoparticles after intraperitoneal administration in Dalton’s lymphoma bearing mice. AAPS J. 8, 2 (2006). 60. L. H. Reddy, R. K. Sharma, K. Chuttani, A. K. Mishra, and R. R. Murthy, Etoposide-incorporated tripalmitin nanoparticles with different surface charge: formulation, characterization, radiolabeling, and biodistribution studies. AAPS J. 6, 3 (2004). 61. M. K. Lee, S. J. Lim, and C. K. Kim, Preparation, characterization and in vitro cytotoxicity of paclitaxel-loaded sterically stabilized solid lipid nanoparticles. Biomaterials 28, 12 (2007). 62. J. You, F. Wan, F. de Cui, Y. Sun, Y. Z. Du, and F. Q. Hu, Preparation and characteristic of vinorelbine bitartrate-loaded solid lipid nanoparticles. Int. J. Pharm. 343, 1 (2007). 63. Y. Hattori and Y. Maitani, Folate-linked lipid-based nanoparticle for targeted gene delivery. Curr. Drug Deliv. 2, 3 (2005). 64. K. Tabatt, C. Kneuer, M. Sameti, C. Olbrich, R. H. Muller, C. M. Lehr, and U. Bakowsky, Transfection with different colloidal systems: Comparison of solid lipid nanoparticles and liposomes. J. Control. Rel. 97, 2 (2004). 65. K. Tabatt, M. Sameti, C. Olbrich, R. H. Muller, and C. M. Lehr, Effect of cationic lipid and matrix lipid composition on solid lipid nanoparticle-mediated gene transfer. Eur. J. Pharm. Biopharm. 57, 2 (2004). 66. J. Tian, X. Pang, K. Yu, L. Liu, and J. Zhou, Preparation, characterization and in vivo distribution of solid lipid nanoparticles loaded with cisplatin. Pharmazie 63, 8 (2008). 67. P. Sharma, S. Ganta, W. A. Denny, and S. Garg, Formulation and pharmacokinetics of lipid nanoparticles of a chemically sensitive nitrogen mustard derivative: Chlorambucil. Int. J. Pharm. (2008). 68. Z. Xu, L. Chen, W. Gu, Y. Gao, L. Lin, Z. Zhang, Y. Xi, and Y. Li, The performance of docetaxel-loaded solid lipid nanoparticles targeted to hepatocellular carcinoma. Biomaterials 30, 2 (2009). 69. A. Miglietta, R. Cavalli, C. Bocca, L. Gabriel, and M. R. Gasco, Cellular uptake and cytotoxicity of solid lipid nanospheres (SLN) incorporating doxorubicin or paclitaxel. Int. J. Pharm. 210, 1 (2000). 70. G. P. Zara, R. Cavalli, A. Bargoni, A. Fundaro, D. Vighetto, and M. R. Gasco, Intravenous administration to rabbits of non-stealth and stealth doxorubicin-loaded solid lipid nanoparticles at increasing concentrations of stealth agent: pharmacokinetics and distribution of doxorubicin in brain and other tissues. J. Drug Target. 10, 4 (2002).
20
Mendoza Estella-Hermoso de Mendoza et al.et al. 71. G. P. Zara, A. Bargoni, R. Cavalli, A. Fundaro, D. Vighetto, and M. R. Gasco, Pharmacokinetics and tissue distribution of idarubicin-loaded solid lipid nanoparticles after duodenal administration to rats. J. Pharm. Sci. 91, 5 (2002). 72. L. Serpe, M. Guido, R. Canaparo, E. Muntoni, R. Cavalli, P. Panzanelli, C. Della Pepal, A. Bargoni, A. Mauro, M. R. Gasco, M. Eandi, and G. P. Zara, Intracellular accumulation and cytotoxicity of doxorubicin with different pharmaceutical formulations in human cancer cell lines. J. Nanosci. Nanotechnol. 6, 9 (2006). 73. M. Trotta, F. Debernardi, and O. Caputo, Preparation of solid lipid nanoparticles by a solvent emulsification-diffusion technique. Int. J. Pharm. 257, 1 (2003). 74. A. Del Pozo-Rodriguez, M. A. Solinis, A. R. Gascon, and J. L. Pedraz, Short- and long-term stability study of lyophilized solid lipid nanoparticles for gene therapy. Eur. J. Pharm. Biopharm. (2008). 75. R. Paliwal, S. Rai, B. Vaidya, K. Khatri, A. K. Goyal, N. Mishra, A. Mehta, and S. P. Vyas, Effect of lipid core material on characteristics of solid lipid nanoparticles designed for oral lymphatic delivery. Nanomedicine: Nanotechnology, Biology and Medicine. Corrected Proof (2008), In Press. 76. R. Cortesi, E. Esposito, G. Luca, and C. Nastruzzi, Production of lipospheres as carriers for bioactive compounds. Biomaterials 23, 11 (2002). 77. M. Uner, Preparation, characterization and physico-chemical properties of solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC): their benefits as colloidal drug carrier systems. Pharmazie 61, 5 (2006). 78. Y. Wang, Y. Deng, S. Mao, S. Jin, J. Wang, and D. Bi, Characterization and body distribution of beta-elemene solid lipid nanoparticles (SLN). Drug Dev. Ind. Pharm. 31, 8 (2005). 79. E. Zimmermann, R. H. Muller, and K. Mader, Influence of different parameters on reconstitution of lyophilized SLN. Int. J. Pharm. 196, 2 (2000). 80. W. N. Charman and V. J. Stella, Lymphatic Transport of Drugs, CRC Press (1992). 81. J. Kreuter, Physicochemical characterization of polyacrylic nanoparticles. Int. J. Pharm. 14, 1 (1983). 82. M. Nidhin, R. Indumathy, K. J. Sreeram, and B. U. Nair, Synthesis of iron oxide nanoparticles of narrow size distribution on polysaccharide templates. Bull. Mater. Sci. 31, 1 (2008). 83. P. Ahlin, J. Kristl, A. Kristl, and F. Vrecer, Investigation of polymeric nanoparticles as carriers of enalaprilat for oral administration. Int. J. Pharm. 239, 1 (2002). 84. J. Stecova, W. Mehnert, T. Blaschke, B. Kleuser, R. Sivaramakrishnan, C. C. Zouboulis, H. Seltmann, H. C. Korting, K. D. Kramer, and M. Schafer-Korting, Cyproterone acetate loading to lipid nanoparticles for topical acne treatment: particle characterisation and skin uptake. Pharm. Res. 24, 5 (2007). 85. J. Zhang, Y. Fan, and E. Smith, Experimental design for the optimization of lipid nanoparticles. J. Pharm. Sci. (2008). 86. N. Anton, J. P. Benoit, and P. Saulnier, Design and production of nanoparticles formulated from nano-emulsion templates-a review. J. Control. Rel. 128, 3 (2008). 87. C. Freitas and R. H. Muller, Effect of light and temperature on zeta potential and physical stability in solid lipid nanoparticles (SLN) dispersion. Int. J. Pharm. 168 (1998). 88. N. Gadegaard, Atomic force microscopy in biology: technology and techniques. Biotech. Histochem. 81, 2 (2006). 89. H. Zhang, W. Cui, J. Bei, and S. Wang, Preparation of poly(lactideco-glycolide-co-caprolactone) nanoparticles and their degradation behaviour in aqueous solution. Polym. Degrad. Stab. 91, 9 (2006). 90. A. Radomska-Soukharev, Stability of lipid excipients in solid lipid nanoparticles. Adv. Drug Deliv. Rev. 59, 6 (2007). 91. T. Kramer, D. M. Kremer, M. J. Pikal, W. J. Petre, E. Y. Shalaev, and L. A. Gatlin, A procedure to optimize scale-up for the primary drying phase of lyophilization. J. Pharm. Sci. 98, 1 (2009).
J. Biomed. Nanotechnol. 5, 1–21, 2009
22
Introduction. Lipid nanomedicines for anticancer drug therapy
Estella-Hermoso Mendoza et al. de Mendoza et al.
107. A. Bargoni, R. Cavalli, O. Caputo, A. Fundaro, M. R. Gasco, and G. P. Zara, Solid lipid nanoparticles in lymph and plasma after duodenal administration to rats. Pharm. Res. 15, 5 (1998). 108. N. L. Trevaskis, W. N. Charman, and C. J. Porter, Lipid-based delivery systems and intestinal lymphatic drug transport: a mechanistic update. Adv. Drug Deliv. Rev. 60, 6 (2008). 109. G. P. Zara, R. Cavalli, A. Fundaro, A. Bargoni, O. Caputo, and M. R. Gasco, Pharmacokinetics of doxorubicin incorporated in solid lipid nanospheres (SLN). Pharmacol. Res. 40, 3 (1999). 110. H. Yuan, J. Chen, Y. Z. Du, F. Q. Hu, S. Zeng, and H. L. Zhao, Studies on oral absorption of stearic acid SLN by a novel fluorometric method. Colloids Surf. B Biointerfaces 58, 2 (2007). 111. L. Hu, X. Tang, and F. Cui, Solid lipid nanoparticles (SLNs) to improve oral bioavailability of poorly soluble drugs. J. Pharm. Pharmacol. 56, 12 (2004). 112. S. Peltier, J. M. Oger, F. Lagarce, W. Couet, and J. P. Benoit, Enhanced oral paclitaxel bioavailability after administration of paclitaxel-loaded lipid nanocapsules. Pharm. Res. 23, 6 (2006). 113. H. Chen, X. Chang, D. Du, W. Liu, J. Liu, T. Weng, Y. Yang, H. Xu, and X. Yang, Podophyllotoxin-loaded solid lipid nanoparticles for epidermal targeting. J. Control. Rel. 110, 2 (2006). 114. S. Yang, J. Zhu, Y. Lu, B. Liang, and C. Yang, Body distribution of camptothecin solid lipid nanoparticles after oral administration. Pharm. Res. 16, 5 (1999). 115. D. E. Owens, 3rd and N. A. Peppas, Opsonization, biodistribution, and pharmacokinetics of polymeric nanoparticles. Int. J. Pharm. 307, 1 (2006). 116. C. E. Soma, C. Dubernet, G. Barratt, S. Benita, and P. Couvreur, Investigation of the role of macrophages on the cytotoxicity of doxorubicin and doxorubicin-loaded nanoparticles on M5076 cells in vitro. J. Control. Rel. 68, 2 (2000). 117. J. Gualbert, P. Shahgaldian, and A. W. Coleman, Interactions of amphiphilic calix[4]arene-based Solid Lipid Nanoparticles with bovine serum albumin. Int. J. Pharm. 257, 1 (2003). 118. N. Iodoshima, C. Udagawa, T. Ando, H. Fukuyasu, T. Watanabe, and S. Nakabayashi, Lipid nanoparticles for delivering antitumor drugs. Int. J. Pharm. 146 (1997). 119. J. X. Wang, X. Sun, and Z. R. Zhang, Enhanced brain targeting by synthesis of 3! ,5! -dioctanoyl-5-fluoro-2! -deoxyuridine and incorporation into solid lipid nanoparticles. Eur. J. Pharm. Biopharm. 54, 3 (2002). 120. L. H. Reddy, K. Vivek, N. Bakshi, and R. S. Murthy, Tamoxifen citrate loaded solid lipid nanoparticles (SLN): preparation, characterization, in vitro drug release, and pharmacokinetic evaluation. Pharm. Dev. Technol. 11, 2 (2006).
REVIEW
92. P. Yadava, M. Gibbs, C. Castro, and J. A. Hughes, Effect of lyophilization and freeze-thawing on the stability of siRNAliposome complexes. AAPS Pharm. Sci. Tech. 9, 2 (2008). 93. Y. Bensouda and A. Laatiris, The lyophilization of dispersed systems: influence of freezing process, freezing time, freezing temperature and RBCs concentration on RBCs hemolysis. Drug Dev. Ind. Pharm. 32, 8 (2006). 94. R. Seetharam, Y. Wada, S. Ramachandran, H. Hess, and P. Satir, Long-term storage of bionanodevices by freezing and lyophilization. Lab. Chip. 6, 9 (2006). 95. R. Cavalli, O. Caputo, and M. R. Gasco, Preparation and characterization of solid lipid nanospheres containing paclitaxel. Eur. J. Pharm. Sci. 10, 4 (2000). 96. C. Schwarz and W. Mehnert, Freeze-drying of drug-free and drugloaded solid lipid nanoparticles (SLN). Int. J. Pharm. 157, 2 (1997). 97. R. H. Müller, S. Maassen, C. Schwarz, and W. Mehnert, Solid lipid nanoparticles (SLN) as potential carrier for human use: interaction with human granulocytes. J. Control. Rel. 47, 3 (1997). 98. N. Scholer, H. Hahn, R. H. Muller, and O. Liesenfeld, Effect of lipid matrix and size of solid lipid nanoparticles (SLN) on the viability and cytokine production of macrophages. Int. J. Pharm. 231, 2 (2002). 99. H. Weyhers, S. Ehlers, H. Hahn, E. B. Souto, and R. H. Muller, Solid lipid nanoparticles (SLN)–effects of lipid composition on in vitro degradation and in vivo toxicity. Pharmazie 61, 6 (2006). 100. M. D. Joshi and R. H. Muller, Lipid nanoparticles for parenteral delivery of actives. Eur. J. Pharm. Biopharm. 71, 2 (2009). 101. P. K. Dudeja, K. M. Anderson, J. S. Harris, L. Buckingham, and J. S. Coon, Reversal of multidrug resistance phenotype by surfactants: relationship to membrane lipid fluidity. Arch. Biochem. Biophys. 319, 1 (1995). 102. D. M. Woodcock, M. E. Linsenmeyer, G. Chojnowski, A. B. Kriegler, V. Nink, L. K. Webster, and W. H. Sawyer, Reversal of multidrug resistance by surfactants. Br. J. Cancer. 66, 1 (1992). 103. E. Allard, C. Passirani, E. Garcion, P. Pigeon, A. Vessieres, G. Jaouen, and J. P. Benoit, Lipid nanocapsules loaded with an organometallic tamoxifen derivative as a novel drug-carrier system for experimental malignant gliomas. J. Control. Rel. 130, 2 (2008). 104. D. B. Chen, T. Z. Yang, W. L. Lu, and Q. Zhang, In vitro and in vivo study of two types of long-circulating solid lipid nanoparticles containing paclitaxel. Chem. Pharm. Bull. (Tokyo). 49, 11 (2001). 105. L. H. Reddy, R. K. Sharma, K. Chuttanib, A. K. Mishrab, and R. S. R. Murth, Influence of administration route on tumor uptake and biodistribution of etoposide loaded solid lipid nanoparticles in Dalton’s lymphoma tumor bearing mice. J. Control. Rel. 105 (2005). 106. N. Scholer, C. Olbrich, K. Tabatt, R. H. Muller, H. Hahn, and O. Liesenfeld, Surfactant, but not the size of solid lipid nanoparticles (SLN) influences viability and cytokine production of macrophages. Int. J. Pharm. 221, 1 (2001).
Lipid Nanomedicines for Anticancer Drug Therapy
Received: xx Xxxx xxxx. Accepted: xx Xxxx xxxx.
J. Biomed. Nanotechnol. 5, 1–21, 2009
21
23
Hypothesis and Objectives -‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐-‐ Hipótesis y Objetivos
25
Hypothesis and Objectives
HYPOTHESIS AND OBJECTIVES The synthetic analogues, ether phospholipids, also known as alkyl-‐ lysophospholipids (ALP), among which 1-‐O-‐octadecyl-‐2-‐O-‐methyl-‐rac-‐glycero-‐3-‐ phosphocholine (ET-‐18-‐OCH3, edelfosine) is the prototypical compound, constitute a promising class of oral anti-‐tumor substances which do not have DNA as their target, but rather act at the level of the membrane. However, due to their chemical structure, these molecules present a dose depending hemolysis and gastrointestinal toxicity when they are administered at high doses, as well as low oral bioavailability. On the other hand, lipid nanoparticles (LN) have arisen as an interesting proposal for the oral administration of drugs, since due to their characteristics they facilitate the solubilization of highly lipophilic substances in addition to favoring the absorption of the drugs in the intestine. In fact, these particles have been used to increase or improve the oral bioavailability of numerous active principles, since due to their small size and composition LN can penetrate into the spaces between the villi of the intestinal mucosa and thus reach the blood stream. The use of these nanotransporters for the transport of anti-‐tumor agents additionally offers an important advantage: their tumor specificity due to the EPR effect (“enhanced permeability and retention effect”) and therefore their lower toxicity. On the basis of the foregoing, the initial hypothesis of this work is: The use of LN of edelfosine may produce a highly significant increase in the anti-‐ tumor efficacy of this compound, both because it may propitiate absorption in the intestine, and because it would reduce the dose of active principle needed, since high daily doses would not be necessary to maintain significant drug levels in the blood, which would also lessen any possible toxic effects.
27
Hypothesis and Objectives
In short, it is expected that oral administration of edelfosine inside these nanotransporters will result in an increase in the bioavailability of the drug, and a reduction in both the required dose and the possible adverse reactions associated with this drug. The general aim of this study is to design, develop and evaluate lipid nanotransporters containing edelfosine both in vitro and in vivo.
More specifically, the partial objectives are as follows: 1. Determination of the pharmacokinetic profile of free edelfosine and its biodistribution in mantle cell lymphoma-‐bearing and non mantle cell lymphoma-‐bearing animal models. 2. Development, formulation and physico-‐chemical characterization of lipid drug nanotransporters. 3. Determination of the pharmacokinetic profile of edelfosine when vectorized into LN and its biodistribution in animal models. 4. Evaluation of the (therapeutic) efficacy of the selected nanomedicines in animal models of cancer (mantle cell lymphoma and glioma).
28
Hipótesis y Objetivos
HIPOTESIS Y OBJETIVOS Los análogos sintéticos éter fosfolípidos, o también conocidos como alquil-‐ lisofosfolípidos entre los que el 1-‐O-‐octadecil-‐2-‐O-‐metil-‐rac-‐glicero-‐3-‐fosfocolina (ET-‐18-‐OCH3, edelfosina) es su compuesto prototípico, constituyen una prometedora clase de substancias antitumorales orales que no tienen al ADN como diana, sino que actúan al nivel de la membrana. Sin embargo, y debido a su naturaleza química, estas moléculas poseen una toxicidad hemolítica dosis dependiente, además de toxicidad en el tracto gastrointestinal cuando se administran a dosis elevadas y baja biodisponibilidad oral. Por otro lado, las nanopartículas lipídicas (NL) han surgido como una propuesta prometedora para la administración oral de fármacos, ya que por sus características facilitan la solubilización de sustancias muy lipófilas además de favorecer la absorción de los fármacos a nivel intestinal. De hecho, estas partículas han sido utilizadas para aumentar o mejorar la biodisponibilidad oral de numerosos principios activos ya que debido a su pequeño tamaño y a su composición, las LN pueden penetrar en los espacios entre las villi de la mucosa intestinal y llegar al torrente circulatorio. El uso de estos nanotransportadores para el transporte de agentes antitumorales ofrece además una importante ventaja: su especificidad tumoral debido a un efecto realzado de la permeabilidad y la retención en la zona peritumoral y por tanto, una menor toxicidad. Teniendo en cuenta todo lo anterior, la hipótesis de partida de este trabajo es la siguiente: El uso de nanopartículas lipídicas de edelfosina podría suponer un incremento muy significativo de la eficacia antitumoral de este compuesto, tanto por la posibilidad de aumentar la absorción a nivel intestinal como por la de disminuir la dosis de principio activo, no requiriéndose por tanto una dosis diaria elevada para mantener niveles de compuesto farmacológicamente significativos en sangre, lo que a su vez disminuirá los posibles efectos tóxicos. 29
Hipótesis y Objetivos
En definitiva, se espera que la administración por vía oral de la edelfosina vehiculizada en estos nanotrasportadores se traduzca en un aumento de la biodisponibilidad del fármaco, así como en la disminución de la dosis administrada y posibles reacciones adversas asociadas a éste fármaco. El objetivo general de este trabajo es el diseño, desarrollo y evaluación in vitro e in vivo de nanotransportadores lipídicos conteniendo edelfosina. Los objetivos parciales son específicamente: 1. Determinación del perfil farmacocinético de la edelfosina y su biodistribución en modelos animales tanto con linfoma de manto como sin él. 2. Desarrollo,
formulación
y
caracterización
físico-‐química
de
nanotransportadores lipídicos del fármaco. 3. Determinación del perfil farmacocinético de la edelfosina vectorizada en nanopartículas lipídicas (NL) y su biodistribución en modelos animales. 4. Evaluación de la eficacia (terapéutica) de los nanomedicinas seleccionadas en modelos animales de cáncer (linfoma de manto y glioma).
30
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
Chapter 1 Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐ bearing immunosuppressed mice Ander Estella-‐Hermoso de Mendoza1, Miguel A. Campanero2, Janis de la Iglesia-‐ Vicente3, Consuelo Gajate3,4, Faustino Mollinedo3, María J. Blanco-‐Prieto1 1Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia,
University of Navarra, E-‐31008, Spain 2Servicio de Farmacología Clínica, Clínica Universitaria, E-‐31080 Pamplona, Spain 3Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain 4Unidad de Investigación, Hospital Universitario de Salamanca, Campus Miguel de
Unamuno, E-‐37007 Salamanca, Spain Running Title: Biodistribution and pharmacokinetics of edelfosine Key Words: edelfosine, biodistribution, mantle cell lymphoma, bioavailability, pharmacokinetics Corresponding author: Dr. María J. Blanco-‐Prieto, Department of Pharmacy and Pharmaceutical Technology, School of Pharmacy, University of Navarra, C/Irunlarrea 1, E-‐31080 Pamplona, Spain, Office phone: + 34 948 425 600 ext. 6519, Fax: + 34 948 425 649, e-‐mail:
[email protected]
Clinical Cancer Research, 2009. 15(3): p. 858-‐64
31
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
Abstract Purpose: The present study investigates and compares the dose-‐dependent pharmacokinetics and oral bioavailability of edelfosine in healthy, immune deficient and tumour-‐bearing immune suppressed mouse animal models, as well as edelfosine uptake and apoptotic activity in the Z-‐138 mantle cell lymphoma (MCL) cell line. Experimental design: Biodistribution study of edelfosine was performed in both BALB/c and severe combined immune deficiency (SCID) mice, and then the in vivo behaviour of the drug after intravenous and oral administration was monitored. Results: We found that edelfosine is incorporated and induces apoptosis in the Z-‐138 human mantle cell lymphoma cell line, whereas normal resting peripheral blood human lymphocytes were not affected. In vivo biodistribution studies revealed that accumulation of edelfosine in the tumour of a MCL-‐bearing mouse animal model was considerably higher (p < 0.01) than in the other organs analysed. Besides, no statistical differences were observed between the pharmacokinetic parameters of BALB/c and SCID mice. Edelfosine presented slow elimination and high distribution to tissues. Bioavailability for a single oral dose of edelfosine was less than 10 %, but a multiple dose oral administration increased this value up to 64 %. Conclusion: Our results show that edelfosine is widely scattered across different organs, but that it is preferentially internalised by the tumour both in vitro and in vivo.
33
Introduction. Lipid nanomedicines for anticancer drug therapy
34
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
1. Introduction Edelfosine is the prototype molecule of a promising family of anti-‐tumour compounds collectively known as synthetic alkyl-‐lysophospholipids, which include also clinically relevant drugs such as miltefosine and perifosine, and which can be administered orally [1-‐4]. This class of synthetic anti-‐cancer agents acts at the level of the cell membrane, unlike most currently available chemotherapeutic drugs that target the nuclear DNA, and induces selective apoptosis in malignant cells, sparing normal cells [5-‐7]. Although the precise mechanism of action is not yet fully elucidated, edelfosine-‐induced apoptosis is mediated by Fas/CD95 death receptor [8-‐10], JNK activation [11, 12], endoplasmic reticulum stress [13] and mitochondria [14]. Edelfosine has been applied as a purging agent in autologous bone marrow transplantation [15], and recent data suggest a putative clinical relevance for this agent in cancer [7]. However, there is very little information regarding the disposition and oral bioavailability of edelfosine in vivo. The kinetic behaviour of radioactive edelfosine in rats and mice after intravenous administration has been previously investigated. Arnold et al. [16] administered 15 daily injections of an amount of radioactive edelfosine of 1 · 106 cpm (counts per minute) intravenously to methylcholantrene-‐induced fibrosarcoma-‐bearing mice. Results showed that 1 day after the last injection an overall activity of only 2 · 106 cpm was recovered. This indicated that edelfosine did not accumulate in the organism. However, measurement of total radioactivity does not reflect the pharmacokinetic behaviour or the tissue distribution of the parent drug edelfosine, since the radiolabel may be included in some metabolites. Kötting et al. studied biodistribution in rats after oral administration of edelfosine [17]. Quantification was performed with high performance thin-‐layer chromatography (HPTLC). After 24 h, no significant levels of edelfosine were detected either in faeces or urine. This information led them to conclude that approximately 96 % of the drug was absorbed in the first 24 h [17]. As yet no precise pharmacokinetic parameters have been determined for free edelfosine. Characterisation of the pharmacokinetics of edelfosine is crucial to
35
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
understand the in vivo concentration-‐effect and concentration-‐toxicity relationships and to choose dosing regimens. Furthermore, knowledge of the in vivo concentration range, oral bioavailability and pharmacokinetics of free edelfosine is crucial to evaluate the effects of edelfosine delivery systems on the pharmacokinetic behaviour of this anti-‐neoplastic agent. In order to determine levels of edelfosine either in plasma, tissue or tumour, sensitive and selective techniques must be used to detect only the compound of interest and not others, like metabolites or very similar molecules. We have recently developed a simple and highly selective and sensitive HPLC-‐MS technique with a quantitation limit of 0.3 ng, which avoids the use of radiolabeled compounds [18]. With this technique it is possible to quantify edelfosine concentrations either in plasma, tissues or tumour accurately and sensitively in order to determine pharmacokinetic parameters. The present study investigates and compares the dose-‐dependent pharmacokinetics and oral bioavailability of edelfosine in healthy, immune deficient and tumour-‐bearing immune suppressed mouse animal models. Severe combined immune deficiency (SCID) mice are routinely used as hosts for malignant cells and for in vivo testing of new anti-‐tumour agents. SCID mice are characterised by the complete inability of the adaptive immune system to mount an appropriate immune response, due to absence of functional lymphocytes as a result of defects in T and B cell development. On these grounds, use of SCID mice permits the long-‐term engraftment of human tumour cells. Because most of the in vivo anti-‐tumour drug testing assays are carried out with SCID mice, accomplishment of the pharmacokinetics in these immune compromised animals is imperative, and so we studied the tissue and tumour drug distribution in these immune deficient mice. In addition, in vitro experiments with the Z-‐138 mantle cell lymphoma (MCL) cell line, used for xenograft assays, were performed in order to determine both edelfosine uptake and apoptotic activity.
36
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
2. Materials and Methods 2.1.
Chemicals.
Edelfosine was from INKEYSA (Barcelona, Spain). Platelet Activating Factor (PAF) and PBS [10 mM phosphate, 0.9 % NaCl] were obtained from Sigma-‐Aldrich (Madrid, Spain). Formic acid 99 % was purchased from Fluka (Madrid, Spain) and methanol was obtained from Merck (Barcelona, Spain). All solvents employed for the analysis were of analytical grade. RPMI-‐1640 cell culture medium, heat-‐ inactivated fetal calf serum (FCS) and antibiotics were from Gibco Invitrogen (Carlsbad, CA).
2.1.
Animal experiments.
The protocol for animal experiments was approved by the University of Navarra Animal Experimentation Ethics Committee (No. of protocol: 060 – 06). BALB/c mice (20 g) were obtained from Harlan Interfauna Ibérica S.L. (Barcelona, Spain). Animals received a standard diet and water ad libitum, except for the animals who received the oral doses, which were fasted for 24 hours prior to administration. Female CB17-‐SCID mice were purchased from Charles River Laboratories (Lyon, France). In xenograft experiments, eight-‐week SCID mice were subcutaneously inoculated into the lower dorsum with 1 x 107 Z-‐138 cells in 100 μL of PBS and 100 μL of Matrigel basement membrane matrix (Becton Dickinson, San Jose, CA).
2.2.
Cell culture and isolation of human peripheral blood lymphocytes.
The Z-‐138 MCL cell line [19] was grown in RPMI-‐1640 culture medium supplemented with 10 % FCS, 2 mM L-‐glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37 °C in humidified 95 % air and 5 % CO2. Lymphocytes were isolated from fresh human peripheral blood by dextran sedimentation, centrifugation on Ficoll-‐Paque density gradients, and monocyte depletion by culture dish adherence as described previously [20].
37
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
2.3.
Edelfosine uptake in cell culture.
Drug uptake was measured as described previously [5] after incubating cells (106) with 10 nmol [3H]edelfosine for 1 h in RPMI-‐1640/10 % FCS, and subsequent exhaustive washing (six times) with PBS + 2 % BSA. [3H]edelfosine (specific activity, 42 Ci/mmol) was synthesised by tritiation of the 9-‐octadecenyl derivative (Amersham Buchler, Braunschweig, Germany).
2.4.
Apoptosis assay.
Quantitation of apoptotic cells, following treatment with edelfosine, was calculated by flow cytometry as the percentage of cells in the sub-‐G1 region (hypodiploidy) in cell cycle analysis, as previously described [14].
2.5.
Pharmacokinetic studies after oral administration.
Three treatment lines were followed. A BALB/c mice group and a SCID non-‐ tumour-‐bearing mice group were treated with a daily oral administration of 30 mg/kg of edelfosine for 6 days. At various time points after the first oral administration (0, 1, 2, 5, 8 and 24 h), blood was collected in EDTA surface-‐coated tubes and then centrifuged at 2,000 × g for 10 min (4 °C) to collect plasma (100 μL). After the administration of the sixth dose, blood samples were collected at the same time intervals (0, 1, 2, 5, 8 and 24 h). Then, animals were sacrificed and spleen, liver, lungs, kidneys, heart, brain, stomach and intestine were collected and weighed. Tissues were homogenised in 1 mL of PBS pH=7.4 using a Mini-‐bead Beater (BioSpect Products, Inc., Bartelsville, Oklahoma, USA) and centrifuged at 10,000 × g for 10 min. Both plasma and tissue supernatants were collected and stored at −80 °C until high-‐performance liquid chromatography-‐mass spectrometry (HPLC-‐MS) analysis was performed. The last group, MCL-‐ bearing SCID mice, received daily oral administration of 30 mg/kg of edelfosine for 6 days. After the administration of the sixth dose, blood samples were collected at the same time intervals (0, 1, 2, 5, 8 and 24 h), and then the animals were sacrificed and liver and kidneys were collected, weighed and processed as previously described. Tumours were collected as the other organs.
38
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
2.6.
Pharmacokinetic study after intravenous administration.
An intravenous single dose of 200 μg (10 mg/kg) was administered to BALB/c mice via the tail vein. At various time points after administration (0, 1, 2, 5, 8 and 24 h), blood was collected in EDTA surface-‐coated tubes and then centrifuged at 2,000 × g for 10 min (4 °C) to separate the plasma (100 μL). After 24 h animals were sacrificed by cervical dislocation and tissueswere collected, weighed and processed as explained before.
2.7.
Linearity study of the dose.
BALB/c mice were administered single intravenous doses of 100, 200, 250 and 600 μg (5, 10, 12.5 and 30 mg/kg) of edelfosine dissolved in PBS via the tail vein. At various time points after administration (0, 0.5, 1 and 2 h), blood was collected as described above.
2.8.
Plasma/tissue extraction procedure for analytical process.
10 μg of Platelet Activating Factor (0.2 mg/mL) used as internal standard were added to 100 μl of plasma or tissue supernatant. 190 μl of a mixture of 1 % formic acid/methanol (5:95, v/v) were added in order to precipitate the proteins. Samples were vortexed for 1 min and after centrifugation (20,000 × g, 10 min), 25 μl of the supernatant were analysed by HPLC-‐MS [18].
2.9.
HPLC-‐MS analysis for edelfosine.
The method used for edelfosine quantitation was a slight modification of an earlier HPLC-‐MS method [18]. Quantitation was achieved by comparing the observed peak area ratios of edelfosine and internal standard of the samples to a regression curve determined from drug-‐fortified plasma and tissue standards. The calibration curve of edelfosine in plasma showed good linearity over the concentration range of 0.1 – 75 μg/mL. Linear range of edelfosine concentrations in lung, kidney, liver, spleen, brain, heart, stomach and intestine homogenate was 0.2 – 31.75 μg/mL. 39
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
2.10. Data analysis. The plasma concentration data were analysed by non-‐compartmental and compartmental analysis using WinNonlin Professional Edition Version 2.1 (Pharsight, Mountain View, CA, USA). Pharmacokinetic analysis was performed for two different doses of 10 and 30 mg/kg with plasma samples obtained from experiments with all mice. The area under the plasma concentration vs. time curve (AUC) was determined using the log-‐linear trapezoidal rule with extrapolation to 24 hours. Clearance (CL) is the volume of plasma completely cleared of a specific compound per unit time by the organism; it was calculated by dividing the dose by AUC. The maximum plasma concentration (Cmax) was determined directly from the plasma concentration-‐time curve. Oral bioavailability (F) was determined by ratio of the dose-‐normalized AUCs following oral and i.v. administration. Volume of distribution at steady-‐state (Vss) is the volume of fluid that would be required to contain the amount of drug in the body if it were uniformly distributed at a concentration equal to that in the plasma. The half-‐life value (t1/2) refers to the time taken for plasma concentration to fall by 50 %. Disposition (t1/2α) and elimination (t1/2β) half-‐lives were determined using the following formulas: t1/2α=ln (2)/α and t1/2β=ln (2)/β, where α and β represent the disposition and elimination constant rates, respectively, calculated from the intercompartmental mass transfer rates (k12, k21) and elimination rates (k10).
2.11. Statistical analysis. Mean values of the tissue/plasma concentration ratio of more than 2 groups were analysed by an analysis of variance (ANOVA) followed by Dunnett's test for a double comparison using Social Package of Statistical Sciences (SPSS). The presence of differences in tissue/plasma ratios was measured by the Mann Whitney U test for double comparisons using the same program. A correlation analysis was performed for the study of the linearity of the dose. A value of p < 0.05 was considered to be statistically significant for all statistical tests.
40
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
3. Results 3.1.
Edelfosine uptake and induction of apoptosis in MCL cells.
Edelfosine has been reported to be selectively incorporated into malignant cells leading to their demise, while normal cells were spared [5, 8]. We found that the Z-‐ 138 MCL cell line took up significant amounts of edelfosine and underwent apoptosis, whereas normal resting peripheral blood lymphocytes were not affected and drug incorporation was scarce (Fig. 1).
Figure 1. Selective uptake and induction of apoptosis in mantle lymphoma cells. (A) Edelfosine uptake was determined after incubating human mantle lymphoma Z-‐138 cells and resting human peripheral blood lymphocytes (PBL) with 10 nmol [3H]edelfosine for 1 h. (B) Z-‐138 cells and resting human PBLs were incubated for 24 h with 10 µM edelfosine, and apoptosis was then quantitated as the percentage of cells in the sub-‐G1 region in cell cycle analysis by flow cytometry.
3.2.
Tissue distribution of edelfosine in tumour-‐bearing SCID mice.
Because Z-‐138 MCL cells incorporated edelfosine, we next analysed the in vivo tissue distribution of the drug in tumour-‐bearing mice. Immune deficient SCID mice were injected subcutaneously with Z-‐138 MCL cells, and xenografts were allowed to establish to an average size of 300 mm3 before drug oral treatment. The tissue distribution expressed as tissue/plasma ratio of edelfosine concentrations after multiple oral dose administration of an edelfosine dose of 30 mg/kg (six-‐day treatment) to MCL-‐bearing SCID mice is shown in Fig. 2A. This oral 30 mg/kg dose was perfectly well tolerated by the mice, and we observed no side-‐effects or body weight loss (data not shown). Mean concentration of edelfosine in plasma 24 h after the sixth dose was 10.69 μg/mL. 41
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
Kidney and liver were selected for drug accumulation, as they are the major drug clearance tissues. These organs showed a two-‐fold higher drug concentration than the corresponding one in plasma (Fig. 2A). Interestingly, the tumour showed a high accumulation of edelfosine 24 h after the dose on day 6 (Fig. 2A). The mean concentration of edelfosine in the tumour at this stage was 138.64 μg/g, which is about thirteen times higher than the plasma drug concentration, and five to six times higher than the drug concentration found in kidney and liver. Moreover, the tumour/plasma concentration ratio of edelfosine was significantly higher (p < 0.01) than the corresponding ratios observed in both kidney and liver. Thus, an in vivo and in vitro comparison indicates that edelfosine is selectively taken up by tumour cells.
Figure 2. A) Tissue/plasma concentration ratios of edelfosine after ( ) multiple dose oral administration of 30 mg/kg of edelfosine for six days to non tumor-‐bearing SCID mice and ( ) multiple dose oral administration of 30 mg/kg of edelfosine for six days to mantle cell lymphoma bearing SCID mice (n=6, Mean ± S.D.); and B) tissue/plasma concentration ratios of edelfosine after ( ) single intravenous dose of 10 mg/kg of edelfosine to healthy BALB/c mice; ( ) multiple dose oral administration of 30 mg/kg of edelfosine for six days to healthy BALB/c mice; ( ) multiple dose oral administration of 30 mg/kg of edelfosine for six days to non tumor-‐bearing SCID mice (n=6, Mean ± S.D.)
42
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
3.3.
Tissue distribution of edelfosine in non-‐tumour-‐bearing SCID mice.
The tissue distribution of edelfosine was also analysed in control non-‐tumour-‐ carrying SCID mice following the same protocol as with tumour-‐bearing SCID mice. The mean plasma concentration of edelfosine at the time of 24 h after the sixth dose was 12.12 μg/mL. Fig. 2B shows a higher distribution of edelfosine to lung, spleen, intestine, liver and kidney. In contrast, low drug concentrations were found in heart, brain and stomach (Fig. 2B). Compared to tumour-‐bearing SCID mice, no statistical differences in liver/plasma concentration ratios (p > 0.05) were found in non-‐tumour-‐bearing SCID mice, whereas the kidney was found to have a statistically highly significant increase (p < 0.01) (Fig. 2A). 3.4.
Tissue distribution in healthy BALB/c mice after oral administration.
A multiple oral dose administration of 30 mg/kg of edelfosine (six-‐day treatment) was performed to healthy BALB/c mice. 24 h after the last dose (day six), the mean plasma concentration was 13.22 μg/mL. As is shown in Fig. 2B, the highest tissue/plasma concentration ratios were found for kidney and intestine. Lower ratios were observed for stomach, spleen, lung and liver. A low amount of edelfosine was detected in the heart and brain. Fig. 2B also shows that no statistically significant differences were appreciated among the tissue/plasma concentration ratios of lung, heart, brain, intestine and liver of BALB/c and non-‐tumour-‐bearing SCID mice. The stomach of BALB/c mice after the multiple oral administration of 30 mg/kg presented a higher ratio than SCID mice after the administration of same dose (p < 0.05) and the spleen/plasma concentration ratio showed a significant decrease (p < 0.05) with regard to SCID non-‐tumour-‐bearing mice. 3.5.
Tissue distribution in healthy BALB/c mice after intravenous
administration. A single intravenous administration of 200 μg of edelfosine (10 mg/kg) was administered to healthy BALB/c mice. After 24 h, edelfosine concentration in 43
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
plasma was around 2.5 μg/mL. The tissue distribution of edelfosine expressed as tissue/plasma ratio can be observed in Fig. 2B. No drug was detectable in heart and very little was found in brain. Kidney and intestine presented the highest tissue/plasma concentration ratios, followed by lung, liver, stomach and spleen. 3.6.
Pharmacokinetic characterization after intravenous administration.
Fig. 3 depicts the concentration of edelfosine in mouse plasma plotted against time after a single dose intravenous administration of 200 μg of edelfosine (10 mg/kg) to BALB/c mice.
Figure 3. Time-‐concentration curve data of edelfosine 24 h after single dose intravenous (♦) administration of 10 mg/kg and oral single dose (▲) administration of 30 mg/kg to BALB/c mice (n=6, Mean ± S.D.)
Pharmacokinetic analysis of edelfosine in blood plasma showed a Cmax of 50.7±28.1 μg/mL and a Cmin of 2.5±1.3 μg/mL, 24 hours after intravenous administration. The obtained pharmacokinetic parameters are listed in Table I.
44
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
Table I. Comparison of pharmacokinetic parameters of edelfosine after administration of different doses to BALB/c and SCID mice by different routes (n=6, Mean ± S.D.). No statistical differences were found among the parameters (p > 0.05)
Plasma concentration-‐time data of edelfosine was well described by a bi-‐ exponential function (model selection criterion, -‐24.12) following intravenous administration. The half-‐lives of distribution (t1/2α) and elimination (t1/2β) phases were 0.286±0.076 and 22.288±14.016 h, respectively. The systemic clearance (CL) and steady-‐state volume of distribution (Vss) were 0.056±0.030 L/h/kg and 1.285±0.556 L/kg, respectively. There was little variability in most of the values of the parameters, indicating a well-‐controlled and reproducible study, except for the elimination phase half-‐life value. 3.7.
Linearity of dose.
Both AUC-‐dose and Cmax-‐dose profiles of edelfosine showed a linear correlation (r2=0.999), as can be seen in Fig. 4, suggesting linear pharmacokinetics with similar elimination and distribution half-‐lives after intravenous administration of doses between 5 and 30 mg/kg.
45
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
Figure 4. Correlation between doses of edelfosine of 100, 200, 250 and 600 µg (5, 10, 12.5 and 30 mg/kg) and the calculated AUC (A) and Cmax (B) (n=6, Mean ± S.D.)
3.8.
Pharmacokinetic characterization after oral administration.
A very low concentration of edelfosine could be found in plasma along the time interval of 0, 1, 2, 5, 8 and 24 h after oral administration of 30 mg/kg of edelfosine in BALB/c mice, as can be observed in Fig. 3. These concentrations were close to the detectable limits of the technique and not sufficient for the calculation of pharmacokinetic parameters. Oral bioavailability for edelfosine was calculated from the ratios of the average values for AUC0→24h for the oral and intravenous doses. Oral bioavailability (F) for edelfosine was found to be less than 10 %. Pharmacokinetic parameters obtained after compartmental analysis of experimental data are shown in Table I. Oral bioavailability increased to 64 %, when the steady-‐state was reached after the sixth day. At this point, a Cmax of 14.46±2.97 μg/mL at a tmax of 3 h and a Cmin of 7.65±2.45 μg/mL were measured 24 hours after the last administration. Due to the fast elimination during the disposition phase, only the β elimination phase could be characterized. AUC presented a value of 258.54±58.16 μg·h/L. The half-‐life of elimination was 46
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
30.404±26.761 h, with a volume of distribution at steady-‐state of 1.563±0.163 L/kg and a systemic clearance of 0.057±0.039 L/h/kg, similar values to those obtained after intravenous administration of edelfosine, with no statistically significant differences. In order to determine possible inter-‐strain pharmacokinetic differences, these results were compared to the results obtained after edelfosine was administered to non tumour-‐bearing SCID mice. After reaching the steady-‐state, pharmacokinetic parameters were estimated (Table I). Whatever the parameters studied (half life of elimination, clearance value, steady-‐state volume of distribution and AUC) no statistically significant differences were observed between SCID and BALB/c mice.
4. Discussion Several attempts have been made over the years accurately to detect the accumulation of edelfosine in tumours, tissues and plasma as well as to estimate its pharmacokinetic parameters. This pharmacokinetic analysis is mandatory in order to accomplish all the preclinical assays prior to a rational clinical use of the drug. Because edelfosine shows a variety of medical applications, a pharmacokinetic study in different murine species, each one appropriate for distinct in vivo assays, is required. We carried out a complete pharmacokinetic analysis in SCID mice, widely used for anti-‐tumour activity testing, and BALB/c, useful for additional putative edelfosine applications [21]. In the present study, we set up a MCL-‐ bearing animal model in immune deficient SCID mice to analyse the distribution of edelfosine in tissues and tumour. We found that the Z-‐138 MCL cell line took up significant amounts of edelfosine, subsequently undergoing apoptosis, whereas normal cells hardly incorporated the drug and were spared. The in vitro and in vivo data reported here indicate a remarkably selective accumulation of edelfosine in tumour cells. In the MCL-‐bearing animal model, kidney and liver were the only organs extracted as they are considered to be the main drug distribution and clearance tissues. Comparing the tissue distribution of edelfosine between these tumour-‐ 47
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
bearing SCID mice and the non-‐tumour-‐bearing SCID mice (Fig. 2A), no statistically significant differences were seen in liver (p > 0.05), while a major decrease (p < 0.01) was found in the kidneys of the tumour-‐bearing mice. This lower renal uptake in the tumor model might be a result of an increased drug uptake by the tumor with less drug available for renal excretion, in addition to the renal failure usually expected at advanced stages of the disease [22]. These results clearly indicate that MCL does not seem to affect the tissue distribution of the drug. Fig. 2B displays that the distribution of edelfosine in mice is mainly predominant in spleen, intestine and kidney. The high presence of edelfosine in spleen can be explained by the incorporation of the drug into the membrane of erythrocytes inducing a haemolytic effect [17]. These red blood cells end up being cleared from the blood in the spleen. We found a higher concentration of edelfosine in the spleen of SCID mice as compared to BALB/c mice, and this could be due to putative differences in the spleen of immune deficient mice. No statistically significant differences were observed between tissue/plasma concentration ratios of edelfosine in lung, heart, brain, intestine liver and kidney of immune suppressed mice (SCID mice) and BALB/c mice (Fig. 2B). Moreover, no significant differences were observed in mean plasma concentrations of edelfosine 24 h after the sixth dose of the multiple oral administration of 30 mg/kg to tumour-‐bearing SCID mice, non-‐tumour-‐bearing SCID mice and healthy BALB/c mice (10.69 μg/mL, 12.12 μg/mL and 13.22 μg/mL, respectively). As a result, a similar pharmacokinetic profile would be expected in both strains. Therefore, we performed the characterisation of edelfosine pharmacokinetic profile in healthy BALB/c mice. Knowledge of the pharmacokinetic parameters of a drug is a must in order to ensure that exposure is sufficient to evaluate its pharmacodynamic properties, to predict the pharmacokinetics in other species and to develop an appropriate dosage form. Following intravenous administration of a dose of 10 mg/kg of edelfosine to BALB/c mice, mean blood concentrations declined biphasically, with an initial half life of around 0.3 h (Table I). After 2 h, plasma concentrations appeared to decline 48
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
more slowly. Given that blood concentrations had declined about six-‐fold during the initial phase, the contribution of the secondary phase to the total drug exposure is of little relevance. According to these half-‐life values, edelfosine presented a rapid distribution to central compartment tissues whereas a slower distribution to deep compartments was appreciated. The rapid distribution is mostly observed to organs that are more highly irrigated, like kidney, liver, intestine and lung. Steady-‐state volume of distribution was 1.26 L/kg, much greater than that of the vascular volume in mice and approximately twice its total body water [23], when normalized to body weight, suggesting that edelfosine is highly distributed extra-‐ vascularly. In fact, k12 presented a value five to six times higher than the corresponding k21 value. Edelfosine showed very low clearance values in mice compared to its respective liver and kidney blood flow. It is interesting to note that this value of 0.056 L/h/kg is translated to barely 1 % of the liver blood flow [23], suggesting a much lower intrinsic hepatic clearance value. We verified that as the drug dose increases the area under the curve (AUC) increases equivalently, meaning that edelfosine exhibits linear pharmacokinetics. In this study we demonstrate that edelfosine shows a linear correlation not only between the administered dose and the resulting AUC, but also between the administered dose and the corresponding Cmax. This fact suggests that there is no saturation of the elimination process of edelfosine at the concentration range between 5 and 30 mg/kg. Even though the mechanism for its clearance is still unknown, renal elimination of the drug might well happen, as edelfosine is present in kidney. Previous studies have established that amphiphilic cationic drugs are efficiently removed by the liver from the blood [24]. Additionally, amphiphilic cationic drugs were detected in other secretion organs of cationic amphiphilic drugs like intestinal mucosa [25] and kidney. These data are also in good agreement with our results and the ones obtained by Marschner et al. [26].
49
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
The first determinations of oral absorption of 15 μmol of edelfosine (38 mg/kg) in 1992 suggested that about 96 % of edelfosine was absorbed by rats within 24 hours [17]. Conversely, our study shows very low plasma concentrations of edelfosine 24 hours after the oral administration of a single dose of 30 mg/kg, revealing very low absorption of edelfosine by the gastrointestinal tract in mice. However, oral bioavailability showed a significant increase (64 %) after daily administration of 30 mg/kg of edelfosine for 6 days. As can be seen from our studies, edelfosine showed a moderate volume of distribution and a rapid and varied distribution through the organism when it is administered by an intravenous route. This information, along with the amphiphilic properties, may lead us to consider the possibility of passive diffusion through the intestinal epithelium as mechanism of absorption for edelfosine. However, this suggestion does not correlate with the low absorption rate of the drug found using the oral route. Moreover, it has been shown that the absorption of certain cationic drugs, such as quaternary ammonium compounds, shows a reduced absorption rate due to the presence of efflux transport systems on the apical membrane (e.g. P-‐gp). Indeed, the alkylphosphocholine miltefosine has been reported to interact with P-‐gp [27], and it has also been reported that edelfosine is a substrate for P-‐glycoprotein [28]. This issue would only explain the lack of absorption at a single oral dose, and the high oral bioavailability achieved after a multiple oral dose administration. In summary, edelfosine showed a remarkable apoptotic effect in Z-‐138 MCL cells in vitro and a high and rather selective uptake in MCL tumour cells in vitro and in vivo. This fact, along with knowledge of the biodistribution and pharmacokinetic behaviour of edelfosine, permits the establishment of treatment regimens as well as the development of drug delivery systems in order to treat MCL and other tumours. In addition, since no statistically significant differences were noticed among the pharmacokinetic parameters of all three treatment lines, it can be concluded that there are no differences between BALB/c and SCID pharmacokinetic profiles. Thus, further studies with edelfosine delivery systems will be able to be performed either in one animal model or the other.
50
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
These data indicate that multiple oral administration of edelfosine is required to reach a clinically important plasma concentration. Our results also show that edelfosine is widely scattered across different organs, but that it is preferentially internalised by the tumour both in vitro and in vivo.
Financial support Caja Navarra Foundation, the Spanish Ministry of Science and Innovation (SAF2007-‐61261, SAF2005-‐04293, PCT-‐090100-‐2007-‐27) the “Fondo de Investigación Sanitaria” and European Commission (FIS-‐FEDER 06/0813) and RD06/0020/1037 from Red Temática de Investigación Cooperativa en Cáncer (RTICC), Instituto de Salud Carlos III (ISCIII), Ministerio de Sanidad. Consuelo Gajate is supported by the Ramón y Cajal Program from the Spanish Ministry of Science and Innovation. Ander Estella-‐Hermoso de Mendoza is supported by the research grant from the Department of Education of the Basque Government (BFI06.37).
51
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
References 1.
Gajate, C. and F. Mollinedo, Biological Activities, Mechanisms of Action and Biomedical Prospect of the Antitumor Ether Phospholipid ET-‐18-‐OCH3 (Edelfosine), A Proapoptotic Agent in Tumor Cells. Curr Drug Metab, 2002. 3(5): p. 491-‐525.
2.
Mollinedo, F., et al., ET-‐18-‐OCH3 (edelfosine): a selective antitumour lipid targeting apoptosis through intracellular activation of Fas/CD95 death receptor. Curr Med Chem, 2004. 11(24): p. 3163-‐84.
3.
Houlihan, W.J., et al., Phospholipid antitumor agents. Med. Res. Rev., 1995. 15(3): p. 157-‐223.
4.
Munder, P.G. and O. Westphal, Antitumoral and other biomedical activities of synthetic ether lysophospholipids. Chem. Immunol., 1990. 49: p. 206-‐35.
5.
Mollinedo, F., et al., Selective induction of apoptosis in cancer cells by the ether lipid ET-‐ 18-‐OCH3 (Edelfosine): molecular structure requirements, cellular uptake, and protection by Bcl-‐2 and Bcl-‐X(L). Cancer Res, 1997. 57(7): p. 1320-‐8.
6.
Ruiter, G.A., et al., Alkyl-‐lysophospholipids as anticancer agents and enhancers of radiation-‐ induced apoptosis. Int J Radiat Oncol Biol Phys, 2001. 49(2): p. 415-‐9.
7.
Gajate, C. and F. Mollinedo, Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and downstream signaling molecules into lipid rafts. Blood, 2007. 109(2): p. 711-‐ 9.
8.
Gajate, C., et al., Intracellular triggering of Fas, independently of FasL, as a new mechanism of antitumor ether lipid-‐induced apoptosis. Int J Cancer, 2000. 85(5): p. 674-‐82.
52
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
9.
Gajate, C. and F. Mollinedo, The antitumor ether lipid ET-‐18-‐OCH3 induces apoptosis through translocation and capping of Fas/CD95 into membrane rafts in human leukemic cells. Blood, 2001. 98(13): p. 3860-‐3.
10.
Gajate, C., et al., Intracellular Triggering of Fas Aggregation and Recruitment of Apoptotic Molecules into Fas-‐enriched Rafts in Selective Tumor Cell Apoptosis. J Exp Med, 2004. 200(3): p. 353-‐65.
11.
Gajate, C., et al., Involvement of c-‐Jun NH2-‐terminal kinase activation and c-‐ Jun in the induction of apoptosis by the ether phospholipid 1-‐O-‐octadecyl-‐2-‐O-‐ methyl-‐rac-‐glycero-‐3-‐phosphocholine. Mol Pharmacol, 1998. 53(4): p. 602-‐ 12.
12.
Ruiter, G.A., et al., Alkyl-‐lysophospholipids activate the SAPK/JNK pathway and enhance radiation-‐induced apoptosis. Cancer Res, 1999. 59(10): p. 2457-‐63.
13.
Nieto-‐Miguel, T., et al., Endoplasmic Reticulum Stress in the Proapoptotic Action of Edelfosine in Solid Tumor Cells. Cancer Res, 2007. 67(21): p. 10368-‐10378.
14.
Gajate, C., et al., Involvement of mitochondria and caspase-‐3 in ET-‐18-‐OCH3-‐ induced apoptosis of human leukemic cells. Int J Cancer, 2000. 86(2): p. 208-‐ 18.
15.
Vogler, W.R., et al., A phase II trial of autologous bone marrow transplantation (ABMT) in acute leukemia with edelfosine purged bone marrow. Adv Exp Med Biol, 1996. 416: p. 389-‐96.
16.
Arnold, B., R. Reuther, and H.U. Weltzien, Distribution and metabolism of synthetic alkyl analogs of lysophosphatidylcholine in mice. Biochim. Biophys. Acta., 1978. 530(1): p. 47-‐55.
17.
Kotting, J., et al., Hexadecylphosphocholine and octadecyl-‐methyl-‐glycero-‐3-‐ phosphocholine: a comparison of hemolytic activity, serum binding and tissue distribution. Prog. Exp. Tumor Res., 1992. 34: p. 131-‐42.
18.
Blanco-‐Prieto, M.J., M.A. Campanero, and F. Mollinedo, Quantitative determination of the antitumor alkyl ether phospholipid edelfosine by 53
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
reversed-‐phase liquid chromatography-‐electrospray mass spectrometry: application to cell uptake studies and characterization of drug delivery systems. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci., 2004. 810(1): p. 85-‐92. 19.
Medeiros, L.J., Z. Estrov, and G.Z. Rassidakis, Z-‐138 cell line was derived from a patient with blastoid variant mantle cell lymphoma. Leuk Res, 2006. 30(4): p. 497-‐501.
20.
Cabaner, C., et al., Induction of apoptosis in human mitogen-‐activated peripheral blood T-‐ lymphocytes by the ether phospholipid ET-‐18-‐OCH3: involvement of the Fas receptor/ligand system. Br J Pharmacol, 1999. 127(4): p. 813-‐25.
21.
Mollinedo, F., Antitumor ether lipids: proapoptotic agents with multiple therapeutic indications. Expert Opin Ther Patents, 2007. 17(4): p. 385-‐405.
22.
Davies, J., et al., Acute renal failure due to mantle cell lymphoma-‐-‐a case report and discussion of the literature. Clin. Nephrol., 2007. 67(6): p. 394-‐6.
23.
Davies, B. and T. Morris, Physiological parameters in laboratory animals and humans. Pharm. Res., 1993. 10(7): p. 1093-‐5.
24.
Smit, J.W., et al., Hepatobiliary and intestinal clearance of amphiphilic cationic drugs in mice in which both mdr1a and mdr1b genes have been disrupted. Br. J. Pharmacol., 1998. 124(2): p. 416-‐24.
25.
Kim, M.K. and C.K. Shim, The transport of organic cations in the small intestine: current knowledge and emerging concepts. Arch. Pharm. Res., 2006. 29(7): p. 605-‐16.
26.
Marschner, N., et al., Distribution of hexadecylphosphocholine and octadecyl-‐ methyl-‐glycero-‐3-‐phosphocholine in rat tissues during steady-‐state treatment. Cancer Chemother. Pharmacol., 1992. 31(1): p. 18-‐22.
27.
Rybczynska, M., et al., MDR1 causes resistance to the antitumour drug miltefosine. Br. J. Cancer., 2001. 84(10): p. 1405-‐11.
54
Chapter 1. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-‐bearing immunosuppressed mice
28.
Ruetz, S., et al., Functional interactions between synthetic alkyl phospholipids and the ABC transporters P-‐glycoprotein, Ste-‐6, MRP, and Pgh 1. Biochemistry, 1997. 36(26): p. 8180-‐8.
55
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Chapter 2
Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Ander Estella-‐Hermoso de Mendoza1, Marta Rayo1, Faustino Mollinedo2 and María J. Blanco-‐Prieto1* 1Department of Pharmacy and Pharmaceutical Technology, School of Pharmacy,
University of Navarra, C/Irunlarrea 1, E-‐31080 Pamplona, Spain 2Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain
Key words: Lipid nanoparticles; Edelfosine; Drug delivery; Atomic force microscopy; Differential scanning calorimetry; X–ray diffractometry. Corresponding author: Dr. María J. Blanco-‐Prieto, Department of Pharmacy and Pharmaceutical Technology, School of Pharmacy, University of Navarra, C/Irunlarrea 1, E-‐31080 Pamplona, Spain, Office phone: + 34 948 425 600 ext. 6519, Fax: + 34 948 425 649, e-‐mail:
[email protected] European Journal of Pharmaceutics and Biopharmaceutics, 2008. 68(2): p. 207-‐13
57
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Abstract The
ether
lipid
1-‐O-‐octadecyl-‐2-‐O-‐methyl-‐rac-‐glycero-‐3-‐phosphocholine,
edelfosine (ET-‐18-‐OCH3) is the prototype molecule of a promising class of antitumour drugs named alkyl–lysophospholipid analogues (ALPs) or antitumor ether lipids. This drug presents a very important drawback as can be the dose depending haemolysis when administered intravenously. Lipid nanoparticles have been lately proposed for different drug encapsulation as an alternative to other controlled release delivery systems, such as liposomes or polymeric nanoparticles. The aim of this study was to develop a lipid nanoparticulate system that would decrease systemic toxicity as well as improve the therapeutic potential of the drug. Lipids employed were Compritol® 888 ATO and stearic acid. The nanoparticles were characterized by photon correlation spectroscopy for size and size distribution, and atomic force microscopy (AFM) was used for the determination of morphological properties. By both differential scanning calorimetry (DSC) and X-‐ray diffractometry, crystalline behaviour of lipids and drug was assessed. The drug encapsulation efficiency and the drug release kinetics under in vitro conditions were measured by HPLC–MS. It was concluded that Compritol® presents advantages as a matrix material for the manufacture of the nanoparticles and for the controlled release of edelfosine.
59
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
1. Introduction Edelfosine
(ET–18–OCH3,
1-‐O-‐octadecyl-‐2-‐O-‐methyl-‐rac-‐glycero-‐3-‐
phosphocholine) is the prototype of a promising class of antitumor agents, collectively known as alkyl-‐lysophospholipid analogues (ALP) or antitumor ether lipids, that do not target the DNA, but affect the cell membrane and the apoptotic machinery of the cancer cell [1]. Phase I studies have shown a good tolerability of the drug but with haematological and systemic side effects [1, 2]. However, although edelfosine has been shown to exert potent antineoplastic effects in vitro [3, 4], the antitumor activity in phase II clinical studies has been only moderate [5]. Moreover, ALPs show manifold biological effects in addition to their antineoplastic actions, including an antiparasitic effect on Leishmania [6] as well as an inhibition of the cell membrane phospholipid turnover [7] and a potent inhibition of neovascularization [8], protein kinase C and Na+/K+-‐ATPase [9]. Edelfosine has also been given intravenously, but this provoked haemolysis as a major side effect [10]. This led to the only formulation developed so far, the TLC ELL-‐12, in which edelfosine was included into liposomes [11] to avoid the haemolytic toxicity of the drug. However, the main inconvenience of liposomes is their rapid clearance from plasma in comparison with other delivery systems. Lipid nanoparticles have been proposed as an alternative for the existing traditional particulate systems, such as previously mentioned liposomes or polymeric nanoparticles. These particulate systems made from solid lipids started being developed in the early nineties. They provide physical stability, controlled release and a wide variety of application routes (parenteral, oral, dermal, ocular, pulmonary and rectal) [12-‐16]. Lipid nanoparticles are basically composed of a high melting point lipid that acts as a solid core, covered by surfactants. Lipids used to form these matrices are biodegradable raw materials that are physiologically tolerated [17]: triglycerides (i.e. tristearin), partial glycerides (i.e. Compritol), fatty acids (i.e. stearic acid), steroids (i.e. cholesterol) or waxes (i.e. cetyl palmitate) [18].
61
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
The formulation methods are also diverse. Emulsion formation and solvent evaporation method has been widely used for particle formation, although other methods like high pressure homogenization [18], solvent diffusion methods in aqueous solutions [19] or hot emulsion methods [18, 20] have also been employed. Drawbacks associated to this kind of formulations, like limited drug loading capacity, adjustment of drug release profile and potential drug expulsion during storage have been reported [21]. Besides, drug loading capacity is limited by the solubility of the drug in the lipid melt, the structure of the lipid matrix and the polymorphic state of the lipid matrix. Comparing with the previously mentioned liposomes, the main improvement of the lipid nanoparticles is their physical and chemical long-‐term stability up to 12 – 24 months [13], even though an increase in particle size has been reported in a lesser time [22]. As a feasible solution for this setback, the freeze-‐drying process has shown to increase physicochemical stability of lipid particles over large periods of time [23]. Taking into account this information, the aim of this study was to develop a new formulation that would provide a controlled release for the antitumor lipid edelfosine, in order to improve its therapeutic activity.
2. Materials and Methods 2.1.
Materials
Edelfosine was from INKEYSA (Barcelona, Spain). Compritol® 888 ATO was a gift of Gattefossè (Cedex, France). Stearic acid and Tween® 80 were purchased from Roig Farma (Barcelona, Spain). Phosphate Buffer Saline (PBS) was provided by Sigma-‐Aldrich (Barcelona, Spain). Chloroform and ethyl acetate were obtained from Panreac Química S.A. (Barcelona, Spain). All solvents employed for the chromatographic analysis were of analytical grade. Formic acid 99 % for mass spectroscopy was purchased from Fluka (Barcelona, Spain) and methanol was obtained from Merck (Barcelona, Spain). All other chemicals were of reagent grade and used without further purification. Amicon Ultra–15 centrifugal filter devices were purchased from Millipore (Cork, Ireland). 62
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
2.2.
Preparation of lipid nanoparticles
Lipid nanoparticles were prepared by the emulsification/solvent evaporation method. For the simple emulsion solvent evaporation method, edelfosine and the lipid (100 mg of either Compritol® or stearic acid) were dissolved in 2 mL of chloroform. This solution was emulsified with 10 mL of a 0.5 % or 1 % Tween 80 solution by ultrasonication using a MicrosonTM ultrasonic cell disruptor (NY, USA). The O/W emulsion formed was magnetically stirred for 45 minutes and subjected to low vacuum rotary evaporation for the complete elimination of the organic solvent. Particles were centrifuged at 4500 g for 40 minutes using an Amicon Ultra–15 filter device and washed twice with distilled water. The obtained particular suspension was fast frozen under –80 °C for at least 3 hours and freeze-‐ dried in order to store it at 4 °C. For the double emulsion (W/O/W), edelfosine was dissolved in 100 μl of a 1 % Tween 80 solution and emulsified with 2 ml of ethyl acetate. This first W/O emulsion was then emulsified with 10 ml of a 1 % Tween 80 solution. Following steps were the same to the ones of the simple emulsion solvent evaporation method.
2.3.
Encapsulation efficiency
Edelfosine was extracted by dissolving 10 mg of nanoparticles in 1 ml of chloroform and then mixed with 3 ml of ultra pure water. The mixture was vortexed for 1 minute and then centrifuged at 9500 g for 10 minutes. The supernatant was analysed by a HPLC–MS method, which is a slight modification of a previously developed method [24]. The apparatus used for the HPLC analysis was a Model 1100 series LC coupled with an atmospheric pressure (AP)-‐ electrospray ionization (ESI) mass spectrometer (HP 1100 with MSD VL, Waldbronn, Germany). Data acquisition and analysis were performed with a Hewlett-‐Packard computer using the ChemStation G2171 AA programme. Separation was carried out at 50 °C on a reversed-‐phase, 150 mm
3 mm column
packed with C18, 5 μm silica reversed-‐phase particles (Gemini®) obtained from Phenomenex® (Torrance, CA, USA). This column was preceded by a reversed-‐ phase, C18, 5 μm guard column (SecurityGuard®, 20 mm
4 mm, Phenomenex®,
63
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Torrance, CA, USA). The mobile phase was a mixture of methanol–1 % formic acid (95:5, v/v). Separation was achieved by isocratic solvent elution at a flow-‐rate of 0.5 ml/min. The mass spectrometer was operated in the positive ESI mode. The detection of edelfosine was performed by selected ionization monitoring (SIM) mode. The mass spectrometer was programmed to monitor the ion of edelfosine at m/z 524.40. Typical retention time was 3.65 minutes.
2.4.
Nanoparticle characterization
2.4.1. Particle size, size distribution and zeta potential Particle size and distribution of the nanoparticles were measured by photon correlation spectroscopy (PCS) using a Zetasizer Nano (Malvern Instruments, UK). Each sample was diluted 30 fold in distilled water until the appropriate concentration of particles was achieved to avoid multiscattering events. The obtained homogenous suspension was examined to determine the volume mean diameter, size distribution and polydispersity and repeated three times for each sample. The data are expressed as a mean value ± standard deviation. Similarly, the zeta potential was measured using the same equipment described previously with a combination of laser Doppler velocimetry. Samples were diluted with distilled water and each experiment was repeated three times.
2.4.2. Morphology Atomic force microscopy (Cervantes AFM System, Nanotec Electrónica, S.L., Spain) was employed to determine the shape and surface morphology of the nanoparticles. AFM was conducted with Nanoscope IIa IIIa in the tapping mode. The nanoparticle sample was mounted on a metal stab and scanned by the AFM maintained in a constant temperature and vibration time environment.
2.4.3. Thermal analysis of freeze-‐dried lipid nanoparticles The thermal characteristics of selected batches of nanoparticles were determined by differential scanning calorimetry thermal analysis using a 2920 DSC (Universal V3.6C TA Instruments, USA). The scan rate was 10 °C/min in the 64
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
temperature range from –10 °C to 275 °C and a N2 flow of 20 L/min. An empty pan was used as reference standard. Indium (purity ≥99.95 %, Fluka, Switzerland) was employed to check the calibration of the calorimetric system. 2.4.4. X-‐ray studies of freeze-‐dried lipid nanoparticles X-‐ray diffraction measurements were performed in order to clearly elucidate the solid state of both lipids and drug in lipid nanoparticles, using a Bruker D8 Advance X-‐ray diffractometer (Bruker Biosciences Española, S.A., Spain). The X-‐ray diffractogram was scanned with the diffraction angle increasing from 2° to 40°, 2θ angle, with a step angle of 0.02° and a count time of 1 s at a constant temperature of 25 °C. 2.5.
In vitro release studies
The release rate of edelfosine from lipid nanoparticles was measured in PBS medium (pH 7.4). Briefly, 5 mg of nanoparticles were dispersed in 1 ml of buffer solution and maintained at 37 °C under stirring (260 rpm). At appropriate time intervals, samples were centrifuged (23500 g, 10 minutes), supernatants were filtrated with a 0.45 μm pore diameter filter and kept at –20 °C until further HPLC– MS analysis was conducted as previously described. Three samples were employed for each time and the study was performed in triplicate. The mean in vitro dissolution time (MDT), a model-‐independent in vitro parameter that shows the meantime for edelfosine to release from the lipid nanoparticles under in vitro release conditions, was calculated according to the equation: Eq. (1) where ABCin vitro is the area between the release curve and its asymptote, calculated by the trapezoidal rule from time zero to the last measured time point, and M∞ is
65
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
the total amount of released drug at this time point. The release rate constant (kd) was calculated by the expression kd = 1/MDT.
3. Results and discussion 3.1.
Particle size, size distribution and zeta potential
Structurally, edelfosine has a structure of an amphifile, with a part of the molecule exhibiting hydrophobicity and another part exhibiting hydrophilicity, like a surface–active agent (Fig.1).
Figure 1. Chemical structure of edelfosine
Edelfosine-‐loaded lipid nanoparticles were obtained by either simple or multiple emulsion solvent evaporation method, and freeze-‐dried. These nanoparticles were then characterized to assess the effect of the different lipids and surfactant concentrations on mean particle size, size distribution and surface charge. Using lipids as matrices for the particles, different characteristics can be obtained by optimizing the formulation parameters such as type of lipids, surfactants, organic solvents and emulsifying procedure chosen [18]. The presence of a non-‐ionic surfactant is important to reduce the dynamic interfacial tension and to stabilize the nanosuspension. The surfactant is adsorbed on the nanoparticle surface, increasing the steric repulsion between particles. In this study, Tween® 80 was tested at two different concentrations (w/v), being sufficient to obtain small lipid nanoparticles and permitting the removal of its excess by centrifugation and washing.
66
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
The average size and polydispersity indices of lipid nanoparticles formulated with different lipids are reported in Table 1. All data are expressed as mean value ± standard deviation.
Table 1. Average size, polydispersity index (PDI) and zeta potential of lipid nanoparticles prepared either with methylene chloride (MC) or trichloromethane (TCM). Lipid
Solvent
Stabilizing agent
Size (nm)
PDI
ζ Potential (mV)
Stearic acid
MC
0.5% Tween 80
TCM Compritol® 888 ATO
TCM
480 ± 10.79
0.204 ± 0.01
–32,75
1% Tween 80
611 ± 5.29
0.626 ± 0.13
–35,15
0.5% Tween 80
415 ± 30.09
0.451 ± 0.09
–29,33
1% Tween 80
312 ± 7.51
0.225 ± 0.04
–32,4
1% Tween 80
372 ± 18.84
0.354 ± 0.06
–32,3
For all lipids, it was possible to obtain submicron sized lipid nanoparticles with two different concentrations of Tween® 80. In fact, all edelfosine loaded formulations showed a mean diameter in the range of 300 – 600 nm. There was a decrease in size of particles when formulated with edelfosine (data not shown), probably due to the effect of edelfosine as a surfactant agent. Smallest particles were obtained using stearic acid, chloroform and an aqueous solution of Tween® 80 at a concentration of 1 %. The polydispersity index (PDI) value was in the range of 0.2 – 0.6 for all lipid nanoparticles investigated. Zeta potential can make a prediction about the stability of colloid dispersions. A high zeta potential (> 30 mV) can provide an electric repulsion to avoid the aggregation of particles [25]. The incorporation of edelfosine into lipid nanoparticles had no significant influence on the zeta potentials of particles, which was negative in all cases. However, as the concentration of surfactant was increased in the formulation the zeta potential was found to be more negative (Table 1). This could be due to the presence of oleic acid traces in Tween on the particle surface, forming a denser surfactant film, thus eliciting an increased electrophoretic mobility.
67
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
3.2.
Drug entrapment efficiency and loading capacity
Over the past few years many different drugs had been successfully incorporated in lipid nano– or microparticles [20, 26-‐28]. Relatively high drug encapsulation efficiency for hydrophobic drugs was one of the major advantages of lipid nanoparticles [20]. It is also known that the lipid crystalline structure related to the chemical nature of the lipid is a key factor to determine whether a drug will be expelled or firmly incorporated into the carrier systems. In the nanoparticle structure, the lipid forming highly crystalline state with a perfect lattice would lead to drug expulsion. On the other hand, imperfections (lattice defects) of the lipid structure could offer space to accommodate the drugs [29]. As a result, the structure of less ordered arrangement in the nanoparticles would be beneficial to the drug loading capacity like the samples in this study.
Table 2. Encapsulation efficiency (EE) of the different nanoparticles prepared either with methylene chloride (MC) or trichloromethane (TCM).
Lipid
Solvent
Stabilizing agent
%EE
Stearic acid
MC
Tween 80 0,5%
5,10
Tween 80 1%
4,00
Tween 80 0,5%
9,20
Tween 80 1%
6,40
Tween 80 1%
84,35
TCM Compritol® 888 ATO
TCM
From the results listed in Table 2, it can be observed that the entrapment efficiency of edelfosine in the lipid nanoparticles prepared by the simple emulsion formation solvent evaporation method ranged from about 4 to 10 % for stearic acid nanoparticles. On the other hand, nanoparticles formulated using Compritol® encapsulated more than 80 % of the drug (Table 2). This high encapsulation efficiency in comparison to the stearic acid nanoparticles is likely to be due to the partially amorphous state of the Compritol® in the formulation, which allows more edelfosine to be incorporated among lipid chains. It can also be observed that the emulsifier concentrations investigated in this study (0.5 % and 1 % Tween 80) do not affect the encapsulation efficiency.
68
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
On the other hand, lipid nanoparticles prepared by the multiple emulsion solvent evaporation method showed very low encapsulation efficiency, less than 1 % (data not shown). 3.3.
Morphology
In order to investigate the shape and surface morphology of the Compritol® nanoparticles, atomic force microscopy was employed. The AFM images reveal the fine structure of the Compritol® lipid nanoparticle surface (Fig. 2A). They give clear 3D morphological images of spherical nanoparticles of sub–400 nm diameter and they also confirm that there was no aggregation or adhesion among the nanoparticles (Fig. 2C). Furthermore, the surface morphology of the nanoparticles could be seen closely from the AFM images. It was noticeable from the zoom–in picture (Fig. 2B) the smooth surface morphology of the nanoparticles.
A
B
C
Figure 2. Atomic force microscopy images of freeze-‐dried Compritol® lipid nanoparticles: multi-‐ particles (A); zoom-‐in of the selected area of A (B); 3D morphological image (C).
69
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
3.4.
DSC and X-‐ray diffractometry assays
Lipid nanoparticles were analysed by DSC and X-‐ray diffractometry to investigate the crystal pattern of both edelfosine and lipids, because this aspect could influence the in vitro and in vivo release of the drug from the systems. To probe this effect, analysis were performed on the following samples: edelfosine; Compritol® 888 ATO; stearic acid; edelfosine loaded stearic acid nanoparticles; edelfosine loaded Compritol® nanoparticles; unloaded stearic acid nanoparticles; unloaded Compritol® nanoparticles.
Figure 3. DSC thermograms of: (a) edelfosine, (b) stearic acid, (c) unloaded stearic acid lipid nanoparticles and (d) edelfosine-‐loaded stearic acid lipid nanoparticles (Fig. 3A); (a) edelfosine, (b) Compritol®, (c) unloaded Compritol® lipid nanoparticles and (d) edelfosine-‐loaded Compritol® lipid nanoparticles (Fig. 3B).
70
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Fig. 3 depicts the DSC thermograms of edelfosine loaded and unloaded lyophilised nanoparticles. As Compritol® is not composed of pure triacylglycerols, the observed melting peak at 72.43 °C might be due to a mixture of metastable polymorphic β and β’ forms. The heating run showed the just mentioned melting event at 72.43 °C and a relatively small endothermic shoulder at around 51 °C. This small shoulder corresponds to the melting of a very unstable modification of Compritol®, which is the α modification [30], that clearly disappears after the treatment with organic solvents for the nanoparticle preparation. DSC analysis of edelfosine-‐loaded lipid nanoparticles showed that the drug melting peak at 223°C is present neither in the stearic acid lipid nanoparticles nor in the Compritol® lipid nanoparticles whereas for the pure drug, the melting peak occurs before its decomposition. This thermal behaviour may be ascribed to the presence of edelfosine in an amorphous form or molecularly dispersed, but it may also be related to a possible solubilisation of the drug in the molten lipid when the DSC assay was performed. Besides, X-‐ray diffraction studies support this theory showing a partial crystalline state of the drug. This effect on the crystalline habit of edelfosine may be related to the preparative method of the lipid nanoparticles, in which edelfosine may be turned from a crystalline state to an amorphous one by the use of organic solvents like chloroform or methylene chloride. Thermal behaviour of lipids can also explain the different encapsulation efficiency of edelfosine. Unloaded stearic acid nanoparticles showed the melting peak of the stearic acid at 59 °C, indicating its presence in crystalline state, thus letting less amount of drug to be incorporated among its lipidic chains. On the other hand, Compritol® nanoparticles seem to lose part of their crystalline state transforming from a mixture of β and β’ polymorphs to the most stable β polymorph, permitting edelfosine to fit in the molecular gaps. These findings were confirmed by X-‐ray diffractometry assays.
71
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Figure 4. X-‐ray diffractograms of: (a) edelfosine, (b) stearic acid, (c) unloaded stearic acid lipid nanoparticles and (d) edelfosine-‐loaded stearic acid lipid nanoparticles (A); (a) edelfosine, (b) Compritol®, (c) edelfosine-‐loaded Compritol® lipid nanoparticles and (d) unloaded Compritol® lipid nanoparticles (B).
Unloaded stearic acid nanoparticles (Fig. 4A) show two sharp peaks corresponding to those of the stearic acid, whereas the diffraction pattern of bulk Compritol® (Fig. 4B) showed two main typical signals at 21.5 (2θ) and 23.5 (2θ) that are significantly modified when formulated into nanoparticles. Besides, when Compritol® nanoparticles are formulated, another signal arises at 19.4 (2θ) (Fig. 4B) which corresponds to the most stable polymorphic form of triacylglycerols β [30]. These results might indicate that the final formulation is composed of the 72
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
most stable polymorphic state of Compritol®. Comparing these results with edelfosine loaded lipid nanoparticles, some crystalline drug signal could still be detected, indicating a possible coexistence of edelfosine in both crystalline and amorphous states, being this last one the predominant. As a result, edelfosine would enrich the particle surface when formulated with stearic acid, whereas it would be incorporated among the lipid chains of Compritol®. 3.5.
In vitro release studies
The amount of edelfosine released from lipid nanoparticles was determined by an in vitro release assay, in an effort to assess whether edelfosine–incorporating lipid nanoparticles might be useful as a sustained-‐release dosage form. Fig. 5 displays the release profiles for nanoparticles fabricated with stearic acid or Compritol® using 1 % Tween 80 as emulsifier. When edelfosine was incorporated in stearic acid nanoparticles, a 63 % burst release of the drug was observed within 20 minutes (Fig.5). Conversely, the initial release burst for nanoparticles prepared using Compritol® was less than 40 % within the first 20 minutes. Figure 5. In vitro release profiles of edelfosine from stearic acid lipid nanoparticles (■) and Compritol® 888 ATO nanoparticles (▲).
The reason for the high initial release of edelfosine from the stearic acid nanoparticles could be the diffusion release of edelfosine distributed near the
73
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
surface and in the outer portion of the nanoparticles [12]. These results were in accordance with the observations made by DSC and X-‐ray diffractometry. Afterwards, the release rate slowed for both formulations, reaching 55 % after 24 h for nanoparticles prepared with Compritol®, and 98 % for the stearic acid nanoparticles. The release of a drug from the dosage form implies a crucial step, which is the dissolution of the drug. This process is ruled by a release rate constant (kd) that can be easily estimated and it is different depending, among other factors, on the composition of the particle. In our study, this constant was estimated for the release of edelfosine from lipid nanoparticles formulated either with stearic acid or Compritol®. Results showed that there is a different release profile depending on the lipid. Stearic acid presents a higher release rate constant (kd = 0.208) than Compritol® (kd = 0.178), indicative of a faster release. Besides, there seems to be no influence of the amounts of surfactant employed in the release kinetics of the drug, since the release rate constant from nanoparticles prepared with different surfactant concentrations (0.5 or 1 % Tween 80) were equivalent. Nevertheless, the in vitro release rates cannot be directly extrapolated to predict behaviour of these systems in biological environments. The differences between in vitro and in vivo degradation of the nanoparticles can be ascribed to the presence of enzymes [31] and the prolonged release of the drug from the nanoparticulate systems has to be quantified in an in vivo animal model.
4. Conclusions The present research paper proposed a novel formulation for edelfosine by using lipid nanoparticles. It can be concluded that Compritol® presents advantages as a matrix material for the manufacture of the nanoparticles and the controlled release of edelfosine. Current studies are aimed at evaluating the in vitro (in cell lines) and in vivo efficacy of these newly developed formulations.
74
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
Acknowledgements This work was supported by “Fundación Caja Navarra”. The authors are grateful to Nanotec Electrónica S.L. and Dr. Blanca Galar for her assistance in X–Ray diffraction studies. The first author acknowledges a fellowship from the Department of Education of the Basque Government (BFI06.37).
75
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
References 1.
Gajate, C. and F. Mollinedo, Biological activities, mechanisms of action and biomedical prospect of the antitumor ether phospholipid ET-‐18-‐OCH3 (edelfosine), a proapoptotic agent in tumor cells. Curr. Drug. Metab., 2002. 3(5): p. 491-‐525.
2.
Berdel, W.E., U. Fink, and J. Rastetter, Clinical phase I pilot study of the alkyl lysophospholipid derivative ET-‐18-‐OCH3. Lipids, 1987. 22(11): p. 967-‐9.
3.
Mollinedo, F., et al., Selective induction of apoptosis in cancer cells by the ether lipid ET-‐18-‐OCH3 (Edelfosine): molecular structure requirements, cellular uptake, and protection by Bcl-‐2 and Bcl-‐X(L). Cancer Res., 1997. 57(7): p. 1320-‐8.
4.
Gajate, C. and F. Mollinedo, Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and downstream signaling molecules into lipid rafts. Blood, 2007. 109(2): p. 711-‐ 9.
5.
Mollinedo, F., Antitumour ether lipids: proapoptotic agents with multiple therapeutic indications. Expert Opin. Therap. Patents., 2007. 17(4): p. 385-‐ 405.
6.
Azzouz, S., et al., Leishmanicidal activity of edelfosine, miltefosine and ilmofosine. Basic Clin Pharmacol Toxicol, 2005. 96(1): p. 60-‐5.
7.
Boggs, K.P., C.O. Rock, and S. Jackowski, Lysophosphatidylcholine and 1-‐O-‐ octadecyl-‐2-‐O-‐methyl-‐rac-‐glycero-‐3-‐phosphocholine inhibit the CDP-‐choline pathway of phosphatidylcholine synthesis at the CTP:phosphocholine cytidylyltransferase step. J. Biol. Chem., 1995. 270(13): p. 7757-‐64.
8.
Vogler, W.R., et al., The anticancer drug edelfosine is a potent inhibitor of neovascularization in vivo. Cancer Invest, 1998. 16(8): p. 549-‐53.
76
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
9.
Zheng, B., et al., Inhibition of protein kinase C, (sodium plus potassium)-‐ activated adenosine triphosphatase, and sodium pump by synthetic phospholipid analogues. Cancer Res, 1990. 50(10): p. 3025-‐31.
10.
Ahmad, I., et al., Enhanced therapeutic effects of liposome-‐associated 1-‐O-‐ octadecyl-‐2-‐O-‐methyl-‐sn-‐glycero-‐3-‐phosphocholine.
Cancer
Res.,
1997.
57(10): p. 1915-‐21. 11.
Bhamra, R., et al., Activity, pharmacokinetics and tissue distribution of TLC ELL-‐12 (liposomal antitumor ether lipid) in rats with transplantable, s.c. methylnitrosourea-‐induced tumors. Anticancer Drugs, 2003. 14(6): p. 481-‐6.
12.
zur Muhlen, A., C. Schwarz, and W. Mehnert, Solid lipid nanoparticles (SLN) for controlled drug delivery-‐-‐drug release and release mechanism. Eur. J. Pharm. Biopharm., 1998. 45(2): p. 149-‐55.
13.
Siekmann, B. and K. Westesen, Sub-‐micron sized parenteral carrier systems based on solid lipids. Pharm. Pharmacol. Lett., 1992. 1: p. 123.
14.
Schwarz, C. and W. Mehnert, Solid lipid nanoparticles (SLN) for controlled drug delivery. II. Drug incorporation and physicochemical characterization. J. Microencapsul., 1999. 16(2): p. 205-‐13.
15.
de Vringer, T. and H.A. de Ronde, Preparation and structure of a water-‐in-‐oil cream containing lipid nanoparticles. J. Pharm. Sci., 1995. 84(4): p. 466-‐72.
16.
Almeida, A.J., S. Runge, and R.H. Muller, Peptide-‐loaded solid lipid nanoparticles (SLN): Influence of production parameters. Int. J. Pharm., 1997. 149(2): p. 255-‐265.
17.
Wissing, S.A., O. Kayser, and R.H. Muller, Solid lipid nanoparticles for parenteral drug delivery. Adv. Drug Deliv. Rev., 2004. 56(9): p. 1257-‐72.
18.
Mehnert, W. and K. Mader, Solid lipid nanoparticles: production, characterization and applications. Adv. Drug Deliv. Rev., 2001. 47(2-‐3): p. 165-‐96.
77
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
19.
Hu, F.Q., et al., Preparation and characterization of stearic acid nanostructured lipid carriers by solvent diffusion method in an aqueous system. Colloids Surf. B Biointerfaces., 2005. 45(3-‐4): p. 167-‐73.
20.
Jaspart, S., et al., Solid lipid microparticles as a sustained release system for pulmonary drug delivery. Eur. J. Pharm. Biopharm., 2007. 65(1): p. 47-‐56.
21.
Muller, R.H., M. Radtke, and S.A. Wissing, Nanostructured lipid matrices for improved microencapsulation of drugs. Int. J. Pharm., 2002. 242(1-‐2): p. 121-‐ 8.
22.
Freitas, C. and R.H. Muller, Effect of light and temperature on zeta potential and physical stability in solid lipid nanoparticles (SLN) dispersion. Int. J. Pharm., 1998. 168: p. 221-‐229.
23.
Cavalli, R., et al., Sterilization and freeze -‐drying of drug-‐free and drug-‐loaded solid lipid nanoparticles. Int. J. Pharm., 1997. 148: p. 47-‐54.
24.
Blanco-‐Prieto, M.J., M.A. Campanero, and F. Mollinedo, Quantitative determination of the antitumor alkyl ether phospholipid edelfosine by reversed-‐phase liquid chromatography-‐electrospray mass spectrometry: application to cell uptake studies and characterization of drug delivery systems. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci., 2004. 810(1): p. 85-‐92.
25.
Levy, M.Y., et al., Characterization of diazepam submicron emulsion interface: role of oleic acid. J. Microencapsul., 1994. 11(1): p. 79-‐92.
26.
Chen, H., et al., Podophyllotoxin-‐loaded solid lipid nanoparticles for epidermal targeting. J. Control. Rel., 2006. 110(2): p. 296-‐306.
27.
Hu, F.Q., et al., Preparation of solid lipid nanoparticles with clobetasol propionate by a novel solvent diffusion method in aqueous system and physicochemical characterization. Int. J. Pharm., 2002. 239(1-‐2): p. 121-‐8.
28.
Lee, M.K., S.J. Lim, and C.K. Kim, Preparation, characterization and in vitro cytotoxicity of paclitaxel-‐loaded sterically stabilized solid lipid nanoparticles. Biomaterials, 2007. 28(12): p. 2137-‐46.
78
Chapter 2. Lipid nanoparticles for alkyl lysophospholipid edelfosine encapsulation: development and in vitro characterization
29.
Hou, D., et al., The production and characteristics of solid lipid nanoparticles (SLNs). Biomaterials, 2003. 24(10): p. 1781-‐5.
30.
Souto, E.B., W. Mehnert, and R.H. Muller, Polymorphic behaviour of Compritol888 ATO as bulk lipid and as SLN and NLC. J. Microencapsul., 2006. 23(4): p. 417-‐33.
31.
Olbrich, C. and R.H. Muller, Enzymatic degradation of SLN-‐effect of surfactant and surfactant mixtures. Int. J. Pharm., 1999. 180(1): p. 31-‐9.
79
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
Chapter 3 Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against mantle-‐cell lymphoma Ander Estella-‐Hermoso de Mendoza1, Miguel A. Campanero2, Hugo Lana1, Janny A. Villa-‐Pulgarin3,4, Janis de la Iglesia-‐Vicente3, Faustino Mollinedo3, María J. Blanco-‐ Prieto1
1Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia,
University of Navarra, E-‐31008, Spain 2Servicio de Farmacología Clínica, Clínica Universitaria, E-‐31080 Pamplona, Spain 3Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain 4APOINTECH, Centro Hispano-‐Luso de Investigaciones Agrarias (CIALE), Parque
Científico de la Universidad de Salamanca, C/ Rio Duero 12, E-‐37185 Villamayor, Salamanca, Spain
Key Words: edelfosine, lipid nanoparticles, bioavailability, pharmacokinetics, biodistribution, lymphoma Corresponding author: Dr. María J. Blanco-‐Prieto, Department of Pharmacy and Pharmaceutical Technology, School of Pharmacy, University of Navarra, C/Irunlarrea 1, E-‐31080 Pamplona, Spain, Office phone: + 34 948 425 600 ext. 6519, Fax: + 34 948 425 649, e-‐mail:
[email protected]
Submitted for evaluation.
81
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
Abstract To increase the oral bioavailability of the antitumor drug edelfosine, lipid nanoparticles (LN) made of synthetic lipids Compritol® 888 ATO and Precirol® ATO 5 were developed. Average size of particles was 110.4 ± 2.1 nm and 103.1 ± 2.9 nm, for Compritol® and Precirol® drug-‐loaded LN. Zeta potential was close to -‐ 21 mV in all cases, and encapsulation efficiency of edelfosine reached 85 % for both type of lipids. Pharmacokinetic and biodistribution profiles of the drug were studied after intravenous and oral administration of edelfosine-‐containing LN. Results showed that encapsulation of edelfosine in LN decreases hemolytic toxicity of the drug in 90 % and provides an increase in relative oral bioavailability of 1500 % after a single oral administration of drug-‐loaded LN, maintaining edelfosine plasma levels over 7 days in contrast to a single oral administration of edelfosine solution, which presents a relative oral bioavailability of 10 %. Pharmacokinetic parameters confirmed the slow elimination of the drug and a wide distribution throughout the body. Moreover, edelfosine-‐loaded LN showed a high accumulation of the drug in lymph nodes and showed slower tumor growth than the free drug in a murine lymphoma xenograft model, as well as potent extranodal dissemination inhibition.
83
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
1. Introduction Edelfosine is considered the prototype of a promising class of antitumor agents, collectively known as alkyl-‐lysophospholipid analogues or antitumor ether lipids. These agents present the singular characteristic of not targeting the DNA, but affecting the cell membrane and the apoptotic machinery of the cancer cell [1]. Recent in vitro studies have shown that edelfosine is preferentially uptaken by tumoral cells, sparing normal cells [2]. However, edelfosine presents some drawbacks when administered intravenously, as dose-‐dependent hemolysis hampers its administration at certain doses [3]. In addition, edelfosine presents bioavailability values below 10 % after a single oral administration of 30 mg/kg; however, this bioavailability increased to 64 % after multiple oral administration of the same dose after six days [2]. Owing to the drawbacks of this molecule, there has been an attempt to design new drug delivery systems that can modify the absorption rate, selectively transport the drug to the target, modifying the drug distribution profile and extending the drug release time in order to improve drug bioavailability, and decrease its toxicity. Among the different lipid-‐made colloidal carriers, edelfosine was incorporated into liposomes [4] and lipid nanoparticles (LN) made of biocompatible lipids [5]. The liposomal formulation was able to prevent the hemolytic toxicity of the drug, but the main inconvenience found was its rapid clearance from plasma. Edelfosine-‐loaded liposomes showed both in vivo and in vitro activity against methylnitrosourea-‐induced tumors, and it was approximately 4 -‐ 8 times less acutely toxic than free edelfosine. After a slow intravenous (i.v.) bolus injection of 12.5 mg/kg of edelfosine delivered as edelfosine-‐loaded liposomes to rats, several parameters were determined. Terminal half-‐life (t1/2) value was approximately 13.1 h. Clearance (CL) and volume of distribution at the steady-‐state (Vss) values were estimated at 0.0161 L/h/kg and 0.203 L/kg, respectively [4]. Edelfosine-‐loaded LN developed by our group were considered another alternative to deliver the drug to the organism [5]. These carriers are colloidal transporters composed of a biocompatible and biodegradable lipid matrix. They combine advantages of liposomes, polymeric nanoparticles and 85
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
emulsions, while diminishing possible drawbacks associated with them [6]. Lipids employed to form these lipid cores are biodegradable raw materials that are biocompatible: triglycerides (i.e. tristearin), partial glycerides (i.e. Compritol® 888 ATO and Precirol® ATO 5), fatty acids (i.e. stearic acid), steroids (i.e. cholesterol) or waxes (i.e. cetyl palmitate) [7]. The formulation methods are also diverse [7, 8]. However, most techniques employ organic solvents, which may imply regulatory and toxicity issues. Moreover, an improvement of LN over the liposomes is their physical and chemical long-‐term stability up to 12 -‐ 24 months [9]. The freeze-‐ drying process of LN has been shown to increase their physicochemical stability over long periods of time [10]. LN may also be used to target drugs at particular organs without surface modification because of their small size (50 -‐ 200 nm). Besides, LN have attracted rising interest for their ability to overcome certain biological barriers, resulting in increased therapeutic efficacy of the encapsulated drug and increase in tumor accumulation [11]. Mantle cell lymphoma (MCL) is a B-‐cell malignancy that comprises about 7 % of all non-‐Hodgkin’s lymphomas (NHLs), which is characterized clinically by extranodal disease in older male patients who present at an advanced stage [12]. Even though most patients initially gain a benefit from systemic treatments, the responses obtained are generally of limited duration. As a result, patients generally relapse with less responsive disease, showing a consistently bad outcome with a median overall survival from diagnosis of 43 months [13, 14]. Several approaches using more severe combination chemotherapy, namely stem cell transplantation, have shown higher response rates and more lasting remission in selected patients, but the greater part of MCL patients are not candidates for such dose-‐intensive regimens [15, 16]. MCL has recently become an area of intense clinical research, and it appears that median overall survival may be improving, but MCL is still considered incurable with current treatments [17]. Although several cytotoxic combinations including cisplatin, gemcitabine, carmustine and other alkylating drugs have been employed, their inherent toxicity is considered the main drawback at the time of election [18].
86
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
In a previous study, we determined that edelfosine displays a biexponential pharmacokinetic behavior in mice, presenting no significant differences regardless of the mouse strain employed [2]. The tissue distribution of edelfosine in mice shows that the drug is widely scattered across different organs, although it is preferentially internalized by the tumor both in vitro and in vivo. The present work tries to point out how the biodistribution and pharmacokinetic profile of edelfosine is altered compared to that of the free drug, when it is encapsulated in LN. The efficacy of the chemotherapeutic potential of edelfosine loaded LN via the oral route in experimental murine lymphoma xenograft model was also evaluated.
2. Materials and Methods 2.1.
Chemicals
Edelfosine was from APOINTECH (Salamanca, Spain). Compritol® 888 ATO and Precirol® ATO 5 were a gift from Gattefossé (Lyon, France). Tween® 80 was obtained from Roig Farma (Barcelona, Spain). Platelet Activating Factor (PAF) and PBS (10 mM phosphate, 0.9 % NaCl) were obtained from Sigma-‐Aldrich (Madrid, Spain). Chloroform was purchased from Panreac (Madrid, Spain) and methanol was obtained from Merck (Barcelona, Spain). All other solvents were of analytical grade. 2.2.
Cell culture
The human mantle-‐cell lymphoma cell line JVM-‐2 (DSMZ, Germany) was grown in RPMI-‐1640 containing 10 % heat-‐inactivated fetal calf serum (GIBCO/BRL, Carlsbad, CA, USA), 2 mM of L-‐glutamine, 100 μg/ml streptomycin sulphate (Sigma) and 100 U/ml penicillin (Sigma), at 37 °C in a humidified atmosphere of air containing 5 % CO2.
87
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
2.3.
Preparation of LN incorporating edelfosine
LN were prepared by a warm oil-‐in-‐water (o/w) microemulsion followed by high shear homogenization and ultrasonication. The lipid phase consisted of either Compritol® 888 ATO or Precirol® ATO 5 along with edelfosine, while the aqueous phase consisted of a 2 % Tween® 80 aqueous solution. The aqueous phase was heated at about 5 °C above the melting point of the lipid and added to the melted lipid phase at the same temperature. The mixture was dispersed with the help of a MicrosonTM ultrasonic cell disruptor (NY, USA) for 1 minute. The preformed emulsion was then homogenised in an Ultraturrax® (IKA-‐Werke, Germany) for 1 minute and sonicated again with the MicrosonTM ultrasonic cell disruptor (NY, USA) for 1 minute. The nanoparticle suspension was cooled in an ice bath and washed twice with filtered water by diafiltration with Amicon Ultra-‐15 filters of 10,000 dalton molecular weight cut-‐off membrane (Millipore®, Cork, Ireland) to remove the excess of surfactant. Nanoparticles were then resuspended in PBS for animal administration or in 10 % trehalose solution for freeze-‐drying. 2.4.
Characterization of edelfosine loaded LN 2.4.1. Particle size and zeta potential
The average particle size and polydispersity index of edelfosine loaded LN were determined by photon correlation spectroscopy (PCS) using a Zetasizer Nano (Malvern Instruments, UK). Each sample was diluted 30 fold in distilled water until the appropriate concentration of particles was achieved to avoid multiscattering events. The obtained homogenous suspension was examined to determine the volume, mean diameter, size distribution and polydispersity and repeated three times for each sample. Similarly, the zeta potential was measured using the same equipment with a combination of Laser Doppler (LD) velocimetry. Samples were diluted with distilled water and each experiment was performed in triplicate. All data are expressed as a mean value ± standard deviation.
88
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
2.4.2. Loading capacity Edelfosine was quantified by an ultra high-‐performance liquid chromatography tandem mass spectrometry (UHPLC-‐MS/MS) method that had been previously validated [19]. The drug was extracted from a sample of 10 mg of lyophilized nanoparticles, to which 1 ml of chloroform was added in order to dissolve them. 10 μL of the internal standard PAF (0.2 mg/mL) were then spiked to the samples. 3 mL of methanol were added to the mixture, and after vortex mixing for 1 min at room temperature and centrifuging at 20,000 × g for 10 min, 2 μL aliquots of the supernatant were injected into the chromatographic system. 2.5.
Hemolysis experiments
Erythrocytes from fresh human blood were separated from plasma by centrifugation (4,000 × g, 7 min) and washed three times with PBS. 4 mL of the washed erythrocyte suspension were diluted to 100 mL with PBS. 1.5 mL of this suspension were treated with 0.5 mL of tested samples: free edelfosine (10 μg/mL), edelfosine loaded Compritol® and Precirol® nanoparticles (10 μg/mL) and drug free Compritol® and Precirol® nanoparticles. Absorbance was measured in an Agilent 8453 UV-‐visible spectrophotometer (Agilent, Palo Alto, CA, USA) at 540 nm 1 h after the treatment. 2.6.
Animal experiments
Animal handling was conducted in compliance with the regulations of the Ethical Committee of the University of Navarra as well as with the European Community Council Directive Ref. 86/609/EEC. For pharmacokinetic studies, BALB/c mice (20 g) were obtained from Harlan Interfauna Ibérica S.L. (Barcelona, Spain). For efficacy studies, SCID mice (Janvier, Genest St Isle, France) were employed. Animals received a standard diet and water ad libitum, except for the animals that received the oral doses, which were fasted for 24 hours prior to administration. 89
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
2.6.1. Pharmacokinetic and biodistribution studies after intravenous administration An i.v. single dose of edelfosine-‐loaded LN (50 mg/kg) was administered to BALB/c (n=8 per group) mice via the tail vein. Group 1 received Compritol® 888 ATO LN and group 2, Precirol® ATO 5 LN. At various time points after administration (0, 1, 2, 5, 8, 24 and 31 h for Compritol® group and 0, 1, 2, 5, 8, 24, 48, 72, 96, 120, 144 and 168 h for Precirol® group), blood was collected in EDTA surface-‐coated tubes and then centrifuged at 2,000 × g for 10 min (4 °C) to separate the plasma (100 μL). Then, animals were sacrificed and spleen, liver, lungs, kidneys, heart, stomach and intestine were collected and weighed. Tissues were homogenised in 1 mL of PBS pH=7.4 using a Mini-‐bead Beater (BioSpect Products, Inc., Bartelsville, Oklahoma, USA) and centrifuged at 10,000 × g for 10 min. Both plasma and tissue supernatants were collected and stored at -‐80 °C until UHPLC-‐MS/MS analysis was performed. 2.6.2. Pharmacokinetic and biodistribution studies after oral administration Two BALB/c mice groups were treated with a single oral administration of edelfosine-‐loaded LN (edelfosine concentration of 50 mg/kg, n=8 per group). Group 1 received Compritol® 888 ATO LN and group 2, Precirol® ATO 5 LN. At various time points after the administration (0, 1, 2, 5, 8, 24, 48, 72, 96, 120, 144 and 168 h for Compritol® group and 0, 1, 2, 5, 8, 24, 48, 72, 96, 120, 144, 168, 192 and 216 h for Precirol® group), blood was collected in EDTA surface-‐coated tubes and then centrifuged at 2,000 × g for 10 min (4 °C) to collect plasma (100 μL). After sacrifice by cervical dislocation, tissues were collected, weighed and processed as explained above. 2.6.3. Lymphatic absorption studies A group of BALB/c mice received an oral dose of both types of LN (edelfosine concentration of 50 mg/kg, n=8). 24 hours later, an oily emulsion (milk) was orally administered 1 hour prior to sacrifice, in order to make the lymph ducts and nodes
90
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
more visible, animals were sacrificed and mesenteric lymph nodes were extracted and analyzed by UHPLC-‐MS/MS. 2.6.4. Efficacy studies Eight-‐week SCID mice were subcutaneously inoculated into the lower dorsum with 1 x 107 JVM-‐2 cells in 100 μL of PBS and 100 μL of Matrigel basement membrane matrix (Becton Dickinson, San Jose, CA). Animals received a standard diet and water ad libitum. When tumors were palpable, mice were randomly assigned to the treatment groups. Six groups of mice (n=8 per group) were treated orally: group 1: PBS; group 2: edelfosine solution (30 mg/kg dissolved in PBS); group 3: edelfosine-‐loaded Compritol® 888 ATO nanoparticles (edelfosine concentration of 30 mg/kg); group 4: edelfosine-‐loaded Precirol® ATO 5 nanoparticles (edelfosine concentration of 30 mg/kg); group 5: blank Compritol® 888 ATO nanoparticles (10 mg/mL lipid concentration); and group 6: blank Precirol® ATO 5 nanoparticles (10 mg/mL lipid concentration, equivalent to a 30 mg/kg edelfosine dose). The treatments were administered by oral gavage every four days. The experiment ended when control group tumors reached a volume of 5.0±0.5 cm3. At this point, animals were sacrificed and tumors were collected for the determination of their volume and weight. Axilary, inguinal and mesenteric lymph nodes were also extracted from MCL-‐bearing mice treated with either free or vectorized edelfosine and macroscopically analyzed. 2.7.
Data analysis
Pharmacokinetic analysis was performed with plasma samples obtained from experiments with all mice. All these plasma concentration data were analyzed by non-‐compartmental and compartmental analysis using WinNonlin Professional Edition Version 2.1 (Pharsight, Mountain View, CA, USA). The area under the plasma concentration vs. time curve (AUC) was determined using the log-‐linear trapezoidal rule with extrapolation to infinitum and normalized against the dose. The CL value is the volume of plasma completely cleared of a specific compound
91
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
per unit time by the organism; it was calculated by dividing the dose by AUC. The maximum plasma concentration (Cmax) was calculated from the plasma concentration-‐time curve and normalized against the dose. Oral bioavailability (F) was determined by ratio of the dose-‐normalized AUC following oral and i.v. administration. Vss is the volume of fluid that would be required to contain the amount of drug in the body if it were uniformly distributed at a concentration equal to that in the plasma. The t1/2 value refers to the time taken for plasma concentration to fall by 50 %, and it was determined using the following formula: t1/2=ln (2)·Vss/CL. 2.8.
Statistical analysis
The presence of differences in tissue/plasma ratios and pharmacokinetic parameters was measured by the Mann Whitney test for double comparisons using Social Package of Statistical Sciences (SPSS). Student’s t test was used for measuring differences in efficacy and tumor dissemination inhibition studies. A value of p < 0.05 was considered to be statistically significant for all statistical tests.
3. Results and discussion
Traditionally, nanoparticulated drug delivery systems have been formulated using organic solvents as solubilizing agents. However, toxicity of organic solvents is a special concern as it may lead to the dismissal of any proposed formulation regardless of potential benefits. After administration, nanoparticles are generally uptaken by macrophages or the mononuclear phagocytes in the organism, leading to generation of inflammatory and tissue response. Thus, the solvents used, degradation products of the nanoparticles, adhesion of these degradation products to the membrane and consequent stimulation of cells and release of inflammatory mediators may be the cause of the toxicity of nanoparticles. The avoidance of organic solvents during the preparation of LN by the warm oil-‐in-‐water (o/w)
92
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
microemulsion followed by high shear homogenization and ultrasonication method decreases the mentioned potential toxicity. Edelfosine-‐loaded LN were produced by a warm oil-‐in-‐water (o/w) microemulsion followed by high shear homogenization and ultrasonication method, and freeze-‐dried. This solvent-‐free method does not entail the utilization of toxic organic solvents with obvious implications from a biocompatibility point of view.
3.1.
Particle size, size distribution and zeta potential
The peroral route is the most preferred route of administration, but this path is limited for many substances as they present poor oral bioavailability due to biopharmaceutical (low solubility, low permeability, and/or instability in gastrointestinal environment) and pharmacokinetic (extensive first pass metabolism and/or rapid clearance) drawbacks in their delivery approach. Therefore, the development of delivery systems that would be able to overcome these drawbacks is essential to ensure the effectiveness of such molecules. The physical-‐chemical characteristics of the developed nanoparticles are compiled in Table I. All data are expressed as mean value ± standard deviation.
Table I. Average size, PDI, zeta (ζ) potential, encapsulation efficiency (EE) and drug loading of edelfosine-‐loaded LN (n=20) prepared by the warm microemulsion formation followed by high shear homogenization and ultrasonication method.
ζ Potential (mV)
% EE
Drug loading (μg edelfosine/mg form.)
Drug-‐free 130.6 ± 3.1 0.275 ± 0.021 Compritol® LN
-‐28.6 ± 2.1
−
−
Drug-‐loaded 110.4 ± 2.1 0.241 ± 0.050 Compritol® LN
-‐21.2 ± 1.5 84.68 ± 7.18
LN
Size (nm)
PDI
Drug-‐free Precirol® LN
117.7 ± 2.4 0.243 ± 0.026
-‐29.1 ± 1.7
Drug-‐loaded Precirol® LN
103.1 ± 2.9 0.231 ± 0.012
-‐22.4 ± 2.0 82.62 ± 5.73
−
17.57 ± 1.97 − 13.95 ± 0.79
93
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
Drug-‐free LN presented diameters of 130.6 ± 3.1 and 117.7 ± 2.4 for Compritol® and Precirol® LN, respectively. The average diameter of edelfosine-‐loaded Compritol® LN was 110.4 ± 2.1 nm, while drug-‐loaded Precirol® LN presented a mean diameter of 103.1 ± 2.9 nm, suggesting that edelfosine might be responsible for the reduction in size of LN, as it is a surfactant structured molecule. These particles present a smaller size than those prepared by the emulsion formation and solvent evaporation method previously developed [5]. This method therefore provides smaller particles, which are organic solvent-‐free, in less formulation time. PDI was below 0.3 in all cases indicating that the LN were homogenous in size. These particles present an appropriate size for their oral administration, since sizes below 300 nm are suitable for intestinal transport to the thoracic duct [20]. Zeta (ζ) potential can make a prediction about the stability of colloid dispersions. A high ζ potential (>|30| mV) can provide an electric repulsion to avoid the aggregation of particles [21]. In our case, the ζ-‐potential values measured in double-‐distilled water were negative. The mean ζ potentials of edelfosine-‐loaded and drug-‐free Compritol® LN were -‐21.2 ± 1.5 mV and -‐28.6 ± 2.1 mV, respectively. Precirol® LN showed similar values of -‐22.4 ± 2.0 mV and -‐29.1 ± 1.7 mV for edelfosine-‐loaded and drug-‐free Precirol® LN, respectively. The incorporation of edelfosine into LN had little influence on the zeta potentials of nanoparticles. This negative surface charge could be due to the presence of oleic acid traces in Tween® 80 on the particle surface, forming a denser surfactant film, and thus eliciting increased electrophoretic mobility. Besides, the steric impediment of Tween® 80 might be another effect which would increase the stability of colloidal dispersions [22].
3.2.
Loading capacity
It is well known that the crystalline state in the LN structure leads to faster drug expulsion. However, lattice defects of the lipid structure offer space to accommodate the drugs [23]. As a result, the structure of less ordered arrangement in the nanoparticles should be beneficial to the drug loading capacity, as in the case of the particles developed in this work.
94
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
The entrapment efficiency of edelfosine in the LN prepared by the warm microemulsion followed by high shear homogenization and ultrasonication was similar for nanoparticles prepared with both types of lipids (Table I). Nanoparticles formulated using Compritol® encapsulated 84.68 ± 7.18 % of edelfosine (equivalent to 17.57 ± 1.97 μg edelfosine/mg formulation), while Precirol® LN encapsulated 82.62 ± 5.73 % of the drug (13.95 ± 0.79 μg edelfosine/mg formulation). This high encapsulation efficiency is likely to be due to the partially amorphous state of the lipids in the formulation, which allows more edelfosine to be incorporated among lipid chains [5]. 3.3.
Pharmacokinetic
characterization
and
biodistribution
after
intravenous administration Figures 1A and 1B show the concentration of edelfosine in mouse plasma plotted against time after a single i.v. administration of edelfosine-‐loaded LN (concentration ranging 30 -‐ 60 mg/kg) to BALB/c mice.
Figure 1. Time-‐concentration curve data of edelfosine after a single intravenous administration of edelfosine loaded (A) Compritol® and (B) Precirol® LN to BALB/c mice (n=8, mean ± S.D.)
Dose-‐normalized pharmacokinetic analysis of edelfosine in blood plasma showed a Cmax of approximately 0.3 μg/mL for both types of LN, showing no statistical differences between them. All obtained pharmacokinetic parameters were dose-‐normalized and are listed in Table II.
95
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
Table II. Comparison of pharmacokinetic parameters of edelfosine after intravenous administration of edelfosine loaded LN (50 mg/kg bw, n=8 per group, mean ± SD). No statistical differences were found among parameters p>0.05. Parameters
Compritol® 888 ATO LN i.v. Precirol® ATO 5 LN i.v. administration administration
t½α (h)
0.395 ± 0.124
0.505 ± 0.151
t½β (h)
16.970 ± 5.775
23.718 ± 17.743
Cmax / D (μg/mL/μg)
0.290 ± 0.087
0.341 ± 0.044
CL (L/h/kg)
0.105 ± 0.021
0.065 ± 0.023
Vss (L/kg)
1.668 ± 0.730
1.313 ± 0.838
MRT (h)
17.154 ± 6.517
26.096 ± 20.598
AUCinf / D (μg/mL/μg)
0.573 ± 0.053
0.894 ± 0.399
Plasma concentration-‐time data of edelfosine in LN were well described by bi-‐ exponential functions following i.v. administration of both types of LN. The distribution half-‐lives (t1/2α) of the two formulations were around 0.4 h, while the elimination half-‐lives (t1/2β) were 17 h for Compritol® LN and 23 h for Precirol® LN, suggesting a much slower elimination of these last nanoparticles. These parameters are higher than those of the liposomes described by Bhamra et al. [4], indicating that these LN circulate in plasma for a longer period of time. The rest of the pharmacokinetic parameters showed no statistical differences between the two types of LN. The mean systemic CL and Vss values for edelfosine-‐ loaded LN were around 0.08 L/h/kg and 1.5 L/kg, respectively. When edelfosine was loaded into liposomes, the Vss was 7 times lower than this value, suggesting that LN allow a broader distribution of the drug in the body. There was little variability in most of the values of the pharmacokinetic parameters, indicating a well-‐controlled and reproducible study, except for the elimination phase half-‐life value. AUC values were between 0.6 and 0.9 μg·h/mL/μg, similar to those obtained in a previous research work [2] after the i.v. administration of edelfosine solution,
96
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
indicating that the i.v. administration of edelfosine in LN presents a bioavailability of 100 %. Figure 2 depicts the scattering of the drug through the mouse body after i.v. administration of edelfosine-‐loaded LN, expressed as tissue/plasma ratios. Figure 2. Tissue/plasma concentration ratios of edelfosine after a single intravenous dose of edelfosine-‐loaded Compritol® and Precirol® LN to BALB/c mice (n=8)
Whichever the type of nanoparticle employed, the highest accumulations of the drug were achieved in kidney, intestine and liver, followed by spleen, stomach and lung, with no statistical differences between the ratios. 3.4.
Pharmacokinetic characterization and biodistribution after oral
administration Pharmacokinetic studies were performed after a single oral administration of 50 mg/kg of edelfosine-‐loaded in LN. This oral dose was well tolerated by the mice and no hemolytic side effects or body weight loss was observed (data not shown). Figure 3 shows the concentration of edelfosine in mouse plasma plotted against 97
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
time after a single oral administration of edelfosine-‐loaded LN to BALB/c mice. The endpoint of the experiment was the day after the concentration of edelfosine in plasma reached 0.5 μg/mL.
Figure 3. Time-‐plasma concentration curve data of edelfosine obtained with the WinNonLin program after a single oral administration of edelfosine solution and edelfosine loaded Compritol® and Precirol® LN to mice (n=8 per group)
It can be observed that unlike edelfosine solution, drug-‐loaded LN enhanced the absorption of the drug maintaining detectable concentrations for over seven days. As a result, the maintenance times in plasma achieved with the developed LN were much longer than those observed by different studies recently published with the same Compritol® lipid, which did not last longer than 24 h [24, 25]. All dose-‐normalized pharmacokinetic parameters obtained after the administration of edelfosine in LN are summarized in Table III.
98
Chapter 3. Lipid nanoparticles loaded with alkyl-‐lysophospholipid edelfosine: Pharmacokinetic profile, biodistribution studies and in vivo efficacy against MCL
Table III. Comparison of pharmacokinetic parameters of edelfosine after oral administration of edelfosine loaded LN (50 mg/kg bw, n=8 per group, mean ± SD). Asterisks indicate significantly different values between the two types of LN, p99.9%) was provided by
Praxair (Madrid, Spain). Chemical structures of edelfosine and PAF are shown in Fig. 1. 2.2. Instruments and analysis conditions 2.2.1. HPLC–MS method This method is based on a previous method developed in our group [14]. The apparatus used for the HPLC analysis was a Model 1100 series LC coupled with an atmospheric pressure–electrospray ionization (ESI) single quadrupole mass spectrometer equipped with a collision-induced dissociation cell (HP 1100 with MSD VL, Waldbronn, Germany). Separation was carried out at 50 ◦ C on a reversed-phase, 150 mm × 3 mm column packed with C18 , 5 !m silica reversed-phase particles (Gemini® ) obtained from Phenomenex® (Torrance, CA, USA). This column was preceded by a reversed-phase, C18 , 5 !m guard column (SecurityGuardTM , 20 mm × 4 mm, Phenomenex® , Torrance, CA, USA). The mobile phase was a mixture of methanol–1% formic acid (95:5, v/v). Separation was achieved by isocratic solvent elution at a flow rate of 0.5 mL/min. The MS was operated in the positive ESI mode. The detection of edelfosine and the internal standard was performed by selected ionization monitoring (SIM) mode. ESI-MS conditions were as follows: source temperature 350 ◦ C, capillary voltage 4 kV, and collision-induced dissociation voltage 140 V. Nitrogen was used as the desolvation gas with a flow rate of 12 L/min and a pressure of 30 psi (1 psi = 6894.76 Pa). Optimization of the interface variables, such as gas flows and voltages was done manually during direct infusion of 10 !g/mL of the target analyte dissolved in methanol. The spectrometer was programmed to monitor both the ion of edelfosine at m/z 524.4 and platelet activating factor at m/z 574.4. Under these conditions, edelfosine and I.S. were eluted at 3.65 and 3.50 min, respectively. Data acquisition and analysis were performed with a Hewlett-Packard computer using the ChemStation G2171 AA program (Agilent, Palo Alto, CA, USA). 2.2.2. UHPLC–MS/MS method The UHPLC system was composed of an Acquity UPLCTM system (Waters Corp., Milford, MA, USA) with thermostatized autosampler and column compartment. Separation was carried out on an Acquity UPLCTM BEH C18 column (50 mm × 2.1 mm, 1.7 !m; Waters Corp., Milford, MA, USA) with isocratic elution using a mobile phase composed of 5% of a 1% formic acid aqueous solution and 95% of methanol. Column temperature was maintained at 50 ◦ C. The flow rate was set at 0.5 mL/min. The autosampler was conditioned at 4 ◦ C and the injection volume was 2 !L using partial loop mode for sample injection. Triple-quadrupole tandem mass spectrometric detection was performed on an AcquityTM TQD mass spectrometer (Waters Corp., Milford, MA, USA) with an electrospray ionization (ESI) interface. The mass spectrometer operated in positive mode was set up for multiple reaction monitoring (MRM) to monitor the transition of m/z 524.3 → 104.2 for edelfosine and the transition of m/z 552.3 → 184.2 for the internal standard, with the dwell time of 0.1 s per transition. To optimize the MS parameters, standard solutions of both the analyte and internal standard were infused into the mass spectrometer. The following optimized MS parameters were employed: 4 kV capillary voltage, 60 V cone voltage for edelfosine and 30 V for the internal standard, 150 ◦ C source temperature and 350 ◦ C desolvation temperature. Nitrogen was used for the desolvation and as cone gas at a flow rate of 650 and 50 L/h, respectively. Argon was used as the collision gas. The optimized collision energy for edelfosine was 30 and 20 eV for the internal standard. Under these conditions, edelfosine and I.S. were eluted at 1.23 ± 0.01 and 1.19 ± 0.02 min, respectively. Data acquisition and analysis were performed using the MassLynxTM
260
Annex II. Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems
A. Estella-Hermoso de Mendoza et al. / J. Chromatogr. B 877 (2009) 4035–4041
4037
Fig. 1. Chemical structure of (A) Edelfosine and (B) platelet activation factor.
NT 4.1 software with QuanLynxTM program (Waters Corp., Milford, MA, USA). 2.3. Preparation of standard and quality control (QC) samples 2.3.1. Standard solutions and QC samples for plasma Stock solutions of edelfosine were prepared in methanol. The stock solution of PAF was prepared in methanol at a concentration of 0.2 mg/mL. For the HPLC, two calibration ranges had to be established for sample quantitation between 0.1 and 75 !g/mL. The calibration curves for concentration range from 0.1 to 1 !g/mL were
prepared by adding 50 !L of the standard solutions of 0.2, 1 and 2 !g/mL to 100 !L of blank mouse plasma. Effective concentrations of edelfosine were 0.1, 0.5 and 1 !g/mL. Calibration curves for a concentration range from 1 to 75 !g/mL were prepared by adding 50 !L of the standard solutions of 2, 30, 60 and 150 !g/mL to mouse plasma. Effective concentrations of edelfosine were 1, 15, 30 and 75 !g/mL. The QC samples were pooled at concentrations of 0.1, 5, 10 and 50 !g/mL. For the UHPLC, calibration curves for a concentration range from 0.0075 to 75 !g/mL were prepared by adding 50 !L of the standard solutions of 0.015, 6, 60 and 150 !g/mL to 100 !L of blank mouse plasma. Effective concentrations of edelfosine were
Fig. 2. Representative chromatograms of: (A) blank mouse plasma sample; (B) edelfosine resulting from the analysis of mouse plasma obtained at 1 min after intravenous administration of 200 !g of edelfosine; (C) Platelet Activating Factor (PAF) resulting from the analysis of mouse plasma after intravenous administration of 200 !g of edelfosine (PAF concentration = 0.2 mg/mL); (D) blank mouse plasma sample; (E) plasma sample after an intravenous administration of 200 !g of edelfosine to a BALB/c mouse and (F) Platelet Activating Factor (PAF) resulting from the analysis of mouse plasma after intravenous administration of 200 !g of edelfosine (PAF concentration = 0.2 mg/mL).
261
Annex II. Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems
4038
A. Estella-Hermoso de Mendoza et al. / J. Chromatogr. B 877 (2009) 4035–4041
0.0075, 3, 30 and 75 !g/mL. The QC samples were pooled at concentrations of 0.075, 0.75, 15 and 60 !g/mL. The spiked plasma samples (standard and QC) were then processed following the application of the extraction procedure.
at 20,000 × g for 10 min, 200 !L of the supernatant were mixed with 800 !L of methanol and then, 5 and 2 !L aliquots were injected into the HPLC and UHPLC systems, respectively. 2.5. Method validation
2.3.2. Standard solutions and QC samples for tissues Stock solutions of edelfosine were prepared in methanol. The stock solution of PAF was prepared in methanol at a concentration of 0.2 mg/mL. For the HPLC, calibration curves were prepared by adding 50 !L of the standard solutions of 0.4, 4, 8, 16 and 63.5 !g/mL to tissue homogenate in buffer (see Section 2.6). Effective concentrations of edelfosine were 0.2, 2, 4, 8 and 31.75 !g/mL. The QC samples were pooled at concentrations of 0.1, 1.72, 17.19 and 31.75 !g/mL. For the UHPLC, calibration curves for a concentration range from 0.0075 to 75 !g/mL were prepared by adding 50 !L of the standard solutions of 0.015, 6, 60 and 150 !g/mL to the tissue homogenate. Effective concentrations of edelfosine were 0.0075, 3, 30 and 75 !g/mL. The QC samples were pooled at concentrations of 0.075, 0.75, 15 and 60 !g/mL. The spiked tissue samples (standard and QC) were then processed after application of extraction procedure. 2.3.3. Standard solutions and QC samples for lipid nanoparticles The stock solution of PAF was prepared in methanol at a concentration of 0.2 mg/mL. Stock solutions of edelfosine were prepared in methanol. For the HPLC, effective concentrations of edelfosine were 1, 15, 30 and 75 !g/mL. The QC samples were pooled at concentrations of 1, 5, 10 and 50 !g/mL. For the UHPLC, effective concentrations of edelfosine were 0.0075, 3, 30 and 75 !g/mL. The QC samples were pooled at concentrations of 0.075, 0.75, 15 and 60 !g/mL. The edelfosine standards and QCs were then injected into the chromatographic system. 2.4. Sample preparation 2.4.1. Plasma samples Blood was collected in EDTA surface-coated tubes and then centrifuged at 2000 × g for 10 min (4 ◦ C) to separate the plasma. A portion of 100 !L of mouse plasma was transferred to a 1.5-mL tube and then 10 !L of PAF (0.2 mg/mL), used as internal standard (I.S.) were spiked to the samples. Then, 190 !L of mobile phase (methanol–1% formic acid (95:5, v/v)) were added and the mixture was vortex-mixed at room temperature for 1 min for precipitation. After centrifuging at 20,000 × g for 10 min, 200 !L of the supernatant were mixed with 800 !L of methanol and then, 5 and 2 !L aliquots were injected into the HPLC and UHPLC systems, respectively. 2.4.2. Tissue samples A portion of 100 !L of tissue homogenate in buffer (see Section 2.6) was transferred to a 1.5-mL tube and then, 10 !L of I.S. (0.2 mg/mL) were spiked to the samples and 10 !L of 10% TCA were added to the mixture and vortex-mixed for 10 s for protein precipitation. Finally, 180 !L of mobile phase were added to the mixture. After vortex mixing for 1 min at room temperature and centrifuging at 20,000 × g for 10 min, 200 !L of the supernatant were mixed with 800 !L of methanol and then, 5 !L and 2 !L aliquots were injected into the HPLC and UHPLC systems, respectively. 2.4.3. Lipid nanoparticles A sample of 10 mg of lyophilized nanoparticles was weighed in a 10-mL tube and then, 1 mL of chloroform was added in order to dissolve the nanoparticles. 10 !L of I.S. (0.2 mg/mL) were spiked to the samples. Then, 3 mL of mobile phase were added to the mixture. After vortex mixing for 1 min at room temperature and centrifuging
The selectivity of the assay was determined by the individual analysis of blank samples. The retention times of endogenous compounds in the matrix were compared with those of edelfosine and PAF. LOD was defined as the sample concentration resulting in a peak area of three times the noise level. LOQ was defined as the lowest drug concentration, which can be determined with an accuracy and precision < 20%. In this work LOD of the assay method was determined by analysis of the peak baseline noise in ten blank samples. Plasma samples were quantified using the internal standard method. Standard curves were calculated using linear least squares regression between theoretical edelfosine concentration on calibrator samples and the chromatographic peak area ratios of edelfosine to that of the internal standard. To evaluate linearity, calibrator samples were prepared and analyzed in duplicate on 3 separate days. Accuracy and precision were also determined by replicate measurements (n = 6) of quality control samples at four concentration levels on five different validation days. The accuracy was expressed as (real concentration − theoretical concentration)/(theoretical concentration) × 100 and the precision by the CV (%) of the measured concentration values obtained after analysis of the quality control samples with different nominal concentration values. The absolute extraction recoveries of edelfosine at three QC levels were evaluated by measuring the samples as described above and comparing the peak areas of the edelfosine and the I.S., and then comparing with those obtained from direct injection of the compounds dissolved in the supernatant of the processed blank biological samples (plasma or the different tissues). The matrix effect was evaluated by comparing the peak area of the analyte dissolved in the reconstituted residues of processed blank plasma with the standard solutions at the same concentration dissolved in mobile phase. The matrix effect was evaluated at three different concentration levels, with three samples analyzed in each set. The matrix effect of the internal standard was evaluated at the concentration in plasma samples using the same method. 2.6. Application of the method The present method has been successfully applied to the quantitation of edelfosine in biological matrices and lipid nanocarriers. To demonstrate the reliability of this method for the study of edelfosine pharmacokinetics, this assay was applied to the quantitation of edelfosine in plasma samples obtained from 6 BALB/c mice treated with an intravenous dose of 200 !g of an edelfosine solution. Tissue distribution of edelfosine was also studied. Blood samples were withdrawn at 0, 1, 2, 5, 8 and 24 h postadministration in EDTA surface-coated tubes and then centrifuged at 2000 × g for 10 min (4 ◦ C) to separate the plasma (100 !L). Plasma was stored frozen (−80 ◦ C) until analysis. Then, the animals were sacrificed and the spleen, liver, lungs, kidneys, heart, brain, stomach and intestine were collected and weighed. Tissue samples were homogenized in 1 mL of Phosphate Buffered Saline (PBS) using a Mini-bead Beater (BioSpect Products, Inc., Bartelsville, OK, USA) and centrifuged at 10,000 × g for 10 min. Supernatant was separated and stored frozen (−80 ◦ C) until analysis. This method was also employed for the assessment of encapsulation efficiency in lipid nanoparticles. Compritol® 888 ATO is the lipid employed for the formulation of nanoparticles. The method for
262
Annex II. Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems
4039
A. Estella-Hermoso de Mendoza et al. / J. Chromatogr. B 877 (2009) 4035–4041 Table 1 Standard calibration curves of edelfosine in plasma, tissue and particulate homogenate samples calculated by the HPLC method. Range (!g/mL) 0.1–1 Plasma 1–75 Kidney Liver Lung Heart Spleen Brain Stomach Intestine Particulate homogenate
0.2–31.75 0.2–31.75 0.2–31.75 0.2–31.75 0.2–31.75 0.2–31.75 0.2–31.75 0.2–31.75 1–75
Regression equation
r
y = 0.1342x + 0.0001 y = 0.1013x + 0.0051 y = 0.1159x + 0.0018 y = 0.0221x − 0.0471 y = 0.0345x + 0.0972 y = 0.0267x − 0.0474
0.999 1.000 0.999 0.999 0.999 0.999
y = 0.1387x − 0.0149 y = 0.1327x − 0.0059 y = 0.1410x − 0.0603 y = 0.1862x − 0.0038 y = 0.1721x − 0.0662 y = 0.1519x − 0.0715 y = 0.1251x + 0.0069 y = 0.1244x + 0.0008 y = 0.1276x + 0.0096
0.999 0.999 0.999 1.000 1.000 0.999 0.999 1.000 0.999
r: correlation coefficient.
lipid nanoparticle formulation is the emulsion formation/solvent evaporation method. Briefly, lipid and edelfosine were dissolved in chloroform and homogenized by ultrasonication with an aqueous solution of 2% Tween® 80. The obtained emulsion was then subjected to mechanical stirring for the organic solvent evaporation and consequent lipid nanoparticle formation. Particles were then centrifuged at 4500 × g for 10 min using an Amicon Ultra-15 filter device and washed twice with distilled water. The obtained particular suspension was fast frozen at −80 ◦ C for at least 3 h and freeze-dried in order to store it at 4 ◦ C [24]. 3. Results and discussion A sensitive method for edelfosine detection in plasma and tissue samples was needed for their concentration-time course measurements during dose escalation and other pharmacokinetic preclinical studies. To achieve this aim it is critical to optimize the chromatographic conditions to obtain symmetrical peak shapes and a short chromatographic analysis time with high sensitivity and selectivity. Previously, we developed a HPLC–MS method for the quality control of edelfosine drug delivery systems and the edelfosine quantitation in cell internalization studies. Under these chromatographic conditions edelfosine is eluted as tailing and band-broadening chromatographic peaks (values of USP asymmetry factor between 1.1 and 1.3), with insufficient chromatographic efficiency to measure this drug at low concentrations [14]. Chemically, edelfosine is an anionic amphiphilic compound that is positively charged at acidic pH, which is the pH of the mobile phase. The observed peak tailing may be a result of an ionic interaction of residual package silanols and positively charged nitrogen of edelfosine. One of the first efforts was the attempt to reduce the value of the limit of quantitation while maintaining the employed extraction procedure. In our previous work, edelfosine determination was performed with polymerically bounded C18 reversed-phase narrow-bore column packed with double encapped spherical silica particles (Alltima® ). To overcome the technical difficulties observed, a substantial improvement of the chromatographic performance of the method is the only valid solution. For the HPLC method, we have used a Gemini® column instead of the Alltima® package. Gemini® column has been developed with a technology that grafts additional silica–organic layers onto the surface of the internal base silica. These additional layers protect the particle from ionic interactions with ionized compounds, such as edelfosine at acidic pH, and as a result, sharp, symmetrically chromatographic peaks are obtained.
Table 2 Standard calibration curves of edelfosine in plasma, tissue and particulate homogenate samples calculated by the UHPLC method. Range (!g/mL)
Regression equation
R
Plasma
0.0075–75
y = 0.03831x − 0.0004 y = 0.04029x + 0.0005 y = 0.03551x − 0.0007
0.999 0.999 0.999
Kidney Liver Lung Heart Spleen Brain Stomach Intestine Particulate homogenate
0.0075–75 0.0075–75 0.0075–75 0.0075–75 0.0075–75 0.0075–75 0.0075–75 0.0075–75 0.0075–75
y = 0.0321x + 0.0004 y = 0.0413x + 0.0002 y = 0.0429x − 0.0005 y = 0.0526x + 0.0001 y = 0.0520x − 0.0003 y = 0.0397x + 0.00007 y = 0.0762x + 0.0003 y = 0.0447x + 0.0005 y = 0.0313x + 0.0009
0.999 0.999 0.999 0.999 0.999 1.000 0.999 0.999 0.999
r: correlation coefficient.
On the other hand, and according to the van Deemter equation, one helpful way to improve the efficiency and analysis time of the HPLC column is to decrease the particle size. However, the improved column efficiency gained from using small particles comes along with a tremendous increase in the column pressure, which is prohibitive for traditional HPLC hardware, but not for UHPLC hardware, which can easily resist pressure values up to 15,000 psi [15,19]. For an ideal comparison, we would prefer to use columns with identical chemistry. However, at the time of the study, columns packed with Gemini® C18 particles < 2 !m were unavailable. Thus, Acquity BEH UPLCTM C18 columns packed with bridged ethylsiloxane-silica particles were employed. Fig. 2 shows the typical chromatograms of (A) a blank and (B) a spiked plasma sample with edelfosine and (C) the internal standard analyzed by the HPLC technique and (D) a blank and (E) a spiked plasma sample with edelfosine and (F) the internal standard analyzed by the UHPLC technique. The typical retention time for edelfosine and internal standard was 3.65 and 3.50 min for the HPLC–MS method, and 1.23 and 1.19 min for edelfosine and internal standard, respectively, for the UHPLC–MS/MS method. Compared with HPLC, UHPLC reduced the retention times threefold on average. The resolution between the chromatographic peaks of the edelfosine and internal standard was 0.08 and 0.09 for the HPLC–MS and the UPLC–MS/MS methods, respectively. Therefore, we can affirm that there is little difference in column selectivity between the two types of package employed. The HPLC asymmetry factor for edelfosine was 2, whereas it was 1.2 for the UHPLC method. The main improvement in the chromatographic behavior has been reflected in method sensibility. In fact, a sixfold increase in the LOQ value was observed for the UHPLC–MS/MS method compared to that of the HPLC–MS method, which made edelfosine quantitation possible in small samples, such as that obtained in pharmacokinetic studies in mice, after intravenous administration of sub-milligram doses. Validation data for edelfosine quantitation by the HPLC and UHPLC methods are compared in Tables 1–3. Assay performance of the present methods was assessed by all the following criteria: selectivity, linearity, accuracy, precision, LOD, LOQ, applicability to quantitation of edelfosine in different matrices and quantitation of edelfosine in lipid nanoparticles. Selectivity was assessed by the comparison of the chromatograms of six different batches of blank mouse plasma with the corresponding spiked plasma. There was no relevant interference from endogenous substances observed at the retention times of the analytes, as can be seen in Fig. 2. The HPLC–MS assay exhibited linearity divided into two intervals between the response (y) and the corresponding concentration of edelfosine (x) from 0.1 to 1 !g/mL in the small concentration interval and from 1 to 75 !g/mL in the high concentration
263
Annex II. Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems
4040
A. Estella-Hermoso de Mendoza et al. / J. Chromatogr. B 877 (2009) 4035–4041
Table 3 Accuracy, precision and between- and within-day measured concentrations for analysis of edelfosine by the HPLC and UHPLC methods. Conc. (!g/mL)
Accuracy (bias %)
Precision (%RSD) Between-day
Measured conc. (!g/mL, mean ± SD) Within-day
Between-day
Within-day
HPLC method QC1 QC2 QC3 QC4
0.1 5 10 50
4.88 7.13 −5.83 −0.62
7.77 7.11 5.57 2.04
5.42 4.23 1.66 3.22
0.10 5.22 9.50 50.16
± ± ± ±
0.01 0.37 0.53 1.02
0.09 5.49 9.28 48.90
± ± ± ±
0.01 0.23 0.15 1.58
UHPLC method QC1 QC2 QC3 QC4
0.075 0.75 15 60
−6.84 −4.06 6.49 2.73
10.91 7.20 7.70 4.59
12.23 8.33 6.60 3.72
0.069 0.70 15.80 62.06
± ± ± ±
0.007 0.05 1.19 2.84
0.070 0.73 15.97 61.29
± ± ± ±
0.008 0.06 1.03 2.28
interval for plasma samples. Tissue and nanoparticulate system samples presented linearity from 0.2 to 31.75 !g/mL and from 1 to 75 !g/mL, respectively. The UHPLC–MS/MS method showed a much higher sensitivity and calibration range than the obtained with the HPLC method. A linear range was achieved for all types of samples from 0.0075 to 75 !g/mL and the limit of quantitation was 0.0075 !g/mL. Results are shown in Tables 1 and 2. For each point of calibration standards, the concentrations were back-calculated from the equation of the regression curves and relative standard deviations (%RSD) were measured. %RSD did not exceed 15% in any case. For all calibration curves, linear regression provided r values greater than 0.999. As can be seen in Tables 1 and 2, the slope of the calibration curves from each tissue sample was similar to that obtained after the chromatographic analysis of plasma calibrator samples in a similar concentration range. It is clear that the developed method is adequate to quantify edelfosine in different biological samples, where an adequate extraction procedure has been applied. The lower limit of quantitation for edelfosine with the HPLC method was 0.1 !g/mL (S/N ≥ 5) with 5 !L injected into the chromatographic column with accuracy within ±20% and %RSD lower than 20%. The UHPLC method showed a limit of quantitation of 0.0075 !g/mL with 2 !L injected into the chromatographic column, presenting similar accuracy values. Compared with the HPLC method, the present UHPLC method gave sixfold higher sensitivity. The high sensitivity of the UHPLC method could be attributed to the peak sharpness produced by the column package and the lower analyte dilution in the column. To determine recovery, concentrations of edelfosine and PAF in extracted plasma and tissue QC samples were compared to standards prepared in blank matrix extract. The recovery was eval-
Fig. 3. Time-concentration curve data of edelfosine over 24 h after single dose intravenous administration of 200 !g (10 mg/kg) to BALB/c mice (n = 6, Mean ± S.D.) calculated by the HPLC method.
uated in triplicate and presented acceptable values ranging from 85.11% and 100.93%, and from 83.64% to 107.48% for edelfosine and internal standard, respectively, by the HPLC method, and from 95.25% to 99.35% and from 93.43% to 93.79% for edelfosine and PAF, respectively, by the UHPLC method. Matrix ionization suppression is considered to be a problem when using the protein precipitation method for sample preparation. Nevertheless, this method has been chosen as the sample preparation procedure in our work due to its simplicity and the lack of impact on the accuracy of the assay. The matrix effect values of edelfosine and the I.S. in both plasma and tissue homogenate samples were 15.55 ± 1.79% and 9.03 ± 3.96% for the HPLC method, and 6.17 ± 2.77% and 3.71% for the UHPLC method. It is interesting to note that the matrix effect is slightly reduced when the UHPLC analysis is performed. This may be due to the fact that the signal suppression is diminished drastically in the UHPLC–MS/MS method compared with the HPLC–MS method. Similar data were obtained by the HPLC and UHPLC methods after accuracy and precision evaluation (Table 3). Accuracy values were within acceptable limits ranging between −6.84% and 7.13%. The results for intra-batch and inter-batch precision for the samples ranged between 1.66% and 12.23%. The precision and accuracy of the present method is in accordance with the criteria for the analysis of biological samples according to the guidance of the FDA, where the precision (expressed as %RSD) determined at each concentration level is required not to exceed 15%. The applicability of this method has been demonstrated in vivo by the determination of edelfosine in plasma samples from BALB/c mice treated with 200 !g of edelfosine. Fig. 3 depicts the concentration of edelfosine in mouse plasma plotted against time after a single-dose intravenous administration of 200 !g of edelfosine (10 mg/kg) to BALB/c mice determined by the HPLC method. Edelfosine in blood plasma showed a Cmax of 50.7 ± 28.1 !g/mL and a Cmin of 2.5 ± 1.3 !g/mL, 24 h after intravenous administration. Tissue levels of edelfosine were also measured and compared to those
Fig. 4. Tissue distribution of edelfosine 24 h after a single intravenous administration of 200 !g to mice (10 mg/kg) calculated by the UHPLC method.
264
Annex II. Comparative study of a HPLC–MS assay versus an UHPLC–MS/MS for anti-‐tumoral alkyl lysophospholipid edelfosine determination in both biological samples and in lipid nanoparticulate systems
A. Estella-Hermoso de Mendoza et al. / J. Chromatogr. B 877 (2009) 4035–4041
of plasma. Fig. 4 depicts the distribution of edelfosine to different organs in mice after intravenous administration of 200 !g of edelfosine determined by the UHPLC method. No drug was detectable in the heart and very little was found in the brain. The edelfosine levels present in the kidney and intestine were statistically very significant compared to the plasma levels of edelfosine (p < 0.01). Edelfosine levels in the lung and heart were statistically different from plasma levels (p < 0.05). No significant differences were found between plasma and liver, spleen, brain and stomach levels (p > 0.05). The drug was also extracted from previously formulated lipid nanoparticles and quantified. Nanoparticulate systems made of lipid material Compritol® 888 ATO showed an encapsulation efficiency of about 95%. The same samples were analyzed with both the HPLC–MS and UHPLC–MS/MS method to ensure the interchangeability of both methods. The edelfosine results obtained from liver samples analyzed by the UHPLC–MS/MS method were highly correlated (r2 = 0.91; y = 0.9706x + 0.2123) with those from the HPLC–MS method. A good relationship between both techniques was found over the concentration range of 0.2–31.75 !g/mL for the HPLC–MS and UHPLC–MS/MS. 4. Conclusion Two liquid chromatographic methods, an HPLC–MS method and a UHPLC–MS/MS method for the bio-analysis of edelfosine were developed and evaluated. The UHPLC–MS/MS method developed in this work was more sensitive for the quantitation of edelfosine in plasma, tissue and lipid nanoparticles than the HPLC–MS method. Under the UPLC conditions we are able to achieve a shorter chromatographic run time while still avoiding a matrix ion suppression problem. UHPLC and HPLC are valuable methods for the determination of the pharmacokinetic behavior of edelfosine and bio-distribution in mice after intravenous administration of a dose of 10 mg/kg of edelfosine and the quality control of lipid nanoparticulate systems. Acknowledgements The authors would like to thank Caja Navarra Foundation and the Spanish Ministry of Science and Innovation (SAF2007-61261,
4041
PCT-090100-2007-27, SAF2008-02251 and RD06/0020/1037) for their support. Ander Estella-Hermoso de Mendoza is supported by the research grant from the Department of Education of the Basque Government (BFI06.37). References [1] C. Gajate, F. Mollinedo, Curr. Drug. Metab. 3 (2002) 491. [2] F. Mollinedo, C. Gajate, S. Martin-Santamaria, F. Gago, Curr. Med. Chem. 11 (2004) 3163. [3] W.J. Houlihan, M. Lohmeyer, P. Workman, S.H. Cheon, Med. Res. Rev. 15 (1995) 157. [4] P.G. Munder, O. Westphal, Chem. Immunol. 49 (1990) 206. [5] F. Mollinedo, J.L. Fernandez-Luna, C. Gajate, B. Martin-Martin, A. Benito, R. Martinez-Dalmau, M. Modolell, Cancer Res. 57 (1997) 1320. [6] G.A. Ruiter, M. Verheij, S.F. Zerp, W.J. van Blitterswijk, Int. J. Radiat. Oncol. Biol. Phys. 49 (2001) 415. [7] C. Gajate, F. Mollinedo, Blood 109 (2007) 711. [8] W.R. Vogler, W.E. Berdel, R.B. Geller, J.A. Brochstein, R.A. Beveridge, W.S. Dalton, K.B. Miller, H.M. Lazarus, Adv. Exp. Med. Biol. 416 (1996) 389. [9] R. Bhamra, L.E. Bolcsak, I. Ahmad, J. Schupsky, P. Roberts, R. Stevens, C. Cavanaugh, C.E. Swenson, Anticancer Drugs 14 (2003) 481. [10] J. Kotting, N.W. Marschner, W. Neumuller, C. Unger, H. Eibl, Prog. Exp. Tumor Res. 34 (1992) 131. [11] J. Coene, M. Ghis, P. Van den Eeckhout, P. Van den Bossche, J. Sandra, J. Chromatogr. 553 (1991) 285. [12] J. Coene, E. Van den Eeckhout, P. Herdewijn, P. Sandra, J. Chromatogr. 612 (1993) 21. [13] W.F. Steelant, E.A. Bruyneel, M.M. Mareel, E.G. Van den Eeckhout, Anal. Biochem. 227 (1995) 246. [14] M.J. Blanco-Prieto, M.A. Campanero, F. Mollinedo, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 810 (2004) 85. [15] M.I. Churchwell, N.C. Twaddle, L.R. Meeker, D.R. Doerge, J. Chromatogr. B: Analyt. Technol. Biomed. Life Sci. 825 (2005) 134. [16] I.S. Lurie, J. Chromatogr. A 1100 (2005) 168. [17] I. Citova, L. Havlikova, L. Urbanek, D. Solichova, L. Novakova, P. Solich, Anal. Bioanal. Chem. 388 (2007) 675. [18] D. Guillarme, D.T. Nguyen, S. Rudaz, J.L. Veuthey, Eur. J. Pharm. Biopharm. 66 (2007) 475. [19] Y. Hsieh, C.J. Duncan, S. Lee, M. Liu, J. Pharm. Biomed. Anal. 44 (2007) 492. [20] J. Olsovska, M. Jelinkova, P. Man, M. Koberska, J. Janata, M. Flieger, J. Chromatogr. A 1139 (2007) 214. [21] N. Stephanson, A. Helander, O. Beck, J. Mass Spectrom. 42 (2007) 940. [22] J.C. Van De Steene, W.E. Lambert, J. Am. Soc. Mass Spectrom. 19 (2008) 713. [23] K. Yu, D. Little, R. Plumb, B. Smith, Rapid Commun. Mass Spectrom. 20 (2006) 544. [24] A. Estella-Hermoso de Mendoza, M. Rayo, F. Mollinedo, M.J. Blanco-Prieto, Eur. J. Pharm. Biopharm. 68 (2008) 207.
. 265
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Annex III In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts Faustino Mollinedo1, Janis de la Iglesia-‐Vicente1, Consuelo Gajate1,2, Ander Estella-‐ Hermoso de Mendoza3, Janny A. Villa-‐Pulgarin1, Mercè de Frias4, Gaël Roué5, Joan Gil4, Dolors Colomer5, Miguel A. Campanero6, and Maria J. Blanco-‐Prieto3
1Instituto de Biología Molecular y Celular del Cáncer, Centro de Investigación del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain; 2Unidad de Investigación, Hospital Universitario de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain; 3Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia, Universidad de Navarra, E-‐31008 Pamplona, Spain; 4Departament de Ciències Fisiològiques II, Institut d'Investigació Biomèdica de Bellvitge (IDIBELL)–Universitat de Barcelona, E-‐ 08907 L'Hospitalet de Llobregat, Spain ; 5Unitat d’Hematopatologia, Hospital Clínic, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS)-‐Universitat de Barcelona, E-‐08036 Barcelona, Spain; 6Servicio de Farmacología Clínica, Clínica Universitaria, E-‐31080 Pamplona, Spain;.
Clinical Cancer Research, 2010. 16(7): p. 2046-‐2054
267
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Clinical Cancer Research
Cancer Therapy: Preclinical
In vitro and In vivo Selective Antitumor Activity of Edelfosine against Mantle Cell Lymphoma and Chronic Lymphocytic Leukemia Involving Lipid Rafts Faustino Mollinedo1, Janis de la Iglesia-Vicente1, Consuelo Gajate1,2, Ander Estella-Hermoso de Mendoza3, Janny A. Villa-Pulgarin1, Mercè de Frias5, Gaël Roué6, Joan Gil5, Dolors Colomer6, Miguel A. Campanero4, and Maria J. Blanco-Prieto3
Abstract Purpose: Mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL) remain B-cell malignancies with limited therapeutic options. The present study investigates the in vitro and in vivo effect of the phospholipid ether edelfosine (1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine) in MCL and CLL. Experimental Design: Several cell lines, patient-derived tumor cells, and xenografts in severe combined immunodeficient mice were used to examine the anti-MCL and anti-CLL activity of edelfosine. Furthermore, we analyzed the mechanism of action and drug biodistribution of edelfosine in MCL and CLL tumor-bearing severe combined immunodeficient mice. Results: Here, we have found that the phospholipid ether edelfosine was the most potent alkyllysophospholipid analogue in killing MCL and CLL cells, including patient-derived primary cells, while sparing normal resting lymphocytes. Alkyl-lysophospholipid analogues ranked edelfosine > perifosine ≫ erucylphosphocholine ≥ miltefosine in their capacity to elicit apoptosis in MCL and CLL cells. Edelfosine induced coclustering of Fas/CD95 death receptor and rafts in MCL and CLL cells. Edelfosine was taken up by malignant cells, whereas normal resting lymphocytes hardly incorporated the drug. Raft disruption by cholesterol depletion inhibited drug uptake, Fas/CD95 clustering, and edelfosine-induced apoptosis. Edelfosine oral administration showed a potent in vivo anticancer activity in MCL and CLL xenograft mouse models, and the drug accumulated dramatically and preferentially in the tumor. Conclusions: Our data indicate that edelfosine accumulates and kills MCL and CLL cells in a rather selective way, and set coclustering of Fas/CD95 and lipid rafts as a new framework in MCL and CLL therapy. Our data support a selective antitumor action of edelfosine. Clin Cancer Res; 16(7); 2046–54. ©2010 AACR.
Authors' Affiliations: 1Instituto de Biología Molecular y Celular del Cáncer, Centro de Investigación del Cáncer, Consejo Superior de Investigaciones Cientificas–Universidad de Salamanca, Campus Miguel de Unamuno; 2Unidad de Investigación, Hospital Universitario de Salamanca, Campus Miguel de Unamuno, Salamanca, Spain; 3Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia, Universidad de Navarra; 4Servicio de Farmacología Clínica, Clínica Universitaria, Pamplona, Spain; 5 Departament de Ciències Fisiològiques II, Institut d'Investigació Biomèdica de Bellvitge–Universitat de Barcelona, L'Hospitalet de Llobregat, Spain; and 6Unitat d'Hematopatologia, Hospital Clínic, Institut d'Investigacions Biomèdiques August Pi i Sunyer–Universitat de Barcelona, Barcelona, Spain Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). Corresponding Author: Faustino Mollinedo, Instituto de Biología Molecular y Celular del Cáncer, Centro de Investigación del Cáncer, Consejo Superior de Investigaciones Cientificas–Universidad de Salamanca, Campus Miguel de Unamuno, E-37007 Salamanca, Spain. Phone: 34-923-294806; Fax: 34-923-294795; E-mail:
[email protected] or Consuelo Gajate, Unidad de Investigación, Hospital Universitario de Salamanca, Campus Miguel de Unamuno, E-37007 Salamanca, Spain. E-mail:
[email protected]. doi: 10.1158/1078-0432.CCR-09-2456 ©2010 American Association for Cancer Research.
2046
Chronic lymphocytic leukemia (CLL) and mantle cell lymphoma (MCL) are two major B-cell–derived neoplasias for which current therapy is not satisfactory, leading in most cases to relapse and eventually to a fatal outcome. This lack of efficient therapy underscores the need for a continued search for novel chemotherapeutic agents. CLL is the most common adult leukemia and is characterized by the progressive accumulation of mature CD5+ B lymphocytes in the peripheral blood, bone marrow, and secondary lymphoid organs. New treatment combinations have incorporated the use of purine analogue (fludarabine)–based regimens together with monoclonal antibodies rituximab (anti-CD20) and alemtuzumab (anti-CD52), leading to improved complete response rates and prolonged progression-free survival, but a long-term survival benefit has not been shown (1, 2). MCL is characterized by the chromosomal translocation t(11;14)(q13;q32), resulting in the overexpression of cyclin D1 in mature B
Clin Cancer Res; 16(7) April 1, 2010
269
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Antitumor Activity of Edelfosine against MCL and CLL
Translational Relevance Mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL) lack effective therapy. Synthetic alkyl-lysophospholipid analogues constitute a family of promising anticancer drugs, including miltefosine, perifosine, edelfosine, and erucylphosphocholine, which promote apoptosis in a variety of tumor cells. Here, we have found that edelfosine behaves as the most potent alkyl-lysophospholipid analogue in inducing cell death in MCL and CLL cells through coclustering of Fas/CD95 and rafts. Edelfosine induced a higher apoptotic response than perifosine in MCL and CLL patient-derived cells. Oral administration of edelfosine showed a strong in vivo anti-MCL and anti-CLL activity in xenograft mouse models. The drug accumulated in a dramatic and preferential way in the tumor, leading to drastic tumor regression. Our data reported here show a rather selective action of edelfosine against tumor cells and provide the proof of principle and rationale for further clinical evaluation of edelfosine to improve patient outcome in MCL and CLL.
lymphocytes that have a striking tendency to disseminate throughout the body (3). MCL is an aggressive lymphoma with a poor survival outcome and a median survival time of 3.5 years. Conventional chemotherapeutic regimens have been the standard treatment of MCL until the recent incorporation of rituximab, which increases overall survival as well as the response rate and duration. The introduction of stem cell transplantation improves survival, although this therapeutic modality is only applied to younger and fit patients (4). Currently, allogeneic bone marrow transplantation represents the only therapy with the potential for a curative approach, although it is associated with a high rate of complications (4). Therefore, development of novel therapeutic strategies is urgently needed to improve survival in patients with the above B-cell malignancies. In the last years, new strategies have been developed that target crucial cellular pathways, and proteasome inhibition with bortezomib has recently been approved in relapsed/ refractory MCL (5). MCL and CLL cells share a relatively low proliferative index and a poor apoptotic rate (6, 7), and therefore, the transforming event is likely a failure in death regulation rather than a loss of growth control. This implies that a therapeutic potential for these diseases may lie in potentiating apoptosis. Synthetic alkyl-lysophospholipid (ALP) analogues constitute a family of promising anticancer drugs, including edelfosine, miltefosine, perifosine, and erucylphosphocholine, which act at the cell membrane level (8–12). The phospholipid ether edelfosine (1-O-octadecyl-2-Omethyl-rac-glycero-3-phosphocholine, ET-18-OCH3), considered the ALP prototype, induces selectively apoptosis in tumor cells (13) by recruiting Fas/CD95 death receptor in
www.aacrjournals.org
membrane raft clusters (13–18). Edelfosine was the first antitumor agent shown to act through lipid rafts (15), and these membrane domains have been recently involved in the chemotherapeutic action of some additional antitumor agents (17, 19–23). Lipid rafts are membrane microdomains highly enriched in cholesterol and sphingolipids, and the proteins located in these microdomains are limited in their ability to freely diffuse over the plasma membrane, affecting protein function (24). Recent findings suggest that lipid rafts act as scaffolds, where Fas/ CD95 and downstream signaling molecules are recruited to trigger apoptosis (16–18, 21, 25). On these grounds, we hypothesized that MCL and CLL, showing a defective cell death control, could be considered stark examples for the particular proapoptotic features of ALPs. Thus, we addressed the possibility that these B-cell malignancies could be treated by these agents, and examined the putative involvement of lipid rafts in their antitumor action against MCL and CLL.
Materials and Methods Drugs. Edelfosine was from Inkeysa and Apointech. Miltefosine was from Calbiochem. Perifosine and erucylphosphocholine were from Zentaris. Cell lines and primary cells. Detailed information on the cell culture conditions for human MCL (JVM-2 and Z-138) and CLL (EHEB) cell lines and primary cells from CLL and MCL patients is included in Supplementary Data. Apoptosis assay. Quantitation of apoptotic cells was determined by flow cytometry as the percentage of cells in the sub-G1 region (hypodiploidy) in cell cycle analysis as previously described (26). To analyze apoptosis in CLL/MCL patient-derived samples, 5 × 105 cells were incubated for 48 h with the indicated agents. Cells were then washed in Annexin-binding buffer and incubated in 50 μL Annexin-binding buffer with allophycocyanin-conjugated anti-CD3 and phycoerythrinconjugated anti-CD19 antibodies from Becton Dickinson for 10 min in the dark. Cells were then diluted with Annexin-binding buffer to a volume of 150 μL and incubated with 1 μL FITC-labeled Annexin V (Bender MedSystems) for 15 min in the dark. A total of 10,000 stained cells were then analyzed by flow cytometry on a FACSCalibur flow cytometer using CellQuest software (Becton Dickinson). Western blot. Proteins (50 μg) were separated on 12% SDS-PAGE and transferred onto Immobilon-P membranes (Millipore). Membranes were probed with anti– cyclin D1 (DCS-6, Cell Signaling Technology) and anti–β-actin (Sigma) antibodies. Antibody binding was detected using the enhanced chemiluminescence detection system (Amersham). Confocal microscopy. Cells were settled onto poly- Llysine–coated slides and analyzed with a Zeiss LSM 510 laser scan confocal microscope for membrane raft and Fas/CD95 visualization using FITC-labeled cholera toxin B subunit (Sigma) and anti-human Fas/CD95 SM1/1 IgG2a mouse monoclonal antibody (Bender MedSystems)
Clin Cancer Res; 16(7) April 1, 2010
2047
270
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Mollinedo et al.
Fig. 1. Induction of apoptosis in MCL and CLL cells by ALPs. MCL (JVM-2 and Z-138) and CLL (EHEB) cell lines were incubated for the indicated times with 10 μmol/L of the distinct ALPs edelfosine, perifosine, miltefosine, and erucylphosphocholine (Erucyl-PC; A) or for 24 h with different concentrations of the ALPs (B). Apoptosis was then quantitated as percentage of cells in the sub-G1 region by flow cytometry. Untreated control cells were run in parallel. Data are means ± SE of four independent determinations. C, cells were untreated or treated with 10 μmol/L perifosine or edelfosine for the indicated times and analyzed by Western blot with anti–cyclin D1 and anti–β-actin antibodies. Immunoblotting of β-actin was used as an internal control for equal protein loading in each lane. Blots are representative of three experiments done.
followed by CY3-conjugated anti-mouse antibody (Pharmacia), as described (15). Colocalization assays were analyzed by excitation of both fluorochromes in the same section. Negative controls, lacking the primary antibody or using an irrelevant antibody, showed no staining. Edelfosine uptake. Drug uptake was measured as described previously (13) after incubating 106 cells with 10 nmol [3H]edelfosine for 2 h in RPMI 1640/10% fetal bovine serum and subsequent exhaustive washing (six times) with PBS + 2% bovine serum albumin to eliminate the loosely cell surface–bound ether lipid. [3H]Edelfosine (specific activity, 42 Ci/mmol) was synthesized by
2048
Clin Cancer Res; 16(7) April 1, 2010
tritiation of the 9-octadecenyl derivative (Amersham Buchler). Cholesterol depletion. Cells (2.5 × 105/mL) were pretreated with 2.5 mg/mL methyl-β-cyclodextrin (MCD) for 30 min at 37°C in serum-free medium. Cells were then washed thrice and resuspended in complete culture medium before edelfosine addition. Xenograft mouse model. CB17–severe combined immunodeficient (SCID) mice (Charles River Laboratories), kept and handled according to institutional guidelines, complying with Spanish legislation under a 12/12-h light/dark cycle at a temperature of 22°C, received a standard diet
Clinical Cancer Research
271
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Antitumor Activity of Edelfosine against MCL and CLL
and acidified water ad libitum. CB17-SCID mice were inoculated s.c. into their lower dorsum with 107 Z-138 or EHEB cells in 100 μL PBS and 100 μL Matrigel basement membrane matrix (Becton Dickinson). When tumors were palpable, mice were randomly assigned into cohorts of 8 to 10 mice each, receiving a daily oral administration of edelfosine (30 mg/kg) or an equal volume of vehicle (PBS). The shortest and longest diameter of the tumor were measured with calipers at 2- to 5-d intervals, and tumor volume (mm3) was calculated using the following standard formula: (the shortest diameter)2 × (the longest diameter) × 0.5. Animals were sacrificed, according to institutional guidelines, when the diameter of their tumors reached 3 cm or when significant toxicity was observed. Animal body weight and any sign of morbidity were monitored. Drug treatment lasted for 21 d (MCL) and 34 d (CLL), and mice were killed 24 h after the last drug administration. Then, tumor xenografts were extirpated, measured, and weighed, and a necropsy analysis involving tumors and distinct organs was carried out. Plasma/tissue extraction procedure for edelfosine biodistribution studies. MCL- or CLL-bearing SCID mice were treated with a daily oral administration of edelfosine (30 mg/kg) for 21 d (MCL) or 34 d (CLL). Twenty-four hours after the last drug oral administration, blood was collected in EDTA surface-coated tubes and then centrifuged at 2,000 × g for 15 min (4°C) to collect plasma (100 μL). Then, animals were sacrificed, and distinct organs and tumors were collected and weighed. Tissues and tumors were homogenized in 1-mL PBS (pH 7.4) using a Mini-bead Beater (BioSpect Products, Inc.) and centrifuged at 10,000 × g for 10 min. Both plasma and tissue supernatants
Table 1. Selective killing of patient-derived CLL cells by edelfosine, sparing normal cells Treatment
% Cell viability CD19+ CLL cells CD3+ T cells
Edelfosine (10 μmol/L) Edelfosine (20 μmol/L) Perifosine (10 μmol/L) Perifosine (20 μmol/L) Staurosporine (0.5 μmol/L)
60.3 43.7 73.1 61.0 17.8
± ± ± ± ±
5.6 4.9 4.8 3.7 6.7
97.8 86.5 98.4 90.9 22.5
± ± ± ± ±
1.9 3.2 1.5 2.8 7.5
NOTE: Primary lymphocyte cultures from CLL patients were incubated with edelfosine or perifosine for 48 h at the indicated concentrations. Staurosporine was used as a positive control of apoptosis. Percentage of cell viability was measured as nonapoptotic CD3+/CD19− T cells or CD3−/CD19+ B cells from CLL samples in Annexin V analysis by flow cytometry. Untreated control CD19+ and CD3+ cells were run in parallel and showed a cell viability of >93% and >98%, respectively. Data are shown as mean values ± SE of five independent CLL patients.
were collected and stored at −80°C until high-performance liquid chromatography–mass spectrometry (HPLC-MS) analysis was done. Ten micrograms of platelet-activating factor (1 mg/mL), used as internal standard, were added onto 100 μL of plasma or tissue/tumor supernatant. A mixture (190 μL) of 1% formic acid/methanol was added to precipitate proteins. Samples were vortexed for 1 min, and after centrifugation (20,000 × g, 10 min), 25 μL of the supernatant were analyzed by HPLC-MS. Quantitative determination of edelfosine by HPLC-MS analysis. The technique used was a slight modification of a previously described method (27) and is described in detail in Supplementary Data. Statistical analysis. All values are expressed as means ± SE. Between-group statistical differences were assessed using the Mann-Whitney test or the Student's t test. A P value of 85% in tumor weight and volume in both MCL and CLL animal models (Fig. 5B). Organ examination at necropsy did not reveal any apparent toxicity (data not shown), and there was an evident difference between the highly vascularized tumors from drug-free mice and the pale poorly vascularized tumors from edelfosine-treated mice (Fig. 5C). In addition, MCL tumors were bulky in drug-free mice but resulted rather flat after edelfosine treatment (Fig. 5C). No significant differences in mean body weight were observed between drug-treated and control animals during the in vivo assay (3-5% of body weight loss in the treated groups versus control groups). A drug biodistribution study showed that edelfosine dramatically accumulated in the MCL and CLL tumors (Fig. 5D). Tumor/plasma concentration ratio of edelfosine in the tumor was significantly higher than that detected in both kidney and liver after completion of the experiment in MCL and CLL animal models (Fig. 5D), with a drug mean concentration in plasma of 5.64 μg/ mL. In the CLL animal model, we examined the content of edelfosine in a wide variety of distinct organs and found that the drug was dramatically accumulated in the tumor compared with lung, heart, spleen, liver, intestine, or kidney (Fig. 5D). Taken together, our data indicate a preferential accumulation of edelfosine in the tumor.
Discussion The data reported here show that edelfosine behaves as the most potent ALP in killing MCL and CLL cells via a raft-mediated process. Our data indicate that edelfosine is a powerful antitumor agent against MCL and CLL as assessed by in vitro, ex vivo, and in vivo evidences. In addition, we found a rather selective and dramatic accumulation of edelfosine in MCL and CLL tumor cells in animal models. Here, we found that edelfosine induces the recruitment of Fas/CD95 death receptor in raft aggregates in MCL and CLL cells. Raft disruption by cholesterol depletion in MCL and CLL cells inhibited both edelfosine uptake and druginduced apoptosis, as well as Fas/CD95 clustering, thus suggesting a major role of rafts in the uptake and antitumor action of edelfosine. Previous reports have shown that MCL and CLL cells express Fas/CD95, but a deficient apoptotic response to the external stimulation of Fas/CD95 by agonistic anti-Fas/CD95 antibodies was reported (7, 31). Unlike the natural ligand FasL/CD95L or agonistic antiFas/CD95 antibodies that act through their interaction with
Clin Cancer Res; 16(7) April 1, 2010
2051
274
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Mollinedo et al.
the exogenous portion of the Fas/CD95 death receptor, edelfosine induces activation of Fas/CD95 from within the cell independently of its ligand (16, 32). We have previously found that edelfosine is even more efficient than FasL/CD95L in promoting programmed cell death through Fas/CD95 activation by its recruitment in membrane rafts enriched in downstream signaling molecules (14, 16–18, 25, 33). Thus, edelfosine might induce Fas/CD95 activation, although the receptor is not triggered by its natural ligand or agonistic antibodies. Using distinct MCL and CLL xenograft mouse models, we found that edelfosine accumulates in high amounts in the tumor tissue and shows a remarkable antitumor activity, leading to dramatic tumor regression. In addition, we consistently found in the MCL and CLL xenograft
animal models that tumors became smaller and poorly vascularized. This could be in agreement with reports showing an antiangiogenic effect of edelfosine (34, 35). Thus, further insight into the effect of edelfosine on angiogenesis and how this action affects cancer development is warranted. Following edelfosine oral administration in non–tumorbearing SCID mice, we have recently found a rather wide drug distribution pattern to several tissues, including lung, spleen, intestine, liver, and kidney (36). In this study, we also found that edelfosine showed a preferential accumulation in the tumor in a MCL-bearing mouse animal model (36). Now, we have largely extended this initial study and analyzed the in vivo effect of edelfosine in MCL and CLL animal models. Interestingly, we found here that when
Fig. 5. Edelfosine inhibits human MCL and CLL cell growth in vivo. CB17-SCID mice were inoculated s.c. with 107 Z-138 or EHEB cells. Daily oral administration of edelfosine (30 mg/kg, n = 10 for MCL animal model and n = 8 for CLL animal model) started after the development of a palpable tumor. Tumor size was recorded every 2 to 5 d. A, edelfosine significantly (P < 0.01; from day 15 of treatment until the end of the experiment) inhibited MCL and CLL tumor growth compared with the control group treated with vehicle (PBS, n = 10 for MCL animal model and n = 8 for CLL animal model). Data are means ± SE (n = 10, MCL; n = 8, CLL). B, after completion of the in vivo assay (22 d for MCL and 35 d for CLL), control and edelfosine-treated mice were sacrificed and tumors were measured in the distinct mice. The tumor size and weight values of each single animal (dots) and the average values of each experimental group (horizontal bars) are shown. Asterisks indicate that tumor weight and size values were significantly lower after edelfosine treatment, compared with control drug-free mice, at P < 0.01 (**). C, remarkable growth inhibition of MCL and CLL tumors was observed after edelfosine treatment (30 mg/kg). D, tissue/plasma concentration ratios, in liver, kidney, and tumor, of edelfosine after daily oral administration of edelfosine (30 mg/kg) for 3 wk in MCL-bearing SCID mice (mean ± SE, n = 5), and tissue/plasma concentration ratios, in lung, heart, spleen, liver, intestine, kidney, and tumor, of edelfosine after daily oral administration of edelfosine (30 mg/kg) for 34 d in CLL-bearing SCID mice (mean ± SE, n = 5). Asterisks indicate that the tumor/plasma ratio values are significantly different from the other tissue/plasma ratios at P < 0.01 (**).
2052
Clin Cancer Res; 16(7) April 1, 2010
Clinical Cancer Research
275
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Antitumor Activity of Edelfosine against MCL and CLL
SCID mice contained a MCL or CLL tumor, edelfosine distribution was dramatically and significantly shifted toward the tumor (tissue/plasma concentration ratios >16; P < 0.01), suggesting a preferential tumor location for edelfosine. Our herein reported in vivo data, together with our present and previous in vitro determinations in a wide number of malignant and normal cells (14, 16, 17), suggest a rather selective edelfosine uptake and cytotoxic action in tumor cells. The selective action of edelfosine on tumor cells supports its low toxicity. We did not find any apparent damage in the distinct organs analyzed following necropsy analysis in the in vivo studies reported here. Lack of toxicity of edelfosine in a rat model has been reported, including no significant cardiotoxicity, hepatotoxicity, or renal toxicity (37). Our biodistribution data in the murine models reported here showed a mean concentration of edelfosine in plasma of 5.64 μg/mL (10.77 μmol/L; edelfosine molecular mass, 523.7). Thus, the herein reported in vitro effects, rendered by 10 μmol/L edelfosine, were detected at a pharmacologically relevant drug concentration. Our data constitute the first in vitro and in vivo evidence for the antitumor action of edelfosine in MCL and CLL, two hematologic malignancies with poor survival outcome. Taken together, the results reported here provide the proof of principle and rationale for further clinical evaluation of edelfosine to improve patient outcome in MCL and CLL. The results reported here also highlight the involvement of lipid rafts in the action of edelfosine on B-cell malignancies, such as MCL and CLL.
Disclosure of Potential Conflicts of Interest No potential conflicts of interest were disclosed.
Acknowledgments We thank Alberto Gomez-Alonso, Javier Garcia-Criado, and Jose F. Martin-Martin for their help in the in vivo assays and for allowing us to use their animal facilities during part of this study.
Grant Support Ministerio de Ciencia e Innovación grants SAF2005-04293, SAF20068850, SAF2007-61261, SAF2007-60964, PCT-090100-2007-27, PS09/ 01915, and SAF2008-02251; Red Temática de Investigación Cooperativa en Cáncer grants RD06/0020/0014, RD06/0020/0097, and RD06/0020/ 1037 (Instituto de Salud Carlos III, cofunded by the Fondo Europeo de Desarrollo Regional of the European Union); Fondo de Investigación Sanitaria and European Commission grant FIS-FEDER 06/0813; Junta de Castilla y León (CSI01A08, GR15-Experimental Therapeutics and Translational Oncology Program, and Biomedicine Project 2009); Fundación de Investigación Médica Mutua Madrileña; Fundación “la Caixa” grant BM05-30-0; Caja Navarra Foundation; Department of Health of the Government of Navarra (“Ortiz de Landázuri, 2009” project); and AGAUR-Generalitat de Catalunya grant 2005SGR-00549. C. Gajate is supported by the Ramón y Cajal Program from the Ministerio de Ciencia e Innovación of Spain. A. Estella-Hermoso de Mendoza is supported by Department of Education of the Basque Government research grant BFI06.37. M. de Frias is a recipient of a fellowship from the AGAUR-Generalitat de Catalunya. G. Roué holds a Miguel Servet research contract from Instituto de Salud Carlos III. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Received 09/08/2009; revised 12/27/2009; accepted 12/30/2009; published OnlineFirst 03/16/2010.
References 1.
Christian BA, Lin TS. Antibody therapy for chronic lymphocytic leukemia. Semin Hematol 2008;45:95–103. 2. Montserrat E, Moreno C. Chronic lymphocytic leukaemia: a short overview. Ann Oncol 2008;19 Suppl 7:vii320–5. 3. Jares P, Colomer D, Campo E. Genetic and molecular pathogenesis of mantle cell lymphoma: perspectives for new targeted therapeutics. Nat Rev Cancer 2007;7:750–62. 4. Schmidt C, Dreyling M. Therapy of mantle cell lymphoma: current standards and future strategies. Hematol Oncol Clin North Am 2008;22:953–63. 5. Leonard JP, Williams ME, Goy A, et al. Mantle cell lymphoma: biological insights and treatment advances. Clin Lymphoma Myeloma 2009;9:267–77. 6. Rummel MJ, de Vos S, Hoelzer D, Koeffler HP, Hofmann WK. Altered apoptosis pathways in mantle cell lymphoma. Leuk Lymphoma 2004;45:49–54. 7. Proto-Siqueira R, Panepucci RA, Careta FP, et al. SAGE analysis demonstrates increased expression of TOSO contributing to Fasmediated resistance in CLL. Blood 2008;112:394–7. 8. Gajate C, Mollinedo F. Biological activities, mechanisms of action and biomedical prospect of the antitumor ether phospholipid ET18-OCH3 (edelfosine), a proapoptotic agent in tumor cells. Curr Drug Metab 2002;3:491–525. 9. Jendrossek V, Handrick R. Membrane targeted anticancer drugs: potent inducers of apoptosis and putative radiosensitisers. Curr Med Chem Anti-Canc Agents 2003;3:343–53. 10. Mollinedo F, Gajate C, Martin-Santamaria S, Gago F. ET-18-OCH3 (edelfosine): a selective antitumour lipid targeting apoptosis through intracellular activation of Fas/CD95 death receptor. Curr Med Chem 2004;11:3163–84. 11. Vink SR, van Blitterswijk WJ, Schellens JH, Verheij M. Rationale and
www.aacrjournals.org
12. 13.
14.
15.
16.
17.
18.
19.
20.
clinical application of alkylphospholipid analogues in combination with radiotherapy. Cancer Treat Rev 2007;33:191–202. Mollinedo F. Antitumor ether lipids: proapoptotic agents with multiple therapeutic indications. Expert Opin Ther Patents 2007;17:385–405. Mollinedo F, Fernandez-Luna JL, Gajate C, et al. Selective induction of apoptosis in cancer cells by the ether lipid ET-18-OCH3 (edelfosine): molecular structure requirements, cellular uptake, and protection by Bcl-2 and Bcl-XL. Cancer Res 1997;57:1320–8. Gajate C, Fonteriz RI, Cabaner C, et al. Intracellular triggering of Fas, independently of FasL, as a new mechanism of antitumor ether lipid-induced apoptosis. Int J Cancer 2000;85:674–82. Gajate C, Mollinedo F. The antitumor ether lipid ET-18-OCH3 induces apoptosis through translocation and capping of Fas/CD95 into membrane rafts in human leukemic cells. Blood 2001;98:3860–3. Gajate C, Del Canto-Janez E, Acuna AU, et al. Intracellular triggering of Fas aggregation and recruitment of apoptotic molecules into Fasenriched rafts in selective tumor cell apoptosis. J Exp Med 2004;200: 353–65. Gajate C, Mollinedo F. Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and downstream signaling molecules into lipid rafts. Blood 2007;109: 711–9. Gajate C, Gonzalez-Camacho F, Mollinedo F. Involvement of raft aggregates enriched in Fas/CD95 death-inducing signaling complex in the antileukemic action of edelfosine in Jurkat cells. PLoS One 2009;4:e5044. Lacour S, Hammann A, Grazide S, et al. Cisplatin-induced CD95 redistribution into membrane lipid rafts of HT29 human colon cancer cells. Cancer Res 2004;64:3593–8. Delmas D, Rebe C, Micheau O, et al. Redistribution of CD95, DR4 and DR5 in rafts accounts for the synergistic toxicity of resveratrol
Clin Cancer Res; 16(7) April 1, 2010
2053
276
Annex III. In vitro and in vivo selective antitumor activity of edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts
Mollinedo et al.
21.
22.
23.
24. 25.
26.
27.
28.
2054
and death receptor ligands in colon carcinoma cells. Oncogene 2004;23:8979–86. Gajate C, Mollinedo F. Cytoskeleton-mediated death receptor and ligand concentration in lipid rafts forms apoptosis-promoting clusters in cancer chemotherapy. J Biol Chem 2005;280:11641–7. Mollinedo F, Gajate C. Fas/CD95 death receptor and lipid rafts: new targets for apoptosis-directed cancer therapy. Drug Resist Updat 2006;9:51–73. Reis-Sobreiro M, Gajate C, Mollinedo F. Involvement of mitochondria and recruitment of Fas/CD95 signaling in lipid rafts in resveratrolmediated antimyeloma and antileukemia actions. Oncogene 2009; 28:3221–34. Simons K, Toomre D. Lipid rafts and signal transduction. Nat Rev Mol Cell Biol 2000;1:31–9. Gajate C, Gonzalez-Camacho F, Mollinedo F. Lipid raft connection between extrinsic and intrinsic apoptotic pathways. Biochem Biophys Res Commun 2009;380:780–4. Gajate C, Santos-Beneit AM, Macho A, et al. Involvement of mitochondria and caspase-3 in ET-18-OCH3-induced apoptosis of human leukemic cells. Int J Cancer 2000;86:208–18. Blanco-Prieto MJ, Campanero MA, Mollinedo F. Quantitative determination of the antitumor alkyl ether phospholipid edelfosine by reversed-phase liquid chromatography-electrospray mass spectrometry: application to cell uptake studies and characterization of drug delivery systems. J Chromatogr B Analyt Technol Biomed Life Sci 2004;810:85–92. Gajate C, Barasoain I, Andreu JM, Mollinedo F. Induction of apoptosis in leukemic cells by the reversible microtubule-disrupting agent 2-methoxy-5-(2′,3′,4′-trimethoxyphenyl)-2,4,6-cycloheptatrien-1one: protection by Bcl-2 and Bcl-XL and cell cycle arrest. Cancer Res 2000;60:2651–9.
Clin Cancer Res; 16(7) April 1, 2010
29. Schon A, Freire E. Thermodynamics of intersubunit interactions in cholera toxin upon binding to the oligosaccharide portion of its cell surface receptor, ganglioside GM1. Biochemistry 1989;28:5019–24. 30. Harder T, Scheiffele P, Verkade P, Simons K. Lipid domain structure of the plasma membrane revealed by patching of membrane components. J Cell Biol 1998;141:929–42. 31. Plumas J, Jacob MC, Chaperot L, Molens JP, Sotto JJ, Bensa JC. Tumor B cells from non-Hodgkin's lymphoma are resistant to CD95 (Fas/Apo-1)-mediated apoptosis. Blood 1998;91:2875–85. 32. Mollinedo F, Gajate C. FasL-independent activation of Fas. In: Wajant H, editor. Fas signaling, chapter 2. Georgetown (TX): Landes Bioscience and Springer Science; 2006, p. 13–27. 33. Mollinedo F. Death receptors in multiple myeloma and therapeutic opportunities. In: Lonial S, editor. Myeloma therapy. Pursuing the plasma cell, chapter 25. Totowa (NJ): Humana Press; 2008, p. 393–419. 34. Candal FJ, Bosse DC, Vogler WR, Ades EW. Inhibition of induced angiogenesis in a human microvascular endothelial cell line by ET18-OCH3. Cancer Chemother Pharmacol 1994;34:175–8. 35. Zerp SF, Vink SR, Ruiter GA, et al. Alkylphospholipids inhibit capillary-like endothelial tube formation in vitro: antiangiogenic properties of a new class of antitumor agents. Anticancer Drugs 2008;19:65–75. 36. de Mendoza AE, Campanero MA, de la Iglesia-Vicente J, Gajate C, Mollinedo F, Blanco-Prieto MJ. Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-bearing immunosuppressed mice. Clin Cancer Res 2009;15: 858–64. 37. Mollinedo F, Gajate C, Morales AI, et al. Novel anti-inflammatory action of edelfosine lacking toxicity with protective effect in experimental colitis. J Pharmacol Exp Ther 2009;329:439–49.
Clinical Cancer Research
277
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Annex IV Lipid raft-‐targeted therapy in multiple myeloma Faustino Mollinedo1, Janis de la Iglesia-‐Vicente1, Consuelo Gajate1,2, Ander Estella-‐ Hermoso de Mendoza3, Janny A. Villa-‐Pulgarin1, Miguel A. Campanero4, and Maria J. Blanco-‐Príeto3
1Centro de Investigación del Cáncer, Instituto de Biología Molecular y Celular del
Cáncer, CSIC-‐Universidad de Salamanca, Campus Miguel de Unamuno, E-‐37007 Salamanca, Spain 2Unidad de Investigación, Hospital Universitario de Salamanca, Campus Miguel de
Unamuno, E-‐37007 Salamanca, Spain
3Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia,
Universidad de Navarra, E-‐31008 Pamplona, Spain
4Servicio de Farmacología Clínica, Clínica Universitaria, E-‐31080 Pamplona, Spain
Oncogene, 2010. 29(26): p. 3748-‐57
279
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Oncogene (2010) 29, 3748–3757
& 2010 Macmillan Publishers Limited All rights reserved 0950-9232/10 www.nature.com/onc
ORIGINAL ARTICLE
Lipid raft-targeted therapy in multiple myeloma F Mollinedo1, J de la Iglesia-Vicente1, C Gajate1,2, A Estella-Hermoso de Mendoza3, JA Villa-Pulgarin1, MA Campanero4 and MJ Blanco-Prieto3 1
Centro de Investigacio´n del Ca´ncer, Instituto de Biologı´a Molecular y Celular del Ca´ncer, CSIC-Universidad de Salamanca, Campus Miguel de Unamuno, Salamanca, Spain; 2Unidad de Investigacio´n, Hospital Universitario de Salamanca, Campus Miguel de Unamuno, Salamanca, Spain; 3Departamento de Farmacia y Tecnologı´a Farmace´utica, Facultad de Farmacia, Universidad de Navarra, Pamplona, Spain and 4Servicio de Farmacologı´a Clı´nica, Clı´nica Universitaria, Pamplona, Spain
Despite recent advances in treatment, multiple myeloma (MM) remains an incurable malignancy. By using in vitro, ex vivo and in vivo approaches, we have identified here that lipid rafts constitute a new target in MM. We have found that the phospholipid ether edelfosine targets and accumulates in MM cell membrane rafts, inducing apoptosis through co-clustering of rafts and death receptors. Raft disruption by cholesterol depletion inhibited drug uptake by tumor cells as well as cell killing. Cholesterol replenishment restored MM cell ability to take up edelfosine and to undergo drug-induced apoptosis. Ceramide addition displaced cholesterol from rafts, and inhibited edelfosineinduced apoptosis. In an MM animal model, edelfosine oral administration showed a potent in vivo antimyeloma activity, and the drug accumulated preferentially and dramatically in the tumor. A decrease in tumor cell cholesterol, a major raft component, inhibited the in vivo antimyeloma action of edelfosine and reduced drug uptake by the tumor. The results reported here provide the proofof-principle and rationale for further clinical evaluation of edelfosine and for this raft-targeted therapy to improve patient outcome in MM. Our data reveal cholesterolcontaining lipid rafts as a novel and efficient therapeutic target in MM, opening a new avenue in cancer treatment. Oncogene (2010) 29, 3748–3757; doi:10.1038/onc.2010.131; published online 26 April 2010 Keywords: lipid raft; cholesterol; phospholipid ether; edelfosine; apoptosis; MM
Introduction Synthetic alkyl-lysophospholipid analogs (ALPs) constitute a promising family of anticancer drugs that act at Correspondence: Dr F Mollinedo, Centro de Investigacio´n del Ca´ncer, Instituto de Biologı´ a Molecular y Celular del Ca´ncer, CSICUniversidad de Salamanca, Campus Miguel de Unamuno, Salamanca E-37007, Spain. E-mail:
[email protected] or Dr C Gajate, Unidad de Investigacio´n, Hospital Universitario de Salamanca, Campus Miguel de Unamuno, Salamanca E-37007, Spain. E-mail:
[email protected] Received 23 October 2009; revised 12 January 2010; accepted 7 February 2010; published online 26 April 2010
the membrane level (Gajate and Mollinedo, 2002; Mollinedo et al., 2004; Mollinedo, 2007). The ALP prototype edelfosine (1-O-octadecyl-2-O-methyl-racglycero-3-phosphocholine, ET-18-OCH3) induces apoptosis in tumor cells (Mollinedo et al., 1997) through co-capping of the Fas/CD95 death receptor with membrane rafts (Gajate et al., 2000a, 2004; Gajate and Mollinedo, 2001, 2007). In vitro studies have shown that this phospholipid ether is preferentially taken up by tumor cells, leading to the intracellular activation of the Fas/CD95 death receptor, independently of its ligand FasL/CD95L (Mollinedo et al., 1997; Gajate et al., 2000a, 2004; Gajate and Mollinedo, 2001). Edelfosine-induced apoptosis is mediated by the recruitment of Fas/CD95 and downstream signaling molecules into membrane rafts (Gajate and Mollinedo, 2001, 2007; Gajate et al., 2004). Lipid rafts are membrane microdomains highly enriched in cholesterol and sphingolipids, and the proteins located in these microdomains are limited in their ability to freely diffuse over the plasma membrane, affecting protein function (Simons and Toomre, 2000). Recent evidence suggests that membrane rafts constitute the linchpin from which Fas/CD95 signaling is launched (Gajate et al., 2004, 2009a; Gajate and Mollinedo, 2005, 2007). Multiple myeloma (MM) is the second most prevalent blood cancer, accounting for 10% of all hematological malignancies (Hussein et al., 2002). MM is characterized by the accumulation of malignant plasma cells in the bone marrow with frequent occurrence of lytic bone disease. Although therapeutic advances in the treatment of MM have improved remission duration and overall survival, relapse of the disease is inevitable for nearly all patients, and MM remains a deadly B-cell malignancy (Mihelic et al., 2007). Therefore, development of novel therapeutic strategies is urgently needed to improve survival in MM patients. MM cells show a relatively low proliferative index and a poor apoptotic rate (Dimberg et al., 2005), and hence it is likely that the transforming event is a failure in death regulation rather than a loss of growth control. This implies that a therapeutic potential for MM may lie in potentiating apoptosis, and thereby we hypothesized that MM might be suitable for the particular proapoptotic features of edelfosine. Our previous in vitro data indicated that edelfosineinduced apoptosis in cancer cells, including MM, was mediated by the reorganization of raft protein
281
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3749
Results Edelfosine induces Fas/CD95-membrane raft co-clustering and selective killing in MM patientderived cells, sparing normal cells We have previously reported that edelfosine induced apoptosis in human MM cell lines, mediated by coclustering of the Fas/CD95 death receptor in lipid rafts (Gajate and Mollinedo, 2007). Here, we found that edelfosine also induced apoptosis and co-aggregation of Fas/CD95 and rafts in CD138 þ malignant cells derived from MM patients (Figures 1a and b), whereas CD138" normal cells from the same patients, and human normal resting peripheral blood lymphocytes (PBLs), isolated from healthy volunteers, were spared (Figure 1a). MM patient-derived cells took up significant amounts of edelfosine (392±43 pmol per 106 cells after 1 h of incubation with 10 nmol [3H]edelfosine; n ¼ 5), whereas CD138" normal cells from the same patients and normal PBLs did not incorporate the drug (o20 pmol per 106 cells after 1 h of incubation with 10 nmol [3H]edelfosine; n ¼ 3). In addition, normal human cells with distinct proliferation rates, such as purified B cells, human umbilical vein endothelial cells, hepatocytes, CD34 þ hematopoietic progenitor cells and adipose-derived stem cells, were resistant to edelfosine (o3% apoptosis following treatment with 10 or 20 mM edelfosine for 24 h, n ¼ 3) and did not incorporate significant amounts of drug (o20 pmol per 106 cells after 1 h of incubation with 10 nmol [3H]edelfosine; n ¼ 3). By using the raft marker fluorescein isothiocyanate (FITC)-labeled cholera toxin (CTx) B subunit that binds to ganglioside GM1 (Schon and Freire, 1989), mainly found in rafts (Harder et al., 1998), we observed that 10 mM edelfosine promoted a potent co-aggregation of Fas/CD95 and rafts in malignant cells derived from MM patients (Figure 1b). However, no co-clustering of Fas/CD95 and rafts was detected in normal CD138" cells, resting PBLs or purified B cells following incubation with the phospholipid ether (data not shown). Furthermore, incubation of MM patient-derived cells with the cholesterol-depleting agent MCD (Christian et al., 1997) disrupted rafts (Gajate and Mollinedo, 2001), as well as inhibited clustering of Fas/CD95 (data not shown) and apoptosis (41±5 vs 12±3 apoptosis in
a
Control
50
EDLF
% Apoptosis
composition (Gajate and Mollinedo, 2001, 2007; Gajate et al., 2004, 2009a, b), but here we studied whether edelfosine was acting by directly targeting rafts both in vitro and in vivo, using a clinically relevant disease. By using in vitro, ex vivo and in vivo approaches, we show here that edelfosine targets tumor cell lipid rafts and accumulates in these membrane domains, inducing selective apoptosis in MM cells through its preferential uptake in tumor cell rafts. Our data set raft-targeted therapy as a new framework in MM treatment, and identified edelfosine as the first raft-targeted antitumor drug, thus opening a new avenue in cancer chemotherapy.
40 30 20 10 0
b
CD138Raft
CD138+ Fas/CD95
PBLs Merge
Control
Edelfosine
Figure 1 Selective edelfosine-induced apoptosis and Fas/CD95 raft clusters in patient-derived primary MM cells. (a) Freshly isolated CD138 þ tumor MM cells and normal CD138" mononuclear cells from MM patients, as well as normal resting PBLs, were incubated for 24 h with 10 mM edelfosine (EDLF) and analyzed for apoptosis, assessed by the percentage of cells in the sub-G1 region following cell-cycle analysis by flow cytometry. Untreated control cells were run in parallel. Data shown are means±s.e. of five independent determinations. (b) Freshly isolated MM cells from patients were either untreated (control) or treated with 10 mM edelfosine for 9 h, and then stained with FITC-CTx B subunit to identify lipid rafts (green fluorescence) and with a specific anti-Fas/CD95 monoclonal antibody, followed by CY3-conjugated anti-mouse Ig antibody (red fluorescence). Areas of colocalization between membrane rafts and Fas/CD95 in the merge panels are yellow. Images shown are representative of three independent experiments. Bar, 8 mm.
cells untreated vs pretreated with MCD, and then incubated with 10 mM edelfosine for 24 h; n ¼ 3). Thus, the patient-derived MM cells behaved similarly to MM cell lines (Gajate and Mollinedo, 2007) in their capacity to undergo co-clustering of Fas/CD95 rafts as the initial trigger for apoptosis (Mollinedo and Gajate, 2006b; Mollinedo, 2008) on edelfosine treatment. These data suggest that edelfosine shows a remarkable selectivity for primary tumor cells and cancer cell lines in its capacity to promote co-aggregation of Fas/CD95 and rafts and to induce apoptosis. Accumulation of edelfosine in lipid raft clusters enriched in death receptors is critical for MM cell killing We next asked whether edelfosine interacted with MM cell membrane rafts. We tested this directly by Oncogene
282
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3750
incubating the MM cell line MM144 with [3H]-edelfosine followed by raft isolation. [3H]-edelfosine accumulated in GM1-containing rafts (Figure 2a). In addition, we found that the fluorescent analog all-[E]-1-O-[150 phenylpentadeca-80 ,100 ,120 ,140 -tetraenyl]-2-O-methyl-racglycero-3-phosphocholine (PTE-edelfosine) (Gajate et al., 2004; Quesada et al., 2004; Nieto-Miguel et al., 2006) accumulated in well-defined patches that colocalized with raft clusters enriched in Fas/CD95 and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) receptors, death receptor (DR) 4 and DR5 (Figure 2b). PTEedelfosine induced apoptosis in MM144 cells (38.6±4.1 and 59.8±5.2% apoptosis following 24-h incubation with 10 and 20 mM PTE-edelfosine, respectively; n ¼ 3) at a similar concentration range as the parent drug edelfosine (51.9±4.3 and 65.6±6.4% apoptosis following 24-h incubation with 10 and 20 mM edelfosine, respectively; n ¼ 3), while sparing normal resting PBLs (o3% apoptosis following 24-h incubation with 10 or 20 mM PTE-edelfosine). Fluorescent PTE-edelfosine incorporation into the MM cell was blocked by addition of the parent drug (data not shown), indicating the specificity in the uptake, and PTE-edelfosine behaves as a bona fide analog to analyze edelfosine localization in the cell. Raft disruption by MCD treatment (Christian et al.,
Fraction
1997) inhibited edelfosine-induced apoptosis (Figure 2c) and clustering of Fas/CD95 and TRAIL receptors (data not shown) in MM144 cells. Similar results were obtained with the MM cell line MM1S (data not shown). These results indicate that edelfosine binds to rafts forming aggregates of edelfosine-rich rafts that recruit death receptors and convey apoptotic signaling. Cholesterol is required for edelfosine uptake and drug-induced apoptosis We found that MM1S and MM144 cells took up edelfosine in the range 415–482 pmol per 106 cells after 1 h of incubation with 10 nmol [3H]edelfosine. To assess the role of cholesterol, a critical raft component, in mediating drug uptake and edelfosine-induced apoptosis, we treated MM cell lines with the cholesterol-depleting agent MCD (Christian et al., 1997). Measurements of [3H]cholesterol levels in cells that had been pre-equilibrated with this lipid showed that treatment with 2.5 mg/ml MCD for 30 min removed >67% of the cell cholesterol, in agreement with previous estimates (Barnes et al., 2004; Gajate et al., 2009b). Disruption of lipid rafts by MCD inhibited both drug uptake (Figure 3a) and edelfosine-induced apoptosis
Raft
Fas/CD95
PTE-EDLF
DIC
Raft
DR4
PTE-EDLF
DIC
Raft
DR5
PTE-EDLF
DIC
3H-EDLF (%)
25 20 15 10 5 0
1 2 3 4 5 6 7 8 9 10 11 12 GM1 128
128
Control
EDLF (51.0%)
Relative number of cells
(1.4 %)
0
0
128
0
1023 MCD (2.5 %)
0
1023
0
0
128
0
1023 MCD + EDLF (9.6%)
0
1023
Relative DNA content (PI Fluorescence)
Figure 2 Edelfosine accumulates in membrane rafts from MM cells enriched in death receptors, leading to apoptosis. (a) MM144 cells treated with 10 mM [3H]edelfosine (0.1 mCi) for 15 h were lysed in 1% Triton and fractionated by centrifugation on a discontinuous sucrose density gradient. An equal volume of each collected fraction was counted for radioactivity and subjected to SDS polyacrylamide gel electrophoresis, and the distributions of [3H]edelfosine (upper panel) and GM1-containing rafts (lower panel) (enriched in fractions 3–5) over the gradient fractions are shown. (b) MM144 cells were incubated with 10 mM PTE-edelfosine (PTEEDLF) (blue fluorescence) for 15 h, washed with PBS–2% BSA, and then its colocalization with membrane rafts and death receptors Fas/CD95, DR4 and DR5 was examined using the FITC-CTx B subunit (green fluorescence for rafts) and specific antibodies against the distinct death receptors, followed by CY3-conjugated antibodies (red fluorescence for death receptors). Differential interference contrast (DIC) images are also shown. Bar, 6 mm. (c) MM144 cells were untreated (control), treated with 10 mM edelfosine for 24 h (EDLF), or pretreated with 2.5 mg/ml MCD for 30 min and then incubated for 24 h in culture medium in the absence (MCD) or presence of 10 mM edelfosine (MCD þ EDLF), and apoptosis was analyzed by flow cytometry. The percentage of apoptotic cells, assessed by a DNA content less than G1 (sub-G1), is shown in parentheses in each histogram. The position of the G1 peak is indicated by arrows. Data shown are representative of three separate experiments. Oncogene
283
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3751 Control MCD MCD + CHOL
Control C8-Cer
60 40 20
MM1S
MM144 Control MCD CHOL EDLF MCD + EDLF MCD + CHOL + EDLF
70 60 % Apoptosis
60 40 20 0
50 40
EDLF
60
80
% Apoptosis
% EDLF uptake
% EDLF uptake
80
C8-Cer
70
100
100
0
Control
C8-Cer + EDLF
50 40 30 20 10
MM1S
0
MM1S
Figure 4 C8-ceramide inhibits edelfosine uptake and edelfosineinduced apoptosis in MM cells. (a) MM1S cells were untreated (control), or treated with 5 mM C8-ceramide for 30 min (C8-Cer), and then drug uptake was determined after incubation with 10 mM [3H]edelfosine (0.1 mCi) for 1 h. Data are shown as the percentage of edelfosine uptake, considering the drug uptake in untreated control cells as 100%. (b) MM1S cells were untreated (control), treated with 10 mM edelfosine for 24 h (EDLF), or treated with 5 mM C8-ceramide for 30 min, and then incubated for 24 h in culture medium in the absence (C8-Cer) or presence of 10 mM edelfosine (C8-Cer þ EDLF), and apoptosis was analyzed by flow cytometry as the percentage of hypodiploid (sub-G1) cells following cell-cycle analysis. Data shown are means±s.e. of three independent determinations.
30 20 10 0
MM1S
MM144
Figure 3 Involvement of cholesterol-containing rafts in edelfosine uptake and edelfosine-induced apoptosis in MM cells. (a) MM1S and MM144 cells were untreated (control) or treated with 2.5 mg/ ml MCD for 30 min (MCD), and then drug uptake was determined after incubation with 10 mM [3H]edelfosine (0.1 mCi) for 1 h. Cells were also treated with 2.5 mg/ml MCD for 30 min, followed by cholesterol repletion (MCD þ CHOL) and then drug uptake was examined as above. Data are shown as the percentage of edelfosine uptake, considering the drug uptake in untreated control cells as 100%. (b) MM1S and MM144 cells were untreated (control) and treated with MCD or MCD þ CHOL as above, and then incubated for 24 h in the presence of 10 mM edelfosine (EDLF). Apoptosis was analyzed by flow cytometry as the percentage of hypodiploid (subG1) cells following cell-cycle analysis. Data shown are means±s.e. of three independent determinations.
(Figure 3b). Replenishment of cholesterol cell content by exposure of MCD-treated cells to 100 mg/ml cholesterol for 1 h led to the recovery of B109% cell cholesterol, thus reconstituting cellular cholesterol levels, and cells regained the capacity to take up edelfosine and underwent apoptosis following edelfosine treatment (Figures 3a and b). These results indicate that the presence of cholesterol in rafts is essential for the uptake and apoptotic action of edelfosine. Ceramide inhibits edelfosine-induced apoptosis N-octanoyl-sphingosine (C8-ceramide) is a cell-permeable analog of naturally occurring ceramides that closely
mimics the effects of physiological long-chain ceramide (Karasavvas et al., 1996). Previous reports have shown that both natural and synthetic ceramides specifically displace sterols from lipid rafts affecting their properties and molecular composition (Megha and London, 2004). In this regard, cholesterol is replaced by ceramide in cholesterol-containing rafts to form ceramide-rich rafts (Megha and London, 2004). C8-ceramide has been reported to displace cholesterol from raft domains and destabilize them (Nybond et al., 2005; RouquetteJazdanian et al., 2007). We found that the pretreatment of MM1S cells with C8-ceramide removed cholesterol from the rafts (31±5 vs 9±2 mg cholesterol per mg protein in untreated vs C8-ceramide-treated cells; n ¼ 3), and inhibited both edelfosine uptake (Figure 4a) and edelfosine-induced apoptosis (Figure 4b). These results are in agreement with a critical requirement of cholesterol in rafts for the uptake of edelfosine into MM cells and the subsequent drug-induced apoptotic response. Edelfosine accumulates in MM tumor and inhibits human MM cell growth in vivo We next determined the in vivo antimyeloma activity of edelfosine using a severe combined immune deficiency (SCID) mouse plasmacytoma model. Following toxicity analyses (data not shown), we found that a daily oral administration dose of 30 mg/kg edelfosine was well tolerated, 45 mg/kg being the maximum tolerated dose. CB17-SCID mice were inoculated with 3 " 107 MM1S cells in 100 ml phosphate-buffered saline (PBS), together with 100 ml Matrigel. On the development of a palpable tumor, mice were randomized into drug-treated Oncogene
284
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
Control
3000 2000 1000
6 5 4 3 2 1 0
EDLF 3 5 9 11 13 16 18 20 23 25 27 30 days
**
Control
EDLF
6 5 4 3 2 1 0
**
Control
Tissue/plasma ratio
0
Tumor weight (g)
Control EDLF
4000
Tumor volume (cm3)
Tumor volume (mm3)
3752
EDLF
**
20 15 10 5 0
Liver
Kidney
Tumor
Figure 5 Edelfosine inhibits human MM cell growth in vivo. CB17-SCID mice were inoculated subcutaneously with 3 " 107 MM1S cells. Daily oral administration of edelfosine (30 mg/kg) started after the development of a palpable tumor. Caliper measurements of longest and shortest diameters of each tumor were carried out at the indicated times. Edelfosine (EDLF) significantly (Po0.01; from day 11 of treatment until the end of the experiment) inhibited MM tumor growth compared with the vehicle-treated control group (a). Data shown in panel a are means±s.e. (n ¼ 12). (b) A remarkable MM tumor growth inhibition was observed after 1-month treatment with edelfosine. Representative tumors isolated from two drug-free MM-bearing mice (control, upper panel) and from three drug-treated MM-bearing mice (EDLF, lower panel) are shown. (c) After completion of the in vivo assay (30 days), drug-free control and edelfosine-treated (EDLF) mice were killed and tumor weight and volume were measured. The tumor size and weight values of each single animal (dots) and the average values of each experimental group (horizontal bars) are shown. **Indicate values that are significantly different at Po0.01. (d) Tissue/plasma concentration ratios, in the liver, kidney and tumor, of edelfosine after daily oral administration of edelfosine (30 mg/kg) for 1 month in MM-bearing SCID mice (means±s.e., n ¼ 8). **Indicate values that are significantly different from liver/plasma and kidney/plasma ratios at P o0.01.
(30 mg/kg edelfosine, daily oral administration for 29 days) and control (water vehicle) groups of 12 mice each. Serial caliper measurements were made to determine the rate at which the tumor grew (Figure 5a). A comparison of tumors isolated from nontreated control and drugtreated MM-bearing mice, at the end of the 1-month treatment, rendered a remarkable and statistically significant (Po0.01) antimyeloma activity of edelfosine, with a reduction of B88% in tumor weight and volume, and four tumor-free mice (Figures 5b and c). Organ examination at necropsy did not reveal any apparent toxicity (data not shown). Compared with the large, highly vascularized tumors from drug-free mice, the small tumors from edelfosine-treated mice were pale and poorly vascularized (Figure 5b). No significant differences in mean body weight were observed between drug-treated and drug-free control animals (3–5% of body weight loss in the treated group vs control group). At the end of the in vivo assay, we found a mean concentration of edelfosine in plasma of 5.42±0.49 mg/ml (n ¼ 12), using high-performance liquid chromatography–mass spectrometry (HPLC–MS) analysis. Interestingly, edelfosine was remarkably more concentrated in the MM tumor (136.83±18.41 mg/g, n ¼ 8) than in the liver (7.91±1.04 mg/g, n ¼ 8) and kidney (13.46±0.26 mg/g, n ¼ 8). Tumor/plasma concentration ratio of edelfosine was dramatically higher than that detected in both kidney and liver, organs that are directly implicated in drug excretion (Figure 5d). Thus, our drug biodistribution studies indicate a rather selective accumulation of edelfosine in the MM tumor in vivo.
Tumor cholesterol levels affect edelfosine antimyeloma activity in vivo To analyze the putative role of rafts in edelfosine action in vivo, the 3-hydroxy-3-methylglutaryl-CoA reductase inhibitor pravastatin was given to mice (1.5 mg/kg) to reduce cholesterol levels. An in vivo MM-bearing CB17SCID mouse model assay was conducted involving four groups: untreated control mice, pravastatin-treated mice, edelfosine-treated mice, and mice treated with pravastatin and edelfosine. Pravastatin has been reported to reduce de novo cholesterol synthesis in the liver and in a number of tissues (Koga et al., 1990; Elahi et al., 2008). MM tumors derived from pravastatintreated mice contained a lower content of cholesterol than tumors isolated from untreated mice, 6.9 vs 13.4 mg cholesterol per g of tumor, respectively. Pravastatin significantly inhibited the in vivo antitumor action of edelfosine (Figures 6a, c and d) and drastically decreased the tumor/plasma concentration ratio of edelfosine (Figure 6b). This suggests that tumor cholesterol content has a major role in edelfosine uptake and in the antimyeloma activity of edelfosine in vivo. Although statins are lipid-lowering drugs that block cholesterol biosynthesis, they exert a wide array of effects by inhibiting prenylation of proteins involved in signal transduction (Perez-Sala and Mollinedo, 1994; Perez-Sala et al., 1995; Auer et al., 2002; Jasinska et al., 2007). On these grounds, we cannot rule out the possibility that the above effect of pravastatin could be due, at least in part, to prenylation inhibition.
Oncogene
285
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
4000 3500 3000 2500 2000 1500 1000 500 0
2
9
23
30 days
2 1 0
20
10 5 0
Liver
EDLF
Pravastatin Pravastatin + EDLF
Tumor
4 3 2 1 0
Control
Kidney
*
*
3
**
15
Tumor volume (cm3)
Tumor weight (g)
4
16
EDLF Pravastatin+EDLF
25 Tissue/plasma ratio
Tumor volume (mm3)
3753 Control EDLF Pravastatin Pravastatin + EDLF
Control
EDLF Pravastatin Pravastatin + EDLF
Figure 6 Pravastatin inhibits the in vivo antitumor action of edelfosine. MM1S-bearing CB17-SCID mice were randomly assigned to four cohorts of eight mice, each receiving a daily oral administration of water vehicle (control), 30 mg/kg edelfosine (EDLF), 1.5 mg/kg pravastatin and pravastatin þ EDLF for 29 days. Tumor shortest and longest diameters were measured weekly to calculate the apparent tumor volume in the above four experimental groups (a). Data shown in panel a are means±s.e. of eight mice in each group. (b) Tissue/plasma concentration ratios, in the liver, kidney and tumor, of edelfosine content after daily oral administration of edelfosine (EDLF) and pravastatin þ EDLF for 1 month in MM-bearing SCID mice (means±s.e., n ¼ 5). **Indicate that the tumor/ plasma ratio values in EDLF and pravastatin þ EDLF are significantly different at Po0.01. Mice were killed 30 days after the initiation of drug treatment, and tumor weight (c) and volume (d) were measured, showing that pravastatin hampered the antitumor effect of edelfosine. *Denotes values that are significantly different between the indicated groups at Po0.05.
Discussion The in vitro and in vivo data reported here identify lipid rafts as a novel and effective therapeutic target in MM, setting a novel framework in cancer chemotherapy. Our results show edelfosine as the first raft-targeted drug, accumulating in lipid rafts and triggering the demise of MM cells. Hence, edelfosine can be considered as the paradigm of this new raft-targeted therapy. The accumulation of edelfosine in rafts might be explained by its high affinity for cholesterol (Ausili et al., 2008; Busto et al., 2008), because of geometry compensation of the ‘cone shape’ of sterols and the ‘inverted cone shape’ of edelfosine that leads to a stable bilayer (Busto et al., 2008). Edelfosine targets membrane rafts of malignant cells, inducing raft aggregates that act as scaffolds for the recruitment and concentration of Fas/ CD95 and TRAIL receptors. Previous reports have shown that MM cells express Fas/CD95, but are rather resistant to undergo apoptosis in response to the external stimulation of Fas/CD95 by agonistic antiFas/CD95 antibodies (Dimberg et al., 2005). However, edelfosine activates Fas/CD95 and induces its aggregation in rafts from within the cell independently of its ligand (Gajate et al., 2004), triggering subsequently downstream extrinsic and intrinsic signaling pathways that eventually lead to apoptosis (Gajate et al., 2000a, b, 2004, 2009a, b; Gajate and Mollinedo, 2001, 2007).
The results reported here, together with our previous findings (Gajate and Mollinedo, 2007; Gajate et al., 2009b), suggest the following cascade of events in edelfosine-induced apoptosis in MM cells: drug accumulation in lipid raft - raft reorganization Fas/CD95 activation - DISC - mitochondrial cytochrome c release - apoptosome. The initial accumulation of edelfosine in lipid rafts is critical to trigger the whole apoptotic pathway, and raft disruption prevents edelfosine uptake and the subsequent activation of both extrinsic and intrinsic pathways (Gajate and Mollinedo, 2007; Gajate et al., 2009b). Although there is increasing evidence that Fas/CD95 apoptotic signaling is initiated through the recruitment of Fas/CD95 receptor in lipid rafts (Mollinedo and Gajate, 2006a, b; Gajate and Mollinedo, 2001, 2007; Mollinedo, 2008), it is not clear whether these domains, initially enriched in cholesterol, must turn into ceramide-rich rafts to deliver apoptotic signals (Grassme et al., 2003). In this regard, edelfosine does not activate sphingomyelinase and no ceramide is generated on its action (Gajate et al., 2004), suggesting that edelfosine acts through cholesterol-rich rafts. As a matter of fact, we found here that the presence of ceramide, which is known to displace cholesterol from lipid rafts (Megha and London, 2004), removes cholesterol from MM cell lipid rafts and inhibits both edelfosine uptake and edelfosine-induced apoptosis. On these grounds, we Oncogene
286
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3754
conclude that the lipid rafts involved in both the uptake and apoptotic action of edelfosine are cholesterol-rich rafts and not ceramide-rich rafts. Our biodistribution data in the MM murine model showed a mean concentration of edelfosine in plasma of 5.42 mg/ml (10.35 mM; edelfosine molecular mass, 523.7). Thus, the in vitro effects reported herein, rendered by 10 mM edelfosine, were detected at a pharmacologically relevant concentration. The edelfosine-induced death receptor concentration in specific domains of the cell surface reported here has interesting clinical consequences, as it might either induce apoptosis by itself, through subsequent recruitment of downstream signaling molecules (Gajate and Mollinedo, 2001, 2005, 2007; Gajate et al., 2004, 2009a; Mollinedo and Gajate, 2006a, b), or sensitize cancer cells to death receptor ligands, including TRAIL (Gajate and Mollinedo, 2007), currently under clinical trials. Using an MM xenograft mouse model, we have found that edelfosine accumulates in high amounts in the tumor, highlighting the rather selective affinity of edelfosine for tumor cells. In addition, edelfosine shows a remarkable in vivo antimyeloma activity, leading to drastic tumor regression. Following edelfosine oral administration in nontumor-bearing SCID mice, we have found a rather wide drug distribution pattern to several tissues, including the lung, spleen, intestine, liver and kidney (de Mendoza et al., 2009). Interestingly, we found here that when SCID mice contained MM tumors, edelfosine distribution was dramatically shifted toward the tumor (tumor/plasma concentration ratio D20), indicating a rather selective tumor targeting for edelfosine. We did not find any apparent damage in the distinct organs analyzed following necropsy in the in vivo studies reported here, and have recently found that edelfosine lacked toxicity in rats using histological, functional and biochemical parameters (Mollinedo et al., 2009). Furthermore, the small tumors isolated from edelfosine-treated mice were pale and poorly vascularized. This might be in agreement with reports showing an antiangiogenic effect of edelfosine (Zerp et al., 2008). Interestingly, pharmacological inhibition of endogenous cholesterol synthesis, which diminishes the cholesterol content of membrane rafts (Zhuang et al., 2005), led to a reduction in tumor cholesterol content as well as to a significant decrease in edelfosine antitumor activity and drug accumulation in the tumor. A decrease in cholesterol level might affect lipid rafts, thus hampering edelfosine uptake. Indeed, cholesterol depletion in MM cells inhibited both edelfosine uptake and drug-induced apoptosis. However, cholesterol repletion restored the ability of tumor cells to take up edelfosine and to undergo edelfosine-induced apoptosis. These cholesterol depletion/replenishment assays show that lipid rafts are critical for the uptake of edelfosine and for the triggering of apoptosis by this drug. The finding that edelfosine binds to cholesterol-containing rafts is of particular importance as the cholesterol content in tumor cells has been shown to be higher than in normal cells (Dessi et al., 1994; Kolanjiappan et al., 2003;
Freeman and Solomon, 2004; Tosi and Tugnoli, 2005). Furthermore, cancer cells have been reported to display higher levels of cholesterol-rich lipid rafts than their normal counterparts (Li et al., 2006). Edelfosine shows a high affinity for cholesterol, and for cholesterol-enriched membranes such as rafts (Ausili et al., 2008; Busto et al., 2008), because of the complementarity of the molecular geometrics of sterols and edelfosine (Busto et al., 2008), and this feature may have interesting biomedical applications. The high cholesterol content in tumor cells, together with the avidity for cholesterol of edelfosine, might contribute in part to the accumulation of edelfosine in tumor cells. Nevertheless, the uptake of edelfosine by tumor cells has been suggested to be mediated by a putative protein transporter that remains to be identified (Mollinedo et al., 1997; Hanson et al., 2003). It might be envisaged that tumor cells could undergo a series of changes in lipids, including cholesterol, and proteins, including transporters, which might putatively affect raft organization in a way that facilitates edelfosine uptake. The results reported here provide the proof-ofprinciple and rationale for further clinical evaluation of edelfosine and for this new raft-targeted therapy to improve patient outcome in MM.
Materials and methods Cell lines and primary cells Detailed information on the MM cell lines, primary tumor cells, and normal primary cells used in this study is included in the Supplementary data. Apoptosis assay Quantitation of apoptotic cells was determined by flow cytometry as the percentage of cells in the sub-G1 region (hypodiploidy) in cell-cycle analysis as previously described (Gajate et al., 2000b). Edelfosine was from INKEYSA (Barcelona, Spain) and APOINTECH (Salamanca, Spain). Confocal microscopy Cells were settled onto poly-L-lysine-coated slides and analyzed with a Zeiss LSM 510 laser scan confocal microscope (Oberkochen, Germany) for membrane raft and Fas/CD95 visualization, using FITC-CTx B subunit (Sigma Chemical, St Louis, MO, USA) and anti-human Fas/CD95 SM1/1 IgG2a mouse monoclonal antibody (Bender Medsystems, Vienna, Austria) followed by CY3-conjugated antimouse antibody (Pharmacia, Uppsala, Sweden) as described in Gajate and Mollinedo (2001). Colocalization assays were analyzed by excitation of both fluorochromes in the same section. Negative controls, lacking the primary antibody or using an irrelevant antibody, showed no staining. Edelfosine localization by fluorescence microscopy Cells were treated for 15 h with 10 mM fluorescent analog PTEedelfosine, provided by AU Acun˜a and F Amat-Guerri (Consejo Superior de Investigaciones Cientı´ ficas, Madrid, Spain), and then incubated with the raft marker FITC-CTx B subunit to label lipid rafts, and with an anti-Fas/CD95 SM1/ 1 monoclonal antibody (Bender Medsystems) or specific monoclonal antibodies against the extracellular domains of
Oncogene
287
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3755 DR4 and DR5 (R&D Systems, Minneapolis, MN, USA), followed by the corresponding CY3-conjugated secondary antibodies as previously described (Gajate et al., 2004; Gajate and Mollinedo, 2007). Colocalization of the distinct fluorochromes was analyzed using a fluorescence microscope (Axioplan 2; Carl Zeiss MicroImaging) and a digital camera (ORCA-ER-1394; Hamamatsu, Shizuoka, Japan). Negative controls, lacking the primary antibody or using an irrelevant antibody, showed no staining. Edelfosine uptake Drug uptake was measured as previously described (Mollinedo et al., 1997) after incubating 106 cells with 10 nmol [3H]edelfosine for 1 h in RPMI-1640/10% heat-inactivated fetal bovine serum, and subsequent exhaustive washing (six times) with PBS þ 2% bovine serum albumin to remove the ether lipid loosely bound to the cell surface. Lipid raft isolation Lipid rafts were isolated by using lysis conditions and centrifugation on discontinuous sucrose gradients as previously reported (Gajate and Mollinedo, 2001). The location of GM1-containing lipid rafts was determined by western blotting using CTx B subunit conjugated to horseradish peroxidase as previously described (Cheng et al., 1999). Cholesterol depletion and repletion To deplete cholesterol, 2.5–3 " 105 cells per ml were pretreated with 2.5 mg/ml MCD (Sigma) for 30 min at 37 1C in serum-free medium. Cells were then washed three times and resuspended in complete culture medium before the addition of edelfosine. For measurement of relative cholesterol contents, cells were preincubated at 37 1C for 14 h with 0.5 mCi/ml [3H]cholesterol (Perkin Elmer NEN Dupont, Boston, MA, USA) in cellculture medium (Barnes et al., 2004; Gajate et al., 2009b). After incubation without or with MCD, cells were washed, lysed and the radioactivity present in the lysates was then quantified by liquid-scintillation counting. For cholesterol repletion, 100 mg/ml of cholesterol balanced with MCD (watersoluble cholesterol; Sigma) was added for 1 h, and then the cells were washed and resuspended in culture medium for drug treatment. Cholesterol concentration was determined in cells and isolated membrane rafts as previously described (Hope and Pike, 1996; Zhuang et al., 2005), following lipid extraction with chloroform/methanol/HCl, using the Infinity reagent (Thermo Scientific, Worthing, UK). Xenograft mouse models CB17-SCID mice (Charles River Laboratories, Lyon, France), kept and handled according to institutional guidelines, complying with Spanish legislation under a 12/12 h light/dark cycle at a temperature of 22 1C, received a standard diet and acidified water ad libitum. CB17-SCID mice were inoculated subcutaneously into their lower dorsum with 3 " 107 MM1S cells in 100 ml PBS and 100 ml Matrigel basement membrane matrix (Becton Dickinson). When tumors were palpable, mice were randomly assigned to cohorts of 12 mice each, receiving a daily oral administration of edelfosine (30 mg/kg) or an equal volume of vehicle (water). The shortest and longest diameter of the tumor were measured with calipers at the indicated time intervals, and tumor volume (mm3) was calculated using the following standard formula: (the shortest diameter)2 " (the longest diameter) " 0.5. Animals were killed, according to institutional guidelines, when the diameter of their tumors reached 3–4 cm or when significant toxicity was observed.
Animal body weight and any sign of morbidity were monitored. Drug treatment lasted for 29 days, and mice were killed 24 h after the last drug administration. Then, tumors were extirpated, measured and weighed, and a necropsy analysis involving tumors and distinct organs was carried out. For in vivo analysis of the effect of pravastatin, four cohorts of eight CB17-SCID were assigned, receiving daily oral administration of pravastatin (1.5 mg/kg), edelfosine (30 mg/ kg), pravastatin þ edelfosine or an equal volume of vehicle (water). Pravastatin-treated animals received a 1-week pretreatment of pravastatin. Then, analysis was as above. Cholesterol concentration was determined in isolated tumors using a commercial kit (Calbiochem, San Diego, CA, USA) according to the manufacturer’s instructions. Plasma/tissue extraction procedure for edelfosine biodistribution studies MM-bearing SCID mice were treated with a daily oral administration of edelfosine (30 mg/kg) for 29 days. At 24 h after the last drug oral administration, blood was collected in EDTA surface-coated tubes and then centrifuged at 2000 g for 15 min (4 1C) to collect the plasma (100 ml). Then, the animals were killed and distinct organs and tumors were collected and weighed. Tissues and tumors were homogenized in 1 ml water (pH 7.4) using a Mini-bead Beater (BioSpect Products, Inc, Bartlesville, OK, USA) and centrifuged at 10 000 g for 10 min. Both plasma and tissue supernatants were collected and stored at #80 1C until HPLC–MS analysis was carried out. A 10 mg volume of platelet-activating factor (1 mg/ml), used as internal standard, was added onto 100 ml of plasma or tissue/tumor supernatant. A volume of 190 ml of a mixture of 1% formic acid/methanol was added to precipitate proteins. Samples were vortexed for 1 min and, after centrifugation (20 000 g, 10 min), 25 ml of the supernatant were analyzed by HPLC–MS. Quantitative determination of edelfosine by HPLC–MS analysis The technique used was a slight modification of a previously described method (Blanco-Prieto et al., 2004), and is detailed in the Supplementary data. The range of linear response of this technique was 5–250 ng for plasma samples and 5–685 ng for tissue samples. Edelfosine concentrations (mg/ml and mg/g) were calculated dividing the amount of edelfosine measured in samples by the volume of plasma or the weight of tissue used for the analysis. Statistical analysis All values are expressed as means±s.e. Between-group statistical differences were assessed using Mann–Whitney’s test or Student’s t-test. A P-value of o0.05 was considered statistically significant.
Conflict of interest The authors declare no conflict of interest.
Acknowledgements This work was supported by grants from Ministerio de Ciencia e Innovacio´n (SAF2007-61261, SAF2008-02251, PCT-0901002007-27, RD06/0020/1037 from Red Tema´tica de Investigacio´n Cooperativa en Ca´ncer, Instituto de Salud Carlos III), Oncogene
288
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3756 Fondo de Investigacio´n Sanitaria and European Commission (FIS-FEDER 06/0813, PS09/01915), Junta de Castilla y Leo´n (GR15-Experimental Therapeutics and Translational Oncology Program, and Biomedicine Project 2009) and Caja Navarra Foundation, Department of Health of the Navarra
Government (‘Ortiz de Landa´zuri, 2009’ project). CG is supported by the Ramo´n y Cajal Program from the Ministerio de Ciencia e Innovacio´n of Spain. AEHdM is supported by a research grant (BF106.37) from the Department of Education of the Basque Government.
References Auer J, Berent R, Weber T, Eber B. (2002). Clinical significance of pleiotropic effects of statins: lipid reduction and beyond. Curr Med Chem 9: 1831–1850. Ausili A, Torrecillas A, Aranda FJ, Mollinedo F, Gajate C, CorbalanGarcia S et al. (2008). Edelfosine is incorporated into rafts and alters their organization. J Phys Chem B 112: 11643–11654. Barnes K, Ingram JC, Bennett MD, Stewart GW, Baldwin SA. (2004). Methyl-beta-cyclodextrin stimulates glucose uptake in Clone 9 cells: a possible role for lipid rafts. Biochem J 378: 343–351. Blanco-Prieto MJ, Campanero MA, Mollinedo F. (2004). Quantitative determination of the antitumor alkyl ether phospholipid edelfosine by reversed-phase liquid chromatography-electrospray mass spectrometry: application to cell uptake studies and characterization of drug delivery systems. J Chromatogr B Analyt Technol Biomed Life Sci 810: 85–92. Busto JV, del Canto-Jan˜ez E, Gon˜i FM, Mollinedo F, Alonso A. (2008). Combination of the anti-tumour cell ether lipid edelfosine with sterols abolishes haemolytic side effects of the drug. J Chem Biol 1: 89–94. Cheng PC, Dykstra ML, Mitchell RN, Pierce SK. (1999). A role for lipid rafts in B cell antigen receptor signaling and antigen targeting. J Exp Med 190: 1549–1560. Christian AE, Haynes MP, Phillips MC, Rothblat GH. (1997). Use of cyclodextrins for manipulating cellular cholesterol content. J Lipid Res 38: 2264–2272. de Mendoza AE, Campanero MA, de la Iglesia-Vicente J, Gajate C, Mollinedo F, Blanco-Prieto MJ. (2009). Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-bearing immunosuppressed mice. Clin Cancer Res 15: 858–864. Dessi S, Batetta B, Pulisci D, Spano O, Anchisi C, Tessitore L et al. (1994). Cholesterol content in tumor tissues is inversely associated with high-density lipoprotein cholesterol in serum in patients with gastrointestinal cancer. Cancer 73: 253–258. Dimberg LY, Dimberg AI, Ivarsson K, Stromberg T, Osterborg A, Nilsson K et al. (2005). Ectopic and IFN-induced expression of Fas overcomes resistance to Fas-mediated apoptosis in multiple myeloma cells. Blood 106: 1346–1354. Elahi MM, Cagampang FR, Anthony FW, Curzen N, Ohri SK, Hanson MA. (2008). Statin treatment in hypercholesterolemic pregnant mice reduces cardiovascular risk factors in their offspring. Hypertension 51: 939–944. Freeman MR, Solomon KR. (2004). Cholesterol and prostate cancer. J Cell Biochem 91: 54–69. Gajate C, Del Canto-Janez E, Acuna AU, Amat-Guerri F, Geijo E, Santos-Beneit AM et al. (2004). Intracellular triggering of Fas aggregation and recruitment of apoptotic molecules into Fas-enriched rafts in selective tumor cell apoptosis. J Exp Med 200: 353–365. Gajate C, Fonteriz RI, Cabaner C, Alvarez-Noves G, AlvarezRodriguez Y, Modolell M et al. (2000a). Intracellular triggering of Fas, independently of FasL, as a new mechanism of antitumor ether lipid-induced apoptosis. Int J Cancer 85: 674–682. Gajate C, Gonzalez-Camacho F, Mollinedo F. (2009a). Involvement of raft aggregates enriched in Fas/CD95 death-inducing signaling complex in the antileukemic action of edelfosine in Jurkat cells. PLoS ONE 4: e5044. Gajate C, Gonzalez-Camacho F, Mollinedo F. (2009b). Lipid raft connection between extrinsic and intrinsic apoptotic pathways. Biochem Biophys Res Commun 380: 780–784.
Gajate C, Mollinedo F. (2001). The antitumor ether lipid ET-18-OCH3 induces apoptosis through translocation and capping of Fas/CD95 into membrane rafts in human leukemic cells. Blood 98: 3860–3863. Gajate C, Mollinedo F. (2002). Biological activities, mechanisms of action and biomedical prospect of the antitumor ether phospholipid ET-18-OCH3 (edelfosine), a proapoptotic agent in tumor cells. Curr Drug Metab 3: 491–525. Gajate C, Mollinedo F. (2005). Cytoskeleton-mediated death receptor and ligand concentration in lipid rafts forms apoptosis-promoting clusters in cancer chemotherapy. J Biol Chem 280: 11641–11647. Gajate C, Mollinedo F. (2007). Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and downstream signaling molecules into lipid rafts. Blood 109: 711–719. Gajate C, Santos-Beneit AM, Macho A, Lazaro M, Hernandez-De Rojas A, Modolell M et al. (2000b). Involvement of mitochondria and caspase-3 in ET-18-OCH3-induced apoptosis of human leukemic cells. Int J Cancer 86: 208–218. Grassme H, Cremesti A, Kolesnick R, Gulbins E. (2003). Ceramidemediated clustering is required for CD95-DISC formation. Oncogene 22: 5457–5470. Hanson PK, Malone L, Birchmore JL, Nichols JW. (2003). Lem3p is essential for the uptake and potency of alkylphosphocholine drugs, edelfosine and miltefosine. J Biol Chem 278: 36041–36050. Harder T, Scheiffele P, Verkade P, Simons K. (1998). Lipid domain structure of the plasma membrane revealed by patching of membrane components. J Cell Biol 141: 929–942. Hope HR, Pike LJ. (1996). Phosphoinositides and phosphoinositideutilizing enzymes in detergent-insoluble lipid domains. Mol Biol Cell 7: 843–851. Hussein MA, Juturi JV, Lieberman I. (2002). Multiple myeloma: present and future. Curr Opin Oncol 14: 31–35. Jasinska M, Owczarek J, Orszulak-Michalak D. (2007). Statins: a new insight into their mechanisms of action and consequent pleiotropic effects. Pharmacol Rep 59: 483–499. Karasavvas N, Erukulla RK, Bittman R, Lockshin R, Zakeri Z. (1996). Stereospecific induction of apoptosis in U937 cells by Noctanoyl-sphingosine stereoisomers and N-octyl-sphingosine. The ceramide amide group is not required for apoptosis. Eur J Biochem 236: 729–737. Koga T, Shimada Y, Kuroda M, Tsujita Y, Hasegawa K, Yamazaki M. (1990). Tissue-selective inhibition of cholesterol synthesis in vivo by pravastatin sodium, a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor. Biochim Biophys Acta 1045: 115–120. Kolanjiappan K, Ramachandran CR, Manoharan S. (2003). Biochemical changes in tumor tissues of oral cancer patients. Clin Biochem 36: 61–65. Li YC, Park MJ, Ye SK, Kim CW, Kim YN. (2006). Elevated levels of cholesterol-rich lipid rafts in cancer cells are correlated with apoptosis sensitivity induced by cholesterol-depleting agents. Am J Pathol 168: 1107–1118; quiz 1404-1105. Megha, London E. (2004). Ceramide selectively displaces cholesterol from ordered lipid domains (rafts): implications for lipid raft structure and function. J Biol Chem 279: 9997–10004. Mihelic R, Kaufman JL, Lonial S. (2007). Maintenance therapy in multiple myeloma. Leukemia 21: 1150–1157. Mollinedo F. (2007). Antitumor ether lipids: proapoptotic agents with multiple therapeutic indications. Expert Opin Ther Pat 17: 385–405.
Oncogene
289
Annex IV. Lipid raft-‐targeted therapy in multiple myeloma
Lipid raft-targeted therapy in multiple myeloma F Mollinedo et al
3757 Mollinedo F. (2008). Myeloma Therapy. Pursuing The Plasma Cell. Lonial S (ed). Humana Press: Totowa, NJ, pp Chapter 25, pp. 393–419. Mollinedo F, Fernandez-Luna JL, Gajate C, Martin-Martin B, Benito A, Martinez-Dalmau R et al. (1997). Selective induction of apoptosis in cancer cells by the ether lipid ET-18-OCH3 (edelfosine): molecular structure requirements, cellular uptake, and protection by Bcl-2 and Bcl-XL. Cancer Res 57: 1320–1328. Mollinedo F, Gajate C. (2006a). Fas/CD95 death receptor and lipid rafts: new targets for apoptosis-directed cancer therapy. Drug Resist Updat 9: 51–73. Mollinedo F, Gajate C. (2006b). Fas Signaling. Wajant H (ed). Landes Bioscience and Springer Science: Georgetown, TX, pp Chapter 2, pp 13–27. Mollinedo F, Gajate C, Martin-Santamaria S, Gago F. (2004). ET-18OCH3 (edelfosine): a selective antitumour lipid targeting apoptosis through intracellular activation of Fas/CD95 death receptor. Curr Med Chem 11: 3163–3184. Mollinedo F, Gajate C, Morales AI, del Canto-Janez E, Justies N, Collia F et al. (2009). Novel anti-inflammatory action of edelfosine lacking toxicity with protective effect in experimental colitis. J Pharmacol Exp Ther 329: 439–449. Nieto-Miguel T, Gajate C, Mollinedo F. (2006). Differential targets and subcellular localization of antitumor alkyl-lysophospholipid in leukemic versus solid tumor cells. J Biol Chem 281: 14833–14840. Nybond S, Bjorkqvist YJ, Ramstedt B, Slotte JP. (2005). Acyl chain length affects ceramide action on sterol/sphingomyelin-rich domains. Biochim Biophys Acta 1718: 61–66.
Perez-Sala D, Collado-Escobar D, Mollinedo F. (1995). Intracellular alkalinization suppresses lovastatin-induced apoptosis in HL-60 cells through the inactivation of a pH-dependent endonuclease. J Biol Chem 270: 6235–6242. Perez-Sala D, Mollinedo F. (1994). Inhibition of isoprenoid biosynthesis induces apoptosis in human promyelocytic HL-60 cells. Biochem Biophys Res Commun 199: 1209–1215. Quesada E, Delgado J, Gajate C, Mollinedo F, Acuna AU, AmatGuerri F. (2004). Fluorescent phenylpolyene analogues of the ether phospholipid edelfosine for the selective labeling of cancer cells. J Med Chem 47: 5333–5335. Rouquette-Jazdanian AK, Pelassy C, Breittmayer JP, Aussel C. (2007). Full CD3/TCR activation through cholesterol-depleted lipid rafts. Cell Signal 19: 1404–1418. Schon A, Freire E. (1989). Thermodynamics of intersubunit interactions in cholera toxin upon binding to the oligosaccharide portion of its cell surface receptor, ganglioside GM1. Biochemistry 28: 5019–5024. Simons K, Toomre D. (2000). Lipid rafts and signal transduction. Nat Rev Mol Cell Biol 1: 31–39. Tosi MR, Tugnoli V. (2005). Cholesteryl esters in malignancy. Clin Chim Acta 359: 27–45. Zerp SF, Vink SR, Ruiter GA, Koolwijk P, Peters E, van der Luit AH et al. (2008). Alkylphospholipids inhibit capillary-like endothelial tube formation in vitro: antiangiogenic properties of a new class of antitumor agents. Anticancer Drugs 19: 65–75. Zhuang L, Kim J, Adam RM, Solomon KR, Freeman MR. (2005). Cholesterol targeting alters lipid raft composition and cell survival in prostate cancer cells and xenografts. J Clin Invest 115: 959–968.
Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)
Oncogene
290
Annex V. Commentary: The Rafts of the Medusa: cholesterol targeting in cancer therapy
Annex V
Commentary: The Rafts of the Medusa: cholesterol targeting in cancer therapy MR Freeman, D Di Vizio and KR Solomon Oncogene, 2010. 29: p. 3745-‐47
291
Annex V. Commentary: The Rafts of the Medusa: cholesterol targeting in cancer therapy
Oncogene (2010) 29, 3745–3747
& 2010 Macmillan Publishers Limited All rights reserved 0950-9232/10 www.nature.com/onc
COMMENTARY
The Rafts of the Medusa: cholesterol targeting in cancer therapy MR Freeman1,2,3,4, D Di Vizio1,2,3 and KR Solomon1,2,5 1 Urological Diseases Research Center, Children’s Hospital Boston, Boston, MA, USA; 2Department of Urology, Children’s Hospital Boston–Harvard Medical School, Boston, MA, USA; 3Department of Surgery, Children’s Hospital Boston–Harvard Medical School, Boston, MA, USA; 4Department of Biological Chemistry and Molecular Pharmacology, Children’s Hospital Boston–Harvard Medical School, Boston, MA, USA and 5Department of Orthopaedic Surgery, Children’s Hospital Boston–Harvard Medical School, Boston, MA, USA
In this issue of Oncogene, Mollinedo and co-workers present promising evidence that cholesterol-sensitive signaling pathways involving lipid rafts can be therapeutically targeted in multiple myeloma. Because the pathways considered in their study are used by other types of tumor cells, one implication of this report is that cholesterol-targeting approaches may be applicable to other malignancies. Oncogene (2010) 29, 3745–3747; doi:10.1038/onc.2010.132; published online 3 May 2010
Cholesterol is a sterol that serves as a metabolic precursor to other bioactive sterols, such as nuclear receptor ligands, and also has a major role in plasma membrane structure. Cholesterol and longchain sphingolipids are believed to accumulate in ‘liquid-ordered’ patches termed lipid rafts, which are distinct in dynamic behavior from cholesterol-poor regions of the membrane. Biophysical, biochemical and imaging studies have provided strong evidence that lipid rafts exist in biological membranes, although details of their precise form and stability are still hotly debated. Cholesterol-rich microdomains have been shown to sequester a variety of membraneassociated signaling proteins, while excluding others, and a variety of approaches have shown that cholesterol itself has a role in transmitting cell growth, survival and differentiation signals. There has been speculation that cholesteroldependent membrane dynamics, such as receptor clustering, may be
Correspondence: Dr MR Freeman, Urological Diseases Research Center, Enders 1161, Children’s Hospital Boston, 300 Longwood Avenue, Boston, MA 02115, USA. E-mail:
[email protected] Received 14 March 2010; accepted 14 March 2010; published online 3 May 2010
targeted therapeutically in the case of certain malignancies. Published evidence suggests that a cholesterol-focused approach might work in some clinical scenarios. Cholesterol-lowering drugs that inhibit the enzyme HMG-CoA reductase, which catalyzes the ratelimiting step in cholesterol synthesis (these drugs are generically termed ‘statins’), have been reported to inhibit cancer incidence or progression in some studies. Although there is much controversy, buttressed by claims and counterclaims, in the various population-based reports of the effects of statins on cancer, recent evidence published by several groups examining large prospective series suggest that prostate cancer progression is likely to be inhibited by long-term cholesterol-lowering therapy (Platz et al., 2006; Mondul et al., 2010). These promising results in humans are in agreement with animal model data, in which cholesterol is raised or lowered and prostate tumor growth thereby promoted or inhibited, respectively (Zhuang et al., 2005; Solomon et al., 2009). Prostate cancer may be a special case, however, because recent observations from several groups indicate that tumor cells are capable of de novo androgen synthesis from cholesterol, obviating the need to acquire the hormone from the circulation. Because
androgens are generally thought to promote prostate cancer disease progression, the relative clarity of the epidemiological data in prostate cancer in comparison to other organ sites may arise from the role of cholesterol as the synthetic precursor of androgens. The paper by Mollinedo et al. in this issue, along with previous studies from this group and from other labs, provide persuasive evidence that membrane cholesterol itself is a critical mediator of cell survival signaling mechanisms that can be effectively targeted with clinically relevant drugs. Because this has been shown in multiple myeloma, a malignancy without the hormone dependence of prostate cancer, the in situ production of steroid hormones is an unlikely explanation for the sensitivity to cholesterol-targeting manipulations. Edelfosine, and the related compound, perifosine, are prototypic members of a family of synthetic lipids with tumor-killing properties known as alkyl-lysophospholipid analogs. Both drugs are presently in human trials and significant clinical responses have been reported. In a series of studies, edelfosine was shown to accumulate in cell membrane rafts and cause co-clustering of components of the death-inducing signaling complex, resulting in tumor cell apoptosis.
293
Annex V. Commentary: The Rafts of the Medusa: cholesterol targeting in cancer therapy
The Rafts of the Medusa MR Freeman et al
3746
Raft disruption by cholesterol depletion, using a variety of distinct approaches, inhibits drug uptake by tumor cells, preventing apoptosis. This inhibition is reverted by cholesterol replenishment. in vivo, edelfosine accumulates in mouse plasmacytomas, inducing apoptosis. However, the drug uptake by the tumor is reduced if tumor cholesterol levels are lowered by statin therapy, resulting in treatment failure. These and supporting observations strongly suggest that tumor apoptosis induced by the alkyllysophospholipid compounds is cholesterol dependent. Edelfosine was the first compound reported to promote apoptosis by ligand-independent activation of Fas/CD95 death receptors. However, this class of drug may also operate by cholesterol-sensitive mechanisms distinct from death-inducing signaling complex formation. Perifosine was shown to potently inhibit activation of the serine-threonine kinase Akt (Hideshima et al., 2006), which usually performs a prosurvival function. In other studies, Akt has been identified as a cholesterol-sensitive protein that localizes to lipid rafts (Adam et al., 2007), consistent with a mechanism of action where perifosine alters raftmediated signaling from Akt. Collectively, these results suggest that cholesterol/lipid raft targeting may cooperatively activate both prodeath and inhibit pro-survival pathways simultaneously. One important issue is that edelfosine is taken up specifically by CD138 þ malignant cells obtained from patients with multiple myeloma, whereas CD138" normal cells from the same patients and normal peripheral blood lymphocytes did not incorporate the drug. Similarly, edelfosine induced clustering of Fas/CD95 and TRAIL receptors in lipid rafts in tumor cells, but not in normal cells from the same patients. In vivo, edelfosine concentrated in the multiple myeloma tumors but the drug localized poorly in liver and kidney. These observations suggest that tumor cell membranes are substantially altered in lipid composition, a consequence of malignant transformation that
could be exploited with lipid-targeting agents. Lipid targeting in cancer therapy is of growing interest because of a reawakened appreciation for the role of metabolic pathways as essential features of malignant transformation and progression (Vander Heiden et al., 2009). Tumor cells consume high levels of glucose but often simultaneously suppress ATP production, diverting carbon atoms and reducing equivalents toward pathways that synthesize macromolecules, particularly lipids required for new membrane formation. Upregulation of fatty acid synthase, the major source of long-chain fatty acids (principally palmitate) in tumor cells, is a prominent characteristic of this metabolic profile (Menendez and Lupu, 2007). Lipogenesis is also activated by Akt upregulation, another common event in malignancy. Preclinical data have shown that fatty acid synthase is effectively targetable in vivo and efforts are underway to improve the toxicity profile associated with the available anti-fatty acid synthase compounds. Inhibiting various components of the growth factor receptor-PI3kinase-Akt-mTOR pathway, an active area in pharmaceutical R&D, will produce major alterations in lipid metabolism, only some of which are known. The landscape of tumor cell membranes is drastically altered from normal as a result of shifts in metabolism arising from malignant transformation; however, characterization of membrane lipid composition in tumor cells is still at a fairly primitive stage. Given the absolute requirement for cholesterol in the synthesis of mammalian cell membranes, it follows that rapidly proliferating tumor cells require more cholesterol than normal cells. Moreover, the ability of cancer cells to metastasize may depend on the formation of cholesterol-rich cell extensions called invadopodia, which may not form in the absence of excess cholesterol (Caldieri and Buccione, 2010). Thus, metastatic cells’ dependence on abnormal levels of cholesterol may prove to be their undoing if vulnerabilities in lipid metabolism can be identified
and exploited. As the tumor metabolism picture becomes clearer over the next decade, lipid-targeting strategies should prove relevant to the clinical situation, at least for some classes of tumors. Because of shared metabolic consequences of cell transformation that arise from the many varieties of genetic alterations in multiple signal transduction pathways, one can even imagine that lipid-targeting approaches may ultimately prove to be more widely efficacious than is currently the case for drugs directed toward the major protein signal transduction targets, such as ErbB family receptors and other kinases. There are some caveats that should be considered when evaluating the fortunes of cholesterol targeting. Modeling cholesterol reduction in rodents is challenging because statin drugs do not lower circulating cholesterol in mice and rats, as they do in humans. Cholesterol lowering in rodent tumors will thus depend on whether the drug can penetrate sufficiently into the tumor bed, a scenario that does not adequately model the conventional cholesterol reduction situation in humans, in which the level of extrahepatic statin is low and of short duration. However, other approaches to cholesterol lowering in mice have recently been developed (Solomon et al., 2009) and these might be exploited to identify cholesterol-sensitive tumor types by way of xenografts or transgenic models. Whether the unusually high level of selectivity toward tumor cells seen with edelfosine will be replicated with other lipid-targeting agents is an open question. In addition, high concentrations of cholesterol are naturally found in the brain, ilium, liver and prostate. Consequently, a cholesterol-targeting approach may not be selective for tumors at certain anatomical sites. It is also not clear whether edelfosine can cross either the blood–brain or blood–testes barriers, so malignancies at these locations may not be targetable by the alkyl-lysophospholipid class of compounds. A not unfitting metaphor for cancer, Medusa was a mythical
Oncogene
294
Annex V. Commentary: The Rafts of the Medusa: cholesterol targeting in cancer therapy
The Rafts of the Medusa MR Freeman et al
3747
creature, highly feared, with snakes in place of hair, which could turn someone to stone with a single glance. The Raft of the Medusa (Le Radeau de la Me´duse) is a famous oil painting by The´odore Ge´ricault, which depicts a tragic escape from a wrecked French naval vessel of the same name.
Although most of the castaways lost their lives in the escape attempt, some were saved by the hastily constructed raft. It is intriguing that the metaphor of the Raft of the Medusa providing sanctuary and life may someday extend to patients, where the cholesterol- and lipid-rich membrane environment
of tumors might provide effective therapeutic opportunities.
cancer pathogenesis. Nat Rev Cancer 7: 763–777. Mondul AM, Clipp SL, Helzlsouer KJ, Platz EA. (2010). Association between plasma total cholesterol concentration and incident prostate cancer in the CLUE II cohort. Cancer Causes Control 21: 61–68. Mollinedo F, de la Iglesia-Vicente J, Gajate C, Estella-Hermoso de Mendoza A, VillaPulgarin JA, Campanero MA et al. (2010). Lipid raft-targeted therapy in multiple myeloma. Oncogene 29: 3748–3757. Platz EA, Leitzmann MF, Visvanathan K, Rimm EB, Stampfer MJ, Willett WC et al. (2006). Statin drugs and risk of advanced
prostate cancer. J Natl Cancer Inst 98: 1819–1825. Solomon KR, Pelton K, Boucher K, Joo J, Tully C, Zurakowski D et al. (2009). Ezetimibe is an inhibitor of tumor angiogenesis. Am J Pathol 174: 1017–1026. Vander Heiden MG, Cantley LC, Thompson CB. (2009). Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324: 1029–1033. Zhuang L, Kim J, Adam RM, Solomon KR, Freeman MR. (2005). Cholesterol targeting alters lipid raft composition and cell survival in prostate cancer cells and xenografts. J Clin Invest 115: 959–968.
Conflict of interest The authors declare no conflict of interest.
References Adam RM, Mukhopadhyay NK, Kim J, Di Vizio D, Cinar B, Boucher K et al. (2007). Cholesterol sensitivity of endogenous and myristoylated Akt. Cancer Res 67: 6238–6246. Caldieri G, Buccione R. (2010). Aiming for invadopodia: organizing polarized delivery at sites of invasion. Trends Cell Biol 20: 64–70. Hideshima T, Catley L, Yasui H, Ishitsuka K, Raje N, Mitsiades C et al. (2006). Perifosine, an oral bioactive novel alkylphospholipid, inhibits Akt and induces in vitro and in vivo cytotoxicity in human multiple myeloma cells. Blood 107: 4053–4062. Menendez JA, Lupu R. (2007). Fatty acid synthase and the lipogenic phenotype in
Oncogene
295