EULALIO GRACIA CORTÉS Ciudad Real, Septiembre De 20

UNIVERSIDAD DE CASTILLA-LA MANCHA FACULTAD DE CIENCIAS Y TECNOLOGÍAS QUÍMICAS DEPARTAMENTO DE INGENIERÍA QUÍMICA

IMPREGNATION AND FUNCTIONALIZATION OF BIODEGRADABLE

POLYMERS VIA CLICK CHEMISTRY IN SUPERCRITICAL CO2

Memoria que para optar al grado de Doctor en Ingeniería Química presenta:

EULALIO GRACIA CORTÉS

Directores: Dr. Antonio de Lucas Martínez Dr. Ignacio Gracia Fernández

Composición del tribunal: Dr. Juan Francisco Rodriguez Romero Dra. Lourdes Calvo Garrido Dr. Louis Adriaenssens

Profesores que han emitido informes favorable de la tesis: Dr. Renata Adami Dr. Giuseppe Caputo

Ciudad Real, Septiembre de 2019

D. Antonio de Lucas Martínez y D. Ignacio Gracia Fernández, Catedráticos de Ingeniería Química de la Universidad de Castilla-La Mancha

certifican que:

Eulalio Gracia Cortés ha realizado bajo su dirección el trabajo titulado “Impregnation and functionalization of biodegradable polymers via click chemistry in supercritical CO2”, en el Departamento de Ingeniería Química de la Facultad de Ciencias y Tecnologías Químicas de la Universidad de Castilla-La Mancha. Considerando que dicho trabajo reúne los requisitos para ser presentado como Tesis Doctoral expresan su conformidad con dicha presentación.

Ciudad Real, a de de 2019

D. Antonio de Lucas Martínez D. Ignacio Gracia Fernández

TABLE OF CONTENTS

RESUMEN 1 DESCRIPCIÓN DEL TRABAJO REALIZADO 9 A. Introducción 12 A.1. Evolución de los polímeros 12 A.2. Aplicaciones principales de los biopolímeros 14 A.2.1. Aplicaciones industriales 15 A.2.2. Aplicaciones en medicina 16 A.2.2.1. Liberación controlada 19 A.3. Ácido poliláctico (PLA) y uso en liberación controlada 24 A.4. Compuestos orgánicos con propiedades farmacológicas 26 A.4.1. Curcumina 26 A.4.2. Cumarina 28 A.5. Funcionalización de polímeros 29 A.5.1. Química click 31 A.6. Tecnología supercrítica 33

A.6.1. CO2 como fluido supercrítico 35 A.7. Objetivo del presente trabajo 36 A.8. Bibliografía 38 B. Materiales y métodos 48 B.1. Materiales 48 B.2. Instalaciones experimentales 51 B.2.1. Instalación de polimerización 51 B.2.2. Instalación de modificación de compuestos orgánicos 53 B.2.3. Instalación de reacción en medio supercrítico 53 B.2.4. Instalación de liberación in vitro 56 B.3. Procedimientos 57 B.3.1. Instalación de polimerización 57

B.3.2. Proceso de síntesis de 4-azidometil-7-metoxicumarina y 3- 57 azidometilcumarina.

i B.3.3. Impregnación de principios activos 59 B.3.4. Instalación en medio supercrítico 60 B.3.5. Instalación de liberación in vitro 60 B.3.6. Elaboración de la disolución tampón 61 B.4. Bibliografía 62 C. Presentación de resultados 63

CHAPTER 1. Supercritical CO2 feasibility for the impregnation 67 and release of curcumin in PLGA 1.1. Introduction 70 1.2. Experimental 72 1.2.1. Materials 72 1.2.2. PLGA copolymerization set-up 73 1.2.3. Supercritial carbon dioxide impregnation set-up 73 1.2.4. Study of curcumin solubility 74

1.2.5. Impregnation of PLGA with curcumin using organic 74 solvents 1.2.6. Impregnation of PLGA using supercritical carbon dioxide 75 (scCO2) 1.2.7. In vitro release study 75 1.2.8. Characterization measurements 75 1.3. Results and discussion 77 1.3.1. Synthesis of PLGA 77 1.3.2. PLGA curcumin impregnation 79 1.3.3. Bulk impregnation of PLGA with curcumin 80

1.3.4. ScCO2 impregnation of PLGA with curcumin 83 1.3.5. In vitro release 87 1.4. Conclusions 91 1.5. Bibliography 93

CHAPTER 2. Functionalization and optimization of PLA with 97 coumarin via click chemistry in supercritical CO2 2.1. Introduction 100 2.2. Experimental 103 2.2.1. Materials 103 2.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin 103 2.2.3. Synthesis of click product at atmospheric pressure 104 2.2.4 Synthesis of click product at supercritical conditions 104 2.2.5. Characterization measurements 105 2.3.Results and discussion 105 2.4. Conclusions 113 2.5. Bibliography 114

CHAPTER 3. Feasibility of copper wire as heterogeneous 119 catalyst in click chemistry in scCO2 122 3.1. Introduction 3.2. Experimental 124 3.2.1. Materials 124 3.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin 124

3.2.3. Synthesis of click product in scCO2 124 3.2.4. Purification of click product 125 3.2.5. Cleaning process of copper wire catalyst 125 3.2.6. Characterization measurements 125 3.3. Results and discussion 126 3.4. Conclusions 147 3.5. Bibliography 149

CHAPTER 4. Application of click chemistry in scCO2 to PLA- 153 derived coumarin-release systems 4.1. Introduction 156

4.2. Experimental 158 4.2.1. Materials 158 4.2.2. Synthesis of of 3-(bromomethyl)coumarin 158 4.2.3. Synthesis of 3-azidomethyl coumarin 159

iii 4.2.4. Synthesis of click product in ScCO2 159 4.2.5. Characterization measurements 159 4.3. Results and discussion 160 4.3.1. Design of the PLA-coumarin adducts 160

4.3.2. 1H-NMR spectroscopic analysis 163 4.3.3. Maldi-Tof analysis 165 4.4. Conclusions 168 4.5. Bibliography 169

CHAPTER 5. Drug release profile of click products in 173 comparison with impregantions carried out in scCO2 5.1. Introduction 176

5.2. Experimental 177 5.2.1. Materials 177 5.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin 178 5.2.3. Synthesis of 3-azidomethyl coumarin 178

5.2.4. Synthesis of click product in scCO2 178

5.2.5. Impregnation of coumarin in scCO2 178 5.2.6. Purification of click products 179 5.2.7. In vitro release study 179 5.2.8. Characterization measurements 179 5.3. Results and discussion 180 5.4. Conclusions 191 5.5. Bibliography 193

Conclusions and recommendations 197 Conclusions 199

Recommendations 201

203 List of publications and conferences 205 Publications 206 Conferences iv

Annex I. Impregnation of curcumin into biodegradable support (poly-lactic-coglycolic acid, PLGA), to transfer its well known in 209 vitro effect to an in vivo prostate cancer model

Annex II. Scientific articles published 233

Resumen

El desarrollo de nuestra sociedad ha conllevado el desarrollo de nuevos materiales que sean capaces de satisfacer nuestras necesidades. En este ámbito aparecen los polímeros biodegradables, materiales muy versátiles que ofrecen un amplio rango aplicaciones. Los beneficios tanto ambientales como sociales de estos polímeros han permitido su uso en distintas aplicaciones, entre las que destacan las del ámbito médico, donde se usan en ingeniería de tejidos biológicos como soporte físico, en vendajes o la que en la actualidad está incrementando su estudio en investigación, la liberación controlada.

El uso de polímeros biodegradables para liberación controlada consiste en que un polímero actúa como soporte, el cual se combina con un principio activo de manera que éste se libera acorde a un modo prediseñado. Así se resuelven problemas actualmente existentes, como ciertas limitaciones para administración oral o intravenosa. Mediante los sistemas de liberación controlada (SLC) se consigue mejorar la biodisponibilidad eliminando así el riesgo de una degradación prematura, liberando más eficazmente los principios activos.

Dentro de los polímeros biodegradables se ha elegido el uso del ácido poliláctico (PLA), el segundo polímero más comercializado en la actualidad, fundamentalmente por su biodegradabilidad y biocompatibilidad, destacando en aplicación biomédicas como la liberación controlada de fármacos. En cuanto a los compuestos orgánicos naturales con propiedades farmacéuticas estudiados en esta tesis doctoral se han seleccionado la curcumina y cumarina. La curcumina es conocida por sus propiedades antiinflamatorias, antioxidantes, antimicrobianas, anticancerígenas o en la

Resumen

lucha contra el Alzheimer. Con respecto a la cumarina, este compuesto también exhibe propiedades antiinflamatorias, actividad anticoagulante, actividad antiviral en el tratamiento del VIH o en la lucha contra el cáncer de mama.

La combinación entre polímeros y fármacos para su uso en liberación controlada es un tema con gran interés en la actualidad. En concreto, son reseñables las ventajas que aporta la impregnación en medio supercrítico utilizando CO2 como disolvente, ya que permite una distribución homogénea del principio activo en la estructura del polímero y que la liberación del mismo sea controlada conforme el polímero se degrada.

En esta tesis doctoral se aborda el estudio de la viabilidad de la tecnología supercrítica para la mejora de la impregnación y liberación de curcumina en ácido poliláctico-co-glicólico (PLGA). Para ello se llevó a cabo un primer estudio comparativo de impregnación de curcumina en el polímero a presión atmosférica y en medio supercrítico, pudiendo así comparar el resultado obtenido en ambos medios. En este estudio se pudo comprobar que las impregnaciones llevadas a cabo en medio supercrítico doblaron prácticamente los rendimientos en comparación a las realizadas a presión atmosférica. La liberación del compuesto orgánico se llevó a cabo entre 9 y 12 días, dato que mejora sustancialmente los productos existentes.

Una vez comprobada la viabilidad de la tecnología supercrítica en compuestos destinados a aplicaciones farmacéuticas el próximo objetivo en esta tesis doctoral fue el estudio de la unión del compuesto orgánico, como la cumarina, y polímero biodegradable (ácido poliláctico) mediante un enlace covalente; validando en este capítulo la química click como método

Resumen de funcionalización para llevar a cabo esta unión de compuestos. En este apartado se pudo comprobar que el CO2 como disolvente actúa positivamente en esta reacción, utilizando un catalizador en fase homogénea. Las condiciones de operación optimizadas permitieron obtener rendimientos de funcionalización superiores al 90%.

El cambio del catalizador homogéneo por uno en fase heterogénea en la funcionalización mediante química click fue el siguiente apartado de estudio llevado a cabo en esta tesis doctoral. El empleo de hilos de cobre supone una ruta que permite obtener productos más puros mediante un proceso medioambientalmente más limpio. El proceso de purificación del producto final asociado supone un gran avance, ya que es posible su eliminación mediante una simple etapa de purificación. Para obtener las óptimas condiciones operacionales con este catalizador, diferentes parámetros como relación superficie/volumen, tiempo de reacción, reutilización de catalizador fueron estudiados obteniendo así polímeros funcionalizados con un rendimiento superior al 90%. Una vez determinada la viabilidad de este catalizador, se realizó un estudio de costes, donde se pudo comprobar que el uso de hilos de cobre reutilizados en medio supercrítico es la alternativa más económica, con un coste de producción de 108,97 €/kg.

En cuanto a la estructura química de cumarina que permita un abordaje más eficaz a la hora de llevar a cabo la reacción click, se llevó a cabo un estudio empleando el software Maestro, donde se estudió la interacción de las distintas posiciones para funcionalizar la cumarina considerando posibles impedimentos estéricos, llegándose a la conclusión que otra posición factible de poder ser funcionalizada era el carbono en posición 3. 3

Resumen

Finalmente se realizó un estudio de liberación in vitro que permitiera comparar muestras de cumarina impregnadas con muestras de cumarina funcionalizadas por química click en medio supercrítico. Mediante este estudio se pudo comprobar que aunque en ambos casos el mecanismo de liberación estaba en su mayoría condicionada a la fase de degradación del polímero, en el caso donde la cumarina había sido funcionalizada la liberación total del principio activo se alargó en 10 días si se compara con la cumarina impregnada sin ningún tipo de enlace covalente que le una al polímero biodegradable.

Por último, en colaboración con investigadores de la Universidad Europea de Madrid y de la Universidad de L’Aquila, una vez comprobado su buen rendimiento en liberación, las muestras impregnadas en medio supercrítico se utilizaron en ratones, pudiéndose esta vez determinar cómo la curcumina liberada del polímero en animales afectados con tumores genera una reducción del volumen del mismo, siendo más activa esta curcumina que la administrada por vía oral.

Resumen

Summary

The development in our society has required new materials which are able to cover the new necessities. In this way, biodegradable polymers appear as a versatile material which can be used in a wide range of applications. The environmental and social benefits related to these materials have allowed them to be used in different applications. In this way can be found industrial applications like applications in medicine where they are used in tissue engineering as scaffolds, where they work as physical and biological support, polymers also find application in dressing and one of the most important, whose investigation is increasing in the last years, controlled release.

The use of biodegradable polymer in controlled release consists on the polymer works as a support, it is combined with a drug which is released according to a prearranged way. The polymers application for controlled release is a solution for different limitations which can be found nowadays. Between this difficulties are the oral administration in which the drugs have a higher dissolution rate in comparison with gastrointestinal transit time or via intravenous where it is needed an organic solvent due to the low drug solubility. Through the drug delivery systems is possible to improve the bioavailability, removing the risk of a fast degradation. In this way the drug will be efficiently released.

Within biodegradable polymers is highlighted the use of polylactic acid (PLA), whose low production cost, and properties allows it to be the second most produced polymer nowadays. Two of its most important properties are its biodegradability and bioavailability, which make it to be highlighted in biomedical applications like drug release. About the natural organic compounds with pharmacological properties studied in this 5

Resumen doctoral thesis are found curcumin and coumarin. Curcumin is well known because of its versatility, which allows it to be used because of its anti- inflammatory, antioxidant o anticarcinogenic. With respect to coumarin, this compounds has also anti-inflammatory, anticoagulant, antiviral and anticarcinogenic properties being used in the treatment of breast cancer.

The combination between polymers and drugs for their correct dosage is one of the most interesting topics nowadays. In this way the polymer impregnation with drugs in supercritical media using CO2 as solvent, is of extreme importance, because it allows a homogeneous drug distribution in the polymer structure, allowing the drug release in a controlled way according to the polymer is being degraded.

In this doctoral thesis has been studied the feasibility of supercritical technology for the impregnation and release of curcumin in poly(lactic-coglycolic acid) (PLGA). For that purpose a comparative study between atmospheric and supercritical impregnations was carried out. In this study was checked that the impregnations performed in supercritical media practically doubled the yield obtained in comparison to the ones carried out at atmospheric pressure. The release of organic compound was carried out in a period between 9 and 12 days.

Once the supercritical technology feasibility in organic compounds destined to pharmaceutical applications was checked, the next target in this doctoral thesis was the study of the linkage between the polymer (polylactic acid) and the organic compound (coumarin) through a covalent bound. In this way click chemistry was confirmed in this chapter as the functionalization method to carry out the linkage of compounds. In this chapter was also checked that the CO2 as solvent has a great effect in this

Resumen reaction using a catalyst in homogeneous phase. The optimized operational conditions made possible to obtain functionalization yields over 90%.

The use of a catalyst in heterogeneous phase in spite of a catalyst in homogeneous phase in click chemistry was the following study carried out in this doctoral thesis. The use of copper wires make possible to obtain purer products in a cleaner environmental process. The purification process with the final product supposes a great advance because it is possible to achieve its total removal with a simple purification step. In order to obtain the optimum operational conditions with this catalyst, different parameters like surface/volume, reaction time or catalyst reusing were studied obtaining in this way functionalized polymer with a yield higher than 90%. Once the feasibility of this catalyst was checked, it was made a cost study where it was checked that the use of reused copper wires in supercritical media is the most economical alternative, with a manufacturing cost of 108.97€/kg.

About the coumarin structure, it was achieved to synthetize two different compounds, whose one of the main differences was related to the carbon in which the triazol ring would be formed. For that purpose a interaction study was carried out for the determination of the best position to modify the coumarin structure, being the carbon in position 3 the most convenient.

Finally, a drug release study in vitro was carried out. In this study the main purpose was to make the comparison between coumarin impregnate samples and the functionalized ones. It was observed that in both cases the drug release was controlled by the degradation of the polymer phase, but in the case of functionalized coumarin the whole release of the organic

Resumen compound was lasted 10 days more in comparison with the impregnated coumarin without any type of covalent link.

Descripción del trabajo realizado

A. Introducción A.1. Evolución de los polímeros A.2. Aplicaciones principales de los biopolímeros A.3. Ácido poliláctico (PLA) y uso en liberación controlada A.4. Compuestos orgánicos con propiedades farmacológicas A.5. Funcionalización de polímeros A.6. Tecnología supercrítica A.7. Objetivo del presente trabajo A.8. Bibliografía B. Materiales y métodos B.1. Materiales B.2. Instalaciones experimentales B.3. Procedimientos B.4. Bibliografía C. Presentación de resultados

ste trabajo forma parte de un amplio programa de investigación sobre la funcionalización de biopolímeros con compuestos E orgánicos caracterizados por sus propiedades farmacéuticas mediante tecnología supercrítica para su aplicación en fines terapéuticos, que se viene desarrollando durante los últimos años en el Departamento de Ingeniería Química de la Universidad de Castilla-La Mancha.

En particular, esta Tesis Doctoral tiene como objetivo el uso de polímeros biodegradables para su impregnación y funcionalización vía química click mediante la utilización de tecnología supercrítica. Este trabajo ha sido financiado por el Ministerio de Economía y Competitividad a través del proyecto titulado: “Funcionalización de Biopolímeros mediante Química Click en Medio Supercrítico”.

Descripción del trabajo realizado

A. Introducción

A.1. Evolución de los polímeros

A medida que la sociedad ha progresado en el ámbito tecnológico, se ha puesto de manifiesto la necesidad de desarrollar y optimizar nuevos materiales que sean capaces de satisfacer nuestras necesidades. Dentro de estos materiales, los polímeros han cobrado una gran importancia en su investigación debido a propiedades como su gran versatilidad, posibilidad de modificarlos, o incluso combinarlos con otros materiales.

Dentro de los polímeros se puede encontrar una clasificación en función de su procedencia, distinguiéndose así polímeros sintéticos y biopolímeros.

Los polímeros sintéticos aun siendo los más fabricados en la actualidad, presentan dos grandes inconvenientes: su procedencia de un recurso limitado y no renovable como es el petróleo, y su segunda limitación son los residuos generados a partir de estos, ya que aunque una pequeña parte de los plásticos se recicla, la mayoría se lleva a vertederos, donde pueden llegar a tardar cientos de años en descomponerse.

El segundo tipo de polímeros son los biopolímeros o bioplásticos, cuya obtención procede de recursos naturales renovables. Estos polímeros se clasifican a su vez en:

 Biopolímeros naturales. Los cuales son polímeros sintetizados por los seres vivos, encontrándose en este grupo los ácidos nucleicos, proteínas, polisacáridos, politerpenos o polihidroxialcanoatos (PHA).  Biopolímeros sintéticos. Son polímeros sintetizados artificialmente, entre ellos destacan: poliuretanos (PU), siliconas (Si), polimetilmetacrilato (PMMA), policaprolactona

Descripción del trabajo realizado

(PCL), poliácido glicólico y polivinilalcohol o alcohol polivinílico (PVA).  Biopolímeros derivados: Estos biopolímeros se caracterizan por ser sintetizados artificialmente pero a partir de sustancias naturales. Entre los más importantes destacan el ácido poliláctico (PLA), polietileno derivado del etanol de la caña de azúcar o los celuloides.

La importancia adquirida por los bioplásticos en los últimos años queda latente en la evolución del mercado producción. Así en el año 2018, la producción de bioplásticos fue de 2,1 millones de toneladas, dato que prevé ser incrementado aproximadamente un 25% en los próximos cinco años.

1% 1% 9% 16%

5% 19%

16% 10% 4% 1% 3% 8% 7%

Figura A.1. Producción global de bioplásticos estimada para el año 2023.

Una propiedad que cobra gran interés desde el punto de vista medioambiental es la biodegradabilidad de estos compuestos, proceso mediante el cual se produce la desintegración del polímero en pequeños

Descripción del trabajo realizado fragmentos como consecuencia de la ruptura de enlaces en su cadena principal. Durante el año 2018 un 43,2% de los bioplásticos producidos fueron biodegradables, destacando entre ellos la aportación del ácido poliláctico con un 10,3%. Este interés en los bioplásticos biodegradables no es algo puntual, tal y como muestran las estimaciones para los próximos años donde se prevé un incremento de su producción en un 6%, alcanzando el 49,2% con respecto a la cantidad total de bioplásticos producidos, tal y como se muestra en la Figura A.2.

3000

2500

2000 1328 1244 1200 1190 1179 1500 1175 1200

toneladas) 1000

1113 1288 500 885 912 994 1026 1026

Producción de bioplásticos (milesdeProducciónbioplásticos 0 2017 2018 2019 2020 2021 2022 2023

Biodegradable No biodegradable

Figura A.2. Producción y estimaciones en la producción de bioplásticos en miles de toneladas.

A.2. Aplicaciones principales de los biopolímeros.

Los beneficios ambientales y sociales derivados del uso de los bioplásticos pueden sopesarse frente a los problemas de durabilidad, ya que la diferencia entre el grado de biodegradabilidad entre polímeros con origen en el petróleo y los biopolímeros, es considerable. Un ejemplo significativo se encuentra en la comparación de vida útil entre el polietlen- tereftalato (PET), cuya vida útil se sitúa en 125 años, por los 2 años 14

Descripción del trabajo realizado estimados en el ácido poliláctico (PLA). El uso de polímeros biodegradables podría suponer una disminución del problema derivado de los residuos plásticos con su consecuente problema ambiental. La creciente tendencia en la industria de los biopolímeros es observable en las numerosas aplicaciones donde son utilizados. Así se puede observar que cada vez es más numerosos su uso tanto aplicaciones industriales como en medicina.

A.2.1. Aplicaciones industriales.

Entre las aplicaciones industriales más importantes donde los biopolímeros son utilizados se encuentran el tratamiento de aguas residuales, electrónica, agricultura o empaques.  Tratamiento de aguas residuales: En la actualidad uno de los adsorbentes más utilizados para llevar a cabo la eliminación de iones de metales pesados de las aguas residuales es el quitosano [1]. Este biopolímero posee gran capacidad de adsorción y excelentes propiedades tanto físicas como mecánicas, ausencia de toxicidad, biodegradabilidad, biocompatibilidad o bioactividad, que le hacen idóneo para la eliminación de contaminantes existentes en las aguas residuales [2].  Envasado: El uso de polímeros en la formación de películas da un valor añadido al producto debido a sus propiedades tanto físicas como mecánicas. En la industria de empaquetado los envases de plásticos utilizados para aplicaciones tanto alimentarias como no alimentarias son generalmente no biodegradables, procediendo en su mayoría del petróleo. Sin embargo, la investigación se encuentra focalizada actualmente en el desarrollo de envases de alimentos biodegradables a partir de biopolímeros. Entre de los

Descripción del trabajo realizado

biopolímeros más utilizados en esta aplicación se encuentran el almidón, celulosa, quitosano, caseína, PLA, PHA, PVA o PGA. Dentro de los biopolímeros anteriormente mencionados, el almidón es el polímero natural más utilizado en la producción de películas biodegradables debido a su gran disponibilidad, precio y biodegradabilidad [3]. El PLA es también objeto de investigación para la sustitución de polietieleno (PE) en el empaquetado de recipientes destinados para comida. La limitada estabilidad térmica del PLA, así como sus propiedades mecánicas y procesabilidad hacen necesario el uso de aditivos, como es el caso de colágeno, para poder mejorar su rendimiento en estas aplicaciones [4].  Agricultura: En los últimos años la búsqueda de métodos biológicos que permitan limitar o incluso eliminar el uso de productos químicos tanto para combatir las enfermedades en las plantas, como para incrementar la productividad de cultivos en la agricultura, ha conllevado la investigación de biopolímeros para cubrir estas necesidades. El uso de quitosano, biopolímero previamente comentado, se situa como una de las alternativas con mayor número de ventajas, debido a su accesible precio y posibilidad de ser combinado con otros compuestos que le permitan, como anteriormente en el caso del PLA, mejorar su rendimiento [5].

A.2.2. Aplicaciones en medicina.

El esfuerzo actual dedicado al desarrollo de polímeros biodegradables y biocompatibles ha permitido su aplicación en el mundo farmacológico. Así, avances en nanotecnología, fermentación bateriana,

Descripción del trabajo realizado recombinación de polímeros y avances en la ingeniería genética, entre otros, permiten añadir ciertas propiedades específicas a los polímeros naturales ya existentes, de manera que pueden ser utilizados con fines terapéuticos específicos. Entre las aplicaciones más utilizadas para biopolímeros en medicina se encuentra la ingeniería de tejidos biológicos y la liberación controlada de medicamentos. Siendo utilizados en menor medida en vendajes.  Ingeniería de tejidos biológicos: Su principal objetivo es el diseño de estructuras de material poroso (conocida por el término inglés “scaffold”), que sirvan como soporte físico y biológico, de manera que ayuden al crecimiento de las células, hasta una completa regeneración de tejidos u órganos. Para ello, el diseño de estas estructuras es fundamental, y es ahí donde los polímeros protagonizan una función esencial debido a su buen rendimiento. Actualmente, el principal reto a conseguir dentro del amplio rango de biopolímeros existentes es encontrar el correcto balance de propiedades que permita un buen funcionamiento en aplicaciones biológicas. Una de estas propiedades es la biodegradabilidad. La importancia de la biodegradabilidad reside en la facilidad con la que el polímero es degradado por acción biológica, de manera que se consiga evitar una posterior cirugía para eliminar el implante o una infección a largo plazo [6]. Generalmente, los polímeros sintéticos presentan mayores periodos de degradación, pudiendo variar entre 1-24 meses, frente al periodo de degradación de los biopolímeros, el cual oscila entre 12 horas y 5 meses [7, 8]. Otro parámetro de gran importancia son las propiedades mecánicas del scaffold, las cuales guardan relación directa con el

Descripción del trabajo realizado

peso molecular y cristalinidad del polímero, afectando a su vez a su velocidad y mecanismo de degradación.

Propiedades mecánicas Mantienen la integridad estructural y proporciona resistencia junto al tejido regenerado.

Biodegradable Sincronización entre la velocidad de degradación del scaffold y la regeneración del tejido, asegurando que los productos de degradación no sean tóxicos.

Biocompatible Soporte de actividad celular sin generación de inflamaciones crónicas o severas ni respuestas inmunogénicas.

Figura A.3. Propiedades requeridas en biopolímeros para su implantación en ingeniería de tejidos.

Un ejemplo donde las propiedades mecánicas del “scaffold” son esenciales es en los implantes diseñados para enfermedades óseas, mediante los cuales se puede conseguir la completa regeneración del hueso dañado [9].  Vendajes: Los tratamientos llevados a cabo en animales mediante el uso de biopolímeros como quitosina o quitina mostraron una sustancial reducción del tiempo de tratamiento y formación de cicatrices. Siendo también modificados mediante la

Descripción del trabajo realizado

incorporación agentes antimicrobianos a su estructura formando microgeles [10]. Otros polímeros, como el alcohol polivinílico (PVA), fue mezclado con quitosano, almidón y gelatina, pudiéndose observar mejoras en las actividades biológicas del PVA así como propiedades óptimas para vendaje y cicatrización de heridas [11].

A.2.2.1. Liberación controlada.

La liberación controlada de fármacos se produce cuando un material que actúa como soporte, y que en este caso es un polímero, es combinado con un principio activo, de manera que éste se libera acorde a un modo prediseñado. La dosificación de principios activos conlleva un gran número de problemas como consecuencia de las limitaciones de los mismos. Una de las limitaciones más importantes reside en la baja solubilidad de algunos principios activos en agua. Este hecho limita el uso de fármacos por administración oral, ya que las drogas prácticamente insolubes tienen una velocidad de disolución menor que el tiempo de tránsito gastrointestinal, resultando problemas graves de biodisponibilidad [12]. Otra vía de administración que presenta limitaciones es la vía intravenosa, en la cual los principios activos con baja solubilidad necesitan de la aplicación de un disolvente orgánico que permita la disolución de los mismos con anterioridad a la administración [13]. El uso de sistemas de liberación controlada (SLC) mediante polímeros permite reducir las limitaciones existentes con la consecuente mejora en biodisponibilidad. De esta manera se consigue eliminar el riesgo de una degradación prematura, así como poder llevar a cabo liberaciones eficaces de los principios activos.

Descripción del trabajo realizado

Tabla A.1. Problemás farmacológicos de los principios activos y las soluciones conseguidas a través de los sistemas de liberación controlada (SLC).

Problema Consecuencia Efecto de SLC

SLC como micelas o Un fármaco hidrofóbico liposomas puede puede precipitar en proporcionar medio acuoso. La Baja solubilidad hidrofilicidad o toxicidad está asociada hidrofobicidad, con el uso de aumentando la excipientes solubilidad del fármaco El principio activo Extravasación liberado del SLC puede Tejido dañado en accidental de fármacos reducir o eliminar el extravasación citotóxicos conlleva dañado de tejidos por dañado de tejidos extravasaciones accidentales Rápida descomposición SLC evita Pérdida de actividad degradaciones de del fármaco in vivo del fármaco fármaco prematuras Fármaco eliminado SLC puede alterar Desfavorable demasiado rápido, sustancialmente la requiriendo altas dosis farmacocinética farmacocinética del o continuas fármaco dosificaciones La naturaleza del SLC Fármacos con alta permite reducir el distribución en el volumen de cuerpo pueden afectar distribución y ayuda a Baja biodistribución a los tejidos, reducir los efectos provocanfo efectos secundarios en tejidos secundarios que distintos al tejido limitan la dosificación diana

Esta liberación eficaz consiste en la dosificación de medicamento dentro de los niveles denominados “efectivos” tal y como se puede observar en la Figura A.4, eliminando así una concentración en la dosificación por debajo

Descripción del trabajo realizado del nivel mínimo efectivo, o sobredosis, que podrían conllevar

intoxicaciones.

Máximo nivel deseado Mínimo nivel efectivo Dosificación tradicional

Dosificación controlada Concentración Concentración fármaco del

Tiempo

Figura A.4. Niveles de principios activos en sangre con liberación controlada.

La correcta dosificación del principio activo es una preocupación que queda latente en los actuales sistemas de tratamiento para algunas enfermedades. Tratamientos consolidados como la quimioterapia para tratamiento de cáncer son un ejemplo de ello [14, 15]. En este caso uno de los principales inconvenientes que surgen es la baja eficiencia de liberación conseguida en el tumor objetivo, dañando por consiguiente tejidos sanos donde erróneamente se libera el principio activo.

Descripción del trabajo realizado

Principio activo

Una mayor Solamente cantidad de parte del principio activo principio activo actúa en el tumor consigue llegar al tumor

Sistema de Tumor liberación controlada

Figura A.5. Esquema de comparación entre tratamiento tradicional y sistema de liberación controlada [16].

En la Figura anterior, se puede observar la diferencia existente entre la liberación llevaba a cabo en un tratamiento tradicional, donde parte del principio activo es liberado erróneamente en lugares distintos al tumor, con respecto a los sistemas de liberación controlada (SLC), mediante la cual el implante facilita la liberación en la zona específica a tratar. Este mecanismo de liberación permite obtener un mejor posicionamiento para llevar a cabo el tratamiento la zona afectada, lo que conlleva una reducción significativa de la cantidad de principio activo requerido, así como una reducción de los potenciales efectos secundarios. Otras ventajas derivadas de su uso son la posibilidad de conseguir una concentración de principio activo estable durante el tiempo de dosificación y la protección frente a la 22

Descripción del trabajo realizado degradación del principio activo hasta que no es totalmente liberado [15, 17, 18].

Los materiales poliméricos utilizados para la liberación controlada pueden clasificarse en dos categorías diferentes en función del momento en el que adoptan su forma, distinguiéndose sistemas prediseñados o moldeados con anterioridad y sistemas que adoptan su forma “in situ” en el lugar de tratamiento [19].

 Los sistemas prediseñados son caracterizados por tener una forma predefinida anteriormente a su implantación. Esta característica les permite tener tanto prolongadas liberaciones como prolongados perfiles de liberación. Su principal inconveniente reside en la inevitable necesidad de cirugía para llevar a cabo su implantación en el cuerpo. Los materiales empleados en este método de liberación son biopolímeros termoplásticos como PLA, PLGA o PCL, así como hidrogeles tanto naturales como sintéticos, son los soportes más utilizados [20-22].  Los sistemas “in situ” se caracterizan por estar compuestos de una solución o mezcla semisólida de un polímero y el principio activo, adoptando su forma previa inyección en el tumor objetivo [23, 24]. El proceso mediante el cual el complejo adopta su forma difiere dependiendo del tipo de biopolímero utilizado, pudiéndose clasificar a su vez en: o i) precipitación in situ, en la cual el polímero precipita de la disolución como consecuencia de la eliminación del disolvente [25], cambio de temperatura [26] o pH [27]; o ii) entrecruzamiento in situ, donde las cadenas del polímero se entrecruzan en el momento de la inyección a consecuencia

Descripción del trabajo realizado

de un enlace covalente [28], o no covalente de la cadena polimérica [29].

A.3. Ácido poliláctico (PLA) y uso en liberación controlada.

El ácido poliláctico (PLA) ocupa un lugar muy importante en la industria polimérica. Su bajo coste de producción, así como sus destacadas propiedades lo permite estar situado como el segundo polímero más comercializado actualmente [30]. Dos de sus propiedades más características son su biodegradabilidad y biocompatibilidad, las cuales le permiten destacar en aplicaciones biomédicas como la liberación controlada de fármacos [31, 32].

Lactida Ácido poliláctico

Figura A.6. Síntesis química de ácido poliláctico.

La síntesis de este biopolímero es llevada a cabo a partir de su monómero (lactida), el cual permite en función de su quiralidad obtener PLA procedente de D-Lactida, D/L-Lactida o L-Lactida. La adopción de las dos primeras configuraciones otorga al polímero una mayor velocidad en su degradación, y por tanto un perfil más acentuado en la liberación controlada de fármacos [33]. Estas configuraciones también le permiten ser semipermeable tanto a oxígeno como al agua, consiguiendo mayor susceptibilidad a la biodegradación en comparación con otros polímeros destinados para fines biomédicos [34]. 24

Descripción del trabajo realizado

El principal mecanismo de degradación del PLA es el proceso de hidrólisis llevado a cabo por la presencia de grupos ester en la cadena polimérica. Este aspecto le permite ser degradado en productos no tóxicos que le confieren la biocompatibilidad necesaria para ser utilizado en aplicaciones biomédicas.

Parámetros como el pH o la temperatura también tienen gran influencia en la velocidad de degradación [35]. Según estudios bibliográficos, el PLA muestra un periodo de degradación de 100 horas a pH neutro, mientras que en un pH ácido 3 no se muestra aparente degradación después de 400 horas de seguimiento [36]. Con respecto a la temperatura, se puede observar una velocidad de degradación 4 veces mayor al aumentar la temperatura de 25ºC a 37ºC.

Además de su aplicación en implantes para regeneración de tejidos, este biopolímero encuentra aplicación en la liberación controlada de fármacos en enfermedades tan importantes como el cáncer. Un ejemplo de aplicación es la síntesis de micelas con PLA cargadas con un principio activo anticancerígeno como es Paclitaxel, que junto a marcadores tumorales, permiten conocer la biodistribución y focalización de las nanopartículas en el tumor, así como estudiar la respuesta terapéutica al tratamiento [37]. Otra ventaja en el uso de PLA en liberación controlada es su modularidad, la cual permite combinar PLA con otros polímeros para llevar a cabo la liberación del principio activo [38]. De esta manera se consigue potenciar las ventajas derivadas de cada uno de los polímeros utilizados para este fin, disminuyendo sus limitaciones.

Descripción del trabajo realizado

A.4. Compuestos orgánicos con propiedades farmacológicas.

Además de los principios activos comercializados con fines farmacéuticos comentados anteriormente, los compuestos naturales con propiedades farmacológicas ocupan un lugar importante en la medicina actual por sus amplios beneficios y pocos efectos secundarios.

Estos compuestos, obtenidos a partir de plantas son caracterizados por poseer una gran variedad de propiedades beneficiosas entre las que destacan en función del compuesto las antioxidantes, antiinflamatorias o incluso antitumorales.

A.4.1. Curcumina.

Este compuesto natural es obtenido a partir del rizoma de la cúrcuma, una planta cultivada en el sudeste de Asia tropical [39]. Es característico su color amarillo dorado, color que viene determinado precisamente por la presencia de la curcumina en su composición.

Figura A.7. Planta de cúrcuma y estructura química de curcumina.

La estructura correspondiente a la curcumina se trata de un polifenol hidrofóbico. Esta estructura puede variar en función del medio donde se encuentre, así en medio ácido y neutro presenta predominantemente su forma ceto, mientras que en medio alcalino ésta se encuentra en su forma enol.

Descripción del trabajo realizado

Figura A.8. a) Estrucutra ceto de curcumina, b) Estructura enol de curcumina.

El uso de la curcumina es muy versátil, pudiéndose encontrar en productos alimenticios tales como la cúrcuma, en la cual la curcumina ha sido identificada como su principio activo, atribuyéndole así las propiedades antiinflamatorias que exhibe [40]. Además de su aplicación antiinflamatoria en alimentos, la curcumina también encuentra aplicación por sus funciones tanto antioxidantes, como antimicrobianas, anticancerígenas o en la lucha contra el Alzheimer. Su actividad como compuesto anticancerígeno ha sido probada en distintos tipos de tumores tales como tumores de colon, pulmón, páncreas o de próstata entre otros [40-45].

Sin embargo, este compuesto natural presenta una gran limitación para ser aplicado directamente, la cual es su baja solubilidad en agua, y por tanto su biodisponibilidad. Por esta razón, se ha llevado el estudio del uso de curcumina aplicando diferentes tecnologías como liposomas, nanopartículas poliméricas, micelas o adición con otras moléculas con el objetivo de poder minimizar esta limitación [46]. Además de un aumento de la solubilidad de curcumina, la implementación de estas tecnologías ayuda a proteger al principio activo de la inactivación por hidrólisis, consiguiendo de esta manera una liberación más prolongada de curcumina. 27

Descripción del trabajo realizado

A.4.2. Cumarina

La cumarina es un compuesto orgánico de origen natural perteneciente a la familia de las benzopironas, cuya estructura base es 2H-cromen-2-ona. La adición de distintos elementos o grupos funcionales a esta molécula constituye la familia de las cumarinas.

Figura A.9. Estructura química de cumarina.

Los distintos tipos de cumarinas se diferencian en función de su estructura química, pudiéndose distinguir así entre cumarinas simples [47, 48], furano cumarinas [48], dihidrofurano cumarinas[49], pirano- cumarinas (lineales y angulares) [50, 51], fenil-cumarinas o biocumarinas.

Con respecto a la actividad farmacológicas de esta molécula, estas sustancias exhiben actividad antiinflamatoria, donde son utilizadas por ejemplo en el tratamiento de edemas[52], actividad anticoagulante donde intervienen como antagonistas de vitamina K[53] o actividad antiviral, en el tratamiento de VIH. Otra de las aplicaciones de la cumarina se encuentra en la lucha contra el cáncer, donde ha sido probado su uso en tratamiento de cáncer de mama, así como su actividad ha sido testada en la reducción de efectos secundarios surgidos en la aplicación de la radioterapia [54].

Descripción del trabajo realizado

A.5. Funcionalización de polímeros.

La combinación polímero-fármaco para su correcta dosificación es un tema de investigación con gran desarrollo en la actualidad. El adecuado diseño del sistema de liberación permite mejorar las propiedades farmacológicas del principio activo seleccionado, modificando a su vez su perfil farmacocinético y biodistribución [55].

La unión polímero-fármaco conseguida mediante el enlace covalente de un compuesto funcionalizado confiere a dicho compuesto una mayor estabilidad en comparación a las impregnaciones físicas. Este hecho es transcendental de cara a la futura liberación del fármaco, asegurando dosificaciones del fármaco mucho más estables en el tiempo en comparación a los estudios clásicos de liberación [56-58].

Otra ventaja derivada de la funcionalización de polímeros consiste en que distintos grupos funcionales pueden ser eficientemente insertados en distintos lugares de la cadena polimérica, otorgando una gran variedad de posibilidades en su funcionalización.

En la actualidad existen diferentes tipos de reacciones mediante las cuales se lleva a cabo la funcionalización de polímeros:

 Modificación de Tiol-X: Este tipo de funcionalización consiste en la reacción entre compuestos tioles y distintos tipos de insaturaciones. Esta ruta de funcionalización es ampliamente reconocida debido a que su funcionamiento se asemeja a la de la química click, donde se obtienen productos altamente selectivos y eficientes con ausencia de productos secundarios [59]. Dentro de este tipo de funcionalización podemos distinguir entre modificaciones radicales y nucleófilas. Pudiendo observar así dentro del primer grupo adiciones tioéter con alquenos y alquinos, y dentro del segundo grupo adiciones X-ester,

Descripción del trabajo realizado

amida o ciano, adiciones de isocianato, oxirano o substitución halógena.  Modificaciones de isiocianato. Este tipo de adición nucleófila previamente mencionada, consiste en la adición de alcoholes, aminas, y tioles a isocianatos. Sin embargo, este tipo de funcionalización no se encuentra tan extendida debido a la toxicidad y sensibilidad que presentan sus reacciones, viéndose así limitada la eficacia de las mismas.  Reacciones de apertura de anillo. Las reacciones de apertura de anillo son a pesar de su reciente inclusión en modificaciones macromoleculares, unas de las reacciones clásicas dentro del mundo de los polímeros. Estas reacciones son termodinámicamente generadas por la liberación de la tensión del anillo, convirtiéndolas en un tipo de reacción muy versátil usadas en multitud de transformaciones. De hecho, las reacciones nucleófilas de apertura de anillo son consideradas como una de las primeras reacciones que cumplen los criterios para ser denominadas reacciones “click”. Dentro de los distintos grupos funcionales utilizados en este tipo de reacciones destaca el uso de epóxidos, así como el uso de aziridinas y azlactonas.  Reacciones de cicloadición: Las reacciones catalizadas por cobre mediante adición azida-alquino conocidas tanto como “química click” como por el acrónimo inglés “CuAAC”, son categorizadas entre las más importantes dentro de la funcionalización de polímeros en la actualidad. La utilización de cobre como catalizador en la funcionalización de polímeros fue introducida por Meldal y colaboradores [60], este hecho supuso un punto de inflexión en la funcionalización de polímeros debido a su eficiencia y modularidad.

Descripción del trabajo realizado

A.5.1. Química click.

El concepto “química click” fue introducido por el científico Sharpless en el año 2001. El objetivo de esta nuevo tipo de reacción química pretende conseguir mediante un número de reacciones eficientes que permitiesen llevar a cabo enlaces C-heteroátomo, la unión de moléculas sencillas en diversas estructuras complejas.

Las reacciones click son reacciones caracterizadas por cumplir una serie de criterios, como son la modularidad, eficiencia, selectividad o esteroespecificidad, haciendo posible la purificación del producto mediante una metodología sencilla.

Dentro del amplio rango de reacciones químicas que podrían cumplir estos criterios, existen cuatro grandes tipos de reacciones que podrían encajar perfectamente dentro del marco de la química “click”:

- Reacciones de cicloadición 1,3-dipolares, o cicloadiciones (4+2) hetero Diel-Alder. - Reacciones de sustitución nucleofílica/apertura de anillo. - Reacciones del grupo carbonilo de tipo no aldólica. - Reacciones de adición a enlaces múltiples C-C, como epoxidación, dihidroxilación, aziridinación y adiciones de haluros nitrosilo y sulfonilo.

Entre los tipos de reacción previamente comentados, la reacción de cicloadición 1,3-dipolar entre azidas y alquinos ha sido establecida como la reacción prototipo dentro de las clasificadas como “click”. Esta reacción fue descrita por Huisgen en 1960, constituyendo la ruta más eficiente para llevar a cabo la formación de 1,2,3-triazoles [61].

Descripción del trabajo realizado

Figura A.10. Esquema de reacción de cicloadición 1,3-dipolar.

El uso de cobre como catalizador supuso un punto de inflexión para el uso de esta reacción. Así, anteriormente a su uso eran necesarias altas temperaturas, lo que sumado a tiempos de reacción de entre 12 y 60 horas permitían obtener una mezcla de regioisómeros (1,4 y 1,5) en proporción equimolar.

El gran impacto producido por el uso de cobre (I) en esta reacción permitió aumentar su velocidad hasta 107 veces, junto a la formación exclusiva del isómero 1,4, lo que se traduce en una gran regioselectividad.

Las fuentes de cobre utilizadas en el catalizador son muy diversas. Así se pueden encontrar sales de cobre (ioduros, bromuros, cloruros, acetatos) y compuestos de coordinación complejos. Otras fuentes de cobre utilizadas son sales de cobre (II), como sulfato de cobre (II) pentahidratado, o acetato de cobre (II), en presencia de un agente reductor que genere in situ cobre (I).

Además del estudio de cobre como catalizador en química click, varios metales han sido estudiados con resultados satisfactorios, pudiendo encontrar entre ellos Rutenio (RuAAC) o Iridio (IrAAC).[62, 63].

Descripción del trabajo realizado

A.6. Tecnología supercrítica.

La tecnología supercrítica es posiblemente una de las tecnologías más reconocidas dentro de las técnicas de alta presión debido a sus excelentes resultados.

Entre sus principales distintivos esta tecnología se caracteriza por su nula toxicidad, inflamabilidad, no reactividad, accesibilidad económica y baja emisión contaminante. Estas características le han permitido encontrar aplicación en campos tan importantes como el farmacéutico, el textil o la industria alimentaria [64]. Las principales razones para el uso de la tecnología supercrítica no son solamente sus beneficios medioambientales, sino que además es una tecnología muy interesante desde el punto de vista económico[65].

Las ventajas anteriormente comentadas han contribuido a la implantación de esta tecnología como medida de solución a las limitaciones presentes en tecnologías convencionales tales como como la degradación térmica y química de principios activos, uso excesivo de disolventes orgánicos o presencia de residuos procedentes de los disolventes en el producto final [66]. En particular, la tecnología supercrítica ha permitido la sustitución de métodos actualmente utilizados, como por ejemplo, el “spray-drying” [67, 68].

Los fluidos utilizados en esta tecnología presentan características muy particulares, ya que sus condiciones de operación, tanto temperatura como presión, son aquellas que se encuentran por encima de sus puntos críticos respectivamente, hecho que les permite atribuirles el término de fluidos “supercríticos”. Estas condiciones de operación permiten a los fluidos supercríticos poder adoptar valores de densidad, difusividad y viscosidad intermedios entre las que equivaldrían a un gas y las de un líquido,

Descripción del trabajo realizado pudiendo estas además ser modificadas con el ajuste de los valores de presión y temperatura [69-71].

Presión (P, Mpa) (P, Presión

Tc Temperatura (T, ºC)

Figura 11. Diagrama de fases de una sustancia.

Tal y como puede observarse en el diagrama de fases (Figura 11), se pueden observar los distintos estados por los que pasa una sustancia en función de sus condiciones de operación. Así, se puede observar cómo coexisten las 3 fases en el denominado “punto triple”, punto en el cual se encuentran en equilibrio el estado sólido, líquido y gaseoso de una sustancia. A partir de este punto se puede determinar la temperatura y presión de vapor del compuesto.

A medida que aumenta la presión y temperatura se llega a un nuevo punto de equilibrio entre líquido y gas, denominado “punto crítico”. Es a partir de este punto donde el fluido deja de comportarse como líquido o gas para entrar a formar parte de la región supercrítica donde poseerá propiedades como se comentó anteriormente, intermedias entre ambas fases[72]. 34

Descripción del trabajo realizado

La cantidad de fluidos que pueden ser utilizados en medio supercrítico es extensa, así se puede encontrar dióxido de carbono, agua, acetona, mezclas CO2/etanol, clorodifluorometano, dietileter, óxido nitroso, propano o triofluorometano entre otros[73, 74].

A.6.1. CO2 como fluido supercrítico.

Entre los distintos fluidos usados en medio supercrítico destaca el uso del dióxido de carbono (CO2). A pesar de no ser el fluido con menores valores de temperatura y presión crítica, su accesible punto crítico (Pc: 7,38

MPa y Tc: 31,1 ºC) junto a su baja densidad, aplicabilidad a sustancias termosensibles, su nula reactividad, el no dejar residuos después de evaporación, su posibilidad de ser reciclado o su compatibilidad medioambiental le hacen uno de los disolventes más utilizados.

Disolventes como el Xenón (Xe) o Fluoruro de azufre (SF6) presentan valores más bajos en sus puntos críticos, pero tienen por inconveniente su alto coste de producción, hecho que les supone una gran limitación para su aplicabilidad.

En cuanto al uso de CO2 supercrítico (scCO2) este es muy variado, así puede ser utilizado como disolvente, antidisolvente o agente plastificante en síntesis, modificación y purificación de tanto polímeros naturales como sintéticos. Además de estos usos el scCO2 es usado también para el procesado de polímeros, así este disolvente encuentra aplicación en modificación de polímeros, reacciones de copolimerización, impregnaciones, o funcionalizaciones entre otros [75].

Además, el scCO2 ostenta el reconocimiento por parte de la Administración de Alimentos y Medicamentos de Estados Unidos (US- FDA) para su uso en procesos farmacéuticos como la liberación controlada de medicamentos debido a su baja toxicidad, hecho que sumado a las ventajas anteriormente comentadas de este disolvente, le hace ser 35

Descripción del trabajo realizado utilizado dentro de la tecnología supercrítica en la mayoría de los procesos con fines farmacéuticos [71, 76].

A.7. Objetivo del presente trabajo

El objetivo global del presente trabajo es la impregnación y funcionalización de biopolímeros biocompatibles mediante química click, empleando tecnología supercrítica con CO2 para su uso en aplicaciones farmacológicas. De este modo, se desarrolló un programa de trabajo con las siguientes etapas:

- Revisión bibliográfica y puesta a punto de las distintas instalaciones experimentales y técnicas de caracterización. - Impregnación del copolímero PLGA con curcumina tanto a presión atmosférica como en medio supercrítico. - Estudio de liberación in vitro de curcumina con el compuesto previamente impregnado en medio supercrítico. - Investigación de la funcionalización PLA con cumarina en medio supercrítico con objeto de llevar a cabo la reacción click. - Estudio estereoquímico para predecir la mejor estructura de cumarina que permita llevar a cabo la funcionalización en aquel enlace con menor impedimento estérico de cara a una futura liberación del compuesto orgánico. - Obtención y caracterización del compuesto obtenido mediante reacción click - Estudio de la viabilidad de la funcionalización utilizando un catalizador en fase heterogénea. - Estudio de comparación entre liberación in vitro de cumarina impregnada y funcionalizada mediante química click.

Descripción del trabajo realizado

“Functionalization and optimization of PLA with coumarin via click chemistry in supercritical CO2” “Copper wire as a clean and efficient catalyst for click chemistry in supercritical CO2” “Improvement of PLGA loading and “Feasibility of copper wire as release of curcumin by supercritical heterogeneous catalyst in click chemistry technology” at supercritical conditions” Impregnación y funcionalización Impregnaciones Química click Obtenido Cumarina PLA Cumarina PLGA Curcumina Esperable Catalizador de cobre

scCO2

PLA acetileno

“Improvement of PLGA loading and release of curcumin by supercritical technology”

Administración de fármacos Liberación del fármaco in vitro Perfil de via tradicional liberación vía 1ª dosis 2ª dosis 3ª dosis 1ª dosis química click Obtenido Rango tóxico

Rango terapéutico Esperable sangre Liberación

Rango sub-terapéutico controlada y eficaz

en una sola dosis Concentración de fármaco Concentración en

Tiempo Tiempo

“Impregnation of Curcumin into Biodegradable Support to Transfers its Well Known in vitro Effect to an in vivo Prostate Cancel Model” Pruebas clínicas

Obtenido

- Tratamiento de Esperable cáncer - Tratamiento de efectos de radioterapia

Figura 12. Graphical abstract de los temas tratados en la presente tesis doctoral.

Descripción del trabajo realizado

A.8. Bibliografía

 1. Wan Ngah, W.S., L.C. Teong, and M.A.K.M. Hanafiah, Adsorption of dyes and heavy metal ions by chitosan composites: A review. Carbohydrate Polymers, 2011. 83(4): p. 1446-1456.  2. Sarode, S., et al., Overview of wastewater treatment methods with special focus on biopolymer chitin-chitosan. International Journal of Biological Macromolecules, 2019. 121: p. 1086-1100.  3. Katarzyna L., G.L., Polymer Biodegradation and Biodegradable Polymers – a Review. Polish Journal of Environmental Studies, 2010. 19(2): p. 255-266.  4. Cheng, H.Y., et al., Modification and extrusion coating of polylactic acid films. Journal of Applied Polymer Science, 2015. 132(35).  5. Malerba, M. and R. Cerana, Recent advances of chitosan applications in plants. Polymers, 2018. 10(2).  6. Farokhi, M., et al., Silk fibroin/hydroxyapatite composites for bone tissue engineering. Biotechnology Advances, 2018. 36(1): p. 68-91.  7. Gunatillake, P.A., R. Adhikari, and N. Gadegaard, Biodegradable synthetic polymers for tissue engineering. European Cells and Materials, 2003. 5: p. 1-16.  8. Freier, T., et al., Controlling cell adhesion and degradation of chitosan films by N-acetylation. Biomaterials, 2005. 26(29): p. 5872-5878.  9. Oryan, A., S. Monazzah, and A. Bigham-Sadegh, Bone injury and fracture healing biology. Biomedical and Environmental Sciences, 2015. 28(1): p. 57-71.

Descripción del trabajo realizado

 10. Lee, Y.M., et al., β-Chitin-based wound dressing containing silver sulfurdiazine. Journal of Materials Science: Materials in Medicine, 2000. 11(12): p. 817-823.  11. Kamoun, E.A., et al., Crosslinked poly(vinyl alcohol) hydrogels for wound dressing applications: A review of remarkably blended polymers. Arabian Journal of Chemistry, 2015. 8(1): p. 1-14.  12. Stegemann, S., et al., When poor solubility becomes an issue: From early stage to proof of concept. European Journal of Pharmaceutical Sciences, 2007. 31(5): p. 249-261.  13. Medlicott, N.J., et al., Comparison of the Effects of Potential Parenteral Vehicles for Poorly Water Soluble Anticancer Drugs (Organic Cosolvents and Cyclodextrin Solutions) on Cultured Endothelial Cells (HUV-EC). Journal of Pharmaceutical Sciences, 1998. 87(9): p. 1138-1143.  14. Miller, K.D., et al., Cancer treatment and survivorship statistics, 2016. CA Cancer Journal for Clinicians, 2016. 66(4): p. 271-289.  15. Conde, J., N. Shomron, and N. Artzi, Biomaterials for Abrogating Metastasis: Bridging the Gap between Basic and Translational Research. Advanced Healthcare Materials, 2016. 5(18): p. 2312-2319.  16. Talebian, S., et al., Biopolymers for Antitumor Implantable Drug Delivery Systems: Recent Advances and Future Outlook. Advanced Materials, 2018. 30(31).  17. Bastiancich, C., et al., Anticancer drug-loaded hydrogels as drug delivery systems for the local treatment of glioblastoma. Journal of Controlled Release, 2016. 243: p. 29-42.

Descripción del trabajo realizado

 18. Kearney, C.J. and D.J. Mooney, Macroscale delivery systems for molecular and cellular payloads. Nature Materials, 2013. 12(11): p. 1004-1017.  19. Exner, A.A. and G.M. Saidel, Drug-eluting polymer implants in cancer therapy. Expert Opinion on Drug Delivery, 2008. 5(7): p. 775-788.  20. Indolfi, L., et al., A tunable delivery platform to provide local chemotherapy for pancreatic ductal adenocarcinoma. Biomaterials, 2016. 93: p. 71-82.  21. Seib, F.P. and D.L. Kaplan, Doxorubicin-loaded silk films: Drug-silk interactions and in vivo performance in human orthotopic breast cancer. Biomaterials, 2012. 33(33): p. 8442- 8450.  22. Yan, M., et al., Immunoglobulin G Expression in Human Sperm and Possible Functional Significance. Scientific Reports, 2016. 6.  23. Agarwal, P. and I.D. Rupenthal, Injectable implants for the sustained release of protein and peptide drugs. Drug Discovery Today, 2013. 18(7): p. 337-349.  24. Krukiewicz, K. and J.K. Zak, Biomaterial-based regional chemotherapy: Local anticancer drug delivery to enhance chemotherapy and minimize its side-effects. Materials Science and Engineering C, 2016. 62: p. 927-942.  25. Wang, S., et al., Injectable 2D MoS2-Integrated Drug Delivering Implant for Highly Efficient NIR-Triggered Synergistic Tumor Hyperthermia. Advanced Materials, 2015. 27(44): p. 7117-7122.  26. Huang, P., et al., Bridging the Gap between Macroscale Drug Delivery Systems and Nanomedicines: A Nanoparticle-Assembled 40

Descripción del trabajo realizado

Thermosensitive Hydrogel for Peritumoral Chemotherapy. ACS Applied Materials & Interfaces, 2016. 8(43): p. 29323-29333.  27. Li, L., et al., Injectable and Biodegradable pH-Responsive Hydrogels for Localized and Sustained Treatment of Human Fibrosarcoma. ACS Applied Materials & Interfaces, 2015. 7(15): p. 8033-8040.  28. Wu, X., et al., Synergistic therapeutic effects of Schiff's base cross-linked injectable hydrogels for local co-delivery of metformin and 5-fluorouracil in a mouse colon carcinoma model. Biomaterials, 2016. 75: p. 148-162.  29. Sun, J., et al., Synergistic in vivo photodynamic and photothermal antitumor therapy based on collagen-gold hybrid hydrogels with inclusion of photosensitive drugs. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2017. 514: p. 155-160.  30. Lactic Acid Market Analysis By Application (Industrial, F&B, Pharmaceuticals, Personal Care) & Polylactic Acid (PLA) Market Analysis By Application (Packaging, Agriculture, Transport, Electronics, Textiles), And Segment Forecasts. Grand View Research, 2017. 2014–2025.  31. Avérous, L. and E. Pollet, Biodegradable polymers. Environmental Silicate Nano-Biocomposites, 2012: p. 13-39.  32. Swed, A., et al., Sustained release of TGF-β1 from biodegradable microparticles prepared by a new green process in CO2 medium. International Journal of Pharmaceutics, 2015. 493(1-2): p. 357-365.  33. Jamshidian, M., et al., Poly-Lactic Acid: Production, applications, nanocomposites, and release studies.

Descripción del trabajo realizado

Comprehensive Reviews in Food Science and Food Safety, 2010. 9(5): p. 552-571.  34. Lehermeier, H.J., J.R. Dorgan, and J.D. Way, Gas permeation properties of poly(lactic acid). Journal of Membrane Science, 2001. 190(2): p. 243-251.  35. Kumar, S., et al., Controlled drug release through regulated biodegradation of poly(lactic acid) using inorganic salts. International Journal of Biological Macromolecules, 2017. 104: p. 487-497.  36. Xu, L., K. Crawford, and C.B. Gorman, Effects of temperature and pH on the degradation of poly(lactic acid) brushes. Macromolecules, 2011. 44(12): p. 4777-4782.  37. Yang, C., et al., Theranostic poly(lactic-co-glycolic acid) nanoparticle for magnetic resonance/infrared fluorescence bimodal imaging and efficient siRNA delivery to macrophages and its evaluation in a kidney injury model. Nanomedicine: Nanotechnology, Biology, and Medicine, 2017. 13(8): p. 2451- 2462.  38. Perez, C., et al., Poly(lactic acid)-poly(ethylene glycol) nanoparticles as new carriers for the delivery of plasmid DNA. Journal of Controlled Release, 2001. 75(1-2): p. 211-224.  39. Shishodia, S., G. Sethi, and B.B. Aggarwal, Curcumin: Getting Back to the Roots. Annals of the New York Academy of Sciences, 2005. 1056(1): p. 206-217.  40. Aggarwal, B.B., A. Kumar, and A.C. Bharti, Anticancer potential of curcumin: Preclinical and clinical studies. Anticancer Research, 2003. 23(1 A): p. 363-398.

Descripción del trabajo realizado

 41. Maheshwari, R.K., et al., Multiple biological activities of curcumin: A short review. Life Sciences, 2006. 78(18): p. 2081- 2087.  42. Duvoix, A., et al., Chemopreventive and therapeutic effects of curcumin. Cancer Letters, 2005. 223(2): p. 181-190.  43. Wang, Z., et al., Notch-1 down-regulation by curcumin is associated with the inhibition of cell growth and the induction of apoptosis in pancreatic cancer cells. Cancer, 2006. 106(11): p. 2503-2513.  44. Lev-Ari, S., et al., Curcumin synergistically potentiates the growth inhibitory and pro-apoptotic effects of celecoxib in pancreatic adenocarcinoma cells. Biomedicine and Pharmacotherapy, 2005. 59(SUPPL. 2): p. S276-S280.  45. Aggarwal, B.B., et al., Curcumin suppresses the paclitaxel- induced nuclear factor-κB pathway in breast cancer cells and inhibits lung metastasis of human breast cancer in nude mice. Clinical Cancer Research, 2005. 11(20): p. 7490-7498.  46. Naksuriya, O., et al., Curcumin nanoformulations: A review of pharmaceutical properties and preclinical studies and clinical data related to cancer treatment. Biomaterials, 2014. 35(10): p. 3365-3383.  47. Piller, N.B., A comparison of the effectiveness of some anti- inflammatory drugs on thermal oedema. British journal of experimental pathology, 1975. 56(6): p. 554-560.  48. Chiang, C.-C., et al., A Novel Dimeric Coumarin Analog and Antimycobacterial Constituents from Fatoua pilosa. Chemistry & Biodiversity, 2010. 7(7): p. 1728-1736.  49. Chakthong, S., et al., Alkaloid and coumarins from the green fruits of Aegle marmelos. Phytochemistry, 2012. 75: p. 108-113. 43

Descripción del trabajo realizado

 50. Basile, A., et al., Antimicrobial and antioxidant activities of coumarins from the roots of Ferulago campestris (Apiaceae). Molecules (Basel, Switzerland), 2009. 14(3): p. 939-952.  51. Patil, A.D., et al., The inophyllums, novel inhibitors of HIV-1 reverse transcriptase isolated from the Malaysian tree, Calophyllum inophyllum Linn. Journal of Medicinal Chemistry, 1993. 36(26): p. 4131-4138.  52. Hadjipavlou-Litina, D.J., K.E. Litinas, and C. Kontogiorgis, The Anti-inflammatory Effect of Coumarin and its Derivatives. Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2007. 6(4): p. 293-306.  53. Hirsh, J., et al., Oral Anticoagulants: Mechanism of Action, Clinical Effectiveness, and Optimal Therapeutic Range. CHEST, 2001. 119(1): p. 8S-21S.  54. Luo, K., et al., Anticancer Effects of Imperatorin Isolated from Angelica dahurica: Induction of Apoptosis in HepG2 Cells through both Death-Receptor- and Mitochondria-Mediated Pathways. Chemotherapy, 2011. 57(6): p. 449-459.  55. Yun, Y.H., B.K. Lee, and K. Park, Controlled Drug Delivery: Historical perspective for the next generation. Journal of Controlled Release, 2015. 219: p. 2-7.  56. Cabezas, L.I., et al., Novel Model for the Description of the Controlled Release of 5-Fluorouracil from PLGA and PLA Foamed Scaffolds Impregnated in Supercritical CO2. Industrial & Engineering Chemistry Research, 2014. 53(40): p. 15374- 15382.  57. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with 5-fluorouracil in supercritical CO2. The Journal of Supercritical Fluids, 2013. 80: p. 1-8. 44

Descripción del trabajo realizado

 58. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with indomethacin in supercritical CO2. The Journal of Supercritical Fluids, 2012. 63: p. 155-160.  59. Brosnan, S.M. and H. Schlaad, Modification of polypeptide materials by Thiol-X chemistry. Polymer, 2014. 55(22): p. 5511- 5516.  60. Tornøe, C.W., C. Christensen, and M. Meldal, Peptidotriazoles on solid phase: [1,2,3]-Triazoles by regiospecific copper(I)-catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. Journal of Organic Chemistry, 2002. 67(9): p. 3057- 3064.  61. Tron, G.C., et al., Click chemistry reactions in medicinal chemistry: Applications of the 1,3-dipolar cycloaddition between azides and alkynes. Medicinal Research Reviews, 2008. 28(2): p. 278-308.  62. Ding, S., G. Jia, and J. Sun, Iridium-Catalyzed Intermolecular Azide–Alkyne Cycloaddition of Internal Thioalkynes under Mild Conditions. Angewandte Chemie International Edition, 2014. 53(7): p. 1877-1880.  63. Kappe, C.O. and E. Van der Eycken, Click chemistry under non-classical reaction conditions. Chemical Society Reviews, 2010. 39(4): p. 1280-1290.  64. Kompella, U.B. and K. Koushik, Preparation of drug delivery systems using supercritical fluid technology. Critical Reviews in Therapeutic Drug Carrier Systems, 2001. 18(2): p. 173-199.  65. Hauthal, W.H., Advances with supercritical fluids [review]. Chemosphere, 2001. 43(1): p. 123-135.

Descripción del trabajo realizado

 66. Yeo, S.D. and E. Kiran, Formation of polymer particles with supercritical fluids: A review. Journal of Supercritical Fluids, 2005. 34(3): p. 287-308.  67. Vemavarapu, C., et al., Design and process aspects of laboratory scale SCF particle formation systems. International Journal of Pharmaceutics, 2005. 292(1-2): p. 1-16.  68. Wais, U., et al., Nanoformulation and encapsulation approaches for poorly water-soluble drug nanoparticles. Nanoscale, 2016. 8(4): p. 1746-1769.  69. Pasquali, I. and R. Bettini, Are pharmaceutics really going supercritical? International Journal of Pharmaceutics, 2008. 364(2): p. 176-187.  70. Kalani, M. and R. Yunus, Application of supercritical antisolvent method in drug encapsulation: a review. International journal of nanomedicine, 2011. 6: p. 1429-1442.  71. Davies, O.R., et al., Applications of supercritical CO2 in the fabrication of polymer systems for drug delivery and tissue engineering. Advanced Drug Delivery Reviews, 2008. 60(3): p. 373-387.  72. Clifford, A.A. and J.R. Williams, Introduction to Supercritical Fluids and Their Applications, in Supercritical Fluid Methods and Protocols, J.R. Williams and A.A. Clifford, Editors. 2000, Humana Press: Totowa, NJ. p. 1-16.  73. Byrappa, K., S. Ohara, and T. Adschiri, Nanoparticles synthesis using supercritical fluid technology – towards biomedical applications. Advanced Drug Delivery Reviews, 2008. 60(3): p. 299-327.  74. Ajzenberg, N., F. Trabelsi, and F. Recasens, What's new in industrial polymerization with supercritical solvents? A short 46

Descripción del trabajo realizado

review. Chemical Engineering and Technology, 2000. 23(10): p. 829-839.  75. Nalawade, S.P., F. Picchioni, and L.P.B.M. Janssen, Supercritical carbon dioxide as a green solvent for processing polymer melts: Processing aspects and applications. Progress in Polymer Science (Oxford), 2006. 31(1): p. 19-43.  76. Padrela, L., et al., Screening for pharmaceutical cocrystals using the supercritical fluid enhanced atomization process. Journal of Supercritical Fluids, 2010. 53(1-3): p. 156-164.

Descripción del trabajo realizado

B. Materiales y métodos

B.1. Materiales

Los materiales que se detallan a continuación son todos aquellos utilizados para el desarrollo del proyecto de investigación que engloba a la presente Tesis Doctoral (capítulos 1-5). A continuación se muestra una lista de todos los productos utilizados en la experimentación.

Tabla B.1. Compuestos químicos utilizados en experimentos de polimerizaciones, impregnaciones, modificación de compuestos, funcionalizaciones y liberaciones.

Producto Fórmula química Procedencia Pureza Polimerizaciones 3,6-dimetil-1,4-dioxano- C6H8O4 Purac Biochem 99,5% 2,5-diona (D,L-Lactida) 1,4-dioxano-2,5-diona C4H4O4 Purac Biochem 99,5% (Glicolida) Octoato de estaño SnOct2 Sigma-Aldrich 95%

Metanol CH3OH Sigma-Aldrich 99,8% Impregnaciones

Dióxido de carbono CO2 Carburos metálicos 99,5%

Curcumina C21H20O6 Sigma-Aldrich 98%

Cumarina C9H6O2 Sigma-Aldrich 99%

Ácido poliláctico (C3H4O2)n Corbion 99,5%

Etanol C2H5OH Sigma-Aldrich 99,5%

Ácido acético CH3COOH Sigma-Aldrich 99,5%

Etil lactato C5H10O3 Sigma-Aldrich 98%

Butil lactato C7H14O3 Sigma-Aldrich 98%

Acetona C3H6O VWR 99,8%

Descripción del trabajo realizado

Modificación de compuestos orgánicos

Nitrógeno N2 Carburos metálicos 99,5%

4-Bromometil-7- C21H20O6 Sigma-Aldrich 97% metoxiCumarina Azida sódica NaN3 Sigma-Aldrich 99,5%

Sulfato de sodio Na2SO4 Sigma-Aldrich 99%

DIPEA C8H19N Sigma-Aldrich 99%

Anhídrido propiónico C6H10O3 Sigma-Aldrich 99% Hidróxido sódico NaOH Sigma-Aldrich 99%

Cloruro amónico NH4Cl Sigma-Aldrich 99%

Peróxido de benzoilo C14H10O4 Sigma-Aldrich 98%

Salicilaldehído (CH3CH2CO)2CO Sigma-Aldrich 98%

Propionato de sodio CH3CH2COONa Sigma-Aldrich 99%

N-Bromosuccinimida C4H4BrNO2 Sigma-Aldrich 99%

Acetonitrilo CH3CN Sigma-Aldrich 99,8%

Tetrahidrofurano C4H8O Sigma-Aldrich 99,9%

Acetona C3H6O VWR 99,8%

Heptano C7H16 Sigma-Aldrich 99% Funcionalizaciones

Ácido poliláctico H(C3H4O2)nC3H3O Specific Polymers 98,3% acetileno Ioduro de cobre CuI Sigma-Aldrich 99%

Acetato de cobre Cu(CO2CH3)2·H2O Sigma-Aldrich 99% monohidratado Hilos de cobre Cu Sigma-Aldrich 99%

Dióxido de carbono CO2 Carburos metálicos 99,5%

Descripción del trabajo realizado

Liberaciones Cloruro de sodio NaCl Panreac 99% Cloruro de potasio KCl Panreac 99%

Fosfato KH2PO4 Panreac 99% monopotásico

Hidrógeno fosfato Na2HPO3 Panreac 99% de disodio

Tabla B.2. Listado de patrones de calibración utilizados en la investigación.

Patrón Fórmula química Mna (g/mol) Procedencia

Poliestireno [C8H8]x 370 Waters

Poliestireno [C8H8]x 474 Waters

Poliestireno [C8H8]x 996 Waters

Poliestireno [C8H8]x 2770 Waters

Poliestireno [C8H8]x 6520 Waters

Poliestireno [C8H8]x 9730 Waters

Poliestireno [C8H8]x 17800 Waters

Poliestireno [C8H8]x 44200 Waters

Poliestireno [C8H8]x 120000 Waters

Poliestireno [C8H8]x 177000 Waters a.Peso molecular en número promedio.

Descripción del trabajo realizado

Tabla B.3. Listado de disolventes utilizados para la caracterización de las muestras.

Producto Fórmula química Procedencia Pureza

Cloroformo CDCl3 Sigma-Aldrich 100% deuterado Tetrahidrofurano C4H8O Sigma-Aldrich 99,9% Ácido clorhídrico HCl Sigma-Aldrich 99% Ácido fluorhídrico HF Sigma-Aldrich 99%

B.2. Instalaciones experimentales

A continuación se detallan los equipos utilizados, sus componentes y las diferentes configuraciones usadas en los mismos.

B.2.1. Instalación de polimerización

Los experimentos de polimerización en masa se han llevado a cabo en un reactor discontinuo de tipo tanque agitado. Este sistema fue seleccionado tras la realización de experimentos previos por parte del grupo de Tecnología Química del Departamento de Ingeniería Química de la Universidad de Castilla-La Mancha en tubos de ensayo, en los que se puso de manifiesto la necesidad de agitación debido a la heterogeneidad del sistema bajo ciertas condiciones de reacción.

El equipo utilizado se muestra en la Figura B.1, la cual dispone de un vaso de reacción de 500 ml de capacidad, termostatizado con polietilenglicol 200, que se calienta con una placa calefactora Selecta, con control de temperatura, así como control de agitación. En la parte superior se dispone de una toma de nitrógeno para mantener atmósfera inerte y un cabezal de agitación digital Heidolph RZR 2041, con un intervalo de agitación de 0 a 2.000 rpm, que acciona un agitador tipo ancla. La toma de

Descripción del trabajo realizado muestras durante la reacción se lleva a cabo por medio de una espátula a través de una de las bocas de la tapa.

Figura B.1. Reactor a escala laboratorio: a) vista y b) esquema.

Sobre la escala a la que se han realizado los experimentos, hay que destacar que la mayoría de estudios previos en bibliografía han sido realizados a muy pequeña escala, empleando no más de un gramo de sustancia y se han utilizado ampollas como dispositivos de reacción. Además, los monómeros, catalizadores e iniciadores han sido sometidos a estrictas purificaciones para evitar impurezas[1]. En este estudio, por contra, se ha utilizado un reactor de tanque agitado, donde se han cargado 100 g de monómeros y los reactivos han sido utilizados sin purificación, lo que hace que el sistema sea más similar a las condiciones industriales. En este sentido, el sistema de reacción en masa se puede considerar una planta semi-piloto. Como medida de seguridad, dada la posible emisión de vapores procedentes de la oxidación del fluido calefactor empleado, la instalación se ha situado en una vitrina con sistema de extracción de gases.

Descripción del trabajo realizado

B.2.2. Instalación de modificación de compuestos orgánicos

Las modificaciones llevadas a cabo en la cumarina fueron realizadas en la instalación mostrada en la Figura B.2.

Figura B.2. Instalación para la modificación de compuestos orgánicos a presión atmosférica.

Esta instalación está compuesta de una placa calefactora y agitadora (Selecta-01), un soporte, regulador de temperatura, baño de aceite, condensador, línea de nitrógeno, y tapones dispuestos en el matraz con la función de poder introducir reactivos y mantener atmósfera inerte durante la reacción. Finalmente se disponía un globo lleno de nitrógeno en uno de los tapones con el fin de ir suministrando nitrógeno en el caso de que hubiese una fuga del nitrógeno existente dentro del matraz.

B.2.3. Instalación de reacción en medio supercrítico

Tanto las impregnaciones como las funcionalizaciones de polímero fueron llevadas a cabo en una instalación a escala de laboratorio en una planta de reacción en condiciones supercríticas diseñada y montada en el propio laboratorio. 53

Descripción del trabajo realizado

En la Figura B.3 se muestra un esquema del sistema descrito. Está construido con materiales diseñados para la alta presión de acero inoxidable 316 suministrados por:

Autoclave Engineers, Hoke, Swagelok y Parker. El sistema se divide en tres módulos: módulo de alimentación, módulo de reacción y módulo de despresurización.

E-1 V-1

P-1

C-1

Figura B.3. Esquema de la unidad de laboratorio para reacción en medio supercrítico.

El módulo de alimentación consta de:

- Dos intercambiadores de calor de carcasa y tubos calorifugados (E-1). Cada uno de ellos tiene 14 tubos con una longitud de 210 mm por tubo. El diámetro externo y espesor de los tubos es de 6,35 mm (1/4” O.D) y 0,0889 mm (0,035”), respectivamente. El modo de operación es flujo en contracorriente.

- Bomba Mil Royal D dosificadora de CO2 (P-1). Modelo MD140G4M500 / ND VV2 Z de desplazamiento positivo, 0,187/1,87 L h-1, 500 bar. Válvulas antirretorno, filtros y otros componentes.

Descripción del trabajo realizado

- Baño de refrigeración Selecta (-30°C) para enfriar la alimentación de

CO2 a la bomba y al cabezal de la misma, así como el reactor.

- Regulador de presión GO, tipo Back Pressure 66 (BPR). Máxima Presión de entrada 690 bar, temperatura máxima 298°C. Intervalo de regulación de presión de entrada de 0 a 414 bar. Cv: 0,04 (caudal bajo).

- Medidor másico de caudal Rheonik tipo Coriolis para CO2 líquido. Modelo RHM 03 GNT, tubos en paralelo, transmisor tipo RHE 08. Temperatura máxima 120°C, presión máxima 300 bar. Intervalo de medida de 154,14 a 4923,07 g/min.

- Manómetros (intervalo de medida de 0 a 400 bar), válvulas de media y alta presión y otros componentes.

Los principales componentes del módulo de reacción son:

- Reactor tipo tanque agitado de Berghof (C-1), 1200 mL, presión máxima 200 bar a 230 ºC. La agitación se realiza mediante motor eléctrico con control de velocidad, intervalo de operación de 50 a 2000 rpm. Agitador de paletas.

- Camisa calefactora eléctrica (máxima temperatura de operación 300ºC) asociada a un controlador de temperatura con entrada termopar de NiCr/Ni.

- Indicador de velocidad de agitación, manómetros (intervalo de medida de 0 a 400 bar), válvulas de alta presión y otros componentes.

El módulo de despresurización está compuesto de:

- Válvula micrométrica de regulación HOOKE (V-1). Presión máxima 344 bar a temperatura de 232°C y orificio 1,19 mm.

- Flujómetro vidrio PYREX 250 mL.

Descripción del trabajo realizado

- Hilo calefactor, manómetros (intervalo de medida de 0 a 400 bar), válvulas de alta presión y otros componentes.

B.2.4. Instalación de liberación in vitro

Los experimentos de liberación in vitro se llevaron a cabo en la instalación mostrada en la Figura B.4:

Figura B.4. Instalación de liberación in vitro.

La instalación consta de una placa agitadora (WH620), sobre la cual se sitúa un baño de metacrilato. Este baño es rellenado con agua termostatizada a 37 ºC y la agitación se ajustó a 100 rpm. Los botes donde se lleva a cabo la liberación son botes de cristal perfectamente sellados y cubiertos por papel de aluminio de la luz exterior.

El equipo incluye sistema de regulación del valore de agitación (SBS, modelo ANM-10009), así como regulación de temperatura mediante un termostato introducido en el agua del baño (SBS, modelo TI-02).

Descripción del trabajo realizado

B.3. Procedimientos

B.3.1. Operación de la instalación de polimerización

El procedimiento consiste en la carga del reactor con los monómeros, y el inicio de la calefacción con una agitación de 100 rpm hasta la temperatura de operación en presencia de atmósfera inerte (N2). Una vez alcanzada ésta se añaden la cantidad necesaria de catalizador y de iniciador, a través de una de las bocas de la tapa con unas jeringas de vidrio. La adición del sistema catalítico se considera tiempo cero de la reacción. A continuación, se procede a la toma de muestras a distintos intervalos de tiempo. Las muestras recogidas se enfrían rápidamente en hielo, se almacenan en viales de vidrio con atmósfera de nitrógeno y se guardan en el frigorífico para su posterior análisis.

B.3.2. Síntesis de 4-azidometil-7-metoxicumarina y 3- azidometilcumarina

a) El procedimiento seguido para la síntesis del compuesto 4-azidometil- 7-metoxicumarina fue realizado siguiendo el procedimiento descrito en bibliografía [2].

En un matraz de 1 litro se añade etilenglicol y se pone a calentar sobre una placa calefactora. Por otra parte, se pesa en un matraz de tres bocas (100 o 250 mL de capacidad) la cantidad de 4-bromometil-7- metoxicumarina que se quiere hacer reaccionar (0.5 g o 1 g) y la azida sódica en un exceso del 20% en masa (0.6 g o 1.2 g). Como disolvente se emplea una mezcla acetona/acetonitrilo (1:1) de manera que la concentración del 4-bromometil-7-metoxicumarina sea de 0,03M. De este modo, se forma una suspensión en la que se introduce un agitador

Descripción del trabajo realizado magnético. Finalmente, el matraz se sitúa en el baño caliente y se fija la velocidad de agitación en 500 rpm.

La temperatura de reacción es de 50ºC, ya que se mejora el rendimiento de la misma. Es importante controlar que ésta no supere los 50ºC debido al bajo punto de ebullición de la acetona. Además, para evitar sobrepresiones en el interior del recipiente y conseguir condiciones de reflujo, se coloca un pequeño condensador en una de las bocas del matraz permaneciendo las otras dos cerradas.

A las 24 horas del inicio de la reacción se añade un exceso de azida sódica (0.6 ó 1.2 g). Terminada la reacción a las 48 horas, se eliminan los disolventes en el rotavapor. A continuación, se emplea agua milli Q (0,5 - 1 L) para disolver la azida sódica y precipitar una mezcla de 4-bromometil-7- metoxicumarina (reactivo sin reaccionar) y 4-azidometil-7- metoxicumarina. Se filtra a vacío con un kitasato y se recupera el sólido. A éste, se le añade heptano caliente, para separar la mezcla de productos sólidos, en el que la 4-azidometil-7-metoxicumarina es totalmente insoluble y se filtra en caliente con un filtro de pliegues. Así, se elimina el 4- bromometil-7-metoxicumarina (soluble en heptano caliente).

El producto sintetizado se introduce en la estufa a 80ºC alrededor de 1 hora en un vidrio de reloj con el fin de eliminar los restos de heptano y agua. Una vez seco, se vuelve a pesar y se determina el rendimiento de la reacción.

b) El procedimiento de síntesis del producto 3-azidometilcumarina se realizó acorde al procedimiento descrito en bibliografía [3].

En un matraz de 250 ml se depositan 0,5 moles de salicilaldehído, 0,25 moles de CH3CH2COONa, 0,75 moles de (CH3CH2CO)2CO y 0,25 moles de trietilamina. La mezcla de reactivos se mantuvo en condiciones de reflujo

Descripción del trabajo realizado durante 8 horas y 50ºC. Posteriormente se procedió al enfriado del producto a temperatura ambiente, siendo lavada con NaOH al 40%,

NH4Cl, H2O y agua saturada de sal, respectivamente.

Una vez lavado el producto de reacción, se procedió a su secado con

Na2SO4, y recristalizada con etanol al 95%. A continuación se mezclan 30 mmol del producto obtenido a 120 ml de CH3CN en un matraz de 250 ml, a una temperatura de 85ºC. 30 mmoles de NBS y 250 mg de BPO se añaden posteriormente a la reacción, la cual se encuentra en modo reflujo durante 24 horas. Después de volver a enfriarla a temperatura ambiente, el producto de reacción se lavó utilizando los mismos reactivos descritos anteriormente y secado con Na2SO4. Finalmente el producto se recristaliza con tolueno.

En ambos procedimientos de modificación de compuestos orgánicos se utilizó atmósfera inerte con N2. El procedimiento para la intertización del medio de reacción consiste en conseguir la estanqueidad de la instalación mediante el uso de tapones, los cuales se pinchan con una aguja que mediante una bomba de vacío extrae el aire existente dentro. Una vez extraído se pincha un globo lleno de nitrógeno en el tapón, rellenando el hueco existente en la instalación con nitrógeno. Este proceso se repite 3 veces para asegurar la completa eliminación del aire dentro del matraz y su llenado con nitrógeno.

B.3.3. Impregnación de principios activos

Las impregnaciones llevadas a cabo a presión atmosférica se llevan a cabo mediante la preparación de una disolución del principio activo junto al disolvente correspondiente y el polímero. Una vez homogeneizada la mezcla, se deja evaporar el disolvente a temperatura atmosférica, obteniendo así el polímero sin disolvente. En caso de que la impregnación se lleve a cabo en medio supercrítico la preparación de la disolución se 59

Descripción del trabajo realizado lleva a cabo igual que la anteriormente descrita para posteriormente impregnarse y separarse el disolvente mediante el contacto con el CO2 a través del cual se obtiene el polímero impregnado del principio activo.

B.3.4. Operación de la instalación en medio supercrítico

El procedimiento de utilización de esta instalación es el siguiente: Una vez introducidos los reactivos en el reactor (C-1), se procede al sellado del mismo mediante la tapa dispuesta en su parte superior. Conocidas las condiciones de operación requeridas, se abren las válvulas que permiten el paso del CO2 de manera que mediante la ecuación de Bender ser puede conocer la temperatura a la que se necesita calentar el reactor para alcanzar la presión deseada. En caso de que el valor de presión fuese inferior al requerido, se acciona la bomba (P-1), la cual bombea el CO2 elevando el valor de presión hasta el propuesto inicialmente. Una vez alcanzadas las condiciones de operación se procede al cierre de válvulas y programación del tiempo de reacción, tiempo a partir del cual el controlador de temperatura deja de mantener la temperatura en valor constante y se procede a la etapa de despresurización.

En la etapa de despresurización se procede a la apertura de la válvula situada en la parte superior del reactor, estando la salida conectada a un caudalímetro que indica el caudal adoptado en la despresurización.

Mediante esta etapa de despresurización se consigue eliminar el CO2 presente en el reactor, obteniendo un producto libre de disolventes.

B.3.5. Operación de la instalación de liberación in vitro

Los experimentos de liberación de compuestos orgánicos con propiedades farmacéuticas se llevaron a cabo en una instalación con temperatura controlada a 37ºC mediante agua, y agitación constante de 100 rpm.

Descripción del trabajo realizado

Las muestras fueron introducidas en botes donde se mezclaban con una disolución tampón fosfato salina (PBS) de donde se iba cogiendo muestra de aproximadamente 10 ml periódicamente para analizar la cantidad de compuesto orgánico que había sido liberado, siendo posteriormente devuelta al bote de origen. La cantidad de compuesto orgánico liberado era determinada mediante UV, a la longitud de onda característica del compuesto a analizar. Los perfiles de liberación fueron calculados en términos de porcentaje de compuesto orgánico liberado con respecto al tiempo, siendo todos los experimentos realizados por duplicado.

B.3.6. Elaboración de la disolución tampón

La disolución fosfato salina (PBS) usada para las liberación in vitro fue preparada siguiendo el método de preparación dictado por Cold Spring Harbor Protocols (2006) y resumido en la Tabla B.4.

Tabla B.4. Tabla resumen de la preparación de la disolución PBS.

Producto Cantidad NaCl 8,0 g KCl 0,2 g

Na2HPO4 1,44 g

KH2PO4 0,24 g Agua desionizada 800 mL

Los reactivos anteriormente mencionados fueron mezclados en aproximadamente 800 ml de agua desionizada. Después de su total disolución, se ajustó el pH de la disolución a valor 7,4 con HCl 0,1N y fue añadida la cantidad de agua restante hasta 1 l.

Descripción del trabajo realizado

B.4. Bibliografía

 1. Stjerndahl, A., A.F. Wistrand, and A.C. Albertsson, Industrial utilization of tin-initiated resorbable polymers: Synthesis on a large scale with a low amount of initiator residue. Biomacromolecules, 2007. 8(3): p. 937-940.  2. Velencoso, M.M., et al., Click-ligation of coumarin to polyether polyols for polyurethane foams. Polymer International, 2013. 62(5): p. 783-790.  3. Tan, Y., et al., Novel one-pot asymmetric cascade approach toward densely substituted enantioenriched α-methylene-γ- lactams. Tetrahedron Letters, 2014. 55(44): p. 6105-6108.

Descripción del trabajo realizado

C. Presentación de resultados.

En este apartado se presentan los diferentes aspectos tratados en esta Tesis Doctoral, así como las vías elegidas para llevar a cabo la consecución de estos objetivos. Adicionalmente, se representa una esquematización de todos los aspectos tratados en este trabajo de investigación (Figura 1).

Liberación Liberación in vitro in vivo

Impregnaciones

scCO2

Compuestos orgánicos Polímeros con propiedades biodegradables farmacéuticas

scCO2 Compuestos Catalizador en Catalizador en orgánicos fase homogénea fase heterogénea modificados

Química click

Liberación in vitro

Figura 1. Esquema de los temas tratados en cada uno de los capítulos de resultados.

Descripción del trabajo realizado

- Así en el capítulo 1 se llevó a cabo el estudio de la viabilidad de la tecnología supercrítica para la impregnación de compuestos orgánicos con propiedades farmacéuticas en polímeros biodegradables, siendo estudiada su liberación in vitro. El capítulo que engloba estos resultados está recogido en un artículo publicado, el cual se titula “Improvement of PLGA loading and release of curcumin by supercritical technology”. - En el capítulo 2, una vez comprobada la viabilidad de la tecnología supercrítica en el uso de compuestos con propiedades farmacéuticas, se estudió la funcionalización mediante química click de polímeros biodegradables mediante el uso de un catalizador en fase homogénea en medio supercrítico. El capítulo que engloba estos resultados está recogido en un artículo publicado, el cual se titula “Functionalization and optimization

of PLA with coumarin via click chemistry in supercritical CO2”. - El uso de catalizador en fase heterogénea como hilos de cobre se trata en el capítulo 3, en este apartado se estudian distintos parámetros para obtener las condiciones operacionales óptimas para la funcionalización del polímero. Posteriormente se realiza un estudio económico donde se determina la alternativa más económica para llevar a cabo el proceso de producción mediante química click. El capítulo que engloba estos resultados está recogido en dos artículos, uno publicado y otro mandado a publicar, los cuales se titulan “Copper wire as a clean and

efficient catalyst for click chemistry in supercritical CO2” y “Feasibility of copper wire as heterogeneous catalyst in click chemistry at supercritical conditions”, respectivamente. - En el capítulo 4 se estudia cuál es la estructura óptima para llevar a cabo la funcionalización de cumarina sin afectar a su

Descripción del trabajo realizado

rendimiento por impedimentos estéricos. El capítulo que engloba estos resultados se titula “Application of click chemistry

in scCO2 to PLA-derived Coumarin-release systems”. - El último capítulo de resultados en esta tesis doctoral engloba la comparación entre los resultados obtenidos en liberación in vitro cuando las muestras de cumarina son impregnadas y funcionalizadas en medio supercrítico. El capítulo que engloba estos resultados se titula “Drug release profile of click products

in comparison with impregnations carried out in scCO2”. - Una vez finalizados los capítulos de resultados se presenta un anexo donde se presentan los resultados obtenidos en colaboración con otro grupo de investigación mediante el cual se llevó a cabo la liberación in vivo de curcumina. Los resultados obtenidos en este anexo están recogidos en un artículo mandado a publicar titulado “Impregnation of Curcumin into Biodegradable Support (poly-lactic-coglycolic acid, PLGA), to Transfer its Well Known in vitro Effect to an in vivo Prostate Cancel Model”.

Supercritical CO2 feasibility for the impregnation and release of curcumin in Chapter

PLGA 1

1.1. Introducción 1.2. Experimental 1.2.1. Materials 1.2.2. PLGA copolymerization set-up 1.2.3. Supercritial carbon dioxide impregnation set-up 1.2.4. Study of curcumin solubility 1.2.5. Impregnation of PLGA with curcumin using organic solvents 1.2.6. Impregnation of PLGA using supercritical carbon dioxide (scCO2) 1.2.7. In vitro release study 1.2.8. Characterization measurements 1.3. Results and discussion 1.3.1. Synthesis of PLGA 1.3.2. PLGA curcumin impregnation 1.3.3. Bulk impregnation of PLGA with curcumin 1.3.4. ScCO2 impregnation of PLGA with curcumin 1.3.5. In vitro release 1.4. Conclusions 1.5. Bibliography

A continuación se muestran de forma esquematizada los contenidos del capítulo 1 de resultados.

Liberación in vivo

scCO2

scCO2 Compuestos Catalizador en Catalizador en orgánicos fase homogénea fase heterogénea modificados

Química click

Liberación in vitro

Abstract

he study of supercritical technology feasibility for the improvement of the loading and release of curcumin in PLGA was carried out in T this chapter. The impregnation of curcumin in PLGA was made at atmospheric and supercritical conditions in order to compare the results obtained in both technologies.

High-pressure impregnation performances practically doubled bulk results, leading to values up to 86.1% obtaining a product free of solvent in a sponge form.

The release kinetics of the samples constantly delivered more than 90% of curcumin between 9 and 12 days. Compared to other technologies our samples improved significantly the combined loading and release characteristics, indicating that supercritical technology is an interesting alternative for curcumin loading and controlled delivery in medical applications. Encouraging results are expected from in vivo experiments carried out where at first instance could be observed an important reduction of tumor volume in mice, being PLGA-impregnated curcumin significantly more active respect to oral curcumin.

Chapter 1

1.1. Introduction

Biodegradable polymers are defined as any substance, material or combination of both, that can be used as a part of a treatment, replacement of tissues, organs or any organism functions. An important use of this type of polymers is the medical application, where aliphatic polyesters, and those derived from lactic acid (LA) and glycolic acid (GA) in particular appear to be the most promising because of their excellent biocompatibility and variable degradability[1].

Between all the currently polyesters, polylactide acid (PLA) and glycolic acid (PGA) are the most interesting due to they can be used in for a wide range of applications [2]. However, PLA with PGA present several limitations for medical applications when they are used separately. Two of limitations solved by copolymerization are term stability or degradation time [3]. These limitations are solved by copolymerization of lactide and glycolide (PLGA). PLGA is one of the polymers with higher potential as a drug delivery carrier because of its tuneable properties as degradation, processing and mechanical strength [4]. The main advantage of this polymer is the property of varying degradation rate depending on the ratio of monomers (PLA/PGA) used to carry out the polymerization. According to previous research, a molar composition of polylactide (PLA) in PLGA between 75 and 100% provide a variation of copolymer half-life from 2 weeks to 6 months[5].

Nowadays several drugs are studied to carry out drug delivery in polymers [6-8]. One of the drugs whose importance is increasing recently is curcumin [9]. It is a yellowish orange colour substance found in the rhizome of Curcuma longa. This herb is composed of three different species

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA called curcuminoids in different proportions: curcumin (77%), demethoxycurcumin (17%) and bis-demethoxycurcumin (6%) [10].

At first, curcumin was used as colouring agent and as a food additive, but applications for this drug have changed to pharmaceutical uses due to the excellent results obtained in several studies[11, 12]. Properties as antioxidant, anti-inflammatory, antimicrobial and anticarcinogenic make curcumin an excellent candidate to perform the polymer impregnations [13, 14].

There are some different alternatives to carry out supercritical polymer impregnations like drug and dye impregnations[15]. In this work, low pressure and supercritical carbon dioxide (scCO2) were chosen to impregnate curcumin in the PLGA previously synthesized. Supercritical fluid technology was tested for curcumin impregnation into PLGA like an interesting alternative because of its excellent properties, like mass transfer, lack of residual solvent in the products or plasticization of polymers [16-18]. These properties are crucial for pharmaceutical applications, because the solvent must be removed completely from the polymer. Plasticization of polymers is another required property to increase the performance of impregnation, because scCO2 swells the polymer achieving to increase its free volume, so the amount of drug loaded in the polymer is higher than if the impregnation had been carried out in bulk [19-21]. Zabihi et al. [17, 18] used supercritical anti-solvent (SAS) process to impregnate PLGA carriers with curcumin. The aforementioned process was conducted on the relatively low pressures (9- 9.6 MPa) and temperatures of 33-35 °C. Obtained PLGA nano-sized particles contained up to 45% of curcumin. The same authors [18] also

Chapter 1 demonstrated a controlled release of curcumin from PLGA carried in PBS at 37 °C for 10 h without bursting effect.

The aim of this study was to improve loading the PLGA with curcumin and its prolonged release in order to widen pharmaceutical application of the curcumin-loaded PLGA carriers. For that purpose, polymerization of the PLGA support and influence of PLA molar composition in PLGA on the curcumin loading using supercritical CO2 was studied. Once the PLGA is synthetized, a study of curcumin impregnation was carried out varying the pressure and solvent to get to know the best conditions in which the drug is impregnated[17]. Once the best solvent and molar composition of PLGA were determined, an in vitro drug delivery test was performed to study the kinetic release of curcumin from PLGA in a phosphate-buffered saline solution (PBS).

1.2. Experimental

1.2.1. Materials

Glycolide (G) (1,4-dioxane-2,5-dione; Purac Biochem bv, The Netherlands) and D,L-lactide (L) (3,6-dimethyl-1,4-dioxane-2,5-dione; Purac Biochem bv, Netherlands) both with a purity higher than 99.5%, Tetrahydrofuran (THF) (HPLC grade; SDS S.A., Spain), carbon dioxide (Carburos metálicos, S.A., Spain) with a purity of 99,5% Stannous octoate (tin(II) 2-ethylhexanoate (Sigma-Aldrich Química, S.A., Spain) were used for polymer synthesis. Methanol anhydre (MeOH) (SDS S.A., Spain) with purity higher than 99.85%, ethanol (Panreac Química S.L.U., Spain) with purity higher than 99,60%, acetic acid (Panreac Química S.L.U., Spain) with purity higher than 99,4%, ethyl Lactate, butyl lactate and curcumin (Sigma-Aldrich Química, S.A., Spain) with analytical grade were used for study of curcumin solubility. 72

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

1.2.2. PLGA copolymerization setup

Experiments were carried out in a set-up consisting on a glass stirred- tank reactor with a volume of 500 ml and put into an inert atmosphere of nitrogen. The copolymerization process consists on placing monomers in a bulk reactor with catalyst and initiator at high temperature to form a viscous growing prepolymer, which after a 4 h reaction led to a solid polymer finished when cooled to room temperature. Three different D,L- Lactide:Glycolide (LA:GA) molar ratios of 20:80, 50:50 and 80:20 were tested to build the polymer support in which curcumin will be impregnated in the second step. Temperature was controlled by means of a temperature controller with a sensor inside the reaction melted mixture. Operational conditions for polymerization were: atmospheric pressure (1 atm), temperature 130 ºC, agitation of 100 rpm and a total mass of 100 g of monomers in the reactor. Two mass ratios (90:1 and 1:2) of monomer- catalyst (Stannous octoate) to catalyst-initiator were used for polymerization of PLGA. The total time of polymerization was 4 hours and samples were taken every 30 minutes to analyse the evolution of synthetized polymer during the reaction.

1.2.3. Supercritical carbon dioxide impregnation setup

Experiments were carried out in a lab scale installation divided into three modules: feed system, reactor, and depressurization line, respectively.

Feed system consisted of two heat exchangers, one positive displacement type pump for liquid CO2, (model MD140G4M500 / ND VV2

Z) , refrigerator unit for cooling feed and CO2 pump head; and back pressure regulator (GO) for controlling pressure in reactor.

Chapter 1

Stirring tank reactor had a nominal volume of 1200 ml, and a maximum pressure of 200 bar at 230 ºC and it was equipped with a magnetically coupled mechanical stirrer. This reactor is also equipped by a heater with temperature control by a PID, and for cooling a serpentine refrigerator inside was used.

Depressurization line was heated with an electrical heating tape and two pressure regulators with a valve to prevent freezing of CO2 by Joule- Thompson effect during the depressurization stage.

The procedure in supercritical carbon dioxide is composed of four steps: At first, the sample is introduced in the vessel and reactor is closed to avoid CO2 leakage, secondly the reactor is loaded with the required pressure for each experiment, and heater is connected until the pressure and temperature values are achieved. After exposing the polymer sample to scCO2 for a given time, system is depressurized, where the CO2 is removed from the reactor. Finally, once reactor is depressurized, it is opened and the sample is taken. Further information about experimental set up can be found in reference[22].

1.2.4. Study of curcumin solubility

The solubility values were obtained experimentally. Saturated solutions of curcumin were prepared to determine the maximum amount of drug that is solubilized in each solvent. This process consists on the solubilisation of maximum amount of curcumin in the correspondent solvent until we obtain a saturated solution.

1.2.5. Impregnation of PLGA with curcumin using organic solvents

Bulk impregnation consisted on 1500 solid PLGA mg mixed with a solution of saturated curcumin for 24 h. Once the polymer was

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA impregnated, the solvent was separated from the mixture PLGA-curcumin through evaporation at room temperature.

1.2.6 Impregnation of PLGA using supercritical carbon dioxide (scCO2)

The quantity of polymer used was 1500 mg as in impregnations carried out at atmospheric pressure. The procedure for high pressure impregnations was the same that those described for impregnation in the organic solvent saturated solutions of curcumin [22]. Saturated solutions of the curcumin in acetone, ethanol and methanol were placed in the high pressure vessel and the CO2 was charged until 150 bar and 45º C were reached. Contact time was different depending on the solvent used to ensure the total solubilization of solvent in scCO2. Finally, reactor is depressurized at the rate of 1.67 bar/min as was explained previously.

1.2.7. In vitro release study

Polymer impregnated with curcumin were suspended in a phosphate saline solution (PBS) 0.1 M (pH 7.4, 1 M), placed in the middle of a 100 mL flask hermetically closed and preserved from light, stirred at 100 rpm, and incubated in a shaking water bath at 37 °C. 5 ml solution was periodically removed from the flask in order to measure by UV spectrophotometry the quantity of curcumin released.

1.2.8. Characterization measurements

Infrared (IR) spectra of synthetized and impregnated polymers were obtained with a spectrophotometer Varian model 640-IR in range from 4000 to 400 cm-1, with a resolution of 4.0 cm-1 and 64 scanning, using the software Varian Resolution in ATR mode.

The amount of drug released was determined using a UV–Vis apparatus (Shimadzu UV-1603, Germany) with a spectral range from 190 to 1100 nm, 75

Chapter 1 halogen and deuterium lamps and a silicon photo diode detector. It was provided with the software UVPC Personal Spectroscopy Software, Version 3.6.

Molecular weight of polymers was determined by gel permeation chromatography on GPC cromatrograph (Waters, Spain) model 717. It is equipped by one column Viscotek, whose interval of molecular weight is 500-2000 g/mol, two peristaltic pumps, electric oven and a refractive index detector. The eluent used was tetrahydrofuran (THF) at 35ºC (flow: 1 mL·min-1; injection volume of 100 µL. Samples were dissolved in THF at a concentration of 1.5 mg·mL-1 and filtered before injection.

PLGA and mixture PLGA-curcumin compositions were determined by thermogravimetric analysis (TGA apparatus, model SDT 2960). In this analysis is possible to determine the amount of solvent, residue and drug impregnated which is present in every sample. The amount of each one is calculated by weight (%) decrease in every sample with respect to temperature.

The calorimetric analysis was determined by DSC model Q100, equipped by a refrigeration system (TA Instruments). Samples of 3-10 mg were prepared in aluminium capsules. This analysis was carried out in 3 stages according to is shown in Table 1. The sample was heated until 280 °C with a ramp of 10 °C/min, followed by a cooling until −50 °C with de same ramp and, finally, was heated until 280 °C again with the same ramp.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

1.3. Results and discussion

The main objective of this work is to determine the feasibility of supercritical technology for improvement of the curcumin loading in PLGA and its prolonged release for pharmaceutical applications. For that purpose, the synthesis of the PLGA support with determining the effect of PLGA molar composition was first studied. Once PLGA polymers were synthetized, impregnations of PLGA using organic solvents or scCO2 were studied. Finally, a study of drug delivery in vitro corresponding to the samples impregnated previously in supercritical CO2 was performed. Samples showing better results regarding curcumin loading and release time were compared to other bibliography studies.

All the experiments were carried out by duplicate with an error of 2.67%.

1.3.1. Synthesis of PLGA

In order to determine the best PLGA composition for curcumin impregnation, in this study three different (D,L-polylactide:glycolide) molar ratios were synthetized according to a procedure previously described in bibliography by this research group [20, 21]. For a lower ratio of lactide (20:80), a solid unreacting block was formed in the first minutes of reaction, so that this PLGA relation monomer was excluded for the impregnation of curcumin. For this reason, only PLGA with LA:GA molar composition of 50:50 and 80:20 were considered in the rest of the work. Obtained PLGA samples were characterized according to techniques described in experimental section. Table 1.1 shows the results for the molecular weight for the selected polymers using GPC analysis. Bulk polymers showed a molecular weight distribution in which it was achieved the theorical molecular weight Mw - 77

Chapter 1 weighed contribution of each monomer- for all PLGA relations synthetized during reaction. This fact indicates that the synthesis of PLGA support was performed as expected. Due to the polidispersity, the averaged molecular weight for the polymer, Mn, was lower to Mw as previosly observed [23]. Main functional groups of PLGA were analysed in IR spectra, where different groups were observed : C=O (1760 cm-1), CO (1000 and 1215 cm-1),

CH- (671 cm-1), -CH3 (1520 cm-1) and CH (2353 cm-1)[24]. Figure 1.1 shows the polymer IR spectra, where the main PLGA functional groups are shown. Table 1.1. Molecular weight and Tg of PLGA in bulk polymerization.

Molar Mw Sample composition Mn (g/mol) Tg (ºC) (g/mol) (LA:GA) PLGA80:20 80:20 14425 6875 51.90

PLGA50:50 50:50 12811 7165 48.59

Evolution of glass transition temperature (Tg) was measured with DSC, where it was observed in Table 1 an increasing of Tg when increasing LA proportion from PLGA50:50 to PLGA80:20. This evidence is due to the higher LA molecular weigh as observed previously in GPC analysis [25].

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

PLGA (50:50) PLGA (80:20)

1000-

1760 1215

2353 Transmitance (u.a)

4000 3600 3200 2800 2400 2000 1600 1200 800 400

Wave number (cm-1)

Figure 1.1. FTIR spectra of PLGA obtained in bulk polymerization.

1.3.2. PLGA curcumin impregnation

Curcumin impregnations in PLGA were performed using bulk and scCO2. A previous solubility study was performed to select solvents with both, higher curcumine solubility and affinity to scCO2. After solubility test, impregnation process for both alternatives was carried out. In this last part, the effect of supercritical technology in the polymer impregnation was studied to determine those conditions leading to the maximum quantity of loaded drug.

The selection of a solvent with a high solubility value with curcumin is one of the main requests to carry out the impregnations. Table 1.2 shows the solubility results for the different solvents chosen in this study.

Chapter 1

Table 1.2. Curcumin solubility in different solvents.

Solvent Solvent purity Solvent solubility CO2 affinity (%) (mg/ml) Acetone 99.80 78.80 High Acetic acid 99.50 5.10 High Water 100.00 0.60 Low Ethanol 99.70 6.20 High Ethyl lactate 98.00 17.40 Low Butyl lactate 98.00 15.00 Low Methanol 99.95 8.05 High

According to Table 1.2 results, the three solvents chosen to carry out the impregnations were acetone, ethanol and methanol. In spite of higher values of solubility in ethyl lactate and butyl lactate, they present low affinity to CO2. This fact will mean higher residual content in the loaded polymer in the high pressure processing because of the lower solubility in

CO2. For this reason, they were discarded.

1.3.3. Bulk impregnation of PLGA with curcumin Once solubility study was carried out, a relation of PLA bulk impregnations with curcumin are shown.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

Table 1.3. Relation of PLGA bulk impregnations carried out at atmospheric pressure.

Sample Polymer Solvent Curcumin Impregnatio Tg (ºC) (mg) n efficiency (%) I-01 PLGA 80:20 Acetone 170 38.64 37.56 (14.38) I-02 PLGA 80:20 Methanol 40 32.50 31.24 (20.70) I-03 PLGA 80:20 Ethanol 31 31.93 31.03 (20.91) I-04 PLGA 50:50 Acetone 170 35.04 35.06 (13.53) I-05 PLGA 50:50 Methanol 40 28.86 30.85 (17.74) I-06 PLGA 50:50 Ethanol 31 27.94 29.79 (18.80)

As can be observed in Table 1.3, six impregnations were made, using the three different solvents and two PLGA relations selected. Column #4 gives the total amount of curcumin contained in the initial solution, of which only a part will be finally impregnated in the solid polymeric matrix. Impregnation performance was obtained by thermogravimetric analysis (TGA), which determined the amount of drug impregnated in the polymeric matrix. In the last column two values of Tg are given: Tg of non- impregnated PLGA and values in parenthesis correspond to ∆Tg (Tg non- impregnated PLGA – Tg impregnated PLGA).

According to the results in Table 1.3, there is higher impregnation efficiency for acetone due to its higher solubility (Table 1.2). In addition, it can be observed that molar ratio of PLGA has also influence on the impregnation efficiency. A higher molar fraction of LA in PLGA80:20 favoured the impregnation of curcumin using tested organic solvents (Table 1.3). This fact is related to the molecular weight of the polymer, as it was previously described [26].

Chapter 1

The success of impregnation was confirmed by FTIR. The IR spectra exhibit characteristics bands related to curcumin and bands related to PLGA. As example, Figure 1.2 represents IR spectra corresponding to PLGA80:20 impregnated in bulk using acetone as solvent. The most characteristic absorbance band arising from the PLGA impregnated is the peak at 1760 cm-1 characteristic of the carbonyl group (C=O), whose size is higher than the absorbance band of pure polymer (Figure 1.2). From 3000 cm-1 there are observed some tiny absorbance bands that correspond to the existence of solvent in the polymeric matrix which was not totally removed in the evaporation process, what represents a limitation for medical applications of the samples. Similar results were obtained for the rest of impregnations carried out at atmospheric pressure.

Glass transition temperature (Tg) of curcumin-loaded PLGA samples determined by DSC analysis are given in Table 1.3. As can be seen, curcumin-loaded PLGA samples obtained using various solvents had different Tg values. All the Tg values of curcumin-loaded PLGA samples were lower than those corresponding to unloaded samples (Table 1.3). The lower values indicate that a significant amount of solvent is placed in the polymeric matrix. The Tg reduction with respect Table 1.3 (values in parenthesis) is proportional to the residual solvent content in the polymer, not removed in the evaporation step. It can be concluded that acetone produced the lowest amount of residual solvent in the loaded sample, while in the case of methanol and ethanol a higher percentage of solvent is kept in the polymeric matrix. For that reason, acetone is the most suitable solvent for loading curcumin in PLGA 80:20.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

C=O

Figure 1.2. IR spectra correspondent to PLGA 80:20 impregnated in bulk using acetone as solvent.

1.3.4. ScCO2 impregnation of PLGA with curcumin

Next task in this work studied the impregnation of PLGA with curcumin using scCO2 as an impregnation medium, comparing these results with organic solvent impregnation and other methods for curcumin loading into solid carriers described in the literature. This study was aimed to justify use of supercritical impregnation for fabrication of system for controlled and prolong release of curcumin for medical applications. To determine whether the advantages of this technology can improve the characteristics of these polymers, curcumin impregnation was studied at high pressure. The value of pressure chosen for impregnation in scCO2 (150 bar) was the same used in a previous study about PLGA polymerization using supercritical carbon dioxide [20, 21].

To use all the scCO2 advantages a polymer temperature close to Tg is required. The scCO2 acts as a molecular lubricant, decreasing the Tg of the PLGA[15]. High pressure Polymer Tg was calculated using Chow equation 83

(1) Chapter 1

(1) and values of CO2 solubility in PLGA obtained from bibliography [27, 28].

푇푔 ln( ) = 훽 ∙ [휃 ∙ 푙푛휃 + (1 − 휃) ∙ ln(1 − 휃)] 푇푔,0

where  and  are obtained from equations (2) and (3), respectively

푀 휔 휃 = 푚 1 (2) 푧푀푑 1−휔1

푧푅 훽 = (3) 푀푢∆퐶푝푝

In these equations Tg is the glass transition temperature of the polymer containing a weight fraction, ω1, of the dissolved component; Tg,0 is the glass transition temperature of the pure polymer; Mm is the molar mass of the polymer repeat unit; Md is the molar mass of the dissolved component;

R is the gas constant; ΔCpp is the excess transition isobaric specific heat of the pure polymer, and z is the lattice coordination number. In this study z=1; ΔCpp=0.336 J/(g K); and Tg,0= 58 ºC.

For comparison purposes, the same number of impregnations was carried out in scCO2 as in previous bulk tests. Results are shown in Table 1.4.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

Table 1.4. High pressure CO2 impregnations of PLGA

Impregnatio Loading Sample Polymer Solvent n efficiency Tg (ºC) time (h) (%) I-07 PLGA 80:20 Acetone 8 84.3 50.78 (1.16) I-08 PLGA 80:20 Methanol 16 66.5 40.57 (11.37) I-09 PLGA 80:20 Ethanol 16 65.8 40.32 (11.62) I-10 PLGA 50:50 Acetone 8 58.2 47.01 (1.58) I-11 PLGA 50:50 Methanol 16 50.1 39.86 (8.73) I-12 PLGA 50:50 Ethanol 16 51.3 38.79 (9.80) I-13 PLGA 80:20 Acetone 8 86.1 50.70 (1.35)

Table 1.4 shows that impregnations were accomplished in a maximum time of 16 hours when ethanol and methanol were used as solvents, being reduced to 8 hours . This time was selected according to solubility values of theses solvents in scCO2 which ensure their total solubilization in scCO2. This fact supposes an operational advantage with respect to impregnations carried out at low pressure, for which impregnation time was 24 hours.

Glass transition and amount of curcumin impregnated in the polymer were analysed with TGA and DSC. As it can be observed in Table 4, the samples loaded with supercritical CO2 had an impregnation yield almost two times higher than those obtained in bulk conditions. The highest impregnation efficiency was evidenced for the PLGA80:20 independently of solvent used for first time, whereby use of the acetone yield the highest curcumin loading (84.3-86.1%). This curcumin loading is up to two times higher than reported for PLGA nano-sized PLGA particles obtained using

Chapter 1

SAS process with scCO2 used as a solvent at 9-9.6 MPa and 33-35 °C (up to 44.73%) [18].

DSC analysis showed low residual solvent content of samples (wt%).

Curcumin-loaded PLGA obtained using scCO2 with acetone solution of curcumin had similar Tg to non-impregnated PLGA what practically means that the complete solvent was removed from the polymeric matrix. This is advantageous for the loaded samples could be used for prevention and treatment of cancer in market medical formulations without additional processing, avoiding any further concentration or solvent elimination step [19].

The IR spectra analysis of curcumin-loaded PLGA80:20 using scCO2 and the acetone solution of curcumin presents the same functional groups as observed for the same sample impregnated at atmospheric pressure previously. This evidence indicates that impregnation was carried out satisfactorily again. This finding is shown in Figure 1.3, which compares the IR spectra correspondent to PLGA80:20 impregnated at high pressure using acetone (a) to PLGA obtained in bulk conditions (b). Similar results were obtained for the rest of impregnations.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

b) Transmittance (u.a.) Transmittance

4000 3600 3200 2800 2400 2000 1600 1200 800 400 Wavernumber (cm-1)

Figure 3. Comparison of IR spectra in transmittance correspondent to a)

PLGA80:20 impregnated in scCO2 using acetone solution of curcumin; b) PLGA80:20 impregnated with curcumin dissolved in acetone at atmospheric condition.

1.3.5. In vitro release

Once the best molar ratio of LA:GA co-monomers, solvent and procedure were chosen, the following task studied the curcumin release profile from the polymer. Two experiments were performed for this study varying the quantity of curcumin impregnated in the polymer as it can be seen in Table 1.5.

Table 1.5. Drug delivery study of samples I-07 and I-13.

Initial Polymer Impregnation kdegr Impregnation curcumin (mg) efficiency (%) (cm-1) (mg) I-07 1500 170 84.3 0.114(0.0026) I-13 1500 270 86.1 0.151(0.0015)

Chapter 1

The amount of impregnated curcumin was determined by UV spectra at 421cm-1. A calibration line was used to get to know the concentration of curcumin in the sample impregnated according to the value of absorbance registered. Release profiles were calculated in terms of the cumulative release percentage of curcumin.

According to bibliography several theoretical mechanisms for controlled release of drug from biocompatible polymers have been described [29]. These mechanisms are composed up to 3 steps, where the first one corresponds to the initial burst of drug release of the most accessible drug, generally located in the surface of the particles and controlled by the diffusion in the film. The second step is controlled by the internal diffusion into the most tortuous or narrow pores. The last step is the step controlled by the degradation of the polymer. Due to the homogeneous distribution of the drug into the polymer matrix consequence of the easy access using CO2 in addition to the tailored biodegradability of the PLGA, this study showed only one long constant-high release stage (Figure 1.4). This stage corresponds to a degradation of the complex polymer-drug in the PBS. Compared to a classical 3 steps profile, Figure 1.4 (c), associated to a heterogeneous distribution of the drug mainly located in the particles surface, supercritical loaded samples showed a more interesting release profile for medical applications. According to the results obtained in this work, a minimum of 9 and 12 days are necessary to constant release more than 90% of drug impregnated in the polymer. Release time depended on the quantity of curcumin impregnated in the polymer, being necessary a higher number of days in the experiment where 270 mg of curcumin was used.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

100 90 80 70 60 50 40 30 20 10

Cumulative percentagerelease (%) 0 0 1 2 3 4 5 6 7 8 9 10 11 Time (day)

Figure 1.4. Drug release profile of curcumin in PLGA at 37º C of a) (●) I-07 using 170 mg of curcumin; b) (▲) I-13 using 270 mg of curcumin; c) (—) Typical profile correspondent to 3 steps release.

Release kinetics was modelled using equation (4), where M0 and Mf represent the total mass at the beginning of the release and at the end of the experiment respectively, R0 is the initial radius of the spherical foam

(0.2 cm) and kdegr is the pseudo-first kinetic constant of degradation for the PLGA foam.

1 푀 푘 ( 푓)3 = 1 − 푑푒푔푟 (4) 푀0 푅0

Using the equation 4 the constants correspondent to degradation stage for both experiments can be determined, as indicated Table 1.5. There were obtained a value of 0.114 cm for I-07 and 0.151 cm for I-13 (values in parenthesis correspond to standard deviation), respectively, in a drug 89

Chapter 1 release profile which corresponds to the monophasic release from a single homogeneous phase [30]. This trend is expected to be the desirable behaviour for pharmacological applications of constant, durable and high dosage release.

Finally, our results were compared in Table 6 to other works where different techniques are used to improve curcumin loading and release [31- 33].

Table 1.6. Proposed methods for increased curcumin loading and release.

Release Curcumin Method Reference time (days) loaded (mg) Encapsulation 1 348.75 Sherbini, et al[31] Vapor induced phase 1.5 0.45 Bajpai et al[32] inversion Coating stent 18 0.16 Pan et al[33] [18] The Journal of Supercritical Supercritical Fluids Anti-Solvent 10 h 90 Volume 89, May 2014, (SAS) process Pages 99-105 This work 10 170-270 -

As can be observed in Table 6, supercritical technology allows an important improvement of both, drug loading and time of release. Comparing this work with bibliography [18], we can observe drug release time is up to 60 times higher, what supposes a longer and more controlled release profile. This fact suggests that supercritical technology is an

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA interesting alternative for curcumin loading and controlled delivery in medical applications.

Impregnation procedure of curcumin with PLGA in scCO2 shown in this chapter was chosen for in vivo experiments with mice receiving prostate cancer xenografts. The way of administration of curcumin was as a powder directly in a subcutaneous sutured pocket. This preparation prolonged disease free survival and reduce tumour growth. No immunologic side effects seemed to be present during the experimental time. After 19 days tumour volumes, indeed, were strongly reduced when compared with controls and with commercial orally administered encapsulated curcumin. These experiments were carried out in collaboration with Department of Biotechnological and Applied Clinical Sciences, University of L'Aquila (see Annex 1).

1.4. Conclusions

This work justifies use of suggested scCO2-assisted process for an increased curcumin loading into PLGA carrier and its prolonged release (9- 12 days). Compared to classical bulk impregnations at atmospheric conditions , by immersion of PLGA in curcumin solutions in organic solvents, impregnations aided by use of scCO2 at 150 bar and 45ºC yielded up to 2 times higher curcumin loading in PLGA. The best results regarding curcumin loading were evidenced when PLGA80:20 was used as a carrier and acetone as a solvent. Final PLGA80:20 foams loaded with curcumin dissolved in acetone by using scCO2 had no residual solvents. This makes the presented technology and the obtained promising for use in commercial formulations without need for further purification or concentration step. In addition, these samples showed a single long constant-high release of curcumin which is the most desired profile for medical applications. 91

Chapter 1

Comparison to other impregnation technologies showed that our samples had higher curcumin loading (50.1-86.1%) and had more advantageous release profile for medical applications like for example blood stream. Encouraging results were obtained in in vivo experiments in mice where could be observed an important reduction of tumour volume as a consequence of curcumin released in about 19 days.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

1.5. Bibliography

1. Middleton, J.C. and A.J. Tipton, Synthetic biodegradable polymers as orthopedic devices. Biomaterials, 2000. 21(23): p. 2335-2346. 2. Athanasiou, K.A., et al., Orthopaedic applications for PLA-PGA biodegradable polymers. Arthroscopy: The Journal of Arthroscopic & Related Surgery, 1998. 14(7): p. 726-737. 3. Ulery, B.D., L.S. Nair, and C.T. Laurencin, Biomedical Applications of Biodegradable Polymers. Journal of polymer science. Part B, Polymer physics, 2011. 49(12): p. 832-864. 4. Auras, R., B. Harte, and S. Selke, An overview of polylactides as packaging materials. Macromolecular Bioscience, 2004. 4(9): p. 835-864. 5. Miller, R.A., J.M. Brady, and D.E. Cutright, Degradation rates of oral resorbable implants (polylactates and polyglycolates): Rate modification with changes in PLA/PGA copolymer ratios. Journal of Biomedical Materials Research, 1977. 11(5): p. 711-719. 6. Liechty, W.B., et al., Polymers for Drug Delivery Systems. Annual review of chemical and biomolecular engineering, 2010. 1: p. 149-173. 7. Priya James, H., et al., Smart polymers for the controlled delivery of drugs – a concise overview. Acta Pharmaceutica Sinica B, 2014. 4(2): p. 120-127. 8. Srivastava, A., et al., Polymers in Drug Delivery. Journal of Biosciences and Medicines, 2016. Vol.04No.01: p. 16. 9. Das, R.K., N. Kasoju, and U. Bora, Encapsulation of curcumin in alginate-chitosan-pluronic composite nanoparticles for delivery to cancer cells. Nanomedicine: Nanotechnology, Biology and Medicine. 6(1): p. 153- 160.

Chapter 1

10. Kunnumakkara, A.B., P. Anand, and B.B. Aggarwal, Curcumin inhibits proliferation, invasion, angiogenesis and metastasis of different cancers through interaction with multiple cell signaling proteins. Cancer Letters. 269(2): p. 199-225. 11. Gupta, S.C., S. Patchva, and B.B. Aggarwal, Therapeutic Roles of Curcumin: Lessons Learned from Clinical Trials. The AAPS Journal, 2013. 15(1): p. 195-218. 12. Nanjwade, B., et al., Curcumin: Nutraceutical and Pharmaceutical Applications. Vol. 1. 2015. 17-26. 13. Johnson, J.J. and H. Mukhtar, Curcumin for chemoprevention of colon cancer. Cancer Letters. 255(2): p. 170-181. 14. Goel, A., A.B. Kunnumakkara, and B.B. Aggarwal, Curcumin as “Curecumin”: From kitchen to clinic. Biochemical Pharmacology, 2008. 75(4): p. 787-809. 15. Kikic, I. and F. Vecchione, Supercritical impregnation of polymers. Current Opinion in Solid State and Materials Science, 2003. 7(4): p. 399- 405. 16. Takada, M., S. Hasegawa, and M. Ohshima, Crystallization kinetics of poly(L-lactide) in contact with pressurized CO2. Polymer Engineering & Science, 2004. 44(1): p. 186-196. 17. Zabihi, F., et al., High yield and high loading preparation of curcumin-PLGA nanoparticles using a modified supercritical antisolvent technique. Industrial and Engineering Chemistry Research, 2014. 53(15): p. 6569-6574. 18. Zabihi, F., et al., Polymeric coating of fluidizing nano-curcumin via anti-solvent supercritical method for sustained release. Journal of Supercritical Fluids, 2014. 89: p. 99-105.

Supercritical CO2 feasibility for the impregnation and release of curcumin in PLGA

19. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with indomethacin in supercritical CO 2. Journal of Supercritical Fluids, 2012. 63: p. 155-160. 20. Mazarro, R., et al., Influence of the operative conditions on the characteristics of poly(D,L-lactide-co-glycolide) synthesized in supercritical carbon dioxide. Macromolecular Symposia, 2010. 287(1): p. 111-118. 21. Mazarro, R., et al., Copolymerization of D,L-lactide and glycolide in supercritical carbon dioxide with zinc octoate as catalyst. Journal of Biomedical Materials Research - Part B Applied Biomaterials, 2008. 85(1): p. 196-203. 22. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with indomethacin in supercritical CO2. The Journal of Supercritical Fluids, 2012. 63: p. 155-160. 23. Mazarro, R., Síntesis de materiales poliméricos biocompatibles para la liberación controlada de fármacos mediante tecnología supercrítica. 2008. 24. Shaker, M., et al., Enhanced photodynamic efficacy of PLGA- encapsulated 5-ALA nanoparticles in mice bearing Ehrlich ascites carcinoma. Applied Nanoscience, 2014. 4(7): p. 777-789. 25. Rosas, J.E. and J.L. Pedraz, Microesferas de PLGA: un sistema para la liberación controlada de moléculas con actividad inmunogénica. Revista Colombiana de Ciencias Químico - Farmacéuticas, 2007. 36: p. 134-153. 26. Makadia, H.K. and S.J. Siegel, Poly Lactic-co-Glycolic Acid (PLGA) as Biodegradable Controlled Drug Delivery Carrier. Polymers, 2011. 3(3): p. 1377-1397. 27. Cao, G.-P., T. Liu, and G.W. Roberts, Predicting the effect of dissolved carbon dioxide on the glass transition temperature of poly(acrylic acid). Journal of Applied Polymer Science, 2010. 115(4): p. 2136-2143. 95

Chapter 1

28. Oliveira, N.S., et al., Gas solubility of carbon dioxide in poly(lactic acid) at high pressures: Thermal treatment effect. Journal of Polymer Science Part B: Polymer Physics, 2007. 45(5): p. 616-625. 29. Cabezas, L.I., et al., Validation of a mathematical model for the description of hydrophilic and hydrophobic drug delivery from biodegradable foams: Experimental and comparison using indomethacin as released drug. Industrial and Engineering Chemistry Research, 2014. 53(21): p. 8866-8873. 30. Xu, Y., et al., Polymer degradation and drug delivery in PLGA- based drug–polymer applications: A review of experiments and theories. Journal of Biomedical Materials Research Part B: Applied Biomaterials, 2016: p. n/a-n/a. 31. El-Sherbiny, I.M. and H.D.C. Smyth, Controlled Release Pulmonary Administration of Curcumin Using Swellable Biocompatible Microparticles. Molecular Pharmaceutics, 2012. 9(2): p. 269-280. 32. Bajpai, S.K., N. Chand, and S. Ahuja, Investigation of curcumin release from chitosan/cellulose micro crystals (CMC) antimicrobial films. International Journal of Biological Macromolecules, 2015. 79: p. 440-448. 33. Pan Ch, J., et al., Preparation, characterization and anticoagulation of curcumin-eluting controlled biodegradable coating stents. J Control Release, 2006. 116(1): p. 42-9.

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

Chapter 2

2.1. Introducción 2.2. Experimental 2.2.1. Materials 2.2.2. Synthesis of 4-azidomethyl-7- methoxycoumarin 2.2.3. Synthesis of click product at atmospheric pressure 2.2.4. Synthesis of click product at supercritical conditions 2.2.5. Characterization measurements 2.3. Results and discussion 2.4. Conclusions 2.5. Bibliography

A continuación se muestran de forma esquematizada los contenidos del capítulo 2 de resultados.

Liberación Liberación in vitro in vivo

Impregnaciones

scCO2

scCO2 Compuestos Catalizador en orgánicos fase heterogénea modificados

Liberación in vitro Abstract

nce scCO2 technology has been checked as a suitable media for

Oreactions in which a pharmaceutical compound was impregnated, the next step in this research is to link the biodegradable polymer and another compound with pharmaceutical properties through a covalent bound, being the functionalization with click chemistry the one chosen for this work. This is the first work where click chemistry is employed using supercritical technology with CO2 as solvent. For this purpose the process will be optimized in order to obtain the best conditions where polymer can be functionalized in scCO2.

Chapter 2

2.1. Introduction

Polymer functionalization has emerged as one of the most important ways in the chemistry of polymers destined to pharmaceutical applications [1, 2]. The linkage through a covalent bond of the polymer to the organic compound with pharmaceutical properties is a key advantage in the controlled drug release because of a more controlled profile can be obtained in comparison with impregnation methods [3, 4].

In this way, the selection of the right solvent for the synthesis of pharmaceutical related products is critical. Beyond the obvious function of solvents to allow proper reactivity in solution, they should be easily separated from the Active Pharmaceutical Ingredient (API), reducing their economic and environmental impact on the product production [5]. The usage of harmful solvents also brings the disadvantage of solvent incorporation into the API. If they cannot be removed, the amount must be controlled or limited to levels safe to the patient [6]. Moreover, the presence of toxic residues related with raw materials excess, by-products or solvents from the production process or further manufacture is not acceptable in such medical related products [7]. Most of the drug related allergies is suspected that could be in relation with a deficient removal of synthesis and manufacture residues. This issue is even more critical in polymers destined to controlled drug delivery because the objective is that have a prolonged stay in the body until being biodegraded and bioabsorbed by the body [8]. By now, significant efforts in substitution of traditional organic solvents are focused on using green solvents to carry out environmental friendly processes which can eliminate this problem [9].

100

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

Supercritical fluids are finding application in the production of pharmaceutical related products due to their ability to solve difficult process, particulation and formulation problems [10, 11]. Supercritical fluids exhibit a pressure-tuneable dissolving power with a liquid-like density and gas-like transport properties [12]. They possess the attractive property of easy separation from the substrate once the synthesis is complete, dragging with it most of the low molecular weight residues by simply and quickly venting [13].

In this context, the use of supercritical carbon dioxide (scCO2) as solvent appears as a solution to carry out environmental and patient friendly processes due to its lack of reactivity, high diffusivity, zero surface tension, good transport properties and sterilization capacity [5, 14]. In relation with polymeric materials, the employment of scCO2 produces a significant reduction of the glass transition temperature (Tg) and polymer swelling that allows the proper and homogeneous diffusion and dispersion of molecules into the polymer network [15]. In order to get a polymer based material for controlled drug release there are two main alternatives: physical entrapping of the API (absorbed or encapsulated) or covalent bonding to the polymer backbone. Both have advantages and disadvantages, while the first one exhibits a prompt delivery of great part of the drug, the second one present a more time stable release linked with the polymer biodegradation.

Although the employment of scCO2 as impregnation media and/or carrier for polymer-drug presentations (micro and nanocapsules, scaffolds, etc) has been widely described in literature, the functionalization of biopolymers in scCO2 for medical application is a field waiting to be explored [15-19].

101

Chapter 2

Among different techniques for polymer functionalization where toxic organic solvents are used, click chemistry has emerged as one of the most promising reactions because it is classified as a very specific, efficient and versatile reaction which allow to obtain high products yields[20]. Within the reactions included in the field of click chemistry, Huisgen 1,3-dipolar cycloaddition is the most employed in polymer chemistry. It consists on the reaction of an azido group to an alkyne group (AAC) catalyzed by copper (CuAAC) in organic media, where DFM or THF are the most common solvents used to achieve the functionalization of chemical product [21, 22]. The employment of the “click” route to get functionalized polymers using scCO2 as reaction media is almost unexplored and can open an absolutely green road to lots of pharmaceutical preparations. In fact, there are only three papers in the literature describing “click” reactions performed into scCO2 and any of them describes polymer functionalization without the use of a ligand [23-25]. The polymer chosen to be functionalized in this work is Polylactic acid (PLA). PLA is an aliphatic polyester approved by the US Food and Drug Administration (FDA) for contact with biological fluids[26]. This polymer presents a big variety of properties including renewability, biocompatibility, processability and energy saving [27]. Among PLA properties, its bioresorbability and biocompatibility in the human body make this polymer an excellent candidate to be used in biomedical field as manufacture tissue engineering scaffolds, delivery system materials or different bioabsorbable medical implants [28-33]. The organic compound chosen for PLA functionalization in this work is the coumarin. This substance is a plant-derived natural product which consists of an aromatic ring fused to a condensed lactone ring [34]. Coumarin is well-known by its pharmacological properties such as anti- 102

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2 inflammatory where it is able to remove protein and oedema fluid from injured tissues, anticoagulant activity due to coumarin is vitamin K antagonist producing anticoagulant effect or antiviral activity because this compound is considered as anti-HIV agent [35]. In this work, the functionalization of polylactic acid (PLA) acetylene by click chemistry in supercritical carbon dioxide is studied for first time. The catalytic activity of the system and the purity of the products obtained is investigated to let clear the great advantages of using scCO2 as reaction media for “click” functionalization.

2.2. Experimental

2.2.1. Materials

Sodium azide (> 99,5%, Sigma Aldrich), 4-Bromomethyl-7-methoxycoumarin (97%, Sigma Aldrich), Copper(I) iodide (CuI) (> 99%, Sigma- Aldrich), N,N-Diisopropylethylamine (DIPEA;>99%, Sigma-Aldrich), Polylactic Acid acetylene (98.3%, Specific polymers), Tetrahydrofuran (THF) (HPLC grade; SDS S.A., Spain) Copper(II) acetate monohydrate

(Cu(CO2CH3)2·H2O) (> 99%, Sigma Aldrich) and carbon dioxide (Carburos metálicos, S.A., Spain) with a purity of 99.5%. All other reagents and solvents used in the study were of analytical grade and used as delivered.

2.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin

The synthesis of this compound was carried out according to bibliography [36]. A mixture of NaN3 (1,2 g) and 4-bromomethyl-7- methoxycoumarin (1 g) in acetone/acetonitrile (1:1, 120 ml) solution was added to a 250 ml flask. The mixture was stirred at 50º C for 48 h. Then, solvents were removed under vacuum. The organic extracts were washed 103

Chapter 2 with water to precipitate the 4-bromomethyl-7-methoxycoumarin which did not react. Then, product was filtered and washed with heptane and dried under vacuum.

2.2.3. Synthesis of click product at atmospheric pressure

Click product at atmospheric pressure was synthetized using CuI as catalyst where DIPEA was used as nitrogen base.

PLA-coumarin click was synthetized as follows. A solution of 4- azidomethyl-7-methoxycoumarin (30 mg, 0.13 mmol) and PLA acetylene (258.31 mg, 0.13 mmol) in tetrahydrofuran (THF) (4,06 ml) was purged with nitrogen. Then, copper iodide (CuI) (6.19 mg, 0.032 mmol) and DIPEA (0.0057 ml, 0.032 mmol) were added under nitrogen atmosphere. Then reaction mixture was stirred for 24 h at 40 ºC. After the reaction, the solvent was removed under vacuum, to obtain a click product with the minimum quantity of organic solvent.

2.2.4. Synthesis of click product at supercritical conditions

Click product at supercritical conditions was synthetized using Copper

(II) acetate monohydrate (Cu(OAc)2·H2O) as catalyst.

The procedure of synthesis is the following: An equimolar quantity of PLA acetylene and 4-azidomethyl-7-methoxycoumarin were added into a stirring tank reactor using a low loading of catalyst (1-12% mol). Once sample is introduced into the reactor, it is heated and loaded with CO2 until temperature and pressure conditions are reached. When reaction time is completed (24-48 h), the reactor is depressurized to remove CO2 from the reactor in order to obtain click product without any solvent.

104

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

2.2.5. Characterization measurements

Infrared (IR) spectra were recorded on a Varian 640-IR Fourier transform IR spectrophotometer with 16 scans per experiment at a resolution of 32 cm-1 in the range 4000 to 400 cm-1, using the software Varian Resolution.

Nuclear Magnetic Resonance (NMR) was measured with Varian Gemini

FT-400 spectrometer using CDCl3 as solvent.

Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) was carried out using a Bruker Autoflex II

TOF/TOF spectrometer (Bremen, Germany) using CDCl3 as solvent and dithranol (1,8,9-trihydroxyanthracene) as matrix material.

Table 2.1. Temperature intervals in DSC analysis.

Polymer Ramp (°C/min) Temperature intervale (°C) First heating 40 40 to 200 Cooling 10 200 to -50 Second heating 10 -50 to 200

2.3. Results and discussion

In order to demonstrate that scCO2 is a suitable media to perform the functionalization of and alkyne-PLA with an organic compound with pharmacological properties the azido coumarin, the reaction was carried 105

Chapter 2

out in parallel at 1 bar into conventional solvents and in scCO2. Both reagents were previously checked to ensure the existence of alkyne group in the polymer and azide group in the coumarin (Figure 2.1). In both cases CuAAC type of Huisgen 1,3-dipolar cycloaddition based on the formation of a triazol ring was chosen for linkage of PLA acetylene and 4-azidomethyl- 7-methoxycoumarin [37]. The first one was performed using THF as solvent, where CuI is partiality soluble allowing the functionalization of the product without degradation. A nitrogen base (DIPEA) and inert atmosphere to maintain the oxidation value of catalyst, which in this case is +1 referred to CuI. The quantities of catalyst, nitrogen base and solvent used were calculated in relation to the amount of azide chosen for the reaction, based on the conditions optimized in a previous work [38].

H3C Catalyst O O O 24-48 h P

40-47 ºC

N N N O O OH

n H3C H3C O O O

Figure 2.1. Scheme of click reaction.

Maldi-Tof mass spectra of the PLA acetylene and the coumarin functionalized one are shown in Figure 2.2. It was analysed to get to know the exactly mass distribution of the polymer and its molecular weight to

106

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2 observe the displacement of molecular weight when click reaction was carried out.

IV Mw: 1958 III Mw: 1815 V Mw: 2102 II Mw: 1670 VI Mw: 2246

VII Mw: 2390 I Mw: 1525

III* - Mw:2045 II* - Mw:1902 IV* - Mw:2190

V* - Mw:2334 I* - Mw:1756

VI* - Mw:2478

VII* - Mw:2622

III* - Mw:2045 IV* - Mw:2190

II* - Mw:1902 V* - Mw:2334

I* - Mw:1756 VI* - Mw:2478

VII* - Mw:2622

Figure 2.2. MALDI-TOF mass spectrum of a) starting PLA acetylene (initial product); b) coumarin functionalized PLA at atmospheric pressure, t: 24 h; c)

coumarin functionalized PLA in scCO2, t: 48 h.

107

Chapter 2

In Figure 2.2a it can be confirmed the repetition of monomeric units of PLA acetylene every 144 g/mol for an average molecular weight of 1987 g/mol. In the Figures 2.2b and 2.2c can be confirmed that the peaks marked as (I*-VII*) have suffered a displacement of molecular weight of 231 g/mol in comparison with the corresponding peaks of original PLA acetylene marked with the same numbers without * (I-VII). This displacement corresponds exactly with the 4-azidomethyl-7- methoxycoumarin molecular weight, which is repeated every 144 g/mol correspondent to the monomeric unit of PLA acetylene, confirming that the click reaction was successfully carried out. The yield of functionalized product is calculated through the comparison of intensities corresponding to the peaks to PLA acetylene (I-VII) with respect to the peaks of coumarin functionalized PLA (I*-VII*) (Figures 2.2b and 2.2c). These mass spectrums distribution confirms the reaction was successfully carried out in a yield higher than 95 % in both cases.

To have a complete characterization of the product in order to be compared to the obtained in supercritical conditions FTIR spectra of both products were performed, to check that the peak correspondent to azide at 2110 cm-1 disappears (Fig. 2.3), what demonstrate the triazol ring was formed and the reaction was carried out successfully.

108

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

b) 2110 cm-1

4000 3600 3200 2800 2400 2000 1600 1200 800 400 Wavenumber (cm-1)

Figure 2.3. FTIR spectra corresponding to a) 4-azidomethyl-7-methoxycoumarin, t: 48h; b) click product synthetized at atmospheric pressure, t: 24h; c) click product

synthetized in scCO2, t: 48 h.

Figure 2.4 shows 1H NMR of PLA acetylene and click product synthesized at atmospheric and supercritical conditions. In this Figure can be observed the signals correspondent to functionalized PLA. These signals are at 5.30 ppm due to the CH2 neighbouring the 1,2,3-triazole group and the coumarin ring and the signal at 7.6 ppm correspondent to the proton of the 1,2,3-triazole ring. As it is demonstrated, the reaction has been carried out successfully at atmospheric pressure and in scCO2 with high yield.

109

Chapter 2

Figure 2.4. 1H NMR spectra of a) PLA Acetylene (initial product); b) click product

synthesized at atmospheric pressure t: 24 h; c) click product synthesized in scCO2 t: 48 h.

110

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

As can be observed in Figures 2.3c and 2.4c click reaction at supercritical conditions was successfully performed. Signals at 7.6 correspondent to 1,2,3-triazol ring can be observed in Figure 4c. Whereas in Figure 2c a displacement of 231 g/mol in click product is also checked by Maldi-Tof analysis with a high yield, according to the intensity of signals correspondent to PLA acetylene compared to the intensity of functionalized product.

Taking into consideration the previous results, the next step in this work was to study the influence of the operational variables to determine optimized conditions in order to obtain the maximum yield in the shorter reaction time.

Different reactions at supercritical conditions were carried out to determine the yield achieved when operational conditions and catalyst loading were modified. The range of temperatures used for this study was from 40 to 47 ºC according to PLA acetylene Tg, and the values of pressure were established between 80 and 110 bar. A low loading of catalyst

(Cu(OAc)2·H2O) was used (1-10% mol) and a time between 24 and 48 hours was taken for each experiment. The different click reactions at supercritical conditions studied in this work are shown in Table 2.2.

111

Chapter 2

Table 2.2. Relation of click reactions carried out at supercritical conditions and atmospheric pressure.

Entry Catalyst loading (%) P (bar) Time (h) Temperature (ºC) Yield (%)

1 11.5 110 48 47 96.9 2 100 92.6 3 80 24 89.61 4 12.3 40 79.7 5 6.31 71.5 6 2.83 69.1 7 1.11 9.3 8 0 0 9 12.3 1 24 40 96.2

As can be observed in Table 2.2, different click reactions at supercritical conditions were carried out obtaining yields close to 70% when the molar percentage of catalyst was higher than 2.8%. In all the cases studied in this work the catalyst loading of Cu(OAc)2·H2O was not higher than 12.3%, what demonstrate the efficiency of this catalyst in supercritical media.

According to the results shown in Table 2.2, the highest yields were achieved when values of pressure were established close to 100 bar and temperature was 47 ºC, obtaining efficiencies in clickation over 90 % (Entries 1 and 2) in both cases according to Maldi-Tof and 1H NMR analysis (Figures 2.2 and 2.4), being necessary a time of 48 h for each reaction. When reaction time was reduced to 24 h maintaining the catalyst loading to 11.5% and reducing pressure to a value close to critical point (Entry 3), the yield only decreased a 3%, demonstrating that reaction rate is quite fast also in scCO2.

112

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

The two parameters which more affect in terms of yield are temperature and quantity of catalyst. It can be observed that when the value of temperature was decreased to 40 ºC, yield was approximately a 10% lower than that of the reaction carried out at the same value of pressure and time, in spite of using a tiny quantity more of catalyst. The other parameter which affects considerably to the performance of the reaction is the amount of catalyst. It was observed a limit for which, if the molar percentage of catalyst used decreases from 2.83% to 1.11%, the yield of the reaction decreases a 58.9%, being a critical parameter.

2.4 Conclusions

The functionalization via click chemistry of polylactic acid (PLA) with coumarin in supercritical conditions CO2 has been achieved being observed that is possible to obtain similar yields, higher than 95% in both cases.

The highest yields were achieved at 100 bar and 47 ºC obtaining efficiencies in clickation about 95 % in 48 h of reaction. When reaction time was reduced to 24 h, maintaining the catalyst loading to 11.5% and reducing pressure to a value close to critical point, the yield only decreased a 3%, demonstrating that reaction rate is quite fast also in scCO2.

113

Chapter 2

2.5. Bibliography

1. Neamtu, I., et al., Basic concepts and recent advances in nanogels as carriers for medical applications. Drug Delivery, 2017. 24(1): p. 539-557. 2. Hu, X., et al., Surface functionalization of hydrogel by thiol-yne click chemistry for drug delivery. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2016. 489: p. 297-304. 3. Paka, G.D. and C. Ramassamy, Optimization of Curcumin-Loaded PEG-PLGA Nanoparticles by GSH Functionalization: Investigation of the Internalization Pathway in Neuronal Cells. Molecular Pharmaceutics, 2017. 14(1): p. 93-106. 4. Mauri, E., F. Rossi, and A. Sacchetti, Tunable drug delivery using chemoselective functionalization of hydrogels. Materials Science and Engineering C, 2016. 61: p. 851-857. 5. Grodowska, K. and A. Parczewski, Organic solvents in the pharmaceutical industry. Acta Poloniae Pharmaceutica - Drug Research, 2010. 67(1): p. 3-12. 6. Bohrer, D., Sources of Contamination in Medicinal Products and Medical Devices. Sources of Contamination in Medicinal Products and Medical Devices. 2012. 1-571. 7. Argentine, M.D., P.K. Owens, and B.A. Olsen, Strategies for the investigation and control of process-related impurities in drug substances. Advanced Drug Delivery Reviews, 2007. 59(1): p. 12-28. 8. Dixit, K., R.B. Athawale, and S. Singh, Quality control of residual solvent content in polymeric microparticles. Journal of Microencapsulation, 2015. 32(2): p. 107-122.

114

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

9. Henderson, R.K., et al., Expanding GSK's solvent selection guide - Embedding sustainability into solvent selection starting at medicinal chemistry. Green Chemistry, 2011. 13(4): p. 854-862. 10. Fages, J., et al., Particle generation for pharmaceutical applications using supercritical fluid technology. Powder Technology, 2004. 141(3): p. 219-226. 11. Jung, J. and M. Perrut, Particle design using supercritical fluids: Literature and patent survey. Journal of Supercritical Fluids, 2001. 20(3): p. 179-219. 12. Martín, A. and M.J. Cocero, Micronization processes with supercritical fluids: Fundamentals and mechanisms. Advanced Drug Delivery Reviews, 2008. 60(3): p. 339-350. 13. Eckert, C.A., B.L. Knutson, and P.G. Debenedetti, Supercritical fluids as solvents for chemical and materials processing. Nature, 1996. 383(6598): p. 313-318. 14. Lang, Q. and C.M. Wai, Supercritical fluid extraction in herbal and natural product studies — a practical review. Talanta, 2001. 53(4): p. 771- 782. 15. Mazarro, R., et al., Copolymerization of D,L-lactide and glycolide in supercritical carbon dioxide with zinc octoate as catalyst. Journal of Biomedical Materials Research - Part B Applied Biomaterials, 2008. 85(1): p. 196-203. 16. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with indomethacin in supercritical CO 2. Journal of Supercritical Fluids, 2012. 63: p. 155-160.

115

Chapter 2

17. Mazarro, R., et al., Influence of the operative conditions on the characteristics of poly(D,L-lactide-co-glycolide) synthesized in supercritical carbon dioxide. Macromolecular Symposia, 2010. 287(1): p. 111-118. 18. Cabezas, L.I., et al., Novel model for the description of the controlled release of 5-fluorouracil from PLGA and PLA foamed scaffolds impregnated in supercritical CO2. Industrial and Engineering Chemistry Research, 2014. 53(40): p. 15374-15382. 19. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with 5-fluorouracil in supercritical CO2. Journal of Supercritical Fluids, 2013. 80: p. 1-8. 20. Hein, C.D., X.-M. Liu, and D. Wang, Click Chemistry, A Powerful Tool for Pharmaceutical Sciences. Pharmaceutical Research, 2008. 25(10): p. 2216-2230. 21. Rostovtsev, V.V., et al., A Stepwise Huisgen Cycloaddition Process: Copper(I)-Catalyzed Regioselective “Ligation” of Azides and Terminal Alkynes. Angewandte Chemie International Edition, 2002. 41(14): p. 2596- 2599. 22. Meldal, M., Polymer "clicking" by CuAAC reactions. Macromolecular Rapid Communications, 2008. 29(12-13): p. 1016-1051. 23. Grignard, B., et al., First example of "click" copper(i) catalyzed azide-alkyne cycloaddition in supercritical carbon dioxide: Application to the functionalization of aliphatic polyesters. Green Chemistry, 2009. 11(10): p. 1525-1529. 24. Zhang, W., et al., Cu(OAc)2·H2O - An efficient catalyst for Huisgen- click reaction in supercritical carbon dioxide. Tetrahedron Letters, 2015. 56(19): p. 2472-2475.

116

Functionalization and optimization of PLA with coumarin via click chemistry in

supercritical CO2

25. Jiang, Y., et al., Metallic copper wire: a simple, clear and reusable catalyst for the CuAAC reaction in supercritical carbon dioxide. RSC Advances, 2015. 5(90): p. 73340-73345. 26. Jain, R.A., The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices. Biomaterials, 2000. 21(23): p. 2475-2490. 27. Anderson, J.M. and M.S. Shive, Biodegradation and biocompatibility of PLA and PLGA microspheres. Advanced Drug Delivery Reviews, 1997. 28(1): p. 5-24. 28. Hamad, K., et al., Properties and medical applications of polylactic acid: A review. Express Polymer Letters, 2015. 9(5): p. 435-455. 29. Zeng, X., et al., Cholic acid-functionalized nanoparticles of star- shaped PLGA-vitamin E TPGS copolymer for docetaxel delivery to cervical cancer. Biomaterials, 2013. 34(25): p. 6058-6067. 30. Zeng, X., et al., Docetaxel-Loaded Nanoparticles of Dendritic Amphiphilic Block Copolymer H40-PLA-b-TPGS for Cancer Treatment. Particle & Particle Systems Characterization, 2015. 32(1): p. 112-122. 31. Zhu, D., et al., Docetaxel (DTX)-loaded polydopamine-modified TPGS-PLA nanoparticles as a targeted drug delivery system for the treatment of liver cancer. Acta Biomaterialia, 2016. 30: p. 144-154. 32. Tao, W., et al., Polydopamine-Based Surface Modification of Novel Nanoparticle-Aptamer Bioconjugates for In Vivo Breast Cancer Targeting and Enhanced Therapeutic Effects. Theranostics, 2016. 6(4): p. 470-484. 33. Wang, T., et al., DTX-loaded star-shaped TAPP-PLA-b-TPGS nanoparticles for cancer chemical and photodynamic combination therapy. RSC Advances, 2015. 5(62): p. 50617-50627.

117

Chapter 2

34. Jain, P.K. and H. Joshi, Coumarin: Chemical and pharmacological profile. Journal of Applied Pharmaceutical Science, 2012. 2(6): p. 236-240. 35. Venugopala, K.N., V. Rashmi, and B. Odhav, Review on natural coumarin lead compounds for their pharmacological activity. BioMed Research International, 2013. 2013. 36. Velencoso, M.M., et al., Click-ligation of coumarin to polyether polyols for polyurethane foams. Polymer International, 2013. 62(5): p. 783- 790. 37. Riva, R., et al., Combination of ring-opening polymerization and "click" chemistry towards functionalization of aliphatic polyesters. Chemical Communications, 2005(42): p. 5334-5336. 38. Bolognesi, A., et al., Towards Semiconducting Graft Copolymers: Switching from ATRP to “Click” Approach. Macromolecular Chemistry and Physics, 2010. 211(13): p. 1488-1495.

118

3 Feasibility of copper wire as heterogeneous catalyst in click chemistry in

scCO2

Chapter 3

3.1. Introducción 3.2. Experimental 3.2.1. Materials 3.2.2. Synthesis of 4-azidomethyl-7- methoxycoumarin 3.2.3 Synthesis of click product in scCO2 3.2.4. Purification of click product 3.2.5. Cleaning process of copper wire catalyst 3.2.6. Characterization measurements 3.3. Results and discussion 3.4. Conclusions 3.5. Bibliography

A continuación se muestran de forma esquematizada los contenidos del capítulo 3 de resultados.

Liberación Liberación in vitro in vivo

Impregnaciones

scCO2

scCO2 Compuestos Catalizador en orgánicos fase homogénea modificados

Liberación in vitro Abstract

he use of copper wire as catalyst in the functionalization of

biopolymers by click chemistry in supercritical CO2 (scCO2)

T constitutes an interesting route to obtain purer products and a greener processes. The use of a heterogeneous catalysis allows a cleaner work-up procedure, allowing a complete removal of the catalyst through a simple purification step. Different parameters as relation surface/volume of catalyst, reaction time or catalyst reusability were studied to determine the way in which they affect the yield of the functionalization process, where coumarin was employed as organic compound in order to produce the PLA-pharmaceutical linked product, obtaining yields over 90%. Finally a scaled costs analysis of this process was carried out. Different click chemistry reactions were evaluated varying the technology and catalysts used. Being concluded that the use of reused and cleaned copper wire in scCO2 conditions supposes the most economical process with a manufacturing cost of 108.97 €/kg.

121

Chapter 3

3.1. Introduction

Total elimination of harmful or allergenic catalysts residues in functionalized products destined for biomedical applications is a critical issue. Depending on the mechanism of reaction selected, a different type of catalyst is required [1]. In the case of polymers functionalized using click chemistry a copper compound is the most frequently catalyst used. This reaction consists on the linkage of an azido group to an alkyne group (AAC) catalyzed by copper (CuAAC), being Huisgen 1,3-dipolar cycloaddition the most employed in polymer chemistry. In the last years, it has emerged as one of the most promising reactions because it is considered as a very specific, efficient and versatile reaction which allows btaining high product yields [2-4].

In Chapter 2, click chemistry in scCO⁠2 was described for first time for the functionalization of a biodegradable polymer eliminating the need of an organic solvent apart from CO⁠2 in the reaction, obtaining yields higher than 95% without any catalyst purification step [5]. In the conventional click chemistry cooper is employed as homogeneous catalyst. Apart from the limitations of its complete elimination from the final clicked product, the use of copper as a catalyst presents some additional difficulties depending on its state of oxidation. The first limitation is fresh Cu⁠+1 is required for each new click reaction. Second limitation is Cu⁠+1 is oxidized in the presence of air to Cu+2⁠ . Finally the third limitation is the removal of Cu⁠+1 and Cu⁠+2 from the reactional medium, fact which is very tedious and difficult for air-sensitive systems [6]. Both problems are solved in this work by the use of copper wire bits as catalyst (Cu (0) wire) in scCO⁠2. The selection of copper wire in this reaction allows the elimination of the whole amount of catalyst charged in the

122

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 reaction with a sieving step or simply with the use of tweezers. This catalyst adds also the possibility of reusing the same catalyst in several reactions, what could mean a decrease of the operational cost of this process [7, 8]. Additionally, the use of copper wire presents an important advantage which resides in the possibility of being recovered easily and reused in successive reactions. In case of large scale applications, this fact supposes an interesting economic saving due to facility of copper recovery and the negligible amount of catalyst that would have to be replenished [7, 8]. In this Chapter the click chemistry reaction between polylactic acid (PLA) and coumarin is performed in presence of copper wire as catalyst. The use of this heterogeneous catalyst provides a simple purification step which could mean an interesting decrease of manufacturing costs. In this work, different parameters will be studied in order to determine what of them could affect the most to the yield obtained. Among these parameters, it is remarkable the possibility of reusing copper wire as catalyst in several cycles. Additionally, the catalyst activity after being reused will be checked to establish the possible loss of yield in the click functionalization. Finally, it was performed an economical study about manufacturing cost in order to determine the feasibility and benefits associated to this alternative process, being compared to the classical homogeneous one. This study will make possible to determine the best configuration of catalyst in our process and validate this procedure like a profitable alternative to the classical procedure.

123

Chapter 3

3.2. Experimental

3.2.1. Materials

Sodium azide (> 99,5%, Sigma Aldrich), 4-Bromomethyl-7-methoxycoumarin (97%, Sigma Aldrich), Polylactic Acid acetylene (98.3%, Specific polymers), Copper wire (99%, Sigma-Aldrich), Nitrogen (99.5%, Carburos metálicos, S.A., Spain) and Carbon dioxide (99,5%, Carburos metálicos, S.A., Spain) with a purity of 99.5%. All other reagents and solvents used in the study were of analytical grade and used as delivered.

3.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin

The synthesis of this compound was carried out following a bibliography route detailed in Chapter 2 (1.2.1) [9].

3.2.3. Synthesis of click product in scCO2 Click product at supercritical conditions was synthetized using copper wire as catalyst. The procedure of synthesis is the following: An equimolar quantity of PLA acetylene and 4-azidomethyl-7-methoxycoumarin were added into a stirring tank reactor using copper wire as a catalyst. The quantity of catalyst in each reaction is indicated through the catalyst loading (mass %), this parameter was normalized in order to make comparable all the experiments performed. It was calculated like the weight of catalyst per surface and volume of catalyst. Once sample is introduced into the reactor, it is heated and loaded with CO2 until temperature (50 ºC) and pressure (130 bar) conditions are reached. When reaction time is completed (24–48 h), the reactor is depressurized to remove CO2 from the reactor in order to obtain click product without any solvent.

124

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

3.2.4. Purification of click product

Once click product was obtained from click chemistry reaction it is separated from copper wire through a sieving step where the whole quantity of catalyst is removed from click product.

3.2.5. Cleaning process of copper wire catalyst

Copper wire was introduced in a beaker where was washed with 50 ml of an organic solvent, acetone, with a constant stirring of 500 rpm for 1 h at 25ºC.

3.2.6. Characterization measurements

Cu content of the copper wires was determined by atomic absorption spectrophotometry, using a SPECTRA 220FS analyzer. The sample (ca. 0.5 g) was treated in 2 mL HCl, 3 mL HF and 2 mL H2O2 followed by microwave digestion (T = 250 ºC).

Copper wire was also characterized before reaction tests by X-Ray Diffraction (XRD) analysis with a Philips PW-1710 instrument, using Ni- filtered Cu Kα radiation (λ = 1.5404 Å). The samples were scanned at a rate of 0.02º·step-1 over the range 20º ≤ 2θ ≤ 80º (scan time 2 s·step-1) and the diffractograms were compared with the JCPDS-ICDD references.

Synthesized lightweight gypsum composites were depicted by means of Scanning Electron Microscopy (SEM) by using a FEI QUANTA 200.

Nuclear Magnetic Resonance (NMR) was measured with Varian Gemini FT-400 spectrometer using CDCl3 as solvent.

Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) was carried out using a Bruker Autoflex II

125

Chapter 3

TOF/TOF spectrometer (Bremen, Germany) using CDCl3 as solvent and dithranol (1,8,9-trihydroxyanthracene) as matrix material.

Table 3.1. Temperature intervals in DSC analysis.

Polymer Ramp (°C/min) Temperature intervale (°C) First heating 40 40 to 200 Cooling 10 200 to -50 Second heating 10 -50 to 200

3.3. Results and discussion

In order to check if copper wire is a suitable catalyst for the click functionalization with coumarin of polylactide, several reactions were carried out using CO2 in supercritical conditions. This reaction can be observed in the scheme shown in Figure 3.1.

Figure 3.1. Scheme of click reaction using copper wire as a catalyst.

The main novelty in this chapter is the substitution of the catalyst previously used in scCO2, copper (II) acetate monohydrate, by copper wire bits. This fact supposes a great advance from purification point of view

126

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 because a chromatographic column will not be necessarily set up to perform the purification. In this case a simple sieve step or the use of tweezers is enough for click product separation.

Initially, a first click chemistry reaction was carried out in order to check if copper wire works as catalyst using a low loading of catalyst when it is used with a biodegradable polymer. The operational conditions were chosen according to optimized conditions studied in Chapter 2 [5].

1H NMR spectrum of click product using copper wire as a catalyst is shown in Figure 3.2. In the spectrum can be identified the characteristic signal corresponding to the proton of the 1,2,3-trizole ring at 7.70 ppm and the corresponding to CH2 neighbouring the 1,2,3-trizole group at 5.30 ppm. These observations confirms that click ligation of coumarin to PLA has been achieved using only copper wire as catalyst in absence of solvents other than scCO2.

To complete the characterization of the product and get stronger confirmation about the formation of the target product and no side byproducts, Maldi-Tof mass spectrum of the click product was carried out following the same procedure as in Chapter 2. In Figure 3.3 it can be confirmed that the peaks of the coumarin-PLA appear spaced at the proper molecular weight distance, corresponding to the molecular weight of the coumarin compound.

127

Chapter 3

Figure 3.2. 1H NMR spectra of click product synthetized in scCO2 using copper wire as catalyst.

Figure 3.3. Maldi-Tof mass spectrum of coumarin functionalized PLA in

scCO2.

128

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

Yield was determined according to procedure explained in Chapter 2. In this reaction the yield reached was 26.36%. This yield is somewhat low compared to the yield higher than 95% that was obtained in a previous work where copper acetate monohydrate was employed as catalyst in scCO2.

Copper concentration in the final product was determined by ICP analysis. The presence of copper in the coumarin-PLA was undetectable. This result confirms that copper wire can be an excellent and non-harmful catalyst for the preparation of functionalized biopolymers for drug delivery by a click procedure. The amount of catalyst employed was modified in order to check if this increase improves the yield of the reaction. In Table 3.2, are shown the yields of the click reaction employing growing amounts of catalyst using copper wire bites of the same diameter.

Table 3.2. Study of catalyst loading using copper wire as catalyst.

Catalyst loading* Catalyst mass Remaining Cu Reaction Yield (%) (mg·mm-2·mm-3) (mg) (ppm) 1 0.011 181 0.00 71.28 2 0.010 198 0.00 85.64 3 0.009 215 0.00 93.02

*Catalyst loading was normalized: mg of catalyst/(catalyst surface · catalyst volume)

As can be observed in Table 3.2, an increase of catalyst loading means an increase of the yield in the reaction. The highest yield reached was 93.02%, it was obtained for the highest quantity of copper wire. The amounts of residual copper in the final product were measured by ICP and as can be observed copper wire is totally absent in the final product, independently of the amount of catalyst used. This fact supposes an 129

Chapter 3 important improvement with respect to the previous click reactions where copper catalyst in powder form was used [5].

The success of product obtained in reaction 3, was also checked with the comparison of polymer functionalized and the raw polymer glass transition temperature (Tg) through differential scanning calorimetry. It was found that the polymer functionalization can greatly enhance Tg of functionalized polymer with respect to the non-functionalized one [10, 11]. According to Figure 3.4, in this study PLA glass transition temperature (Tg) exhibited a relation with the functionalization of PLA. Meanwhile PLA Acetylene Tg is 24.94 ºC, PLA functionalized with coumarin in a 93% yield shows an increase of its Tg up to a value of 42.69ºC. The main reason for the large enhancement of Tg is that the chain decorative group provided a stronger interlocking between the coumarin and PLA structure.

Figure 3.4. Differential scanning calorimetry of PLA acetylene and PLA functionalized.

130

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

The high purity of the product obtained in reaction 3 can be appreciated in Figure 3.5, where Maldi-Tof spectra shows the peaks corresponding to the click product with a much higher intensity with respect to the peaks corresponding to PLA acetylene.

Figure 3.5. Maldi-Tof mass spectrum of coumarin functionalized PLA in

scCO2, cat: 215 mg; t:24 h.

As the specific surface of copper wire is substantially smaller than specific surface of the catalyst employed as powder, the yield obtained for the same reaction time is lower. This fact could be an inconvenient since a higher amount of catalyst would be required to obtain the same yield. However, as copper wire is easily and totally removed in the purification step, independently of the catalyst quantity, copper wire can be considered as an excellent alternative catalyst for this reaction.

Once copper wire has been established as an excellent alternative for click chemistry in scCO2, the next step is to study the industrial feasibility

131

Chapter 3 of use of this catalyst in a heterogeneous process. For that purpose it will be necessary to optimize this process in order to establish the economic benefits which this click chemistry reaction will provide in comparison to the technologies previously studied. For that purpose several process parameters affecting the reaction yield were studied. Cu-to-reactants ratio (i) was related to the reaction rate; while as the influence of the surface-to- volume ratio (ii) for the same quantity of catalyst can give information about fluidodinamics, which will affect the packing of the column. In addition the kinetic profile (iii) can give information about if the global process is driven by thermodynamics, equilibrium, or their combination. Finally, one of the most important process steps within the heterogeneous catalysis is the copper wire reutilization (iv), which will provide a substantial benefit in manufacturing costs. Once determined the effect of these parameters, a preliminary economical cost study will determine the feasibility of this technology like an industrially interesting alternative technology.

Table 3.3 shows the results obtained for the different process parameters studied. For that purpose, different quantities of catalyst where chosen (i). The total quantity of catalyst was obtained using 3 different lengths of cooper wires in order to study the influence of the relation surface/volume on the reaction (ii). The 3 lengths used in this work were 1.25, 2.5 and 5 mm, for the same commercial diameter 0.01 mm. All the reactions were carried out at 50º C and the same pressure, 130 bar, according to the optimized results obtained in Chapter 2. Regarding the reaction time, 24 h was chosen from these data in order to ensure the maximum achievable conversion. Another fundamental parameter in this reaction is the temperature. According to bibliography, an increase of 10ºC

132

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 from 40ºC to 50ºC could improve the yield obtained significantly. Nevertheless, the obtained yields when temperature was above 50ºC did not present almost any improvement [29].

Column 3 shows the mass of catalyst, while as column 2 give the specific surface corresponding to the different lengths (mm) for a giver diameter of 0.01 mm. To compare results for the same mass of catalyst, the number of wires was changed in the bed. For example, in the experiments carried out with a copper wire length of 1.25 mm, a higher number of copper wires were needed in order to change the catalyst loading from 0.015 mg/(mm2·mm3) to 0.011 mg/(mm2·mm3).

Table 3.3. Study of different process parameters in click chemistry carried out in

scCO2 using copper wire as catalyst.

Entry Catalyst loading Catalyst Length Time Remaining Yield (mg·mm-2·mm-3) mass (mg) (mm) (h) copper (ppm) (%) 1 0.015 130 1.25 24 0.00 72.59 2 0.011 180 0.00 81.54 3 0.015 130 2.5 0.00 76.82 4 0.011 180 0.00 83.66 5 0.009 215 0.00 85.65 6 0.015 130 5.00 0.00 76.89 7 0.011 180 0.00 85.64 8 0.009 215 0.00 93.02 9 0.006 360 0.00 94.97

From Table 3.3, the increase of the quantity of catalyst (i) produces an increase of the reaction yield for all lengths tested, fact which was previously observed previously [12]. The increase of yield is not linear, it

133

Chapter 3 could be explained because the reaction takes place in a step where equilibrium is not the variable which controls the process and hence they must be taken into account both kinetics and fluid dynamics. This fact could explain the loss of yield because of the lower packaging bed. In Figure 3.6 is shown a comparison of yields obtained for the experiments carried out with coppers wires of 5 mm length (Entries 6-9).

Figure 3.6. Comparative of yield obtained using different catalyst loadings for a copper wire of 5 mm.

Considering the effect of the specific surface (ii), increasing values of this variable in the range analysed produced a decrease in reaction yields. This result, apparently unexpected, must be analysed considering that in this reaction kinetic and fluid dynamics are the limiting steps, as it was previously observed. In consequence, lower specific surface (smaller wires) produces a higher packaging bed density and higher pressure drop producing lower yield. This effect produced the higher yields (95%) for of longer Cu packing beds of 5.00 mm.

134

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

In order to verify if the kinetic is the step which is controlling the reaction, several click functionalizations were carried out selecting different reaction times (iii). Experiments at 6, 12, 24 and 48 hours were performed (Figure 3.7). The values of pressure, temperature and catalyst loading were kept constant with respect to the experiments of Table 3.3.

100 90 80 70 60 50

Yield (%) Yield 40 30 20 10 0 0 4 8 12 16 20 24 28 32 36 40 44 48 Reaction time (h)

Figure 3.7. Study of influence of time for click product in scCO2.

In the light of the results obtained in Figure 3.3, there is a linear shift in the yield when the reactions take from 6 to 24 hours, typically corresponding to an equilibrium-governed reaction. From 24 to 48 hours time it is observable a change of tendency in yield. In this way could be concluded that equilibrium is not the step which is controlling the reaction at this point [13].

The results obtained also show the efficacy and reaction kinetics of copper wire as catalyst in comparison to copper in powder form at supercritical conditions [5]. The use of polymers in click chemistry 135

Chapter 3 functionalization makes this reaction slower in kinetic terms in comparison with other reactions in scCO2 where discrete molecules were employed [14]. In this case a reaction time of 24 h was required in order to obtain yields higher than 90%. With respect to the quantity of remaining copper in the final product, it was analysed by ICP analysis. As could be observed in Table 3.3, the whole quantity of copper was removed from the product once the copper wire was separated from the click compound.

For the previous reasons, the reaction time of 24 h was chosen as the optimum to get the maximum conversion. It is also remarkable that seems to be an “induction time” of 4-6 h and after that the reaction rate increase reaching the maximum conversion at about 24 h. This induction time means that the reaction needs a minimum time in order to achieve an observable conversion. Nevertheless, when this reaction is catalyzed by copper nanoparticles (0.5%mol) at 50ºC and even a lower value of pressure (100 bar) the is achieved in only 30 minutes [15]. This fact could be concluded that the longer the copper wire, the faster reaction time.

Finally, once the reaction conditions were studied the possibility of reusing the copper wire catalyst has been checked. To check the possibility of reusing the copper wire in successive reaction cycles, the whole quantity of the catalyst employed in one of the experiments previously reported. It was removed carefully and completely from the reaction media and employed in a new reaction with the same operational conditions (Figure 3.8).

136

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

Figure 3.8. 1H NMR spectra of click product in scCO2 with reused catalyst at 50 ºC, 130 bar and 24 h.

According to the results obtained in 1H RMN analysis in Figure 3.5, the signal corresponding to triazol ring is appearing at 7.70 ppm. It means that the click reaction was successful using the reused copper wires. In order to calculate the exact mass distribution of polymer and functionalized polymer Maldi-Tof mass spectra analysis was performed (Figure 3.9). The analysis of Maldi-Tof results was carried following the same procedure as in Chapter 2.

137

Chapter 3

IV Mw: 1958 III Mw: 1815 V Mw: 2102

II Mw: 1670 VI Mw: 2246

VII Mw: 2390 I Mw: 1525

III* Mw: 2045 II* Mw: 1902 IV* Mw: 2189 I* Mw: 1756 V* Mw: 2246 VI* Mw: 2390

1.000 1.500 2.000 2.500 3.000 III* Mw: 2045 II* Mw: 1902 IV* Mw: 2189 I* Mw: 1756 V* Mw: 2246 VI* Mw: 2390

1.000 1.500 2.000 2.500 3.000 Figure 3.9. Maldi-Tof mass sprectrum of a) starting PLA acetylene; b) click product using raw copper wire; c) click product reusing copper wire.

138

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

As the used copper wires were not washed or regenerated in any way before being reused, the appearance of the catalyst wires before and after the reaction were compared by SEM analysis, Figure 3.10.

Figure 3.10. SEM images of copper wire a) before the reaction; b) after the reaction; c) after the reaction and cleaning step.

As can be seen in Figure 3.10b), a hard crush of polymer remains onto the copper wire after the click reaction in scCO2 that explains the decrease of the yield observed in Maldi-Toff.

Catalyst was introduced in a beaker and was washed with acetone at constant stirring of 500 rpm for 1 h at 25ºC to remove the product remains. According to the SEM image shown in Figure 3.10 c) can be appreciated how the crush has been completely removed after washing with the organic solvent.

For the complete characterization of copper wire an X-ray diffraction analysis was carried out. As is shown in Figure 3.11 there was no surface oxidation of the wires before and after the reaction. XRD peaks at 43.4º, 50.3º and 74.12º respectively belong to Cu (111), Cu (200) and Cu (220), demonstrate that the catalyst is metallic copper in all the cases.

139

Chapter 3

(220)

(200) (111)

After cleaning step

After reaction Intensity / a.u. / Intensity

Before reaction

35 40 45 50 55 60 65 70 75 80 2 º

Figure 3.11. X-ray diffraction analysis of metallic copper wire.

Finally a reaction was carried out at the same conditions as Entry 8, to confirm the influence of the cleaning step (iv) in the yield of the reused copper wire.

In Figure 3.12 is shown a comparative of yields obtained for the different states of the catalyst, demonstrating that the washing performed to the particle allows to restore the wires to its original activity and the insignificant loss of yield (0.23%) related to reusability of copper wire in several cycles.

140

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

Figure 3.12. Yield of click chemistry reaction with a) new catalyst; b) reused catalyst for first time; c) reused catalyst for first time after the cleaning step; d) reused catalyst for second time after cleaning step.

The final purpose of this technology as it was previously commented is a pharmacological application. In order to carry out its industrial implementation a previous costs analysis must be evaluated to determine the feasibility of the different alternatives in the process. In this way, a comparison of different click reactions with different processes was carried out. The aim of this study is the determination of most interesting industrial process taking into account the different purification processes which each technology will require. In Table 3.4 the relation of the mentioned click chemistry processes is shown.

141

Chapter 3

Table 3.4. Relation of different click chemistry processes for the study of their economic impact.

Reused Process Pressure Catalyst Purification Solvent Ref. catalyst Chromatographic 1 Low CuI No THF [5] column Copper Chromatographic 2 High No CO2 [5] acetate column Copper 3 High No Sieving CO2 [12] wire Copper This 4 High Yes Sieving CO2 wire work Copper Sieving + This 5 High Yes CO2 wire cleaning step work

The processes 1 and 2 shown in Table 3.4 are carried out with copper catalyst in powder form. The main difference between both processes is the technology employed for the functionalization; while Process 1 is carried out at atmospheric pressure (1 atm) using THF as solvent, Process 2 was carried out using supercritical technology with CO2 as solvent. This fact supposes that Process 1 will require a purification process not only to remove the copper catalyst quantity from the product, but also another step to ensure the total elimination of THF. As in Process 2 supercritical technology was employed, the step corresponding to the solvent removal is omitted because of the scCO2 is a recognised solvent in pharmaceutical production processes[16]. With respect to Process 3, it is a process using supercritical technology with scCO2 as solvent. The catalyst used in this

142

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 case was copper wire. This reaction corresponds to a reaction where copper wire is replaced in every reaction with no reutilization. About Process 4, copper wire is reused with no cleaning step. In this way, the yield after being used is 83% as was previously commented. The yield which will be obtained in successive reactions will be lower as the available catalytic surface decreases with the reactions. In this case only a sieving step will be required to obtain copper free product. Finally, Process 5 corresponds to the process previously explained (Process 4), but in this case a catalyst cleaning step is carried out after the reactions. For this reason, the yield obtained in successive reactions will be very similar to the first one. This additional step will suppose an extra cost of purification process but it will be offset by the product purity obtained in each reaction.

A simple estimation of the different processes manufacturing cost (COM) can be calculated according to bibliography [17]. The COM can be determined by the sum of five main costs: raw material, operational labor, utilities, waste treatment, and investment (Table 3.5):

COM = 0.304 FCI + 2.73 COL + 1.23 · (CUT + CWT + CRM) (1)

Where FCI (€/kg) is the fraction of investment, COL (€/kg) is the operational labor cost, CUT (€/kg) is the utility cost, CWT (€/kg) is the waste treatment cost and CRM (€/kg) is the raw material cost.

For the calculation of COM several parameters and assumptions will be established for comparison purposes. Regarding FCI, we used the annual amortization cost of a plant for the main reaction with a period of amortisation of 10 years. For its determination, we used real equipment costs for reactor in process#1 and purification equipment in processes#1-5; while as equipment costs in processes #2-5 were calculated scaling laboratory data to obtain a target industrial production. This production, 143

Chapter 3

400 kg of product per year (50 grams of product by reaction), was scaled according to bibliography [18]. In the case of FCI, in Process 1 it is used an atmospheric reactor, whose total price is 14.000 €. FCI cost will be the same for processes from 2 to 5 due to in all the cases supercritical technology is used. For these cases the investment cost will be 40000 € in 2019.

In all the production processes, COL can be determined considering 4 employees, working 300 hours/year with a cost of 10€/h. Due to CUT are related mainly to purification processes, CUT costs will be integrated in CWT.

Table 3.5. Relation of fraction of investment, labor, utility, waste treatment and raw material costs.

FCI COL CWT CRM Process Yield (%) COM (€/kg) (€/kg) (€/kg) (€/kg) (€/kg) 1 3.5 30 23.75 14.64 0.96 135.33 2 10 30 20 14.64 0.97 131.63 3 10 30 1.25 14.64 0.93 112.32 4 10 30 1.25 13.52 0.88 117.17 5 10 30 1.62 11.53 0.93 108.97

About the waste treatment cost (CWT), it is one of the variables most influential in the process. In the case of low pressure process there will be required a chromatographic column and a rotary evaporator of high capacity leading to the desired year target production valued at 23.75 €/kg, which corresponds to the total price of a chromatographic column and a rotary evaporator according to the estimated rate of production in the plant and the period of its amortization, taking into account these two 144

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 parameters for the rest of CWT values in the different processes. For Process 2 the solvent is not necessary to be removed, so for this reason only a chromatographic column for separation of copper catalyst from product, whose price is 20 €/kg, is required. Processes 3 and 4 will have the same cost. It is because in both of the cases only a sieving step will be necessary, with a price of 1.25 €/kg. Finally the CWT cost of Process 5 is 1.62 €/kg which is corresponding to the sieving and cleaning steps.

In CRM parameter are also observed some differences among the processes. Whereas in the 3 first processes the cost will be same because of the catalyst is not reused. This difference of price is even more remarkable in the case of reused and cleaned copper wire, where CRM is 11.53 €/kg. It is due to in Process 4 the cost related to copper catalyst is lower due to it is reused (6.2 €/kg), being necessary a lower quantity of catalyst in the required reactions. This effect of copper catalyst saving is even more noticeable in the Process 5, because as the catalyst undergone to a washing step and it is reused, a lower quantity of catalyst is required in comparison to the commented previous processes, making possible to save a 32% of copper with respect to Process 4.

145

Chapter 3

As can be observed in Figure 3.13, the least economically interesting process is the one carried out at atmospheric pressure. It is because this process requires a wide waste treatment to separate THF and copper in powder form from the product. It is also remarkable that the process where supercritical technology is employed supposes a lower manufacturing cost in comparison to the process carried out at atmospheric pressure for the same volume of production. The main reason for this decrease of cost is because of the removal of solvent separation step.

140

/kg) 120 € 100

80 CRM 60 CWT COL 40 FCI

20 Process manufacturing Processmanufacturing cost( 0 1 2 3 4 5 Processes

Figure 3.13. Relation of different manufacturing cost processes.

With respect to the processes where the heterogeneous copper wire catalyst is used, it is worth highlighting that the manufacturing cost is as minimum a 10% lower in comparison to the process carried out at supercritical conditions using catalyst in powder form. This difference is mainly due to the reduction of waste treatment cost. The use of copper

146

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2 wire makes possible to separate the catalyst through a sieving step, something too much cheaper than the chromatographic column required in the case of catalyst in powder form.

Finally, according to the results obtained in Figure 3.13, it is observable that the copper wire reusability is an aspect which can increase or decrease the manufacturing cost depending on how it is applied. When copper wire is reused for one time with no cleaning step, the yield obtained in the second step decreases up to a value of 83%. This fact means that the saving made with the reusability of copper wire is lost as a result of the decrease of yield, obtaining higher COM values. Nevertheless, if copper wire is reused and cleaned, the save made in raw material cost supposes a critical factor in the manufacturing cost, being this process the most economically interesting.

3.4. Conclusions

The functionalization via click chemistry of polylactic acid (PLA) with coumarin in supercritical CO⁠2 has been achieved using copper wire as catalyst, obtaining similar yield to the reaction carried out with copper catalyst in powder form obtaining 0.00 ppm of catalyst in the final product. It was also observed that the use of this heterogeneous catalyst can provide some keys advantages which can enhance the click functionalization process in supercritical conditions. In this way, several parameters in this reaction were studied in order to determine the conditions where this reaction is more economically interesting.

A manufacturing cost study was carried out in order to compare different click chemistry reactions using copper catalyst in different forms. Supercritical technology was resulted in the most interesting technology in

147

Chapter 3 comparison to atmospheric pressure due to the removal of solvent elimination step, which supposed 3.75 €/kg. About the comparison between homogeneous or heterogeneous catalyst, it was observable that the savings made in waste treatment cost plays a crucial part in the manufacturing cost, being 16 times higher the waste treatment when a catalyst in powder form was chosen.

To conclude, the reusability and cleaning step of this heterogeneous catalyst was checked as the most interesting parameters in the manufacturing cost study. In this case the savings made in waste treatment and raw material costs made this process the most economically interesting among all the cases studied in this work with a COM of 108.97 €/kg.

148

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

3.5. Bibliography

1. Argentine, M.D., P.K. Owens, and B.A. Olsen, Strategies for the investigation and control of process-related impurities in drug substances. Advanced Drug Delivery Reviews, 2007. 59(1): p. 12-28. 2. Hein, C.D., X.-M. Liu, and D. Wang, Click Chemistry, A Powerful Tool for Pharmaceutical Sciences. Pharmaceutical Research, 2008. 25(10): p. 2216-2230. 3. Franc, G. and A.K. Kakkar, "click" methodologies: Efficient, simple and greener routes to design dendrimers. Chemical Society Reviews, 2010. 39(5): p. 1536-1544. 4. Lutz, J.F. and Z. Zarafshani, Efficient construction of therapeutics, bioconjugates, biomaterials and bioactive surfaces using azide-alkyne "click" chemistry. Advanced Drug Delivery Reviews, 2008. 60(9): p. 958- 970. 5. Gracia, E., et al., Functionalization and optimization of PLA with coumarin via click chemistry in supercritical CO2. Journal of CO2 Utilization, 2017. 20: p. 20-26. 6. A., B.C., et al., Modulating catalytic activity of polymer‐based cuAAC “click” reactions. Journal of Polymer Science Part A: Polymer Chemistry, 2011. 49(21): p. 4539-4548. 7. Urbani, C.N., et al., Convergent Synthesis of Second Generation AB- Type Miktoarm Dendrimers Using “Click” Chemistry Catalyzed by Copper Wire. Macromolecules, 2008. 41(4): p. 1057-1060. 8. Jiang, Y., et al., Metallic copper wire: a simple, clear and reusable catalyst for the CuAAC reaction in supercritical carbon dioxide. RSC Advances, 2015. 5(90): p. 73340-73345.

149

Chapter 3

9. Velencoso, M.M., et al., Click-ligation of coumarin to polyether polyols for polyurethane foams. Polymer International, 2013. 62(5): p. 783- 790. 10. Xue, Q., et al., Glass transition temperature of functionalized graphene–polymer composites. Computational Materials Science, 2013. 71: p. 66-71. 11. Yadav, A., et al., Glass transition temperature of functionalized graphene epoxy composites using molecular dynamics simulation. Vol. 184. 2018. 1-9. 12. Gracia, E., et al., Copper wire as a clean and efficient catalyst for click chemistry in supercritical CO2. Catalysis Today, 2018. 13. Mazarro, R., et al., Copolymerization of D,L-lactide and glycolide in supercritical carbon dioxide with zinc octoate as catalyst. Journal of Biomedical Materials Research - Part B Applied Biomaterials, 2008. 85(1): p. 196-203. 14. Zhang, W., et al., Cu(OAc)2·H2O—an efficient catalyst for Huisgen- click reaction in supercritical carbon dioxide. Tetrahedron Letters, 2015. 56(19): p. 2472-2475. 15. Jiang, Y., et al., An Efficient Synthesis of 1,4-Disubstituted Triazoles in Water via CuCl2/Zn-Catalyzed Huisgen Cycloaddition, in Zeitschrift für Naturforschung B. 2012. p. 226. 16. Hanai-Yoshida, V., et al., Supercritical fluid and pharmaceutical applications. Part I: Process classification. Vol. 10. 2016. 132-144. 17. Rosa, P.T.V. and M.A.A. Meireles, Rapid estimation of the manufacturing cost of extracts obtained by supercritical fluid extraction. Journal of Food Engineering, 2005. 67(1): p. 235-240.

150

Feasibility of copper wire as heterogeneous catalyst in click chemistry in scCO2

18. Perrut, M., Supercritical Fluid Applications: Industrial Developments and Economic Issues. Industrial & Engineering Chemistry Research, 2000. 39(12): p. 4531-4535.

151

Application of click 4 chemistry in scCO2 to PLA-derived Coumarin-release

systems

Chapter 4

4.1. Introducción 4.2. Experimental 4.2.1. Materials 4.2.2. Synthesis of of 3- (bromomethyl)coumarin 4.2.3. Synthesis of 3-azidomethyl coumarin 4.2.4. Synthesis of click product in scCO2 4.2.5. Characterization measurements 4.3. Results and discussion 4.3.1. Design of the PLA-coumarin adducts 4.3.2. 1H-NMR spectroscopic analysis 4.3.3. Maldi-Tof analysis 4.4. Conclusions 4.5. Bibliography A continuación se muestran de forma esquematizada los contenidos del capítulo 4 de resultados.

Liberación Liberación in vitro in vivo

Impregnaciones

scCO2

Liberación in vitro Abstract

T he position where the triazol ring is linked to the organic compound can play an important function in the final product due to the possible release limitations. For this reason in this chapter X-ray structures of co-crystalized coumarin-type ligands were studied in Maestro software in order to get to know the position where the molecule could have less steric hindrance issues in click chemistry.

Once the optimum carbon position was chosen, the click chemistry reaction of the modified coumarin in scCO2 was performed in both of the catalyst studied, copper acetate and copper wire obtaining yields over 91% in homogeneous and heterogeneous catalyst, respectively.

155

Chapter 4

4.1. Introduction

The goal of drug discovery is to find best medicines to prevent, treat and cure diseases quickly and efficiently. In this way, the structure-based drug predictions for drug design is an issue in which it has been made significant progress [1].

Among the wide variety of natural compounds with pharmaceutical properties studied, the natural compound studied in previous chapters is an excellent option to be investigated. The chosen of coumarin is not only due to the variety of positions where this organic compound can be modified, but also the pharmaceutical properties which make this natural compound an effective tool against important diseases like cancer. Coumarins are checked as compounds with antioxidant and cytostatic properties which could even reduce the side effects caused by anticancerous drugs reported in the market with cytotoxic properties [2]. The pharmacological properties displayed by coumarin derive from the 2H- chromen-2-one structure which, while small and simple, can engage in a number of specific interactions with protein targets. The lactone moiety induces ‘push-pull’ electronic delocalisation across the π system [3]. As a consequence of this, the scaffold features local areas of positive and negative electrostatic surface potential which impart selectivity to the hydrophobic π-type interactions [4]. Additionally, the lactone ring comprises two oxygen atoms which can accept hydrogen bonds from various amino acid residues.

Given the potent biological profile of the 2H-chromen-2-one scaffold, research in our laboratories has focused on novel delivery systems capable of controlled release of coumarin-type species. Our previous work 156

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems identified previously acetylene polylactic acid (PLA acetylene) as a potential carrier in Chapter 2. Furthermore, we have previously shown that the alkyne unit in PLA acetylene represents an excellent synthetic handle with which to attach coumarin species to the PLA backbone via a 1,2,3-triazole linker[5].

In previous chapters the coumarin chosen was linked in position 4 to the triazol ring, being 4-azidomethyl-7-methoxycoumarin the compound chosen. The position in which PLA acetylene is linked to the coumarin compound through the triazol ring could interfere in the pharmaceutical properties of the final compound due to steric hindrance issues [6]. For this reason, Schrodinger software Maestro was employed in this work in order to study the best position in coumarin structure to carry out the linkage of triazol ring. The coumarin study through this software made possible to obtain a position where a fewer steric hindrances concerns could be found in the further release step. In addition to the modified coumarin in position 4 previously synthetized in chapter 2, in this chapter position 3 has been determined as an optimum position to carry out the functionalization of this organic compound with the biodegradable polymer due to the interactions observed in X-ray structures in the protein crystal structure data base (PDB) that feature co-crystalized coumarin-type ligands. This study will allow to synthetize an additional coumarin compound which will present the fewest steric hindrances problems among the different possible coumarin structures.

157

Chapter 4

4.2. Experimental

4.2.1. Materials

Sodium azide (> 99,5%, Sigma-Aldrich), Polylactic Acid acetylene (98.3% Specific polymers), Salicylaldehyde (99%, Sigma-Aldrich), Propionic anhydride (99%, Sigma-Aldrich), Sodium propionate (99%, Sigma-Aldrich), Trimethylamine (99%, Sigma-Aldrich), NBS (99%, Sigma-Aldrich), Sodium sulphate (99%, Sigma-Aldrich), Ethanol (99%, Sigma-Aldrich), Ethyl acetate (98%, Sigma-Aldrich), Copper wire (99%, Sigma-Aldrich), Acetonitrile (99.8%, Sigma-Aldrich), Heptane (99%, Sigma-Aldrich), Sodium hydroxide (99%, Sigma-Aldrich), Ammonium chloride (99%,

Sigma-Aldrich), Copper(II) acetate monohydrate (Cu(CO2CH3)2·H2O) (> 99%, Sigma-Aldrich), Nitrogen (99.5%, Carburos metálicos, S.A. Spain) and Carbon dioxide (Carburos metálicos, S.A., Spain) with a purity of 99.5%. All other reagents and solvents used in the study were of analytical grade and used as delivered.

4.2.2. Synthesis of 3-(bromomethyl)coumarin

The synthesis of 3-(Bromomethyl)coumarin) was carried out following a bibliographic procedure [7]. In this procedure A 250-mL round bottom flask was charged with 0.5 mol of 2-Hydroxybenzaldehyde, 0.25 mol of

CH3CH2COONa, 0.75 mol of (CH3CH2CO)2O and 0.25 mol of Et3N, then the mixture was refluxed for 8h. After cooling down to room temperature, the reaction was washed with 40% NaOH, sat. NH4Cl, H2O and brine respectively, and then dried over anhydrous Na2SO4. After concentrated, the product was recrystalyized with 95% EtOH. 3-(methyl)coumarin was obtained in 65% yield. 30 mmol of 3-(methyl)coumarin was added to 120 mL of CH3CN in a 250 mL-round bottom flask, the mixture was then heated to 85 ºC. 30 mmol of NBS (N-bromosuccinimide) and 250 mg of BPO 158

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems

(Peroxidation benzoin formyl) were added to the mixture in portion and the reaction was kept string at reflux for 24 hours. After cooling down to room temperature, the reaction was washed with 40% NaOH, sat. NH4Cl,

H2O and brine respectively, dried over anhydrous Na2SO4. After concentrated, the product was recrystallized with toluene. 4.2.3. Synthesis of 3-azidomethyl coumarin In this case the azidation procedure followed is the same than the one detailed in 2.2.2 [8].

4.2.4. Synthesis of click product in scCO2 It was previously reported the synthesis of click product using 4- azidomethyl-7-methoxycoumarin via the CuAAC coupling of PLA acetylene and the corresponding 4-azidomethyl coumarin in chapters 2 and 3.

Notably, this reaction is performed in supercritical CO2 (scCO2) using copper sulphate and copper wire as the catalyst in the absence of additional ligands. The excellent efficiency of this procedure and the advantages surrounding patient compatibility, sustainability and operational facility associated with the use of scCO2 as a solvent [9-15] and copper metal as a catalyst [16-18].

4.2.5. Characterization measurements

Nuclear Magnetic Resonance (NMR) was measured with Varian Gemini

FT-400 spectrometer using CDCl3 as solvent.

159

Chapter 4

Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) was carried out using a Bruker Autoflex II

TOF/TOF spectrometer (Bremen, Germany) using CDCl3 as solvent and dithranol (1,8,9-trihydroxyanthracene) as matrix material.

4.3. Results and Discussion

4.3.1 Design of the PLA-coumarin adducts Impregnation of the drug within the polymer is expected to deliver rapid dosage via desorption or related mechanisms, whereas covalent linkage of the drug to the polymer backbone delivers the drug through a polymer breakdown pathway and offers the advantage of controlled and time-stable dosage of the drug. Copper-catalyzed azide-alkyne cycloaddition (CuAAC) was previously identified by us as an efficient way to covalently link azide-functionalised coumarin derivatives and PLA acetylene via the triazole linker [19]. Crucially, this linker involves carbon- carbon bonds and the stable triazole ring. As such the coumarin is connected to the linker via a bond of high stability ensuring polymer decomposition and not linker decomposition is the prime route leading to drug release. The position of the triazole linker is expected to be key in maintaining the native activity and selectivity of the released coumarin species. Should the linker be placed such that it interfered with a key π-type interaction or hydrogen bond with the target enzyme, then significant and likely undesirable alterations in activity would be expected. For this reason, when designing the coumarin-PLA adducts we consulted X-ray structures in the Protein crystal structure Data Base (PDB) that feature co- crystalized coumarin-type ligands (Figure 4.1). 160

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems

Figure 4.1. X-ray structure of coumarin obtained from Maestro Schrodinger software.

All co-crystal structures featuring coumarin-type ligands that exhibit minimal structural deviation from native 2H-chromen-2-one were investigated. For example, ligands featuring non-coumarin heterocyclic rings were not included in the study. This study suggested that the all- carbon benzene moiety of the coumarin scaffold is typically buried in the protein structure. As such, linkage to this ring was ruled out. Of the two remaining positions, functionalisation at the 3-position appeared as the best option to minimise disturbance of coumarin-enzyme interactions. However, depending on the co-crystallised enzyme, all possibilities appear to disrupt key hydrogen bonding or hydrophobic interactions to some extent. As such, two coumarin/PLA composite materials were studied, one featuring a triazole linker between the PLA backbone and the 3-position of the coumarin scaffold PLA-C3 and another featuring the same linker to the 4-position PLA-C4 (Figure 4.2). It is envisaged that application of these positional isomers to biological assays will reveal which linkage position is preferred for specific protein targets. Furthermore, from the perspective of this study, these two coumarin species allow us to probe the role, if any, of linker position in coumarin release from the PLA carrier.

161

Chapter 4

Figure 4.2. Covalently linked PLA-coumarin adducts studied in this report.

PLA acetylene was efficiently coupled to 3-azidomethyl coumarin C3. C3 was synthesized according to literature procedures through application of our previously identified conditions [20]. When copper wire was employed as the catalyst PLA-C3 was obtained in an excellent 93% yield (Figure 4.3). Corresponding analysis are shown in the following points 4.3.2 and 4.3.3. PLA-C4 was not synthetized due to the low yield obtained in the reaction of the coumarin compound.

24 h

Figure 4.3. Covalent linkage of coumarin C3 to PLA acetylene to give PLA-C3.

As can be observed in Table 4.1, all the experiments were carried out successfully obtaining more than 90% of yield in the cases of 48 h of click reaction using copper acetate and 24 h of reaction using copper wire. These 162

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems results show the versatility of click chemistry when different coumarin compounds are synthetized. Being the yield obtained in each experiment analysed with 1H NMR and Maldi-Tof methods.

Table 4.1. Relation of experiments carried out in scCO2.

Reaction t (h) T (ºC) P (bar) Catalyst Catalyst loading (%) 1 24 47 110 Copper acetate 12.1 2 48 47 110 Copper acetate 11.6 3 24 50 130 Copper wire 70

4.3.2 1H-NMR spectroscopic analysis

Polymer-coumarin adduct PLA-C3 was analyzed by 1H-NMR spectroscopy to confirm the success of the CuAAC process with both catalysts, copper acetate and copper wire (Figure 4.3). Disappearance of the signal at 2.51 ppm corresponding to the acetylenic proton H2 in PLA acetylene and the presence of the signal at 7.88 ppm corresponding to the triazole proton H2’ in PLA-C3 confirms formation of the triazole linkage. Signals corresponding to PLA and coumarin C3 methylene protons, H1 and H3a&b respectively, are shifted significantly downfield in the polymer adduct PLA-C3. In the case of H1 the shift is over 1 ppm from 4.39 ppm in free coumarin C3 to 5.47 ppm for H1’ in PLA-C3. Diastereotopic methylene protons H3a and H3b in PLA acetylene are shifted from ≈4.75 ppm to ≈ 5.28 ppm for Hs3’a & 3’b in PLA-C3. These notable shifts are a consequence of the ring current imparted by the triazole ring and are further evidence of the successful formation of the triazole linker. To our knowledgement, it is the first time where this software is used for the determination of adducts feasibility in click chemistry.

163

Chapter 4

Figure 4.3. 1H-NMR spectroscopic analysis of a) PLA acetylene, b) C3 and c) PLA-C3. Proton assignments correspond to those shown in Figure 3.

PLA-C3 features two distinct end groups: on one side of the polymer chain the 2° alcohol and on the other the triazole-linked coumarin moiety. The integral ratios for the signals belonging to the coumarin and/or PLA components of both end groups display the expected integer ratios, strongly indicating complete conversion of the acetylene and azide

164

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems functionalities to the triazole linker. The slightly higher than expected integration value for the diastereotopic methylene protons H3’a and H3’b (2.5 vs 2.0 expected) is the consequence of overlap with the signal belonging to the PLA methyne protons. The good match of integrals for signals corresponding to the coumarin moiety (aromatic signals and methylene protons H1) and H4’ is particularly notable as it highlights that PLA-C3 possesses equal populations of the both ‘end groups’. Within the sensitivity limits of 1H-NMR spectroscopy this suggests that >90% of polymer chains within the material comprising PLA-C3 have undergone the CuAAC reaction.

Besides the reaction carried out with copper wire as the catalyst, it was also checked with copper acetate used in Chapter 2 as the catalyst. In this case the reaction was also successfully achieved, being demonstrated the versatility of click chemistry with different catalysts when the organic compound, in this case coumarin, changes its structure.

4.3.3. Maldi-Tof analysis

The yields obtained in each reaction were determined as in previous chapters through Maldi-Tof analysis [5, 21]. In this case it is remarkable that the molecular weight of the organic compound is the corresponding to 3-azidomethyl coumarin (Figure 4.4), which is 201 g/mol, being expected an increase in the click product molecular weight of 201 g/mol in comparison with the peaks corresponding to PLA acetylene.

165

Chapter 4

Figure 4.4. Chemical structure of 3-azidomethylcoumarin.

Maldi-Tof mass spectra of the PLA acetylene and the coumarin functionalized one are shown in Fig. 4.5. In Figure 4.5 b), c) and d) it is observed that there is a displacement of 201 g/mol in all PLA-C3 products, which is exactly the molecular weight of 3-azidomethyl coumarin, in comparison with the peaks of PLA acetylene (Figure 4.5 a)), fact which confirms that the click reactions have been successfully carried out. The procedure to calculate the yield is the same as was used in Chapters 2 and 3, where it was calculated through the comparison of intensities corresponding to the peaks to PLA acetylene with respect to the peaks of coumarin functionalized PLA (Table 4.2).

Table 4.2. Relation of yields obtained in click chemistry reactions.

Catalyst Reaction Catalyst t (h) T (ºC) P (bar) Yield (%) loading (%) 1 Copper acetate 24 47 110 12.1 82.36 2 Copper acetate 48 47 110 11.6 91.74 3 Copper wire 24 50 130 70 92.67

166

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems

a) IV Mw: 1958 III Mw: 1815 V Mw: 2102 II Mw: 1670 VI Mw: 2246

VII Mw: 2390 I Mw: 1525

b) IV* Mw: 2158 III* Mw: 2014 V* Mw: 2302 VI* Mw: 2446 II* Mw: 1870 VII* Mw: 2590 I* Mw: 1726

IV* Mw: 2158 c) V* Mw: 2302 III* Mw: 2014 II* Mw: 1870 VI* Mw: 2446 I* Mw: 1726 VII* Mw: 2590

d) IV* Mw: 2158 III* Mw: 2014 V* Mw: 2302 II* Mw: 1870 VI* Mw: 2446 VII* Mw: 2590 I* Mw: 1726

1.300 1.500 1.700 1.900 2.100 2.300 2.500 2.700 2.900

Figure 4.5. Maldi-Tof analysis of a) PLA acetylene and b), c), d) PLA-C3 products.

167

Chapter 4

As can be observed in Table 4.2 and Figure 4.5 the PLA-C3 products were obtained with high yield using copper acetate as well as copper wire as the catalysts. Reaction conditions were chosen according to the optimized conditions obtained in previous chapters, where both of the catalyst were used.

4.4. Conclusions

This study suggested that the all-carbon benzene moiety of the coumarin scaffold is typically buried in the protein structure. As such, linkage to this ring was ruled out, in this study Maestro software has been checked as an excellent software to predict feasible adduct synthesis of PLA-C3 for coumarin structure. In this way, biodegradable polymer was successfully coupled to 3-azidomethyl coumarin in supercritical conditions using copper acetate as well as copper wire as the catalysts with yields over 91% in both cases.

168

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems

4.5. Bibliography

1. Sun, H. and D.O. Scott, Structure-based drug metabolism predictions for drug design. Chemical Biology and Drug Design, 2010. 75(1): p. 3-17. 2. Marshall, M., K. Butler, and D. Hermansen, Structural modification of coumarin for increased Anti-Coagulation potency. Prostate, 1990. 17: p. 95-108. 3. Torres, F.C., et al., New Insights into the Chemistry and Antioxidant Activity of Coumarins. Current Topics in Medicinal Chemistry, 2014. 14(22): p. 2600-2623. 4. Cockroft, S.L., et al., Electrostatic Control of Aromatic Stacking Interactions. Journal of the American Chemical Society, 2005. 127(24): p. 8594-8595. 5. Gracia, E., et al., Copper wire as a clean and efficient catalyst for click chemistry in supercritical CO2. Catalysis Today, 2018. 6. Pingaew, R., et al., Synthesis, biological evaluation and molecular docking of novel chalcone-coumarin hybrids as anticancer and antimalarial agents. European Journal of Medicinal Chemistry, 2014. 85: p. 65-76. 7. Tan, Y., et al., Novel one-pot asymmetric cascade approach toward densely substituted enantioenriched α-methylene-γ-lactams. Tetrahedron Letters, 2014. 55(44): p. 6105-6108. 8. Velencoso, M.M., et al., Click-ligation of coumarin to polyether polyols for polyurethane foams. Polymer International, 2013. 62(5): p. 783- 790.

169

Chapter 4

9. Grodowska, K. and A. Parczewski, Organic solvents in the pharmaceutical industry. Acta Poloniae Pharmaceutica - Drug Research, 2010. 67(1): p. 3-12. 10. Argentine, M.D., P.K. Owens, and B.A. Olsen, Strategies for the investigation and control of process-related impurities in drug substances. Advanced Drug Delivery Reviews, 2007. 59(1): p. 12-28. 11. Bohrer, D., Sources of Contamination in Medicinal Products and Medical Devices. Sources of Contamination in Medicinal Products and Medical Devices. 2012. 1-571. 12. Dixit, K., R.B. Athawale, and S. Singh, Quality control of residual solvent content in polymeric microparticles. Journal of Microencapsulation, 2015. 32(2): p. 107-122. 13. Henderson, R.K., et al., Expanding GSK's solvent selection guide - embedding sustainability into solvent selection starting at medicinal chemistry. Green Chemistry, 2011. 13(4): p. 854-862. 14. Fages, J., et al., Particle generation for pharmaceutical applications using supercritical fluid technology. Powder Technology, 2004. 141(3): p. 219-226. 15. Jung, J. and M. Perrut, Particle design using supercritical fluids: Literature and patent survey. The Journal of Supercritical Fluids, 2001. 20(3): p. 179-219. 16. Jiang, Y., et al., Metallic copper wire: a simple, clear and reusable catalyst for the CuAAC reaction in supercritical carbon dioxide. RSC Advances, 2015. 5(90): p. 73340-73345. 17. Himo, F., et al., Copper(I)-Catalyzed Synthesis of Azoles. DFT Study Predicts Unprecedented Reactivity and Intermediates. Journal of the American Chemical Society, 2005. 127(1): p. 210-216.

170

Application of click chemistry in scCO2 to PLA-derived Coumarin-release systems

18. Gommermann, N., A. Gehrig, and P. Knochel, Enantioselective Synthesis of Chiral α-Aminoalkyl-1,2,3-triazoles Using a Three-Component Reaction. Synlett, 2005. 2005(18): p. 2796-2798. 19. G., A.S., M.S. R., and P.V. S., Click Chemistry: 1,2,3-Triazoles as Pharmacophores. Chemistry – An Asian Journal, 2011. 6(10): p. 2696- 2718. 20. Ye, X.-W., et al., Synthesis and biological evaluation of coumarin– 1,2,3-triazole–dithiocarbamate hybrids as potent LSD1 inhibitors. MedChemComm, 2014. 5(5): p. 650-654. 21. Gracia, E., et al., Functionalization and optimization of PLA with coumarin via click chemistry in supercritical CO2. Journal of CO2 Utilization, 2017. 20: p. 20-26.

171

Drug release profile of click products in comparison with 5 impregnations carried out in scCO2

Chapter 5

5.1. Introducción 5.2. Experimental 5.2.1. Materials 5.2.2. Synthesis of 4-azidomethyl-7- methoxycoumarin 5.2.3. Synthesis of 3-azidomethyl coumarin 5.2.4. Synthesis of click product in scCO2 5.2.5. Impregnation of coumarin in scCO2 5.2.6. Purification of click products 5.2.7. In vitro release study 5.2.8. Characterization measurements 5.3. Results and discussion 5.4. Conclusions 5.5. Bibliography

A continuación se muestran de forma esquematizada los contenidos del capítulo 5 de resultados.

Liberación in vivo

scCO2

Catalizador en fase homogénea

Abstract

ifferent click products have been previously synthetized using

Dsupercritical carbon dioxide as solvent. In this chapter the main purpose is to study the drug release profile of click products in comparison to a coumarin compound impregnated in order to get to know the advantages given by the covalent bond to the drug release.

Results obtained show that click chemistry products when are undergone to in vitro experiments show a release 33% longer than impregnations where there is not any covalent bond between organic compound an biodegradable polymer.

175

Chapter 5

5.1. Introduction

The design of polymer carriers is of high importance for a range of biomedical applications where is it is highlighted drug delivery application. Depending on the type of linkage between the organic compound with pharmaceutical properties and the biodegradable polymer, different drug release profiles can be obtained [1-5]. An example can be found in the case of polymers impregnated with organic compounds with pharmaceutical properties [6-9], where 3 different steps are expected in the drug release profile[10, 11]. The first step would correspond to the initial burst of drug release of the most accessible drug, which is generally located on the surface of the compound and this process is controlled by diffusion in the film. The second step is in this case controlled by the internal diffusion into the most tortuous or narrow pores. Finally the last step in this type of drug release is controlled by the degradation of the polymer, where according to the degradation rate of polymer the drug will be released in the media[10]. Another example is found in functionalized polymers, where a covalent bond is the responsible of the linking between the drug and the biodegradable polymer[12-14]. This type of bond between both molecules makes possible to expect a more controlled release in comparison to the case previously commented. Among the different types of polymer functionalization, copper-catalyzed 1,3-dipolar cycloaddition between azides and alkynes, also called click chemistry[15, 16], is finding application in drug delivery in different type of molecules, being possible to be carried out under benign conditions[17-20]. The drug release profile obtained is even more stable when the reaction is performed in supercritical conditions, specifically in scCO2[21]. This fact

176

Drug release profile of click products in comparison with impregnations carried out in

scCO2 is mainly due to the plasticization effect which makes possible to the drug being almost in full impregnated in the inside part of the polymer structure[22]. This is an important advantage with respect to the conventional impregnation methods carried out at atmospheric pressure where an important quantity of drug is on the surface of the polymer, being released in the first minutes of the release. Another worth mentioning benefit is the clean atmosphere where the impregnation or functionalization of polymer is carried out. This factor is essential when a product is destined to a pharmaceutical application, being scCO2 a solvent as was previously commented suitable for this purpose because of its lack of reactivity, good transport properties or its sterilization capacity[23, 24]. In this chapter the release profile comparison between impregnated and click chemistry products were carried out. For this purpose the two different compounds of coumarin described in chapters 3 and 4 were chosen, both of them synthetized using copper wire as catalyst. Coumarin is well known for its pharmacological properties[25], something which makes this natural compound an excellent option for its application in drug delivery.

5.2. Experimental

5.2.1. Materials Sodium azide (> 99,5%, Sigma Aldrich), 4-Bromomethyl-7-methoxycoumarin (97%, Sigma Aldrich), Polylactic Acid (99.5%, Corbion), Polylactic Acid acetylene (98.3%, Specific polymers), Propionic anhydride (99%, Sigma-Aldrich), Sodium propionate (99%, Sigma-Aldrich), Trimethylamine (99%, Sigma-Aldrich), NBS (99%, Sigma-Aldrich), Sodium sulphate (99%, 177

Chapter 5

Sigma-Aldrich), Ethanol (99%, Sigma-Aldrich), Ethyl acetate (98%, Sigma- Aldrich), Acetonitrile (99.8%, Sigma-Aldrich), Heptane (99%, Sigma- Aldrich), Sodium hydroxide (99%, Sigma-Aldrich), Ammonium chloride (99%, Sigma-Aldrich) Acetone (99.8%, VWR), Copper wire (99%, Sigma- Aldrich), Nitrogen (99.5%, Carburos metálicos, S.A., Spain) and Carbon dioxide (99,5%, Carburos metálicos, S.A., Spain) with a purity of 99.5%. All other reagents and solvents used in the study were of analytical grade and used as delivered. 5.2.2. Synthesis of 4-azidomethyl-7-methoxycoumarin The synthesis of this compound was carried out following a bibliographic route detailed in Chapter 2 (1.2.1)[26]. 5.2.3. Synthesis of 3-azidomethyl coumarin The synthesis of this compound was carried out following the procedure detailed in Chapter 4 (3.2.2 and 3.2.3).

5.2.4. Synthesis of click product in scCO2

The synthesis of click products in scCO2 was carried out following the same procedure than was described in chapters 3 and 4 (3.2.3 and 4.2.4). The operational conditions were 50ºC and 130 bar, the polymer functionalizations were completed after 24 hours of reaction using copper wire as catalyst for both reactions. Once the polymer was functionalized, it was impregnated in PLA using supercritical conditions in order to obtain a foam structure which is going to be released in in vitro release experiment.

5.2.5. Impregnation of coumarin in scCO2 In order to compare the release of coumarin impregnated and functionalized, the organic compound was impregnated with PLA using supercritical conditions (scCO2). In the case of coumarin impregnated 10.55 g of coumarin were used in the impregnation using 3 ml of acetone and 189.45 g of PLA. 178

Drug release profile of click products in comparison with impregnations carried out in

scCO2

Whereas when functionalized polymer is mixed with PLA, 100 mg of click product was added to 100 mg of pure PLA and 3 ml of acetone, in order to obtain the same coumarin concentration than in the case of coumarin impregnated.

The impregnation procedure in scCO2 is the same than was described previously in chapter 1 (1.2.6). 5.2.6. Purification of click products Purification of click product was carried out following the same procedure described in chapter 3 (3.2.4). 5.2.7. In vitro release study Polymers impregnated with coumarin were suspended in a phosphate saline solution (PBS) 0.1 M (pH 7.4, 1 M), placed in the middle of a 100 mL flask hermetically closed and preserved from light, stirred at 100 rpm, and incubated in a shaking water bath at 37 °C. 5 ml solution was periodically removed from the flask in order to measure by UV spectrophotometry the quantity of coumarin released at an absorbance value of 320 nm. 5.2.8. Characterization measurements The amount of drug released and the quantity of coumarin impregnated was determined using a UV–Vis apparatus (Shimadzu UV- 1603, Germany) with a spectral range from 190 to 1100 nm, halogen and deuterium lamps and a silicon photo diode detector. It was provided with the software UVPC Personal Spectroscopy Software, Version 3.6.

179

Chapter 5

Synthesized lightweight gypsum composites were depicted by means of Scanning Electron Microscopy (SEM) by using a FEI QUANTA 250. EDX analysis was carried out using a EDAX APOLLO X.

Nuclear Magnetic Resonance (NMR) was measured with Varian Gemini

FT-400 spectrometer using CDCl3 as solvent.

Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) was carried out using a Bruker Autoflex II

TOF/TOF spectrometer (Bremen, Germany) using CDCl3 as solvent and dithranol (1,8,9-trihydroxyanthracene) as matrix material.

5.3. Results and discussion

In vitro and in vivo drug releases are the previous steps to the real application of click chemistry with a pharmaceutical purpose. In previous chapters promising results were obtained when curcumin was impregnated in scCO2, claiming supercritical technology as an excellent alternative for impregnation processes[21]. In this way, PLA was functionalized using click chemistry[27], where copper wire was indicated as an excellent catalyst to carry out this reaction because of its total removal from the click product obtaining yields over 90% of functionalized polymer[28].

For the reasons previously commented, in this chapter different experiments were carried out in order to compare release profiles of PLA functionalized via click chemistry using coumarin compounds of chapters 3 and 4 with respect to impregnation of PLA using coumarin in scCO2 .

The first step in this work was the impregnation of PLA with coumarin. In order to obtain approximately the same concentration of coumarin in the impregnation of pure coumarin than in the impregnation of coumarin

180

Drug release profile of click products in comparison with impregnations carried out in

scCO2 functionalized, 10.55 g of coumarin were added to 189.45 mg of pure PLA, obtaining a total amount of 200 mg of product, which was impregnated in scCO2.

Table 1. PLA impregnation with pure coumarin in scCO2.

Experiment PLA (mg) Coumarin (mg) T (ºC) P (bar) t (h) 1 189.45 10.55 50 150 8

The impregnation time chosen for this experiment was 8 h according to the optimized impregnation process previously studied when acetone was used as solvent in curcumin impregnation. In this case the quantity of acetone used is even lower than in the case of curcumin what means that with the same impregnation time the quantity of acetone will be totally removed from the final product in the depressurization process.

The total amount of coumarin impregnated was calculated with UV technology in the typical absorbance value of this organic compound, 320 nm. The impregnation efficacy of coumarin impregnation is shown in Table 2.

Table 2. Impregnation efficacy of PLA with coumarin in scCO2.

Initial Coumarin Impregnation Experiment coumarin (mg) impregnated (mg) efficacy (%) 1 10.55 10.02 94.97

Once coumarin was successfully impregnated in PLA using supercritical conditions with CO2 as solvent, the next step was to carry out the polymer click functionalizations with 4-azidomethyl-7-methoxycoumarin and 3-

181

Chapter 5 azidomethyl coumarin. In Table 3 both experiments with their operational conditions are shown.

Table 3. Click chemistry experiments carried chosen for drug release process.

Experiment Organic compound P (bar) T (ºC) t (h) Yield (%) 4-azidomethyl-7- 2 130 50 24 92.99 methoxycoumarin 3-azidomethyl 3 130 50 24 92.73 coumarin

In both cases the yield obtained in the click product is very similar, around 93%, being this number essential in order to get to know the quantity of click product needed to obtain the same coumarin concentration than in the previous sole-impregnation experiment. Yield was obtained through Maldi-Tof analysis, being 1H NMR also used for the characterization of click products.

182

Drug release profile of click products in comparison with impregnations carried out in

scCO2

ppm

Figure 1. 1H NMR of click product corresponding to a) 4-azidomethyl-7- methoxycoumarin; b) 3-azidomethyl coumarin.

183

Chapter 5

1.300 1.500 1.700 1.900 2.100 2.300 2.500 2.700 2.900 m/z

1.300 1.500 1.700 1.900 2.100 2.300 2.500 2.700 2.900

m/z Figure 2. Maldi-Tof analysis of click products corresponding to a) 4-azidomethyl- 7-methoxycoumarin; b) 3-azidomethyl coumarin

According to results obtained in Figures 2 and 3 the functionalizations were successfully carried out. It is observed that in 1H NMR analysis there

184

Drug release profile of click products in comparison with impregnations carried out in

scCO2 are the same peaks that were found in chapters 3 and 4 when both coumarin click products were synthetized. With respect to Maldi-Tof analysis, it is remarkable in Figure 2 a) the maximum peak corresponds to the signal at Mw: 2045, whereas in b) the maximum peak does not correspond to this value, it is because the different molecular weight of 4- azidomethyl-7-methoxycoumarin and 3-azidomethyl coumarin. Fact which confirms the successful functionalization of both compounds with PLA in scCO2.

The product form is an essential parameter to carry out drug delivery. The powder form presented in click products makes impossible to be released in a comparable manner with coumarin impregnated. For this reason click product was impregnated in pure PLA in order to obtain a foam from which the organic compound will be released, which will mean a dissolution of the organic compound in PBS since the beginning of the release experiment.

In Table 4 are shown the impregnation reactions between PLA functionalized and pure PLA in scCO2.

Table 4. Impregnations of click products with PLA using scCO2.

Impregnation Experiment Click product (mg) PLA (mg) efficacy (%) 4 100 10 94.77 5 100 10 95.56

The amount of click product impregnated in pure PLA in Table 4 was determined on the same way than was previously explained in the case of

185

Chapter 5 coumarin impregnated using UV methodology in its absorbance value, 320 nm.

These click products impregnated were also characterized by scanning electron micrograph (SEM). The desirable pore size depends on the foam final application. In this way, if the foam is used for tissue regeneration, higher pore sizes are preferred in order to encourage cellular infiltration and allow for the proper exchange of nutrients[29, 30]. If the foam purpose is the one desired for this work, smaller pore size are requested in order to obtain sustained drug concentrations and a slow and continue release[31].

In Figure 3 they can be observed the images corresponding to both functionalized coumarin products.

Figure 3. Scanning electron micrograph of a)4-azidomethyl-7-methoxycoumarin and b) 3-azidomethyl coumarin functionalized with PLA.

186

Drug release profile of click products in comparison with impregnations carried out in

scCO2

An important parameter to be studied is the component distribution in the foam. For this purpose EDX analysis was performed studying the distribution in the foam of carbon, hydrogen and nitrogen. The two first compounds are presented in both polymer and organic compounds, but nitrogen is only presented in the coumarin structure, so this component distribution in the foam will be a key parameter for the determination of functionalized polymer distribution in the foam structure.

Carbon Nitrogen Hydrogen a)

Figure 4. EDX analysis of a) 4-azidomethyl-7-methoxycoumarin and b) 3- azidomethyl coumarin.

As can be observed in Figure 4, carbon and hydrogen are presented in the foam strucutre in a higher concentration according to the numer of points related to these components. It was expected to find nitrogen in a

187

Chapter 5 lower concentration because of nitrogen component is only presented in the triazol ring structure.

According to the results obtained for nitrogen, it is observed the proper distribution of functionalized polymer in the whole foam structure, fact which is an indicator for a constant quantity of organic compound released during the in vitro drug release experiment.

Step I Step II Step III

100

80 Step III

Step II

Step I

20 Cumulative drug release (%)

0 Time (days)

Figure 5. Scheme of a typical drug release profile and a) its three different graphical steps from a porous biodegradable polymer [10]; b) the three different steps in a release profile.

188

Drug release profile of click products in comparison with impregnations carried out in

scCO2

In Figure 5 is observed a typical drug release profile which is composed of three different steps as was previously explained. In the case of drugs impregnated in scCO2 it was observed a constant release profile where the most representative phase was the step III, corresponding to the polymer degradation[21], whereas steps I and II are almost inexistent due to the constant release profile showed in the in vitro release experiment.

Hence, the coumarin and coumarin functionalized products experiments showed in this chapter were undergone to an in vitro release study in order to determine their release profile. The cumulative percentage release was studied during 30 days, which is the period of time necessary for a total release of coumarin presented in the foam structures.

100

80 4-azidomethyl-7- methoxycoumari 60 n 3-azidomethyl coumarin

40 Coumarin impregnated in scCO2 20 Typical 3 steps drug release profile

Cumulative percentagerelease (%) 0 0 5 10 15 20 25 30 Time (days)

Figure 6. Cumulative percentage release of impregnated coumarins.

In Figure 6 it is represented the drug release profile of the different impregnated products during 30 days. In the case of a typical drug release 189

Chapter 5 profile composed of 3 steps (green colour), it is observable that in the two first steps the 90% of drug is released, being the whole of drug released by the 10th day of release. However, as it can be observed in the case of coumarin impregnated (purple colour), the whole quantity of coumarin is released approximately after 20 days of almost constant release. About the kinetic release observed for this product, it is remarkable the existence of only one step which corresponds to the degradation of biodegradable polymer, fact which was previously observed in other work using supercritical technology[21].

Regarding the click products impregnated in PLA, it is observable that in both of them coumarin is totally released at 27th day of in vitro experiment. The difference of days about the release with respect to the coumarin impregnated previously commented is mainly due to the type of linking between the organic compound and the biodegradable polymer. In this case the covalent bond related to the click products allows a slower and more controlled and uniform release than in a pure impregnation.

In both cases, pure impregnation and click product impregnated, there is a homogeneous distribution of the drug into the polymeric matrix as consequence of the use of scCO2 in the impregnation process. This advantage means a long constant-high release step.

Release kinetics was modelled using equation (1), where M0 and Mf represent the total mass at the beginning and at the end of the release experiment, respectively. R0 is the initial radius of the spherical foam (0.2 cm) and kdegr is the pseudo-first kinetic constant of degradation for the PLA foam.

1/3 (푀푓/푀0) = 1 − 푘푑푒푔푟/푅0 (1)

190

Drug release profile of click products in comparison with impregnations carried out in

scCO2

The constants corresponding to the degradation step for both experiments was determined using equation 1. Results are shown in Table 5.

Table 5. Determination of kdegr for release kinetic model.

Drug released Release days Kdegr (cm-1) Coumarin 20 0.157(0.0028) 4-methyl-7-methoxycoumarin 27 0.173(0.0019) 3-methyl coumarin 27 0.175(0.0017) 3-steps typical release profile 1 0.00827 [10]

Values of kdegr obtained in Table 5 are higher for experiments where click products were impregnated in comparison with coumarin impregnated. Values into the arrows correspond to standard deviations. This trend was expected to be the desirable behaviour for controlled release in pharmacological applications. This trend is even more remarkable if the comparison is carried out with a release experiment where three different steps are observed.

5.4. Conclusions

Click chemistry has been shown as a useful functionalization method for biodegradable polymer for its excellent results. The results obtained in this chapter regarding the in vitro release study also confirms that click products are excellent candidates to be used in pharmaceutical applications. The use of functionalized products via click chemistry showed a release more controlled in comparison with coumarin impregnated

191

Chapter 5 without any type of covalent bond between biodegradable polymer and organic compound. According to the results obtained for the same concentration of initial coumarin 10 days more were needed in the case of click products to release the whole quantity of coumarin in the initial product. This constant-high release of drug is the most desired profile for medical applications.

192

Drug release profile of click products in comparison with impregnations carried out in

scCO2

5.5. Bibliography

1. Berger, J., et al., Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. European Journal of Pharmaceutics and Biopharmaceutics, 2004. 57(1): p. 19-34. 2. Chourasia, M.K. and S.K. Jain, Pharmaceutical approaches to colon targeted drug delivery systems. Journal of Pharmacy and Pharmaceutical Sciences, 2003. 6(1): p. 33-66. 3. Wu, W., et al., Covalently combining carbon nanotubes with anticancer agent: Preparation and antitumor activity. ACS Nano, 2009. 3(9): p. 2740-2750. 4. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with 5-fluorouracil in supercritical CO2. The Journal of Supercritical Fluids, 2013. 80: p. 1-8. 5. Cabezas, L.I., et al., Production of biodegradable porous scaffolds impregnated with indomethacin in supercritical CO2. The Journal of Supercritical Fluids, 2012. 63: p. 155-160. 6. Mishima, K., Biodegradable particle formation for drug and gene delivery using supercritical fluid and dense gas. Advanced Drug Delivery Reviews, 2008. 60(3): p. 411-432. 7. Türesin, F., I. Gürsel, and V. Hasirci, Biodegradable polyhydroxyalkanoate implants for osteomyelitis therapy: In vitro antibiotic release. Journal of Biomaterials Science, Polymer Edition, 2001. 12(2): p. 195-207. 8. Guney, O. and A. Akgerman, Synthesis of controlled-release products in supercritical medium. AIChE Journal, 2002. 48(4): p. 856-866.

193

Chapter 5

9. Liu, H., N. Finn, and M.Z. Yates, Encapsulation and sustained release of a model drug, indomethacin, using CO 2-based microencapsulation. Langmuir, 2005. 21(1): p. 379-385. 10. Cabezas, L.I., et al., Novel Model for the Description of the Controlled Release of 5-Fluorouracil from PLGA and PLA Foamed Scaffolds Impregnated in Supercritical CO2. Industrial & Engineering Chemistry Research, 2014. 53(40): p. 15374-15382. 11. Cabezas, L.I., et al., Validation of a Mathematical Model for the Description of Hydrophilic and Hydrophobic Drug Delivery from Biodegradable Foams: Experimental and Comparison Using Indomethacin as Released Drug. Industrial & Engineering Chemistry Research, 2014. 53(21): p. 8866-8873. 12. Bhattarai, N., J. Gunn, and M. Zhang, Chitosan-based hydrogels for controlled, localized drug delivery. Advanced Drug Delivery Reviews, 2010. 62(1): p. 83-99. 13. Shuai, X., et al., Core-cross-linked polymeric micelles as paclitaxel carriers. Bioconjugate Chemistry, 2004. 15(3): p. 441-448. 14. Li, J., X. Ni, and K.W. Leong, Injectable drug-delivery systems based on supramolecular hydrogels formed by poly(ethylene oxide)s and α- cyclodextrin. Journal of Biomedical Materials Research - Part A, 2003. 65(2): p. 196-202. 15. Lutz, J.F., 1,3-Dipolar cycloadditions of azides and alkynes: A universal ligation tool in polymer and materials science. Angewandte Chemie - International Edition, 2007. 46(7): p. 1018-1025. 16. Bock, V.D., H. Hiemstra, and J.H. van Maarseveen, CuI-Catalyzed Alkyne–Azide “Click” Cycloadditions from a Mechanistic and Synthetic Perspective. European Journal of Organic Chemistry, 2006. 2006(1): p. 51- 68. 194

Drug release profile of click products in comparison with impregnations carried out in

scCO2

17. Tron, G.C., et al., Click chemistry reactions in medicinal chemistry: Applications of the 1,3-dipolar cycloaddition between azides and alkynes. Medicinal Research Reviews, 2008. 28(2): p. 278-308. 18. Lu, J., M. Shi, and M.S. Shoichet, Click chemistry functionalized polymeric nanoparticles target corneal epithelial cells through RGD-cell surface receptors. Bioconjugate Chemistry, 2009. 20(1): p. 87-94. 19. Elchinger, P.H., et al., Polysaccharides: The "click" chemistry impact. Polymers, 2011. 3(4): p. 1607-1651. 20. Deshayes, S., et al., "Click" conjugation of peptide on the surface of polymeric nanoparticles for targeting tumor angiogenesis. Pharmaceutical Research, 2011. 28(7): p. 1631-1642. 21. Gracia, E., et al., Improvement of PLGA loading and release of curcumin by supercritical technology. Journal of Supercritical Fluids, 2018. 22. Mazarro, R., et al., Copolymerization of D,L-lactide and glycolide in supercritical carbon dioxide with zinc octoate as catalyst. Journal of Biomedical Materials Research - Part B Applied Biomaterials, 2008. 85(1): p. 196-203. 23. Mulia, K., et al., Drug release from PLGA microspheres attached to solids using supercritical CO2. Journal of Biomaterials Applications, 2011. 25(5): p. 401-412. 24. Fages, J., et al., Particle generation for pharmaceutical applications using supercritical fluid technology. Powder Technology, 2004. 141(3): p. 219-226. 25. Venugopala, K.N., V. Rashmi, and B. Odhav, Review on natural coumarin lead compounds for their pharmacological activity. BioMed Res Int, 2013.

195

Chapter 5

26. Velencoso, M.M., et al., Click-ligation of coumarin to polyether polyols for polyurethane foams. Polymer International, 2013. 62(5): p. 783- 790. 27. Gracia, E., et al., Functionalization and optimization of PLA with coumarin via click chemistry in supercritical CO2. Journal of CO2 Utilization, 2017. 20: p. 20-26. 28. Gracia, E., et al., Copper wire as a clean and efficient catalyst for click chemistry in supercritical CO2. Catalysis Today, 2018. 29. Hutmacher, D.W., et al., Mechanical properties and cell cultural response of polycaprolactone scaffolds designed and fabricated via fused deposition modeling. Journal of Biomedical Materials Research, 2001. 55(2): p. 203-216. 30. O’Brien, F.J., et al., The effect of pore size on cell adhesion in collagen-GAG scaffolds. Biomaterials, 2005. 26(4): p. 433-441. 31. Howdle, S.M., et al., Supercritical fluid mixing: preparation of thermally sensitive polymer composites containing bioactive materials. Chemical Communications, 2001(1): p. 109-110.

196

Conclusions and recommendations

The present work has been aimed at the study of the functionalization of biodegradable polymers via click chemistry for pharmaceutical applications. This chapter lists the main conclusions derived from the research performed in this Doctoral Thesis. In addition, some recommendations are suggested to be taken into account in further studies.

Conclusions

The results obtained from the present research work support the following main conclusions:

 Supercritical carbon dioxide (scCO2) used as solvent makes possible to obtain high pure products because of its lack of residual solvent, mass transfer or sterilization capacity,

something which allows us to choose scCO2 as a solvent for impregnation and functionalization of biodegradable polymers destined to pharmaceutical applications.

 The use of scCO2 in impregnation of biodegradable polymers allows us to obtain a better distribution of organic compound in the polymer structure due to the plasticization effect. This effect means a higher drug loading and more controlled release of the organic compound.

 The use of scCO2 in functionalization of biodegradable polymers via click chemistry has been checked as an excellent alternative for pharmaceutical applications due to the possibility of removing organic solvents previously needed in functionalization at atmospheric pressure. In addition, the use of supercritical technology makes possible to carry out click chemistry without

199

Conclusions and recommendations

the need of a ligand, obtaining yields over 93% with a catalyst in homogeneous phase.  The use of a catalyst in heterogeneous phase has been

successfully performed in click chemistry using scCO2. The use in this case of copper wire as catalyst allowed us to obtain a click product free of catalyst, something essential for pharmaceutical applications. In addition, the possibility of reusing copper wire catalyst in successive reactions was determined as the most economical option in comparison with the technologies previously studied.  Different coumarin structures were studied in order to determine different positions where triazol ring could be linked to the organic compound structure. 3-azidomethyl coumarin was synthetized obtaining yields over 91% in the functionalization using catalyst in powder and wire forms.  The study of click chemistry release using both coumarin structures showed important differences with respect to the coumarin impregnated. Coumarin functionalized needed 23% more of time to be totally released in the PBS solution. This fact shows the contribution of covalent bond between organic compound and biodegradable polymer in the release profile of drug for pharmaceutical applications.

200

Conclusions and recommendations

Recommendations

The following proposal can be stated in order to complete and extend this research work:

 To study different organic compounds with pharmacological properties and biodegradable polymers in order to check the good performance of copper wire as catalyst in supercritical carbon dioxide in other click chemistry reactions.  To perform different in vivo experiments in order to get to know the effectiveness of click chemistry release in animals once it has been checked in in vitro experiments.  To study an economical evaluation in high scale when more results in higher laboratory scale are obtained. This information will make possible to determine costs, pharmacokinetic results for the estimation of dosage and a possible final price for the product.

201

List of publications and conferences

Publications

1. Functionalization and optimization of PLA with coumarin via click

chemistry in supercritical CO2. E. Gracia, M.T. García, A.M.

Borreguero, A. De Lucas, I. Gracia, J.F. Rodríguez. Journal of CO2 Utilization. 2017, 20: p. 20-26. 2. Improvement of PLGA loading and release of curcumin by supercritical technology. E. Gracia, M.T. García, J.F. Rodríguez, A. De Lucas, I. Gracia. Journal of Supercritical Fluids. 2018. 141: p. 60-67. 3. Copper wire as a clean and efficient catalyst for click chemistry in

supercritical CO2. E. Gracia, M.T. García, A. De Lucas, J.F. Rodríguez, I. Gracia. Catalysis Today. In press. 4. Feasibility of copper wire as heterogeneous catalyst in click chemistry at supercritical conditions. E. Gracia, M.T. García, A. De Lucas, J.F. Rodríguez, I. Gracia. Submitted to Journal of Macromolecular Science, Part A. 5. Impregnation of Curcumin into Biodegradable Support (poly-lactic- coglycolic acid, PLGA), to Transfers its Well Known in vitro Effect to an in vivo Prostate Cancel Model. E. Gracia, A. Mancini, A. Colapietro, C. Mateo, I. Gracia, C. Festuccia, M. Carmona. Submitted to Nutrients.

205

List of publications and conferences

Conferences

Oral presentations in congresses:

1. Synthesis of biodegradable PLGA highly impregnated with curcumin using supercritical technology. E. Gracia, I. Gracia, J.F. Rodríguez, M.T. García, A. De Lucas. 5th International Congress on Green Process Engineering. Quebec (Canadá), Junio 2016.

2. Use of scCO2 as solvent for PLA functionalization via click chemistry. E. Gracia, I. Gracia, J.F. Rodríguez, M.T. García, A. M. Borreguero, A. De Lucas. 10th World Congress of Chemical Engineering, Barcelona (España), Abril, 2017. 3. Coumarin modelling for PLA functionalization via click chemistry

in supercritical CO2. E. Gracia, I. Gracia, M.T. García, J.F. Rodríguez, A. De Lucas. 6th International Congress on Green Process Engineering. Toulouse (France), Junio, 2018. 4. Greener click chemistry for the functionalization of biopolymers

using supercritical CO2 as solvent. J.F. Rodríguez, E. Gracia, A.M. Borreguero, M.T. García, A. De Lucas, M.P. Garrido. World Congress on Medical Phsysics & Biomedical Engineering. Praga (República Checa), Junio, 2018. 5. Funcionalización de PLA mediante química click en medio supercrítico. E. Gracia, I. Gracia, M.T. García, J.F. Rodríguez, A. De Lucas. IX Reunión de Expertos en Tecnologías de Fluidos Comprimidos (Flucomp). Madrid (España), Junio, 2018. 6. Click Functionalization of Polymers destined to Pharmaceutical

Applications in scCO2. E. Gracia, M.T. García, A. De Lucas, J.F. Rodríguez, I. Gracia. 17th European Meeting on Supercritical Fluids (EMSF). Ciudad Real (España), Abril, 2019. 206

List of publications and conferences

Poster presentations in congresses:

1. Impregnación de PLGA con curcumina mediante CO2 supercrítico. E. Gracia, C. Gutiérrez, M.T. García, A. De Lucas, J.F. Rodríguez, I. Gracia. VIII Reunión de Expertos en Fluidos Comprimidos (Flucomp). Cádiz (España), Septiembre, 2015. 2. Influence of polymer degradation on the Indomethacin from PLGA and plga foams. E. Gracia, C. Gutiérrez, M.T. García, A. De Lucas, J.F. Rodríguez, I. Gracia. 6th International Conference on Biobased and Biodegradable Polymers. San Sebastián (España), Octubre, 2015. 3. PLA functionalization via click chemistry at supercritical conditions. E. Gracia, I. Gracia, M.T. García, A.M. Borreguero, A. De Lucas, J.F. Rodríguez. 16th European Meeting on Supercritical Fluids (EMSF). Lisboa (Portugal), Abril, 2017.

207

Impregnation of Curcumin into Biodegradable Support

(poly-lactic-coglycolic acid, PLGA), to Transfer its Well

Known in vitro Effect to an in vivo Prostate Cancer Model I Annex I

1. Introducción 2. Experimental 2.1. Materials 2.2. Impregnation of PLGA Using Supercritical Carbon Dioxide (scCO2) 2.3.1. Fourier-transform Infrared Spectroscopy (FTIR) 2.3.2. Ultraviolet-Visible Spectrophometry (UV-Vis) 2.3.3. Differential Scanning Calorimetry (DSC) 2.4. Xenograft Model 2.5.Treatments for in vivo Experiments 2.6. Evalutaion of Treatment Response in vivo 2.7. Immunohistochemical Analyses 2.8. Statitical Analysis 3. Results 4. Discussion 5. Conclusions 6. Bibliography

A continuación se muestran de forma esquematizada los contenidos correspondientes al anexo 1.

scCO2

scCO2 Compuestos Catalizador en Catalizador en orgánicos fase homogénea fase heterogénea modificados

Química click

Liberación in vitro

Annex I

Impregnation of Curcumin into Biodegradable Support (poly-lactic-coglycolic acid, PLGA), to Transfer its Well Known in vitro Effect to an in vivo Prostate Cancer Model Eulalio Gracia 1, Andrea Mancini 2, Alessandro Colapietro 2, Cristina Mateo 3, Ignacio Gracia 1, Claudio Festuccia 2, Manuel Carmona 3,*

1 Institute of Chemical and Environmental Technology (ITQUIMA), Department of Chemical Engineering, University of Castilla-La Mancha, Ciudad Real 13071, Spain; [email protected] (E.G.); [email protected] (I.G.) 2 Laboratory of Radiobiology, Department of Biotechnological and Applied Clinical Sciences, University of L'Aquila, L'Aquila, Italy; [email protected] (A.M.); [email protected] (A.C.); [email protected] (C.F.) 3 Food Technology Lab, School of Architecture, Engineering and Design, Universidad Europea de Madrid, Villaviciosa de Odón, Madrid 28670, Spain; [email protected] (C.M.) * Correspondence: [email protected]; Tel.: +34-912115155

Abstract: Prostate cancer (PCa) is one of the most common cancers in older men with high mortality rates. Despite advances in screening for and early detection of PCa, a large percentage of men continue to be diagnosed with metastatic disease including about 20 % of men affected with a high tumor grade and stage. Medicinal plants show a great potential to prevent/treat PCa, as well as to reduce its incidence/prevalence and to improve survival rate of PCa. One of the most promising compounds is curcumin, a major, nontoxic, bioactive compound extracted from Curcuma longa plant. Curcumin is strongly active as antitumor agent in vitro, however suffers from a low intestinal absorption and poor in vivo bioavailability. Many strategies to circumvent this problem have been explored. In this report, curcumin was impregnated into a biodegradable poly(lactic-coglycolic) acid (PLGA) support and tested in subcutaneous PCa xenografts (PC3, 22rv1 and DU145) comparing its effects with those observed with a commercial curcumin preparation. Our results indicate that PLGA-impregnated curcumin was significantly more active respect to oral curcumin preparation supporting to consider this preparation as useful for subcutaneous administration.

Keywords: curcumin; PGLA polymer; supercritical carbon dioxide impregnation; curcumin release; prostate cancer, mice xenograft model.

1. Introduction Prostate cancer (PCa) represents the second major cause of cancer and the fifth leading cause of cancer-associated deaths in older men worldwide [1, 2]. A close connection between oxidative stress, inflammation and risk of progressive PCa exists 211

Annex I

[3, 4]. Oxidative stress, characterized by an imbalance between the production of reactive oxygen species (ROS) and the capability of biological system to counterbalance the effects of reactive free radicals or restore oxidative damage, play a critical role in PCa progression and therapy responses [5, 6]. Optimal levels of intracellular ROS are critical in maintaining cellular signaling and homeostatic redox balance. High levels of ROS, indeed, can cause significant reduction in the antioxidant defense mechanisms resulting in DNA, protein and lipid damage [5-7]. Oxidative stress contributes to the initiation and progression of PCa by regulating transcription factors, cell cycle regulators and DNA [5, 8-9]. Antioxidant therapy may prevent PCa by combating oxidative stress [10-13], using for it the phytochemicals present in the plants that present in general a low toxicity [14]. Of the different families of compounds naturally present in plants, it is the polyphenols that have shown greater activity against PCa. Some examples are silibinin, luteolin, ellagic acid, epigallocatechin gallate and other catechins, or trans-resveratrol [15], compounds widely spread in food sources like grapes, pomegranate or green tea. But above all, the most promising compound for control and reduce PCa development is curcumin. Curcumin has been used as a food additive and in traditional medicine in many Asian countries. It has a wide range of biological activities, such as anti‐inflammatory, anti‐oxidative, anti‐metastatic, as well as multi‐drug resistance reversing properties [16-20]. Curcumin is a major yellow pigment component of turmeric (Curcuma longa), a spice often found in curry powder. It works by blocking cell signaling and inhibiting cell division through specific types of enzymes and growth factors that are directly involved in cancer development [21]. Several studies have shown anti‐cancer potential of curcumin in preclinical studies through its effects on androgen receptor (AR) signaling and numerous downstream targets (eg, VEGF, PTEN, NF‐kB) [22-25]. Curcumin was shown to down‐regulate AR expression [22], limit AR binding to the androgen response element of the prostate‐specific antigen (PSA) gene, and reduce the expression of PSA in LNCaP cells [26]. The United States Food and Drug Administration has approved curcumin as being generally recognized as safe, and it is now being used as a supplement in several countries [27]. Phase I clinical studies demonstrated that oral curcumin is not toxic to humans at doses up to 8000 mg/day for 3 months [28]. However, curcumin has not yet been approved as a therapeutic agent because of its limitations due to its poor bioavailability (such as poor absorption, limited tissue distribution and rapid metabolism) [29, 30]. When ingested, this substance travels to the light of the intestine where it is absorbed by passive diffusion through the plasma membrane of the endothelial cells of the intestine. Undergoes numerous biotransformations by the action of bacteria and enzymes present in it and is recognized as xenobiotic by certain transmembrane proteins belonging to the family of ABC transporters (P-gp, MRP and BCRP), which are responsible for returning the curcumin from the interior of the enterocyte to the

212

Annex I intestinal light where it is eliminated by fecal excretion [31]. Therefore, curcumin has low absorption and thus poor bioavailability. Generally speaking, the three most common strategies for overcoming curcumin poor bioavailability are: 1) the concomitant consumption of curcumin together with other natural compounds such as piperine [32] that inhibit the activity of ABC transporters, preventing the return of curcumin to intestinal light [33]; 2) chemical modification of the curcumin molecule to obtain analogues in the form of hydrazinocurcumin [34], diphenyl difluoroketone [35], or PAC (3,5-bis(4-hydroxy-3- methoxybenzylidene)-N-methyl-4-piperidone) [36], more soluble and bioavailable compounds; or 3) encapsulation resulting in different types of pharmaceutical forms: nanoparticles or solid molecules of small dimensions between 1 and 600 nm [37], micelles formed by a monolayer of phospholipids [38], liposomes with several layers of amphipathic lipids [39], or niosomes formed by surfactants or non-ionic surfactants that form vesicles without the presence of phospholipids and cholesterol [40]. Of all these possibilities, the most interesting is encapsulation with biocompatible polymers. The poly(lactic-coglycolic) acid (PLGA) is one of the matrices most used today to encapsulate all types of compounds with biological activity, also antitumor such as doxorubicin [41] or paclitaxel [42]. The interest in this biodegradable copolymer is that it is approved by the US Food and Drug Administration (FDA) for contact with biological fluids, with potential applications in wound closure and surgical sutures, delivery carriers, tissue engineering scaffolds or various types of implants [43-44]. In case of curcumin, PLGA has been used to produce particles and nanoparticles by very diverse techniques such as nanoprecipitation [45], coaxial electrospray process [46], liquid-driven co-flow focusing [47], or by means of techniques that allow the particles obtained to have superparamagnetism [48]. In the present work, supercritical carbon dioxide approach (scCO2) is used to obtain a PLGA polymer impregnated with curcumin from previous tests that have demonstrated the absence of residues and a high impregnation efficiency [49]. Its in vitro release has been characterized and, from the results obtained, a periodic administration has been designed every 8-9 days to three PCa xenograft models (PC3, DU145 and 22rv1 cells). At the same time, an improved commercial preparation (MicroActive© curcumin) which is designed to overcome the bioavailability difficulties, has been tested.

213

Annex I

2. Materials and Methods

2.1. Materials For the production of the poly(lactic-coglycolic) acid (PLGA) support, glycolide (G) and D,L-lactide (L) with a purity higher than 99.5 % were purchased from Purac Biochem bv (The Netherlands), and curcumin used for impregnation had purity higher than 98 % (Wellgreen Technology Co. Ltd, Xi'an, China). MicroActive® curcumin 25 % was supplied by Comercial Química Massó S.A. Other reagents used were: carbon dioxide (Carburos Metálicos, S.A., Spain) with a purity of 99.5 %, stannous octoate (Sigma-Aldrich Química, S.A., Spain) and acetone of analytical quality (VWR, S.A., Spain).

2.2. Impregnation of PLGA Using Supercritical Carbon Dioxide (scCO2) Polymerization of PGLA in a molar rate L:G 80:20 and the supercritical impregnation of polymer with the curcumin was carried out according to [49], but slightly modified as: 1000 mg of solid PLGA was mixed with a solution of acetone containing 170 mg of curcumin and placed in a high-pressure vessel where CO2 was charged until 150 bar and 45 ºC were reached. The contact time for this impregnation was 8 hours to ensure the total solubilisation.

2.3. Impregnated Polymer Characterizations

2.3.1. Fourier-transform Infrared Spectroscopy (FTIR) IR spectra of the impregnated polymer was obtained with a spectrophotometer Varian model 640-IR (Varian Inc., Palo Alto, California, USA) in range from 4000 to 400 cm-1, with a resolution of 4.0 cm-1 and 64 scanning, using the software Varian Resolution (Varian Inc., Palo Alto, California, USA).

2.3.2. Ultraviolet–Visible Spectrophotometry (UV-Vis) UV-Vis spectra was used to determine the yield obtained in the PLGA impregnation step, as well as to establish the in vitro release of curcumin. In the first case the high solubility of curcumin in acetone was used to calculate it. An 8-point calibration curve dissolving pure curcumin (0.025-0.250 mg/ml) in acetone was carried out (y= 7.664x-0.0857; r2=0.999). For the in vitro release experiment, 30 mg of polymer impregnated with curcumin was suspended in a PBS solution 0.1 M (pH 7.4,), in a 100 ml flask hermetically closed and preserved from light, stirred at 100 rpm, and incubated in a shaking water bath at 37 ºC. 5 ml solution was periodically taken from the flask in order to measure by UV-Vis spectrophotometry the quantity of curcumin released. The measurements were carried out using a dual beam UV

214

Annex I spectrophotometer Shimadzu UV-1603 (Kyoto, Japan) at an absorption maximum of 421 nm using the UVPC Personal Spectroscopy Software, version 3.6 (Shimadzu, Kyoto, Japan).

2.3.3. Differential Scanning Calorimetry (DSC) The calorimetric analysis was determined by a DSC Q100 equipment (TA Instruments, New Castle, USA). Samples of 3-10 mg were prepared in aluminium capsules. The sample was heated until 280 °C with a ramp of 10 °C/min, followed by a cooling until -50 °C at the same rate and, finally, and heated again until 280 °C with the same ramp.

2.4. Xenograft Model Male CD1 nude mice (Charles River, Milan, Italy) were maintained under the guidelines established by our Institution (University of L’Aquila, Medical School and Science and Technology School Board Regulations, complying with the Italian government regulation n.116 January 27 1992 for the use of laboratory animals). All mice received S.C. flank injections of 1 x 106 PC3, DU145 or 22v1 cells. Tumour growth was assessed by bi-weekly measurement of tumor diameters with a Vernier calliper (length x width). Tumour weight was calculated according to the formula: TW (mg) = tumor volume (mm3) = d2 x D/2, where d and D are the shortest and longest diameters, respectively. The effects of the treatments were examined as previously described [50].

2.5. Treatments for in vivo Experiments. Before starting of treatment, animals were randomized into three groups of treatment as follows: Group 1: mice (n=10) receiving oral administration of 100µl PBS (Control); Group 2: mice (n=10) receiving oral administration of 100 mg/kg of MicroActive® commercial preparation containing 25 % of curcumin (Maypro Industries LLC; Westchester, NY, USA); Group 3: mice (n=10) receiving subcutaneous administration of PLGA-impregnated curcumin of 800 mg/kg at 1, 9, 18 and 27 day. A scheme of the protocol followed is shown in Figure 1.

215

Annex I

Figure 1. Protocol for drugs supplying to the mice xenograft model in both presentations: MicroActive curcumin by oral administration (blue colour), and impregnated curcumin by subcutaneous administration (grey colour).

At the end of experiments (35 days after the start of treatments) animals were sacrificed by carbon dioxide inhalation and tumors were subsequently removed surgically for immunohistochemical analyses.

2.6. Evaluation of Treatment Response in vivo The following parameters were used to quantify the antitumor effects upon different treatments as previously described [51]: (1) tumor volume measured during and at the end of experiments; (2) tumor weight measured at the end of experiment; (3) tumor progression (TP or doubling time) defined as an increase of greater than 100 % of tumor volume with respect to baseline; (4) time to progression (TTP) defined as the time for tumor progression. All together, these parameters provide the data on the percentage of tumors in progression used to generate Kaplan Meyer curves.

2.7. Immunohistochemical Analyses Indirect immunoperoxidase staining was performed on 4 μm paraffin-embedded tissue sections. Ki67 labeling index was determined by counting 500 cells at 100X and determining the percentage of cells staining positively for Ki67. Apoptosis was measured as the percentage of tunnel positive cells +/- standard deviation (SD) measured on five random fields (400X) by using TACS Blue Label kit (R&D Systems, Inc., Minneapolis, MN, USA). A consensus judgment was adopted according to [51], for immunohistochemical scoring of tumors based on the strength of positivity: 216

Annex I negative (score 0), weak (score 1), moderated (score 2), or strong staining (score 3). In each category, the percentage of positive cells was assessed by scoring at least 1000 cells in the area with the highest density of antigen positive cells. Percentage of cells was graded as follow: 0 = absence of cells; 1 = < 10% positive cells; 2 = positive cells in a range of 10 - 50%; and 3 = > 50% positive cells. Overall expression was defined by the Staining Index (SI) and was ranged between 0 and 9, with an SI ≤ 4 indicating a low expression. Martius yellow-brilliant crystal scarlet blue stain and trichromic staining were used to stain erythrocytes, and consequently, the presence of micro-thrombi and bleeding zones and fibrous stroma.

2.8. Statistical Analysis Continuous variables were summarized as mean and S.D. or 95 % CI for the mean. Statistical comparisons between controls and treated groups were established by carrying out the ANOVA test or by Student's t test for unpaired data (for two comparisons). Dichotomous variables were summarised by absolute and/or relative frequencies. For dichotomous variables, statistical comparisons between control and treated groups were established by carrying out the exact Fisher's test. For multiple comparisons, the level of significance was corrected by multiplying the P value by the number of comparisons performed (n) according to Bonferroni correction. Overall survival was analysed by Kaplan–Meier curves and Gehan's generalised Wilcoxon test. When more than two survival curves were compared, the Logrank test for trend was used. This tests the probability that there is a trend in survival scores across the groups. All tests were two-sided and were determined by Monte Carlo significance. P values at least <0.05 were considered statistically significant. In the figures in which statistical analysis was performed, significance was indicated by an asterisk. SPSS (statistical analysis software package, IBM Corp., Armonk, NY, USA) version 10.0 and StatDirect (version. 2.3.3., StatDirect Ltd, Altrincham, Manchester, UK) were used for statistical analysis and graphic presentation.

3. Results

3.1. Characterization of PLGA Impregnated with Curcumin in scCO2

The impregnated PLGA sample with curcumin in scCO2 made following a bibliographic route (1) was characterized using three different techniques. Figure 2a shows the IR spectra where different peaks corresponding to PLGA-curcumin structure are shown. The appearance of the most characteristic absorbance band of PLGA corresponding to the carbonyl group (1760 cm-1), increased after impregnation due to the contribution of the C=O group of curcumin, demonstrating that the process had taken place properly.

217

Annex I

0.25

0.20

0.15

0.10 T : 51.02ºC

0.05 C=O (W/g) Flow Heat 0.00

-0.05 -100 -50 0 50 100 150 200 250 300 Temperature (ºC)

(a (b) ) Figure 2. Characterization of PLGA impregnated with curcumin in scCO2 by (a) FTIR and (b) DSC techniques.

Curcumin-loaded PLGA sample was also characterized by DSC in order to determine its glass transition temperature (Tg), a very useful measure to establish the purity of the resulting product and the level of residual contamination with the solvent used during the process. As it is shown in Figure 2b, the glass transition temperature exhibited by PLGA impregnated with curcumin was 51.02 ºC. The calibration curve carried out on acetone made it possible to determine by means of UV-Vis spectrophotometry that of the 170 mg of curcumin used in impregnation, 159.4 mg were actually impregnated, resulting in an impregnation yield of 93.8 %.

3.2. Curcumin in vitro Release The amount of released curcumin was determined by UV-Vis spectra at 421 nm. A calibration curve in PBS was carried out in order to determine the release sample profile (Figure 3).

218

Annex I

Figure 3. Curcumin release from impregnated PLGA with curcumin in scCO2 when suspended in PBS solution at 37 ºC during 10 days.

According to the results shown in Figure 3, 8 days are necessary for the release of more than 90 % of drug impregnated. It is after 10 days when almost total curcumin is released to the PBS solution. From the previously determined content (159.4 mg of curcumin per 1000 mg of impregnated PGLA) and these results from the in vitro release, the experiment to be carried out with the animal model was designed. During the 5 weeks of the experiment, Group 2 (MicroActive curcumin) was treated 5 days per week at the established concentration (100 mg/kg) as shown in Figure 1, until the day before slaughter (day 34). This involved treating each mouse for 26 days with an equivalent amount of curcumin of 650 mg/kg (2600 mg/kg of MicroActive preparation containing 25 % curcumin). Group 3 (impregnated curcumin) was treated 5 times within the 5-week period, with a cadence of 8-9 days, the frequency established by the release trials. It meant treating the mice with a slightly lower amount of curcumin equivalent (636 mg/kg), as the 800 mg/kg dosed on each occasion contained a lower proportion of curcumin (4000 mg/kg of PGLA impregnated with 15.9 % of curcumin).

3.3. Antitumor Effect of Curcumin Preparations in PC3, DU145 and 22rv1 Subcutaneous Xenografts

The two different preparations of curcumin: 1) MicroActive as oral preparation, and 2) PLGA-impregnated curcumin as subcutaneous administration, showed antitumor effects when administered in nude mice receiving PC3, DU145 and 22rv1 subcutaneous xenografts. Both produced noticeable reduction on tumor size, although PLGA-impregnated curcumin possessed higher antitumor effectiveness when 219

Annex I compared with the commercial MicroActive preparation of curcumin. In Figure 4 are shown the changes in tumor weight by treatments with reduction of 29 % in tumors treated with commercial curcumin and 71 % in tumors treated with PLGA- impregnated curcumin, respectively. These differences were statistically significant towards control and between them (Figure 4).

Figure 4. Aspect and weight of PC3 tumors (shown as an example for the analyses carried out for the three xenografts), evaluated at the end of treatment and their statistical comparison by groups. * Statistically significant differences with the Control group (P<0.05); # Statistically significant differences with MicroActive group (P<0.05).

In order to reduce the probability of biases due to differences in tumor engraftment we analyzed the tumor progression through the parameter "Time to Progression (TTP)", defined as the time (days) necessary to double the volume for each tumor, comparing the percentage of tumor in progression in the time by Kaplan Meyer distribution. In Figure 5 analysis on the Kaplan Meyer curves demonstrates that PLGA-impregnated curcumin was able to reduce significantly the progression of the PC3 tumors with a Hazard ratio (HR) of 4.51, while MicroActive oral preparation did it with a HR value of 2.65. This suggests that the PLGA-impregnated curcumin preparation was 2.47 times (Hazard ratio for the comparison between the two curcumin preparations) more active than oral preparation which also showed an interesting profile of activity with respect to the control. Same experiment was repeated with DU145 (Figure 4E) and 22rv1 (Figure 4) xenografts with similar results. 220

Annex I

Figure 5. Percentage of mice Tumor Progression (TP) plotted for the time by using Kaplan–Meyer analysis for (a) PC3, (b) DU145 and (c) 22rv1 xenografts. At the right side, statistical analyses of Hazard ratios of paired Kaplan Meyer curves.

3.4. Immunohistochemical Analyses

The immunohistochemical preparations confirm the stronger effectiveness of PLGA-impregnated curcumin (Figure 6). 221

Annex I

Figure 6. Immunohistochemical analyses using three different staining techniques for tumors growth after PC3 cells subcutaneously injection in male nu/nu mice (xenograft model): (a) Trichromic staining, (b) Ki67 staining, and (c) Tunel fluorescent staining; and statistical analysis for the three models of PCa xenografts. * Statistically significant differences with the Control group (P<0.05); # Statistically significant differences with MicroActive group (P<0.05). Magnifications are indicated in the single panels as size bar.

222

Annex I

Trichromic staining performed on tumor tissues derived from untreated and treated animals revealed an increased deposition of collagens and following fibrotic reaction (ranged between 20 and 50 % of slide area) after curcumin administration (both treatments). The effects were higher when PLGA-impregnated curcumin was administered and compared to oral preparations of curcumin. In Figure 6a is shown the appearance of trichromic staining performed in PC3 xenografts. Reduced monocyte infiltration was also noticed suggesting a reduced inflammation. Reduced proliferation was associated with reduced Ki67 staining (proliferating cells) (Figure 6b). Also in this case, the percentage of cells positive for Ki67 was significantly reduced by curcumin administration and to a greater extent when we administered PLGA-impregnated curcumin. Parallelly, the percentage of tunel positive cells (apoptotic dead cells) was greater in PLGA-impregnated curcumin as shown in Figure 6c for PC3 xenografts.

4. Discussion

4.1. Characterization of PLGA Impregnated with Curcumin in scCO2 Supercritical technology, specifically supercritical carbon dioxide (scCO2) as solvent, is presented as a suitable media for pharmaceutical applications due to its excellent properties [52]. This solvent also acts on the polymer matrix allowing a phenomenon called plasticization which consists on the decrease of Tg of a polymer, making possible a better interaction between the pharmaceutical compound and the polymer [53]. Technology that when applied to the polymer used in this work, the copolymer obtained from the copolymerization reaction between lactide and glycolide (PLGA), allows to produce scaffolds with a high porosity [54] improving its drug loading capacity by providing more free space [49]. This biodegradable copolymer approved by the US Food and Drug Administration (FDA) for contact with biological fluids presents a wide variety of properties which make it an excellent candidate to be used in biomedical field as manufacture tissue engineering scaffolds, delivery system material or different bioabsorbable medical implants [43, 44]. The glass transition temperature value reached by the impregnated polymer of 51.02 ºC was a high value, very similar to the observed for the pure polymer (51.90 ºC) [49]. This fact means that acetone was totally remove from the polymeric matrix, and being an adequate candidate to be used for prevention and treatment of cancer in market formulations without any additional purification step [55]. The impregnation yield reached as well, a high value (93.8 %), even better than the data of previous work with the same technique [49]. It is among the highest values ever achieved when PLGA is used to encapsulate curcumin, well above the 70 % achieved using liquid-driven co-flow focusing (LDCF) technique [47], in the range of the 89-94 % obtained by single emulsion-solvent evaporation method

223

Annex I

[56], and close to the highest efficiency achieved so far, 97.5 % reached by Anand 2010 [57], although these authors obtained a very low % of drug loading, only 0.4%. This is another key factor to take into account during PLGA impregnation process, as the higher the percentage the lower the quantity dose of the final product to administer the desired amount of drug. In this sense, the results obtained in this work (15,9 % drug loading) are well above the usual 5-10 % [45,47], and only below the 30 % reached by electrospray coaxial technique [46].

4.2. Curcumin in vitro Release The mechanism for controlled release of drug from biocompatible polymers consists of 3 steps [7]: 1) solubilization of the drug located in the surface of the particles, being the most accessible (burst stage), 2) diffusion of the drug through the narrow or tortuous pores (internal diffusion), 3) solubilization of the entrapped drug when polymer network is hydrolyzed (polymer degradation). Taking as a reference this drug delivery scheme, and in view of the results shown in Figure 3, the burst stage in our case was much less intense than other PLGA-curcumin applications [45, 47, 56], probably because of the plasticizing effect already mentioned. There would be a better integration of curcumin in the polymer. The first stage during the first hours hardly separates from the linear evolution that would suppose the long-term release, whereas in other systems it is much more accentuated. For example, while in our study there is a release of around 20 % in the first 24 hours, the same percentage is reached by other authors within the first hour [56], reaching 40 % in the first 4 hours [45] or in the first day [47]. It seems that the effect of scCO2 technique in the impregnation of curcumin allows better homogeneous distribution of curcumin within the polymer matrix, leaving accessible on the outside a small quantity. This is a behaviour more characteristic of the release of a hydrophobic substance (such as curcumin) impregnated in PLGA, whereas the release of curcumin impregnated in PLGA by other techniques [45, 47, 56], is more similar to the behaviour of a hydrophilic drug [7]. In summary, the high efficiency of impregnation, the high drug loading, and the better progressive release, without a burst stage so accentuated than in other techniques of encapsulation / impregnation of curcumin in PLGA, make it an excellent choice in this application. Also, it would allow it to be considered for the treatment of other pathologies different from PCa, such as for example, in locoregional applications in unresectable cancers where intraperitoneal therapy may be a suitable delivery route. An approach that has shown better prognosis and longer survival rate for small tumors in ovarian cancer patients [58]. The stable and continuous release would avoid the toxicity problems normally associated with transferring approved intravenous protocols to intraperitoneal bolus of the entire dose all-at-once [59]. Using in this case,

224

Annex I a biodegradable polymer that is well tolerated in locoregional applications in humans, which is considered the polymer of choice for intraperitoneal treatments [60], and a drug that has been shown a very low toxicity in a healthy status [28]. It would also be a suitable approach for solid tumors of pancreas to which it is difficult to reach chemotherapy for their hypovascular and poorly perfused nature [61].

4.3. Antitumor Effect of Curcumin Preparations in PC3, DU145 and 22rv1 Subcutaneous Xenografts Curcumin has demonstrated in the past to have high anticancer effects by using in vitro tumor models and relatively low antitumor effects in vivo. This could be due to its low bioavailability which restricts its application. In our report, we demonstrated that the PLGA polymer impregnation of curcumin represents a good strategy to increase the effectiveness of curcumin. We observed, indeed, that impregnated curcumin shows higher antitumor effects of the MicroActive curcumin preparation resulting in an increased effectiveness of about 2.0 fold when compared with effects shown after MicroActive curcumin administration. We observed also that animals healthy seemed to be not modified as indicated to the relatively body weight loss (<10%, data not shown). This was due probably to higher concentration of curcumin in the tumor tissue which increase the local antitumor effects as indicated by reduced Ki67 and increased tunnel staining. Apoptosis observed in several report using curcumin is associated also in an increased fibrotic reaction in the treated tumor as indicated by trichromic staining as well as by further analyses (as reduced CD31 staining and vessel formation) which have not been insert in this report.

5. Conclusions Curcumin is a natural anti-cancer compound utilized on a wide variety of human cancer cell lines and animal carcinogenesis models. Although chemotherapy is a conventional treatment for cancers in clinical practice, its efficacy is usually restricted due to an insufficient concentration of the drug at the malignant tissue and its undesired side effects. Although curcumin, was reported to be an active antineoplastic agent the major hurdle in chemotherapy lies in its low bioavailability which restricts its application. Several approaches have been developed to circumvent this problem without interfere with the antitumor effectiveness of curcumin. PLGA polymer impregnation of curcumin represents good strategy to increase the effectiveness of curcumin. The procedure for generating this kind of polymer impregnated with curcumin is relatively simple, the conditions are well established and allow reproducibility and high impregnation yields to be achieved. This approach produces a slow curcumin release which may maintain theoretically elevated, up to 225

Annex I almost ten days, the curcumin concentration in the tumor counteracting the curcumin loss by metabolism and scarce intestinal uptake. Herein we report curcumin impregnated of PLGA polymer administered subcutaneously to PCa xenograft model from three different cell lines. The obtained results clearly suggest a powerful antitumor effect, where the bioavailability and controlled release of the PLGA-impregnated curcumin could be responsible for such increase in antitumor responses. The PLGA particles are a potential curcumin delivery system and encourage, in future, for planning in vivo studies towards other cancer types that might benefit from the curcumin effects.

Author Contributions: Conceptualization, M.C., I.G.; methodology, M.C., C.F.; formal analysis, A.M., C.F.; investigation, E.G., A.M., A.C.; writing—original draft preparation, E.G., C.F.; writing—review and editing, M.C., C.M., I.G.; visualization, C.M.; supervision and project administration, M.C.

Acknowledgments: M.C. thanks to the Spanish Ministry of Science, Innovation and Universities for the Ramón y Cajal Fellowships RyC-2014-16307. Authors wants to thank Comercial Química Massó S.A. company for providing a free sample of MicroActive® curcumin.

Conflicts of Interest: The authors declare no conflict of interest.

226

Annex I

References

1. Thurtle, D., Rossi, S.H., Berry, B., Pharoah, P., Gnanapragasam, V.J. Models predicting survival to guide treatment decision-making in newly diagnosed primary non-metastatic prostate cancer: a systematic review. BMJ Open 2019, 9(6), e029149. 2. Miller, K.D., Nogueira, L., Mariotto, A.B., Rowland, J.H., Yabroff, K.R., Alfano, C.M., Jemal, A., Kramer, J.L., Siegel, R.L. Cancer treatment and survivorship statistics. CA Cancer J Clin 2019, 0, 1-23. 3. Lavender, N., Hein, D.W., Brock, G., Kidd, C.R. Evaluation of Oxidative Stress Response Related Genetic Variants, Pro-oxidants, Antioxidants and Prostate Cancer. AIMS Med Sci. 2015, 2(4), 271-294. 4. Hojan, K., Kwiatkowska-Borowczyk, E., Leporowska, E., Milecki, P. Inflammation, cardiometabolic markers, and functional changes in men with prostate cancer. A randomized controlled trial of a 12‐month exercise program. Pol Arch Intern Med. 2017, 127(1), 25-35. 5. Lv, S., Wen, H., Shan, X., Li, J., Wu, Y., Yu, X., Huang, W., Wei, Q. Loss of KMT2D induces prostate cancer ROS-mediated DNA damage by suppressing the enhancer activity and DNA binding of antioxidant transcription factor FOXO3. Epigenetics 2019, 28,1-15. 6. Höll, M., Koziel, R., Schäfer, G., Pircher, H., Pauck, A., Hermann, M., Klocker, H., Jansen- Dürr, P., Sampson, N. ROS signaling by NADPH oxidase 5 modulates the proliferation and survival of prostate carcinoma cells. Mol Carcinog 2016, 55(1), 27-39. 7. Wang, G., Zhang, D., Yang, S., Wang, Y., Tang, Z., Fu, X. Co-administration of genistein with doxorubicin-loaded polypeptide nanoparticles weakens the metastasis of malignant prostate cancer by amplifying oxidative damage. Biomate Sci. 2018, 6(4), 827-835. 8. Dougan, J., Hawsawi O., Burton, L.J., Edwards, G., Jones, K., Zou, J., Nagappan P., Wang, G., Zhang, Q., Danaher, A., Bowen, N., Hinton, C., Odero-Marah, V.A. Proteomics- Metabolomics Combined Approach Identifies Peroxidasin as a Protector against Metabolic and Oxidative Stress in Prostate Cancer. Int J Mol Sci. 2019, 20(12), E3046. 9. Miller, D.R., Ingersoll, M.A., Chatterjee, A., Baker, B., Shrishrimal, S., Kosmacek, E.A., Zhu, Y., Cheng, P.W., Oberley-Deegan, R.E., Lin, M.F. p66Shc protein through a redox mechanism enhances the progression of prostate cancer cells towards castration- resistance. Free Radic Biol Med. 2019, 139,24-34. 10. Elkady, A.I. Targeting prostate cancer cell proliferation, stemness and metastatic potential using Costus speciosus derived phytochemicals. Am J Transl Res. 2019, 11(4), 2550-2569. 11. Cagnon, V., Lima, E., Lamas, C., Baseggio, A., Kido, L., Marostica, M. Prostatic Adenocarcinoma Progression Delay After Brazilian Berry Extract Intake in Transgenic Mice (Tramp) Submitted to a High- Fat Diet (P05-016-19). Curr Dev Nutr. 2019, 3(Suppl 1), pii: nzz030.P05-016-19. 12. Costea, T., Nagy, P., Ganea, C., Szöllősi, J., Mocanu, M.M. Molecular Mechanisms and Bioavailability of Polyphenols in Prostate Cancer. Int J Mol Sci. 2019, 20(5), E1062. 13. Soundararajan, P., Kim, J.S. Anti-Carcinogenic Glucosinolates in Cruciferous Vegetables and Their Antagonistic Effects on Prevention of Cancers. Molecules 2018, 23(11), E2983.

227

Annex I

14. Kallifatidis, G., Hoy, J.J., Lokeshwar, B.L. Bioactive natural products for chemoprevention and treatment of castration-resistant prostate cancer. Semin Cancer Biol. 2016, 40-41, 160- 169. 15. Lin, P.-H., Aronson, W., Freedland, S.J. An update of research evidence on nutrition and prostate cancer. Urologic Oncology: Seminars and Original Investigations 2019, 37 (6), 387-401. 16. Bahrami, A., Majeed, M., Sahebkar, A. Curcumin: a potent agent to reverse epithelial-to- mesenchymal transition. Cell Oncol. 2019, 42(4), 405-421. 17. Yang, J., Wang, C., Zhang, Z., Chen, X., Jia, Y., Wang, B., Kong, T. Curcumin inhibits the survival and metastasis of prostate cancer cells via the Notch-1 signaling pathway. APMIS 2017, 125(2),134-140. 18. Burge, K., Gunasekaran, A., Eckert, J., Chaaban, H. Curcumin and Intestinal Inflammatory Diseases: Molecular Mechanisms of Protection. Int J Mol Sci. 2019, 20(8), E191. 19. Hassan, F.U., Rehman, M.S., Khan, M.S., Ali, M.A., Javed, A., Nawaz, A., Yang, C. Curcumin as an Alternative Epigenetic Modulator: Mechanism of Action and Potential Effects. Front Genet. 2019, 10, 514. 20. Batra, H., Pawar, S., Bahl, D. Curcumin in combination with anti-cancer drugs: A nanomedicine review. Pharmacol Res. 2019, 139 ,91-105. 21. Mokbel, K., Mokbel, K. Chemoprevention of Breast Cancer With Vitamins and Micronutrients: A Concise Review. In Vivo 2019, 33(4), 983-997. 22. Chou, F.J., Chen, Y., Chen, D., Niu Y., Li, G., Keng, P., Yeh, S., Chang, C. Preclinical study using androgen receptor (AR) degradation enhancer to increase radiotherapy efficacy via targeting radiation-increased AR to better suppress prostate cancer progression. EBioMedicine 2019, 40, 504-516. 23. Mohebbati, R., Anaeigoudari, A., Khazdair, M.R. The effects of Curcuma longa and curcumin on reproductive systems. Endocr Regul. 2017, 51(4), 220-228. 24. Qiang,. Z., Meng, L., Yi, C., Yu, L., Chen, W., Sha, W. Curcumin regulates the miR- 21/PTEN/Akt pathway and acts in synergy with PD98059 to induce apoptosis of human gastric cancer MGC-803 cells. J Int Med Res. 2019, 47(3), 1288-1297. 25. Tian. M., Tian, D., Qiao, X., Li, J., Zhang, L. Modulation of Myb-induced NF-kB -STAT3 signaling and resulting cisplatin resistance in ovarian cancer by dietary factors. J Cell Physiol. 2019, in press. 26. Hong, J.H., Lee, G., Choi, HY. Effect of curcumin on the interaction between androgen receptor and Wnt/β-catenin in LNCaP xenografts. Korean J Urol. 2015, 56(9), 656-65. 27. Soleimani, V., Sahebkar, A,, Hosseinzadeh, H. Turmeric (Curcuma longa) and its major constituent (curcumin) as nontoxic and safe substances: Review. Phytother Res. 2018, 2(6), 985-995 28. Santos Filho, E.X.D., Arantes, D.A.C., Oton Leite, A.F., Batista, A.C., Mendonça, E.F., Marreto, R.N., Naves, L.N., Lima, E.M., Valadares, M.C. Randomized clinical trial of a mucoadhesive formulation containing curcuminoids (Zingiberaceae) and Bidens pilosa Linn (Asteraceae) extract (FITOPROT) for prevention and treatment of oral mucositis - phase I study. Chem Biol Interact. 2018, 291, 228-236.

228

Annex I

29. Bilia , A.R., Piazzini, V., Guccione, C., Risaliti, L., Asprea, M., Capecchi, G., Bergonzi, M.C. Improving on Nature: The Role of Nanomedicine in the Development of Clinical Natural Drugs. Planta Med. 2017, 83(5), 366-381. 30. Bar-Sela, G., Epelbaum, R., Schaffer, M. Curcumin as an anti-cancer agent: review of the gap between basic and clinical applications. Curr Med Chem. 2010, 17(3),190-7. 31. Berginc ,K., Trontelj, J., Basnet, N.S., Kristl, A. Physiological barriers to the oral delivery of curcumin. Die Pharmazie. Int. J. Phar Sci. 2012, 67(6), 518-524. 32. Shoba, G., Joy D., Joseph, T., Majeed, M., Rajendran, R., Srinivas, P. Influence of piperine on the pharmacokinetics of curcumin in animals and human volunteers. Planta Med. 1998, 64(04), 353-356. 33. Li, S., Lei Y., Jia, Y., Li, N., Wink, M., Ma, Y. Piperine, a piperidine alkaloid from Piper nigrum re-sensitizes P-gp, MRP1 and BCRP dependent multidrug resistant cancer cells. Phytomedicine 2011, 19(1), 83-87. 34. Subramaniam, D., May, R., Sureban, S.M., Lee, K.B., George, R., Kuppusamy, P., Ramanujam, R.P., Hideg, K, Dieckgraefe, B.K., Houchen, C.W., Anant, S. Diphenyl difluoroketone: a curcumin derivative with potent in vivo anticancer activity. Cancer Res 2008, 68(6),1962-1969. 35. Rathore, R., Jain, J.P., Srivastava, A., Jachak, S., Kumar, N. Simultaneous determination of hydrazinocurcumin and phenol red in samples from rat intestinal permeability studies: HPLC method development and validation. J Pharm Biomed Anal 2008, 46(2), 374-380. 36. Al-Hujaily, E.M., Mohamed, A.G., Al-Sharif, I., Youssef, K.M., Manogaran, P.S., Al-Otaibi B. PAC, a novel curcumin analogue, has anti-breast cancer properties with higher efficiency on ER-negative cells. Breast Cancer Res Treat. 2011, 128(1), 97-107. 37. Yallapu, M.M., Gupta, B.K., Jaggi, M., Chauhan, S.C. Fabrication of curcumin encapsulated PLGA nanoparticles for improved therapeutic effects in metastatic cancer cells. J Colloid Interface Sci. 2010, 351(1), 19-29. 38. Yu, Y., Zhang, X., Qiu, L. The anti-tumor efficacy of curcumin when delivered by size/charge-changing multistage polymeric micelles based on amphiphilic poly (β-amino ester) derivates. Biomaterials 2014, 35(10), 3467-3479. 39. Mahmud, M., Piwon,i A., Filipczak, N., Janicka, M., Gubernator, J. Correction: Long- Circulating Curcumin-Loaded Liposome Formulations with High Incorporation Efficiency, Stability and Anticancer Activity towards Pancreatic Adenocarcinoma Cell Lines In Vitro. PLoS One 2017, 12(3), e0173728-e0173728. 40. Tavano, L., Muzzalupo, R., Picci, N., de Cindio, B. Co-encapsulation of antioxidants into niosomal carriers: gastrointestinal release studies for nutraceutical applications. Colloids Surf. B Biointerfaces 2014, 114, 82-88. 41. Chronopoulou, L., Domenici, F., Giantulli, S., Brasili, F., D'Errico, C., Tsaouli, G., Tortorella, E., Bordi, F., Morrone, S., Palocci, C., Silvestri, I. PLGA based particles as “drug reservoir” for antitumor drug delivery: characterization and cytotoxicity studies. Colloids Surf. B Biointerfaces 2019, 180, 495-502.

229

Annex I

42. Mittal, P., Vardhan, H., Ajmal, G., Bonde, G.V., Kapoor, R., Mittal, A., Mishra, B. Formulation, optimization, hemocompatibility and pharmacokinetic evaluation of PLGA nanoparticles containing paclitaxel. Drug Dev Ind Pharm. 2019, 45 (3), 365-378. 43. Auras, R., Harte, B., Selke, S. An overview of polylactides as packaging materials. Macromol Biosci. 2004, 4(9), 835-864. 44. Jain, R.A., The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices. Biomaterials, 2000, 21(23), 2475-2490. 45. Joseph, A., Wood, T., Chen, C.C., Corry, K., Snyder, J.M., Juul, S.E., Nance, E. Curcumin- loaded polymeric nanoparticles for neuroprotection in neonatal rats with hypoxic- ischemic encephalopathy. Nano Res 2018, 11(10), 5670-5688. 46. Yuan, S., Lei F., Liu, Z., Tong, Q., Si, T., Xu, R.X. Coaxial Electrospray of Curcumin-Loaded Microparticles for Sustained Drug Release. PLoS ONE 2015, 10(7), e0132609. 47. Dwivedi, P., Yuan, S., Han, S., Mangrio, F.A., Zhu, Z., Lei, F., Ming, Z., Cheng, L., Liu, Z., Si, T., Xu, R.X. Core–shell microencapsulation of curcumin in PLGA microparticles: programmed for application in ovarian cancer therapy. Artif Cells Nanomed Biotechnol. 2018, 46 (sup3), S481-S491. 48. Ayyanaar, S., Kesavan, M.P., Sivaraman, G., Maddiboyina, B., Annaraj, J., Rajesh, J., Rajagopal, G. A novel curcumin-loaded PLGA micromagnetic composite system for controlled and pH-responsive drug delivery. Colloids Surf. A Physicochem. Eng. Aspects 2019, 573, 188-195. 49. Gracia, E., Garcia, M.T., Rodríguez J.F. Improvement of PLGA loading and release of curcumin by supercritical technology. J. Sup. Fluids 2018, 141, 60-67. 50. Gravina, G.L., Mancini A., Colapietro, A., Marampon, F., Sferra, R., Pompili, S., Biordi, L.A., Iorio, R., Flati, V., Argueta, C., Landesman, Y., Kauffman, M., Shacham, S., Festuccia, C. Pharmacological treatment with inhibitors of nuclear export enhances the antitumor activity of docetaxel in human prostate cancer. Oncotarget. 2017, 8(67),111225-111245. 51. Gravina GL, Mancini A, Muzi P, Ventura L, Biordi L, Ricevuto E, Pompili S, Mattei C, Di Cesare E, Jannini EA, Festuccia C. CXCR4 pharmacogical inhibition reduces bone and soft tissue metastatic burden by affecting tumor growth and tumorigenic potential in prostate cancer preclinical models. The Prostate. 2015; 75:1227-46. 52. Fages, J., Lochard, H., Letourneau, J.J., Sauceau, M., Rodier. E. Particle generation for pharmaceutical applications using supercritical fluid technology. Powder Tech. 2004, 141(3), 219-226. 53. Mazarro, R., de Lucas, A., Gracia, I., Rodríguez, J.F. Copolymerization of D,L-lactide and glycolide in supercritical carbon dioxide with zinc octoate as catalyst. J Biomed Mater Res B Appl Biomater. 2008, 85(1), 196-203. 54. Zhang, J., Yang, S., Yang, X., Xi, Z., Zhao, L., Cen, L., Lu, E., Yang, Y. Novel Fabricating Process for Porous Polyglycolic Acid Scaffolds by Melt-Foaming Using Supercritical Carbon Dioxide. ACS Biomater. Sci. Eng, 2018, 4 (2), 694-706. 55. Cabezas, L.I., Gracia, I., de Lucas, A., Rodríguez, J.F. Validation of a Mathematical Model for the Description of Hydrophilic and Hydrophobic Drug Delivery from Biodegradable Foams: Experimental and Comparison Using Indomethacin as Released Drug. Ind. Eng. Chem. Res. 2014, 53(21), 8866–8873. 230

Annex I

56. El-Sherbiny, I.M., Smyth, H.D.C. Controlled Release Pulmonary Administration of Curcumin Using Swellable Biocompatible Microparticles. Mol. Pharm. 2012, 9(2),269-280. 57. Anand P, Nair HB, Sung B, Kunnumakkara AB, Yadav VR, Tekmal RR, et al. Design of curcumin loaded PLGA nanoparticles formulation with enhanced cellular uptake, and increased bioactivity in vitro and superior bioavailability in vivo. Biochem Pharmacol. 2010, 79(3), 330–8. 58. Lu Z, Tsai M, Lu D, Wang J, Wientjes MG, Au JL. Tumor penetrating microparticles for intraperitoneal therapy of ovarian cancer. J Pharmacol Exp Ther. 2008, 327, 673–82. 59. Li, J., Wientjes, M.G., Au, J.L.-S. Pancreatic cancer: Pathobiology, treatment options, and drug delivery AAPS J. 2010, 223-232 60. Au, J.L.-S., Lu, Z., Wientjes, M.G. Versatility of Particulate Carriers: Development of Pharmacodynamically Optimized Drug-Loaded Microparticles for Treatment of Peritoneal Cancer. AAPS J. 2015, 17(5), 1065-1079. 61. Lu, Z., Tsai, M., Wang, J., Cole, D.J., Wientjes, M.G., Au, J.L.S. Activity of drug-loaded tumor-penetrating microparticles in peritoneal pancreatic tumors. Curr. Cancer Drug Targets 2014, 14 (1), 70-78.

231

Scientific articles published

Annex II II II.a. Improvement of PLGA loading and release of curcumin by supercritical technology

II.b. Functionalization and optimization of PLA with coumarin via click chemistry in supercritical CO2

II.c. Copper wire as a clean and efficient catalyst for click chemistry in supercritical CO2

Annex II

235

Annex II

236

Annex II

237

Annex II

238

Annex II

239

Annex II

240

Annex II

241

Annex II

242

Annex II

243

Annex II

244

Annex II

245

Annex II

246

Annex II

247

Annex II

248

Annex II

249

Annex II

250

Annex II

251

Annex II

252

Annex II

253