Smart biomimetic nanosystems for stimuli-responsive drug delivery carriers

INTERNATIONAL DOCTORAL SCHOOL OF THE USC Enrica Soprano PhD Thesis Smart biomimetic nanosystems for stimuli-responsive drug delivery carriers Santiago de Compostela, 2022 Doctoral Programme in Materials Science DOCTORAL THESIS SMART BIOMIMETIC NANOSYSTEMS FOR STIMULI-RESPONSIVE DRUG DELIVERY CARRIERS Enrica Soprano INTERNATIONAL PHD SCHOOL OF THE UNIVERSITY OF SANTIAGO DE COMPOSTELA PHD PROGRAMME IN MATERIAL SCIENCE SANTIAGO DE COMPOSTELA, 2022 DECLARACIÓN DEL AUTOR/A DE LA TESIS D./Dña. Enrica Soprano Título de la tesis: Smart biomimetic nanosystems for stimuli-responsive drug delivery carriers Presento mi tesis, siguiendo el procedimiento adecuado al Reglamento y declaro que: 1) La tesis abarca los resultados de la elaboración de mi trabajo. 2) De ser el caso, en la tesis se hace referencia a las colaboraciones que tuvo este trabajo. 3) Confirmo que la tesis no incurre en ningún tipo de plagio de otros autores ni de trabajos presentados por mí para la obtención de otros títulos. 4) La tesis es la versión definitiva presentada para su defensa y coincide la versión impresa con la presentada en formato electrónico. Y me comprometo a presentar el Compromiso Documental de Supervisión en el caso que el original no esté depositado en la Escuela. En Santiago de Compostela, 22 de febrero de 2022. Firma electrónica AUTORIZACIÓN DEL DIRECTOR DE LA TESIS Smart Biomimetic nanosystems for stimuli-responsive drug delivery carriers Dª. Beatriz Pelaz Dª. Ester Polo INFORMAN: Que la presente tesis, se corresponde con el trabajo realizado por Dª. Enrica Soprano, bajo mi dirección, y a utorizo su presentación , considerando que reúne l os r equisitos exigidos en el R eglamento de Estudios de Doctorado de la USC, y que como director de esta no incurre en las causas de abstención establecidas en la Ley 40/2015. De acuerdo con lo indicado en el Reglamento de Estudios de Doctorado, declara también que la presente tesis doctoral es idónea para ser defendida en base a la modalidad de Monográfica con reproducción de publicaciones, en los que la participación de la doctoranda fue decisiva para su elaboración y las publicaciones se ajustan al Plan de Investigación. En Santiago de Compostela, 22 de Febrero de 2022. AUTORIZACIÓN DEL TUTOR DE LA TESIS Smart Biomimetic nanosystems for stimuli-responsive drug delivery carriers D. Pablo Taboada Antela INFORMA: Que la presente tesis, se corresponde con el trabajo realizado por Dª. Enrica Soprano, bajo mi tutorización. y a utorizo su presentación , considerando que reúne l os r equisitos exigidos en el R eglamento de Estudios de Doctorado de la USC, y que como director de esta no incurre en las causas de abstención establecidas en la Ley 40/2015. De acuerdo con lo indicado en el Reglamento de Estudios de Doctorado, declara también que la presente tesis doctoral es idónea para ser defendida en base a la modalidad de Monográfica con reproducción de publicaciones, en los que la participación del doctoranda fue decisiva para su elaboración y las publicaciones se ajustan al Plan de Investigación. En Santiago de Compostela, 22 de Febrero de 2022. 7 PTX Paclitaxel QDs Quantum dots RBC Red blood cell RBCM Red blood cell membrane RBCNP Red blood cell coating PLGA RES Reticuloendothelial system RGD Arg-Gly-Asp peptide ROI Region of interest RPPN Red blood cell-coated paclitaxel-loaded polymeric core nanoparticle RT Room temperature SDSPAGE Sodium Dodecyl SulphatePolyAcrylamide gel electrophoresis SEM Scanning electron microscopy siRNA Short interfering RNA SMC Smooth muscle cells SpA Staphylococcus aureus protein A SPION Superparamagnetic ironoxide nanoparticles SPR Surface plasmon resonance SSC Side scatter STM Scanning Tunneling Microscope TEMED N,N,N′,N′- Tetramethylethylenediamine TNF Tumor necrosis factor TNF-α Tumor necrosis factor α TRAIL Tumor necrosis factorrelated apoptosis factor VLP Virus-like particles WBC White blood cell 8 Acknowledgments Me gustaría agradecer a mis directoras de esta Tesis doctoral, Dra. B. Pelaz y Dra. E. Polo, y al Dr. P. del Pino. Gracias por todas las enseñanzas y por darme la oportunidad, el apoyo y el tiempo que necesitaba para entender que quería emprender este camino. Sabed que no fue mi gran amor por Santiago lo que me animó, sino vuestro ejemplo de pasión y entusiasmo por este trabajo. Espero seguir buscando este tipo de ilusión para el resto de mi carrera profesional. También quería agradecer al Dr. P. Taboada, a la Dra. S. Barbosa y al Grupo de física de coloides y polímeros (GFCP) por su colaboración y apoyo. Así como al Instituto de investigación sanitaria de Santiago de Compostela (IDIS), y a la colaboración con el Dr. M. Collado. Agradecer a la Agencia Estatal de investigación, (CTQ2017-89588-R), y comisión europea (0624_2IQBIONEURO_6_E-1, y ERC-St SPACING 95041) por la financiación que hizo posible mi Tesis doctoral. Quiero también agradecer a todo el personal de RIAIDT por el análisis de microscopia electrónica que aparecen a lo largo de este trabajo. De igual manera, gracias a la Universidad de Santiago de Compostela, al CiQUS y a todos sus técnicos por poner a mi disposición sus medios de apoyo para el correcto desarrollo de esta Tesis doctoral. 9 Abstract The research conducted in this thesis finds its purpose in the development of cell-derived biomimetic nanosystems, aiming to further broaden nanomedicine´s tools towards advanced therapeutics in the drug delivery field. Nanotechnology offers a plethora of opportunities to design nanodelivery platfoms capable of crossing biological barriers and achieving specific targeting. Nowadays, nanoparticles (NPs) have gained widespread attention as drug delivery vehicles. Thanks to their physical and chemical properties, NPs can be tuned with multiple features, which makes them promising candidates for personalized nanomedicines. Despite the remarkable progress of the last decades in the field, multi-ligand strategies for targeted NPs delivery still suffer from poor in vivo efficacy caused by inadequate interfacing of synthetic NPs with biological environments. The main drawbacks are low specificity in vivo, fast renal clearance, short permanence time in blood torrent and the activation of the immune response. An innovative approach that exploits the natural biological membrane of living cells has been proven to be a successful strategy to improve the in vivo behavior of synthetic nanocarriers. Designing nanostructures by replicating the natural (bio)-physical properties and highly complex functionalities of the cell surface provides delivery nanosystems with effective biointerfacing, preserving the key functionalities of the origin cells. Depending on the type of the cell source, specific functionalities can be exploited for the development of tailored systems with different therapeutic outcomes, which regulate signaling, transport process and immune responses. Specific targeting of cancer cells represents an important challenge for nanomedicine to promote tumor regression while reducing the side effects of anti-cancer drugs. Potentially, cancer cell-derived NPs can exploit their homotypic affinity towards cancer cells from which they are extracted. It is known that cancer cells shown altered expressions of adhesion molecules on their surface that promote changes in their adhesive properties. In this thesis, cancer cells were used to obtain biomimetic membrane based nanosystems, with the aim of exploiting natural cell membrane features to enhanced biocompatibility, obtain high specific targeting efficacy and improve immune evading capability. 10 The first part of this work was aimed to design and develop bio-synthetic cell-derived NPs (named cellsomes, CSMs) using plasmatic membranes of cancer cells. An easily scalable bottom-up process was optimized to obtain CSMs from different cell lines. Their physicochemical and biological properties were comprehensively characterized. To this end, a large variety of techniques were used to study the physiochemical properties of CSMs such as their morphology, hydrodynamic size, colloidal stability in different media, and concentration, among others. The presence of cell membrane components and specific surface biomarkers inherited from the cell source was assessed. The interaction between CSMs and different cell lines was investigated in vitro. Selective targeting capabilities of CSMs for the homotypic tumor cells were demonstrated across several tumoral and nontumoral cell lines. Hybrid CSMs composed of tumoral and non-tumoral cell membranes fragments proved the possibility of translating the targeting capabilities of the tumoral derived CSMs into other non-cancer cell-derived CSMs, thereby expanding their versatility and potential applications. To study the possibility of applying biomimetic coatings to synthetic materials, solid NPs (i.e., polystyrene NPs) were coated with CSMs membranes. The biomimetic effect on the biointerfacing behaviour of coreshell NPs was studied in 2D and 3D cell culture models. Specifically, it was observed that the CSMs-coated NPs inherently mimic the surface properties of the source cells and thus: (i) acquire homotypic targeting ability; (ii) decreased uptake by macrophage cells and; (iii) present a more efficient penetration into 3D tumor model. Remarkably, the possibility to escape from the immune system by avoiding the rapid recognition by the RES, confers an enhanced therapeutic efficacy to synthetic NPs. Based on the studies on their biomimetic properties and their exciting capability to interact with cells, the CSM carrier proved to be a promising candidate for the development of an effective drug delivery nanosystem. Two kinds of smart CSMs-based nanocarriers were set up in this study, focusing on the development of innovative strategies to achieve the intracellular delivery of different cargoes that otherwise would not overcome the cell membrane barrier or would be trapped in endo/lysosome vesicles. The first nanocarrier design was based on a hybrid stimuli-responsive drug delivery nanostructure. The CSMs were combined with photoresponsive NPs (gold nanorods, GNRs), which upon near infrared (NIR) activation, rapidly convert the absorbed energy into thermal energy, 11 mediating the heating of the local environment. The GNRs-tagged CSMs presented several advantages inherited from the cell-membrane nature, such as the ability to target specific cell populations (homotypic targeting) and preventing cargo´s degradation. On the other hand, due to the thermoplasmonic properties of plasmonic GNRs, a spatiotemporal-controlled intracellular release of cargoes into the cytosol of living cells was obtained under NIR stimulation. NIR-triggered cargo release was demonstrated either at the level of single cells (micrometer NIR spot) and at the level of thousands of cells (energy-homogeneous collimated NIR excitation area of ∼0.33 cm2; circular spot with a diameter of ~0.65 cm). Through these procedures, photo-controlled intracellular delivery of non-permeant antibodies (anti-Tubulin antibody as a proof of concept) was achieved without compromising cell viability nor the antibody´s function. These results set the stage for the development of photoactive cell-derived nanocarriers, which in addition to cell specific functions, promise straightforward access to spatiotemporal-controlled intracellular delivery of antibodies for application in different immunotherapies. A second alternative strategy to achieve intracellular cargo delivery, escaping the lysosomal entrapping, was successfully developed by surfaceengineering the cell-derived nanocarrier to obtain fusogenic CSMs (FCSMs). The direct fusion of the nanocarrier with the plasma membrane of the cells is a straightforward strategy to achieve the intracellular release of the carried cargos into the cytosol. To this aim, the lipidic composition of CSMs was modified by integrating an optimal combination of cationic and dye-labeled neutral lipids (DOTAP and DOPE-Atto) without compromising the cellderived membrane features. FCSMs were efficiently loaded with manifold types of cargo, from small hydrophobic molecules such the bisbenzimide compound Hoechst H 33258 (HOE) to large molecular mass macromolecules such dye-labeled phalloidin, dextran polymers or polystyrene beads. Upon entering in contact with living cells, in contrast to non-fusogenic CSMs, FCSMs were proved to be able to fuse with the cell membrane of living cells, thereby leading to the release of the encapsulated compounds into the cytosol of cells, avoiding the endocytosis pathway and lysosomal entrapping of the cargo. This technology could represent a powerful tool for fast cytoplasmatic delivery of sensitive drugs, especially proteins and nucleic acids, enabling of designing a new generation of carriers for nanovaccines. 12 The results of the research conducted in this thesis underline the importance of the biomimetic cell-derived coating technology, which offers a versatile tool for developing improved drug delivery nanovectors by easily recreating natural scenarios at the bionano-interface. These biomimetic interfaces have emerged to overcome some of the main drawbacks inherent to synthetic nanomaterials by translating specific complex functionalities from the cell surfaces that remain complex to replicate synthetically. 13 Resumo A investigación realizada nesta tese ten como obxectivo o desenvolvemento de nanosistemas biomiméticos derivados de células para ampliar aínda máis as ferramentas de nanomedicina cara a terapias avanzadas no campo da administración de fármacos. Nas últimas décadas, a nanotecnoloxía experimentou un crecemento exponencial que continúa na actualidade, introducindo un novo enfoque científico-tecnolóxico que fai uso da materia á nanoescala, dotándoa de propiedades e características especiais, relativas ás propiedades ópticas, eléctricas e magnéticas, á reactividade química e á fluorescencia. Grazas a ditas propiedades, as nanopartículas (NP) adquiriron unha ampla atención como vehículos de administración de fármacos. O uso de nanotecnoloxías farmacéuticas permite superar as propiedades críticas (liberación, penetración, estabilidade) e unha farmacocinética desfavorable das moléculas activas. As vantaxes de utilizar NPs como vehículos para a administración de fármacos están relacionadas co seu pequeno tamaño, permitíndolles atravesar barreiras biolóxicas e pequenos capilares, alcanzando así destinos de interese como tecidos, tumores ou células individuais. Ademais, a súa estrutura ofrece a posibilidade de encapsular o fármaco ou conxugalo na superficie, mediante adsorción ou enlace químico, protexendolo deste xeito da prematura degradación e/ou eliminación in vivo e, ao mesmo tempo, garantindo a súa solubilidade no medio biolóxico. As dimensións dos nanosistemas presentan unha ampla superficie en comparación cos materiais de maiores dimensións, o que mellora a capacidade de unión de moléculas na superficie que atribúen unha funcionalidade específica ás NPs. Esta característica ofrece a capacidade de alcanzar dianas biolóxicas grazas á conxugación cun ligando específico (anticorpo, péptido, etc.). Ademais, o emprego de NP como vectores favorece a acumulación do fármaco no lugar de interese terapéutico, reducindo a súa dispersión no organismo. En consecuencia, permite non só diminuír a frecuencia de dosificación, senón tamén os efectos secundarios, resultando nun sistema máis beneficioso para o paciente. A pesar do notable progreso deste campo nas últimas décadas, as estratexias multiligandos para a administración dirixida de NPs sofren dunha escasa eficacia in vivo causada pola interacción inadecuada das NP sintéticas con ambientes biolóxicos. Tras a administración, as NP enfróntanse a un ambiente complexo que presenta numerosas barreiras físicas e biolóxicas 14 para a súa acción. A primeira delas é ofrecida polo torrente circulatorio, xa que o sangue é rico en proteínas que se adsorben na superficie das NPs formando unha coroa biomolecular. Esta coroa confire ás NPs unha nova identidade biolóxica, desempeñando un papel importante na súa distribución e posiblemente comprometendo a súa acción. A adsorción de proteínas plasmáticas (como apolipoproteínas, compoñentes do complemento e inmunoglobulinas) na superficie da NP facilita o seu recoñecemento e a súa eliminación do sangue (opsonización), accións levadas a cabo polos fagocitos circulantes, así como polos macrófagos dos tecidos (principalmente as células de Kupffer hepáticas e os macrófagos do bazo), tamén coñecido como sistema fagocítico mononuclear (MPS). Este rápido recoñecemento e secuestro representan o principal desafío para o desenvolvemento de función eficaces in vivo das nanoformulacións. Ademais, a exposición de certos dominios proteicos pode desencadear de xeito potencial vías específicas de recoñecemento celular, dando como resultado a activación de determinados procesos biolóxicos como a activación do sistema inmunitario. Un enfoque innovador que explota a utilización da membrana biolóxica natural das células vivas demostrou ser unha estratexia exitosa para mellorar o comportamento in vivo dos nanoportadores sintéticos. Deseñar as nanoestruturas replicando as propiedades (bio)-físicas naturais e as funcionalidades altamente complexas da superficie celular proporciona aos nanosistemas de entrega unha biointerfaz eficaz, preservando as funcionalidades clave das células de orixe. Dependendo do tipo celular utilizado, pódense transferir funcionalidades específicas para o desenvolvemento de sistemas personalizados con diferentes resultados terapéuticos. Deste xeito, é posible regular a sinalización, os procesos de transporte e as respostas inmunitarias. Por exemplo, as NPs recubertas de membranas de glóbulos vermellos (RBC, tamén coñecidas como eritrocitos) asócianse cun maior tempo de circulación sanguínea, reducindo a súa eliminación polo sistema inmunitario. As NPs derivadas de células nai presentan boas propiedades de orientación y emerxen como posibles terapias contra o cancro debido á predisposición destas células a migrar cara o tecido afectado e inflamado. Normalmente, estas rexión caracterízanse por condicións ambientais como a hipoxia ou á interacción con receptores ou citocinas tipo Toll. 15 A decoración da membrana celular dos macrófagos promove a captación das NPs polas células do cancro de mama metastático mediante a interacción entre a integrina α4 e a molécula de adhesión celular vascular-1 (VCAM-1), o cal se viu que produce un notable efecto inhibitorio sobre a metástase pulmonar. Os leucocitos están implicados na protección do organismo contra enfermidades contaxiosas e invasores foráneos. Polo tanto, as membranas dos leucocitos teñen a propiedade de evadir o sistema inmunitario e de localizar os tecidos ou diana, mostrando así a súa capacidade de orientación mediante interaccións célula-célula. As NPs recubertas de membranas de células cancerosas presentan unha excelente internalización por parte das liñas celulares orixinais (afinidade homotípica). En particular, a interacción específica coas células tumorais representa un reto importante na nanomedicina, co obxectivo de promover a regresión do tumor ao mesmo tempo que se reducen os efectos secundarios de fármacos anticanceríxenos. As células tumorais agloméranse facilmente o que desencadea a formación de tumores sólidos debido aos mecanismos de adhesión celular mediados por proteínas específicas (proteínas de adhesión focal, integrinas, etc.) da superficie celular. Desta maneira, ditas interaccións homotípicas entre as células tumorais pódense aplicar como estratexia de especificidade de cara ao tumor mediante revestimentos biomiméticos que permiten o autorecoñecemento. Nesta tese, as células cancerosas utilízanse para obter nanosistemas baseados en membranas biomiméticas co obxectivo de aproveitar as características naturais das membranas celulares para mellorar a biocompatibilidade, obter unha alta especificidade, e promover a capacidade de evasión inmunolóxica. A primeira parte do traballo realizado ao longo desta tese, tivo como obxectivo deseñar e producir NPs bio-sintéticas derivadas de células denominadas cellsomas (CSMs) utilizando membranas plasmáticas de células cancerosas. Optimizouse un proceso ascendente facilmente escalable para obter CSMs de diferentes liñas celulares. Os CSMs producíronse mediante a combinación de dous métodos de perturbación física capaces de romper as membranas usando diferentes forzas externas como a calor e a presión. A lise celular completa levouse a cabo mediante (i) un choque osmótico producido mediante unha solución tampón hipotónica e, (ii) lise térmica realizada mediante ciclos repetidos de conxelación-desconxelación. 16 A solución hipotónica proporciona a presión osmótica necesaria para expansión celular. A concentración de sal que rodea a célula diminúe rapidamente, polo que a membrana celular faise permeable á auga debido ao efecto osmótico. A combinación da presión osmótica e o efecto de choque térmico provocado polos ciclos de conxelación-desconxelación induce aberturas transitorias na membrana celular para baleirar a célula dos seus compoñentes. Para minimizar a posible desnaturalización das proteínas de membrana, a extracción da membrana plasmática foi o máis suave posible, co obxectivo de manter a súa bioactividade. As proteínas solubles, as biomacromoléculas intracelulares e o núcleo foron eliminados mediante centrifugación. Posteriormente, os fragmentos de membrana purificados son presionados contra pasar a través de membranas de policarbonato con poros nanométricos para obter CSMs autoensamblados cun tamaño que oscila entre 100 e 400 nm. As nanoestruturas caracterizáronse completamente mediante unha gran variedade de técnicas coa finalidade de avaliar as características fisicoquímicas dos CSM: Scanning Electron Microscopy (SEM) para a determinación da morfoloxía, Dynamic Light Scattering (DLS) para o tamaño hidrodinámico e a estabilidade coloidal, anemometría láser Doppler para o potencial ζ, Nanoparticle Tracking Analysis (NTA) para a análise de concentración e distribución de tamaño e fluorescencia espectroscopia para cuantificar a eficiencia de carga de moléculas fluorescentes, entre outras. Realizouse un estudo das propiedades biolóxicas e avaliouse a presenza de compoñentes da membrana celular, como biomarcadores específicos de superficie transferidos da liña celular orixinal. A interacción entre CSMs e diferentes liñas celulares investigouse mediante estudos in vitro. A capacidade dos CSMs de levar a cabo un mecanismo de interacción homotípica coas células tumorais móstrase en varias liñas de células tumorais e non tumorais. Por outra banda, CSMs híbridos compostos por fragmentos de membranas celulares tumorais e non tumorais demostraron a posibilidade de trasladar as capacidades interacción específica dos CSMs derivados do tumor a outros CSMs, ampliando así a súa versatilidade e aplicacións potenciais. Para estudar a posibilidade de aplicar revestimentos biomiméticos a materiais sintéticos, recubríronse NPs sólidas (neste caso, NPs de poliestireno) con membrana CSMs. O comportamento e efecto biomimético das NPs formadas por núcleo e cuberta foi estudado en modelos de cultivo celular 2D e 3D. En concreto, observouse que as NP revestidas de CSMs 23 The term "nanotechnology" was actually coined only in 1974 by the researcher Norio Taniguchi, of the University of Tokyo, in relation to the ability to accurately manipulate materials on a nanometric scale, the nanomaterials.3 The term was introduced in the scientific and social field in 2000, on the occasion of the National Nanotechnology Initiative (NNI).4 Subsequently (2004), the Royal Society & Royal Academy of Engineering (UK) defined nanoscience as the study of phenomena and manipulation of materials on the atomic, molecular and macromolecular scales, at which materials acquire chemical-physical properties that differ significantly from those of materials on a larger scale. Nanotechnologies were defined as a discipline for the design, characterization, production and application of structures, devices and systems at the nanoscale.5 Nanotechnologies have undergone an exponential growth that continues today, introducing a new scientific-technological approach that exploits matter at the nanoscale. At the nanoscale, manifold materials present optical, electrical and magnetic properties, chemical reactivity and fluorescence that diverge from these at the macroscopic scale.6 Along with the development of strategies to synthesize nanomaterials, an effort has been made to characterize the unique size-dependent properties of nanoparticles (NPs). Indeed, the radical change in material dimensions affects various physicochemical properties to a large extent, which needs to be understood to exploit them. An important peculiarity of NPs is their high surface to volume ratio. Cutting a 1 cm3 cube into 1021 cubes, each of 1 nm side, will result in the same overall volume and mass, but with an increase of the surface area by a 10 million factor.7 This aspect is fundamental if we consider the NPs in a biological system,8 since it allows to have a greater surface of interaction with the molecular targets, as well as a wider functionalization surface, with an exponential increase in the molecules’ reactivity. In addition, the high surface to volume ratio implies great differences in their electronic, optical, and chemical properties as well as in their mechanical characteristics. For instance, the force of gravity at nanoscale range becomes less influential, being directly proportional to the body weight, while weak interaction forces, such as Van der Waals forces and surface tension, become more relevant. The classical physics laws are no longer appropriate to describe the behavior of nanoscale matter and are replaced by those of quantum physics. The high confinement of electrons causes an increase in the bandgap of semiconductors, optical absorption, and photo-luminescence. Furthermore, mechanical properties such as elasticity, hardness and ductility also change. The distinctive properties of NPs open the way to a wide range of 24 applications in multiple fields, from industry to medicine, from information technology to cosmetics (Figure 1.1.2).9 Figure 1.1.2. The major fields of use of nanotechnology and percentage of incidence. Among the various sectors of interest in which nanotechnology is involved, in the field of nanomedicine, it can offer unique powerful solutions to fight manifold diseases such as cancer, immunological diseases and to develop new vaccines or novel personalized nanomedicines.10-12 Nanomaterials are similar in scale to biologic molecules as nanoscale systems and colloidal syntheses have enabled the engineering of NPs with innumerable designs (i.e., composition, structure, surface chemistry and functionalities), making the nanotechnology potentially useful in several biomedical applications.13 The field of nanomedicine aims to use the properties and physico-chemical characteristics of nanomaterials for the diagnosis and treatment of diseases at the molecular level. 1.1.1. Nanomaterial classification A wide range of NPs have been designed and described for different applications in nanomedicine. They can be classified based on different parameters, namely, their origin (natural or anthropogenic), chemical composition (organic and inorganic), formation (biogenic, geogenic, anthropogenic, and atmospheric), their size, shape and characteristics, their applications in research and industry, etc. As for chemical composition, the first category is represented by inorganic NPs, consisting mainly of metals and their oxides, such as gold NPs, iron oxide, quantum dots (QDs) and silica NPs (Figure 1.1.3). Their 25 nature defines different properties such as high electron density and strong optical absorption (in particular Au),14 photoluminescence in terms of fluorescence (semiconductor QDs, e.g., CdSe or CdTe)15 or phosphorescence (doped oxide materials, e.g., Y2O3), or magnetic moment (e.g., iron oxide or cobalt NPs).16 These properties are widely used for diagnostic and therapeutic purposes.17 Figure 1.1.3. Some examples of inorganic and organic NPs. Organic NPs are made from totally biodegradable materials. For this aspect they are very useful for applications in the medical field, given their compatibility with biological materials. This class of NPs includes liposomes, dendrimers, micelles, and protein-based NPs (e.g., ferritin), mostly having spherical three-dimensional morphology. They not only have a surface that can be exploited for anchoring molecules to increase specificity and selectivity of action, but also have an internal cavity that is mainly used to encapsulate organic material such as drugs, nucleic acids or other molecules that may be insoluble in water. These characteristics determine their main use as vehicles for the transport of molecules to the target of interest (drug delivery systems).11,18,19 The third category is represented by hybrid (core-shell) NPs, so named for their composition combining organic and inorganic material (Figure 1.1.4). They can have an internal core, usually of an inorganic nature, which 26 constitutes the origin of the magnetic or fluorescence signal, and an external shell, i.e., the shell of organic nature, which covers the internal portion and makes the NP biocompatible.9,17 The outer shell can be functionalized so that the surface exposes recognition molecules, such as antibodies or ligands for specific receptors.20,21 The functionalization of these elements leads to the formation of nanocompounds which, thanks to their magnetic and electronic properties, can be exploited as nanodevices. Furthermore, hybrid NPs offer a tool for the study of biological interactions,22 an innovative drug delivery system and the possibility of new theranostic techniques.23,24 Figure 1.1.4. Illustration of multifunctional nanocompounds consisting of multiple inorganic NP cores, functionalized organic coating shells and different surface conjugated biomolecules (adapted with permission from ref.24 Copyright 2013, Royal Society of Chemistry). In the last decade, there has been considerable interest to develop bioinspired nanomaterials derived from biological entities already present in nature. Thanks to their biointerfacing capabilities, cellular structures (i.e., erythrocytes, leukocytes, platelets, and exosomes) or invasive pathogens (i.e., bacteria and viruses) can be exploited to design a new class of nanomaterials in order to overcome the limitations of synthetic NPs.25 Biomimetic membrane-derived NPs will be described in the next sections. 27 1.1.2. Nanobiotechnology in medicine In the context of biomedical application, nanobiotechnology is an extended term that can be defined as the manipulation of materials at molecular levels, which aims to produce non-toxic bioactive devices that have specificity toward the desired tissue type and location. Nanobiotechnology can find its niche in a multitude of applications, wherein nanomaterials provide great advantages in developing small detection and theranostic systems given their small size, thereby allowing the targeting and detection at the single cell or even at molecular level.26,27 1.1.3. Nanodiagnostic To date, the creation of multifunctional tools that allow simultaneous diagnosis and therapy falls within the scope of theranostics, for which nanomaterials are proving excellent candidates. Nanotechnology-based sensors (e.g., nanowires, nanotubes, NPs, cantilevers and micro /nanoarrays) can enable fast and high throughput detection of disease biomarkers under non-invasive conditions, allowing the early identification of the disease and, ideally, the detection of the predisposition to it. Thanks to their small size and high functional versatility, NPs prove to be excellent candidates as contrast agents to obtain sharper and easier-to-analyze images.28 In fact, they are used as diagnostic devices for in vivo applications as contrast agents for medical imaging such as Positron Emission Tomography (PET) and Magnetic Resonance Imaging (MRI)) of various pathologies.29 Nanotechnology also offers innovative solutions for early detection of viruses, bacteria and circulating tumor cells, as well as for single cell analysis. 30,31 Subsequent to diagnosis, the same NPs can be conjugated with a pharmacologically active agent and used for therapy, as detailed in the next section. Certain nanomaterials have unique therapeutic properties that differ from conventional drugs and therefore, can be directly used to treat diseases. For instance, hafnium oxideand gold-based NPs can greatly enhance X-ray therapy;32 gold nanoshells/nanorods, carbon nanotubes, magnetic NPs can induce hyperthermia to kill cancer cells;33 and nanocrystalline silver is used as antimicrobial agent. Some other possible opportunities include reporting in vivo efficacy of therapeutics and tracking nanocarrier biodistribution in the body. 28 1.2. Drug delivery system Drug delivery systems (DDS) represent alternative systems to today's drug delivery methods that allow for the spatial and temporal control of drug release within the body.34 The use of pharmaceutical nanotechnologies allows to overcome the critical biopharmaceutical properties (release, permeation, stability) and an unfavorable pharmacokinetics of the active molecules. Thus, substances that show low bioavailability can be conveyed by particles in such a way as to be directed to the specific site of action, modulating their release, and increasing their absorption. Many of these systems can use polymers with particular chemical and physical characteristics, such as biodegradability, bioabsorbability and sensitivity to changes in temperature and pH. The advantages of using NPs as DDS are related to their small size, which allows them to cross biological barriers and small capillaries thus reaching targets of interests such as tissues, tumor or individual cells.35 Furthermore, their structure offers the possibility of encapsulating the drug or conjugating it on the surface, by adsorption or chemical bond, thus protecting it from premature degradation and / or elimination in vivo and, at the same time, guaranteeing its solubility in the biological environment.36 The dimensions of the nanosystems allow for a high surface area compared to bulk materials, which improves the ability to bind to the surface molecules that attribute specific functionality to NPs. This feature offers the great advantage to reach biological targets thanks to the conjugation with a specific ligand (antibody, peptide, etc.). In addition, the use of NPs as drug delivery vectors favors the accumulation of the drug in the site of therapeutic interest and reduces its dispersion in the body. Consequently, it allows not only to decrease dosage frequency, but also to reduce side effects, favoring patient compliance.34 Based on the mechanisms through which the diseased tissue is reached, the DDS are broadly classified into passive and active targeting (Figure 1.2.1). 29 Figure 1.2.1. Representation of passive and active targeting. a) By enhanced permeability and retention effect, NPs passively diffuse through the blood vessels and accumulate in tumor tissue. b) In active targeting, NPs are functionalized with targeting ligands, triggering the receptor-based endocytosis and the enhanced accumulation in tumor tissues (adapted with permission from ref.37 Copyright 2019, Springer Nature Limited). Passive targeting is driven simply by the size and shape of the NPs, which determine their biodistribution and accumulation at the tissue level.38 For this reason, passive targeting is mainly used to treat pathologies that alter the characteristics of the body's tissues, as in the case of tumors. The physiology of solid tumors has been shown to differ from that of normal tissues in several aspects, including abnormal blood vessels, with slowflowing dilated capillaries.39 The endothelium of the afferents that are formed present large fenestrations with a diameter of 600-800 nm, which allows the extravasation of NPs up to 600 nm in size from the bloodstream to the tumor tissue.40 Furthermore, the low efficacy of lymphatic drainage in these tissues allows prolonged retention times of NPs at the tumor level. Passive permeation at the tumor tissue level is defined as the Enhanced Permeability and Retention effect (EPR) (Figure 1.2.2).41 30 Figure 1.2.2. Representation of the EPR and vascular pathophysiology. a) and c) microcirculation of a healthy tissue, b) and d) microcirculation of a tumor tissue with numerous blood vessel afferents (adapted with permission from ref.42 Copyright 2013, Royal Society of Chemistry). Although the EPR effect is pronounced in preclinical models of solid tumor and in humans too, the potential to harness it therapeutically in patients remains unclear, and most nanotherapeutics investigated in clinical trials showed no improvement in relation to it, compared to conventional chemotherapy.43,44 Because passive targeting does not rely on biochemical identification, it has the disadvantage of low target specificity. In fact, it can affect normal cells or inflamed areas that, in hypoxic conditions or in a pathological state, may be characterized by vessels with the same characteristics as in tumor tissue, reducing the possibility of a selective transport system.40 The disadvantages of passive targeting can be overcome by functionalizing the surface of the NPs with specific molecules such as peptides, proteins or antibodies. This approach is called active targeting and is based on the capability to identify and bind to cell-specific ligands that are expressed on the plasma membrane.45,46 Trafficking processes, access to barriers and into specific cellular locations are regulated by active biological recognitions. To be successful, active targeting requires that the receptor of interest be exclusive of, or overexpressed by, the cells of the target tissues,47 b) c) d) a) 31 in order to achieve a preferential drug accumulation in the diseased tissue and, thus, a selective therapeutic system. Targeting moieties such as antibodies,48 aptamers,49 transferrin,50 biotin51 and folic acid,52 among others, are employed as ligands to promote the specific recognition of the cellular plasma membrane components. Targets can also be part of the intracellular or extracellular matrix (ECM), such as mitochondria,53 nuclei54 or lysosomes,55 with specific ligands (nuclear localization signal (NLS) peptide, ceramide chain, etc.) being selected for their recognition. 1.2.1. Cargo release The release of the drug carried by the NPs (cargo) is a key step in nanotheranostic applications. Among other advantages, nanocarriers can be designed to respond to the microenvironment changes that trigger the drug release in the target site. A number of infection and inflammation-derived conditions, associated to different pathologies, can be exploited as stimuli to promote the disassemble of the nanocarrier, such as variations in pH,56,57 oxygen,58 specific enzymes59 or biomolecules.60,61 In this way the drug is released specifically inside the target cells or in the target tissue where it has to play its role. For example, in acid-responsive DNase I nanocapsules the polymeric shell are shed at low pH, resulting in self-degradation and promoting the release of Doxorubicin (DOX) for enhanced therapeutic efficacy.62 The drug release can also be triggered by external physical stimuli such as ultrasounds,63,64 light65,66 or magnetic field67,68 in a spatiotemporal controlled way. Several strategies are developed in the NP formulation in order to combine the biocompatible properties of organic nanomaterial with the physicochemical properties of inorganic nanomaterials. Magnetic NPs such as superparamagnetic iron-oxide NPs (SPION) under an external alternating magnetic field can increase the surrounding temperature, a property that can be exploited to improve tumor penetration combined with hyperthermia or anti-cancer drug delivery. Furthermore, magnetic nanocarriers can reach deep into the tumor and exhibit a switchable drug release through an external alternating magnetic field.69 Ultrasound-responsive NPs can progress from unstable large-sized particles (100 to 300 μm) to smaller particles of 1-8 μm, which can cross the tumor barrier for image-guided drug delivery. This kind of particles include 32 gas-generated NPs, which at the tumor can form large numbers of CO2 bubbles that efficiently absorb ultrasound at the specific site and timing for ultrasound-actuated drug delivery.70 Moreover, Correa-Paz et al. recently encapsulated a protein drug (the serine protease tissue plasminogen activator, typically used in the treatment of ischemic stroke) and demonstrated ultrasound-activated in vivo delivery of such protein drug from the cavity of polymer nanocapsules.71 Many researchers applied light as external stimuli to promote the triggered release and to enhance drug distribution. Plasmonic NPs such as gold NPs (GNPs) are particularly interesting in this contest because they can absorb and diffuse incident light through the collective oscillation of conduction electrons confined to the surface of the NP, a phenomenon called localized surface plasmon resonance (LSPR).14 Using a light radiation with a frequency strongly superimposed on the SPR absorption band of the NPs, the absorbed energy is rapid dissipated and converted into thermal energy, mediating the heating of the local environment. The great advantage of GNPs is that their LSPR can be extended to the near infrared window (NIR) region by tuning their size, shape and functionalization. The NIR wavelength range (650-900 nm) is the so-called "biological window" since the absorption of light radiation by water, hemoglobin and, therefore, the tissues of the human body is minimal and the light is able to penetrate deeply into the soft tissues.72 These properties thus open up great possibilities for the use of GNPs in the biomedical field which can be used not only for the drug delivery and the cargo release,73 but also in cancer therapy by hyperthermia.74,75 In pioneer studies conducted by Parak and co-workers in this field, plasmonic NPs were embedded in a multilayer polymer microcapsule (≈3–5 µm). Upon photoactivation (by focused NIR micrometric beam) such microcarriers internalized by cells allowed for cytosolic delivery of manifold macromolecules, including small drugs,76 proteins,77 fluorescence reporters,78 and nucleic acids.79 It is worth nothing that in these experiments cargo delivery was demonstrated at individual cell level, ‘opening’ the microcapsules one by one with a NIR focused pointer (i.e., optical setup equivalent to optical tweezers), thus leading, for instance, to fluorescence labeling of different intracellular compartments.80 39 poly(lactic-co-glycolic acid) (PLGA) NPs combined with RBCM purified from fresh RBCs showed a half-life significantly longer than PEGylated NPs (39.6 vs. 15.8 h) with an in-blood retainment even 72 h long after injection.112 Improved pharmacokinetic behavior of RBCM-camouflaged NPs is mainly due to the expression of specific membrane proteins, such as CD47 that inhibit the phagocytosis by macrophages residing in the RES system (liver, spleen and lungs). In fact, CD47 is also known as a do not eat me signal and is responsible of the RBCM ability to escape the recognition by the immune system and minimize the premature blood clearance, phenomena observed in PEG functionalized NPs.113 Besides the biological advantages, the RBCM coated-NPs have shown to improve structural rigidity and particle stability, also leading to a more reliable cargo/dye encapsulation.114,115 RBCMs have recently been used for cancer immunotherapy, leveraging host anti-cancer immune reaction.116 Liang et al. developed biomimeticbased photothermal cancer immunotherapy using a nanoformulation of RBCM-derived black phosphorus (BP) QDs nanovesicles (BPQD-RMNVs) combined with Programmed Death-1 antibody (aPD-1); they induced apoptosis in triple-negative breast cancer cells, in situ by NIR laser irradiation and neoantigen release-mediated immune system activation to eliminate residual and metastatic cancer cells. The NIR mediated apoptosis promoted the recruitment of dendritic cells (DC) and the neoantigens release. Subsequently, an intensive tumor-specific CD8+ T cells response was activated against primary and secondary tumor growth (Figure 1.4.3). BPmediated photothermal therapy (PTT) combined with checkpoint antibody treatment has promise as a potential clinical treatment for breast cancer.117 40 Figure 1.4.3. Photothermal cancer immunotherapy via BPQD-RMNVs and aPD-1. a) Schematic representation of preparation of BPQD-RMNVs by the extraction of RBCM. b) administration of BPQD-RMNV-mediated PTT and aPD-1 to a mouse with a 4T1 tumor. c) In situ induction of NIR-mediated apoptosis in tumoral cells and DC recruitment for the cytotoxic T lymphocyte (CTLs) activation d) against the distal tumor. e) Preferentially accumulation of BPQD-RMNVs into the tumor tissue compared to bare BPQD resulted by in vivo images (1) Tumor, (2) heart, (3) lungs, (4) liver, (5) spleen, and (6) kidney). f) Infrared (IR) thermographic maps of temperature increase in the 4T1 tumor-bearing mice upon NIR irradiation. Mice were treated with PBS, RMNVs, bare BPQDs and BPQDRMNVs (adapted with permission from ref.117 Copyright 2019, published by Elsevier B.V.). 41 1.4.1.2. Platelets Another type of cell membrane that can be extremely useful in biomimetic nanotechnology development are the platelets-derived ones. The platelets originate in the bone marrow and are involved in numerous vital process as homeostasis, tissue repair and thrombosis as well as inflammation and adaptive and innate immune responses.118,119 Platelets have a crucial role in cardiovascular disease and carcinogenesis too.120,121 In fact, platelets membrane (PM) expresses proteins such as D-selectin that recognizes and interacts with CD44-overespressing circulating tumor cells (CTCs), which are strictly involved with tumor metastasis and angiogenesis.122 Hu et al. developed PM-coated core-shell nanovesicles (PM-NVs) loaded with (i) tumor necrosis factor (TNF)-related apoptosis factor (TRAIL) and (ii) DOX. In this formulation the TRAIL was efficiently delivered toward cancer cell membrane where it activated the extrinsic apoptosis signaling pathway. Simultaneously, equipped with an acid-responsive encapsulation matrix, the PM-NV can be digested after endocytosis and enhanced the DOX accumulation at the nuclei for activation of the intrinsic apoptosis pathway (Figure 1.4.4).123 Figure 1.4.4. In vivo targeting and antitumor efficacy of PM-NVs. a) Enhanced accumulation in tumor of PMNVs resulted by region-of-interest (ROI) analysis of fluorescent intensities taken out form ex vivo images of tumors and normal tissues. b) Growth inhibition of MDA-MB-231 tumors after treatment with different TRAIL/Dox formulations at day 16 (image of tumors treated with saline (1), TRAIL-Dox-NV (2), TRAIL-PM-NV (3), Dox-loaded PM-NV (4), TRAIL-Dox-PM-NV (5). c) Inhibition of metastatic nodules after i) saline, ii) TRAIL-Dox-NV and iii) TRAIL-Dox-PM-NV treatments. Adapted with permission from ref.123 Copyright 2015, John Wiley and Sons. 42 The same group, to improve the drug accumulation in tumor, developed a new strategy combining two nanocarriers. For the first one, they used ArgGly-Asp (RGD) peptide to decorate tumor necrosis factor α (TNF-α) loaded nanovesicles (NCA). The RGD peptide selectively bind the integrins such as ανβ3, overexpressed in tumor blood vessels, while TNF-α, inflammationinduced cytokine, is applied to trigger tumor vascular damage. The second nanocarrier is a PM-coated acid-responsive dextran nanostructure (NCB) loaded with the chemotherapeutic agent Paclitaxel (PTX). The study showed that PM proteins as CD36, CD42d, P-Selectin and CD40L were efficiently transferred with their origin membranes and enriched the NP coating. Thanks to the specific interaction between the L-Selectin and the CD44 receptor of the tumoral cells, the authors demonstrated that the presence of the PM give to the NPs the ability to target the myeloma cells with high internalization, high intracellular drug localization and decreased side effect.124 1.4.1.3. White blood cells Biomimetic coating for NPs has recently also been obtained from membranes extracted from white blood cells (WBCs), which are recruited into the tumor site in relation to chronic inflammation. First studies showed that this approach can enhance immune evasion and inflammation targeting.125 ‘Leukolike vectors’ (LLVs) retain many critical leukocyte transmembrane proteins from the cell donor,126 which by clustering reduce MPS uptake; others are involved in the adhesion to inflamed endothelium and tumor targeting, or in immune tolerance and interaction with platelets.127,128 Thanks to the presence of lymphocyte function-associated antigen 1 (LFA1 or CD11a), coated NPs bind actively the TNFα-activated endothelium as evidenced by clustering of endothelial intracellular adhesion molecule 1 (ICAM-1). Furthermore, a transwell chamber assay showed the high suitability of LLVs to cross the layer of inflamed endothelium.125 Leukocyte membrane coating was also used for designing nanoformulations for imaging and photothermal therapies. Xhuan et al.129 successfully developed macrophages membrane coated (MPCMs) Au Nanoshells (AuNSs) for PTT cancer therapy (Figure 1.4.5). In mice, the macrophagic coating demonstrated significant biocompatibility increase, opsonization reduction, circulating time prolongation and tumor-tropic 43 accumulation of MPCM-AuNSs enhancement. Moreover, in vivo PTT showed the inhibition of tumor growth upon NIR irradiation and even its disappearance after 25 days. Figure 1.4.5. a) Schematic illustration of the preparation of MPCM-AuNSs and b) in vivo PTT (reused with permission from ref.129 Copyright 2016, American Chemical Society). 1.4.1.4. Mesenchymal cells Mesenchymal cells (MSCs) are multipotent stem cells with the ability to differentiate in other types of cells such as adipocytes, fibroblast, osteoblasts, chondroblasts, and pericytes. Furthermore, the MSCs migrate to the injured and inflamed tissue under environmental conditions such as hypoxia, and interaction with Toll-like receptors or cytokines. Since tumor is considered to be a chronical inflammation disease, MSCs membrane coated NPs strategy has been harnessed as a biomimetic approach for targeted delivery of cancer drugs to tumors.130,131 Toledano Furman et al.132 developed MSCsderived nanoghosts carriers (NGs) as model platform entrapping therapeutics and achieving specific tumor targeting and tumor growth inhibition. On the other hand, synthetic liposomes as negative control did not show analogous results. It was suggested that the NGs ability to bind and fuse to the tumor cell surface was due to the presence of specific MSC integrins, retained on the membrane coating that can mediate NGs interaction with the tumorinfiltrating immune cells, blood vessel endothelium, and tumor-associated fibroblasts.132 Similarly, Changyong and co-workers used MSCs membrane to coat a gelatin nanogel loaded with DOX. Their studies demonstrated high tumor-targeting capability both in vitro and in vivo of their NGs. The MSCs 44 membrane coating significantly improved the cellular uptake, intratumoral accumulation and penetration comparing with gelatin-DOX and free-DOX administration.133 1.4.1.5. Cancer cells Besides RBCs, platelets, WBCs, and MSCs, cancer cells present exciting advantages to be exploited as membrane coating against tumors. Camouflaging strategies based on their use take advantage of innate homotypic aggregation properties and the immune escape ability. Homotypic targeting is the intrinsic ability of cancer cell coating nanoformulation to interact preferentially and strongly with the same cells from which they are originated. This feature provides unique asset for any specific targeted drug delivery strategies against cancer. The homotypic affinity between cancer cells can be attributed to the interaction between galectin-3 and carcinoembryonic antigen expressed on cancer cells.134 Fang et al.135 first studied the homotypic targeting of MDA-MB-435 cancer membrane-coated PLGA NPs (CCNPs) as DDS. CCNPs showed a 20-fold increase accumulation in MDA-MB-435 cells compared with the bare PLGA NPs, while no difference was observed in human foreskin fibroblasts. Therefore, analogous formulation was performed using a non-specific RBC coating PLGA NPs (RBCNPs) and it showed reduced particle binding to the cancer cells, suggesting that the cancer cell coating enhanced particle-cell adhesion.135 A similar approach was applied by Sun and colleagues developing a 4T1 cell membrane-coated paclitaxel-loaded polymeric core as biomimetic (CPPNs) drug delivery system against breast cancer and its metastasis in lungs.136 4T1 coated CPPNs preferentially targeted its tumoral cells but not lung fibroblast WML2 cells or macrophage RAW264.7 cells. The accumulation of that nanoformulation in primary tumor and metastasized site in lungs increased by 3.3and 2.5-fold when delivered with CPPNs instead of bare NPs (Figure 1.4.6). Furthermore, RBC-coated PPNs (RPPNs) and synthetic liposome vesicle coated PPNs (LPPNs) were used as negative control, and they exhibited a rather lower uptake than the CPPNs, suggesting that the enhanced internalization of the CPPNs was probably caused by the 4T1-tumor-cell-membrane proteins obtained from the source cells. The membrane proteins including TF-antigen and E-cadherin associated with the adherence capabilities during the colonization of metastasis lesion also have 45 effects on the homotypic interactions among the tumor cells.137-140 CD44 and CD326, the surface adhesion molecules on the 4T1 cells, have been recognized as surface markers which also played main roles in the adherence of the metastatic cells to the distant sites.141-142 Figure 1.4.6. a) Schematic illustration of CPPNs synthesis procedure. b) Targeting effect of CPPNs resulted from quantitative analysis by flow cytometry after 1 h incubation in different cell lines. c) Ex vivo tissue distribution in the main organs of PPNs and CPPNs (from the left: heart, liver, spleen, lung, kidney, and tumor). d) In vivo bioluminescence imaging of the mice bearing lung metastasis of the 4T1 bloodstream metastasis model with different treatments (adapted with permission from ref.136 Copyright 2016, WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim). Many other results regarding the exciting specific interaction capabilities leverage by cancer cell coating have recently been reported in the literature.143-145 Besides cancer targeted drug delivery, cell membrane coated NPs could also be exploited for developing novel bio-synthetic nanocarriers for vaccines. Because many tumor antigens are surface markers, tumoral cell membrane can activate immune system to recognize and kill malignant tumor cells based on variant antigen expression.146,147 Approaches based on a single tumor-associated antigen can be inadequate when facing the high heterogeneity and mutation rate of cancer cells.148 On the other hand, when cell lysates are used in multiantigen-based strategies to prime the immune system, the large presence of intracellular, housekeeping proteins may divert focus away from the relevant antigens, which compose a small percentage of the total protein, thus compromising the treatment efficacy.149 Therefore, cancer cell coated NPs represent a good approach to combine the homotypic capability to recognize the tumoral cells and the active delivery tumor- 46 associated antigens to dendritic cells for immune processing, which allowed for the subsequent stimulation of tumor antigen-specific T-cells (Figure 1.4.7).135 Figure 1.4.7. Schematic representation of the synthesis of CCNPs. The presence of tumor-associated antigens can be exploited to deliver the antigen to APCs and mediate the activation of immune response against cancer. In the other hand, the selective tumor affinity of the nanoformulation allows to homotypically target the source cancer cells (reused with permission from ref.135 Copyright 2014, American Chemical Society). Despite the growing number of research aimed at developing these new biomimetic systems, the mechanisms involved in biointerfacing with the biological environment, i.e., how these membrane coatings interact with cells at molecular level, still require in-depth studies. In addition, because in most cases the cargo consists of drugs that need to reach the cytosolic target, it is crucial to investigate the intracellular delivery mechanisms. 1.5. Intracellular delivery In addition to the general in vivo improvements in terms of avoided sequestration by MPS, prolonged circulation time and specific cell or tissue targeting allowed by the use of cell membrane coated technology, NPs 47 internalization pathways must be taken into account, as cytosolic delivery is a fundamental prerequisite for obtaining successful drug delivery systems. In most cases, NPs are taken up through the processes of endocytosis, by which cells internalize them in vesicles, the endosomes. Depending on the entry pathway, endosomes can be recycled, exocytized or transported to organelles such as lysosomes, Golgi and mitochondria (Figure 1.5.1).150 Figure 1.5.1. Illustration of the main fates of the endosome. In some cases, the NPs, and their cargos, get trapped inside the endosomes and undergo protease-mediated degradation and exocytosis, resulting in a very limited fraction of successfully delivered molecules achieving their cytosolic target. At this stage, specific methods of drug intracellular delivery are required, which represents one of the most relevant challenges for the protein-based therapies. Intracellular delivery can be achieved by a range of carrier-based or membrane-disruption-based techniques (Figure 1.5.2). 48 Figure 1.5.2. Schematic representation of substance transportation into the cells, four involving molecular cargo (grey), membrane (double lines), and carrier material (purple). physical-induced membrane disruption mechanisms (via permeabilization or direct penetration), and cargo transportation through endocytosis allowed by biochemical assemblies and viral vectors. If a Fusogenic potential of the carrier allows it to enter through membrane fusion. Membrane permeabilization is facilitated by detergents and pore-forming proteins. Schematics at the top show the four subcategories with molecular cargo (grey), membrane (double lines), and carrier material (purple). Physical and mechanical methods are considered the conventional approach to permeabilize the cell membrane. Microinjection, sonoporation, electroporation and other techniques have been developed as membranedisruption modalities to induce transient discontinuities in the plasma membrane using mechanical, electrical, thermal, optical or chemical forces. These approaches found success in several in vitro proposes.151-153 However, the use of these methods in large scale treatments and pharmaceutical applications is severely limited by their low-throughput and disruptive techniques that require sophisticated and expensive instrumentations. As an alternative to mechanical methods, carrier-based delivery nanosystems have inherent advantages making them attractive but, as explained above, when the nanocarriers are taken up by cells via endocytosis the cargo release from the endosome vesicles is required. 55 In particular, the work described in chapter 3.2 aims to provide a photoresponsive drug delivery biomimetic nanocomposite, aiming to achieve intracellular spatiotemporally controlled cargo release. The specific objectives here can be summarized as: • to functionalize CSMs with gold nanorods (GNRs) thereby leading to plasmonic CSMs having thermoplasmonic properties; • to characterize the thermoplasmonic properties of GNRs-tagged CSMs; • to use plasmonic CSMs as photo-responsive biomimetic nanocarriers for intracellular delivery of macromolecules (e.g., from small to large molecules and macromolecules); • to perform in vitro studies of NIR light-controlled release of nonpermeant Abs with spatiotemporal resolution, either at the level of single cells and at the level of thousands of cells, without impairing cell viability. As an alternative drug delivery strategy, the work described in chapter 3.3 aims to achieve intracellular cargo delivery by direct fusion mechanism between the plasma membrane and the phospholipidic structure of CSMs. The specific objectives are: • to derive fusogenic CSMs (FCSMs) by modifying the biomimetic coating composition with different combinations of cationic and neutral lipids; • to verify the biomimetic properties of FCSMs; • to investigate the intracellular delivery of manifold types of cargo, from small hydrophobic molecules to large molecular mass macromolecules or solid NPs from tunable FCSMs. 56 Materials and Methods 57 2. Materials and Methods 2.1. Synthesis and characterization of CSMs 2.1.1. Synthesis Method The synthesis of cellsomes (CSMs) was carried out following the protocol described in the literature,185,186 with a procedure to obtain monodispersed vesicles of variable size ranging between 100 and 400 nm. Synthesized CSMs are derived from different cell lines (tumoral cell lines such as cervix epithelial carcinoma cells, HeLa cell, and adenocarcinomic human alveolar basal epithelial cells, A549 cells; healthy embryonic adrenal cell line, HEK cells; and macrophages, RAW cells), which were purchased from ATCC®. To prepare the CSMs, cells were incubated in 75 cm2 or 175 cm2 cell culture flasks in Dulbecco´s Modified Eagle Medium (DMEM, high glucose (4.5 g·L-1) and pyruvate, GIBCO™ #41966052), supplemented with 10% fetal bovine serum (FBS, GIBCO™ #10270106) and Penicillin-Streptomycin (50 U·mL-1 - 50 µg·mL-1) (GIBCO™ #15140122) (complete DMEM, cDMEM) in a humidified chamber at 37 °C under 5% CO2. Cells were grown in their preferred environment as they approached 70 – 80% surface coverage. Then, the cells were harvested after trypsinization for 2 min with 2 mL of 0.25% Trypsin–EDTA (GIBCO™ #24200056). 10 mL of cDMEM was added to recover the cells and transferred to a 50 mL sterile tube. The cells were collected after centrifugation at 500 g for 5 min. The collected cells (10-20 · 106 cells) were washed with precooled phosphate-buffered saline (PBS, pH 7.4 Thermo Fisher # 14190169) and centrifuged at 600 g for 5 min. The cell pellet was re-suspended in 10 mL of hypotonic buffer (0.25X PBS) containing 1X Protease inhibitor cocktail (PIC, Sigma-Aldrich™ #P2714BTL) and incubated in an ice bath for 10 min (Figure 2.1.1, 1). Then the cell lysis was carried out using a freeze-thaw method consisting of 4 cycles of freezing in liquid nitrogen for 1 min followed by thawing at 37 °C for 10 min. Finally, the solution was placed in a bath sonicator for 5 min (Figure 2.1.1, 2). To purify the cell membrane fragments, the solution was subjected to several centrifugation steps. First the solution was centrifuged at 700 g for 10 min at 4 °C to discard the nuclei or whole cells. Then the cell membrane fragments remaing in the supernatant were precipitated by centrifugation at 15,000 g for 30 min at 4 °C. To allow for self-assemblage of the membrane 58 fragments into CSMs, a mechanical extrusion process was applied. The pellet was dispersed in 1 mL of PBS buffer (or 1 mL of 20 mM of 2-(4-(2hydroxyethyl)-1-piperazinyl)-ethanesulfonic acid (HEPES) buffer) and subject to 10 cycles of extrusion by using an Avanti® Mini extruder with 800 nm polycarbonate membrane (Figure 2.1.1, 3). Selecting a specific pore size of the membrane used during the extrusion process can modulate the size distribution of the final CSM product. Figure 2.1.1. Schematic representation of the synthesis procedure for CSMs preparation. 2.1.2. Physical Chemical Characterization Dynamic Light Scattering (DLS). DLS was used to characterize the colloidal properties of the CSMs by measuring their hydrodynamic size and polydispersity index (PDI). All the samples were analysed in PBS or HEPES buffer by using a DLS Malvern Zetasizer Nano ZSP (Malvern Instrument Ktd.) equipped with a 10 mW He-Ne laser operating at a wavelength of 633 nm laser and fixed scattering angle at of 173º. Disposable cells for DLS were used. All measurements were carried out in PBS buffer at 37 ºC. Cells 500 g 5min PBS 1x 500 g 5min PBS 0.2x Cell lysis: Freeze / Thaw method + PIC ice 10min Liquid N2 1min 37 ºC 10 min 5 min Extrusion cell membrane fragments Purification pellet pellet 0.8 mm 59 ζ potential. ζ potential of the MilliQ water dispersed CSMs was measured with laser Doppler anemometry (LDA) by using the same Malvern Zetasizer Nano ZSP instrument. Disposable capillary zeta cell for ζ potential were used. Nanoparticle tracking analysis (NTA). NP tracking analysis was used to characterize the colloidal properties of the CSMs and determine the CSMs concentration in solution. All the samples were analysed by using a NanoSight NS300 (Malvern Instruments, UK) equipped with a 488 nm laser module, a sCMOS camera and a syringe pump was used for all NTA measurements. All measurements were carried out at 24 ºC. A CSMs were diluted in filtered MilliQ water to a final volume of 1mL and loaded in the measurement chamber with a flow rate of 50 µL/min. Flow mode measurements were obtained recording 3 videos of 60 s for each measurement. The NanoSight NS300 software was used to analyze the sample (10-100 particles/frame). Scanning Electron Microscopy of CSMs. Scanning electron microscopy (SEM) analysis allowed characterizing the morphology of the CSMs. The sample solution was stained by using a 2% Uranyl acetate solution. All SEM images in this section were obtained using a scanning electron microscope ZEISS FESEM ULTRA Plus. The CSMs were deposited from a diluted solution onto a 3-4 nm thick film of amorphous carbon supported by a 400 µm mesh copper grid (Ted Pella Inc., #01822-F). One drop (2 μL) of CSMs suspension was deposited onto the grid, and the solvent was evaporated at RT. 2.1.3. Biological Characterization Bradford Coomassie Assay. The Bradford assay (PierceTM Coomassie Plus Assay Kit; ThermoFisher #23236) was used to determine the protein concentrations. Following the manufacturer’s instructions, 150 µL of diluted standard protein samples (standard bovine serum albumin, BSA stock 2 mg·mL-1) in a working range of 1-25 µg·mL-1 and samples to be analyzed were mixed with 150 µL of Coomassie reagent in a 96 well microplate. The absorbance at 570 nm was read using a microplate reader (TECAN, Infinite® 200 PRO) and the protein content was calculated by the standard curve in Figure 2.1.2. 60 Figure 2.1.2. Bradford assay standard curve of BSA concentration versus absorbance (A) (a.u.) at 570 nm. SDS-PAGE electrophoresis and Mass Spectrometry (MS) analysis. The protein content on the CSMs samples was analysed by Sodium Dodecyl Sulphate-PolyAcrylamide gel electrophoresis (SDS-PAGE) electrophoresis and electrospray liquid chromatography (nLC-MS/MS). SDS-PAGE was used to separate the proteins contained in the CSMs. The protein content was analysed by 10% SDS-PAGE using an ENDUROTM VE10 vertical gel electrophoresis system from Labnet. The CSMs sample was treated with a solution of Triton 0.1% for 10 minutes and then denatured by incubating the samples for 5 min at 100 ºC in 4 X loading buffer (5 mL of 0.5 M Tris-HCl pH 6.8, 1.2 g of SDS, 4 mL of glycerol, 0.03 g of bromophenol blue and 40 mM dithiothreitol, DTT). A 10% Tris-Glycine SDS-PAGE running gel was precast before each experiment by mixing 3.6 mL of MilliQ water, 3.7 mL of 1.5 M Tris – HCl buffer pH 8.9 and 0.1 mL of 10% SDS with 2.5 mL of acrylamide and 5 μL of N,N,N′,N′-Tetramethylethylenediamine (TEMED). As initiator, 50 μL of ammonium persulfate (APS) 10% was added. Stacking gel (5%) was made up with 0.5 mL acrylamide, 1.6 mL 0.4 M Tris-HCL buffer pH 6.8, 50 µL 10% SDS, 2.8 mL MilliQ water, 50 µL APS 10% and 5 µL TEMED. The electrophoresis was run under constant voltage of 150 V for 45 min and the gel was stained with Coomassie blue visible staining. Afterwards, MS analysis was performed to identify the proteins recovered from the CSMs samples. The gel section containing the protein band was removed using a sterile scalpel and transferred to a clean 0.5 mL sample tube. The gel sections were trypsin-digested in gel. The samples were analysed by electrospray liquid chromatography (nLC-MS/MS, Bruker Amazon ETD). 61 Flow cytometry (FC). FC measurements were performed to detect the lipid presence of CSMs and to characterize the fluorescence signal from the cargo encapsulation and any specific surface markers at the CSMs surface composition.83 The lower size-detection limit for light scattering of conventional flow cytometers is typically in the order of 200-500 nm. We can distinguish the signal of the CSMs from the background by fluorescence, forward and the side light scattering detector. The forward and side scattering and fluorescence intensity of the CSMs samples were measured with a Guava® easyCyte BG HT flow cytometer (Millipore®), using a blue laser emitting at 488 nm and a green laser emitting at 532 nm as excitation sources. Background measurements in PBS buffer were performed before each measurement series and the corresponding threshold applied to remove the PBS signal. Forward (FSC) and side scattering (SSC) signals were recorded to gather information of the CSMs concentration and dispersion. Concentrations between 1105 and 1108 CSMs·µL-1 for CSMs and functionalized CSMs (CSMs@GNR, CSMs@Cargo and CSMs@GNRs@Cargo) were analyzed by using a flow rate of 0.12 µL·s-1. Fluorescence and SSC signals were recorded to gather information of the fluorescently labeled encapsulated cargo and to evaluate the presence of membrane components (proteins and lipids) on the CSMs by using different fluorescence reporters: a lipid staining dye, CellMaskTM Deep Red Lipid staining, and the E-Cadherin/ECAD/Cadherin-1/CD324 Protein human (Sigma-Aldrich™ #SRP6426), an Alexa-488 labeled monoclonal anticadherin antibody that recognizes human CD324. 20 µL of CellMaskTM 1X and 2 µL of antibody (1gL-1) were independently added to the CSMs solution and stirred for 15 min at RT. The fluorescence signal was measured from the corresponding channel of the CellMaskTM (620/50 nm channel) and the Alexa-488anti-cadherin antibody (525/30 nm channel). 2.2. Functionalization and modification of CSMs 2.2.1. Functionalization with fluorescent phospholipid. Fluorescently labeled CSMs were produced using fluorescent phospholipids that can be intercalated within the bilayer. In particular, 1,2-Dioleoyl-snglycero-3-phosphoethanolamine labeled with Atto488 (DOPE-Atto488, Sigma-Aldrich™ #67335) or Atto647N (DOPE-Atto647N, Sigma-Aldrich™ #42247) were added to the lipid bilayer of the CSM. 1mL of obtained CSMs dispersed in PBS 1X were mixed with the 2µL of 1 mg·mL-1 of DOPEAtto488 (dissolved in Methanol/Dichloromethane) or DOPE-Atto647N 62 (dissolved in Dichloromethane) and sonicated for 5 min. The resulting CSMs were extruded 10 cycles using an Avanti® Mini extruder (with a 800 nm polycarbonate membrane) (Figure 2.2.1) and purified by dialysis, size exclusion chromatography or ultrafiltration (see section 2.3). After the CSMs modification, size distribution of DOPE-functionalized CSMs was analyzed by DLS and NTA, and the fluorescence signal was determined by a microplate reader (TECAN, Infinite® 200 PRO). Figure 2.2.1. a) Scheme showing the Labeling of CSMs with fluorescent phospholipids (DOPE-Atto488) and b) the experimental process to obtain the CSMs@DOPE-Atto488. 2.2.2. Functionalization with other lipids (Fusogenic CSMs) The fusogenic CSMs (FCSMs) were prepared using a lipids mixture of neutral lipid as 18:1 (Δ9-Cis) PE (DOPE) (Avanti #850725P), positively charged lipid as 18:1 TAP (1,2-dioleoyl-3-trimethylammonium-propane, DOTAP) (Avanti® Polar Lipids #890890) and aromatic/fluorescent lipid as BODIPY™ FL DHPE (N-(4,4-Difluoro-5,7-Dimethyl-4-Bora-3a,4a-Diazas-Indacene-3-Propionyl)-1,2-Dihexadecanoyl-sn-Glycero-3Phosphoethanolamine, Triethylammonium Salt) (Invitrogen™ #D3800). a) b) 63 According to Rejhana Kolašinac et al.,187 the mixture of DOPE/DOTAP/BODIPY-FL-DHPE was used at a ratio of 1/1/0.01–0.1 mol/mol (Table 2.2.1). Precisely, 1mL of obtained CSMs dispersed in 20 mM of HEPES buffer was mixed with 20 µL of DOPE (1 mg·mL-1), 9 µL DOTAP (1 mg·mL-1) and 1.43 µL DHPE (BODIPY-FL-DHPE) (1 mg·mL1). The solution was sonicated for 10 min and extruded 10 cycles using an Avanti® Mini extruder with a 800 nm polycarbonate membrane and purified by ultrafiltration using Amicon® Ultra-0.5 Centrifugal Filter Devices (cutoff of 100 kDa, Millipore Sigma-Aldrich™ #UFC510096) and washing the solution with 20 mM HEPES at least 3 times (see section 2.3) (Figure 2.2.2). The solution of FCSMs was concentrated/resuspended in 100 µL 20 mM HEPES volume. To encapsulate small or large molecules inside the FCSMs, the corresponding volume of the selected cargo (Table 2.2.1) was added to the FCSMs solution before the sonication step. The samples were purified from the excess of free molecule, by size exclusion chromatography (see section 2.3) Table 2.2.1. The different lipids used to prepare the FCSMs. For 1 mL of CSMs (corresponding to 1011 NCs·mL-1) the lipid ratio indicated in the table was used. * Molar ratio lipids:DOTAP. Sample C. (solvent) Mass Vol Ratio* DOPE-Atto488 1 mg·mL-1 (MeOH/CH2Cl2) 2-20 µg 20 µL 1 DOPE-Atto647N 1 mg·mL-1 (CH2Cl2) 2-20 µg 20 µL 1 BODIPY-FL-DHPE 1 mg·mL-1 (EtOH) 1.43 µg 1.43 µL 0.1 DOTAP 10 mg·mL-1 (CHCl3) 9 µg 0.9 µL 1 64 Figure 2.2.2. a) Scheme showing CSMs functionalization with lipids and b) the experimental process to obtain the FCSMs. 2.2.3. Functionalization with cargo. To functionalize the CSMs and the FCSMs with different cargo, such as Alexa545-PEG(4), seminaphtorhodafluor labeled Dextran-70 kDa (DS), fluorescein isothiocyanate labeled Phalloidin (PHA) and HOECHST-33258 (HOE), or biomolecules, such as β Tubulin monoclonal Ab labeled with Alexa Fluor 555 BT7R (Alexa555 BT7R Ab), the same procedure described above was followed. 1 mL of CSMs dispersed in PBS 1X or FCSMs dispersed in HEPES were mixed with the corresponding volume of the selected cargo (see Table 2.2.). a) b) 71 Dialysis. The reaction excess of Alexa545-PEG(4) or Alexa555 BT7R Ab was separated from the functionalized CSMs using the dialysis purification method. Float-A-Lyzer G2 with the cut-off of 1,000 kDa was used for overnight purification. Ultrafiltration. Molecules with molecular weight below than 100 KDa such as DS, PHA, DOTAP and DOPE, can be easily separated from the functionalized CSMs with Amicon® Ultra-0.5 Centrifugal Filter Devices with the appropriate cut-off (3-5-100 kDa) washing the solution with PBS 1X at least 3 times. The limitation of using a filter centrifugation step arises from the unspecific interactions between some molecules and the filter materials. To reduce this effect and the sample loss, the Amicon® filters were washed with 0.2% of BSA in PBS two times before to use them. Purification of Solid NPs by centrifugation. Due to differences in density, the empty CSMs can be separated from the CSMs-coated PSNPs or CSMscoated GNRs by normal centrifugation. In the PSNPs case, the excess of empty CSMs was removed by centrifugation at 12,000 g for 30 min and the collected CSMs@PSNPs were re-suspended in PBS 1X. Regarding the purification of GNRs-tagged CSMs, the excess of GNRs not incorporated into the nanosystem was washed off by centrifugation at 10,000 g for 10 min. At these conditions, the CSMs@GNRs were collected thanks to their increased weight while the free GNRs remained in the supernatant and were easily removed with it. 2.4. Characterization of functionalized CSMs Quantification by Fluorescence Measurements. The amount of cargo molecules loaded into the CSMs was quantified measuring the fluorescence signal by using a Microplate reader (TECAN, Infinite® 200 PRO) equipped with monochromator-based optics and wavelength selection between 280 and 850 nm. In the Table 2.4.1, the λexc/λem ranges used for the different fluorescent molecule are reported. The cargo concentrations in the solution ware determined by interpolation of the measured fluorescence intensity from a previously constructed analytical calibration curve (see Table 2.4.1 and Figure 2.4.1). 72 Table 2.4.1. The concentration of encapsulated cargo calculated by measuring the fluorescence intensity of CSMs@Cargo samples. Cargo λexc/λem (nm) Equation R Squared C.Cargo (µg·ml-1) C. (CSMs·ml-1) Cargo/CSM DS 500/594 Y = 173.2·X + 2712 0.9928 2.58 ± 0.3 2·1011 110 ± 14 HOE 361/497 Y = 1133·X + 5490 0.9358 2.18 ± 0.4 2·1011 12310 ± 1462 Alexa555 BT7R Ab 560/610 Y = 11222·X + 630.6 0.9997 0.15 ± 0.1 1·1011 60 ± 22 Alexa545 -PEG(4) 560/610 Y = 201637·X + 1450 0.9937 0.28 ± 0.1 7·109 24353 ± 8777 PHA 488/520 Y = 4290·X + 974.1 0.9945 1.79 ± 0.9 2·1011 4311 ± 1183 73 Figure 2.4.1. Calibration curves of different cargos in PBS 1X or HEPES. These calibrations curves were used to quantify the concentration or number of the cargo encapsulated into CSMs. Colloidal stability. The functionalized CSMs were characterized with DLS and NTA analysis (see section 2.4) in order to study the colloidal properties of the CSMs after the functionalization reaction and the purification step. Quantification of encapsulated GNRs by ICP-MS. Once the CSMs@GNRs were obtained and purified following the protocol above, 2.40·109 CSMs were used to quantify the GNRs and, thus, the number of incorporated 74 particles for each nanostructure (Table 2.4.2). Concurrently, also 0.05 mL of 10 nM positively charged PEGylated-GNRs was analyzed in order to obtain a more precise quantification. Both samples were digested overnight in aqua regia (200 µL). Then, they were diluited with 2 mL of HCl 2% v/v for ICPMS analysis. The ICP-MS analysis was performed in an Agilent 7700x inductively coupled plasma mass spectrometer. Table 2.4.2. ICP-MS analysis of total GNRs added to the CSMs and purified CSMs@GNRs sample. CSMs@GNRs g Au·L-1 mol·L-1 GNRs·L-1 GNR CSMs GNRs·CSM-1 AV 0.0036 1.48·1010 8.89·1013 8.89·1010 2.40·109 37.06 SD 0.0013 5.28·1011 3.18·1013 3.18·1010 13.25 GNRs g Au·L-1 Mol·L-1 GNRs·L-1 GNR AV 0.0180 7.32·10-10 4.41·1014 4.41·1011 SD 0.0040 1.63·10-10 9.82·1013 9.82·1010 2.5. In vitro studies 2.5.1. Cell culture HeLa, A549, RAW and HEK cell lines were cultured in cDMEM. Cells were maintained under humid conditions at 37 °C and 5% of CO2. Cells were passaged after cleaning DPBS 1X with 0.25% Trypsin-EDTA 1X when the culture reached confluency. 2.5.2. 3D Cell Culture The HeLa and A549 cell lines were used to obtain the 3D spheroids. 100 µL of suspension of 50,000 cells·mL-1 were transferred to a 96 multi-well plate previously treated with agarose as follows: a 1-1.5% agarose solution was prepared in filtered PBS 1X and heated up to 100 °C; 50 µL of agarose solution were used to cover the bottom of the wells. The plate was let to cool at RT in sterile conditions without the lid for at least 30 min before seeding the cells. The 3D spheroids were grown at 37 °C in a humidified 5% CO2 75 atmosphere, with 100 µL of fresh complete media gently added every day, for 3 days in order to obtain spheroids with a  0.5 µm diameter. On the fourth day, the spheroids were washed with PBS 1X twice, carefully avoiding displacing them from the wells. The spheroids were incubated with a 200-300 µL solution of CSMs. After 3, 6 and 24 h, the CSMs solution was removed, and the spheroids were washed twice with PBS 1X. LIVE/DEAD Assay. Each spheroid was individually transferred into the 96 multi well plate without agarose. Then, they were harvested after trypsinization for 2 min with 0.05 mL of 0.25% Trypsin–EDTA. 0.2 mL of cDMEM was added to recover the cells, which were transferred to 1.5 mL sterile tubes. The cells were collected after centrifugation at 500 g for 5 min. The cell pellets were resuspended in 0.2 mL of PBS 1X at 0.1 µM calceinAM (Invitrogen, C1430, C3099, C3100) and 3 µM Propidium Iodide (Sigma-Aldrich™ #25535-16-4). The cells were incubated for 15-20 min at room temperature in the dark. To have a negative control, some of the spheroids were treated with 0.2% Triton solution for 15-20 min. A suspension of 100,000 cells·mL-1 was used as alive cell control. After the incubation period, the stained cells were analyzed by flow cytometry using a 488 nm excitation and measuring green fluorescence emission of calcein (i.e., 530/30 bandpass) and red fluorescence emission of the propidium iodide (i.e., 610/20 bandpass). Confocal Microscopy Assay. Each spheroid was individually transferred in a µ-Slide Angiogenesis plate (81506, IBIDI) and added 40 µL of cDMEM without phenol red. To obtain a more accurate view of the spheroids, CellMask™ Orange, Deep Red, and Green plasma membrane stains (ThermoFisher™) were used. The CellMask™ plasma membrane stain (1,000X concentration) was used at 1X in cDMEM, following the Thermofisher’s protocol. The spheroids were incubated with the CellMask™ 1X for 1-2 min at 37 °C. Thus, the staining solution was removed and the spheroids were washed 3 times with PBS 1X. Fresh cDMEM without phenol red was added. The spheroids were observed in transmitted light and fluorescence microscopy using an Andor Dragonfly spinning disk confocal system with the 20X objective (MRH07241, Nikon). Z-scans consisted of  300 different stack images in the Z-axis. 3D reconstructions were done with  300 stack images (total thickness  150 µm, 76 step thickness  0.5 µm). ImageJ was used to analyze the 2D images. The deconvolution of the Z-scans was done with the Fusion software. Imaris software (Oxford Instruments) was used for the 3D reconstruction. 2.5.3. Flow Cytometric Studies 2.5.3.1. CSMs Uptake To expose the cells to CSMs, the HeLa and A549 cells were seeded in 24well plate at a density of 50,000 cells per well while the MRC-5, RAW and HEK cells were seeded at a density of 60,000 cells per well in 0.5 mL of cDMEM. After 24 h, cells were washed with PBS 1X and the medium was replaced with the freshly prepared CSMs dispersions of interest in cDMEM. Experiments were performed by exposing the cells to CSM dispersions at 2·105 CSMs per cell at 37 °C and 5% CO2 for 10 min, 1, 3 or 6 hours. After CSMs exposure, cells were washed with 1 mL PBS per well. Then, cells were harvested after trypsinization for 2 min with 0.075 mL 0.25% Trypsin– EDTA. 0.15 mL of PBS or cDMEM was added to each well to recover the cells. Cells were pelleted by centrifugation at 1,500 rpm for 3 min, then resuspended in 200 µL of cell media or PBS. Cell fluorescence intensity was measured using a Guava Millipore flow cytometer equipped with a 488 nm blue laser and a 532 nm green laser coupled with 525/30 nm, 583/26 nm, 620/50 and 695/50 filters, depending on the CSMs fluorescence. Results are reported as the median of cell fluorescence intensity. 2.5.3.2. Endocytosis inhibition For inhibition studies, cells were seeded in 48-well plates in 0.3 mL of cDMEM at a density of 25,000 cells per well for the HeLa and A549 cells and of 30,000 cells per well for the MRC-5, RAW and HEK cells. After 24 h, the medium was replaced with the freshly prepared inhibitor solutions of interest in cDMEM at the concentrations reported in Table 2.5.1. 77 Table 2.5.1. Inhibitors used to block the cell endocytosis. Inhibitor C. Pre-treatment (min) (mM) (µg·mL-1) Nystatin (NYS) 0.01 10 30 Dynasore (DYN) 0.08 25.78 30 Chloroquine (CQ) 0.1 32 30 Chlorpromazine (CP) 0.03 10 30 Methyl-β-cyclodextrin (MBCD) 5 6250 20 After 30 min of treatment, the inhibitor solutions were removed and replaced with the same solutions enriched with fluorescent CSMs at a concentration of 2·1011 CSMs·mL-1 at 37 °C and 5% CO2 for 10 min, 1, 3 or 6 h. Then the cells were washed with 1 mL PBS per well and harvested after trypsinization for 2 min with 0.075 mL 0.25% Trypsin–EDTA. 0.15 mL of PBS or cDMEM was added to each well to recover the cells. Cells were pelleted by centrifugation at 1,500 rpm for 3 min then redispersed in 200 µL of cell media or PBS. The CSMs uptake under the different inhibitor conditions was evaluated by cell fluorescence intensity that was measured using a Guava Millipore flow cytometer equipped with a 488 nm blue laser and a 532 nm green laser coupled with 525/30 nm, 583/26 nm, 620/50 and 695/50 filters, depending on the CSMs fluorescence. 2.5.3.3. Homotypic targeting The homotypic targeting experiments were performed using CSMs obtained from different cell lines (HeLa, A549, HEK, see section 2.1) equivalently labeled with DOPE-Atto488 or DOPE-Atto647N. HeLa and A549 cells were seeded in a 48-well plate at a density of 25,000 cells per well and the MRC5, RAW and HEK cells were seeded at a density of 30,000 cells per well in 0.3 mL of cDMEM. After 24 h, cells were washed with PBS and the medium was replaced with the freshly prepared CSMs dispersions of interest (HeLa derived CSMs@DOPE, A549 derived CSMs@DOPE and HEK derived CSMs@DOPE) in cDMEM. Experiments were performed by exposing the cells to CSMs dispersions at 2·1011 CSMs·mL-1 at 37 °C and 78 5% CO2 for 10 min, 1, 3 or 6 h. After CSMs exposure cells then were washed with 1 mL PBS per well. Then the cells were harvested after trypsinization for 2 min with 0.075 mL 0.25% Trypsin–EDTA. 0.15 mL of PBS or cDMEM was added to each well to recover the cells. Cells were pelleted by centrifugation at 1,500 rpm for 3 min and then redispersed in 200 µL of cell media or PBS. Cell fluorescence intensity was measured using a Guava Millipore flow cytometer equipped with a 488 nm blue laser and a 532 nm green laser coupled with 525/30 nm, 583/26 nm, 620/50 and 695/50 filters, depending on the CSMs fluorescence. Results are reported as the median of cell fluorescence intensity. 2.5.4. Confocal Microscopy In order to perform all the confocal imaging experiments with living cells, the HeLa and A549 cell lines were seeded in IBIDI dishes at the concentrations reported in Table 2.5.2. Regardless of the cell line and the plate type, the cells were let to grow in cDMEM at 37 °C and 5% CO2 for 24 hours to reach 70% of confluence. Table 2.5.2. Different seeding condition in confocal microscopy experiments of HeLa, A549 and MRC-5 cells depending on the IBIDI dishes. IBIDI dish HeLa A549 µ-Slide 8 Well (#80826) Cells 25,000 25,000 Vol. (mL) 0.3 0.3 µ-Slide VI 0.4 (#80606) Cells 15,000 15,000 Vol. (mL) 0.03 0.03 µ-Dish 35 mm (#80136) Cells 50,000 50,000 Vol. (mL) 0.5 0.5 µ-Slide 18 Well (#81816) Cells 10,000 10,000 Vol. (mL) 0.1 0.1 The CSMs samples were exposed to the cells at different time treatments (10 min in PBS 1X, 1, 3 or 6 h in cDMEM) at a concentration of 2·1011 CSMs·mL-1 at 37 °C and 5% CO2 condition. Then the cells were washed 3 79 times with fresh PBS 1X in order to remove non-associated CSMs with cells and added with HEPES supplemented media (DMEM without phenol red, 4.5 g·L-1 D-glucose and L-glutamine, 25 mM HEPES and without pyruvate, Gibco, #21063-029). To reach a more accurate view of the cells, different intracellular compartments were stained. CellMask™ Orange, Deep Red, Green plasma membrane stains and others ThermoFisher™ products provide excellent and rapid compartment staining in live cells for 30–90 minutes depending on the cell type and experimental conditions (Table 2.5.3). For the lysosome staining, a LysoTrackerTM Blue DND-22 (ThermoFisher, #L7525) dilution of 50 nM was prepared in cDMEM. 0.1 mL of this staining solution was gently added to the cells and incubated 30 min at 37 °C in dark conditions. For cell membrane staining, a CellMaskTM Deep Red (ThermoFisher, #C10046) solution (1 µL in 1 mL of cDMEM) was added to the cells 3 min at 37 °C in dark conditions. Subsequently, the cells were washed three times with PBS and HEPES supplemented media. To perform the confocal imaging experiments with fixed cells, after the specific cell treatment (CSMs incubations and/or intracellular compartment staining) the cells were washed three times with PBS and incubated with 3.6% Formaldehyde solution (Sigma-Aldrich™ #F8775) in PBS solution for 20-25 min at RT. Then, the formaldehyde solution was removed, and the cells were washed with PBS before adding few drops of IBIDI Mounting Medium (#50001 IBIDI). 80 Table 2.5.3. Thermo Fisher products used for the intracellular compartment staining and the experimental conditions used. Staining Stock solution Working solution Incubation conditions Em/exc (nm) CellMask™ Green plasma Membrane stain Plasma membrane 5 mg·mL-1 (1,000X) 1X in cDMEM 1 min 37 °C 490/525 CellMask™ Orange plasma Membrane stain Plasma membrane 5 mg·mL-1 (1,000X) 1X in cDMEM 1 min 37 °C 556/572 CellMask™ Deep red Plasma Membrane stain Plasma membrane 5 mg·mL-1 (1,000X) 1X in cDMEM 1 min 37 °C 660/680 LysoTracker® Blue DND-22 Lysosomes 1 mM (DMSO) 0.1 µM in cDMEM 30-45 min 37 °C 373/422 MitoTracker® Green FM Mitochondria 1 mM (DMSO) 0.1 µM in cDMEM 15-20 min 37 °C 410/516 Confocal images of living cells were captured on an Andor Dragonfly spinning disk confocal system mounted on a Nikon TiE microscope equipped with a Zyla 4.2 PLUS camera (Andor, Oxford Instruments) and an OKO-lab incubator to keep cells at 37 °C during the whole experiment. The samples were excited with four different lasers (405, 488, 561 and 637 nm lasers) and the emitted fluorescence was collected by the filter wheel (450 (50) nm, 525 (50) nm, 620 (50) nm and 725 (40) nm) with appropriate combinations of them (Table 2.5.4). Images were taken with different magnification objectives (20X, 60X, 100X). All the images were processed with ImageJ®. 87 Figure 3.1.1. Schematic representation of the workflow for CSM preparation and evaluation of the interaction with 2D and 3D cell culture models. 3.1.1. Design and characterization of CSMs The cell membranes, like derived bio-vesicles, are asymmetric phospholipidbased membranes embedded with active surface compounds that are essential for their bio-functions.190 Human cell cultures derived from cervix epithelial carcinoma (HeLa cells) were first selected as a model cancer cells and the cell membrane fragments synthesis was carried out following the method previously reported in the literature by Zhu et al.185 and described in the Materials & Methods, section 2.1. First, a number of cells ranging between 106 and 107 were harvested from culture dishes and collected by centrifugation. CSMs were produced by combination of two physical disruption methods capable of breaking down cell membranes using different external forces such as heat and pressure. The complete cell lysis was carried out by (i) an osmotic shock produced with a hypotonic buffer and, (ii) thermal lysis conducted by repeated freezing-thawing cycles. The hypotonic solution provides the osmotic pressure required for cell swelling. The concentration of salt surrounding the cell is rapidly decreased, so that the cell membrane becomes 88 permeable to water due to the osmotic effect. The combination of the osmotic pressure and the thermal shock effect caused by the freezingthawing cycles induce transient openings in the cell membrane to empty the cell from its components. To minimize the possible denaturation of the membrane-oriented proteins, plasma membrane extraction was as gentle as possible, while protease inhibitors were added to the solution to maintain the bio-activity of membrane.185,191 Soluble proteins, intracellular biomacromolecules, and nucleus were removed via centrifugation, and the purified membrane fragments were subsequently extruded through polycarbonate membranes with nanosized pores to obtain self-assembled CSMs with size ranging between 100 and 400 nm. By selecting specific pore size of the membrane used during the extrusion process, it is possible modulate the size distribution of the final CSMs. This step was found to be a critical point to obtain highly monodisperse CSMs. The number of extrusion cycles required to obtained monodispersed CSMs was optimized by DLS analysis. As reported in Table 3.1.1 and showed in Figure 3.1.2, increasing the number of extrusion steps narrowed the size distribution of the CSMs synthesized (Figure 3.1.2a) leading to a decrease of the Polydispersity Index (PDI) (Figure 3.1.2b). A minimum of 7 cycles of extrusion was required to obtain a monodisperse CSMs sample (Figure 3.1.2b and Table 3.1.1). Therefore, 10 cycles of extrusion were applied for the preparation of the samples. EM was used to confirm the morphology of the CSMs that show similar sizes as the DLS measurements (Figure 3.1.2c). 89 Figure 3.1.2. DLS characterization of CSMs: a) mean hydrodynamic diameter for the DLS distributions in intensity (dH,I), and number (dH,N) of CSMs before (black line) and after (red line) the extrusion steps; b) variation of the CSM hydrodynamic diameter and the PDI after different number of extrusion steps; c) SEM micrographs of CSMs derived from HeLa cells after their staining with uranyl acetate; scale bars: 200 nm. a) b) c) 90 Table 3.1.1. DLS measurements of CSMs after 1 to 7 extrusion steps. [1] Mean hydrodynamic diameter derived from the DLS distributions. dH,I and dH,N refer to the mean average hydrodynamic diameter from the intensity and number DLS distributions. Standard deviation values were calculated from five measurements. Following the same protocol, three more cell lines were used as source of bioactive membranes: the adenocarcinomic human alveolar basal epithelial cells (A549 cells), the healthy embryonic adrenal cell line (HEK cells) and macrophages from Abelson murine leukemia virus-induced tumor (RAW cells). DLS analysis were used to investigate the hydrodynamic size and the colloidal stability of the CSMs obtained from different cell lines. The results revealed highly monodisperse size distribution (PDI values below 0.2) and nearly constant hydrodynamic diameters for all CSM samples obtained from different cell types (around 200 nm in diameter) for both intensity (dH,I) and number distributions (dH,N) (Table 3.1.2 and Figure 3.1.3). These results shown that only one size distribution is present in the CSM sample solution and there were no agglomerates or aggregates. N° extrusion step PDI[1] dH,I (nm)[1] dH,N (nm)[1] Pk 1 dH,I (nm)[1] Pk 2 dH,I (nm)[1] 0 0.67 ± 0.04 1767.3 ± 71.7 117.7 ± 27.5 2731.2 ± 200.3 118.5 ± 26.7 1 0.45 ± 0.02 371.9 ± 149.8 218.2 ± 21.1 233.8 ± 21.1 5399 ± 227.7 2 0.40 ± 0.02 226.2 ± 27.5 188.2 ± 23.2 198.7 ± 26.6 5452.7 ± 186 3 0.38 ± 0.01 231.7 ± 53.0 218.7 ± 50.6 231.7 ± 53.0 5452.7 ± 182.5 4 0.33 ± 0.01 284.1 ± 50.4 247.7 ± 50.0 267.2 ± 54.5 5 0.28 ± 0.01 234.3 ± 18.5 216.0 ± 9.5 234.4 ± 18.5 6 0.29 ± 0.01 328.2 ± 68.2 220.5 ± 8.2 328.2 ± 68.2 7 0.17 ± 0.02 200.5 ± 5 168.5 ± 6.0 200.5 ± 5.0 91 Table 3.1.2. DLS measurements of CSMs obtained from different cell lines: HeLa cells, A549 cells, HEK cells and RAW cells. [1] Mean hydrodynamic diameter derived from the DLS distributions: dH,I and dH,N refer to the mean average hydrodynamic diameter from the intensity and number DLS distributions. PDI refers to polydispersity index. Standard deviation values were calculated from five measurements. Sample dH,I (nm)[1] dH,N (nm)[1] PDI[1] CSMs@HeLa 200.5 ± 5.0 168.5 ± 6.0 0.16 ± 0.008 CSMs@A549 279.0 ± 25.0 161.1 ± 21.2 0.17 ± 0.011 CSMs@HEK 345.3 ± 41.1 211.3 ± 16.1 0.20 ± 0.019 CSMs@RAW 245.7 ± 19.6 221.4 ± 15.5 0.20 ± 0.015 Figure 3.1.3. DLS characterization of the size distribution of CSMs: a) mean hydrodynamic diameter for the DLS distributions in intensity (dH,I), and b) number (dH,N) of CSMs from different cell-type origin (HEK cells, A549 cells, HeLa cells and RAW cells). NTA was used to further characterize the colloidal properties of the CSMs and determine their concentration in solution. The results NTA showed similar CSMs size distribution as the data obtained from the DLS analysis (Figure 3.1.4) and allowed determining the CSMs concentration that are reported in Table 3.1.3. This parameter is crucial to establish and modulate the doses of administered CSMs to the cells in subsequent in vitro studies, and the cargo loading of the different probes used. a) b) 92 Table 3.1.3. Mean diameter derived from the NTA distributions. Standard deviation values were calculated from three measurements of CSM stock samples. Sample C. (CSMs·mL-1 ·1011) Size (nm) CSMs@HeLa 7.57 ± 0.4 155.33 ± 7.1 CSMs@A549 3.93 ± 0.25 167.1 ± 10 CSMs@HEK 4.07 ± 0.11 152.67 ± 11 CSMs@RAW 2.57 ± 0.64 193.1 ± 3 Figure 3.1.4. NTA size distribution of diluted CSMs in water. Representation of the number distribution of the hydrodynamic diameter of individual CSMs. CSMs derived from HeLa (black line), A549 (red line), HEK (blue line) and RAW cells (green line). Nanomaterials for biological applications are not expected to sedimentate or aggregate. However, aggregation after changes in pH or ionic force of the experiment media can frequently occur.192 Thus, it is crucial to analyze the nanocomposite behavior under different conditions. Here, PBS, 2-(N-morpholino)ethanesulfonic acid (MES) and phagolysosomal simulant fluid (PSF) buffers were used to investigate the CSMs stability at different pH conditions (PBS pH 7.4, MES pH 6.0 and 5.5, PSF pH 4.5). Moreover, cDMEM buffer (pH 7.6) was used as complex media to simulate the proteinenriched bioenvironment of cells. The hydrodynamic diameter of the CSMs after dispersion in the corresponding buffer was measured over time (from 0 to 48 h) because the aggregation or stability loss can be a time-dependent process (see Figure 3.1.5 and Table 3.1.4). The results showed that the CSM size from the four different cell lines and therefore their colloidal stability 93 was maintained in physiological conditions (PBS buffer at pH 7.4) as well as in complex media such as cDMEM. As expected, the media proteins of biological media help to stabilize the CSMs dispersion by forming the protein corona and preventing the aggregation and sedimentation.193,194 On the other hand, when the pH decreases bellow 6, the colloidal stability of the CSMs was compromised causing the increase in the hydrodynamic diameter. Table 3.1.4. DLS measurements of the colloidal stability of CSMs under different pH values and time points. [1] Mean hydrodynamic diameter derived from the DLS distribution: Z-Average, and dH,I refers to the mean average hydrodynamic diameter from the intensity distribution. PDI refers to polydispersity index. Standard deviation values were calculated from three measurements. Buffer Z-Average dH,I (nm)[1] PDI CSMs@HeLa PBS pH 7.4 (0 h) 209.2 ± 16.6 185.79 ± 14.8 0.21 ± 0.09 PBS pH 7.4 (24 h) 253.1 ± 25.6 255 ± 9.6 0.2 ± 0.1 PBS pH 7.4 (48 h) 277.5 ± 8.7 296.3 ± 3.8 0.27 ± 0.06 MES pH 6 (0 h) 337.7 ± 139.4 195.7 ± 23 0.42 ± 0.04 MES pH 6 (24 h) 343.3 ± 73.9 320 ± 46.8 0.36 ± 0.18 MES pH 6 (48 h) 363.5 ± 65.4 339.3 ± 102.7 0.33 ± 0.20 MES pH 5.5 (0 h) 610.5 ± 149,4 2832.7 ± 20.4 0.48 ± 0.15 MES pH 5.5 (24 h) 668.2 ± 24.5 2542.3 ± 636 0.62 ± 0.25 MES pH 5.5 (48 h) 924.7 ± 276.4 956 ± 407.4 0.61 ± 0.34 PSF pH 4.5 (0 h) 751.6 ± 69.1 1973.7 ± 1350.6 0.61 ± 0.21 PSF pH 4.5 (24 h) 971.5 ± 246 1007.3 ± 717.7 0.67 ± 0.13 PSF pH 4.5 (48 h) 1074.2 ± 159.4 1110 ± 379.4 0.7 ± 0.07 cDMEM pH 7.4 (h) 194.7 ± 2.5 227.7 ± 22.7 0.31 ± 0.04 cDMEM pH 7.4 (24 h) 157.7± 7.8 182.33 ± 17.5 0.36 ± 0.07 cDMEM pH 7.4 (48 h) 208.5 ± 16.9 324.7 ± 33.2 0.50 ± 0.09 CSMs@A549 94 PBS pH 7.4 (0 h) 196.7 ± 0.6 224 ± 6.1 0.12 ± 0.03 PBS pH 7.4 (24 h) 205.7 ± 2.1 212.7 ± 10 0.21 ± 0.04 PBS pH 7.4 (48 h) 207 ± 2.7 229.5 ± 2.1 0.24 ± 0.02 MES pH 6 (0 h) 405.33 ± 87.9 303.7 ± 32 0.44 ± 0.06 MES pH 6 (24 h) 354.3 ± 72.9 246 ± 29.7 0.56 ± 0.1 MES pH 6 (48 h) 326.3 ± 36.2 259 ± 2.8 0.58 ± 0.06 MES pH 5.5 (0 h) 1704.7 ± 182.4 616 ± 394.6 0.86 ± 0.19 MES pH 5.5 (24 h) 1115 ± 100.5 1299 ± 17 0.73 ± 0.01 MES pH 5.5 (48 h) 945.7 ± 37.8 1176.3 ± 363.5 0.8 ± 0.06 PSF pH 4.5 (0 h) 1674.7 ± 225.9 1137.7 ± 903.3 0.84 ± 0.11 PSF pH 4.5 (24 h) 1123 ± 193.8 1992.3 ± 1246.1 0.63 ± 0.2 PSF pH 4.5 (48 h) 829.7 ± 120 3001.7 ± 116.8 0.9 ± 0.05 cDMEM pH 7.4 (0 h) 202 ± 1 348.3 ± 9.3 0.28 ± 0.01 cDMEM pH 7.4 (24 h) 219.5 ± 2.2 255 ± 23.8 0.27 ± 0.01 cDMEM pH 7.4 (48 h) 179.8 ± 8.36 234 ± 122.2 0.36 ± 0.12 CSMs@HEK PBS pH 7.4 (0 h) 255 ± 29.5 231.6 ± 3.1 0.32 ± 0.01 PBS pH 7.4 (24 h) 303.3 ± 12.1 243 ± 17.7 0.31 ± 0.05 PBS pH 7.4 (48 h) 311.3 ± 50 229 ± 26.5 0.31 ± 0.25 MES pH 6 (0 h) 244 ± 7.9 273.7 ± 40.45 0.29 ± 0.02 MES pH 6 (24 h) 273.7 ± 40.4 158.7 ± 21.55 1 MES pH 6 (48 h) 158.7 ± 21.6 244 ± 7.9 1 MES pH 5.5 (0 h) 681.3 ± 30.7 1601.3 ± 150 0.79 ± 0.02 MES pH 5.5 (24 h) 1343.7 ± 208.7 592 ± 176-5 0.9 ± 0.09 MES pH 5.5 (48 h) 1582.7 ± 100.3 272 ± 55 0.85 ± 0.08 95 PSF pH 4.5 (0 h) 981.73 ± 84.2 658.7 ± 70.8 0.65 ± 0.2 PSF pH 4.5 (24 h) 999.3 ± 234.7 1244 ± 438.2 0.71 ± 0.06 PSF pH 4.5 (48 h) 1532.3 ± 336.2 248 ± 144.2 0.97 ± 0.06 cDMEM pH 7.4 (0 h) 189 ± 24.25 382.2 ± 90.4 0.57 ± 0.12 cDMEM pH 7.4 (24 h) 208.7 ± 24.01 153.3 ± 20 0.22 ± 0.17 cDMEM pH 7.4 (48 h) 382.3 ± 90.4 189 ± 24.25 1 CSMs@RAW PBS pH 7.4 (0 h) 211 ± 6.6 224.7 ± 3.51 0.16 ± 0.02 PBS pH 7.4 (24 h) 229 ± 5.6 258.3 ± 66.4 0.36 ± 0.03 PBS pH 7.4 (48 h) 216.7 ± 0.6 255 ± 9.54 0.31 ± 0.04 MES pH 6 (0 h) 275 ± 8.2 321 ± 41.5 0.22 ± 0.04 MES pH 6 (24 h) 580 ± 17.7 557.5 ± 0.7 0.3 ± 0.05 MES pH 6 (48 h) 914 ± 73 740 ± 140 0.39 ± 0.1 MES pH 5.5 (0 h) 1965.3 ± 295.5 648.5 ± 63 0.32 ± 0.1 MES pH 5.5 (24 h) 1995 ± 35.5 848.9 ± 440 0.93 ± 0.06 MES pH 5.5 (48 h) 3405.7 ± 324.3 1542.3 ± 122.8 0.93 ± 0.1 PSF pH 4.5 (0 h) 1044.7 ± 88.9 1572.7 ± 246.4 0.42 ± 0.04 PSF pH 4.5 (24 h) 2113 ± 114.6 2747 ± 256 0.82 ± 0.02 PSF pH 4.5 (48 h) 2133.7 ± 1083.8 689.3 ± 268 0.71 ± 0.04 cDMEM pH 7.4 (0 h) 214.3 ± 41.2 214.3 ± 268.1 1 cDMEM pH 7.4 (24 h) 363 ± 55 363 ± 55 1 cDMEM pH 7.4 (48 h) 328 ± 51.5 328 ± 51.5 1 96 Figure 3.1.5. Hydrodynamic diameter of CSMs@HeLa, CSMs@A549, CSMs@HEK and CSMs@RAW under different conditions (0.1 M of PBS pH 7.4, 0.1 M of MES pH 6, 0.1 M of MES pH 5.5, PSF pH 4.7 and cDMEM pH 7.4). Mean average values from three independent measurements. Once the colloidal stability was determined, the surface components (lipids, proteins, etc.), the right orientation of surface markers (such as specific transmembrane proteins), and the biological activity inherited from the cell source of CSMs were studied. First, the protein content in the sample was quantified by a colorimetric assay. The Bradford assay which is based on the Coomassie dye-binding reaction showed that the total protein content determined in the final CSM sample is directly correlated to the total number of cells used for the CSM preparation (Figure 3.1.7). 103 The intercalation/functionalization was proved by fluorescence measurements and FC analysis, which showed a significant variation of the fluorescence signal of the CSMs@DOPE samples after purification compared to the non-functionalized CSMs (Figure 3.1.10). Because of the high labeling efficiency, the functionalization with the DOPE phospholipid showed to be an excellent staining strategy to label and analyze the CSM behavior in in vitro studies and follow their interaction with the cells. Figure 3.1.10. Characterization of DOPE-Atto488 labeled CSMs@HeLa. a) Scheme representation of CSM labelling by intercalating DOPE-Atto488. b) Fluorescence intensity of the maximum DOPE-Atto488 added and DOPE-Atto488 intercalated in the CSMs@DOPE-Atto48 after dialysis purification. c) FC characterization of CSMs@DOPEAtto488. 3.1.2. Preferential cancer cell recognition by CSMs 3.1.2.1. CSMs interactions with cells. The interaction of fluorescently labeled CSMs@HeLa with cells at different doses and different time points were studied by FC and confocal microscopy. FC results of CSMs@DOPE incubation in HeLa cells for different times (1, 3, 6 h and overnight) indicated that the highest internalization levels of CSMs@DOPE occurred after 12 h incubation, when cells showed the maximum level of fluorescence (Figure 3.1.11). However, even shorter time incubations were sufficient to achieve an efficient level of a) c) b) 104 CSMs@DOPE incorporation. The intracellular signal from labeled CSMs increased with longer incubation times and higher CSMs concentrations. In conclusion, the internalization of the CSMs is time and concentration dependent. Figure 3.1.11. Preliminary studies of cellular internalization of CSMs@HeLa. Green fluorescence intensity of CSMs@DOPE-Atto48 internalized in HeLa cells and determined by FC. Quantification of the MFI of the cellular uptake of CSMs@DOPE-Atto488 in HeLa cells after 1, 3, 6 h and overnight (O/N) incubations. Data expressed as mean ± s, n=3. Data expressed as mean ± s, n=3. Confocal microscopy images of internalized CSMs@DOPE in HeLa cells after 3 h and overnight were analyzed. The images confirmed that the intracellular signal from labeled CSMs increases with longer incubation times as the FC results. HeLa cells were also stained with CellMask (Figure 3.1.12). There was no significant colocalization of the cell membrane staining (red channel) with that of the CSMs (green channel). The images confirmed that after 3 h and O/N the CSMs were internalized via endocytosis and the CSMs remained intact inside the cells and accumulated preferentially in endosomes as happened for the majority of nanomaterials. Since the CSMs internalization after 3 h of incubation was efficiently revealed by fluorescence signal, this time point was selected for carrying out the next in vitro experiments. 1H 3H 6H O/N 0 10000 20000 30000 MFI Incubation time 105 Figure 3.1.12. Confocal images of HeLa cells incubated with CSMs@DOPE-Atto488 (green labeled) for 3 h and overnight incubations. Cell membranes were stained with CellMask Deep Red. Scale bars: 25 µm. 3.1.2.2. Homotypic targeting Specific targeting of cancer cells represents an important challenge for nanomedicine to promote tumor regression at the same time reducing the side effects of anti-cancer drugs. In this context, homotypic recognition is a new promising targeting strategy that exploits the self-recognition properties of cancer cells. Potentially, biomimetic CSMs system can exploit the 106 homotypic affinity between cancer cells from which they are extracted, mediated by specific membrane proteins, and it relies on the natural properties of cancer cells to develop strong contacts and adhesive interactions.135,185,195 In this section, the self-recognition capability between the CSMs and their source cell line was investigated. Three different cell sources were selected to produce the CSMs: A549 and HeLa cells as tumoral cell lines and HEK cells as model of healthy cell line. The CSMs obtained were equally fluorescently labeled with DOPE-Atto488 in order to assess the uptake in their original cell source by FC analysis. In Figure 3.1.13, internalization profiles of the three different CSMs after 1, 3 and 6 h incubations are reported in HeLa, A549, HEK cell lines and RAW264.7 macrophage-like cell line (RAW). As expected, in all the combinations the HeLa and the A549 cells exhibited higher internalization rates compared with HEK and RAW cells. This difference is justified by the altered cell metabolism provoked by cancer.196 The CSMs@HeLa were more efficiently internalized by HeLa cells than by the other cell lines at all incubation times, probably due to the homotypic adhesive interactions mediated by the adhesion molecules on the cell membrane.128 Likewise, this preferential recognition was also found in the case of CSMs@A549, confirming the homotypic affinity between the CSMs and their respectively cell source. Regarding the HEK cell line, as was said above, they showed a reduced internalization rate for all type of CSMs, CSMs@HEK included. Nevertheless, it should be noticed that the CSMs@HEK did not showed a preferential internalization either by the HeLa or A549 cells, exhibiting similar uptake rate in both cell lines. RAW cells showed a low internalization rate for all the CSMs tested, regardless the cell source from which they derived. This fact might be an indicative of the camouflaging ability of CSMs to escape from macrophages uptake. 107 Figure 3.1.13. a) Schematic representation of the CSMs@DOPE-Atto488 derived from HeLa, A549 and HEK cells. b) Internalization profiles of DOPE-Atto488-modified CSMs@HeLa, CSMs@A549 and CSMs@HEK. FC quantification of the cellular uptake of CSMs in HEK, A549, HeLa and RAW cells after 1 h, 3 h and 6 h incubations. MF values (Green fluorescence intensity) of DOPE-Atto488-modified CSMs internalized by cells expressed as mean ± s, n=3. 3.1.2.3. Hybrid CSMs To further investigate the homotypic affinity of the CSMs, a hybrid nanostructure was prepared by combining cell membrane fragments from two different cell lines. The capacity to introduce additional functionality, in particular the homotypic targeting capabilities described for cancer cell and cancer cell-derived NPs, was studied. The strategy involves transferring the surface properties of cancer cell-derived CSMs to non-tumoral cell-derived NPs. Hybrid CSMs were engineered by mixing cancer cell derived CSMs (CSMs@HeLa) with non-tumoral cell-derived CSMs (CSMs@HEK), previously labeled with two different dyes: CSMs@HeLa with DOPEAtto488 (green fluorescence) and CSMs@HEK with DOPE-Atto647N (red fluorescence). Then, the samples were mixed at different ratios (1:9, 1:3, 1:1, 3:1 and 9:1) and homogenated together by several extrusion steps, obtaining the hybrid CSMs composed of a green labeled fraction of HeLa membrane and a red labeled fraction of HEK membrane. The size distribution of the hybrid CSMs was studied by NTA analysis (Figure 3.1.14). It showed that the hydrodynamic diameter in maintained around 200-300 nm in all type of hybrid CSMs. 108 0 2×1024×1026×102 0 1×106 2×106 3×106 4×106 5×106 Size (nm) CSMsmL-1 10% / 90% CSMs 25% / 75% CSMs 50% / 50% CSMs 75% / 25% CSMs 90% / 10% CSMs CSMs@HeLa CSMs@HEK Figure 3.1.14. NTA hydrodynamic size distribution of hybrid CSMs obtained mixing CSMs@HeLa and CSMs@HEK at different percentage composition. Representation of the number distribution of the hydrodynamic diameter of individual hybrid CSMs. Initially, to determine the stability of hybrid CSMs synthesized (Figure 3.1.15a), confocal microscopy analysis of HeLa cells after CSM internalization were carried out. To define the experiment conditions, the 1:1 hybrid CSMs were used to incubate with HeLa cells. The images in Figure 3.1.15b show the complete colocalization of green and red fluorescence signal after 3 h of CSM incubation with cells. It suggested that at least the majority of the sample was composed by the hybrid CSMs (which are labeled with both green and red fluorescence reporters). Almost any spot of single fluorescence (green or red) is shown in the merged image (last panel, Figure 3.1.15b). Then, the cancer targeting properties of hybrid CSMs was evaluated by FC analysis. HeLa and HEK cells were incubated with the hybrid CSMs (1:1 ratio) for 3 h and the internalization rate was evaluated by the MFI signal (Figure 3.1.15c). The results from the FC analysis showed similar amount of binding to HeLa cells for the hybrid CSMs than CSMs@HeLa. This indicates that the incorporated surface markers derived from CSMs@HeLa to the hybrid CSMs can enhance their targeting capabilities. In addition, hybrid CSMs show a slightly high amount of binding to HEK cells than CSMs@HeLa. These results confirm that it is possible to fuse CSMs derived from different cells, and they suggest that the hybrid CSMs retain both functionalities derived from the fusion of the two different cell-derived CSMs. 109 Figure 3.1.15. Synthesis and cellular uptake studies of hybrid CSMs. a) Scheme of hybrid CSMs preparation by mixing the DOPE-Atto488 labeled CSMs@HeLa and DOPE-Atto647 labeled CSMs@HEK. b) Confocal images of HeLa cells incubated with hybrid CSMs for 3 h in green channel (left), red channel (middle) and both channels merged (right). Scale bars: 25 µm. c) Fluorescence signal (by FC, as MFI) from the HeLa and HEK cellular uptake (3 h incubation) of hybrid CSMs and CSMs@HeLa in the green channel. d) Cellular uptake studies of hybrid CSMs mixed at different ratios. Fluorescence signal (by FC, as MFI) from the HeLa cellular uptake (3 h incubation) in the green channel. Data expressed as mean ± s, n=3. 110 Finally, a comprehensive study by FC was performed. The uptake of hybrid CSMs with different percentages of CSMs@HeLa and CSMs@HEK (1:9, 1:3, 1:1, 3:1 and 9:1 CSMs@HeLa and CSMs@HEK ratio, respectively) was studied. FC analysis was performed with all the types of hybrid CSMs that were incubated in HeLa cells for 3 h (Figure 3.1.15d). The uptake trend showed that the presence of HEK membrane fractions affects the CSMs´ affinity to the cells. In HeLa cells the lowest internalization value corresponded to the higher percentages of HEK membrane fraction (100, 90 and 75 % of HEK). On the other hand, the uptake increased with the prevalence of HeLa membrane fraction (0, 10 and 25 % of HEK), which confirms the higher homotypic affinity of the latter to HeLa cells. The results suggest that a certain amount of HeLa derived membrane fragments is required to introduce the cancer targeting capabilities to CSMs@HEK samples. A ratio of 1:1 (50 %/50 %) is enough to design hybrid CSMs with the maximum affinity for HeLa cells. Although the mechanisms involved in homotypic recognition are not completely elucidated, their understanding could be an opportunity to develop a new generation of nanovectors, functionalized with a combination of most efficient molecules, in order to improve the targeting efficacy of the nanotherapeutics. 3.1.2.4. Cell transport pathway Endocytosis is the cell transport pathway by which the cell-derived nanovesicles are internalized by cells.197 However, studies about the exact mechanism of internalization remain elusive. Due to the nature of these biomimetic systems, the hypothesis that it could occur by the direct fusion of plasma membranes is not to rule out. In fact, there are various examples of cell-derived nanocarriers which inherit the fusion capabilities, such as viruses or exosomes derived nanocarriers.198,199 To gain insights into CSMs behaviour when they are in contact with cells, the internalization mechanism was studied in different cell lines. Firstly, confocal microscopy was used to qualitatively investigate the intracellular fate of CSMs@HeLa. Cells were incubated for 3 h with DOPEAtto488 labeled CSMs@HeLa and, finally, stained with CellMask Deep Red for the cellular membrane and with LysoTracker Blue for the lysosomes (Figure 3.1.16). The confocal images did not show any significant colocalization of the cell membrane staining (red channel) with that of the 111 CSMs@HeLa (green channel), excluding the possibility that CSMs fuse with the plasmatic membrane. On the other hand, a high degree of colocalization with the lysosomes (blue channel) is showed. This internalization profile suggested that the CSMs are incorporated by the cells following the endocytosis pathway, in agreement with the literature.200 Moreover, the CSMs’ signal appeared punctuated at the perinuclear region, which is a further evidence of endocytosis. It indicates that, after entering inside the cells and passing through early endosomes, the CSMs accumulate in the late endosomes and lysosomes, like other “hard” nanomaterials. Figure 3.1.16. Confocal images of intracellular delivery of CSMs@DOPE-Atto488 (green labeled) after 3 h incubation with HeLa cells. CellMask Deep Red is used for cell membrane labeling and LysoTracker Blue for lysosomes. Scale bars: 25 µm. Endocytosis is the process by which cells internalize and transport surface proteins, lipids and other macromolecules enveloped within small membrane vesicles formed by invagination of the plasma membrane into the cell interior. This process is key in the regulation of many essential cellular processes involved in homeostasis and communication, including internalization of transmembrane receptors, uptake of extracellular vesicles, plasma membrane remodeling and cell surface signaling.201 Endocytosis takes place through several distinct pathways including clathrin-mediated endocytosis (CME), a selective mechanism whereby cell surface proteins containing specific sorting sequences are gathered into membrane 112 depressions by associating with adaptor proteins that recruit clathrin. CME endosomes pinch off from the cell surface by recruiting the dynamin GTPase to the bud neck. Another dynamin dependent pathway is the fast endophilinmediated endocytosis (FEME) that is induced upon ligan binging to specific receptors and it is a clathrin independent mechanism. Clathrin-independent endocytosis (CIE) is involved in the internalization of glycolipid-binding toxins, glycosylphosphatidyl inositol-anchored proteins (GPI-AP), and many cell surface proteins (channels, transporters, proteins involved in cell-cell and cell-matrix interactions and in cellular immune function). CIE occurs independently from adaptor proteins and clathrin coats, and mostly does not require dynamin for vesicle scission. It has been shown that the small GTPase Arf6 is associated with the uptake and sorting of many plasma membrane proteins202 while some lipid-raft associated pathways (called “CLIC/GEEC”)203 are involved in endocytosis of GPI-AP. Rho proteins have been implicated in yet another CIE pathway.204 It is likely that the Arf6 and CLIC-GEEC pathways are closely related since they both are clathrinand dynamin-independent, cholesterol-dependent and carry GPI-AP into the cell. Finally, macropinocytosis is a stimulated form of CIE where large pinosomes are brought into the cell interior as a consequence of cellular protrusions in an actin-dependent process (Figure 3.1.17).205 Figure 3.1.17. Overview of the primary endocytic pathway and mechanisms of cell uptake (adapted with permission from ref.206. Copyright 2021, Springer Nature Limited). In this study, the endocytosis mechanism that involves the CSMs internalization was investigated using endocytosis inhibitors for the most 119 Figure 3.1.21. Effect of endocytosis inhibitors on the internalization of CSMs, PSNPs and CSMs@HeLa@PSNPs. a) Green fluorescence intensity of DOPE-Atto488 functionalized CSMs and red fluorescence intensity of PSNPs internalized by HeLa, A549 and RAW cells was determined by FC end represented as relative uptake (%). b) CSMs@HeLa@PSNPs internalization was represented as relative uptake (%) in both fluorescence channels (Green-Blue channel (exc./em.: 488/525 nm) on the left and Yellow-Green channel (exc./em.: 532/583 nm) on the right). All data are expressed as mean ± s, n=3. a) b) 120 3.1.4. Effects of biomimetic coating on the penetration into 3D cell culture The current process for testing the effectiveness of nanosystems relies greatly on animal models. However, the animal models that are readily available such as the human tumor xenograft or genetically engineered mouse models are expensive, technically difficult, time-consuming to use and poorly representative of the clinical disease. Additionally, it is more difficult to achieve the resolution needed to see penetration at the cellular and tissue level, and so these models will be less useful for understanding and, inappropriate for mass screening of nanodelivery systems.222 Beside technical and economic drawbacks, nowadays, the use of animals in research, teaching and testing is an important ethical and political issue, because many of these experiments cause pain to the animals involved or reduce their quality of life. 3D cell cultures could provide an attractive alternative to animal models. 3D cell cultures display a variety of features, which are absent in cell monolayers (2D), such as a complex network of cell-cell contacts and advanced extracellular matrix. They develop pH, oxygen, metabolic and proliferative gradients causing stratification in mature spheroids, which resemble vascular stages of solid tumors223-226 and micrometastases,227 mimicking the situation in vivo. The spheroid structure is driven by nutrient and signal gradients223 resulting in an outer zone of proliferating cells, followed by an inner hypoxic area with quiescent cells,228 which encloses a necrotic core.229 Here, 3D spheroids were developed as in vitro model to mimic the in vivo tumor microenvironment as far as possible. Following the protocol in the section 2.5.2, the 3D spheroids were made with tumoral HeLa cells. The cells were grown in agarose-pretreated plates in order to avoid the cell adhesion to the bottom plate surface. In this way, the cells grow agglomerating one on top of the other rather than forming the typical cell monolayer of the 2D culture. Three days after cell seeding, confocal microscopy studies were performed to investigate the morphology of the HeLa spheroids. Under followed experimental conditions, the structure self-assembled with a spheroidal shape about 494.5 ± 20.7 µm in size. Cell membranes were stained with CellMask™ Green plasma Membrane stain and multiple images 121 were captured at different focal depths in order to get a Z stack imaging and a 3D reconstruction (Figure 3.1.22). The interior of the spheroid appears dark, as light scattering drastically reduces the fluorescence signal due to the opacity of the spheroids. This limitation implies that the signal is only detected from the outermost cells of the spheroid. Figure 3.1.22. 3D projection of confocal Z scan of HeLa spheroid three day after cell seeding. The CellMask™ Green plasma membrane was used to label the membrane of the cells. Three individual scans at different depth are shown in green channel, bright field and both channels merged. Scale bars: 100 µm. 122 The number of seeded cells and the final time point determine the spheroid size achieved, and it controls the spheroid dimensions with low variability among several samples.  500 µm size is sufficient to establish physiological gradients for nutrients, oxygen, pH, and catabolites due to limitations in diffusion through the multicellular layers. The effect of these gradients is the establishment of heterogeneous cell populations with necrotic cells in the core and proliferating cells on the spheroid surface. Cell viability assay was performed in order to evaluate the evolution of the tumor spheroid growth over time. Once the spheroids were made, they were disaggregated after a certain time: day 0 (time 0, 3 days after initial cell seeding), day 1 (24 h time) and day 2 (48 h time). Cell viability was studied by employing a dead/alive cell assay based on two fluorescence cell markers: calcein-AM, which is retained within live cells and produces a strong green fluorescence, and propidium iodide (PI), which binds to DNA of dead and/or necrotic cells, producing red fluorescence signal (see section 2.5.2). In Figure 3.1.23, FC graphs show the population distributions in the quadrant sectioning over the green (exc./em.: 488/525 nm) and red (exc./em.: 532/620 nm) fluorescence channels. Live cells (upper left quadrant) are traceable by a strong green signal but low red signal, while dead cells (lower right quadrant) by lower green signal and strong red signal. By these studies, it was observed that the live cell population % decreased over time from  73% to  60%, while the dead cell population increased from  11% to  13%, accordingly. It is possible that the dead cell % corresponded to the cell population that are located into the inner spheroid area, where they turn into necrotic cells caused to the hypoxic condition and to a less presence of nutrients. However, some studies showed that the cells in the inner part of 10-day-old spheroids are still able to migrate and repopulate after spheroid disaggregation, indicating that the core cells are still viable.230 123 Figure 3.1.23. FC studies of spheroid cell viability by calcein-AM/PI staining. Data are represented by scatter density plots of green fluorescence signal (calcein-AM signal, viable cells, Green-Blue channel) versus red fluorescence (PI signal, non-viable cells, Orange-Green channel). Negative control was treated with triton 1.5 X to lysate the cells. Then, the penetration and uptake of CSMs was studied in 3D spheroids models. Once the spheroids were made, DOPE-Atto647N labeled CSMs@HeLa were incubated for 30 min, 3 and 24 h. FC analysis showed that the CSMs internalization in 3D spheroids is time-dependent (Figure 3.1.24a), in agreement with the previously shown results in previous results with 2D cell cultures. After a short incubation time, the results showed low level of CSMs internalization indicating the poor spheroid penetration of CSMs probably due to that only the outermost cellular layers of the spheroid are truly exposed to the CSMs dispersion. After 24 hours, the CSMs rate increased, suggesting that longer incubation times allow an enhance interaction between the CSMs and the spheroid, leading to an increase of the CSMs´ their penetration into the inner part of the spheroid. Figure 3.1.24b shows the 3D reconstruction of confocal microscopy Z-scan of a HeLa spheroid incubated during 24 h with CSMs@HeLa. The fluorescence of CSMs (in violet) appeared partly colocalized with the cell membrane (stained in green by CellMask staining) of cells of the spheroid rim. However, the presence of isolated fluorescence signals of the CSMs indicated that, after 24 h incubation, CSMs penetrated the inner spheroid area. 124 Figure 3.1.24. CSMs internalization in 3D spheroid models. a) FC dotplot of MFI values (red-green channel exc./em.: 532/695 nm) measuring uptake at different time points. The control represented the autofluorescence of no CSMs treated spheroid. All data are expressed as mean ± s, n=3. b) 3D reconstruction of confocal z-images of spheroid incubated with CSMs for 24 h. In green, the fluorescence of the cell membranes labeled by CellMask staining. In Violet, the DOPE-Atto647N labeled CSMs fluorescence. In last panel, both fluorescence merged. Scale bars: 100 µm. a) b) 125 Further investigations were carried out to study the effect of CSMs@HeLa coating on the penetration and uptake of PSNPs in 3D spheroid models. First, the viability of the spheroid cells was verified with calcein-AM, after PSNPs and CSMs 24 h incubations (Figure 3.1.25). FC analysis showed that the % of live cell population stained with calcein-AM was around 55% when the cells were incubated with CSMs, and around 69% when they ware incubated with PSNPs. These results indicated that the cell viability is not impaired neither upon the CSMs treatment nor upon the PSNPs treatment. Figure 3.1.25. FC studies of spheroid cell viability by calcein-AM staining after 24 h incubation of CSMs and PSNPs. Data are represented by scatter density plots of green fluorescence signal (calcein-AM signal, viable cells, Green-Blue channel) versus side scatter. By confocal microscopy, the fluorescence distributions of spheroids treated with CSMs (Figure 3.1.26a), CSMs@PSNPs (Figure 3.1.26b) and bare PSNPs (Figure 3.1.26c) were studied. 3D reconstruction of the spheroid treated with CSM-coated PSNPs showed that the green fluorescence of CSM coating and the red fluorescence of PSNPs is colocalized (observed as yellow in merged image) indicating that the NPs were efficiently coated by 126 the CSMs and internalized. Moreover, bare PSNPs showed less internalization comparing with the CSM@PSNPs, indicating that the coating significantly improved the interaction of PSNPs with 3D spheroid. Figure 3.1.26. 3D reconstruction of confocal Z-images of spheroids incubated with a) CSMs, b) CSMs@PSNPs and c) bare PSNPs, for 24 h. In green, the fluorescence of the CSMs (DOPE-Atto488 labeling). In orange, the PSNPs fluorescence (Rhodamine labeling). In yellow, both merged signals derived from the colocalization of CSMs and PSNPs fluorescence of CSMs@PSNPs sample. Scale bars: 100 µm. The effect of CSM coating on the PSNPs penetration was studied with more detail. With this purpose, confocal microscopy analysis was performed on 3D HeLa spheroids and they were incubated with bare PSNPs and CSM@HeLa@PSNPs, for 1 day (Figure 3.1.27) and 3 days (Figure 3.1.29). The figures report the 3D-recostraction of spheroids incubated with the two samples and three different planes in the z-axis. After 1 day incubation, the CSMs coating did not seem to have had a significant effect on the PSNPs a) ) a) b) c) 127 penetration, as the fluorescence signal appeared limited at the rim of the spheroids in both cases. It is confirmed by the fluorescent intensity plots obtained from the confocal image of one of the z-plane of the spheroids (Figure 3.1.28). Figure 3.1.27. Set of pictures from a HeLa cell spheroid incubated for 24 h with PSNPs coated with CSMs@DOPE-Atto488, and bare PSNPs. Three individual scans at different depth in the Z-axis are shown. The depth of the plane increases from up to down. Orange: PSNP fluorescence. Yellow: merged PSNPs and CSMs fluorescences. Scale bars: 100 µm. 128 Figure 3.1.28. a) Individual scans of innermost sections of two spheroids incubated with CSMs@PSNPs and bare PSNPs for 24 h. b) The corresponding fluorescence intensity plot derived from the confocal images above. Scale bars: 100 µm. After 3 days incubation, the fluorescence intensity of bare PSNPs increased in peripheral cells, without appreciable penetration in the central region of spheroid. On the other hand, the spheroid incubated with CSMs@PSNPs showed higher level of accumulation in the inner part of the spheroids, suggesting that the CSMs@PSNPs penetration into 3D models is time-depending. The fluorescence intensity profile of the selected z-sections (Figure 3.1.30b) showed higher intensity peaks of CSMs@PSNPs compared to the bare PSNPs, confirming that the coating improved the penetration into the spheroid core. a) b) 135 CSMs were produced by the method described in the first chapter,185,186 and the positively charged PEGylated GNRs and/or cargo (fluorescently labeled molecules, such as Alexa555 BT7R Ab or Alexa545-PEG(4)) were introduced into the nanoconstructs by the fusion process. Briefly, HeLa cells were used for the cell membrane fragments extraction, as described in the Materials & Methods, section 2.2.6. The membrane fragments were selfassembled into CSMs by cycles of extrusion through 800 nm pore size polycarbonate membranes. The resulting CSMs were mixed with GNRs and/or the selected cargo and extruded again through the polycarbonate membrane leading to GNR-tagged CSMs (i.e., CSMs@GNR) with or without cargo. After the purification steps (see section 2.3), the CSMs@GNRs@Alexa555 BT7R Ab were characterized with DLS, EM and NTA analyses. The DLS studies showed that, upon the GNRs fusion and the cargo loading, the hydrodynamic diameter (dH) of the CSMs remained nearly constant at ∼230 nm. As a result, cargo loading or GNR-tagging have no significant effect on the hydrodynamic size of the nanosystem (Figure 3.2.5 and Table 3.2.1). 136 Figure 3.2.5. DLS measurements of CSMs. DLS distributions in number (DH,N)), and intensity (dH,I) of three independent measurements of different CSM samples: CSMs (a, e), CSMs@Cargo (b, f), CSMs@GNRs (c, g) and CSMs@GNRs@Cargo (d, h), where the cargo is Alexa555 BT7R Ab. ζ potential of the distribution graphs of CSMs (i) and CSMs@GNRs (j). Data expressed as mean ± s, n=3. Table 3.2.1. Mean hydrodynamic diameter derived from the DLS distributions of CSMs, CSMS@GNRs and CSMs@GNRs@cargo where the cargo is Alexa555 BT7R Ab. [1] dH,I and dH refer to the mean average hydrodynamic diameter from the intensity and number DLS distributions. PDI refers to polydispersity index. Standard deviation values were calculated from five measurements of the different CSMs samples. Sample Sample dH,I (nm)[1] dH,N (nm)[1] PDI[1] -potential (mV) CSMs Without cargo 288 ± 33 234 ± 13 0.26 -24.1 ± 1.0 Cargo 327 ± 7 295 ± 7 0.14 CSMs@GNRs Without cargo 279 ± 36 210 ± 9 0.16 -12.9 ± 2.5 Cargo 287 ± 7 219 ± 10 0.25 a) b) c) d) e) f) g) h) b) i) j) 137 However, due to the inclusion of the positively charged PEGylated GNRs, the ζ-potential of the CSMs@GNRs system decreased (-15 vs. -25 mV) in comparison to the non-GNRs tagged CSMs (with or without Alexa555 BT7R Ab). NTA showed CSMs size distribution similar to the results obtained from DLS and allowed to define the CSMs concentration (Figure 3.2.6). Figure 3.2.6. NTA measurements. Size distribution analysis of CSMs (red line, mean 267 ± 147 nm, 5.7·108 CSMs·mL-1) and CSMs@GNRs (black line, mean 331 ± 125 nm, 2.4·108 CSMs·mL-1) dispersed in PBS buffer pH 7.4. EM was used to confirm the morphology of the CSMs and the functionalization with GNRs (Figure 3.2.7), which is also indicated by the plasmonic band centered at ~ 800 nm in the UV-Vis spectrum of the purified CSMs@GNRs structure (Figure 3.2.8). 138 Figure 3.2.7. SEM images of CSMs. a) CSMs STEM images after Uranyl acetate negative staining. b) CSMs@GNRs STEM images. Scale bars: 200 nm. Figure 3.2.8. UV-Vis absorption spectra of CSMs@GNRs and GNRs. a) b) 139 In addition, ICP-MS measurements allowed us to confirm and quantify the amount of GNRs intercalated in the CSMs samples (Table 3.2.2). Table 3.2.2. ICP-MS analysis of total GNRs added to the CSMs and purified CSMs@GNRs sample. CSMs@GNRs g Au·L-1 Mol· L -1 GNRs· L -1 GNRs CSMs GNRs·CSM-1 AV 0.0036 1.48·1010 8.89·1013 8.89·1010 2.40·109 37.06 SD 0.0013 5.28·1011 3.18·1013 3.18·1010 13.25 GNRs g Au·L -1 Mol· L -1 GNRs· L -1 GNRs AV 0.0180 7.32·10-10 4.41·1014 4.41·1011 SD 0.0040 1.63·10-10 9.82·1013 9.82·1010 Then, the biological properties of the CSMs functionalized with GNRs were studied. FC was used to evaluate different surface parameters among the different CSM preparations, (Figure 3.2.9); 83 i) the CSM´s size distribution reported by of SSC signal versus FSC signal (Figure 3.2.9 column a, SSC), ii) cargo loading by the corresponding fluorescence signal (Figure 3.2.9 column b, Cargo) where the cargo is Alexa545-PEG(4), iii) the presence of lipids after CellMask staining (Figure 3.2.9 column c, cell membrane), and iv) the presence and correct orientation of some cell adhesion proteins (Figure 3.2.9 column d, protein membrane), using a fluorescence labeled Ab as a reporter for the cadherin-family proteins. The immunolabeling detection of the extracellular domain of the cadherins confirmed a right-side-out membrane orientation on the synthesized CSMs and the GNR-functionalized CSMs. 140 Figure 3.2.9. CSMs dispersion analysis by flow cytometry. The variation of the flow cytometry side scattering signal and the fluorescence signal as a function of the fluorescently labeled CSMs are shown. a) Scatter density plots of SSC signal versus FSC signal for different CSMs (with or without cargo and with or without GNRs), where the cargo is Alexa545-PEG(4). b-d) Scatter density plots of SSC signal versus fluorescence signal (from each corresponding channel: 525/30 nm, 583/26 nm and 695/50 nm) for the CSMs with encapsulated cargo (b, 583/26 nm channel) and with different labels: CellMask Deep Red lipid staining (c, 695/50 nm channel) and Alexa-488-anti-cadherine antibody (d, 525/30 nm)). 141 Furthermore, the protein content of the purified CSM samples was determined by a colorimetric assay, which showed a similar protein content in all final CSM samples produced from the same number of HeLa cells (Figure 3.2.10), independently of the GNRs incorporation to the CSMs. Figure 3.2.10. Protein determination on CSMs and CSMs@GNRs samples by Bradford assay. All data expressed as mean ± s.d., n=3. The efficiency and robustness of the encapsulation method was studied in detail at pH 7.4, using a fluorescent Alexa545-PEG(4) molecule as a model cargo. The final concentration of the encapsulated cargo was obtained using a calibration curve of the fluorescently labeled cargo (see section 2.4) (Table 3.2.3. The concentration of encapsulated cargo (Alexa545-PEG(4) or Alexa555 BT7R Ab) was calculated by measuring the fluorescence intensity of the CSMs@Cargo samples (calibration curves of Figure 2.4.1 were used for the quantification).Table 3.2.3). The stability of the cargo encapsulated in the CSMs and its release were studied by monitoring the fluorescence signal over time (tmax = 72 h) (Figure 3.2.11) using dialysis purification process. The results showed a fast release during the initial hours due to the unspecific adsorption of the cargo onto the CSM surface. After 24 h of dialysis no more free cargo was found in the solution. Table 3.2.3. The concentration of encapsulated cargo (Alexa545-PEG(4) or Alexa555 BT7R Ab) was calculated by measuring the fluorescence intensity of the CSMs@Cargo samples (calibration curves of Figure 2.4.1 were used for the quantification). Cargo Conc. cargo (g·mL-1) Conc. CSMs (CSM·mL-1) Cargo·CSM-1 Alexa545-PEG(4) 0.28 ± 0.1 7·109 24353 ± 8777 Alexa555 BT7R 0.15 ± 0.03 10·109 60 ± 22 142 Figure 3.2.11. Stability study over time of the cargo encapsulated on the CSMs at pH 7.4; the leaking of the cargo over time was measured by the fluorescence signal in the buffer solution. All data expressed as mean ± s.d., n=3. 3.2.2. Biomimetic properties The biomimetic potential of CSMs derived from different type of cell sources such as neutrophils, leukocytes, erythrocytes, etc., along with their ability of targeting activated endothelia, resolving inflammation and repairing tissue damage, have been reported before in the literature.127,135,236,237 Herein, CSMs@HeLa were synthesized, with or without GNRs, which were incubated with two tumoral cell lines, the HeLa (same origin as the CSMs) and the A549 cells, and with HEK cells for 1, 3 or 6 h. The two different types of CSMs (CSMs and CSMs@GNRs) were equivalently labeled with DOPE-Atto488. The uptake of different nanosystems was studied by FC (Figure 3.2.12). The CSMs excess that was not internalized by the cells was removed by three washing steps with PBS. MFI values were normalized by the autofluorescence signal of each cell line, leading to dimensionless fold increase MFI values, which we used to compare cell uptake among the different CSMs and cells. Additionally, as control experiment, we incubated cells with a dose of free DOPE-Atto488 equivalent to the amount labeling the CSMs (equivalent among CSMs and CSMs@GNRs) which, as expected, was more efficiently incorporated by the tumoral cell lines than by the HEK cells, due to the alteration of cancer cell metabolism in comparison to non-tumoral cells.196 We should notice that the uptake of free DOPE-Atto488 is higher in A549 than in HeLa cells, at any incubation time (1, 3 or 6 h). The GNRs incorporation has no significant effect on cell uptake of CSMs, at any incubation time or cell type (for derivatization of the CSMs or as internalization target). In contrast, we found that HeLa-derived CSMs, with 143 or without GNR functionalization, were more efficiently internalized by HeLa cells than in the other cell lines (2-fold and 10-fold MFI than in A549 and HEK cells, respectively), at all the incubation times tested. Figure 3.2.12. Quantification of the MFI of the cellular uptake of CSMs@DOPE-Atto488, CSMs@DOPE@GNRs and free DOPE in HEK, A549 and HeLa cells after 1 h, 3 h and 6 h incubations. Green fluorescence intensity of DOPE-Atto488 functionalized CSMs internalized by HeLa cells was determined by FC. Data expressed as mean ± s, n=3. These results demonstrate that HeLa derived CSMs show a higher affinity to the cell lines from which they are derived (homotypic targeting), which significantly improves the uptake by the homotypic cells. These results were consistent with those reported in the literature in which homotypic cancer cell membranes have been used to achieve highly specific self-recognition to the source cell line, corroborating that strong cellular internalization levels of cell-derived membrane can be obtained for the same cell line in vitro135 and self-tumor targeting in vivo.185,238 The fact that residual biomolecules from the cancer cell membrane remain at the surface of the CSMs leads to preserve their primary targeting capabilities.239 Tumor cells readily agglomerate with strong adhesion to constitute solid tumors, which could be caused by adhesive interactions of specific proteins (focal adhesion proteins, integrins, etc.) of the cell surface. These specific strong adhesions among homotypic tumor cells may be applied for tumor self-targeting strategy by biomimetic coatings enable of selfrecognition.185,238 144 3.2.3. Thermoplasmonic properties After the preparation and bio-characterization of CSMs, the thermoplasmonic properties of the CSMs@GNRs system, as well as the corresponding quantification of cargo (Alexa545-PEG(4)) release induced by NIR photostimulation, were studied with a NIR optical system (see Materials and Methods). For this purpose, solutions of CSMs@GNRs nanostructures in PBS (108 CSMs·mL-1 concentrated) were irradiated with an 808 nm collimated NIR beam (illumination area ~ 0.33 cm2, 30 W·cm2) during different times (0.5 – 4 min). The heating profile is shown in Figure 3.2.13, the temperature of the CSM@GNR solution raised rapidly reaching the maximum at 2 min under laser irradiation (ΔT ~ 50 ºC). Figure 3.2.13. Temperature curve of PBS and CSMs@GNRs dispersions under NIR laser irradiation of 808 nm at 30 W·cm-2. Data expressed as mean ± s, n=3. As expected, the temperature of the solution increased above room temperature (~ 23 ºC) with longer irradiation times, which also resulted in an increase of the cargo release. As shown in Figure 3.2.13, the temperature of the CSMs@GNRs solution raised rapidly reaching the maximum between 2 and 4 min under laser irradiation (ΔT ~ 15 – 50 ºC). Indeed, using low (0.5 min) and high (4 min) NIR treatments allows delivering 40% - 75% of the encapsulated cargo, respectively (Figure 3.2.14).