Stroke is the leading cause of death and adult disability in Portugal and the second cause of death worldwide. In the coming years, this health condition will worsen due to the increase in average life expectancy and the increase in the number of individuals with risk factors for stroke. Currently, the recommended treatment for stroke patients is the intravenous thrombolysis therapy (with alteplase or endovascular thrombectomy), although it has limited efficacy due to the narrow therapeutic window where it can be applied without causing hemorrhagic transformations, limiting the benefit to many patients and highlighting the need to develop alternative and safer therapies. The development of strategies aiming to potentiate endogenous neuroprotective mechanisms has emerged as a safe alternative to thrombolysis, however, these approaches have undervalued the importance of the cerebral vasculature in the post-stroke recovery. Previously, it was demonstrated that cerebral vascular remodeling is correlated with higher survival rates and improved long-term neurological function in stroke patients. Accordingly, in this research work, it was proposed the modulation of the Argonaute (Ago)-2 endogenous mechanism to enhance the brain vascular function compromised by cerebral ischemia and, consequently, contributing to full parenchymal recovery after a stroke. Besides being the major coordinator of microRNA traffic and activity, Ago2 protein is pivotal for cell survival and tube network formation in peripheral endothelial cells. However, these mechanisms and the role of Ago2 in the ischemic brain vasculature remain unknown. In this sense, the potential of this protein in the inflammatory process was first assessed. This pathological process is one of the most important post-stroke pathological events that can lead to short-term mortality or promote long-term disability. Accordingly, in vitro (mouse primary brain cultures of endothelial and glial cells stimulated with lipopolysaccharide) and in vivo (adult mice intraperitoneally injected with lipopolysaccharide) data revealed that exogenous Ago2 application (downregulated in brain vasculature in these conditions) normalized endothelial markers associated with inflammatory-induced damage which, in turn, reestablished previously compromised glial and neuronal function. Then, after confirming the potential of Ago2 for cerebral ischemia studies, a strategy to enhance the Ago2 intracellular levels (also downregulated in the ischemic context) and to promote endothelial cell survival was defined. The controlled release of retinoic acid (a signaling molecule with defined anti-inflammatory and pro-angiogenic properties) via polymeric nanoparticles arose as a promising approach. In vitro data (mouse primary brain cultures of endothelial cells subjected to oxygen and glucose deprivation) revealed the protective potential of RA-NP in ischemic conditions, only observed by restoring basal levels of Ago2 protein (an effect mediated by the nitric oxide signaling), highlighting the importance of Ago2 in vascular function. Additionally, in vivo data (adult mice subjected to transient occlusion of the middle cerebral artery) showed that the systemic administration of this formulation promoted tissue survival, normalized inflammatory response, and promoted neurovascular repair (with a robust pro-angiogenic effect) in the peri-infarct region after stroke. However, further studies are needed to establish a correlation between the observed in vivo outcome and the Ago2 protein, and of RA-NP treatment and to decipher interactions between this protein and retinoic acid. Altogether, the experimental research developed throughout this Thesis provided evidence that the modulation of the Ago2 endogenous mechanism can be envisioned as a neuroprotective approach in post-stroke recovery. Furthermore, it revealed a safe nanotechnological tool able to modulate this system, suitable for intravenous administration, and compatible with pre-existing therapies.

Cervical cancer is one of the leading causes of death amongst women, especially in countries lacking suitable access to health and hygiene care. Contrarily to most cancers, which usually take origin in mutations related with the action of mutagenic or carcinogenic agents, the development of cervical cancer is strongly associated with infection by Human Papillomavirus (HPV). This virus is responsible for the expression of E6 and E7 oncoproteins, which interfere with the cell cycle regulation, affecting the cellular mechanisms of repair and proliferation of infected cells. In the last decades, DNA-based therapies, such as DNA vaccines or gene therapy, have been highly explored by researchers in the search of an efficient non-viral vector for the cervical cancer treatment. Plasmid DNA (pDNA) is a popular non-viral vector, which has been widely studied in the past decades for the development of a variety of DNA vaccines and gene therapy strategies. However, the presence of CpG motifs necessary to its amplification in prokaryotic hosts may lead to the activation of the immune system and to the degradation of this molecule before it can reach the target cells, in a gene therapy perspective. However, these same motifs may contribute to the recognition of this bioproduct by antigen presenting cells, facilitating the processing of pDNA as a DNA vaccine. Minicircle DNA (mcDNA) consists in a non-viral vector whose popularity has been largely increasing in the last years. This vector results from the recombination of a molecule, which is similar to conventional pDNA, named parental plasmid (PP). However, PP presents two recombination sites which, upon induction, give origin to the recombination of this molecule into two different molecules: one containing all the prokaryotic information necessary for PP amplification in prokaryotic hosts (miniplasmid – mP) and another containing all the information necessary for the therapeutic gene expression in eukaryotic cells (mcDNA). Considering its innovative character, it is crucial to develop suitable strategies for mcDNA preparation, with a purity that fits the requirements established by regulatory agencies such as European Medicines Agency (EMA) or Food and Drug Administration (FDA). Thus, this work was performed with two different perspectives regarding cervical cancer treatment, namely DNA vaccines and gene therapy. Firstly, concerning DNA vaccines, the application of an adequate purification strategy allowed to understand that the purity degree of the pDNA sample can significantly contribute for the increased expression of the target antigen. The application of a monolith modified with arginine ligands, comparatively to the use of commercially available techniques, allowed to retrieve a final DNA vaccine sample with higher supercoiled (sc) content, leading to a higher expression of the target proteins. Concomitantly, the development of nanocarriers composed by calcium carbonate and gelatin allowed the delivery of pDNA vaccine to dendritic cells, also known as antigen presenting cells, without detecting cytotoxicity. On the other hand, two different strategies for mcDNA purification were explored. Firstly, the preparation of a monolith modified with cadaverine ligands and its use in the purification of sc mcDNA allowed a recovery yield of 78.6% coupled with 98.4% purity. These results turned out to be very interesting considering the current mcDNA purification scenario in the literature. On the other side, size exclusion chromatography was also explored through the use of Sephacryl SF-1000 matrix. This strategy allowed to obtain a sc mcDNA yield of 66% alongside a purity of 98.35%. Despite 12 times more mass of pure sc mcDNA was obtained in one assay, the assay turned out to be approximately 8 times longer than the assays performed with the cadaverine modified monolith. In the end, two strategies were developed for sc mcDNA purification, each presenting its advantages and disadvantages. Nonetheless, both allowed the purification of mcDNA without resorting to PP specific genetic modifications, which implies that these strategies can be universally used for mcDNA purification. However, the need to implement an analytical methodology that allows to quantify sc mcDNA content in a fast, low cost and effective way led to the study of cadaverine-modified monolith with this purpose. Thus, a chromatographic method was implemented and validated to quantify sc mcDNA concentration present in a complex or pure sample, within a range of 1-25 μg/mL. Lastly, to study the effectiveness of miR-375 in the treatment of cervical cancer, in vitro studies were performed to evaluate the effect displayed by mcDNA-pri-miR-375 in the expression of E6 an E7 oncoproteins. Therefore, CaSki cell line model was used, which consists in metastatic cervical cancer cells infected by HPV-16. Also, a fibroblast cell line was used as a non-carcinogenic model for control. It was possible to identify the correct expression of miR-375, coupled with a decrease in E6 and E7 transcript and protein levels. Furthermore, it was observed that fibroblast viability was not affected, instead observing a decrease in transfected CaSki proliferation and viability. Thus, the application of mcDNA-pri-miR-375 as a possible therapeutic strategy should be more amply studied in the future. In the end, this thesis presents a scientific work that intends to contribute towards the scientific community with useful information for the potential development of new approaches for cervical cancer treatment, through the implementation of two biotechnological platforms. One of the strategies is based on the obtainment of a DNA vaccine, encoding E6 and E7 HPV antigens with the aim of activating two pathways of the immune system, the preventive/humoral and the therapeutic/cellular pathways, allowing the consequent elimination of the antigen-expressing and HPV-infected cervical cancer cells. The second strategy is focused on the obtainment of innovative gene therapy mcDNA vector, encoding pri-miR-375, to silence E6 and E7 HPV oncoproteins, allowing the direct targeting of infected cervical cancer cells, through the re-establishment of tumor suppressor proteins expression or function.

Curcumin is a polyphenol extracted from the rhizomes of Curcuma longa, also known asTurmeric. It has good therapeutic properties such as anti- anti-carcinogenic, chemoprotectiveanti-inflammatory and antioxidant effects. However, despite its benefits, it has low aqueoussolubility, and consequently poor bioavailability. One strategy to overcome this encapsulating the curcumin into a drug delivery system, like micro- or nanoparticles.The nanoparticles (NPs) have increased their importance since they can improve the properties of certain therapeutic active molecules, like their retention time in the local ofadministration, allowing a more controllable drug release, or even providing a more specialized targeting. Among diverse polymers, carbohydrates, and chitosan in particular, have been intensively used. Chitosan results from the deacetylation of chitin and is a polymer consisting of alternating β (1-4) linked N-acetyl-D-glucosamine and D-glucosamine residues. It has many applications based on biological activities such as mucoadhesive and, immunostimulating properties. In the literature, chitosan NPs have been used in areas like, vaccines or cancer treatment, as a delivery system. This work aims to develop chitosan NPs loading curcumin to target the hepatocellular carcinoma. For that the chitosan was chemically associated with the lactobionic acid (LA) as a ligand to bind to the receptor asialoglycoprotein (ASGPR). The thesis described the production of chitosan NPs modified with lactobionic acid encapsulating curcumin. These NPs were then tested on cells and then compared with non-modified chitosan particles.The NPs had a low polydispersity index (PDI), 0.261 for ChiCas, 0.418 for ChiCas-Cur, 0.188for LA-modified ChiCas and 0.193 for LA-modified ChiCas-Cur, with a small size, 326.9 nmfor ChiCas, 499.2 nm for ChiCas-Cur, 207.3 nm for LA-modified ChiCas and 203.3 nm forLA-modified ChiCas-Cur. In terms of the zeta potential, all NPs presented a positive zetapotential with values 37.1 mV for ChiCas, 42.1 mV for ChiCas-Cur, 44.1 mV for LA-modified ChiCas and 40.9 mV for LA-modified ChiCas-Cur, presenting minor variations between batches. Based on the SEM analysis, the NPs also showed a round shape morphology. After the characterization of the NPs, the immunotoxicity and immunomodulatory effects were evaluated in two different cell types, THP-1 cells and HepG2 cells. THP-1 cells were used to study the effect of the NPs on our immune system, and the HepG2 cells were used to study the effect of the NPs with the presence of the LA on the cells of the liver (hepatocytes). The results of the studies showed that none of the NPs were toxic for the cells.Furthermore, the production of HepG2 spheroids, to test the same formulations, were implemented in the laboratory to confirm the results. The results obtained demonstrated that the NPS did not exhibit toxicity towards the spheroids.

Extracellular vesicles (EVs) are membrane-enclosed particles secreted by all cell types, with a role in intercellular communication. The cellular crosstalk mediated by EVs is a result of the transfer of proteins, lipids and nucleic acids and the interaction with surface receptors - a process conserved throughout evolution. The ability of EVs to transfer functional cargo and cross biological barriers has prompted interest in engineering these particles as next-generation drug delivery systems.EVs are a promising platform to deliver large therapeutic proteins and ribonucleoproteins, such as gene editing enzymes. However, given the complexity of gene editing technologies and major bottlenecks in efficiently delivering protein-based payloads transiently, further advancements are needed to establish EVs as next-generation drug delivery systems. To address these challenges, this project aims to engineer EVs to efficiently load gene editing enzymes, optimize cargo release and improve their cellular uptake - ultimately achieving high levels of gene editing in vitro and in the context of Spinocerebellar Ataxia Type 3 (SCA3).In chapter 3, we focused on identifying novel proteins enriched in EVs using mass spectrometry analysis. We aimed at identifying cell-type specific scaffolds enriched in EVs derived from neuronal and mesenchymal cells, broadening the ability to engineer these cells to endogenously load therapeutic proteins. We started by optimizing a protocol based on density gradient ultracentrifugation to isolate highly pure EV samples. After establishing a proteomics pipeline, we identified a subset of proteins highly enriched in mesenchymal-derived EVs and neuronal-derived EVs. These scaffolds are promising candidates for generating recombinant proteins with the ability to endogenously loading proteins of interest in EVs.In chapter 4, we engineered the outer surface of EVs to improve cell uptake and delivery efficiency of gene editing enzymes. To do this, we engineered tetraspanin CD63, a known hallmark of small EVs, to display an arginine- and lysine-rich outer surface, named SPARK. We showed that SPARK EVs have a 6.6-fold improvement in cell uptake. SPARK EVs expressing VSV-G – generating virus-like particles (VLPs) - efficiently deliver Cre recombinase in hard-to-transfect cell types. We demonstrate that SPARK VLPs are able to mediate EMX1 gene knockout in vitro. Additionally, to the best of our knowledge, we are the first to demonstrate VLP-mediated delivery of Bxb1 serine recombinase to facilitate a large gene insertion of a 4kb construct provided by lipid nanoparticles (LNP) mediated delivery.In chapter 5, we developed a controllable release system to deliver CRISPR-Cas9 RNPs in EVs as a therapeutic strategy for treating the monogenic neurodegenerative disorder Spinocerebellar Ataxia Type 3 (SCA3). SCA3 is characterized by an expanded CAG triplet repeat in the ATXN3 gene. For this purpose, we engineered an EV-sorting palmitoylation signal with a photocleavable linker, PhoCl, to promote the loading and light-triggered release of CRISPR-Cas9 RNPs from EVs. We showed that this system efficiently packages SpCas9 nuclease and sgRNAs into EVs. Upon blue light activation, EVs deliver more efficiently CRISPR-Cas9 RNPs to recipient cells, reducing ATXN3 levels in vitro. Finally, we demonstrated that CRISPR-Cas9 loaded EVs expressing VSV-G effectively target mutATXN3 in two mouse models of SCA3.In summary, this project provides significant contributions to the engineering of EVs as next-generation drug delivery systems. By addressing key EV bottlenecks, we developed new tools to: 1) improve protein loading in EVs in a cell-type specific manner; 2) modify the EV surface for improved cell delivery; 3) optimize the controlled release of therapeutic cargo from EVs; 4) deliver functional gene editing payloads, such as Cre recombinase, CRISPR-based systems and Bxb1 recombinase; 5) demonstrate the therapeutic efficacy of EVs in neurodegenerative diseases, namely SCA3. EV-based delivery systems represent a promising platform for delivering gene editing technologies, offering the potential to revolutionize the field.

Poor aqueous solubility, stability and bioavailability of interesting active biomolecules is a challenge in the development of bioactive formulations. Cellulose micro- and nanostructures are promising and sustainable carriers with unique features that may be used in enabling delivery strategies. In the context of sustainable development, lignocellulosic biomass from industrial and agricultural wastes have attracted much attention as cellulose sources. Sugarcane plantations for the sugar and alcohol industries are known for their high volumes and large amounts of residues, such as sugarcane bagasse (SCB), a promising renewable and low-cost source of cellulose. This PhD project aims to valorize a major by-product of the agro-industrial sugarcane industry – SCB – by extracting the cellulose polymer and developing micro- and nanostructured systems for the delivery of lipophilic biomolecules as means to offer an efficient and controlled release of these molecules and promote their biological functions. Initially, a chemical and structural characterization of SCB was performed, followed by the evaluation of different approaches for the extraction and purification of cellulose from SCB. Considering the results obtained through the different and complementary techniques employed, the most promising approach revealed to be autohydrolysis (170 ºC for 1 h), followed by bleaching with hydrogen peroxide (12%, v/v, at 85 ºC for 1.5 h) and sodium chlorite (12%, w/v, at 85 ºC for 1.5 h). Through this method we were able to obtain a white cellulose rich fraction (87.5% ± 0.6 cellulose) with low contaminants, high crystallinity index (73.1% ± 1.1) and typical cellulose functional groups from SCB. Microcrystalline cellulose (MCC), cellulose nanofibers (CNF) and cellulose nanocrystals (CNC) were then produced from the SCB cellulose rich fraction, contributing to the valorization of the agro-industrial byproduct. Optimized MCC, CNF and CNC were produced by mild acid hydrolysis, ultrasonication and acid hydrolysis followed by sonication, respectively. MCC had a particle size of ca. 30 μm, CNF of ca. 600 nm long and 8 nm wide, and CNC of ca. 120 nm long and 6 nm wide, which is in agreement with the definition of these types of structures in the literature. ATR-FT-IR and PXRD results confirm the cellulose structure for the three cellulosic materials. CNC had a higher crystalline fraction than CNF, which is attributed to the removal of the amorphous part of the microfibril during acid hydrolysis. The overall properties of our micro- and nanocellulose materials were found to be similar to those of commercial products. Subsequently micro- and nanocellulose were tested as carrier materials for the delivery of lipophilic biomolecules: curcumin, a model lipophilic biomolecule, and cannabigerol (CBG), a promising cannabinoid. Nanocellulose hydrophobic modifications with the surfactant cetyltrimethylammonium bromide (CTAB) and tannic acid/decylamine (TADA) and modification by TEMPO-mediated oxidation, were also tested and compared with the unmodified structures. The modified cellulose structures were characterized and found to successfully bind curcumin, with encapsulation efficiencies (EE) ranging from 63% to 99%. The CTAB and TADA modifications resulted in highly effective EE of 90-99% for curcumin. These modified systems provide a strategy for stabilizing and delivering curcumin by fixing it in hydrophobic domains, potentially improving its solubility and stability. Among the modifications, CNC-CTAB showed the most promising results, allowing for a sustained release of curcumin (ca. 50% released in 8 hours) and exhibiting high EE (ca. 80%) for both curcumin and CBG after scale-up by spray drying. The safety and biological potential of the CNC-CTAB delivery systems encapsulating curcumin and CBG were evaluated. In vitro cytotoxicity and genotoxicity tests were performed on the systems, which revealed to be safe for intestinal application at certain concentrations. Encapsulation reduced the cytotoxicity of both curcumin and CBG, highlighting the potential benefits of using CNC-CTAB as a delivery system for these compounds. The encapsulated biomolecules demonstrated antioxidant and anti-inflammatory properties, effectively reducing reactive oxygen species and cytokine production by intestinal cells. The delivery systems also exhibited antimicrobial properties against Campylobacter jejuni, suggesting its potential in mitigating inflammation in the gastrointestinal tract. Furthermore, the system showed ability to protect curcumin from degradation and facilitate its interaction with the intestinal epithelium, highlighting the potential of CNC-CTAB as a carrier to enhance the biological functions of curcumin and CBG, particularly in the context of intestinal inflammatory disorders. The valorization approach proposed in this work could be beneficial for the sugarcane industry to improve its environmental and economic sustainability in line with circular bioeconomy. This study emphasizes the potential of hydrophobic modified CNC as delivery systems for lipophilic biomolecules. This system effectively reduced the cytotoxicity of curcumin and CBG, protected from degradation, and enhanced biological activities, allowing for their beneficial effects to last longer and be more effective. This breakthrough has the potential to be applied to other compounds and ingredients, serving as an innovative step in utilizing agro-industrial by-products for value creation.

Mucosal surfaces represent a major gateway to microorganisms which may be harmful to health. The humoral immune response has an important action in the defense of these surfaces, as it is able to prevent the entry of pathogens in the body. Vaccines with local application have been evaluated in order to stimulate an efficient immune response in the mucous membranes, since conventional vaccines, for parenteral application, tend to stimulate a mostly systemic response. Vaccines that use the mucosa as an inoculation route are able to generate an immune response directly in the application mucosa and corresponding mucosa, since the mucosal system is integrated, which represents an important advantage in choosing the inoculation route. This paper aims to illustrate some concepts related to mucosal immunity in general, as well as to gather information about what has been studied in relation to mucosal routes of administration of vaccines, immunomodulators and antigen delivery systems.