Platelet-derived fibrin GEL loaded with mesenchymal stromal cells-extracellular vesicles and methods and uses thereof

Abstract

It forms an object of the present invention a fibrin gel which is a platelet-rich plasma (PRP) gel comprising vesicles (EVs) from adipose tissue mesenchymal stromal cells (ASCs). Form further objects method to obtain said gel and uses thereof.

Description

[0001] "PLATELET-DERIVED FIBRIN GEL LOADED WITH MESENCHYMAL STROMAL CELLS-EXTRACELLULAR VESICLES AND METHODS AND USES THEREOF"

[0002] Titolare : Ospedale Galeazzi S.p.A.

[0003] Inventori : Enrico Ragni, Paola De Luca, Laura De Girolamo, Simona Landoni, Federico Valli

[0004] BACKGROUND

[0005] Articular cartilage is an avascular tissue characterized by low number of cells, mainly chondrocytes, that are embedded in an abundant extracellular matrix (ECM). The combination of these features, avascularity and low cellularity, together with the complex nature of ECM hindering stromal cell migration, prevents an efficient cartilage regeneration. Thus, cartilage, once injured by trauma or degeneration, ends in chondropenia, decreased proteoglycan production and damage to collagen meshwork, causing joint pain, functional impairment and eventually Osteo Arthritis (OA).

[0006] Cartilage lesions are frequent, with several studies reporting an incidence between 30% and 60% in the general population and up to 36% in athletes. Debilitation leads to a significant drop in the functioning of individuals, that may be managed by various procedures depending on symptoms and lesion size.

[0007] In the frame of identifying new factors able to promote hyaline cartilage, extracellular vesicles (EVs) released by mesenchymal stromal cells (MSCs) were recently described.

[0008] MSC-EVs are enriched in several molecules, including miRNAs, with antiinflammatory and cartilage regenerative properties. Also, MSC-EVs can deeply penetrate cartilage thus releasing their cargo and enhancing tissue regeneration. Consistently, in several in vivo studies, MSC-EVs reduced cartilage loss, improved collagen immunostaining and greater expression of chondrogenic genes and caused macrophage M2 phenotype polarization that promotes resolution of inflammation and cartilage repair.

[0009] Clinical trials of MSC-EVs are currently underway for osteoarthritis (Yasunari Matsuzaka and Ryu Yashiro, Therapeutic Strategy of Mesenchymal-Stem Cell-Derived Extracellular Vesicles as Regenerative Medicine. Int J Mol Sci. 2022; 23(12): 6480). However, the injection of particles in the joint cavity necessarily leads to a dispersion of the same, with reduced concentration at the site of lesion. Moreover, EVs are cleared in few days suggesting that, to maintain the therapeutic efficacy over time, several administrations are required.

[0010] Rak J, et al. (Safety reporting on implantation of autologous adipose tissue-derived stem cells with platelet-rich plasma into human articular joints. BMC Musculoskelet Disord. 2013; 14:337) disclose Adipose tissue derived stem-cells (ADSC) in a PRP gel and their use in local injection to articular joints of patients with orthopedic conditions. Yu H, et al. (Research progress in the use of mesenchymal stem cells and their derived exosomes in the treatment of osteoarthritis. Ageing Res Rev. 2022; 80:101684) resume the use of cells in treatment of osteoarthritis. ADSC in PRP are disclosed among the other treatments.

[0011] There is a strongly felt need for improved treatments that leverage the potential of gels loaded with biological material but increase their reproducibility, where the presence of cells described in the gels of the prior art intrinsically introduces unwanted elements of variability.

[0012] DESCRIPTION

[0013] The object of the present invention is a Platelet-Rich Plasma (PRP) derived fibrin gel loaded with adipose MSCs (ASC)-EVs for use in the treatment of cartilage lesions, preferably focal cartilage lesions. The treatment of focal cartilage lesions limits the onset of osteoarthritis (OA) and other joint pathologies.

[0014] BRIEF DRAWINGS DESCRIPTION

[0015] Figure 1: EVs size analysis from Nanoparticle tracking analysis (NTA) merged data of the supernatants of the five donors used in the study. Mean ± SEM are reported.

[0016] Figure 2: EVs release over time from stable PRP-derived fibrin gels. A) EVs released per mg of wet fibrin gel is more pronounced in the first 48 hours, followed by a reduced but constant exit between the second and the fourth week. Mean ± SEM, n = 10. ** for p-value < 0.01 and **** for p-value < 0.0001. Only significant differences are indicated. B) The cumulative EVs release over time is reported with continuous exit in the last 3 weeks leading to significant difference with respect to values at 48 hours. Mean ± SEM, n = 10. * for p-value < 0.05, ** for p-value < 0.01 and **** for p-value < 0.0001. Only significant (p-value < 0.05) differences are indicated.

[0017] Figure 3: Unstable PRP-derived fibrin gels increase EVs release with dissolution. The cumulative EVs release for representative unstable fibrin gels (one per starting week of dissolution) with respect to the mean of stable clots is shown.

[0018] Figure 4: EVs uptake in OA patients' chondrocytes. Mean fluorescence intensity (MFI) of the FITC channel detecting signal of fluorescent EVs is reported for the four different treatments. No statistically significant (p-value < 0.05) difference between purified EVs and EVs released from the loaded fibrin gel was detected. Mean ± SEM, n = 3, ns for not significant, *** for p-value < 0.001 and **** for p-value < 0.0001.

[0019] Figure 5: ASC-EVs, released of PRP-gel and released of PRP-ASC-EVs gel effect on OA chondrocytes transcriptional profile. A) OA-related gene up- or down-regulation in OA chondrocytes exposed to EVs, 48 hours' PRP-gel released (FGR) and 48 hours' PRP-ASC- EVs gel released (FGR-EV). B) The presence of ASC-EVs empowered the protective effect of FGR and reduced the increase of pro-inflammatory mediators. C) The PRP-ASC-EVs gel effect (fold change) was superior to the expected sum of the FGR and ASC-EV single effects calculated under the formula ASC-EV fold change * FGR fold change. When the ratio (obtained / expected) was < 1 a synergistic effect was present. Fold change mean values are shown (n = 5). For column A and B, italics is for p-value < 0.1 and bold for p-value < 0.05.

[0020] Figure 6: OA-patients chondrocytes morphology (IL1 P) with respect to the same cells treated with the release of PRP-ASC gels (GEL + ASC) or PRP-ASC-EVs (GEL + EVs). Three donors are shown.

[0021] DETAILED DESCRIPTION

[0022] The authors of the present invention have demonstrated how the vesicles obtained from adipose tissue (ASC-EVs), known in the art for their wide availability, have characteristics that make them suitable for the treatment of cartilage pathologies.

[0023] In particular, the authors of the present invention have surprisingly demonstrated that EVs can be purified from ASC without influencing their phenotypic characteristics. The miRNA profile associated with these ASC-EVs has been characterized, showing that said miRNAs are involved in the regulation of growth factors involved in cartilage damage, as well as cytokines / chemokines related to the same pathology. miRNAs from these EVs also target brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), which are involved in joint pain. Enzymes that degrade the cartilage matrix are other targets of the miRNAs associated with these EVs. These observations led to the selection, among all the potentially available EVs, of the ASC-EVs. To direct them to the place of action, and to keep them there for enough time to carry out their therapeutic action, said ASC-EVs are loaded into a gel. To gain advantage of using a formulation already approved for clinical use, fibrin gel has been selected.

[0024] The authors of the present invention surprisingly found that platelet-rich plasma (PRP) gel loaded with ASC-EVs (PRP-ASC-EVs gel) has advantageous effects on chondrocytes, with respect to PRP-gel per se.

[0025] The object of the present invention is therefore a fibrin gel which comprises ASC-EVs. The gel according to the present invention is named PRP-ASC-EVs gel.

[0026] A further object of the present invention is a method for obtaining said PRP-ASC-EVs gel. A further object of the present invention is said PRP-ASC-EVs gel for use in the treatment of cartilage lesions.

[0027] Said cartilage lesions are selected among the group comprising i) focal chondral lesions caused by traumatic event, malalignment, overload, juvenile pathologies (including but not limited to osteochondritis dissecant, Osgood-Schlatter disease); ii) diffuse lesions indicating an initial joint degeneration typical of early osteoarthritis.

[0028] In a preferred embodiment, said cartilage lesions are focal cartilage lesions such as joint surface degeneration and articular cartilage lesions that eventually result in OA. Said PRP-ASC-EVs gel is used according to the state-of-the-art surgical approach in cartilage pathologies. As an example, said PRP-ASC-EVs gel is surgically superimposed on a cartilage lesion. Said gel is a PRP gel comprising ASC-EVs. Said ASC-EVs are loaded in said PRP before jollification in a concentration of between 107and 1011particles / ml, preferably about 109or 1010particles / ml.

[0029] PRP-gel comprises naturally occurring plasma EVs. The PRP-ASC-EVs gel according to the present invention therefore comprises plasma EVs and ASC-EVs, wherein the added ASC-EVs are in a lower amount with respect to the naturally occurring plasma EVs. Preferably, making 100% the total of the EVs comprised in the plasma before jollification, said EVs are more than 55% plasma EVs, less than 45% ASC-EVs, or less than 40% ASC-EVs, preferably about 33% ASC-EVs. For example, from 1 ml of plasma the obtained PRP-ASC- EVs gel is 239 mg ± 16 and the PRP-gel is 110 mg ± 10 (mean ± SEM, n = 22, p-value < 0.0001).

[0030] Advantageously, the increase in the mass is not due to the added ASC-EVs, wherein their mass is negligible, but to the increase capability of a gel comprising ASC-EVs to incorporate water. This feature is advantageous in that to an equal volume of plasma correspond a larger volume of gel to be placed at site of lesion, therefore allowing to save plasma and patient's discomfort for obtainment of a clinically relevant product. In an embodiment, said PRP-ASC-EVs gel comprises at least one supplement selected from the group consisting of a cytotoxin or cell proliferation inhibiting compound, an osteogenic compound, a cartilage inducing compound, an antibiotic, an anesthetic, an antiinflammatory compound, a cardiovascular drug, and a steroid.

[0031] Said method for the preparation of the PRP-ASC-EVs gel comprises the following steps:

[0032] - Making available a sample of human blood platelets, thus obtaining a PRP;

[0033] - Making available a suspension of EVs obtained from ASC;

[0034] - Add said ASC-EVs suspension to said PRP;

[0035] - Add a calcium salt, preferably CaCE;

[0036] - Leave to polymerize, preferably for a minimum of 15 minutes to a maximum of 2 hours at 37°C, until jollification of a portion of PRP;

[0037] - collect said jellified portion which is the PRP gel comprising ASC-EVs.

[0038] To note, naturally said jollification process does not continue until complete jollification, leaving a liquid fraction which is discharged.

[0039] Notably, when adding ASC-EVs to PRP, the jellified portion is higher than the jellified portion obtained when using PRP per se.

[0040] This advantageously allows a reduced waste.

[0041] In one embodiment, said PRP is autologous. In one embodiment, said PRP is heterologous.

[0042] In one embodiment, said EVs are obtained from adipose tissue-derived MSCs according to the following method: - Making available a sample of adipose tissue;

[0043] - After enzymatic digestion of the adipose tissue and filtration with a 100 pm filter, cells are seeded at 103-105cells / cm2preferably at 104cells / cm2and ASCs selected for plastic adherence;

[0044] - cells are growth in complete medium, leaving them reaching 70 % - 90 % confluence;

[0045] - culture medium is removed, and serum free medium is added;

[0046] - after 24-72 h, preferably after 48 hours in serum free medium, EVs are collected via ultracentrifugation;

[0047] - EVs are supplemented to PRE at a concentration between 107e 1011cells / ml.

[0048] Cell passage number is a calculation of the number of times cells have been splitted. Each time cells are collected, the passage number increase by 1. Passage before first trypsinization and split is considered to be number 0, while after first trypsinization and split is considered to be number 1.

[0049] In a preferred embodiment, cells are used at passage 1, or 2 or 3, preferably at passage 1. Said ultracentrifugation is preferably carried out at 100,000 g for 3 hours at 4 °C.

[0050] In one embodiment, said PRP-ASC-EVs gel is for use in the treatment of cartilage lesions and eventually OA.

[0051] Conveniently, said gel, once reached the site of action via surgical methods known in the art, release naturally occurring EVs and ASC-EVs together with the other factors comprised into the same, therefore exerting beneficial effects overtime.

[0052] The authors of the present invention have surprisingly found that, among the ASC- EVs miRNA targets, there are several mRNAs coding for pathology -related cytokines / chemokines: interferon gamma (IFNG), tumour necrosis factor (TNF), interleukin 1 alpha / beta (IL1A / B), interleukin 6 (IL6) and interleukin 8 (IL8). Also, other genes encoding pro-inflammatory interleukins (IL13 / 11 / 15 / 18) and chemokines (CCL5 / 8 and CXCL12), alongside TNFSF11, a member of the TNF cytokine family, were among the targets of ASC-EVsmiRNAs. In addition, several mRNAs for growth factors involved with cartilage damage appeared as targets, with vascular endothelial growth factor A (VEGFA) being regulated by 14 miRNAs, transforming growth factor beta 1 / 2 (TGFB1 / 2), connective tissue growth factor (CTGF), hepatocyte growth factor (HGF) and fibroblast growth factor 1 / 2 (FGF1 / 2), all related to cartilage and subchondral bone sufferance. Notably, brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), both involved in joint pain similarly to VEGFA, were among the ASC-EV-miRNA targets. Last, a number of transcripts for cartilage matrix degrading enzymes resulted as validated interactors, encompassing several matrix metalloproteinases (MMP1 / 2 / 3 / 9 / 13 / 14), a disintegrin and metalloprotease (ADAM12 / 17), a disintegrin and metalloproteinase with thrombospondin motif (ADAMTS9), cathepsin K (CTSK) and serpine E 1 (SER-PINE1), and their activators like adenomatous polyposis coli (ARC), plasminogen activator urokinase (PLAU) and plasminogen activator tissue type (PLAT). ASC-EV-miRNA targets comprises also few molecules with protective features, such as the mRNAs for anti-inflammatory IL4, the cartilage protective erythropoietin (EPO) and insulin D.ssa Dorotea RIGAMONTI - 8 - like growth factor 1 / 2 (IGF1 / 2), alongside some tissue inhibitors of MMPs (TIMP1 / 2 / 3).

[0053] Regarding cartilage, the majority among the ASC-EV-miRNAs have a protective function. The most impactful players are miR-24-3p, miR-125b-5p and miR-193b-3p, with an overall weight for this category of 1.476 %. This value heavily dropped for the 7 identified destructive miRNAs, that stands at 0.388 %, for an al-most 4-fold reduction. Also, 3 miRNAs with dual function were identified, for a 0.211 % weight.

[0054] Concerning macrophages, an equal number of miRNAs (6) characterized the categories pro-Ml (pro-inflammatory) and pro-M2 (anti-inflammatory), albeit the protective (anti-inflammatory) players had a 3-fold higher amount (0.739% vs 0.254 %) mainly due to miR-24-3p. The targeted single factors and, at a more general level, the ASC-EV-miRNAs have a potential therapeutic feature for both damaged cartilage and inflammation resolution. This is unique to ASC-EVs, selected for this purpose.

[0055] In addition to the nucleic-acid related features making ASC-EVs the EVs of choice for use in treating OA pathologies, there are also physical features making them the EVs of choice to be incorporated into a gel. In fact, ASC-EVs are bigger than other EVs, as an example, they are bigger than plasma EVs. Therefore, once incorporated into the PRP gel, ASC-EVs are released more slowly, extending the duration of the therapeutic effect.

[0056] The authors of the present invention observed that the addition of ASC-EVs to PRP led to significantly heavier gels compared to both PRP alone (2.2-fold increase) and PRP supplemented with parenting ASCs (PRP-ASCs) (1.8-fold increase).

[0057] This unexpected outcome is advantageous, as it enables the formation of a larger gel volume from the same amount of plasma, reducing the required plasma volume and minimizing patient discomfort while still achieving a clinically relevant product.

[0058] Experimental analyses have also demonstrated that the release of PRP-ASC gels negatively impacts osteoarthritic chondrocytes by increasing cellular distress. This cellular stress can lead to compromised cell viability and impaired functional capacity of the chondrocytes, which are critical for maintaining cartilage integrity and promoting repair after damage. The observed cellular suffering suggests that the release of PRP-ASC gels may exacerbate pathological processes within the osteoarthritic microenvironment, reducing the ability to promote cartilage regeneration.

[0059] Moreover, the release of PRP-ASCs exhibits a reduced anti-inflammatory potential. This is evidenced by its limited ability to downregulate genes associated with osteoarthritis- related inflammation. Gene expression profiling revealed that the release of PRP-ASC gels affects a smaller subset of pro-inflammatory and catabolic genes, indicating a less effective modulation of the inflammatory pathways that drive disease progression. In contrast, the release of PRP-ASC-EVs gels demonstrates an unexpected and more favourable profile for therapeutic use in osteoarthritis with respect to the use of originating cells. Importantly, the release of PRP-ASC-EVs gels does not induce cellular suffering in osteoarthritic chondrocytes, thereby preserving cell viability and functional activity. This characteristic is crucial for enhancing the structural and metabolic balance of the cartilage tissue.

[0060] Furthermore, the release of PRP-ASC-EVs gels shows a stronger anti-inflammatory effect by significantly reducing the expression of a broader range of genes implicated in osteoarthritic pathology. This downregulation of inflammatory mediators suggests that it could effectively mitigate the inflammatory processes that contribute to cartilage breakdown and joint degeneration, and eventually promote regeneration.

[0061] The data reported in the experimental section clearly support the surprising advantages of PRP-ASC-EVs gels over PRP-ASCs gels for treating osteoarthritic conditions and cartilage damage.

[0062] EVs are a subset of the bioactive components released by ASCs, excluding various soluble factors present in the full secretome. According to the subtraction hypothesis, this would suggest a diminished, rather than enhanced, therapeutic effect. However, the distinct biological responses observed emphasize the importance of selecting therapeutic agents that not only minimize cellular stress but also actively counteract inflammatory pathways driving disease progression.

[0063] EXPERIMENTAL SECTION

[0064] Adipose Tissue Collection, ASCs Isolation and Expansion

[0065] Adipose waste material from the abdomen of five healthy female donors (31 ± 5 years old) undergoing aesthetic procedures was processed as previously reported (Lopa et al., Eur Cell Mater. 201427:298-311. doi: 10.22203 / ecm.v027a21). After 30 min digestion at 37°C with type I collagenase (Worthington Biochemical Co., Lake-wood, NJ, USA), tissues were filtered with a 100 pm membrane, centrifuged at 1000 x g for 5 min at room temperature and pellets suspended in ccMEM + 10% FBS before seeding at 5-10 x 103cells / cm2at 37 °C (5% CO2, 95% humidity).

[0066] Adipose MSCs (ASCs) were selected by plastic adherence and used after the first trypsinization (passage 1) for analyses.

[0067] Cartilage Collection, Chondrocytes Isolation and Expansion

[0068] Cartilage was obtained as waste material from 11 osteoarthritis (OA) (Kellgren Lawrence III-IV grade) patients (69 ± 11 years old, six males and five females) undergoing total knee arthroplasty. When chondrocytes were needed, cartilage was digested with 0.15% w / v type II collagenase (Worthington Biochemical, Lakewood, NJ, USA) at 37 °C for 22 h. Then, chondrocytes were isolated for plastic adherence and cultured in DMEM / F12 + 10% FBS and used at passage 1 / 2. OA phenotype in patients' chondrocytes was maintained adding 1 ng / ml Interleukin 1 (3.

[0069] PRP Collection

[0070] Fifty-eight PRP samples obtained with the Endoret® system (BTI, Vitoria, Alava, Spain) were collected from patients (mean age 53 ± 11 years, thirty -one males and twentyseven females) undergoing PRP-based regenerative orthopaedic procedures. Patients' blood and PRP cell content was assessed with a Sysmex XN-2000 haemocytometer (Sysmex, Kobe, Japan).

[0071] ASCs Characterization by Flow Ci / tometri /

[0072] ASCs at passage 1 were analyzed by flow cytometry with a CytoFLEX flow cytometer (Beckman Coulter, Fullerton, CA, USA), collecting at least 30,000 events. Cells were stained for 30 minutes at 4 °C with the following antibodies: anti-CD90-FITC (REA897, Miltenyi Biotec, Bergisch Gladbach, Germany), CD73-PE (REA804, Miltenyi), CD105-PerCP-Cy5.5 (43A3, BioLegend, San Diego, CA, USA) and CD45-PE-Vio770 (REA747, Miltenyi) and analysed after one wash with FACS-buffer.

[0073] ASCs' Secretome Production and EVs Isolation

[0074] ASCs at 90% confluence were washed three times with PBS and fresh ccMEM (12 ml per T175 cell culture flask) without FBS was added. After 48 h, secretome was collected and centrifuged at 4 °C for 15 min at 1000 x g, 2000 x g and twice at 4000 x g to remove broken cells and debris. Secretomes were stored at -80 °C until further use. After secretome removal, ASCs were counted and viability was assessed with a NucleoCounter NC-3000 (ChemoMetec, Allerod, Denmark). Before EVs isolation, secretomes were pooled and the pool centrifuged at 100,000 x g for 3 h at 4°C in a 70Ti rotor (Beckman Coulter, Fullerton, CA, USA). EV pellets were washed with PBS and suspended in PBS or DMEM / F12 without serum. After count with a NanoSight LM10-HS system (NanoSight Ltd., Amesbury, UK), EVs concentration was adjusted to 10 x 1010particles / ml and 10 x 109EVs' aliquots frozen at -80 °C until use.

[0075] EVs Characterization

[0076] Flow cytometry: cleared secretomes were 1:6 diluted while purified EVs were 1:334 diluted with PBS and divided into 3 aliquots: (i) unstained, (ii) 5(6)- carboxyfluorescein- diacetate-succinimidyl-ester (CFDA-SE, Sigma-Aldrich, St. Louis, MO, USA)-stained (1 pM final concentration, 30 min at 37 °C), (iii) after CFDA-SE supplementation and incorporation leading to FITC-fluorescent carboxyfluorescein succinimidyl ester (CFSE), CD9-APC clone HI9A, CD63- APC clone H5C6, CD81-APC clone 5A6, CD73-APC clone AD2 and CD90- APC clone 5E10 (Biolegend, San Diego, CA, USA) stained (30 min at 4 °C). After a further 1:10 dilution with PBS, samples were analysed with a CytoFlex flow cytometer. At least 30,000 events were collected. FITC-fluorescent nanobeads of 100, 160, 200, 240, 300, 500 and 900 nm (Biocytex, Marseille, France) were used as internal control. Nanoparticle tracking analysis (NTA): cleared secretomes were 1:3 diluted while purified EVs were 1:167 diluted in PBS and visualized by Nanosight NS-300 system (NanoSight Ltd., Amesbury, UK) (5 recordings of 60 s). NTA software v3.4 provided both concentration measurements and high-resolution particle size distribution profiles.

[0077] EVs Multiplex ELISA Assay

[0078] The Quantibody® Human Cytokine Array 4000 Kit (https: / / www.raybiotech.com / quantibody-human-cytokine-array-4000 / , accessed on 30 September 2023) was used to detect two-hundred soluble receptors, chemokines, cytokines, growth and inflammatory factors in purified pooled EVs according to the manufacturers' protocol (RayBiotech, Norcross, GA, USA). A 1:1 dilution was performed and for each analyte and the mean of 4 technical replicates is reported. The amount of each factor in pg / m was converted into pg / x 109EVs by dividing the concentration value per ml with the number of EVs expressed in x 109per ml in the particle suspension used for the assay.

[0079] Protein-protein Interaction Networks

[0080] Interactome maps of ELISA-identified proteins were generated with the online tool STRING (http: / / www.string-db.org, database vll.5) with the following properties: (i) organism, Homo sapiens; (ii) meaning of network edges, evidence; (iii) active interaction sources, experiments, and databases; (iv) minimum required interaction scores, medium confidence (0.400).

[0081] EV-miRNA Quantification

[0082] TRIzol reagent was used to dissolve purified pooled EVs and RNA extracted with the miRNeasy Kit and RNeasy CleanUp Kit (Qiagen, Hilden, Germany), following the manufacturer's instruction. Six pg of a non-human synthetic miRNA (Arabidopsis thaliana ath-miR-159a) was added to each sample as a spike-in to monitor the technical variability during the isolation and following reactions for eventual equalization of panels A and B of the OpenArray® platform (Life Technologies). Standard reverse transcription and preamplification procedures with A and B independent kits were used to generate cDNAs. Real-time RT-PCR analysis with the QuantStudio™ 12 K Flex OpenArray® Platform (QS12KFlex) was then performed as previously described. miRNA expression data from the A and B panels, together covering 754 well-characterized human miRNA sequences from the Sanger miRBase v21, were processed with the Gene Expression Suite Software (Life Technologies). Due to the high correlation between samples, the global mean was selected as the normalization method for calls with a CRT < 28, as per manufacturer's protocol. miRNA expression was determined using the relative quantification 2-ACRT.

[0083] Preparation of Fluorescent EVs

[0084] Pooled secretomes were supplemented with CFDA-SE at a concentration of 10 pM and incubated for 1 hour at 37 °C in the dark. For EVs, purification, centrifugation protocol described above was used and 10 x 109 EVs' aliquots frozen at -80 °C until use. Fibrin Gel Preparation and EVs Incorporation

[0085] Before activation, PRP samples were divided into two 1 ml aliquots and each supplemented either with a suspension of ASC-EVs (10 x 109EVs per ml of PRE, 100 pl) or the equivalent volume of EVs' suspension solution (100 pl of DMEM / F12 or PBS) without EVs. PRP samples were then activated with 22.8 mM CaCb, 2 h at 37 °C to form fibrin gel. Supernatants and fibrin gels were collected and used immediately or stored at -80 °C for further analyses.

[0086] Scanning Electron Microscopy of Fibrin Gels

[0087] After clotting, supernatant was removed from fibrin gels with or without EVs that were fixed with NaCacodilate 0.1 M pH 7.2, 2.5 % glutaraldehyde and 3.7 % formaldehyde for 1 hour at RT. After fixation, samples were washed with distilled water ultra-pure for 30 min, dehydrated with an increasing scale of Ethanol (25, 50, 70, 80, 90, 95, 100%) and placed in a mixture of Ethanol and Hexamethyldisilazane (2:1, 1:1, 1:2) for 15 min each and finally transferred in Hexamethyldisilazane 100% overnight till completely evaporation. Dry samples were mounted on an aluminium stub, sputtered with platinum (Sputter ACE600, Leica Microsystems) and observed at a FESEM Sigma (Zeiss) operated a 5KV.

[0088] Visualization of Fluorescent EVs Incorporation into Fibrin Gel

[0089] For standard immunofluorescence analysis, after clotting formation the supernatant was removed and fibrin gels with or without EVs were fixed with 3.7 % formaldehyde overnight. Afterwards, fibrin gels were dehydrated in alcohol series and eventually embedded in paraffin and then cut into 4 pm sections for end-point immunofluorescence analysis. Fluorescent images were collected with an 1X71 Olympus inverted microscope (Olympus, Tokyo, Japan).

[0090] For multimodal microscopy, living and unprocessed samples were used. The technique allowed to collect images with Coherent Anti-Stokes Raman Scattering (CARS) for fibrin and matrix structures and Two-Photon Excitation Fluorescence (TPEF) for the EVs. The depth of EVs penetration together with the average occupied area and volume were computed using a custom-made ImageJ plugins processing the 3D images as already described. For image acquisition, in the microscopy setup CARS signal was targeted at the CH2 stretching mode (2848 cm-1 ), using a Nd:YVO4 laser emitting 10 ps pulses centered at 1064 nm (Picotrain, HighQLaser, Austria) as the Stokes field and the signal output of an Optical Parametric Oscillator (Levante Emerald OPO) (APE, Berlin, Germany) tuned at about 817 nm with a pulse width of 6 ps as the pump field. The laser repetition rate was 76 MHz. The two beams entered an upright microscope (BX51WI, Olympus, Tokyo, Japan) through the scanning unit (FluoView FV300, Olympus, Tokyo, Japan). The same OPO signal output at 817 nm was used as excitation source of the TPEF. The total average power at the sample of the excitation pulses was set to about 40 mW. The excitation beams were focused on living samples using a water immersion objective (LUMPLFLN 40XW NA = 0.8, W.D. = 3.3 mm, Olympus) that was cleaned and sterilized with a solution 70% ethanol in water (v / v) before the experiment. A condenser objective (UPLSAPO 10x objective NA = 0.4, Olympus) was used to collect forward CARS signal at about 663 nm that was optically filtered and then detected using a PMT (R3896, Hamamatsu, Japan). The TPEF signal was acquired in a back- scattering geometry (epi-detection).

[0091] Quantification of EVs Incorporation and Release

[0092] Fibrin gels were removed from clotting solution and weight was measured as well as supernatant volumes. For both empty gels and EVs-supplemented gels, a double volume of PBS or DMEM / F12 without serum was added with respect to the wet weight in mg (e.g., 200 pl volume for 100 mg gel weight). After 24 hours the supernatant was removed and replaced with the same volume of PBS or DMEM / F12. The same procedure was repeated at 48 hours and 1-2-3-4 weeks. Supernatants were analysed immediately for released EVs quantification or stored at -80°C for further use.

[0093] For EVs incorporation and release quantification, an aliquot of PRP before clotting without and with CFSE-labelled EVs, and aliquots of fibrin gel supernatant after clotting and fibrin gel supernatants after PBS or DMEM / F12 supplementation were analysed with CytoFlex flow cytometer. The same acquisition protocol used for EVs immunophenotype characterization was used as described at point 2.6. Samples were diluted in PBS to allow a total events / sec number below 2,000 with a flow rate of 10 pl / sec. Fluorescent EVs were identified as events falling in the FITC channel-gate obtained using PRP without EVs as signal background. To convert events with EVs number the following equation was followed: events / sec in the FITC gate * sec / pl * dilution factor * sample volume (pl) = events in the total volume. This number was then compared between samples to obtain the percentage with respect to input EVs (before clotting formation) and number of EVs in each sample for each time point was calculated considering EVs in PRP before clotting as 10 x 109.

[0094] Fibrin Gel Released EVs Incorporation into OA Chondrocytes in 2D Model

[0095] Chondrocytes at passage 1 from OA patients were seeded at a density of 10,000 cells / cm2. After 24 hours, 4 conditions were set: i) control with 1 ng / ml Interleukin 1 (3 to maintain OA phenotype and this condition with, ii) purified fluorescent EVs (25,000 particles per cell), iii) 48 hours supernatant of fibrin gels with released fluorescent EVs (25,000 particles per cell), iv) 48 hours supernatant of fibrin gels from the same patients without EVs. The equivalent number of EVs in the different conditions was measured by flow cytometry counting FITC-positive events per volume. All conditions were supplemented with 10% ultracentrifuged (100,000 x g, 9 hours, 4 °C) FBS to remove endogenous EVs. After 48 hours, chondrocytes were analysed by flow cytometry with a CytoFlex instrument (Beckman) to detect fluorescence in the FITC channel. At least 30,000 events were acquired. Mean fluorescence intensity (MFI) of the whole populations was recorded.

[0096] Fibrin Gel Released EVs Incorporation into OA Chondrocytes in Microfluidic Model

[0097] Under the same experimental conditions described for the 2D model above, chondrocytes from OA patients were cultured in a microfluidic chip. Briefly, before seeding in the microfluidic device, OA chondrocytes were stained with CellTracker™ Deep Red Dye (Invitrogen, Eugene, Oregon, USA) and suspended at 3 x 106 cells / ml in human thrombin (4 UI / mL diluted in 40 mM CaCL, Tisseel kit, Baxter, Stephenville, Texas, USA), mixed 1:1 with human fibrinogen (20 mg / mL diluted in PBS, Sigma- Aldrich, St. Louis, Missouri, USA), and injected in the microfluidic compartments. The chips were incubated at RT for 7 min to let the gels polymerize. Purified or fibrin gel released EVs were injected in the central channel and after 48 hours chips were fixed with 2% paraformaldehyde for 15 min at RT. Chips not injected with EVs were also fixed and used to define background fluorescence signals. Pictures were taken by a confocal microscope (Leica SP8) acquiring a z-stack of 150 pm corresponding to the entire channel height. Each sliced picture was taken at a distance of 1 pm. To evaluate the interaction between cells and EVs, image analysis was performed by ImageJ software. The macro script developed for a previous work with the same microfluidic devices to automatize the analysis process using as input two-channel z-stack images acquired by confocal microscopy was used.

[0098] EVs Release from Fibrin Gel and Incorporation into OA Cartilage

[0099] Cartilage waste material from OA patients was cut into pieces of 0.5 cm x 0.5 cm size, with a depth of approximately 0.5 to 1 mm and placed in 12-well plate. Fibrin gel without or with fluorescent EVs were cut into pieces of approximately 0.5 cm length and placed on top of the cartilage fragments. DMEM / F12 medium supplemented with 10 % FBS was added to each well to cover the cartilage / fibrin sandwich. Fibrin gels were incubated with cartilage for 48 hours before examination under the microscope as described above for CARS (here for cartilage matrix detection) and TPEF (EVs) signals.

[0100] Activity ofEVs-Eoaded Fibrin Gel Released on OA Patients Chondrocytes

[0101] Chondrocytes from OA patients at passage 1 were seeded as above described with a density of 30,000 cells / cm2and, after 24 hours, treatments were started. When EVs were added, either ultracentrifuge purified or included in the 48 hours fibrin gel released supernatants, 20,000 particles per cell were administered. The equivalent number of EVs in the administered treatments was measured by flow cytometry counting FITC-positive events per volume. All conditions were supplemented with 10% ultracentrifuged (100,000 x g, 9 hours, 4 °C) FBS to remove endogenous EVs. After 48 hours, chondrocytes were dissolved and RNA extracted with the miRNeasy Kit (Qiagen, Hilden, Germany), following the manufacturer's instruction. First-strand cDNAs were synthesized using the iScript cDNA synthesis kit (Bio-Rad Laboratories, CA, USA) and quantifications performed with iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories) in a CFX Opus 96 Real-Time PCR System (Bio-Rad Laboratories). TBP was used as a reference. Primer sequences will be provided upon request. mRNA expression was determined using the relative quantification 2- ACTanj foij change reported.

[0102] EVs, fibrin gel released, and EVs-loaded fibrin gel released effect on cell proliferation was measured with an identical experimental approach, with the difference that OA chondrocytes at passage 1 were seeded with a density of 10,000 cells / cm2. Cell proliferation was measured with the cell counting kit-8 (CCK-8) cellular proliferation assay (Sigma- Aldrich, St. Louis, MO, USA). At 48 hours, CCK-8 solution was administered for 1 h, followed by absorbance measurement of the cell supernatant at 450 nm using a Victor X3 microplate reader (PerkinElmer Life and Analytical Sciences, Shelton, CT, USA). A calibration curve was also prepared at time 0, associating CCK-8 absorbance readouts with those obtained with pre-determined cell numbers, counted with an automated cell counter (NucleoCounter® NC-3000™, ChemoMetec, Allerod, Denmark). Fold change for cells per cm2 with respect to OA chondrocytes at 48 hours is reported.

[0103] Statistical Analysis

[0104] Statistical analysis was performed using GraphPad Prism Software version 8.0.2 (GraphPad, San Diego, CA, US). Shapiro-Wilk normality test (a of 0.01) was used to test normal data distribution. For analysis of absolute values, for 2 conditions a two-tailed parametric t-test was performed. When the normality test was not performed, a two-tailed non-parametric Wilcoxon test was done. For more than two conditions, a repeated measures one-way ANOVA test was performed, with Tukey's post-hoc test. When the normality test was not passed, a non-parametric Friedman test was executed, with Dunn's post-hoc test. For fold changes and ratios, one-sample t-test was performed with hypothesis set as 1. For all tests, the level of significance was set at p-value < 0.05. Values are reported as mean ± SEM.

[0105] Example 1: ASCs and EVs Phenotype Characterization

[0106] ASCs resulted positive for mesenchymal markers (CD73, 99 % ± 0; CD90, 96 % ± 2; CD105, 67 % ± 8. Mean ± SEM, n = 5) and negative for haematological CD45 (0 % ± 0). ASCs released 5,319 ± 1,078 EVs per cell (mean ± SEM, n = 5) in 48 hours. Considering cell culture surface, 41 x 106 ± 5 EVs / cm2 were released. Cell viability was 97 % ± 1 after starvation. Nanoparticle tracking analysis (NT A) technology calculated for the EVs in the supernatants a mean size of 194 nm ± 6 and a mode size of 131 nm ± 6, with a D50 of 167 nm ± 5 (Figure 1). Particles were positive for the EV markers CD63 (92 % ± 1) and CD81 (95 % ± 0), while almost negative for CD9 (9 % ± 1). With respect to MSC markers that were scored as highly present in the secreting ASCs, CD73 and CD90 gave a strong positivity (90 % ± 0 and 89 % ± 1, respectively). After pooling the supernatants of the different donors and ultracentrifugation, EVs maintained similar size and immunophenotype, as shown in Table 1.

[0107] Table 1

[0108] Purified particles resulted again barely positive for CD9, and strongly positive for EV markers CD63 and CD81 and MSC markers CD73 and CD90. These results suggested that pooling and the purification process did not affect EVs features. Example 2: miRNAs Associated with EVs qRT-PCR assay showed that 344 miRNAs could be found associated with purified EVs (data not shown). Eighty-six players fell in the first quartile of expression and covered the 99.9 % of the investigated genetic message. Therefore, for the sake of simplicity, only this subset was further studied. The experimentally validated mRNA targets were selected for analysis and compared to OA regulators expressed in joint tissues and cells such as chondrocytes, synoviocytes, and different immune cells, including macrophages and T cells, as shown in Table 2.

[0109] Table 2: OA-related regulators targeted by first quartile EV-miRNAs

[0110]

[0111] Several pathology -related cytokines / chemokines emerged, encompassing interferon gamma (IFNG), tumour necrosis factor (TNF), interleukin 1 alpha / beta (IL1A / B), interleukin 6 (IL6) and interleu-kin 8 (IL8). Of note, IFNG, TNF and IL1B fell among the five most targeted molecules, with IFNG being at the top of the list. Also, other pro-inflammatory interleukins (IL13 / 11 / 15 / 18) and chemokines (CCL5 / 8 and CXCL12), alongside TNFSF11, a member of the TNF cytokine family, were target of EV-miRNAs. Several growth factors involved with cartilage damage appeared as targets, with vascular endothelial growth factor A (VEGFA) being regulated by 14 miRNAs and thereby having the highest total genetic pressure (1.657 %) among all analysed factors. In this category also transforming growth factor beta 1 / 2 (TGFB1 / 2), whose high levels may promote hypertrophy of cartilage, and connective tissue growth factor (CTGF), hepatocyte growth factor (HGF) and fibroblast growth factor 1 / 2 (FGF1 / 2), all related to cartilage and subchondral bone sufferance, fell. Notably, brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), both involved in joint pain similarly to VEGFA, were EV-miRNA targets. Last, a number of cartilage matrix degrading enzymes resulted as validated interactors, encompassing several matrix metalloproteinases (MMP1 / 2 / 3 / 9 / 13 / 14), a disintegrin and metalloprotease (ADAM12 / 17), a disintegrin and metalloproteinase with thrombospondin motif (ADAMTS9), cathepsin K (CTSK) and serpine E 1 (SERPINE1), and their activators like adenomatous polyposis coli (AEG), plasminogen activator urokinase (PLAU) and plasminogen activator tissue type (PLAT). EV-miRNAs also may target few molecules with protective features, such as the anti-inflammatory IL4, the cartilage protective erythropoietin (EPO) and insulin like growth factor 1 / 2 (IGF1 / 2), alongside some tiss ue inhibitors of MMPs (TTMP1 / 2 / 3).

[0112] Regarding cartilage, the majority (16 out of first quartile EV-miRNAs) have a protective function. The most impactful players are miR-24-3p (0.538 %), miR-125b-5p (0.323 % ) and miR-193b-3p (0.158 % ), with an overall weight for this category of 1.476 % . This value heavily dropped for the 7 identified destructive miRNAs, that stands at 0.388 %, for an almost 4-fold reduction. Also, 3 miRNAs with dual function were identified, for a 0.211 % weight. Concerning macrophages, an equal number of miRNAs (6) characterized the categories pro-Ml (pro-inflammatory) and pro-M2 (anti-inflammatory), albeit the protective players had a 3-fold higher amount (0.739 % vs 0.254 %) mainly due to miR-24-3p. These data are indicative of the fact that EV-miRNAs have a potential therapeutic feature for both damaged cartilage and inflammation resolution.

[0113] Example 3: Factors Associated with EVs ELISA assay showed that 163 factors could be found associated with purified EVs, data are shown in Table 3

[0114] Table 3: Factors associated with purified EVs.

[0115]

[0116]

[0117] Fifteen factors were detected with an amount > 1 ng / 109EVs, with Insulin-like growth factor-binding protein (IGFBP)-4 and IGFBP-3 being the most abundant ones at a concentration > 10 ng / 109EVs (44 and 20, respectively). In this group of abundant molecules, other two IGFBPs were found, namely 2 and 6. Other highly EVs-associated proteins were, among others, two Bone morphogenetic protein (BMP)-5 / 7, two Tissue inhibitor of metalloproteinase (TIMP)-l / 2, two growth factors like Transforming growth factor beta (TGFbl) and Fibroblast growth factor (FGF-4), and two receptors such as Interleukin 6 receptor (IE-6R) and Tumour Necrosis Factor Receptor 2 (TNF RII). Other 43 factors were detected between 1 and 0.1 ng / 109EVs and other 65 between 0.1 and 0.01 ng / 109EVs. Eventually, 40 proteins were barely detectable with an abundance < 0.01 ng / 109EVs. Factors present at a concentration > 0.1 ng / 109EVs were analysed by functional protein association network approach. Two main clusters emerged, encompassing factors with different functions. Notably, all factors of the smaller and tighter cluster and several of the second and loose one were involved in Inflammatory and Immune response. Also, several factors were related to Chemotaxis. Connecting these two areas, many of the identified proteins regulates the movement of immune cell types, including cells involved in joint inflammation and degeneration such as Lymphocytes, Monocytes and Neutrophils. Moreover, some of the factors in the second and loose cluster were involved in Cartilage and Connective tissue development, as also indicated by their ability to shape the Reactome Pathway "Extracellular matrix organization (HSA-1474244)". Altogether, alongside EV- miRNAs, proteins associated with purified EVs are connected with inflammatory and matrix-related processes, both dysregulated during joint diseases and cartilage damage.

[0118] Example 4: ASC-EVs Supplementation Affects Fibrin Gel Soaking with Plasma, Weight and Ultrastructure

[0119] To obtain wet clots (obtained from 1 ml PRP) without excessive soaking of residual plasma, fibrin gels were let leaking for 10 minutes until no more dripping liquid was visible. ASC-EVs addition led to a significantly higher released volume, being 127 pl ± 13 for control gels and 182 pl ± 14 for gels obtained after ASC-EVs addition (mean ± SEM, n = 21, p-value < 0.01). Moreover, resulting wet fibrin gels were heavier when loaded with ASC- EVs, wherein 1 ml PRP, when jellified, resulted in a weight of 110 mg ± 10 (no EVs) and 239 mg ± 16 (with EVs) (mean ± SEM, n = 22, p-value < 0.0001). The increased weight was due both to an augmented imbibition of liquid (40 pl ± 8 no EVs; 139 pl ± 13 with EVs; mean ± SEM, n = 22, p-value < 0.0001) and to a higher weight of fibrin gel matrix (70 mg ± 4 no EVs; 99 mg ± 9 with EVs; mean ± SEM, n = 22, p-value < 0.01). To shed light on the difference in fibrin gel weight and soaking capacity, wet clots were analysed by SEM (data not shown). The presence of fibrin fibrils was clearly visible in both gels. Intriguingly, PRP-gels showed a more regular structure with bundle of fibres whereas PRP- ASC-EVs gels have a more complex organization with a more wrinkled appearance. At higher magnification, fibrins in clots with EVs were more tightly packed with the formation of smaller and more narrow spaces and cavities. Thus, PRP- ASC-EVs gels result in a spongy structure with tighter fibrils and smaller holes justifying both the increased weight and the increased ability to retain plasma. Example 5: ASC-EVs Fully Embed Fibrin Gel Matrix and are released over time

[0120] Fluorescence was analysed in formaldehyde fixed and paraffin embedded slices encompassing both full-diameter (sagittal) and full length (frontal) planes of fibrin gels without and with fluorescent EVs (data not shown). ASC-EVs gave rise to a positive and homogenous signal along the whole length of the fibrin gel and across its section. To corroborate this result, PRE- ASC-EVs gels were analysed by multimodal microscopy that allows to collect images of fluorescence and matrix without fixation and processing of the samples (data not shown). ASC-EVs homogenously permeated the gels. Moreover, ASCEVs and fibrin signals were completely superimposable until the depth of one hundred microns that was the limit of detection of this technique inside the intact fibrin gel. Thus, ASC-EVs fully and homogeneously permeate gel structure. The amount of ASC-EVs incorporated into fibrin gels was 58 % ± 2 (mean ± SEM, n = 17) of the input (10 x 109EVs) added to 1 ml PRP before clot formation. Considering wet fibrin gel weight, 54 x 106± 6 ASC-EVs per mg were included. EVs release over time was monitored for a period up to 4 weeks. An important point to describe EVs release kinetics is that two main subgroups could be identified based on fibrin gel stability over time. The first one, composed of 10 fibrin gels, was characterized by high stability, with clots being solid for the whole time of analysis. The second one, including 7 fibrin gels, was composed of clots that started to dissolve spontaneously during the investigation period. In this group, clot stability was extremely variable, with some starting to dissolve after one week and others at 2, 3 or 4 weeks. Concerning the stable fibrin gels, amount, and kinetics of released EVs per mg of clot are reported in Figure 2A and 2B, showing both single time periods and cumulative data. In the first 24 hours, 6.10 % ± 0 .97 (mean ± SEM, n = 10) of incorporated ASC-EVs exited the clots, followed by a 1.01 % ± 0.18 in the second day. In the 3 to 7 days period, another 0.32 % ± 0.05 of embedded particles was delivered. In the next 3 weeks, a constant release (0.07 % ± 0.01 for week 2, 0.05 % ± 0.01 for week 3 and 0.07 % ± 0.01 for week 4) was monitored. Overall, during the 4 weeks of the analysis, 7.61 % ± 1.15 of incorporated ASC-EVs exited the fibrin gels, with 92.35 % ± 1.47 of this amount in the first 48 hours. The situation was different and donor -dependent for the other fibrin gels with reduced stability. For those gels starting to dissolve at 1 to 3 weeks, a small clot was visible after 28 days. For those ones dissolving at 4 weeks, clot size was similar to stable fibrin gels, although starting to be reduced in dimension. Of note, fibrin gel dissolution was concomitant with an increase in EVs release (Figure 3). Consistently, for gels that started to dissolve at 1 to 3 weeks, the cumulative EVs release was 25.75 % ± 2.57 (mean ± SEM, n = 4) of the incorporated particles, while for those initiating to reduce at 4 weeks, the value was 8.67 % ± 1.48 (n = 3). Thus, major EVs release is at first 48 hours and continues up to 4 weeks, with particles remaining incorporated in the gel that may be further freed with fibrin gel dissolution.

[0121] Example 6: Released EVs Interacts with Chondrocytes and Are Uptaken by Cartilage Since EVs release from both stable and unstable fibrin gels at 48 hours was comparable, particles collected after 2 days were used to test their conserved ability, with respect to purified ASC-EVs, to interact with OA patients' chondrocytes in both 2D and microfluidic 3D models and into OA patients cartilage using ex vivo explants. By flow cytometry, fluorescent EVs interaction with chondrocytes resulted similar between purified particles and fibrin gel released EVs, with no statistical (p-value < 0.05) difference in the mean fluorescent intensity (MFI) values in the FITC channel (Figure 4). Also, released of PRP-gel did not influence background fluorescence of chondrocytes. To confirm similar interaction properties, the same amount of purified ASC-EVs and ASC-EVs released from PRE- ASC-EVs gel were administered to OA patients' chondrocytes grown in a microfluidic model to mimic the 3D environment cells are embedded within in the cartilage (data not shown). Similar to 2D cultures, pictures of 3D cultures confirmed that after 48 hours EV fluorescence was clearly associated with OA chondrocytes. The analysis of the intensity of total EV fluorescence confirmed no significant difference in the two conditions (arbitrary MFI of chondrocytes with pure EVs vs chondrocytes with released of EVs-loaded fibrin gel: 18.2 ± 0.4 vs 19.1 ± 2.0, mean ± SEM, n = 3, p-value of 0.7644), corroborating EVs interaction also when cells are in a 3D environment and suggesting their ability to penetrate ECM matrix. Eventually, under these premises, ability to penetrate cartilage tissue was assessed by multimodal microscopy on ex vivo explants from OA patients (data not shown). PRP gels and PRP- ASC-EVs gels were cut into small pieces and placed on top of cartilage slices from OA patients to monitor the transfer of EVs from the clot to cartilage matrix. After 48 hours, clots were removed, and fluorescence directly analysed on cartilage samples without further processing. Cartilage that was placed under the fibrin gels showed a positive EVs signal up to a depth of > 200 pm. Overall, these results suggested that EVs are able to exit the fibrin gel, permeate the cartilage of OA patients and interact with pathologic chondrocytes.

[0122] Example 7: PRP and ASC-EVs synergistically act on OA Patients' Chondrocytes

[0123] OA patients' chondrocytes were examined for the modulation of expression of genes coding for OA-related cytokines / chemokines (Interleukin (IL)-l (3 / 6 / 8 / 11, Chemokine (C- C motif) ligand (CCL)-5 / 8, Tumor necrosis factor superfamily (TNFSF-11), growth factors (Fibroblast growth factor (FGF)-2, Growth differentiation factor (GDF)-9), enzymes involved in ECM degradation (Cathepsin S (CTSS), Matrix metalloproteinase (MMP)-l / 3) and enzymes regulating cell metabolic phenotype (Indoleamine 2,3-dioxygenase (IDO)-l, Prostaglandin-endoperoxide synthase 2 (COX-2) (Figure 5). Purified EVs were able to significantly (p-value < 0.05) reduce the amount of several mRNAs of all investigated categories. In particular, the genes for the major proinflammatory cytokines IL6 and IL8 and ECM degrading enzymes MMP1 and MMP3 emerged with a decrease between 30 and 60 % . IL1|3 had a tendency (p-value < 0.1) towards down-regulation (20 %). Fibrin gel released (FGR), mostly composed of PRP factors, had a different impact on chondrocytes. After treatment, IL1 (3 and IL11 resulted highly up regulated, 25 and 13 folds, respectively, as well as FGF2 (2.1 fold). On the contrary, among downregulated genes, IL6, CCL5 / 8, TNFSF11 and GDF9 showed reduced mRNAs (between 60 and 98 %). As for EVs, MMP1 and MMP3 reduced their expression at around 50 % of their levels. Of note, released of fibrin gels with EVs had the strongest effect on chondrocytes since all genes downregulated by either single EVs or fibrin gel resulted even more significantly reduced, while those genes upregulated by fibrin gel released had a contraction of their upregulation, reducing the possible pro- inflammatory effects. This was clearly shown by the direct comparison of fibrin gel released without and with EVs, where all analysed factors had a stronger downregulation in presence of EVs (Figure 5). Many factors resulted downregulated of a factor > 50 %, with the genes for the most OA relevant proteins ILI [3 / 8 and MMP1 / 3 falling in this group. Eventually, to understand whether the combined effect of fibrin gel released and EVs was additive or synergistic, we compared the results of the combination with those expected considering a sum of the single outcomes as per the formula = EVs fold change * fibrin gel released fold change. Notably, except for IL6, the gene expression reduction of the combination was lower than predicted, with an average for all genes of 0.69 ± 0.17 (p-value < 0.0001 with respect to hypothesis set as 1), suggesting that the two products together have a synergistic effect possibly stimulating separate pathways.

[0124] Example 8: ASC-EVs Supplementation Affects Fibrin Gel Weight with respect to ASC Supplementation

[0125] Adipose waste material from the abdomen of healthy female donors (37 ± 5 years old) undergoing aesthetic procedures was processed as previously described. Adipose MSCs (ASCs) were selected by plastic adherence and cultured in alpha-MEM + 10% FBS.

[0126] Five PRE samples obtained with the Endoret® system (BTI, Vitoria, Alava, Spain) were collected (mean age 54± 10 years, 3 males and 2 females) undergoing PRP-based regenerative orthopaedic procedures.

[0127] ASCs at 90% confluence were washed three times with PBS and fresh ccMEM (12 ml per T175 cell culture flask) without FBS was added. After 48 h, secretomes were collected, pooled and EVs isolated as previously described. Briefly, secretomes were centrifuged at 100,000 x g for 3 h at 4 °C in a 70Ti rotor (Beckman Coulter, Fullerton, CA, USA). EV pellets were washed with PBS and suspended in DMEM / F12 without serum. After count with a NanoSight LM10-HS system (NanoSight Ltd., Amesbury, UK), EVs concentration was adjusted to 10 x 1010particles / ml and 10 x 109EVs' aliquots frozen at -80° unhl use.

[0128] Before activation, PRP samples were divided into two 1 ml aliquots and each supplemented either with a suspension of ASC-EVs (10 x 109EVs per ml of PRP, 100 pl) or the equivalent number of ASCs (5 x 106ASCs per ml of PRP, 100 pl). PRP samples were then activated with 22.8 mM CaCL, 2 h at 37 °C to form fibrin gel. Fibrin gels were collected and used immediately. Fibrin gels were removed from clotting solution and weight was measured.

[0129] Wet fibrin gels containing ASC-EVs were significantly heavier than those with ASCs, exhibiting a weight ratio of 1.75 ± 0.11 (mean ± SEM, n = 5, p < 0.01). Example 9: The Released of Fibrin Gels Embedded with ASCs Negatively Impact Chondrocytes Morphology

[0130] With the aim to evaluate the effect of the gel comprising the cells (PRP-ASCs) or the EVs ASC derived (PRP-ASC-EVs), the two supplemented gels have been added to cartilage.

[0131] In particular, cartilage was obtained as waste material from three female osteoarthritis (OA) (Kellgren Lawrence III-IV grade) patients (70 ± 9 years old, six males and five females) undergoing total knee arthroplasty. Cartilage was digested with 0.15% w / v type II collagenase (Worthington Biochemical, Lakewood, NJ, USA) at 37 °C for 22 h. Then, chondrocytes were isolated for plastic adherence and cultured in DMEM / F12 + 10% FBS. OA phenotype in patients' chondrocytes was maintained adding 1 ng / ml Interleukin 1 (3.

[0132] For both ASCs- and ASC-EVs-supplemented gels, a double volume of DMEM / F12 without serum was added with respect to the wet weight in mg (e.g., 200 gl volume for 100 mg gel weight). After 48 hours the supernatant was removed and used immediately.

[0133] OA patients' chondrocytes were analysed for morphological changes following 48 hours of treatment with the release from either PRP-ASCs gels or PRP-ASC-EVs gels.

[0134] Representative pictures are reported in Figure 6. In PRP-ASC gels samples, it clearly emerged a reduction in cell number and signs of cell sufferance as granular cytoplasm. This phenotype is not observed in PRP-ASC-EVs samples where the typical morphology is maintained. Compared to control cells, chondrocytes treated with PRP-ASCs gels exhibited a more dispersed distribution and reduced intercellular connections, along with a granular cytoplasm. In contrast, chondrocytes exposed to PRP-ASC-EVs gels maintained a morphology similar to control cells, displaying an even distribution across the culture surface and numerous cell-to-cell connections, with no cytoplasmic signs of cellular distress.

[0135] Example 10: PRP-ASC-EVs have a superior healing potential than PRP-ASCs

[0136] After 48 hours of treatment with the release of PRP-ASC-EVs or PRP-ASCs, chondrocytes with ILlp (used as control when alone) were dissolved and RNA extracted with the miRNeasy Kit (Qiagen, Hilden, Germany), following the manufacturer's instruction. First-strand cDNAs were synthesized using the iScript cDNA synthesis kit (BioRad Laboratories, CA, USA) and quantifications performed with iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories) in a CFX Opus 96 Real-Time PCR System (Bio-Rad Laboratories). TBP was used as a reference. mRNA expression was determined using the relative quantification 2-AACTand fold change reported.

[0137] The expression of genes associated with OA-related cytokines (Interleukin-6 [IL-6]), chemokines (CCL-5 / 8), matrix metalloproteinases (MMP-1 / 3), and enzymes involved in cell metabolic regulation (Indoleamine 2,3-dioxygenase 1 [IDO-1] and Prostaglandinendoperoxide synthase 2 [COX-2]) was measured. Data are reported in Table 4. Treatment with PRP-ASC-EVs gels significantly reduced (p < 0.05) mRNA levels across all investigated gene categories. In contrast, PRP-ASCs gels led to a significant reduction in only a subset of these genes. Notably, PRP-ASC-EVs gels consistently outperformed PRP-ASCs gels, with stronger statistical significance in 3 out of 4 gene categories. Additionally, only PRP-ASC- EVs gels successfully downregulated the mRNA of the stress- and apoptosis-related gene cyclin-dependent kinase inhibitor 1A (CDKN1A).

[0138] Table 4: Gene expression evaluation in OA chondrocytes after treatment with the release of either PRP-ASCs gels or PRP-ASC-EVs gels.

[0139] Mean ratios are shown when significant. N = 3; ns for not significant; * for p-value < 0.05; ** < 0.01; *** < 0.001; **** < 0.0001.

[0140] Example 11: The release of PRP-ASCs gels impacts OA-chondrocytes viability

[0141] Due to the compromised phenotype of OA chondrocytes and the lack of CDKN1A gene downregulation following incubation with PRP-ASC gel release, cell viability was assessed. The PRP-ASC gel release showed a mild yet significant reduction in cell viability compared to PRP-ASC-EV gels (mean reduction of 5% ± 1, p < 0.05, n = 3), while PRP-ASC- EV gels exhibited viability comparable to control cells (p > 0.7).

Claims

CLAIMS1. A fibrin gel which is a platelet-rich plasma (PRP) gel comprising vesicles (EVs) from adipose tissue mesenchymal stromal cells (ASCs) (PRP-ASC-EVs).

2. The fibrin gel according to claim 1, wherein said ASC-EVs are added to PRP before jollification in an amount comprised between 107and 1011particles / ml, preferably about 109or 1010particles / ml.

3. The fibrin gel according to claim 1 or 2, wherein in said gel ASC-EVS are in a lower amount with respect to the naturally occurring plasma EVs.

4. The fibrin gel according to claims 1-3, further comprising at least one supplement selected from the group consisting of a cytotoxin or cell proliferation inhibiting compound, an osteogenic compound, a cartilage inducing compound, an antibiotic, an anesthetic, an anti-inflammatory compound, a cardiovascular drug, a steroid.

5. The fibrin gel according to any one of the claims 1-4 for use in the treatment of cartilage lesion and / or early osteoarthritis.

6. The fibrin gel for use according to claim 5, wherein said cartilage lesions are focal cartilage lesions.

7. A method for the obtainment of a PRP gel comprising ASC-EV s, said method comprising:- making available a sample of human blood platelets, thus obtaining a PRP;- making available a suspension of EVs obtained from ASC;- add said ASC-EVs suspension to said PRP;- add a calcium salt, preferably CaCE;- leave to polymerize, until jollification of a portion of PRP;- collect said jellified portion which is the PRP gel comprising ASC-EVs.

8. The method according to claim 7, wherein said polymerization occurs for a time comprised between 15 minutes and 2 hours, preferably at 37°C.

9. The method according to claim 7, wherein said PRP is autologous.

10. The method according to claim 7, wherein said EVs are supplemented to PRE at a concentration between 107and 1011preferably about 109or 1010EVs / ml.

11. The method according to claims 7-10, said EVs being obtained from adipose tissue derived MSCs according to the following method: - making available a sample of adipose tissue;- after enzymatic digestion of said adipose tissue and filtration with a 100 pm filter, cells are seeded at 103-105cells / cm2, preferably at 104cells / cm2and ASCs selected for plastic adherence;- cells are growth in complete medium at 70 % - 90 % confluence; - culture medium is removed, and serum free medium is added; after 24-72 h, preferably after 48 hours in serum free medium, EVs are collected by ultracentrifugation.

Images