ALGINATE HYDROGEL-GUIDED RAAV-MEDIATED FGF-2 AND TGF-β DELIVERY AND OVEREXPRESSION STIMULATES THE BIOLOGICAL ACTIVITIES OF HUMAN MENISCAL FIBROCHONDROCYTES FOR MENISCUS REPAIR

Meniscus lesions are unsolved problems that impede the functions of stability, load bearing, and shock absorption in the knee joint and for which there is no satisfying therapeutic option to date. Gene therapy using clinically adapted recombinant adeno-associated virus (rAAV) vectors is a powerful tool to enhance meniscus repair, especially when applied in combination with tissue engineering strategies that may allow to counteract a possible neutralization and/or dissemination of the rAAV particles in the recipient. Here, we examined the ability of an alginate (AlgPH155) hydrogel to formulate and deliver rAAV vectors carrying the genes for the reparative fibroblast growth factor ( FGF-2 ) and transforming growth factor beta ( TGF-β ) as a means to trigger the biological activities in human meniscal fibrochondrocytes. The results show that effective rAAV-mediated FGF-2 and TGF-β overexpression via alginate (AlgPH155) hydrogel-guided vector administration equally led to enhanced levels of cell proliferation and of specific matrix deposition (proteoglycans, type-I/-III collagen) in the cells over an extended period of time (21 days, the latest time point evaluated) relative to the control treatments (hydrogel without vector or with a vector carrying the reporter lacZ gene). These effects were associated with an increase in the expression of the contractile alpha smooth muscle actin ( α -SMA) marker, a determinant of the meniscus response injury, and with decreased levels of pro-inflammatory markers (interleukin 1 beta—IL1 β , tumor necrosis factor alpha—TNF-α ). These findings show the potential of alginate hydrogel-guided rAAV-mediated gene therapy as a new, off-the-shelf therapeutic system for meniscus repair.

Gene therapy offers strong tools to manage such pathological changes in torn menisci via the application www.ecmjournal.orgEuropean Cells and Materials Vol.47 2024 (pages 1-14) DOI: 10.22203/eCM.v047a01 of gene sequences coding for reparative factors that may durably enhance meniscal cell proliferation and ECM deposition relative to short-lived recombinant molecules (Evans et al., 2005; Evans and Huard, 2015; Evans and Robbins, 1999; Huard et al., 2003; Lamsam et al., 1998; Madry et al., 2011; Martinek et al., 2000; Shen et al., 2005).Recombinant adeno-associated virus (rAAV) vectors are particularly well adapted to achieve these goals as these constructs are derived from a replication-defective, non-pathogenic human parvovirus (Atchison et al., 1965) capable of stably targeting primary meniscal fibrochondrocytes at very high efficiencies (up to 100 % for at least 4 weeks in vitro) (Arrigoni et al., 2021; Cucchiarini et al., 2009; Madry et al., 2004) even when the cells are embedded in their ECM (up to 75 % for 2 weeks in situ) (Cucchiarini et al., 2009).Such effects are probably due to the small size (20 nm) (Atchison et al., 1965) and long maintenance of rAAVs as stable episomal forms in their targets (months to years) (Xiao et al., 1996) that may avoid the risk of insertional mutagenesis inherent to integrative retro-/lentivial vectors (Goto et al., 2000) and in contrast to nonviral and adenoviral vectors with short-term functionality (Evans et al., 2005; Evans and Huard, 2015; Evans and Robbins, 1999; Huard et al., 2003; Lamsam et al., 1998; Lee et al., 2014; Madry et al., 2011; Martinek et al., 2000; Shen et al., 2005; Steinert et al., 2007), making currently rAAV the most suited gene transfer vehicles for clinical applications (Cucchiarini, 2016; Evans et al., 2009; Evans et al., 2013; Evans et al., 2018; Evans et al., 2006; Mease et al., 2010; Vrouwe et al., 2022).Nevertheless, a number of obstacles still impede the optimal use of rAAV in vivo, in particular the presence of neutralizing antibodies against the AAV capsid proteins in the human population (Abdul et al., 2023; Cottard et al., 2004; Mingozzi et al., 2013) and the possible dissemination of the vectors to unwanted, nontarget sites (Cucchiarini, 2016).A potent approach to circumvent these hurdles is to take advantage of tissue engineering procedures as a means to deliver rAAV via biocompatible materials adapted for meniscus repair (Bilgen et al., 2018; Cucchiarini, 2016; Cucchiarini et al., 2016; Grogan et al., 2020; Huard et al., 2003; Kluyskens et al., 2022; Kwon et al., 2019; Makris et al., 2011; Rey-Rico et al., 2017), based on the pioneering work from Brunger and colleagues who used a scaffold-mediated lentiviral transduction of TGF-β for functional cartilage tissue engineering (Brunger et al., 2014).Hydrogels provide optimal systems to achieve this goal due to their high water contents that make them consistent with the microenvironment of the meniscus and to their ability to deliver drugs and gene vectors in target tissues in a spatiotemporal manner (De Laporte and Shea, 2007; Jo and Tabata, 2015; Lombardo et al., 2021; Rey-Rico et al., 2017; Seidlits et al., 2013).
The aim of this study was to explore the value of using an alginate hydrogel formulating rAAV vectors (Diaz-Rodriguez et al., 2015) as a new, improved gene therapy platform for meniscus repair.Among the various factors reported for their therapeutic benefits in this tissue (connective tissue growth factor-CTGF; insulin-like growth factor I-IGF-I; hepatocyte growth factor-HGF) (Hidaka et al., 2002; Romanazzo et al., 2018; Zhang et al., 2015; Zhang et al., 2009a; Zhang et al., 2009b), we examined the ability of the fibroblast growth factor (FGF-2) (Lee et al., 2014) and of the transforming growth factor beta (TGF-β) (Goto et al., 2000; Steinert et al., 2007) to trigger the reparative activities of human meniscal fibrochondrocytes upon rAAVmediated gene transfer guided via alginate (AlgPH155) hydrogel administration based on our previous findings showing the potential of these two gene vectors when applied directly to these cells (Cucchiarini et al., 2009; Cucchiarini et al., 2015).

Study Design
Human meniscal fibrochondrocytes were prepared from human adult menisci of knee joints from patients undergoing total knee arthroplasty, placed in monolayer culture, and directly incubated with rAAV (candidate FGF-2 or TGF-β versus control lacZ or no vector)/alginate hydrogels for up to 21 days prior to performing the evaluations that included biochemical, immunohistological, and realtime RT-PCR analyses (Fig. 1).

Gene Transfer via the rAAV/alginate Hydrogels
Monolayer cultures of human meniscal fibrochondrocytes (10 4 cells/well in 48-well plates) were kept in growth medium for 24 h at 37 °C with 5 % CO 2 and the rAAV/alginate hydrogels were directly added to the cultures (multiplicity of infection, i.e., MOI = 160) that were kept in growth medium for up to 21 days with medium change every 3-4 days (Cucchiarini et al., 2009; Cucchiarini et al., 2015; Diaz-Rodriguez et al., 2015).Controls included hydrogels without rAAV vectors.Direct application of the vectors in their free form was not performed here as it was already reported in previous work (Cucchiarini et al., 2009; Cucchiarini et al., 2015).

Statistical Analysis
Data are expressed as mean ± standard deviation of separate experiments.Each condition was performed in duplicate in three independent experiments with all patientsʹ samples.Data were obtained by two individuals blinded with respect to the treatment groups.The t-test was employed with p ≤ 0.05 considered statistically significant.

Effects of rAAV-mediated FGF-2 and TGF-β Overexpression on The Biological Activities of Human Meniscal Fibrochondrocytes upon Alginate Hydrogel-guided Vector Delivery
The candidate FGF-2/AlgPH155 and TGFβ/AlgPH155 hydrogels were next tested for their respective ability to stimulate the biological activities (cell proliferation, matrix deposition, inflammation) in human meniscal fibrochondrocytes relative to the control conditions including the reporter lacZ/AlgPH155 hydrogel and hydrogels without rAAV vectors (-/AlgPH155).
These observations were further supported by the results of a real-time RT-PCR analysis showing significantly increased COL1A1 and COL3A1 expression compared with the control conditions (COL1A1: up to 9.4-and 7.1-fold difference with FGF-2/AlgPH155 and TGF-β/AlgPH155, respectively, versus -/AlgPH155 and lacZ/AlgPH155 on day 21, always p ≤ 0.001; COL3A1: up to 3.2-and 2.2-fold difference with FGF-2/AlgPH155   Cells in monolayer cultures were directly incubated with the independently generated rAAV (FGF-2 or TGF-β)/alginate hydrogels as described in Fig. 2 and kept in culture as described in the Materials and Methods to monitor the deposition of type-I, -III, -II, and -X collagen and the expression of α-SMA on day 21 by immunocytochemistry with specific primary antibodies, biotinylated secondary antibodies, and the ABC method with DAB as the chromogen as described in the Materials and Methods (magnification 4×; all representative data).Abbreviations: -/AlgPH155, hydrogel without rAAV vector; lacZ/AlgPH155, rAAV-lacZ/alginate hydrogel; FGF-2/AlgPH155, rAAV-hFGF-2/alginate hydrogel; TGF-β/AlgPH155, rAAV-hTGF-β/alginate hydrogel.
The data first show that the alginate (AlgPH155) hydrogel was capable of significantly and durably promoting the overexpression of the two candidate (FGF-2, TGF-β) genes in human meniscal fibrochondrocytes upon rAAVmediated delivery relative to control treatments (hydrogel without vector or with a reporter rAAV-lacZ vector) (up to 8.9-fold difference for at least 21 days, the longest time point evaluated), probably due to the ability of this hydrogel to support the effective, controlled release of this class of vectors (Diaz-Rodriguez et al., 2015).Overall, the levels of transgene (FGF-2, TGF-β) expression reached here in human meniscal fibrochondrocytes from rAAV via hydrogelguided vector delivery were higher and more prolonged than those achieved when directly applying the same vectors in their free form to the cells in the same culture conditions and at similar vector doses (~55 pg FGF-2/mL/24 h/10 4 cells for 21 days with FGF-2/AlgPH155 here versus 60 pg FGF-2/mL/24 h/10 4 cells for no longer than 10 days with free rAAV-hFGF-2; ~1,000 pg TGF-β/mL/24 h/10 4 cells for 21 days with TGF-β/AlgPH155 here versus only ~210 pg TGF-β/mL/24 h/10 4 cells for 21 days with free rAAV-hTGF-β) (Cucchiarini et al., 2009; Cucchiarini et al., 2015), again probably resulting from the vector controlled release capabilities of this hydrogel (Diaz-Rodriguez et al., 2015).
The results next demonstrate that the effective, durable overexpression of the two candidate (FGF-2, TGFβ) genes from rAAV via alginate hydrogel-guided vector delivery significantly and durably enhanced the levels of cell proliferation and of specific matrix (proteoglycan, type-I/-III collagen) deposition in human meniscal fibrochondrocytes relative to control treatments (hydrogel without vector or with a reporter rAAV-lacZ vector) (up to 1.2-fold difference for cell proliferation and up to 11.8-fold difference for matrix deposition, for at least 21 days, the longest time point evaluated), probably again due to the vector controlled release ability of this hydrogel (Diaz-Rodriguez et al., 2015), while no effects were noted on type-II/-X collagen expression.These observations are in good agreement with the properties of these growth factors applied either as recombinant agents (Adesida et al., 2006; Collier and Ghosh, 1995; Webber et al., 1988) or via less adapted vectors (nonviral, adenoviral, retroviral gene vehicles) (Goto et al., 2000; Lee et al., 2014; Steinert et al., 2007) and with our findings when directly applying the same vectors in their free form to such cells in similar conditions (Cucchiarini et al., 2009; Cucchiarini et al., 2015).Interestingly, compared with control treatments (hydrogel without vector or with a reporter rAAV-lacZ vector), the effects of FGF-2/AlgPH155 and TGF-β/AlgPH155 were associated with an increased expression of contractile α-SMA, a determinant of the meniscus response to injury (Lin et al., 2002) (up to 10.6-fold difference for at least 21 days, the longest time point evaluated) and with a decreased inflammatory response (IL-1β, TNF-α) in the specific conditions applied here (up to 1.6-fold difference for at least 21 days, the longest time point evaluated), concordant with previous observations (Cucchiarini et al., 2009; Cucchiarini et al., 2015; Kuo et al., 2019; Rabie et al., 2023; Zaleskas et al., 2001).Interestingly, the FGF-2/AlgPH155 and TGFβ/AlgPH155 systems were equally potent, in good agreement with previous findings when applying low, optimally safe doses of recombinant FGF-2 and TGF-β factors to these cells (Pangborn and Athanasiou, 2005) as achieved here.

Conclusions
The current study provides the proof of principle of the potential of therapeutic (FGF-2, TGF-β) rAAV gene vectors delivered via an alginate (AlgPH155) hydrogel as a convenient, off-the-shelf system to advantageously trigger the reparative activities in human meniscal fibrochondrocytes to enhance the processes of meniscus repair in future translational applications.Of note, this strategy was used to durably heal cartilage defects in minipigs over one year with a functional rAAV-hIGF-I/AlgPH55 hydrogel system (Maihöfer et al., 2021) from which rAAV is being released in a sustained, controlled manner by diffusion without degradation (Diaz-Rodriguez et al., 2015; Madrigal et al., 2019).To optimize therapeutic outcomes, it will be crucial to define the ideal vector dose provided in the hydrogel for applications in vivo by performing an extensive, comparative analysis in future work.This scaffold-guided gene therapy approach may provide less invasive options compared with the use of genetically modified cells encapsulated in an alginate hydrogel to achieve this goal (Collier and Ghosh, 1995; Cucchiarini et al., 2009; Lee et al., 2014; Rey-Rico et al., 2016; Zhang et al., 2009a).Work is ongoing to test the benefits of the present strategy to stimulate the reparative activities of human meniscal fibrochondrocytes in a three-dimensional (3D) environment (Liang et al., 2020) known to be adapted for the effective transduction and overexpression of rAAV vectors (3D pellet cultures) (Cucchiarini et al., 2011).A study is currently being performed to monitor the effective repair of meniscal lesions in experimental models ex vivo (Cucchiarini et al., 2009; www.ecmjournal.org European Cells and Materials Vol.47 2024(pages 1-14) DOI: 10.22203/eCM.v047a01 Cucchiarini et al., 2015; Rey-Rico et al., 2016) and in relevant animal models in vivo (Goto et al., 1999; Madry et al., 2004; Zhang et al., 2009a) with this approach and to test its value over a direct administration of the rAAV vectors in their free form.Such an approach may further be adapted for a safe delivery of multiple rAAV vectors without interference (Tao et al., 2016) to synergize and extend the benefits reported here with FGF-2 and TGF-β as noted with recombinant factors, with also a potential to mechanistically increase the integration of the material at repair sites (Ionescu et al., 2012).Other therapeutic gene treatments may be also envisaged like CTGF that induces the deposition of a fibrochondrocyte matrix (Romanazzo et al., 2018), provided that rAAV supports its effective overexpression and activities in the cells.Taken together, these results provide motivation to apply alginate hydrogel-guided gene therapy to treat human meniscal lesions.

Fig. 1 .
Fig.1.Study design.Human meniscal fibrochondrocytes were prepared from human adult menisci of knee joints from patients undergoing total knee arthroplasty and placed in monolayer culture as described in the Materials and Methods.The rAAV-hFGF-2 and rAAV-hTGF-β vectors were independently incubated in alginate to form rAAV (FGF-2 or TGF-β)/alginate hydrogels (macroscopic views) and added to the human meniscal fibrochondrocyte monolayer cultures for up to 21 days as described in the Materials and Methods.The obtained cultures were processed to perform biochemical, immunohistological, and real-time RT-PCR analyses as described in the Materials and Methods (created with BioRender.com).

Fig. 3 .
Fig.3.Detection of matrix deposition and markers in human meniscal fibrochondrocytes treated with the rAAV/alginate hydrogels.Cells in monolayer cultures were directly incubated with the independently generated rAAV (FGF-2 or TGF-β)/alginate hydrogels as described in Fig.2and kept in culture as described in the Materials and Methods to monitor the deposition of type-I, -III, -II, and -X collagen and the expression of α-SMA on day 21 by immunocytochemistry with specific primary antibodies, biotinylated secondary antibodies, and the ABC method with DAB as the chromogen as described in the Materials and Methods (magnification 4×; all representative data).Abbreviations: -/AlgPH155, hydrogel without rAAV vector; lacZ/AlgPH155, rAAV-lacZ/alginate hydrogel; FGF-2/AlgPH155, rAAV-hFGF-2/alginate hydrogel; TGF-β/AlgPH155, rAAV-hTGF-β/alginate hydrogel.