rAAV TGF-β and FGF-2 Overexpression via pNaSS-Grafted PCL Films Stimulates the Reparative Activities of Human ACL Fibroblasts

Lesions in the human anterior cruciate ligament (ACL) are frequent, unsolved clinical issues due to the limited self-healing ability of the ACL and lack of treatments supporting full, durable ACL repair. Gene therapy guided through the use of biomaterials may steadily activate the processes of repair in sites of ACL injury. The goal of the present study was to test the hypothesis that functionalized poly(sodium styrene sulfonate)-grafted poly(ε-caprolactone) (pNaSS-grafted PCL) films can effectively deliver recombinant adeno-associated virus (rAAV) vectors as a means of overexpressing two reparative factors (transforming growth factor beta-TGF-β and basic fibroblast growth factor-FGF-2) in primary human ACL fibroblasts. Effective, durable rAAV reporter red fluorescent protein and candidate TGF-β and FGF-2 gene overexpression was achieved in the cells for at least 21 days, especially when pNaSS-grafted PCL films were used versus control conditions, such as ungrafted films and systems lacking vectors or films (between 1.8- and 5.2-fold differences), showing interactive regulation of growth factor production. The expression of TGF-β and FGF-2 from rAAV via PCL films safely enhanced extracellular matrix depositions of type-I/-III collagen, proteoglycans/decorin, and tenascin-C (between 1.4- and 4.5-fold differences) in the cells over time with increased levels of expression of the specific transcription factors Mohawk and scleraxis (between 1.7- and 3.7-fold differences) and without the activation of the inflammatory mediators IL-1β and TNF-α, most particularly with pNaSS-grafted PCL films relative to the controls. This work shows the value of combining rAAV gene therapy with functionalized PCL films to enhance ACL repair.


Introduction
Injuries in the anterior cruciate ligament (ACL), the ligament that provides essential knee stability, are highly prevalent in the human population, affecting more than 200,000 persons every year in the United States and with an incidence of 1/3000 in Europe only, with costs exceeding USD 7 billion per year [1,2], potentially resulting in osteoarthritis and disability [3,4].
The ACL is a dense, hierarchically organized tissue with a rich extracellular matrix (ECM) that is mostly composed of collagen fiber bundles comprised of type-I and also -III, -IV, -V, and -VI collagen and other structural components, such as proteoglycans/decorin, tenascin-C, fibronectin, elastin, and thrombospondin, that surround ECM-producing fibroblasts which are present at a hypocellular level in a cable-like structure [5]. As a result of its very low cell content, the ACL has a limited and slow intrinsic self-repair ability following injury that promotes the formation of scar tissue with poor mechanical strength [3]. While a variety of treatments are available to manage ACL lesions, including conservative regimens like immobilization and bracing, physiotherapy, and corticoid injection, and surgical

Effective, Sustained rAAV-Mediated Gene Expression in hACL Fibroblasts via PCL Film-Guided Vector Delivery
Monolayer cultures of primary human ACL (hACL) fibroblasts were first transduced using the reporter rAAV-RFP vector coated on pNaSS-grafted versus ungrafted PCL films to examine the potential of this biomaterial-assisted gene transfer method to genetically modify these cells over time in vitro relative to control conditions in the absence of vector coating or to film-free rAAV gene transfer. The data reveal the successful, sustained expression of RFP via rAAV from day 1 until at least day 21 (the longest time-point evaluated) regardless of the type of PCL film employed, with signals similar to those achieved when using a film-free rAAV-RFP solution ( Figure 1B) versus conditions in which rAAV-RFP was omitted ( Figure 1A).
to examine the potential of this biomaterial-assisted gene transfer method to genetically modify these cells over time in vitro relative to control conditions in the absence of vector coating or to film-free rAAV gene transfer. The data reveal the successful, sustained expression of RFP via rAAV from day 1 until at least day 21 (the longest time-point evaluated) regardless of the type of PCL film employed, with signals similar to those achieved when using a film-free rAAV-RFP solution ( Figure 1B) versus conditions in which rAAV-RFP was omitted ( Figure 1A). Monolayer cultures of primary hACL fibroblasts were then transduced using the candidate rAAV-hTGF-β and rAAV-hFGF-2 vectors coated on pNaSS-grafted versus ungrafted PCL films to monitor whether such a system was also capable of mediating the overexpression of TGF-β and FGF-2 in these cells over time in vitro relative to control conditions in the absence of vector coating. An estimation of the levels of TGF-β production via an ELISA showed a significant, optimal increase in growth factor synthesis in Figure 1. Detection of transgene (RFP) expression in hACL fibroblast cultures transduced using rAAV-RFP-coated PCL films. The PCL films were coated with rAAV-RFP (40 µL; 8 × 10 5 transgene copies) or left without a vector coating before they were applied to the cultures, and RFP expression was examined under fluorescence microscopy at the denoted time points, as described in the Materials and Methods section ((A): absence of vector; (B): rAAV-RFP gene delivery; top panels: fluorescent photographs; lower panels: corresponding light microscopy photographs; magnification ×10; scale bars: 300 µm; all representative data). Abbreviations: −/−-lack of vector and PCL film; −/NG-uncoated, ungrafted PCL films; −/G-uncoated, pNaSS-grafted PCL films; RFP/−-film-free rAAV-RFP; RFP/NG-rAAV-RFP-coated, ungrafted PCL films; RFP/G-rAAV-RFP-coated, pNaSS-grafted PCL films.

Effects of rAAV-Mediated TGF-β and FGF-2 Overexpression via PCL Film-Guided Vector Delivery on the Biological Activities of hACL Fibroblasts
Monolayer cultures of primary hACL fibroblasts were finally transduced using the candidate rAAV-hTGF-β and rAAV-hFGF-2 vectors coated on pNaSS-grafted versus ungrafted PCL films to evaluate the effects of TGF-β and FGF-2 overexpression on the biological activities of these cells over time in vitro relative to control conditions in the absence of a vector coating.

Figure 7.
Gene expression profiles of inflammatory mediators in hACL fibroblast cultures transduced with rAAV-coated PCL films, evaluated via real-time RT-PCR. The PCL films were coated with rAAV-hTGF-β or rAAV-hFGF-2 or left without a vector coating before they were applied to the cultures as described in Figures 2-6, and the gene expression profiles for IL-1β and TNF-α were measured after 21 days via real-time RT-PCR with GAPDH serving as a housekeeping gene, as described in the Materials and Methods section. Ct values were obtained for each target gene and for Figure 6. Gene expression profiles of transcription factors in hACL fibroblast cultures transduced with rAAV-coated PCL films, evaluated via real-time RT-PCR. The PCL films were coated with rAAV-hTGF-β or rAAV-hFGF-2 or left without a vector coating before they were applied to the cultures as described in Figures 2-5, and the gene expression profiles for Mohawk and scleraxis were measured after 21 days via real-time RT-PCR with GAPDH serving as a housekeeping gene, as described in the Materials and Methods section. Ct values were obtained for each target gene and for GAPDH as a control for normalization, and fold inductions (relative to the −/− condition) were measured using the 2 −∆∆Ct method. Abbreviations: −/−-lack of vector and PCL film; −/NG-uncoated, ungrafted PCL films; −/G-uncoated, pNaSS-grafted PCL films; TGF-β/−-film-free rAAV-hTGF-β; TGF-β/NG-rAAV-hTGF-β-coated, ungrafted PCL films; TGF-β/G-rAAV-hTGF-β-coated, pNaSSgrafted PCL films; FGF-2/−-film-free rAAV-hFGF-2; FGF-2/NG-rAAV-hFGF-2-coated, ungrafted PCL films; FGF-2/G-rAAV-hFGF-2-coated, pNaSS-grafted PCL films. Statistically significant versus the −/− condition (* p ≤ 0.05) (dots represent single replicates).

Discussion
The combination of gene therapy and the use of biocompatible materials is a promising tool for improving the processes of ACL repair upon the delivery of therapeutic gene sequences in a safe and controlled spatiotemporal manner that allows for future noninvasive translation in patients [9]. Among a variety of candidate factors, TGF-β and FGF-2 were selected here in light of their reparative properties [10][11][12] and tested for their competence in activating hACL fibroblasts upon their delivery via clinically adapted rAAV gene transfer vectors guided by the application of functionalized (pNaSS-grafted) PCL films [28,29].
The data first reveal that the PCL films were capable of successfully delivering and expressing reporter (RFP) rAAV vectors in hACL fibroblasts over extended periods of time (at least 21 days, the longest time-point examined), as noted with a film-free vector solution versus control conditions without a vector, a result which is in good agreement with the findings achieved when applying the current rAAV-RFP construct via similar PCL films to human bone marrow aspirates at a similar MOI [28], probably due to the effective release of a vector coating from such a biomaterial [28]. The data next demonstrate that the PCL films were also capable of successfully delivering and expressing the two candidate (TGF-β and FGF-2) rAAV vectors in hACL fibroblasts over extended periods of time (at least 21 days, the longest time-point examined), especially when applying pNaSS-grafted PCL films versus control conditions and when compared with film-free vector solutions, results that are in good agreement with previous findings achieved when applying the current rAAV-hTGF-β construct via similar PCL films to human bone marrow aspirates at a similar MOI [29], which again reflects the effectiveness of the vector coating released over time from this biomaterial [28] and extends our work using the current rAAV-hFGF-2 construct applied as a vector solution to hACL fibroblasts at a similar MOI [12]. Remarkably, an interactive regulation of growth factor production was noted here in hACL fibroblasts in concordance with previous work showing TGF-β/FGF-2 synergistic interactions in articular chondrocytes [30].
The results next show that the successful, durable overexpression of the two candidate (TGF-β, FGF-2) rAAV vectors enhanced the deposition of typical ECM compounds (type-I and -III collagen, proteoglycans/decorin, and tenascin-C) in hACL fibroblasts over extended periods of time (at least 21 days, the longest time-point examined) in a safe manner (95-100% cell viability), especially when applying pNaSS-grafted PCL films versus control conditions and when compared with film-free vector solutions, results which are in good agreement with our previous findings using the current rAAV-hTGF-β construct delivered via similar PCL films to human bone marrow aspirates [29] and extending our work using the current rAAV-hFGF-2 construct applied as a vector solution to hACL fibroblasts [12], which is overall concordant with the properties of the growth factors, especially the TGF-β1 isoform applied in this study [10,[31][32][33][34]. These effects were accompanied by and probably due to increased levels of specific ECM-inducing transcription factors, Mohawk and scleraxis [20,35], in the hACL fibroblasts over extended periods of time (at least 21 days), especially when applying pNaSS-grafted PCL films versus control conditions and when compared with film-free vector solutions, results which are in overall agreement with the properties of these growth factors [36,37] and which extend our work using the current rAAV-hFGF-2 construct applied as a vector solution to hACL fibroblasts [12]. It is equally important that the delivery of rAAV-hTGF-β and rAAV-hFGF-2 via the PCL films had no detrimental effects on inflammatory processes (the production of IL-1β and TNF-α) in the hACL fibroblasts over time (at least 21 days), in concordance with the properties of these growth factors [38][39][40] and with the protective effects of the PCL films, especially following pNaSS grafting [26].
The present study shows the potential value of delivering clinically suited rAAV vectors that code for TGF-β and FGF-2 via pNaSS-grafted PCL films as an off-the-shelf novel system for activating the processes of ACL repair. It remains to be seen whether a concomitant application of both factors via such a material might even be more beneficial to achieving this goal as a combination of rAAV vectors can be conveniently envisaged without interference [41]. There is ongoing work to test the current approach as a therapeutic platform in relevant animal models of ACL lesions in vivo [42] that might be more potent, safe, and durable than the use of recombinant factors to heal sites of ACL injury [43,44].

Isolation and Culture of Primary Human Anterior Cruciate Ligament (hACL) Fibroblasts
Human anterior cruciate ligament (hACL) fibroblasts were obtained from donors undergoing total knee arthroplasty (n = 3, age range 72-78 years) [12]. Only ligaments without tears or visible degenerative changes upon gross examination were used. The study was approved by the Ethics Committee of the Saarland Physicians Council (Ärztekammer des Saarlandes, reference number Bu291/20). All patients provided informed consent before their inclusion in the evaluation, which was performed according to the Helsinki Declaration. After the retrieval of ACL tissue, small tissue pieces (1-2 mm 2 ) were incubated in 0.2% (w/v) collagenase in Dulbecco's Modified Eagle's Medium (DMEM), 2% penicillinstreptomycin (pen-strep) for 24 h at 37 • C under 5% CO 2 [12] ( Figure 8A). The suspension was then centrifuged 5 min at 1500 rpm, and the clot was first resuspended in DMEM for centrifugation (5 min, 1500 rpm) and then in DMEM complemented with 10% fetal bovine serum (FBS) and 1% pen-strep [12]. The cells were maintained in T-75 flasks at 37 • C under 5% CO 2 until confluence was reached ( Figure 8A). The cells were then seeded at a density of 10 4 cells/well in a 24-well plate and maintained at 37 • C under 5% CO 2 for up to 21 days [12].

rAAV Immobilization on PCL Films
The PCL films were sterilized with 70% ethanol (10 min incubation) and washed with PBS prior to incubation with 0.002% poly-L-lysine overnight at 37 • C [28,29]. After the films were washed with PBS twice, the rAAV vectors (40 µL, 8 × 10 5 transgene copies, MOI = 80) were immobilized on the films for 2 h via dropping at 37 • C (rAAV-coated PCL films) [28,29] (Figure 8B). Some PCL films were prepared without the addition of rAAV vectors as controls ( Figure 8B). Controlled release studies were not performed here as we previously reported that all PCL films (grafted and ungrafted) employed herein effectively release rAAV over extended periods of time (at least 21 days) [28].

rAAV-Mediated Gene Transfer
rAAV-coated or uncoated (pNaSS-grafted and ungrafted) PCL films were placed on the bottoms of 24-well plates seeded with hACL fibroblasts (10 4 cells/well) and incubated for 2 h at 37 • C under 5% CO 2 . They were then incubated with DMEM supplemented with 10% FBS, and 1% pen-strep was added for an overnight incubation at 37 • C under 5% CO 2 ( Figure 8B). Some wells were used as control conditions by directly adding film-free rAAV vector solutions to the cells or left without rAAV vectors ( Figure 8B). The medium was replaced after 24 h and every two days for up to 21 days. When provided, the PCL films were left over the entire period of prolonged culture.

Detection of Transgene Expression
RFP expression was monitored under fluorescence microscopy using a 568 nm filter (Olympus CK41; Olympus, Hamburg, Germany) [28,47]. The expression of TGF-β and FGF-2 was monitored at the denoted time points using specific ELISAs according to the manufacturer's instructions [12,29,47,48]. A GENios spectrophotometer/fluorometer (Tecan, Crailsheim, Germany) was used for the measurements [12,29,47,48]. TGF-β and FGF-2 expression was also monitored by immunocytochemistry using specific primary antibodies, biotinylated secondary antibodies, and the ABC method with diaminobenzidine (DAB) as a chromogen for evaluation under light microscopy (Olympus BX45) [12,29,47,48]. To control for secondary immunoglobulins, samples were processed with the omission of the primary antibody.

Gene
Forward Primer Reverse Primer

Statistical Analysis
All experiments were repeated a minimum of three times (five times for the ELISA experiments) using all isolated hACL donor fibroblasts. A nonparametric one-way ANOVA was used for statistical analysis (except for biochemistry data, which were analyzed using Student's t-test), with * p ≤ 0.05, ** p ≤ 0.01, and *** p ≤ 0.001 considered statistically significant.