Induction of iPSC-derived Prg4-positive cells with characteristics of superficial zone chondrocytes and fibroblast-like synovial cells

Lubricin, a proteoglycan encoded by the PRG4 gene, is synthesised by superficial zone (SFZ) chondrocytes and synovial cells. It reduces friction between joints and allows smooth sliding of tendons. Although lubricin has been shown to be effective against osteoarthritis and synovitis in animals, its clinical application remains untested. In this study, we aimed to induce lubricin-expressing cells from pluripotent stem cells (iPSCs) and applied them locally via cell transplantation. To generate iPSCs, OCT3/4, SOX2, KLF4, and L-MYC were transduced into fibroblasts derived from Prg4-mRFP1 transgenic mice. We established a protocol for the differentiation of iPSC-derived Prg4-mRFP1-positive cells and characterised their mRNA expression profile. Finally, we injected Prg4-mRFP1-positive cells into the paratenon, surrounding the Achilles tendons and knee joints of severe combined immunodeficient mice and assessed lubricin expression. Wnt3a, activin A, TGF-β1, and bFGF were applied to induce the differentiation of iPSC-derived Prg4-mRFP1-positive cells. Markers related to SFZ chondrocytes and fibroblast-like synovial cells (FLSs) were expressed during differentiation. RNA-sequencing indicated that iPSC-derived Prg4-mRFP1-positive cells manifested expression profiles typical of SFZ chondrocytes and FLSs. Transplanted iPSC-derived Prg4-mRFP1-positive cells survived around the Achilles tendons and in knee joints. The present study describes a protocol for the differentiation of iPSC-derived Prg4-positive cells with characteristics of SFZ chondrocytes and FLSs. Transplantation of lubricin-expressing cells offers promise as a therapy against arthritis and synovitis.

It diminishes the friction between joints and allows for smooth sliding of tendons [5][6][7][8][9]. Osteoarthritis causes increased chondrocyte destruction, joint friction, and synovial inflammation. When articular cartilage is destroyed in a localised superficial area, chondrocytes are replaced by fibrous cartilage tissue. When the same happens in a large and deep area, chondrocytes do not reproduce and the subchondral bone becomes exposed. Tenosynovitis is a painful condition, which causes increased synovial inflammation and adhesion of the Page 2 of 13 Satake et al. BMC Molecular and Cell Biology (2022) 23:30 tendon. Subsequently, prolonged tenosynovitis causes scarring of the tendon, which results in loss of its smooth sliding.
In the early stage of osteoarthritis, with only synovial inflammation, symptoms can be alleviated via intra-articular injection of hyaluronic acid or synovectomy. Similarly, mild tenosynovitis can be alleviated via injection of steroids in the tendon sheath, while in the case of deep cartilage defect in a limited area, autologous cartilage transplantation can be performed [10,11]. An alternative therapy is cartilage tissue regeneration from mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), and induced pluripotent stem cells (iPSCs) [12][13][14][15][16][17][18]. Owing to their pluripotency and versatility, iPSCs have been applied in regenerative medicine for several tissue and organs, including cartilage and tendon [19][20][21][22].
The therapeutic effect of lubricin administration has also garnered increasing interest. Intra-articular injections of partial recombinant human lubricin (rhPRG4) prevented cartilage degeneration in a rat model of osteoarthritis [6], and full-length rhPRG4 mitigated cartilage damage following destabilisation of the medial meniscus in a minipig model [23]. In addition to lubricating joints and tendons, rhPRG4 exhibited anti-inflammatory effect by inhibiting cytokine (interleukin-1β and tumour necrosis factor-α) secretion from osteoarthritis fibroblasts and cytokine-induced proliferation of synovial cells [24,25]. However, according to the experimental results, repeated injections of rhPRG4 are likely required to obtain treatment effects. Moreover, rhPRG4-based therapy has not been applied clinically. Here, we hypothesised that a therapeutic strategy based on injecting iPSC-derived lubricin-expressing cells could provide an effective treatment for osteoarthritis and tenosynovitis. In this study, we aimed to induce lubricin-expressing cells from murine iPSCs with fluorescent Prg4 reporter system and confirm the engraftment of transplanted cells in vivo.

Teratoma formation and histological analysis
We injected 3 × 10 6 iPSCs in 200 μl of phosphate-buffered saline into the subcutaneous tissue of severe combined immunodeficient (SCID) mice (n = 4) (Charles River). Three weeks after injection, the mice were sacrificed and teratomas were dissected, fixed in 4% paraformaldehyde overnight, and embedded in paraffin. Semi-serial sections were stained with hematoxylin and eosin (H&E).

Induction of iPSC-derived Prg4-positive cells
Before starting the induction of Prg4-positive cells, iPSCs were cultured for 4 days in iPSC maintenance medium in a humidified atmosphere with 5% CO 2 at 37 ℃, until they became subconfluent. Initially, iPSCs were differentiated as embryoid bodies on 96-well plates (Nunclon Sphera; Thermo Scientific) in standard high-glucose DMEM containing 10% foetal bovine serum, penicillin, and streptomycin for 2 days. On day 2, the medium was changed to standard medium containing 9 ng/mL activin A (Peprotech) and 25 ng/mL Wnt3a (R&D) for mesoderm differentiation. On day 3, the medium was changed to standard medium containing 10 ng/mL basic fibroblast growth factor (bFGF; Wako), and the cells were cultured continuously for 2 days. On day 5, embryoid bodies were dissociated and reseeded as a monolayer on collagencoated 6-well plates, containing standard medium supplemented with 1% insulin-transferring-selenium (ITS; Gibco), 10 ng/mL transforming growth factor beta 1 (TGF-β1; Cell Signaling), and 10 ng/mL bFGF. Medium was changed every other day and differentiated cells were cultured until day 21.

Fluorescence-assisted cell sorting analysis
Prg4-mRFP1-positive cells were selected by fluorescenceactivated cell sorting (FACS) on a BD FACS Aria flow cytometer (BD Biosciences). We used ESCs-derived differentiated cells as a negative control. Data were analysed with DIVA software (BD Biosciences), which identified iPSC-derived Prg4-mRFP1-positive cells as having a higher PE-Texas Red signal than negative control ESCderived cells.

Quantitative real-time PCR
Total RNA was isolated with the RNeasy Plus Mini Kit (Qiagen) and reverse-transcribed into cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer's instructions. Quantitative real-time PCR (qRT-PCR) was performed using TB Green Premix Ex Taq II (TaKaRa). The expression of target genes was normalised to that of reference genes (Actb and Gapdh) and is presented as the mean ± standard deviation based on three technical replicates per n, and n = 3 independent experiments. Primers used for qRT-PCR are listed in Table 1.

Injection of iPSC-derived Prg4-mRFP1-positive cells in the paratenon surrounding the Achilles tendon and knee joint of SCID mice
We injected iPSC-derived Prg4-mRFP1-positive cells at 1 × 10 4 cells/10 μl in phosphate-buffered saline in the paratenon surrounding the Achilles tendons and knee joints of SCID mice (n = 10) (Charles River). Three days after injection, the mice were sacrificed, and the Achilles tendons and knee joints were dissected and fixed in 4% paraformaldehyde overnight.

Histological analysis
All tissues were fixed in 4% paraformaldehyde overnight at 4 ℃ and embedded in paraffin. The knee joint tissues of Prg4-mRFP1 transgenic mice and samples of SCID mice, which were injected with iPSC-derived Prg4-mRFP1positive cells, were fixed in 4% paraformaldehyde overnight at 4 ℃ and then decalcified in EDTA (G-Chelate Mild; NIPPON Genetics) for 2 weeks at 4 ℃.

Statistical analysis
Statistical analysis was performed using Graph-Pad Prism 5 software. One-way ANOVA with Turkey's post-test was used for statistical analysis. A P-value < 0.05 was considered statistically significant. Data are presented as the mean ± standard deviation based on three independent experiments.

Generation of Prg4-mRFP1 transgenic mice and iPSCs harbouring the Prg4-mRFP1 reporter system
To create the Prg4-mRFP1 transgene, we inserted the mRFP1 expression cassette, which contained a Kozak sequence and poly-adenylation signal, 3 kb downstream of the Prg4 promoter (Fig. 1a). The transgenes were injected in the pronuclei of fertilised eggs from C57BL/6 mice to generate Prg4-mRFP1 transgenic mice. To verify the successful construction of transgenic mice, mRFP1 fluorescence was detected around the Achilles and patellar tendons (Fig. 1b); whereas immunohistochemistry showed mRFP1 expression in tenocytes, tendon synovium, and SFZ chondrocytes of the articular cartilage (Fig. 1c). These results suggested that the Prg4-mRFP1 expression system was active in vivo and could be used to monitor the production of lubricin. Next, we generated iPSCs harbouring the Prg4-mRFP1 reporter system. iPSC-like cells were established from fibroblasts of Prg4-mRFP1 transgenic mice by retroviral transduction with OCT3/4, SOX2, KLF4, and L-MYC. Six iPSC-like clones were selected, and qRT-PCR revealed a comparable expression of pluripotency-related genes between iPSCs and ESCs (Fig. 2a). Clone iPSC (#2-9) was the most similar to ESCs (Fig. 2b) and was chosen for transplantation into the subcutaneous tissue of SCID mice. Three weeks after transplantation, teratomas comprising ectodermal, mesodermal, and endodermal tissues could be observed (Fig. 2c), confirming the pluripotency of the established iPSCs. Next, we established a protocol for the differentiation of iPSC-derived Prg4-positive cells (Fig. 3a). First, iPSCs (#2-9) were differentiated as embryoid bodies for 2 days. On day 2, activin A and Wnt3a were added to promote differentiation of the primitive streak [24,[27][28][29], followed by that of the paraxial mesoderm and somites [30,31]. Between day 3 and day 5, embryoid bodies were cultured in the presence of bFGF to promote cell proliferation and differentiation of the sclerotome. On day 5, embryoid bodies were dissociated and reseeded as a monolayer supplemented with ITS, TGF-β1, and bFGF. TGF-β1 and bFGF have been shown to promote the differentiation of SFZ chondrocytes and synovial cells from the joint interzone [30,[32][33][34][35][36][37]. On day 21, mRFP1 fluorescence was detected in iPSC-derived differentiated cells. The average of Prg4-mRFP1 positive cell-ratio was 2.2% (maximum 7.0%), confirming the successful generation of Prg4-positive cells from iPSCs ( Fig. 3b and c).

Prg4-mRFP1-positive cells by qRT-PCR
To investigate the differentiation state of iPSC-derived cells generated with our protocol, we analysed the expression of target mRNAs by qRT-PCR over time.

Prg4-mRFP1-positive cells
Having demonstrated the successful induction of Prg4-mRFP1-positive cells by our differentiation protocol, we proceeded with their further characterisation. Based on FACS, Prg4-mRFP1-positive cells constituted 4.9% (maximum 9.3%) of the cells differentiated from the iPSC (#2-9) clone on an average (Fig. 5a). RNA-sequencing of mRFP1-positive cells obtained by FACS revealed that iPSC-derived Prg4-mRFP1-positive cells lost the properties of pluripotent stem cells, mesoderm, and joint interzone; however, they showed the properties of SFZ chondrocytes and FLSs, but not those of deep zone chondrocytes (Fig. 5b).

Injection of iPSC-derived Prg4-mRFP1-positive cells in the paratenon around the Achilles tendons and knee joints of SCID mice
Finally, we injected iPSC-derived Prg4-mRFP1-positive cells obtained by FACS in the paratenon surrounding the Achilles tendons and knee joints of SCID mice. Immunohistochemical analysis showed that 3 days after injection, some iPSC-derived Prg4-mRFP1-positive cells were still present around the Achilles tendons and in the knee joint (Fig. 6). Notably, no teratomas were observed in SCID mice at 3 weeks after cell transplantation (data not shown). These results indicated that transplanted iPSC-derived Prg4-mRFP1-positive cells could produce lubricin in vivo for at least 3 days without the risk of tumour formation.

Discussion
Lubricin is believed to be an effective therapeutic agent for osteoarthritis and synovitis. However, clinical application of recombinant or synthesised lubricin has not been attempted yet. In this study, we successfully developed a protocol to induce lubricin-expressing cells from pluripotent stem cells and confirmed their differentiation into Prg4-expressing cells that shared the characteristics of SFZ chondrocytes and FLSs from the joint interzone.
Prg4 is expressed in SFZ chondrocytes, synovial cells, synovial fluid, tendons, and tendon sheath [3,4]. Such localisation was confirmed in our Prg4-mRFP1 transgenic mice. First, iPSCs were established by transducing ear tip mouse fibroblasts with all four Yamanaka factors [26]. Second, we devised a protocol for the differentiation of iPSC-derived Prg4-positive cells, focusing on differentiation into SFZ cells and FLSs, which formed the synovial joint rather than cartilage. To induce the primitive streak, and subsequent differentiation into paraxial mesoderm and somites, we followed previous reports on embryoid body formation in suspension or adhesion for 4-6 days [11,12,15]. To maximise differentiation, we chose embryoid body formation in suspension for 5 days. Based on known cytokines involved in this process [12,14,15], we selected Wnt3a, activin A, and bFGF to promote differentiation of the sclerotome as a precursor to the synovial joint. Although useful for induction of the primitive streak, bone morphogenetic protein (BMP) 4 was not applied in this case as it could not lead to differentiation of the paraxial mesoderm.
To confirm successful differentiation, we analysed mRNA expression of iPSC-derived differentiated cells over time using qRT-PCR. Wnt9a and Wnt16 are important in the joint interzone, for the development of synovial joints [41][42][43]. Wnt16 also supports the phenotype of SFZ progenitor cells and lubricin expression [56]. During differentiation, Wnt9a and Wnt16 showed significant upregulation on day 12, followed by a drop on day 21. This expression dynamics suggested that iPSCs differentiated in the joint interzone on day 12 and advanced to the next differentiation stage thereafter. Besides Prg4, other genes expressed in SFZ chondrocytes include Erg, Tnc, and Runx1 [44][45][46][47][48]. We confirmed increased expression of Prg4, Erg, and Tnc on day 21, but could hardly detect them on day 12. Cdh11, Col4a, Cd55, Vcam1, Icam1, and Cd248 are expressed in FLSs of synovial joints [49][50][51][52][53][54][55]. Here, SFZ chondrocyte markers Prg4, Erg, and Tnc exhibited a similar expression pattern as FLS markers Cdh11, Col4a, and Vcam1. These results suggest that iPSCs differentiated into Prg4-expressing cells via the joint interzone. Analysis of mRNA expression in ESC-derived differentiated cells produced a similar temporal pattern as in iPSCs, suggesting that our protocol was applicable to all pluripotent stem cells. RNAsequencing confirmed that iPSC-derived Prg4-positive cells (clone #2-9) were more similar to SFZ chondrocytes and FLSs than to deep zone chondrocytes, mesodermal cells, myogenic cells, and MLSs, suggesting that iPSCderived Prg4-positive cells shared the characteristics of both SFZ chondrocytes and FLSs in synovial joints. Indeed, SFZ chondrocytes and FLSs are representative lubricin-expressing cells, showing higher expression than deep chondrocytes, mesoderm cells, myogenic cells, and MLSs. Therefore, the results of this study are significant in that high lubricin-expressing cells with SFZ chondrocytes and FLS characters were induced.
Recent strategies in regenerative therapy favour the use of stem cells rather than autologous cartilage transplantation to replenish missing or damaged cartilage [10,11]. Examples of such an approach include the generation of articular cartilage using human ESCs [15], generation of scaffold-free hyaline cartilage from human iPSCs [12], generation of Col2a1-EGFP iPSCs for monitoring chondrogenic differentiation [13], the formation of stable human articular cartilage using human MSCs [16,17], cartilage repair using a scaffold-free construct derived from porcine synovial MSCs [18], and specification of chondrocytes and cartilage tissue from murine ESCs [14]. The present study aimed to provide lubrication for arthritic joints and tenosynovitis by transplanting Prg4-expressing cells rather than regenerating damaged cartilage tissue. Hence, we injected iPSC-derived Prg4-mRFP1-positive cells into the paratenon surrounding the Achilles tendon and knee joints of SCID mice and confirmed the cells' survival and lubricin expression in vivo. A recent study revealed that Prg4-lineage cells in the joint SFZ differentiated into articular chondrocytes [57], suggesting that injection of iPSCderived Prg4-expressing cells might also have the potential to regenerate articular cartilage. Major advantages of our approach are the potential to obtain a large number of Prg4-expressing cells by using iPSCs, that harbour high capacity of self-renewal. Regarding the cell source, we can use mesenchymal stem cells with Prg4-mRFP1 reporter system and develop differentiation protocol from MSC to Prg4-positive cells. Differentiation efficiency of MSCs to Prg4-positive cells could become better because usage of MSCs can skip the process of iPSC to mesenchymal lineage differentiation. Mature Prg4-expressing cells could also be available, however, it may be difficult to obtain and expand a large number of Prg4-expressing cells in clinical practice, especially in the elderly patients.
Finally, several hurdles need to be overcome prior to the clinical application of Prg4-expressing cells. First, disadvantage of our approach is the insufficient induction efficiency of Prg4-mRFP1 positive cells, with average 4.9% (maximum 9.3%). Also, using the same protocol, we previously induced tenocyte-like cells from Scx-EGFP iPSCs [58]. Tenocytes share the same developmental process through somite to sclerotome with chondrocytes and synovial cells. Given that Prg4-expression has been confirmed in tendons, iPSC-derived Prg4-positive cells in this study might contain tenocytes. Therefore, the differentiation protocol should be optimised to improve the induction efficiency and purity of Prg4-epressing cells presenting SFZ and FLS characteristics. Second, we need to prolong the survival of these cells after injection. Transplanted Prg4-mRFP1-positive cells were detected in the paratenon surrounding the Achilles tendons and knee joints, but only for up to 3 days. Use of scaffold materials or platelet-rich plasma might offer a physical means for improved temporal survival of iPSC-derived Prg4expressing cells [59,60]. Also, condition media containing cytokines, growth factors, and exosomes has been used for regenerative medicine and many studies have shown their efficacy [12]. Indeed, our previous study regarding Scx-EGFP expressing iPSC-derived tenocyte transplantation revealed that cell transplantation had dual role, a direct contribution for regenerative tendon and an indirect effect (paracrine effect) reducing scar formation [58]. Of note, paracrine effect was presumed to contribute more preferentially for regeneration, indicating that cytokines, growth factors, and exosomes released from induced cells have an important role in regeneration. Therefore, the condition media of Prg4-mRFP1 expressing cells could become an alternative regenerative agent against osteoarthritis and synovitis, and might also prolong the survival of Prg4-mRFP1 positive cell after transplantation. Moreover, transplantation of Prg4-expressing cells might be beneficial in terms of such paracrine effect other than lubricin compared to rhPRG4 administration therapy. Third, to confirm engraftment of iPSC-derived Prg4-expressing cells in vivo, we used SCID mice for the transplantation experiment to avoid immune rejection. However, transplantation for osteoarthritis or synovitis model could be required to confirm the treatment effect of iPSC-derived Prg4-expressing cells. We would need further experiments to reveal its therapeutical benefit in the future.

Conclusions
The present study describes the generation of murine iPSCs harbouring a Prg4-mRFP1 reporter system and the establishment of a protocol for the differentiation of pluripotent stem cells into Prg4-expressing cells exhibiting the characteristics of SFZ chondrocytes and FLSs. Injection of lubricin-producing cells in the joints and paratenon around the tendons could become a viable therapeutic approach for osteoarthritis and tenosynovitis.