Skip to main content
NCBI home page
Arthritis Research & Therapy logoLink to Arthritis Research & Therapy
. 2025 Jul 4;27:134. doi: 10.1186/s13075-025-03590-z

IFNγ production during cell interactions distinguishes localized from diffuse pigmented villonodular synovitis and rheumatoid arthritis

Mélissa Noack 1, Pierre Miossec 1,✉
PMCID: PMC12232138  PMID: 40615912

Abstract

Background

Pigmented villonodular synovitis (PVNS) is a rare articular disease characterized by aggressive synovial proliferation, with localized or diffuse forms. PVNS shares features of an inflammatory disease such as rheumatoid arthritis (RA), including immune cell infiltrate. Thus, we aimed to evaluate PVNS synoviocyte response to inflammatory stimulation or cell interactions to better understand their role in pathophysiology. Results were compared with those in RA.

Methods

Synoviocytes were treated with pro-inflammatory cytokines, IL-17 and/or TNF. IL-6 and IL-8 production was evaluated by ELISA in culture supernatants after 48 h. Migratory capacity was evaluated by a cell scraping assay. Peripheral blood mononuclear cells (PBMC) from healthy donors were co-cultured with PVNS or RA synoviocytes during 48 h, in the presence or not of phytohemagglutinin (PHA). Cytokine production (IL-17, IL-6, IFN-γ, IL-10, IL-1β and TNF) was measured by ELISA.

Results

The addition of IL-17 and TNF stimulated IL-6 and IL-8 secretion by both PVNS and RA synoviocytes, with similar responses between PVNS and RA synoviocytes. The highest production of IL-6 and IL-8 was obtained with the combination of IL-17 + TNF. Diffuse PVNS synoviocytes were less potent to cover a scratch area than localized PVNS or RA synoviocytes (p < 0.05). Finally, responses to cell interactions were assessed using co-cultures between synoviocytes and activated immune cells. IL-17, IL-6, IFNγ, IL-10, IL-1β and TNF production was measured after 48 h. Cell interactions induced massive cytokine production, mainly in PHA activated condition. The source of stromal cells affected the secretion resulting from these interactions. Localized and diffuse PVNS synoviocytes induced more IL-17 than RA synoviocytes (p ≤ 0.01). Localized PVNS induced more IFNγ than both diffuse PVNS and RA synoviocytes (p ≤ 0.05). IL-10 production was negatively correlated with IFNγ secretion.

Conclusion

In conclusion, results show differences in synoviocyte profiles or in response to cell interactions depending on synoviocyte source, with changes in IFNγ / IL-10 balance associated with localized PVNS. These differences could be used to adapt the therapeutic strategy to each form of PVNS.

Keywords: Cell interactions, Stromal cells, IFNγ, Inflammation

Background

Pigmented villonodular synovitis (PVNS), also known as tenosynovial giant cell tumor, is a rare articular disease with a wide spectrum of clinical manifestations, including pain, inflammation and joint swelling leading to limited range of motion and joint destruction [1]. PVNS affects mainly young and middle-aged adults and is typically a monoarticular disease with an estimated annual incidence of 1.8 patients per million [1]. The main characteristic of PVNS is a massive proliferation of the synovium of joints, tendon sheaths and bursae. It is a locally aggressive lesion with features of a tumor. PVNS appears in two forms: localized, with a single nodule, and diffuse, with the entire synovium affected [2, 3], and both forms can be intra or extra-articular.

PVNS is characterized by a synovial hypertrophy resulting from an excessive synovial proliferation, notably of stromal cells, and an accumulation of different cell types including giant cells (GC) and mononuclear cells (MNC) [4, 5]. MNC are mainly CD68 + macrophages rich in hemosiderin and GC are rather osteoclasts with a CD68 + TRAP + and CTR + phenotype [6, 7]. This aggressive synovitis could lead to rapid joint degradation through cytokine production (TNF, IL-6, IL-1) that activate osteoclasts and stimulate MMP production, inducing cartilage and bone destruction [6, 8]. This is explained in part by an over-expression of CSF1 that leads to the recruitment of CSF1R positive cells, including GC and MNC [5, 9], and to cell proliferation [10–12].

Stromal cells, as synoviocytes, are key players in chronic inflammation, and their interest as therapeutic targets [13–15] makes them important cells to study in disease pathophysiology. Furthermore, stromal cells have a key role during cell-cell interactions [16] and the source of cells could influence responses to these interactions [16–18] which contribute to maintain inflammation.

Currently, the main option of treatment for PVNS is surgery but is associated with a high risk of recurrence [19]. New therapeutic strategies are emerging as targeted therapies with tyrosine kinase inhibitors or anti-CSF1/CSF1R monoclonal antibodies [1]. Local targeting of synoviocytes could also be of interest in PVNS. We have already shown that PVNS synoviocytes were less sensitive to cadmium(Cd)-induced cell death than RA synoviocytes [20].

In this context, we aimed to evaluate PVNS synoviocyte response to inflammatory stimulation or cell interactions by comparing them with RA synoviocytes. The objective was to better understand PVNS synoviocytes to consider their therapeutic targeting and the repercussions of inhibiting cell interactions in this pathology.

Methods

Samples

RA and PVNS synoviocytes were obtained from synovial tissue of RA and PVNS patients undergoing joint surgery and who fulfilled the American College of Rheumatology criteria for RA [21], 5 RA patients, 5 localized PVNS patients and 5 diffuse PVNS patients. Synovial tissue was minced into small pieces and then adhered in 6-well plates in Dulbecco’s modified Eagle’s medium (DMEM; Eurobio, Courtaboeuf, France) supplemented with 10% fetal bovine serum (FBS; Life Technologies, Carlsbad, USA), 2mM L-glutamine and 100U/ml penicillin/streptomycin. Cells were maintained at 37 °C in a humidified 5% carbon dioxide incubator and used between passages 4 to 9. PBMC from healthy donors were isolated by Ficoll-Hypaque (Eurobio) density-gradient centrifugation. The study protocol was approved by the institutional review board of the Hospices civils de Lyon, France (accession number AC-2016-272) and written informed consent was obtained from the patients.

Cell culture conditions

Synoviocytes were seeded overnight in 96-well plates at a density of 2 × 104 cells/well and were treated with 50 ng/ml of IL-17 A (IL-17; Eurobio, Courtaboeuf, France), 1 ng/ml of TNF (R&D Systems, Minneapolis, USA) alone or in combination. Concentrations were previously used to stimulate synoviocytes [22–24]. Supernatants were collected after 48 h of culture and cytokine level was measured.

Co-culture was initiated by seeding RA or PVNS synoviocytes overnight in 96-well plates at a density of 2 × 104 cells/well in RPMI 1640 medium (Eurobio) supplemented with 10% FBS, 2mM L-glutamine and 100U/ml penicillin/streptomycin (complete RPMI). The next day, PBMC (1 × 105 cells/well) were seeded in complete RPMI corresponding to 5:1 ratio, in the presence or absence of phytohemagglutinin (PHA, 5µg/ml). After 48 h, supernatants and PBMC were collected for analysis.

Cell scraping

Cells were seeded overnight in 12-well plates at a density of 150 × 103 cells/well. The next day, using 1000 tips, the cells were gently scraped and treated or not with IL-17 (50 ng/ml) and TNF (1 ng/ml). The wells were photographed at the same point at 0 h, 24 h, 48 h and 72 h to assess cell migration. Images were then analyzed using ImageJ software (National Institute of Health, Bethesda, MD, USA).

Enzyme-linked immunosorbent assays (ELISAs)

IL-17 A (IL-17), IL-6, IL-8, IFNγ, IL-10, IL-1β and TNF productions were evaluated from culture supernatants with commercially available human ELISA set, according to the manufacturer’s instructions (Diaclone, Besançon, France).

Statistical analysis

Statistical analyses were performed using Mann-Whitney (unpaired results) or Wilcoxon (paired results) tests. All analyses were performed with Graph Pad Prism 6 software. p values less than 0.05 were considered as significant.

Results

Synoviocytes from PVNS and RA responded similarly to the addition of pro-inflammatory cytokines

Firstly, we wanted to test the capacity of PVNS synoviocytes to respond to pro-inflammatory conditions. Pro-inflammatory cytokines activate RA synoviocytes and other cells with additive or synergistic effect when they are combined as IL-17 and TNF [24]. It has been previously demonstrated that IL-17 and TNF contribute to aggressive phenotype of RA synoviocytes and thus to RA progression [25]. In this context, we tested the response of PVNS versus RA synoviocytes to cytokines. Synoviocytes from PVNS and RA patients were cultured during 48 h in presence or not of IL-17  (50 ng/ml) and TNF (1 ng/ml), alone or in combination. The production of IL-6 and IL-8 was measured in supernatants at 48 h.

Cytokines alone induced a significant increase of IL-6 and IL-8 production compared to control, in both PVNS and RA synoviocytes (p = 0.002, Fig. 1). For IL-6, the highest effect was obtained with TNF in PVNS synoviocytes (130.3 ± 30.7 ng/ml vs. 37.5 ± 11.0 ng/ml for control; p = 0.002, Fig. 1) while it was with IL-17  in RA synoviocytes (138.8 ± 48.7 ng/ml vs. 44.2 ± 20.3 ng/ml for control; p = 0.002, Fig. 1). The combination of IL-17  and TNF induced an additive effect on IL-6 production in PVNS synoviocytes (p ≤ 0.01, Fig. 1) while it induced a synergistic effect in RA synoviocytes (p ≤ 0.01, Fig. 1). No significant difference was observed between PVNS and RA synoviocytes regarding IL-6 secretion in each condition.

Fig. 1.

Fig. 1

Open in a new tab

Effect of cytokine interactions on PVNS and RA synoviocytes. PVNS and RA synoviocytes were cultured during 48 h in presence or not of different cytokine treatments: IL-17  (50 ng/ml), or TNF (1 ng/ml), cytokine alone or in combination, IL-17  + TNF. The production of IL-6 and IL-8 was measured by ELISA in the supernatants after 48 h. *p ≤ 0.05, *compares with control, # compares PVNS vs. RA. Results are represented as min to max all points, n = 9

For IL-8, TNF induced a higher release than IL-17  compared to control, in both PVNS and RA synoviocytes (PVNS: 111.0 ± 4.1 ng/ml vs. 11.3 ± 1.3 ng/ml for control; RA: 174.8 ± 31.2 ng/ml vs. 33.6 ± 13.8 ng/ml for control; p = 0.002, Fig. 1). In contrast to IL-6, the combination of cytokines had similar effect in both PVNS and RA synoviocytes; IL-17  and TNF induced a synergistic effect (p ≤ 0.01, Fig. 1). Furthermore, IL-8 production was lower in PVNS synoviocytes compared to RA synoviocytes in almost all conditions (p ≤ 0.01) and a low heterogeneity in IL-8 production was also observed with PVNS synoviocytes compared with IL-6 production and RA synoviocytes.

In conclusion, the addition of pro-inflammatory cytokines stimulated IL-6 and IL-8 secretion by PVNS and RA synoviocytes. The highest production of IL-6 and IL-8 was obtained with the combination of IL-17 and TNF. RA synoviocytes secreted significantly more IL-8 than PVNS synoviocytes while IL-6 production was closely related.

Diffuse PVNS synoviocytes secreted more IL-6 and were less potent to migrate than localized PVNS or RA synoviocytes

We wanted to go further and see whether the response to inflammation, using IL-17 + TNF combination, was dependent on the form of PVNS, diffuse or localized. We analyzed the secretion of IL-6 and their migratory capacity.

Synoviocytes were cultured for 48 h in presence or not of IL-17 (50 ng/ml) and TNF (1 ng/ml) and IL-6 was measured in supernatants (Fig. 2A). Firstly, the stimulation with IL-17 and TNF increased IL-6 production compared to control, for each type of synoviocytes (localized PVNS: 169.4 ± 43.3 vs. 6.4 ± 0.7 ng/ml; diffuse PVNS: 453.1 ± 162.5 vs. 19.8 ± 5.0 ng/ml; RA: 269.3 ± 20.6 vs. 4.9 ± 1.8 ng/ml; p = 0.002; Fig. 2A). In control condition, diffuse PVNS synoviocytes produced significantly more IL-6 compared to localized PVNS and RA synoviocytes (19.8 ± 5.0 vs. 6.4 ± 0.7 ng/ml and 4.9 ± 1.8 ng/ml, respectively; p ≤ 0.005; Fig. 2A). In inflammatory conditions, the difference between diffuse PVNS synoviocytes and the other two did not reach statistical significance but the concentration of IL-6 remained higher than in localized PVNS and RA synoviocytes (453.1 ± 162.5 vs. 169.4 ± 43.3 ng/ml and 269.3 ± 20.6 ng/ml respectively; NS; Fig. 2A).

Fig. 2.

Fig. 2

Open in a new tab

Effect of IL-17 + TNF on the IL-6 secretion and cell migration capacity of localized and diffuse PVNS or RA synoviocytes. PVNS and RA synoviocytes were cultured in presence or not of IL-17  (50ng/ml) and TNF (1ng/ml) during 24 h, 48 h (A) or 72 h (B). In panel A, IL-6 concentration was measured by ELISA in culture supernatants after 48h. * p < 0.05; min to max all points; n = 9 (A). In panel B, cell migration was evaluated by cell scraping. A tip was used to scrap adherent cells and the starting area was marked out under the microscope to have a reference for the following photos. Photos of the area were taken at 0 h, 24 h, 48 h and 72 h and treated using ImageJ software. For area measurements, uncovered areas were traced manually, and the values given by the software. The migration percentage was calculated as follows: 100-[(measured area*100)/starting area]. * p < 0.05; min to max all points; n = 6 (B)

After testing the response to pro-inflammatory cytokines, we were interested in the migratory capacity of synoviocytes. PVNS and RA synoviocytes were cultured on 12-well plates overnight and then, using tips, cells were gently scraped. The tip was held upright, and a line was scraped across the adherent cells. The area was marked out under the microscope so that photos could always be taken in the same spot and the ability of the cells to cover the scraped area could be quantified. Synoviocytes were treated or not with IL-17  + TNF and the recovered space by growing cells was measured over time (24 h, 48 h and 72 h). The photos were treated using ImageJ software. For area measurements, uncovered areas were traced manually, and the values given by the software. The migration percentage was calculated as follows: 100-[(measured area*100)/starting area].

In control condition, the recovery of scraped area was significantly lower in diffuse PVNS synoviocytes than in localized PVNS or RA synoviocytes since 24 h (17.0 ± 2.7% vs. 25.5 ± 2.9% or 30.3 ± 6.4%, respectively; p < 0.05, Figs. 2B and 3). This difference remained significant at 48 h and 72 h (48 h: 49.5 ± 3.3% vs. 69.9 ± 6.8% or 87.2 ± 7.9, p = 0.01; 72 h: 91.3 ± 2.1% vs. 98.3 ± 0.7% or 98.5 ± 1.5, p < 0.004; Figs. 2B and 3). At 48 h, RA synoviocytes also recovered higher area than localized PVNS synoviocytes (87.2 ± 7.9% vs. 69.9 ± 6.8%, p = 0.03, Figs. 2B and 3). In presence of IL-17 + TNF, the significant difference between diffuse PVNS and localized PVNS or RA synoviocytes appeared only from 48 h (24 h: 18.7 ± 3.7% vs. 26.8 ± 4.0% or 22.5 ± 3.9%, NS; 48 h: 43.1 ± 4.9% vs. 66.5 ± 8.1% or 77.0 ± 8.8%, p ≤ 0.02; 72 h: 85.5 ± 3.5% vs. 96.6 ± 3.5% or 95.6 ± 2.3%, p ≤ 0.01, Figs. 2B and 3). Pro-inflammatory cytokines seemed to slightly decrease the speed of recovery of the scraped area, mainly for RA synoviocytes, but without reaching significance.

Fig. 3.

Fig. 3

Open in a new tab

Capacity of cell migration of PVNS and RA synoviocytes after cell scraping. Photos represented the area covered by growing cells after cell scraping and observed by optic microscopy

To conclude, synoviocytes from diffuse PVNS produced more IL-6 than synoviocytes from localized PVNS in both control and inflammatory conditions. Diffuse PVNS synoviocytes were less potent to cover a scratch area than localized PVNS or RA synoviocytes. Inflammatory conditions did not significantly affect migration. Localized PVNS and RA synoviocytes displayed a similar profile, which differed from diffuse PVNS synoviocytes.

Localized and diffuse PVNS synoviocytes were more potent inducers of IL-17 than RA synoviocytes while only localized PVNS synoviocytes were more potent inducers of IFNγ during cell-cell interactions

During chronic inflammation, cell interactions between blood-migrated immune cells and resident synoviocytes play a critical role in local cytokine production [16]. The next step was therefore to test the effect of cell interactions between synoviocytes and immune cells in the context of PVNS vs. RA. Diffuse and localized PVNS or RA synoviocytes were cultured with activated or non-activated PBMC for 48 h, at 1:5 ratio. To compare the impact of stromal cell source, PVNS and RA synoviocytes were co-cultured with the same PBMC for one experiment but several different PBMC were used.

As previously described [26, 27], cell interactions between activated-PBMC and stromal cells highly increased IL-17 production (Fig. 4). In this condition, the IL-17 secretion was significantly higher in co-culture with PVNS synoviocytes, localized or diffuse, than RA synoviocytes (193.0 ± 49.81 or 151.1 ± 51.0 pg/ml vs. 95.0 ± 32.01 pg/ml, p ≤ 0.005, Fig. 4). Localized PVNS synoviocytes even induced higher IL-17 production than diffuse PVNS synoviocytes (p = 0.04, Fig. 4).

Fig. 4.

Fig. 4

Open in a new tab

Effect of interaction between PVNS or RA synoviocytes and PBMC on pro-inflammatory cytokine production. Healthy PBMC were cultured alone or in co-culture with localized or diffuse PVNS or RA synoviocytes at 5:1 ratio for 48 h, in the presence or absence of PHA (5 µg/ml). IL-17 , IL-6, IFNγ, IL-10, IL1β and TNF production was measured in supernatants after 48 h, by ELISA. * p < 0.05; min to max all points; n = 10

For IL-6, cell interactions were sufficient to induce IL-6 production, in both PVNS, localized and diffuse, and RA conditions. No significant difference was observed between PVNS and RA (Fig. 4). PHA activation did not increase IL-6 secretion in PVNS co-cultures compared to control (localized PVNS: 321.2 ± 62.0 vs. 330.1 ± 68.5 ng/ml; diffuse PVNS: 251.2 ± 33.2 vs. 260.3 ± 54.5 ng/ml, NS, Fig. 4) while with RA synoviocytes a significant increase was observed (326.1 ± 50.9 vs. 248.0 ± 40.8 ng/ml, p = 0.02, Fig. 4).

For IFNγ, cell interactions in control slightly induced IFNγ production in both localized and diffuse PVNS and RA co-cultures compared to PBMC alone (237.3 ± 129.0 pg/ml and 638.4 ± 479.8 for localized and diffuse PVNS respectively and 223.3 ± 65.6 pg/ml for RA vs. 131.3 ± 58.3 pg/ml, NS; Fig. 4). As for IL-17, a high release was obtained in activated co-cultures. The more potent inducers of IFNγ were localized PVNS synoviocytes with an IFNγ concentration significantly higher than with diffuse PVNS and RA synoviocytes (18263.7 ± 5034.9 pg/ml vs. 8177.4 ± 2463.7 and 12440.9 ± 3765.6 pg/ml, respectively, p ≤ 0.03, Fig. 4).

For IL-10, in control condition, cell interactions induced its secretion, with a significant higher production with RA and diffuse PVNS synoviocytes compared to localized PVNS (690.5 ± 212.8 pg/ml and 702.2 ± 123.1 pg/ml vs. 346.9 ± 107.2 pg/ml, p ≤ 0.04, Fig. 4). In PHA condition, the IL-10 secretion was the opposite of that of IFNγ. The more potent inducers of IL-10 were diffuse PVNS synoviocytes followed by RA synoviocytes and then localized PVNS synoviocytes (1120.2 ± 171.0 pg/ml; 766.0 ± 168.2 pg/ml and 705.7 ± 121.1 pg/ml; p ≤ 0.009, Fig. 4). In addition, PHA activation increased IL-10 production in PVNS compared to control but not in RA co-cultures (p ≤ 0.003, Fig. 4).

For IL-1β, cell interactions induced a significant increase of IL-1β production compared to PBMC alone, in both PVNS and RA co-cultures in control (p ≤ 0.006, Fig. 4) and PHA condition (p = 0.001, Fig. 4). The production was higher in PHA than in control condition for all cell cultures (p ≤ 0.01, Fig. 4). Regarding the different co-cultures, only a significant difference appeared in the control between localized and diffuse PVNS (1143.5 ± 366.8 pg/ml vs. 1828.5 ± 436.6 pg/ml, p = 0.02, Fig. 4). Otherwise, despite a tendency to higher production with diffuse PVNS, there was no significant difference between the different synoviocytes, PVNS or RA.

For TNF, results were close to those of IL-10. In control condition, cell interactions induced IL-10 production compared to PBMC alone (p ≤ 0.02, Fig. 4), with no significant difference between PVNS and RA synoviocytes. However, there was a tendency to higher production with diffuse PVNS synoviocytes compared to localized PVNS or RA synoviocytes (261.0 ± 78.9 pg/ml vs. 148.8 ± 53.4 pg/ml or 169.2 ± 48.8 pg/ml respectively, NS, Fig. 4). In PHA condition, only diffuse PVNS synoviocytes induced higher TNF production than in PBMC alone (463.8 ± 115.1 pg/ml vs. 209.2 ± 30.0 pg/ml, p = 0.04, Fig. 4). The production with localized PVNS synoviocytes was like that of PBMC alone (212.6 ± 58.4 pg/ml vs. 209.2 ± 30.0 pg/ml, Fig. 4) and tended to decrease with RA synoviocytes (144.4 ± 42.7 pg/ml vs. 209.2 ± 30.0 pg/ml, Fig. 4).

To conclude, cell interactions induced cytokine production, mainly in activated condition. Origin of stromal cells affected the secretion resulting from these interactions. Localized and diffuse PVNS synoviocytes induced more IL-17 than RA synoviocytes while localized PVNS induced more IFNγ and less IL-10 than both diffuse PVNS and RA synoviocytes. In contrast, diffuse PVNS synoviocytes induced more IL-10, IL-1β and TNF than both localized PVNS and RA synoviocytes.

Discussion

PVNS is a rare articular disease sharing characteristics with RA, notably the hyperproliferation of the synovial tissue and joint destruction. Compared to RA, less is known about the pathogenesis of PVNS. Based on the role of synoviocytes in RA and their interest as therapeutic targets, we wanted to evaluate the response of PVNS synoviocytes to pro-inflammatory stimulation and cell interactions to better understand their role in pathophysiology. We showed that PVNS and RA synoviocytes responded similarly to pro-inflammatory cytokines, with a synergistic effect of IL-17 + TNF. PVNS was next divided into two clinical forms, localized or diffuse. Diffuse PVNS synoviocytes displayed the strongest production of IL-6 and the lowest capacity of migration compared to localized PVNS and RA synoviocytes, which displayed a similar profile in these experiments. Finally, as cell interactions play a critical role in inflammation, we compared co-cultures between PBMC and the three types of synoviocytes. Results were dependent on the produced cytokines. PVNS synoviocytes were more potent inducers of IL-17 than RA synoviocytes, and localized even more than diffuse PVNS cells. Localized PVNS synoviocytes were more potent inducers of IFNγ than diffuse PVNS or RA synoviocytes that was inversely correlated with IL-10 production. Diffuse PVNS synoviocytes were more potent inducers of TNF and IL-1β than localized PVNS or RA synoviocytes.

The two forms of PVNS exhibit different behavior. Localized forms are benign, with a well-delineated lesion, easier to control by surgery, while diffuse forms are more aggressive and destructive, associated with joint effusion and involve the large joints, mainly knee, and more difficult to control [28]. Despite these characteristics, our results showed a higher capacity to invade a scraped area for localized versus diffuse PVNS synoviocytes. Indeed, diffuse PVNS synoviocytes were slower to cover the scraped area than localized ones, which may seem surprising given the aggressive and destructive nature of diffuse PVNS. However, these results were consistent with a study showing higher proliferative activity in localized PVNS than diffuse PVNS synoviocytes [29]. These results also suggested a phenotypic difference in synoviocytes reflecting the different characteristics of the two forms of PVNS, as well as a different stage of disease progression. In the end, little is known about the progression of PVNS except for a slow progression suggested by the often-long interval between the first symptoms and diagnosis. It is therefore possible that synoviocytes of the localized form ultimately migrate to invade other areas, presenting a profile like RA synoviocytes, whereas synoviocytes of the diffuse form have already invaded these areas and are therefore less inclined to migrate. Then, diffuse synoviocytes showed a greater capacity to secrete IL–6 than localized or even RA synoviocytes. These results were consistent with the destructive and aggressive side of the diffuse form of PVNS, showing a higher inflammatory capacity of diffuse than localized synoviocytes. Based on these results, either the localized form with more proliferative synoviocytes or the diffuse form with more inflammatory synoviocytes, targeting synoviocytes seem to be a good strategy. However, the different behavior of both cell types, localized or diffuse, lead to consider a different approach to target these cells. Migration inhibition may be envisaged in the case of localized PVNS whereas inhibition of inflammation in the case of diffuse PVNS could be more appropriate.

Currently, surgery remains the main treatment for PVNS, but a high frequency of recurrence is observed. Targeting synoviocytes could be an interesting strategy to avoid surgery or systemic therapy. We showed that PVNS synoviocytes were sensitive to Cd-induced apoptosis [20]. An increased cell death associated with a decreased of cell viability has been observed in PVNS and RA synoviocytes treated with Cd [20]. This effect was dose-dependent and combined with a decrease of IL-6 production. To extend the results, PVNS synovial tissue was exposed to Cd and as in cell culture, a decrease of IL-6 secretion was observed. Furthermore, Cd effect was increased in presence of pro-inflammatory cytokines, allowed to reduce Cd dose [20]. This treatment has been extended in an in vivo RA model, adjuvant-induced arthritis (AIA) [30]. Intra-articular injection of Cd reduced synovitis and bone destruction in AIA. Another model to target synoviocytes hyperproliferation has been developed in RA through gene targeting. In AIA, the pro-apoptotic gene PUMA combined with an adenovirus-baculovirus complex vector induced apoptosis of synoviocytes leading to decrease in joint inflammation, joint damage, and bone loss [31]. The induction of apoptosis targeting directly synoviocytes could be an interesting therapeutic option, notably for PVNS where only a single joint is affected. The development of an animal model for PVNS would be a good step forward to test these therapeutic options.

Stromal cells, as synoviocytes, also have a key role during cell-cell interactions [16] and the source of cells could influence responses to these interactions [16–18]. Both diseases, PVNS and RA, show an infiltrate of immune cells that interact with synoviocytes. To reproduce these cell interactions in vitro, we realized co-cultures between PBMC and synoviocytes from localized or diffuse PVNS, or RA. These experiments showed the importance of the source of stromal cells. PVNS synoviocytes were more potent inducers of IL-17 production than RA synoviocytes, and even more localized than diffuse PVNS synoviocytes. For IL-6, although diffuse PVNS synoviocytes alone were the highest producers, in co-cultures, no significant difference was observed in co-cultures with the three types of synoviocytes. IFNγ production showed an even different profile. Localized PVNS synoviocytes were the most potent inducers of IFNγ, followed by RA and diffuse PVNS synoviocytes. As IL-10 inhibits IFNγ secretion [32], results of IFNγ were coherent with IL-10 production, which was the highest with diffuse PVNS, followed by RA and localized PVNS synoviocytes. Diffuse PVNS synoviocytes were also the most potent inducers of TNF and IL-1β. Cytokines secreted during cell-cell interactions therefore appear to be dependent on the stromal cell source. These results suggested different cytokine-inducing profiles depending on the pathology. A larger number of donors will be needed to validate these profiles, which could be used to guide therapeutic strategies, particularly by inhibiting the cytokine that is predominantly secreted during the course of a disease. Treatment could thus be adapted according to the form of PVNS.

These results therefore showed that different synoviocytes would be linked to profiles depending on the measured cytokine and the cell origin. In addition, these differences may be related to the nature of the immune infiltrate found in tissues at the level of the synovial membrane. In PVNS, a predominance of CD68 + macrophages was observed. The high production of IFNγ with localized PVNS synoviocytes found in co-cultures was in line with the presence and activation of macrophages. A study using diffuse PVNS samples has shown a high immune cell infiltration including CD4 + T cells, notably Th17 cells, and a local activation of osteoclastogenesis [33]. This could be correlated with our co-culture results, showing a high IL-17 production with diffuse PVNS synoviocytes. Furthermore, IFNγ could suppress the formation of osteoclasts [34–36]. Knowing that localized synoviocytes induced more IFNγ than diffuse synoviocytes, results were coherent with the non-destructive characteristic of localized PVNS. In addition, the lower level of IFNγ with diffuse synoviocytes and thus a lower suppression of osteoclast formation was coherent with destructive characteristic of diffuse PVNS. IFNγ can also increase T CD8 + cell infiltration present in synovial sarcoma [37, 38] and in RA, IFNγ-producing CD8 + T cells also produce IL-10 that can establish a negative feedback [39]. This mechanism could be extended to diffuse PVNS, regarding the higher level of IL-10 in co-cultures with diffuse PVNS synoviocytes compared to localized PVNS synoviocytes. Currently, studies of immune infiltrate in PVNS are mostly done in diffuse PVNS. It could be interesting to extend the identification of immune cells infiltrating diffuse and localized PVNS. The nature of the cells infiltrating the synovial membrane is therefore just as important as the origin of the stromal cells in response to cell interactions. It might now be interesting to co-culture with isolated cell types, monocytes/macrophages on one side and T lymphocytes on the other, to study in greater detail the differences induced by the three types of synoviocytes. These results could provide guidance for the development of specific therapeutic targets, appropriate to the pathology. Furthermore, based on current data, the infiltrate profile is closer between RA and diffuse PVNS, whereas the synoviocyte profiles are closer between RA and localized PVNS. The results obtained with RA synoviocytes appear to be somewhere between those of localized PVNS and diffuse PVNS synoviocytes, which suggests a certain linearity between localized PVNS, RA and diffuse PVNS. It would be interesting to place localized PVNS synoviocytes in long-term culture in the presence of inflammatory cytokines to see whether their behavior evolves towards that of RA and diffuse PVNS synoviocytes in co-culture.

Conclusion

In conclusion, PVNS is an articular pathology with hyperproliferation of synoviocytes and an infiltrate of immune cells that may play an essential role in pathological processes, as in RA. Despite the need to increase the number of donors, our results suggest that synoviocyte profiles differ between disease subsets, notably in the response to cell interactions depending on synoviocyte origin, with changes in IFNγ / IL-10 balance associated with localized PVNS. These different profiles could be used to adapt the therapeutic strategy to each form of PVNS.

Acknowledgements

The authors acknowledge the orthopedic surgeons from Hospital Croix Rousse, Lyon (Pr Servien, Pr Lustig) and from Hospital Edouard Herriot, Lyon (Dr Pibarot, Dr Wegrzyn) for providing PVNS and RA biopsies.

Abbreviations

AIA

adjuvant-induced arthritis

GC

giant cells

IFNγ

interferon gamma

IL-

interleukin-

MNC

mononuclear cells

PBMC

peripheral blood mononuclear cells

PHA

phytohemagglutinin

PVNS

pigmented villonodular synovitis

RA

rheumatoid arthritis

TNF

tumor necrosis factor

Author contributions

MN: concept, experiments and writing. PM: concept, writing and final draft. The authors read and approved the final manuscript.

Funding

Not applicable.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the institutional review board of the Hospices civils de Lyon, France (accession number AC-2016-272) and written informed consent was obtained from the patients.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Robert M, Farese H, Miossec P. Update on tenosynovial giant cell tumor, an inflammatory arthritis with neoplastic features. Front Immunol. 2022;13:820046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ofluoglu O. Pigmented villonodular synovitis. Orthop Clin North Am. 2006;37(1):23–33. [DOI] [PubMed] [Google Scholar]
  • 3.O’Connell JX. Pathology of the synovium. Am J Clin Pathol. 2000;114(5):773–84. [DOI] [PubMed] [Google Scholar]
  • 4.Murphey MD, Rhee JH, Lewis RB, Fanburg-Smith JC, Flemming DJ, Walker EA. Pigmented villonodular synovitis: radiologic-pathologic correlation. Radiographics. 2008;28(5):1493–518. [DOI] [PubMed] [Google Scholar]
  • 5.Staals EL, Ferrari S, Donati DM, Palmerini E. Diffuse-type tenosynovial giant cell tumour: current treatment concepts and future perspectives. Eur J Cancer. 2016;63:34–40. [DOI] [PubMed] [Google Scholar]
  • 6.Yoshida W, Uzuki M, Kurose A, Yoshida M, Nishida J, Shimamura T, Sawai T. Cell characterization of mononuclear and giant cells constituting pigmented villonodular synovitis. Hum Pathol. 2003;34(1):65–73. [DOI] [PubMed] [Google Scholar]
  • 7.Darling JM, Goldring SR, Harada Y, Handel ML, Glowacki J, Gravallese EM. Multinucleated cells in pigmented villonodular synovitis and giant cell tumor of tendon sheath express features of osteoclasts. Am J Pathol. 1997;150(4):1383–93. [PMC free article] [PubMed] [Google Scholar]
  • 8.O’Keefe RJ, Rosier RN, Teot LA, Stewart JM, Hicks DG. Cytokine and matrix metalloproteinase expression in pigmented villonodular synovitis May mediate bone and cartilage destruction. Iowa Orthop J. 1998;18:26–34. [PMC free article] [PubMed] [Google Scholar]
  • 9.Ota T, Urakawa H, Kozawa E, Ikuta K, Hamada S, Tsukushi S, Shimoyama Y, Ishiguro N, Nishida Y. Expression of colony-stimulating factor 1 is associated with occurrence of osteochondral change in pigmented villonodular synovitis. Tumour Biol. 2015;36(7):5361–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Cupp JS, Miller MA, Montgomery KD, Nielsen TO, O’Connell JX, Huntsman D, van de Rijn M, Gilks CB, West RB. Translocation and expression of CSF1 in pigmented villonodular synovitis, tenosynovial giant cell tumor, rheumatoid arthritis and other reactive synovitides. Am J Surg Pathol. 2007;31(6):970–6. [DOI] [PubMed] [Google Scholar]
  • 11.Moller E, Mandahl N, Mertens F, Panagopoulos I. Molecular identification of COL6A3-CSF1 fusion transcripts in tenosynovial giant cell tumors. Genes Chromosomes Cancer. 2008;47(1):21–5. [DOI] [PubMed] [Google Scholar]
  • 12.West RB, Rubin BP, Miller MA, Subramanian S, Kaygusuz G, Montgomery K, Zhu S, Marinelli RJ, De Luca A, Downs-Kelly E, et al. A landscape effect in tenosynovial giant-cell tumor from activation of CSF1 expression by a translocation in a minority of tumor cells. Proc Natl Acad Sci U S A. 2006;103(3):690–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Dakin SG, Coles M, Sherlock JP, Powrie F, Carr AJ, Buckley CD. Pathogenic stromal cells as therapeutic targets in joint inflammation. Nat Rev Rheumatol. 2018;14(12):714–26. [DOI] [PubMed] [Google Scholar]
  • 14.Friscic J, Hoffmann MH. Stromal cell regulation of inflammatory responses. Curr Opin Immunol. 2022;74:92–9. [DOI] [PubMed] [Google Scholar]
  • 15.Naylor AJ, Filer A, Buckley CD. The role of stromal cells in the persistence of chronic inflammation. Clin Exp Immunol. 2013;171(1):30–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Noack M, Miossec P. Importance of lymphocyte-stromal cell interactions in autoimmune and inflammatory rheumatic diseases. Nat Rev Rheumatol. 2021;17(9):550–64. [DOI] [PubMed] [Google Scholar]
  • 17.Noack M, Miossec P. Synoviocytes and skin fibroblasts show opposite effects on IL-23 production and IL-23 receptor expression during cell interactions with immune cells. Arthritis Res Ther. 2022;24(1):220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Tout I, Noack M, Miossec P. Differential effects of interleukin-17A and 17F on cell interactions between immune cells and stromal cells from synovium or skin. Sci Rep. 2023;13(1):19223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Mastboom MJL, Palmerini E, Verspoor FGM, Rueten-Budde AJ, Stacchiotti S, Staals EL, Schaap GR, Jutte PC, Aston W, Gelderblom H, et al. Surgical outcomes of patients with diffuse-type tenosynovial giant-cell tumours: an international, retrospective, cohort study. Lancet Oncol. 2019;20(6):877–86. [DOI] [PubMed] [Google Scholar]
  • 20.Farese H, Noack M, Miossec P. Synoviocytes from pigmented villonodular synovitis are less sensitive to cadmium-induced cell death than synoviocytes from rheumatoid arthritis. Sci Rep. 2022;12(1):3832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO 3rd, Birnbaum NS, Burmester GR, Bykerk VP, Cohen MD, et al. 2010 rheumatoid arthritis classification criteria: an American college of rheumatology/european league against rheumatism collaborative initiative. Arthritis Rheum. 2010;62(9):2569–81. [DOI] [PubMed] [Google Scholar]
  • 22.Lavocat F, Osta B, Miossec P. Increased sensitivity of rheumatoid synoviocytes to Schnurri-3 expression in TNF-alpha and IL-17A induced osteoblastic differentiation. Bone. 2016;87:89–96. [DOI] [PubMed] [Google Scholar]
  • 23.Osta B, Roux JP, Lavocat F, Pierre M, Ndongo-Thiam N, Boivin G, Miossec P. Differential effects of IL-17A and TNF-alpha on osteoblastic differentiation of isolated synoviocytes and on bone explants from arthritis patients. Front Immunol. 2015;6:151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Noack M, Beringer A, Miossec P. Additive or synergistic interactions between IL-17A or IL-17F and TNF or IL-1beta depend on the cell type. Front Immunol. 2019;10:1726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hot A, Zrioual S, Lenief V, Miossec P. IL-17 and tumour necrosis factor alpha combination induces a HIF-1alpha-dependent invasive phenotype in synoviocytes. Ann Rheum Dis. 2012;71(8):1393–401. [DOI] [PubMed] [Google Scholar]
  • 26.Noack M, Ndongo-Thiam N, Miossec P. Interaction among activated lymphocytes and mesenchymal cells through Podoplanin is critical for a high IL-17 secretion. Arthritis Res Ther. 2016;18:148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Noack M, Ndongo-Thiam N, Miossec P. Role of Podoplanin in the high interleukin-17A secretion resulting from interactions between activated lymphocytes and psoriatic skin-derived mesenchymal cells. Clin Exp Immunol. 2016;186(1):64–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Gouin F, Noailles T. Localized and diffuse forms of tenosynovial giant cell tumor (formerly giant cell tumor of the tendon sheath and pigmented villonodular synovitis). Orthop Traumatol Surg Res. 2017;103(1S):S91–7. [DOI] [PubMed] [Google Scholar]
  • 29.Berger I, Weckauf H, Helmchen B, Ehemann V, Penzel R, Fink B, Bernd L, Autschbach F. Rheumatoid arthritis and pigmented villonodular synovitis: comparative analysis of cell polyploidy, cell cycle phases and expression of macrophage and fibroblast markers in proliferating synovial cells. Histopathology. 2005;46(5):490–7. [DOI] [PubMed] [Google Scholar]
  • 30.Bonaventura P, Courbon G, Lamboux A, Lavocat F, Marotte H, Albarede F, Miossec P. Protective effect of low dose intra-articular cadmium on inflammation and joint destruction in arthritis. Sci Rep. 2017;7(1):2415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hong SS, Marotte H, Courbon G, Firestein GS, Boulanger P, Miossec P. PUMA gene delivery to synoviocytes reduces inflammation and degeneration of arthritic joints. Nat Commun. 2017;8(1):146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Chomarat P, Rissoan MC, Banchereau J, Miossec P. Interferon gamma inhibits Interleukin 10 production by monocytes. J Exp Med. 1993;177(2):523–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zhao Y, Lv J, Zhang H, Xie J, Dai H, Zhang X. Gene expression profiles analyzed using integrating RNA sequencing, and microarray reveals increased inflammatory response, proliferation, and osteoclastogenesis in pigmented villonodular synovitis. Front Immunol. 2021;12:665442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Takayanagi H, Ogasawara K, Hida S, Chiba T, Murata S, Sato K, Takaoka A, Yokochi T, Oda H, Tanaka K, et al. T-cell-mediated regulation of osteoclastogenesis by signalling cross-talk between RANKL and IFN-gamma. Nature. 2000;408(6812):600–5. [DOI] [PubMed] [Google Scholar]
  • 35.Ji JD, Park-Min KH, Shen Z, Fajardo RJ, Goldring SR, McHugh KP, Ivashkiv LB. Inhibition of RANK expression and osteoclastogenesis by TLRs and IFN-gamma in human osteoclast precursors. J Immunol. 2009;183(11):7223–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Cheng J, Liu J, Shi Z, Jules J, Xu D, Luo S, Wei S, Feng X. Molecular mechanisms of the biphasic effects of interferon-gamma on osteoclastogenesis. J Interferon Cytokine Res. 2012;32(1):34–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Zhang S, Kohli K, Black RG, Yao L, Spadinger SM, He Q, Pillarisetty VG, Cranmer LD, Van Tine BA, Yee C, et al. Systemic Interferon-gamma increases MHC class I expression and T-cell infiltration in cold tumors: results of a phase 0 clinical trial. Cancer Immunol Res. 2019;7(8):1237–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nowicki TS, Akiyama R, Huang RR, Shintaku IP, Wang X, Tumeh PC, Singh A, Chmielowski B, Denny C, Federman N, et al. Infiltration of CD8 T cells and expression of PD-1 and PD-L1 in synovial sarcoma. Cancer Immunol Res. 2017;5(2):118–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Carvalheiro H, da Silva JA, Souto-Carneiro MM. Potential roles for CD8(+) T cells in rheumatoid arthritis. Autoimmun Rev. 2013;12(3):401–9. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

No datasets were generated or analysed during the current study.

Articles from Arthritis Research & Therapy are provided here courtesy of BMC

RESOURCES