Targeting of chondrocyte plasticity via connexin43 modulation attenuates cellular senescence and fosters a pro-regenerative environment in osteoarthritis

Abstract

Osteoarthritis (OA), a chronic disease characterized by articular cartilage degeneration, is a leading cause of disability and pain worldwide. In OA, chondrocytes in cartilage undergo phenotypic changes and senescence, restricting cartilage regeneration and favouring disease progression. Similar to other wound-healing disorders, chondrocytes from OA patients show a chronic increase in the gap junction channel protein connexin43 (Cx43), which regulates signal transduction through the exchange of elements or recruitment/release of signalling factors. Although immature or stem-like cells are present in cartilage from OA patients, their origin and role in disease progression are unknown. In this study, we found that Cx43 acts as a positive regulator of chondrocyte-mesenchymal transition. Overactive Cx43 largely maintains the immature phenotype by increasing nuclear translocation of Twist-1 and tissue remodelling and proinflammatory agents, such as MMPs and IL-1β, which in turn cause cellular senescence through upregulation of p53, p16^INK4a and NF-κB, contributing to the senescence-associated secretory phenotype (SASP). Downregulation of either Cx43 by CRISPR/Cas9 or Cx43-mediated gap junctional intercellular communication (GJIC) by carbenoxolone treatment triggered rediferentiation of osteoarthritic chondrocytes into a more differentiated state, associated with decreased synthesis of MMPs and proinflammatory factors, and reduced senescence. We have identified causal Cx43-sensitive circuit in chondrocytes that regulates dedifferentiation, redifferentiation and senescence. We propose that chondrocytes undergo chondrocyte-mesenchymal transition where increased Cx43-mediated GJIC during OA facilitates Twist-1 nuclear translocation as a novel mechanism involved in OA progression. These findings support the use of Cx43 as an appropriate therapeutic target to halt OA progression and to promote cartilage regeneration.

Fig. 1. OAC reverts to a more progenitor-like or stem cell-like phenotype.

a Mesenchymal stem cell markers (CD90, CD105, CD29, CD44, CD166 and CD73) were tracked by flow cytometry for 7 and 14 days during chondrogenic differentiation of hMSCs in monolayer culture. Data were normalized to the 0-day value. b Dedifferentiation of OAC to a more progenitor-like cell state at different time periods in monolayer culture was quantified with the same stem cell markers and flow cytometry. Data were normalized to those of low passage numbers (<1 month; S₁). Cell populations of unstained cells (Unst; negative control) and during short (<1 month; S₁) and long (>2 months; S₂) passages are shown on the right. The heterogeneous subpopulations of dedifferentiated OAC (red line, S₁ passage) tended to reach a single population of dedifferentiated chondrocytes cells after long passages (blue line, S₂). c Comparative flow cytometry analysis showing that OACs express higher levels of the stem cell marker CD166 compared with chondrocytes isolated from healthy donors (n = 5–8; mean ± s.e.m.; ***P < 0.0001; one-way ANOVA). d Increased CD166 levels detected by flow cytometry in OACs from high-grade (grades III–IV) versus low-grade (grades I–II) OA donors (n = 4–5; mean ± s.e.m.; *P < 0.05; Mann–Whitney test). e Expression of stem markers by flow cytometry when OACs are cultured in chondrogenic medium (CM) for 14 days suggests that dedifferentiated OACs can redifferentiate into mature chondrocytes (n = 2; mean ± s.e.m.; ***P < 0.0001, Student’s t-test). f Cx43 protein levels in OACs cultured for 7 and 14 days in normal growth medium (DMEM with 10% FBS; UT) or in chondrogenic medium (CM) (n = 4; mean ± s.e.m.; *P < 0.05; Student’s t-test). g Representative images of OACs that have undergone chondrogenic differentiation (CM) as 3D micromasses for 30 days, and evaluated with Safranin O/Fast Green staining and Col2A1 immunohistochemistry. h Comparative images of OACs and healthy chondrocytes (N) that have undergone adipogenic (AM, evaluated by Oil red O staining) or osteogenic (OM, evaluated by alizarin red staining) differentiation for 21 days (AM; adipogenic induction medium, OM; osteogenic medium) OACs show more plasticity and ability to differentiate into adipocytes and bone cells (representative of n = 10 experiments; mean ± s.e.m.; *P < 0.05, ***P < 0.0001; Mann–Whitney test). Original magnifications ×20, ×40 and ×100. Scale bars are shown below

Fig. 2. Downregulation of Cx43 GJ plaques and GJIC restores chondrocyte redifferentiation in chondrocytes from patients with OA.

a OACs express higher levels of Cx43 protein (mainly located in GJ plaques, white arrows) compared with chondrocytes from healthy donors (N), (western blot, flow cytometry and immunofluorescence assays (n = 3; mean ± s.e.m.; *P < 0.05; Mann–Whitney test). b OACs show higher levels of GJIC in comparison with healthy chondrocytes in primary cultures measured by SL/DT assay (n = 3; mean ± s.e.m.; *P < 0.05; Mann–Whitney test). c Flow cytometry analysis shows reduced levels of CD105 and CD166 in OACs treated for 7 days with 100 µM CBX, suggesting reduced chondrocyte dedifferentiation upon GJIC inhibition. Graphs show CD105 and CD166 levels detected in OACs exposed to CBX or chondrogenic medium (CM) (n = 4–5; mean ± s.e.m.; *P < 0.05, **P < 0.01, ***P < 0.0001; one-way ANOVA). d OACs in chondrogenic (CM) and normal growth medium (UT, DMEM 10% FBS) for 48 h. GJCI was measured by SL/DT assay (n = 3; mean ± s.e.m.; ***P < 0.0001; Mann–Whitney test). e CBX 50 and 100 µM inhibit GJIC in OACs measured by SL/DT assay (Supplementary Fig. 3a). Immunofluorescence analysis of Col2A1 and Cx43 levels in OACs treated with CBX 50 and 100 µM for 24 h. The graphs show the percentage of positive cells for Col2A1 and the CTCF for Cx43 mainly located in the margin of the cells (gap junction plaques, white arrows) (n = 5; mean ± s.e.m.; *P < 0.05, **P < 0.01; one-way ANOVA). f Adipogenic differentiation of OACs exposed to CBX 50 µM for 21 days was evaluated with oil red O staining. The graph represents the ratio of cells with lipid deposits to the total number of cells (n = 8, mean ± s.e.m.; ***P < 0.0001; Mann–Whitney test). g IL-1ß, IL-6, COX-2, MMP-3 and MMP-13 mRNA expression in OACs treated with 50 and 100 µM CBX for 15 min. Data were normalized to HPRT-1 (n = 3–4; mean ± s.e.m.; *P < 0.05, **P < 0.01, ***P < 0.0001; one-way ANOVA)

Fig. 3. Upregulation of Cx43 induces Twist-1 activation and switches on the dedifferentiation program to gain mesenchymal characteristics.

a The T/C-28a2 chondrocyte cell line transfected with a plasmid to overexpress Cx43 (referred to as Cx43) displays increased Cx43 protein levels (detected by western blot) and GJIC (evaluated by LY transfer in an SL/DT assay). Cx43 levels were compared with those of the non-transfected and empty vector (EV)-transfected cells. The graph represents the quantification of more than 10 images from two independent experiments (*P < 0.05; Mann–Whitney test). Below, immunofluorescence assays indicating that, in chondrocytes overexpressing Cx43, Cx43 (green) is mainly localized to the membrane (red arrows) and perinuclear area (white arrows). Nuclei were stained with DAPI. b Flow cytometry analysis showing that T/C-28a2 cells overexpressing Cx43 (red) expressed significantly higher levels of the stemness markers CD105 (n = 7) and CD166 (n = 5) versus the same cell line transfected with a control plasmid (black) (mean ± s.e.m.; *P < 0.05; Mann–Whitney test). c Downregulation of Cx43 levels in T/C-28a2 cells with the CRISPR-Cas9 knockdown system (Cx43^−/+) (detected by western blot) led to significantly lower levels of GJIC, detected by SL/DT assay (n = 6, mean ± s.e.m.; **P < 0.01; Mann–Whitney test). d Compared with the control cells, the CRISPR-Cas9 knockdown chondrocyte cell line (T/C-28a2, Cx43^−/+) contained lower levels of the stem cell markers CD105 (flow cytometry) and CD106 (flow cytometry, western blot and RT-qPCR). Flow cytometry; n = 7 mean ± s.e.m.; *P < 0.05, **P < 0.01; Mann–Whitney test. Gene expression was normalized to HPRT-1 levels (n = 7, mean ± s.e.m.; ***P < 0.0001; one-way ANOVA). e Upregulation of Cx43 increases the levels of nuclear Twist-1 and nuclear PCNA (western blot) and upregulates the gene expression of Twist-1 and the mesenchymal markers vimentin and N-cadherin (RT-qPCR, below) (n = 3–5; mean ± s.e.m, *P < 0.05; Mann–Whitney test). T/C-28a2 cell line overexpressing Cx43 compared with non-transfected cells or control plasmid (EV). Lamin A was used as a nuclear loading control. f Changes in the vimentin (red) filament network were associated with Cx43 upregulation. T/C-28a2 cells overexpressing Cx43 versus non-transfected cells. Nuclei were stained with DAPI. Scale bar is shown below. g Reduced levels of Cx43 in T/C-28a2 cells (CRISPR-cas9 system) correlated with downregulation of N-cadherin, vimentin and Twist-1 gene expression (n = 3, mean ± s.e.m.; *P < 0.05; Mann–Whitney test). Cx43 levels correlated with nuclear levels of Twist-1 and PCNA, detected by western blot. T/C-28a2 cell line compared to the same cells overexpressing Cx43 (Cx43) or the heterozygous for Cx43 (Cx43^−/+). Lamin A was used as a nuclear loading control (n = 2–3, mean ± s.e.m.; *P < 0.05, **P < 0.01; Student’s t-test)

Fig. 4. Dedifferentiation induced by Cx43 activates senescence and prolongs inflammation and remodelling.

a Flow cytometry analysis with the FDG substrate showing elevated levels of SAβG activity in Cx43-overexpressing T/C-28a2 cells (red) versus empty vector (EV) control cells (black) (n = 5; mean ± s.e.m.; *P < 0.05; Mann–Whitney test). b Western blot (n = 2; mean ± s.e.m.; *P < 0.05; Student’s t-test) and RT-qPCR (n = 4; mean ± s.e.m.; *P < 0.05; Mann–Whitney test) showing that overexpression of Cx43 increased p53 and p16 levels in T/C-28a2 chondrocytes. c Analysis of IL-1β, COX-2, MMP-3 and MMP-13 mRNA levels using real-time PCR in T/C-28a2 cells overexpressing Cx43 (n = 2–4; normalized to HPRT-1 and represented as mean ± s.e.m.; *P < 0.05; Mann–Whitney test). d Nuclear levels of NF-κB detected by western blot in the T/C-28a2 cell line overexpressing Cx43 (Cx43) and in the T/C-28a2 heterozygous for Cx43 (Cx43^−/+), compared to the non-transfected cell line (T/C-28a2). Lamin A was used as a nuclear loading control (n = 2, mean ± s.e.m.; *P < 0.05, **P < 0.01; Student’s t-test). e Downregulation of Cx43 in T/C-28a2 cells (CRISPR-Cas9 knockdown) decreased SAβG (blue) (n = 3–4; mean ± s.e.m. *P < 0.05, **P < 0.01; one-way ANOVA), and increased of p53 and p16 protein levels, detected by western blot (n = 2–3, mean ± s.e.m.; *P < 0.05; Student’s t-test). f Analysis of IL-1β, COX-2, MMP-3 and MMP-13 mRNA levels in the T/C-28a2 heterozygous for Cx43 (Cx43^−/+). Data were normalized to HPRT-1 (mean ± s.e.m. n = 3–4; *P < 0.05; Mann–Whitney test)

Fig. 5. Downregulation of Cx43 in OA fosters a pro-regenerative environment by favouring redifferentiation.

Accumulation of dedifferentiated (2) and senescent (3) chondrocytes and Cx43 overexpression are typical in the cartilage microenvironment from the early stages of OA. Phenotypic changes associated with upregulation of Cx43 during tissue repair may contribute to ECM remodelling and cartilage regeneration (or wound healing) (1–2 and 4, black arrows). However, they result in a chondrocyte fibrogenic phenotype (2, and red arrow) and accumulation of senescent cells (3) in OA, which is unable to support the ECM composition by continuous activation of ECM remodelling (MMPs) and proinflammatory factors, such as IL-1ß, NF-κB and IL-6 a, c. Downregulation of Cx43 or GJIC (e.g., by CRISPR-Cas9 or CBX treatment) leads to redifferentiation of OACs (4) by reducing Twist-1 activation (2), senescence (3 and b) and SASP (c, NF-κB-p65). The results indicate that dedifferentiation and senescence in OA are flexible processes that can be reverted by modulating Cx43. The elimination or inhibition of factors that contribute to the prolongation of cellular dedifferentiation (Twist-1; chondrocyte-mesenchymal transition) or reprogramming (e.g. via IL-6) reduces the burden of dedifferentiated (2) and senescent (3) cells, which in turn decreases the synthesis of catabolic factors (a and c) involved in cartilage degradation in OA and leads to a mature phenotype able to support the cartilage ECM composition