Kindlin-2 inhibits Nlrp3 inflammasome activation in nucleus pulposus to maintain homeostasis of the intervertebral disc

Abstract

Intervertebral disc (IVD) degeneration (IVDD) is the main cause of low back pain with major social and economic burdens; however, its underlying molecular mechanisms remain poorly defined. Here we show that the focal adhesion protein Kindlin-2 is highly expressed in the nucleus pulposus (NP), but not in the anulus fibrosus and the cartilaginous endplates, in the IVD tissues. Expression of Kindlin-2 is drastically decreased in NP cells in aged mice and severe IVDD patients. Inducible deletion of Kindlin-2 in NP cells in adult mice causes spontaneous and striking IVDD-like phenotypes in lumbar IVDs and largely accelerates progression of coccygeal IVDD in the presence of abnormal mechanical stress. Kindlin-2 loss activates Nlrp3 inflammasome and stimulates expression of IL-1β in NP cells, which in turn downregulates Kindlin-2. This vicious cycle promotes extracellular matrix (ECM) catabolism and NP cell apoptosis. Furthermore, abnormal mechanical stress reduces expression of Kindlin-2, which exacerbates Nlrp3 inflammasome activation, cell apoptosis, and ECM catabolism in NP cells caused by Kindlin-2 deficiency. In vivo blocking Nlrp3 inflammasome activation prevents IVDD progression induced by Kindlin-2 loss and abnormal mechanical stress. Of translational significance, adeno-associated virus-mediated overexpression of Kindlin-2 inhibits ECM catabolism and cell apoptosis in primary human NP cells in vitro and alleviates coccygeal IVDD progression caused by mechanical stress in rat. Collectively, we establish critical roles of Kindlin-2 in inhibiting Nlrp3 inflammasome activation and maintaining integrity of the IVD homeostasis and define a novel target for the prevention and treatment of IVDD.

Fig. 1

The expression of Kindlin, Talin, and Vinculin proteins in degenerative human nucleus pulposus (NP) specimens and mouse intervertebral discs (IVDs). a Representative MRI images and general views of NP tissues with different Pfirrmann degrees. Scale bar, 1 cm. b Hematoxylin and eosin (H/E) and Alcian blue staining of human NP samples. Scale bar, 50 μm. c, d Immunofluorescent (IF) staining of KINDLIN-1, 2, 3 (K1, 2, 3), TALIN, and VINCULIN in human NP samples. Scale bar, 50 μm. n = 9. e, f Safranin O and Fast Green (SO&FG) staining and histological scores of lumbar IVDs in young mice (3-month-old) and aged mice (20-month-old). Scale bar, 200 μm. n = 6 (young) and 5 (aged). g–l IF staining of K1, K2, K3, Talin, and Vinculin in young and aged mice. Scale bar, 200 μm. n = 6 (young) and 5 (aged). NS, no statistical significance, **P < 0.01, ***P < 0.001

Fig. 2

Kindlin-2 deletion results in spontaneous and progressive IVDD in lumbar IVDs in mice. a A schematic illustration of the experimental design. b, c IF staining of K2 in lumbar IVDs in control and Kindlin-2 conditionally knockout (cKO) mice at 3 months of age. Scale bar, 200 μm. n = 6. d, f SO&FG staining and histological scores of lumbar IVDs in control and cKO mice at 3, 5 and 8 months of age. Scale bar, 200 μm. n = 6. e Cell numbers of NP in control and cKO mice at 3, 5, and 8 months of age. n = 6. g, h, j, k Western blotting analyses of Aggrecan (Acan), Collagen type II (Col2a1), a disintegrin and metalloproteinase with thrombospondin motif 5 (Adamts5) and matrix metalloproteinase 13 (Mmp13) as well as active Caspase3, Bcl2 and Bax in lumbar IVDs in control and cKO mice at 8 months of age. n = 3. i, l IF staining of Acan, Col2a1, Mmp13 and Adamts5 as well as active Caspase3, Bcl2, and Bax in NP tissues in lumbar IVDs of control and cKO mice at 8 months of age. Scale bar, 50 μm. m, n TUNEL staining of NP tissues in lumbar IVDs of control and cKO mice at 8 months of age. Scale bar, 50 μm. n = 6. NS, no statistical significance, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 3

Kindlin-2 deletion accelerates progression of coccygeal IVDD in the presence of abnormal mechanical stress in mice. a Overview of the experimental design of coccygeal IVDs needle stab (CINS) model. b, c SO&FG staining and histological scores of coccygeal IVDs in control and cKO mice with or without CINS. Scale bar, 100 μm. n = 6. d, e Changes in the disc height index (%DHI) (Co6-7, 7–8) were evaluated by micro-computed tomography (μCT) analyses. n = 6. f–h IF staining of Col2a1 and Mmp13 in NP tissues in coccygeal IVDs of control and cKO mice with or without CINS. Scale bar, 50 μm. n = 6. i, j TUNEL staining of NP tissues in coccygeal IVDs of control and cKO mice with or without CINS. Scale bar, 50 μm. n = 6. **P < 0.01, ***P < 0.001

Fig. 4

Kindlin-2 deletion promotes Nlrp3 inflammasome activation, ECM catabolism and cell apoptosis in NP cells in vitro. a Immunohistochemical (IHC) staining of Nlrp3, Caspase-1(Casp1) and IL-1β in human NP samples. Scale bar, 50 μm. b, c IF staining of Nlrp3, Casp1 and IL-1β in NP tissues in lumbar IVDs of control and cKO mice at 8 months of age. Scale bar, 50 μm. d–f Western blotting analyses of K2, Col2a1, Mmp13, Nlrp3, Casp1, and IL-1β in NP cells, which were transfected with negative control siRNA (si-NC) or K2 siRNA (si-K2), negative control plasmid (NC-PL) or K2 plasmid (K2-PL), and then treated with or without compression loading (CL) treatment. n = 3. g The levels of IL-1β in the conditioned media of cultured NP cells treated as in (d), as detected by ELISA. n = 3. h, i TUNEL staining of NP cells, which were treated as in (d). Scale bar, 50 μm. n = 4. j–m Western blotting analyses of K2, Nlrp3, and Casp1 in NP cells, which were treated with different doses (0, 25, 50, 100 ng·mL⁻¹) of IL-1β for 24 h or 50 ng·mL⁻¹ IL-1β for different time (0, 12, 24, 48 h) with or without CL treatment. n = 3. n, o Western blotting analyses of Casp1, IL-1β, Col2a1, and Mmp13 in NP cells, which were transfected with si-NC or si-K2, pretreated with or without MCC950, and then treated with or without CL treatment. n = 3. p The levels of IL-1β in the conditioned media of cultured NP cells treated as in (n), as detected by ELISA. n = 3. q, r Apoptosis rate of NP cells, which were treated as in (n), was measured by flow cytometry analyses. n = 4. NS no statistical significance, *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 5

MCC950 limits IVDD progression caused by Kindlin-2 deletion and abnormal mechanical stress in mice. a Overview of the experimental set-up of coccygeal IVDs needle stab (CINS) model with or without MCC950 treatment. b, c SO&FG staining and histological scores of coccygeal IVDs in control and cKO mice, which were treated with or without CINS, and then treated with or without MCC950. Scale bar, 100 μm. n = 6. d, e Changes in the disc height index (%DHI) (Co6-7, 7–8) were evaluated by micro-computed tomography (μCT). n = 6. f–h IF staining of Col2a1 and Mmp13 in NP tissues in coccygeal IVDs of control and cKO mice treated as in (b). Scale bar, 50 μm. n = 6. i, j TUNEL staining of NP tissues in coccygeal IVDs of control and cKO mice treated as in (b). Scale bar, 50 μm. n = 6. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 6

Administration of AAV-Kindlin-2 largely alleviates IVDD progression. a, c–f IF staining of K2, Nlrp3, Col2a1, and Mmp13 in human primary NP cells infected with control adeno-associated virus (CTL AAV) or K2 AAV, and then treated with or without compression loading (CL). Scale bar, 50 μm. n = 4. g The levels of IL-1β in the conditioned media of cultured human NP cells treated as in (a), as detected by ELISA. n = 4. b, h TUNEL staining of primary human NP cells, which were treated as in (a). Scale bar, 50 μm. n = 4. i Overview of the experimental set-up of rat coccygeal IVDs compression (CIC) model with or without K2 AAV treatment. j, k Magnetic resonance imaging (MRI) and Pfirrmann grades of coccygeal IVDs in rats treated as in (i). n = 6. l, o SO&FG staining and histological scores of coccygeal IVDs in rats treated as in (i). Scale bar, 500 or 100 μm. n = 6. m, p-r IF staining of K2, Col2a1, and Mmp13 in NP tissues in coccygeal IVDs of rats treated as in (i). Scale bar, 50 μm. n = 6. n, s TUNEL staining of NP tissues in coccygeal IVDs of rats treated as in (i). Scale bar, 50 µm. n = 6. *P < 0.05, **P < 0.01, ***P < 0.001