Calcium calmodulin kinase II activity is required for cartilage homeostasis in osteoarthritis

Abstract

WNT ligands can activate several signalling cascades of pivotal importance during development and regenerative processes. Their de-regulation has been associated with the onset of different diseases. Here we investigated the role of the WNT/Calcium Calmodulin Kinase II (CaMKII) pathway in osteoarthritis. We identified Heme Oxygenase I (HMOX1) and Sox-9 as specific markers of the WNT/CaMKII signalling in articular chondrocytes through a microarray analysis. We showed that the expression of the activated form of CaMKII, phospho-CaMKII, was increased in human and murine osteoarthritis and the expression of HMOX1 was accordingly reduced, demonstrating the activation of the pathway during disease progression. To elucidate its function, we administered the CaMKII inhibitor KN93 to mice in which osteoarthritis was induced by resection of the anterior horn of the medial meniscus and of the medial collateral ligament in the knee joint. Pharmacological blockade of CaMKII exacerbated cartilage damage and bone remodelling. Finally, we showed that CaMKII inhibition in articular chondrocytes upregulated the expression of matrix remodelling enzymes alone and in combination with Interleukin 1. These results suggest an important homeostatic role of the WNT/CaMKII signalling in osteoarthritis which could be exploited in the future for therapeutic purposes.

Figure 1

HMOX1 is a selective target of the WNT/CaMKII pathway in the articular chondrocytes. (a–d) Human articular chondrocytes isolated from OA patients were treated with human recombinant WNT3A in presence of the CaMKII inhibitor KN93 or its inactive control molecule KN92. (a) Volcano plot showing genes regulated by WNT3A (KN92 + WNT3A vs KN92), (b) genes modulated by CaMKII inhibition in the presence of WNT3A (KN93 + WNT3A vs KN92 + WNT3A) and (c) genes activated by WNT3A independently of CaMKII (KN93 + WNT3A vs KN92). (d) Venn diagram showing the number of significantly modulated genes in each group. (e) qPCR confirming downregulation of HMOX1 in response to WNT3A but not in the presence of the CaMKII inhibitor KN93. Microarray data were obtained by testing samples from 4 human donors. For 2 donors all conditions were tested in two independent cartilage explants; for one donor the condition KN92 + WNT3A was tested in a single explant; for the remaining donor all the conditions were tested in a single explant per condition. (f) qPCR showing that HMOX1 is a selective target of the WNT/CaMKII branch and it is not modulated via activation of the WNT/beta-catenin pathway. n = 3 human donors, 2 technical replicates per donor. (a–d) data were analysed with a linear models with 10% FDR p value correction. (e,f) Data were analysed with one-way ANOVA followed by TukeyHSD post-hoc test.

Figure 2

CaMKII and HMOX1 expression are oppositely modulated in osteoarthritis. (a) Relative expression of the four CaMKII mRNA isoforms as assessed by expression microarrays in human cartilage explants. (b) Gene expression analysis by semiquantitative PCR for CaMKIIα, -β, -γ, -δ, in P0 human articular chondrocytes (upper row). A pool of cDNA from brain, placenta and lymph node was used as positive control for the PCR (lower row) (n cycles = 40). All the samples were loaded and run in the same gel whose image was cut for best rendering. A picture of the full original gel can be found in Supplementary Figure 3. (c) RNA expression of CaMKII isoforms in human cartilage retrieved from normal and OA human cartilage as assessed by Soul and colleagues in a previously performed RNAseq. (d) Immunodetection of phospho-CaMKII in normal and OA articular cartilage. Image representative of n = 3. (e) RNA deep sequencing analysis of HMOX1 expression in normal and OA articular cartilage.

Figure 3

Expression of CaMKII isoforms in murine cartilage. (a,b) qPCR analysis of murine femoral heads and patellae removed from 1 month-old 129/Sv male mice (n = 4). Gene expression was normalized for the housekeeping gene beta-actin. (c–o) Immunofluorescence for CaMKII-β, -γ and δ in the murine joint. (p) mRNA expression of the CaMKII isoforms after CaMKIIγ silencing by siRNA in C3H/10T1/2 cells. n = 4. (q) Downregulation of CaMKII by siRNA was sufficient to upregulate HMOX1 in C3H/10T1/2 cells n = 4. Data analysed by t-test. F = femur; T = tibia; M = meniscus; BM = bone marrow; CL = cruciate ligament; SM = synovial membrane.

Figure 4

Effect of CaMKII inhibition in vivo in a murine model of OA. (a) Osteoarthritis was induced by removal of the anterior horn of the medial meniscus and transection of the medial collateral ligament (MLI model). Four weeks after surgery, mice were administered KN93 or PBS vehicle subcutaneously via a minipump for additional 4 weeks (the initial minipump was replaced after 2 weeks). The total volume administered over 4 weeks was equal to 400 µl. n = 15/treatment. (b) Safranin O staining of representative sections of the medial compartment of the knee of mice treated with PBS or KN93. (c) OARSI score of the medial and the lateral compartments of the MLI knees n = 15/treatment. (d) Histomorphometry analysis of the MLI knees. (e) Correlation of the OARSI score and histomorphometry analysis. MF: medial femur; MT: medial tibia; LF: lateral femur; LT: lateral tibia. OARSI scores were analysed with Mann–Whitney test with Wilcoxon post-test. Histomorphometry data were analysed by t-test. Linear regression was used to compare the OARSI scores and histomorphometry data.

Figure 5

CaMKII inhibition decreases subchondral thickness. (a,b) MicroCt analysis of operated and contralateral sham-operated knee joints of mice treated with KN93 or PBS vehicle upon surgical destabilization of the knee joint. (a) Epiphyseal Bone volume/tissue volume of the entire tibial epiphysis of the operated knees. (b) Comparison of the percent change of the BV/TV between the MLI and the sham operated knee in the two treatment groups. (c) Histomorphometrical assessment of osteophyte area, % of cartilage area and histomorphometrical quantification of subchondral bone thickness. Data were analysed by t-test. n = 15/treatment.

Figure 6

CaMKII inhibition enhanced the capacity of IL1B to upregulate catabolic enzymes in the articular chondrocytes. (a–h) mRNA expression of ADAMTS4/5 and MMP3/13 measured by qPCR of bovine articular chondrocytes stimulated for 24 h with IL1B (20 ng/ml) in presence or absence of KN93 or its inactive analogue KN92 (both 10 µM) (a,c,e,g) or in the presence or absence of the CaMKII inhibitor AIP (b,d,f,h) (5 µM). Data were analysed with 2 ways ANOVA and Tukey post-test. n = 3.

Figure 7

Diagram summarising the transcriptional targets modulated by WNT3A in a CaMKII dependent manner and the effect of the pharmacological blockade of CaMKII on the remodelling of joint tissues in osteoarthritis.