Directed differentiation of human pluripotent stem cells into articular cartilage reveals effects caused by absence of <i>WISP3</i>, the gene responsible for progressive pseudorheumatoid arthropathy of childhood.

Abstract

Progressive pseudorheumatoid arthropathy of childhood (PPAC), caused by deficiency of <i>WNT1 inducible signalling pathway protein 3</i> (<i>WISP3</i>), has been challenging to study because no animal model of the disease exists and cartilage recovered from affected patients is indistinguishable from common end-stage osteoarthritis. Therefore, to gain insights into why precocious articular cartilage failure occurs in this disease, we made in vitro derived articular cartilage using isogenic <i>WISP3</i>-deficient and <i>WISP3</i>-sufficient human pluripotent stem cells (hPSCs). We generated articular cartilage-like tissues from induced-(i) PSCs from two patients with PPAC and one wild-type human embryonic stem cell line in which we knocked out WISP3. We compared these tissues to in vitro-derived articular cartilage tissues from two isogenic <i>WISP3</i>-sufficient control lines using histology, bulk RNA sequencing, single cell RNA sequencing and in situ hybridisation. <i>WISP3</i>-deficient and <i>WISP3</i>-sufficient hPSCs both differentiated into articular cartilage-like tissues that appeared histologically similar. However, the transcriptomes of <i>WISP3</i>-deficient tissues differed significantly from <i>WISP3</i>-sufficient tissues and pointed to increased TGFβ, TNFα/NFκB, and IL-2/STAT5 signalling and decreased oxidative phosphorylation. Single cell sequencing and in situ hybridisation revealed that <i>WISP3</i>-deficient cartilage contained a significantly higher fraction (~4 fold increase, p<0.001) of superficial zone chondrocytes compared with deeper zone chondrocytes than did <i>WISP3</i>-sufficient cartilage. <i>WISP3</i>-deficient and <i>WISP3</i>-sufficient hPSCs can be differentiated into articular cartilage-like tissues, but these tissues differ in their transcriptomes and in the relative abundances of chondrocyte subtypes they contain. These findings provide important starting points for in vivo studies when an animal model of PPAC or presymptomatic patient-derived articular cartilage becomes available.