Abstract
An effective treatment for osteoporosis requires both, inhibition of bone resorption by osteoclasts and induction of bone formation by osteoblasts. Human bone marrow derived mesenchymal stromal cells (MSC) have the potential to differentiate into osteoblasts and thus form new bone. Currently, many different proteins and signaling pathways are known to regulate the differentiation of MSCs into osteoblasts, but most signals are not specific enough to induce osteogenesis while preventing off-target effects on other cell types. Recently, the C-type lectin Clec11a was demonstrated to be a potential factor to regulate osteogenesis in MSCs. However, whether previous work in mice would translate into human cells, and how Clec11a signals to cells, remains largely unknown. To study the effect of Clec11a on osteogenesis, we used lentiviral vectors to either over-express or silence Clec11a expression in human bone marrow-derived MSC, derived from multiple donors. We found that MSCs transduced with an shRNA targeting Clec11a showed an increase in calcium precipitation and alkaline phosphatase activity, while over-expression of Clec11a had no significant effect on the cells. To elucidate the underlying mechanism, we used an antibody array to detect the phosphorylation of proteins involved in signaling pathways. In the absence of Clec11a, we found a consistent decrease in the phosphorylation of GSK3 and JNK, proteins involved in canonical and noncanonical Wnt signaling pathways, as well as AKT. These pathways have all shown to regulate osteogenesis in MSCs. Canonical Wnt activity was confirmed through analyzing the gene expression of the canonical Wnt target gene Grem1, which we found consistently upregulated in Clec11a silenced cells. The decrease in GSK3 phosphorylation and increase in Grem1 expression suggests an overall decrease in canonical Wnt signaling, induced in the absence of Clec11a. To assess non-canonical Wnt signaling, we utilized a novel reporter system (Kif26b-GFP) that detects non-canonical Wnt activity through flow cytometry. Flow analyses showed that in the absence of Clec11a, there was indeed a consistent increase of non-canonical Wnt signaling. Although the precise receptor for Clec11a in MSCs remains unknown, we used an in silico generated model to determine potential protein interactions and found PDGFR as the strongest candidate. As experimental evidence for this, we found Grem1 upregulated and downregulated when we inhibited and activated PDGFR expression respectively. We observed an attenuated effect on Grem1 in the absence of Clec11a. Importantly, PDGFR has been shown to regulate phosphorylation of GSK3, AKT and JNK, linking this result to our abovementioned findings. Together, our results implicate PDGFR as a potential receptor of Clec11a. In fact, in MSCs, Clec11a seems necessary for normal PDGFR signaling to occur. Taken together, we believe Clec11a regulates osteogenesis at endogenous expression levels and its absence induces increased osteogenesis of human MSCs in vitro. We propose that this process relies on PDGFR signaling and the regulation of Akt and canonical and non-canonical Wnt signaling pathways.