br Results br Discussion As

Results
Discussion As stem reference 4 of the skeleton, BM-MSC\'s lineage commitment, a crucial step thought to determine the yield of functional osteoblasts, needs to be tightly regulated. Yet how BM-MSC lineage commitment is regulated remains poorly understood. This study revealed that p38α MAPK regulates tri-lineage commitment and proliferation of BM-MSCs. p38α MAPK deficiency compromises osteogenic/chondrogenic differentiation but favors adipogenic differentiation, and results in a decrease in bone mass, a defect in growth plate, and an increase in bone marrow fat. Previous studies have implicated that YAP, C-MAF, and VEGF differentially regulate osteoblast and adipogenic differentiation of BM-MSCs (Hong et al., 2005; Liu et al., 2012; Nishikawa et al., 2010). It will be interesting to test whether there exist links between p38 MAPKs and YAP, C-MAF, or VEGF in the context of BM-MSC lineage commitment. These proteins may present a class of targets for drug development for osteoporosis prevention/treatment. Our data suggest that p38α may promote BM-MSC osteogenic commitment by suppressing the NF-κB pathway. Previous studies suggest that p38α and p38β promote osteoblast differentiation via modifying RUNX2 and OSX (Greenblatt et al., 2013). But lineage-specific transcription factors such as RUNX2 and OSX are in general not expressed in multipotent BM-MSCs, and thus p38 MAPKs-mediated phosphorylation of these proteins may not be a major factor in determining the cell fates of BM-MSCs (Chang et al., 2009; Yu et al., 2014). Thus, p38α may regulate BM-MSC osteogenic commitment and osteoblast maturation with different mechanisms (Figure 6H). Moreover, p38α may promote BM-MSC osteogenic commitment by suppressing the competing adipogenic or chondrogenic commitment through down-regulating PPARγ, C/EBPα, and SOX9, transcription factors essential for commitment of these two lineages (Kozhemyakina et al., 2015). p38α deficiency led to a decrease in the number of osteoblasts on bone sections, which can be caused by defective BM-MSC osteogenic differentiation. Surprisingly, the number of BM-MSCs and the BM-MSC proliferation rate were increased in the absence of p38α. This seeming discrepancy may be explained by the fact that mouse bone marrow may contain MSCs that outnumber the bone multicellular units, where new bone formation is occurring. Although p38α ablation leads to overproliferation of BM-MSCs, a large portion of these cells may not be actively involved in bone formation. Thus, the number of BM-MSCs is not directly translated into the number of osteoblasts. We found that the function of BM-MSC-expressed p38α is not limited to osteoblastogenesis and bone formation. We show that p38α-controlled, CREB-mediated OPG production by BM-MSCs is critical for coupling bone formation and resorption. This is in contrast to the findings that ablation of p38α or p38β in osteoprogenitors/osteoblasts or Dermo1+ cells failed to affect osteoclastogenesis or bone resorption (Greenblatt et al., 2010; Thouverey and Caverzasio, 2012). Thus, only p38α expressed in BM-MSCs but not in osteoblasts is involved in communicating with monocytes/osteoclasts under physiological conditions. One explanation is that in a bone remodeling unit, the newly recruited BM-MSCs may secrete OPG to immediately inhibit further osteoclastogenesis, whereas osteoblast-secreted OPG molecules are embedded in the newly formed bone matrix and are not accessible to monocytes/osteoclasts. Alternatively, osteoblasts may have pathways redundant for p38α-OPG to communicate to osteoclastogenesis under physiological conditions. The p38-CREB-OPG pathway in BM-MSCs is also under the control of estrogen. We show that p38α MAPK can be activated by estrogen and that p38α is also required for optimal expression of ERα. Activated p38α mediates estrogen-induced OPG expression in BM-MSCs, which helps to maintain the bone mass under physiological conditions. Estrogen deficiency leads to a decrease in p38α MAPK activation and OPG synthesis, leading to an increase in osteoclastogenesis and bone resorption (Figure 6H). As such, Prx1+ BM-MSC-specific deletion of p38α renders the mice resistant to estrogen deficiency-induced bone loss. Interestingly, a recent study reported that mice with p38α ablated in osteoblasts are also resistant to ovariectomy-induced bone loss, likely via RANKL and interleukin 6 (IL6) (Thouverey and Caverzasio, 2015b). These results confirm the important role for p38α in BM-MSCs and osteoblasts in communicating to osteoclastogenesis under estrogen-deficient conditions, although these two cells may use different mechanisms. While recent studies suggest that estrogen may promote osteoclast apoptosis and thus inhibit bone resorption to preserve the bone (Krum et al., 2008; Nakamura et al., 2007), these findings suggest that the contribution of BM-MSC/osteoblast-osteoclast coupling to the pathogenesis of postmenopausal osteoporosis may be greater than previously thought.