Severity Level and Duration of Energy Deficit in Mice Affect Bone Phenotype and Bone Marrow Stromal Cell Differentiation Capacity.

Anorexia nervosa is known to induce changes in bone parameters and an increase in bone marrow adiposity (BMA) that depend on the duration and seriousness of the disease. Previous studies have found that bone loss is associated with BMA accumulation. Sirtuin of type 1 (Sirt1), a histone deacetylase that is partly regulated by energy balance, was shown to have pro-osteoblastogenic and anti-adipogenic effects. To study the effects of the severity and duration of energy deficits related to bone loss, a mouse model of separation-based anorexia (SBA) was established. We recently demonstrated that moderate body weight loss (18%) 8-week SBA protocol in mice resulted in an increase in BMA, bone loss, and a significant reduction in Sirt1 expression in bone marrow stromal cells (BMSCs) extracted from SBA mice. We hypothesised that Sirt1 deficit in BMSCs is associated with bone and BMA alterations and could potentially depend on the severity of weight loss and the length of SBA protocol. We studied bone parameters, BMA, BMSC differentiation capacity, and Sirt1 expression after induction of 4 different levels of body weight loss (0%,12%,18%,24%), after 4 or 10 weeks of the SBA protocol. Our results demonstrated that 10 week SBA protocols associated with body weight loss (12%, 18%, 24%) induced a significant decrease in bone parameters without any increase in BMA. BMSCs extracted from 12% and 18% SBA groups showed a significant decrease in Sirt1 mRNA levels before and after co-differentiation. For these two groups, decrease in Sirt1 was associated with a significant increase in the mRNA level of adipogenic markers and a reduction of osteoblastogenesis. Inducing an 18% body weight loss, we tested a short SBA protocol (4-week). We demonstrated that a 4-week SBA protocol caused a significant decrease in Tb.Th only, without change in other bone parameters, BMA, Sirt1 expression, or differentiation capacity of BMSCs. In conclusion, this study showed, for the first time, that the duration and severity of energy deficits are critical for changes in bone parameters, BMSC differentiation, and Sirt1 expression. Furthermore, we showed that in this context, Sirt1 expression could impact BMSC differentiation with further effects on bone phenotype.

Sirtuin 1 deficiency decreases bone mass and increases bone marrow adiposity in a mouse model of chronic energy deficiency.

Sirtuin of type 1 (Sirt1), a class III HDAC, is known to be involved in the regulation of differentiation of skeletal stem cells (SSCs) into osteoblasts and adipocytes. In caloric restriction, it has been shown that the expression and activity of Sirt1 is a tissue-dependent regulation. However, at present, no study has focused on the link between Sirt1, bone marrow adiposity (BMA) and osteoporosis related to anorexia nervosa (AN). Thus, the aims of this work were to (i) determine BMA and bone changes in a mouse model replicating the phenotypes of AN (separation-based anorexia model (SBA)); (ii) determine the expression of Sirt1 in bone marrow stromal cells (BMSCs) extracted from these mice and identify their differentiation capacities; (iii) study the effects of pharmacological activation and inhibition of Sirt1 on the osteoblastogenesis and adipogenesis of these cells and (iiii) delineate the molecular mechanism by which Sirt1 could regulate osteogenesis in an SBA model. Our results demonstrated that SBA protocol induces an increase in BMA and alteration of bone architecture. In addition, BMSCs from restricted mice present a down-regulation of Sirt1 which is accompanied by an increase in adipogenesis at expense of osteogenesis. After a 10-day organotypic culture, tibias from SBA mice displayed low levels of Sirt1 mRNA which are restored by resveratrol treatment. Interestingly, this recovery of Sirt1 levels also returned the BMA, BV/TV and Tb.Th in cultured tibias from SBA mice to normal levels. In contrast of down-regulation of Sirt1 expression induced by sirtinol treatment, stimulation of Sirt1 expression by resveratrol lead to a decrease in adipogenesis and increase in osteogenesis. Finally, to investigate the molecular mechanisms by which Sirt1 could regulate osteogenesis in the SBA model, the acetylation levels of Runx2 and Foxo1 transcription factors were determined. Our data show that this chronic energy deficiency in female mice causes a decrease in BMSC activity, resulting in critical changes to Runx2 and Foxo1 acetylation levels and thus to their activity. Altogether, these data suggest that Sirt1 could be considered as a potential therapeutic target in osteoporosis related to AN.

Cell-specific effects of TNF-α and IL-1ß on alkaline phosphatase: implication for syndesmophyte formation and vascular calcification.

Tumor necrosis factor (TNF)-α and interleukin (IL)-1ß stimulate tissue non-specific alkaline phosphatase (TNAP) activity and mineralization in cultures of vascular smooth muscle cells (VSMCs). They are, therefore, considered as stimulators of vascular calcification in the context of atherosclerosis and diabetes type 2. In contrast, although ankylosing spondylitis (AS) leads to the formation of syndesmophytes, which are ectopic ossifications from entheses (where ligaments, tendons and capsules are attached to bone), anti-TNF-α therapies fail to block bone formation in this disease. In this context, our aims were to compare the effects of TNF-α and IL-1ß on TNAP activity and mineralization in entheseal cells and VSMCs. Organotypic cultures of mouse ankle entheses were treated or not with TNF-α and IL-1ß for 5 days. Micro-computed tomography was performed to determine trabecular bone parameters, and histology to assess TNAP activity and mineralization. Human mesenchymal stem cells cultured in pellets in chondrogenic conditions and human VSMCs were also used to determine the effects of cytokines on TNAP activity and expression, measured by quantitative PCR. In organotypic cultures, TNF-α and IL-1ß significantly reduced the tibia BV/TV ratio. They also inhibited TNAP activity in entheseal chondrocytes in situ, and in mouse and human chondrocytes in vitro. In contrast, TNF-α stimulated TNAP expression and activity in human VSMCs. These differences were likely due to cell-specific effects of peroxisome proliferator-activated receptor γ (PPARγ), which is inhibited by TNF-α. Indeed, in human chondrocytes and VSMCs, the PPARγ inhibitor GW-9662 displayed the same opposite effects as TNF-α on TNAP expression. In conclusion, whereas TNF-α and IL-1ß stimulate TNAP activity in VSMCs, they inhibit it in entheseal cells in situ and on chondrocytes in vitro. The identification of PPARγ as a likely mediator of cytokine effects deserves consideration for future research on the mechanisms of ectopic ossification.