Figure 5

Secondary effects of apoA-1 deficiency on bone and MSC development. All graphs indicate mean±s.d. (a) Quantitative analysis of mRNAs important in MSC, adipocyte, and osteoblast differentiation. All studies here had n=4 for both ApoA-1^−/− and wild-type mRNA isolates. The mRNA levels of Cxcl12, Cxcr4, Anxa2, another regulator of stem cell homing to the osteoblast niche, were significantly reduced in the ApoA-1^−/− animal relative to the wild-type. The osteoblast-specific genes Runx2 and type I collagen Col1a1, were significantly reduced in bone marrow cells from null mice. Runx2 increased <25%, and this is particularly important, so this was studied at the protein level, (b and c) below panels. In contrast, mRNA expression of the Cxcl12 receptor, Cxcr4, was elevated 6.5-fold in apoA-1-deficient mice. The key nuclear receptor regulator of lipoblast differentiation, PPARγ was augmented significantly, but only 20% at the mRNA level in the apoA-1-deficient as compared with the wild-type C57BL/6 mice; this important receptor was also studied at the protein level (b and c, below panels). Two more noteworthy findings were that the CCAAT/enhancer-binding protein-α, cebpa was increased fourfold in the apoA-1-deficient mice; cebpa upregulation is a feature of osteoporosis in other contexts including glucocorticoid exposure. A surprising finding was a large increase in Slc9a3r1, the sodium-hydrogen exchange regulatory factor-1, which is expressed in osteoblasts during the mineral accumulation phase. (b) Expression of the important genes Pparγ and RunX2, significantly but moderately increased and decreased, respectively, at the mRNA level, was studied by Western blotting. Examples of blots and actin reblots are shown on the left, with densitometry summarizing six wild-type and nine control lysates all normalized to actin on the right. These studies showed that Pparγ increased 35%, and Runx2 decreased by half, in the apoA-1^−/− animals relative to the WT. (c) Expression of the important genes Pparγ and Runx2 was confirmed using flow cytometric detection of the proteins. On the left are shown histograms of count and intensity. On the right are summaries of quantitative data on mean PPARγ and Runx2 antibody labeling from five isolates of each genotype in PPARγ, or six WT and nine knock-out mice for Runx2. In parallel with our previous findings the protein levels of Runx2 were significantly reduced, while the protein levels of PPARγ were greatly elevated in the apoA-1 KO as compared with the WT mice.