Bone marrow adipocytes support hematopoietic stem cell survival

Abstract

In bone marrow (BM), hematopoietic elements are mingled with adipocytes (BM-A), which are the most abundant stromal component in the niche. BM-A progressively increase with aging, eventually occupying up to 50% of BM cavities. In this work, the role played by BM-A was explored by studying primary human BM-A isolated from hip surgery patients at the molecular level, through microarray analysis, and at the functional level, by assessing their relationship with primary human hematopoietic stem cells (HSC) by the long-term culture initiating cell (LTC-IC) assay. Findings demonstrated that BM-A are capable of supporting HSC survival in the LTC-IC assay, since after 5 weeks of co-culture, HSC were still able to proliferate and differentiate. Furthermore, critical molecules such as C-X-C motif chemokine 12 (CXCL12), interleukin (IL)-8, colony-stimulating factor 3 (CSF3), and leukaemia inhibitory factor (LIF), were expressed at similar levels in BM-A and in primary human BM mesenchymal stromal cells (BM-MSC), whereas IL-3 was higher in BM-A. Interestingly, BM-A displayed a different gene expression profile compared with subcutaneous adipose tissue adipocytes (AT-A) collected from abdominal surgery patients, especially in terms of regulation of lipid metabolism, stemness genes, and white-to-brown differentiation pathways. Accordingly, analysis of the gene pathways involved in hematopoiesis regulation showed that BM-A are more closely related to BM-MSC than to AT-A. The present data suggest that BM-A play a supporting role in the hematopoietic niche and directly sustain HSC survival.

Yellow adipocytes comprise a new adipocyte sub-type present in human bone marrow

“…This conclusion is supported by unbiased proteomic approaches that indicate a seemingly unidentified cholesterol-oriented metabolism. In contrast to our results, a recent transcriptomic study comparing gene expression of human BM-Ad isolated from the femoral head and SC-Ad show that genes over-represented in human BM-Ad participate in signaling pathways without clear differences in the enzymes involved in lipid metabolism (such as cholesterol metabolism and TG hydrolysis) (Mattiucci et al, 2018). In addition, this report found decreased adiponectin expression, which stand in contrast to our current results and another study that identified BM-Ad as an important source of adiponectin (Cawthorn et al, 2014).…”

“…However, it is unclear whether these differentiated cells recapitulate the phenotype of mature human primary BM-Ad. These in vitro studies suggest a role for BM-Ad in hematopoiesis regulation (Mattiucci et al, 2018; Naveiras et al, 2009), bone remodeling (Hardaway et al, 2015) and cancer progression (Diedrich et al, 2016; Herroon et al, 2013; Liu et al, 2015; Shafat et al, 2017; Tabe et al, 2017). These issues highlight that our knowledge of the physiological phenotype of primary BM-Ad remains limited.…”

Exercise Degrades Bone in Caloric Restriction, Despite Suppression of Marrow Adipose Tissue (MAT)

“…Adipocytes in the bone marrow serve critical roles, regulating homeostasis in hematopoietic niches (63,64) and storing energy for use during exercise. (17) An increase in marrow fat is found during conditions associated with osteoporosis-aging and estrogen deficiency (12) -and the same has been suggested in the unique bone fragility that accompanies caloric restriction.…”

Bone marrow adipose tissue is a unique adipose subtype with distinct roles in systemic glucose homeostasis

“…To determine if similar differences occur in humans, we next analysed the transcriptomes of adipocytes isolated from human femoral BMAT and subcutaneous WAT, which our previous analyses revealed to be globally distinct (Mattiucci et al, 2018). Consistent with our findings in rabbits, human BMAds were not enriched for brown or beige markers and had decreased expression of genes relating to glucose metabolism and insulin responsiveness (Fig.…”

“…1E, Supplemental Fig. 2A-B), ethical approval and subject characteristics are as described previously (Mattiucci et al, 2018). For human subjects in cohort 2 (Fig.…”