The New Obesity Paradox: How Bone Marrow Immunosuppression Leads to Obesity-Related Bone Loss and Peripheral Inflammation

Bone marrow adipocytes have a critical role within the bone marrow niche. However, our understanding of bone marrow adipose tissue expansion with obesity and the role it plays in immune cell regulation and osteoclastogenesis is limited. In humans, we investigated bone marrow metabolism in response to acute overnutrition (high-calorie diet) and calorie deprivation (fasting). Participants underwent a 10-day high-calorie diet followed by a two-week interval of an ad libitum diet and then underwent 10 days of fasting. Bone marrow adipocytes, bone marrow serum, and peripheral blood serum were collected before and after each dietary intervention. Using comprehensive and integrated analyses, we characterized nutritional influences on bone marrow adiposity. Bone marrow adipocytes after the high-calorie diet phase showed an upregulation of FOXP3 (p < 0.0001), the transcription factor that controls the development of Tregs, which are critical in reducing inflammatory immune responses. After fasting, bone marrow adipocytes had an upregulation of inflammatory genes (CP, CFH, and IGFBP3) (p < 0.0001). Proteomic analysis after the high-calorie diet showed that bone marrow serum had an upregulation of proteins related to an inflammatory/complement pathway (PROC, RBP4, and CFI). After fasting, in bone marrow serum, there was a significant downregulation of inflammatory/complement pathway proteins (C1QC and RBP4). Despite both interventions causing bone marrow adipose tissue expansion, the mechanism for adipogenesis appears to be dependent on nutrient availability. After the high-calorie diet, lipid-mediated signaling and lipid storage, and lipid droplet biogenesis were significantly downregulated (p < 0.0001). In contrast, after fasting, lipid-mediated signaling and lipid storage, and lipid droplet biogenesis were significantly upregulated (p < 0.0001). Using a pre-clinical mouse model for diet-induced obesity, we also showed that the expansion of bone marrow adipocytes in obese mice promoted osteoclast differentiation and subsequently led to obesity-related trabecular and cortical bone loss through a stimulatory effect of the PD-1/PD-L1 axis. Bone marrow adipocytes isolated from obese mice had increased Mcp-1 expression, a key regulator of osteoclastogenesis and myeloid cell infiltration. With the increase in bone marrow adipose tissue-derived Mcp-1, we found an increase in immunosuppressive PD-L1+ myeloid cells. Despite these PD-L1+ myeloid cells having an inhibitory effect on T-cells, we found an increase in PD-1-expressing osteoclast precursors. The inhibition of PD-1/PD-L1 signaling during early osteoclastogenesis prevented myeloid cell commitment and resulted in decreased cell fusion, thus suggesting a stimulatory effect of PD-1/PD-L1 signaling within the myeloid lineage. Using a bone marrow adipocyte depletion mouse model (BMAd-Pparg KO), we demonstrated that obese BMAd-Pparg KO mice had decreased bone marrow PD-L1+ myeloid cells as well as decreased PD-1+ osteoclast precursors. The reduction in these cell populations resulted in fewer mature osteoclasts, subsequently leading to improved trabecular bone volume. Since osteoclasts are myeloid cell-derived, these results suggest bone marrow adipocytes are critical for the commitment and differentiation of myeloid cells to osteoclasts. Targeting bone marrow adipogenesis could ameliorate enhanced osteoclastogenesis and provide a novel approach to treat obesity-related bone loss.