Glucocorticoids modulate bone microvascular barrier function via Connexin43/MAPK pathway

Objective: This study sought to elucidate the modulation of bone microvascular barrier function via Connexin 43 (Cx43)/MAPK pathway, utilizing a glucocorticoid (GC)-induced osteoporosis model. Hypothesis: We hypothesized that the pathogenic mechanisms underlying GC-induced osteoporosis include cell adhesion mechanisms that mediate microvascular barrier function and osteoblast-endothelial interactions. Our study particularly focuses on gap junction molecule Cx43, which controls endothelial cell-cell contacts, angiogenesis, and bone cell differentiation. Methods: We engineered a novel 3D bicellular microfluidic platform consisting of a 3D perfusable endothelial channel embedded within collagen I-based matrix surrounded by human osteoblasts (HOBs). Endothelial cells (ECs) and HOBs were treated for 18 days in the presence of 200 nmol/L hydrocortisone (HYD), 100 nmol/L prednisone (PRED), and 100 nmol/L prednisolone (PSL). We first evaluated the osteogenic potential of osteoblasts under normal and GC conditions by alkaline phosphatase (ALP), alizarin red (AR), and Von kossa (VK) staining. Next, we examined the role of GCs in microvascular barrier dysfunction by imaging in real time the diffusion of fluorescently labeled 70kDa dextran with confocal microscopy. Analysis of the mean intensities permitted us to determine the diffusive permeability coefficient (Pd). To investigate the mechanistic underpinnings, protein analysis by immunoblots was performed for phosphorylated and total Cx43 levels. For the in vivo studies, prednisolone pellets were implanted subcutaneously in the dorsal scapular region of 4-6 months-old Swiss Webster mice. 60 days post pellet implantation, the mice were euthanized, the long bones and skull harvested and subjected to histopathological analyses. Cx43 expression and bone microarchitecture were visualized by immunohistochemistry. Data and summary of results: Initially, we demonstrated that GCs decreased osteogenic function in vivo and in vitro. Our in vitro studies demonstrated that the presence of GCs reduced HOB functionality by reducing ALP activity (11-fold), calcium deposition (AR staining) by 4.5-fold, and mineral deposition (VK staining) by 6.4-fold compared to untreated conditions (p<0.05). In addition, our in vivo data showed 5-times decrease in mineral deposition based on VK staining, indicating that GCs diminished the osteoblastic function, coupled with a 10-fold decrease in the number of bone microvessels. Hence, we next investigated the role of GCs in bone microvasculature. Specifically, we showed that vascular leakiness (Pd) increased 5-fold (HYD), 4-fold (PRED), and 7.8-fold (PSL) compared to untreated conditions in vitro. Using the microfluidic device setup, we demonstrated that the osteoblast-endothelial cell interface integrity was disrupted via loss of junctional Cx43 upon GC treatment. Also, Cx43 blockade showed increased vascular leakiness by 4.5-fold (p<0.05), while bone tissues showed downregulation of Cx43 in PSL treated animals, indicating that Cx43 is essential for bone microvascular function. Finally, immunoblotting analysis of EC-HOB co-culture lysates showed that GCs induce activation of the MAPK pathway, which in turn regulates Cx43 phosphorylation, destabilizing cell-cell junctions. Conclusions: Overall, our findings offer mechanistic insights into GC-induced osteoporosis and may pave the way for mitigating drug-induced adverse effects on bone health. A deeper understanding of the intricate interplay between Cx43/MAPK, cell-cell interactions and bone microvasculature is crucial to divulge bone pathophysiology and improve outcomes for patients susceptible to bone disease. the NIH (R01DE031046, R21AR076497, R21CA294025), Georgetown Startup Funds (Assignee: 91252), the MISR/Lombardi Comprehensive Cancer Center grant (P30-CA051008). This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.