Alcohol‑Induced Liver Fibrosis Alters Bone via BMP Signaling: Modeling Hepatic Osteodystrophy in vitro

Abstract

In this study, we established a novel three-dimensional in vitro liver–bone co-culture system that simulated the physiological interactions between liver and bone tissues at the tissue structural and functional levels. The model consists of liver microorganoids composed of hepatocytes (HepaRG), hepatic stellate cells (LX-2), and human umbilical vein endothelial cells (HUVECs) in a 4:2:1 ratio, co-cultured with scaffolds embedded with a human monocytic cell line (THP-1) and mesenchymal stem cell line (SCP-1) on agarose plate. This system demonstrated stable viability and functional for at least 28 days, offering a reliable platform for studying liver–bone crosstalk in vitro. Numerous clinical and animal studies have demonstrated that alcohol intake induces osteoporosis and delays fracture healing. However, the underlying pathological mechanisms remain unclear. Since alcohol mainly affects the liver, and hepatic dysfunction is closely associated with osteodystrophy, it is plausible that liver-driven mechanisms are key contributors to alcohol-related bone disease. This study provides a systematic evaluation of how alcohol exposure affects the liver–bone axis. Specifically, we simulated alcohol exposure by exposing our system to an alcohol concentration that would result from drinking more than four beers (50 mM). This exposure significantly inhibited the bone-forming capacity, as evidenced by a decrease in AP and PINP expression, accompanied by downregulation in the expression of a key osteogenic transcription factor, RUNX2, and upregulation of the chondrogenic marker gene SOX9. Further mechanistic investigation revealed that this effect may be closely related to alteration of the BMP signaling pathway. Specifically, liver microorganoids exposed daily to alcohol showed significantly reduced expression and secretion of BMP-2, which is closely associated with osteogenesis, and markedly increased expression of BMP-13, a change that drives the shift from osteogenic toward chondrogenic differentiation. Although BMP-9 did not show significant changes, this protein exists in different forms stabilized by disulfide bonds, and the correct formation of these bonds is crucial for its biological activity. Alcohol intake has been reported to alter the expression and activity of protein disulfide isomerases in the liver, which are essential for the proper formation of disulfide bonds in secreted proteins. This suggests that, even if the total level of BMP-9 remains unchanged, alcohol exposure may disrupt the balance between its different molecular forms by affecting disulfide bond formation, thereby altering the ratio of D-form BMP-9 to M-form BMP-9. Overall, these findings suggest that these changes affect bone homeostasis along the liver–bone axis and highlight the key role of liver-derived factors in modulating bone metabolism under alcohol-induced pathological state. In conclusion, we have established a long-term, stable in vitro liver fibrosis model by exposing the liver microorganoids to alcohol and have determined the potential molecular mechanism by which alcohol disrupts bone homeostasis through the liver–bone axis. These findings provide a new explanation for alcohol-related bone metabolism disorders and offer a theoretical basis and potential therapeutic targets for the clinical management of alcoholic hepatic osteodystrophy. In addition, our in vitro liver–bone co-culture model can be stably maintained for up to 28 days, offering a controllable and expandable research platform. This model has broad application prospects, including the screening of drug-induced osteotoxicity, investigation of the etiology of liver dysfunction-associated bone diseases, and personalized efficacy and toxicity prediction studies.