Translational boundaries of Txnip-targeted metabolic modulation in bone homeostasis Authors: Zhewei Ma, Yiquan Ji Published: 2026 Dear Editor, We were interested in reading the paper by Cao et al. titled “Txnip regulates glycolysis and tricarboxylic acid cycle balance to maintain bone homeostasis”. The authors have presented the following evidence showing that the thioredoxin-interacting protein (Txnip) is involved in skeletal homeostasis by coordinating glycolysis and the tricarboxylic acid (TCA) cycle. Specifically, the results show that alterations in Txnip may lead to defective bone remodeling and a delayed healing of fractures in older age, and inhibition of Txnip can be applied in the way of autologous bone marrow mesenchymal stromal cell (BMSC)-based regeneration strategies [1]. It is relatively good and suitable for translation. Txnip has not yet been recognized as one of the metabolic targets of bone regeneration, nor have the following three problems been addressed. First, the cell-type specificity of Txnip-mediated metabolic regulation needs to be better known. Coordinated activity of osteoblast-lineage cells, osteoclasts, osteocytes, chondrocytes, blood vessels and immune cells maintains bone homeostasis; it does not occur in a single cell compartment [2]. According to study, TXNIP is found in osteoblasts, osteoclasts, and chondrocytes and can change skeletal cell development and function through redox-dependent and redox-independent pathways. In addition, Txnip has been reported to regulate Rankl-mediated osteoblast-osteoclast crosstalk in aging- and diabetes-related osteoporosis, and TBP-2/TXNIP has been implicated in human osteoclast differentiation through redox-sensitive signaling [3,4]. Therefore, a uniform interpretation of Txnip inhibition may overlook some cells. Although Txnip reduction may increase the osteogenic capacity of senescent BMSCs, its effects on osteoclast differentiation and osteoblast-osteoclast coupling have been partly reported, whereas broader cell-context-dependent effects, including osteocyte-mediated remodeling, remain incompletely defined. Future research may prioritize BMSC/osteoblast-lineage cells and osteoclast-lineage cells as the most likely mediators of the major skeletal effects of Txnip, while subsequently extending the analysis to osteocytes and other bone-niche cells. Conditional Txnip modulation in osteoprogenitor, osteoclast-lineage and osteocyte compartments, in combination with co-culture or bone organoid models, can be used to identify the cell population in which Txnip is therapeutically beneficial rather than biologically compensatory. Second, the flux-level metabolic validation will be carried out on the proposed glycolysis-TCA cycle balance. Altered expression of metabolic enzymes, pathway enrichment or changes in the concentration of steady-state metabolites can all be considered indicative of altered directional carbon flow. Stable isotope tracking and extracellular flux analysis are increasingly employed in bone biology to differentiate glycolytic activation, mitochondrial oxidation, pyruvate entrance into the TCA cycle, and compensatory bioenergetic responses [5,6]. In the future, ^13C-glucose or ^13C-glutamine tracing can be used to determine whether Txnip inhibition has restored the coordination between glycolysis and the TCA cycle by measuring lactate release, ATP production rate, oxygen consumption rate and extracellular acidification rate; further incorporation of ^15N-glutamine or [U-^13C5,^15N2]-glutamine tracing would help determine whether Txnip modulation alters glutamine-derived carbon and nitrogen fluxes into TCA cycle intermediates and whether glycolysis-TCA cycle balance is coupled with glutamine anaplerosis [7,8]; otherwise, it may be that the cells have shifted to a stress-adaptive glycolytic state. Therefore, excessive or off-time glycolysis may not be beneficial in all circumstances, such as late-stage mineralisation or osteoclast activation. Third, one needs to pay attention to the system-level safety boundary of Txnip targeting. TXNIP is a regulator of skeletal metabolism that has also been revealed as a broader cellular stress node involved in thioredoxin-dependent redox regulation, oxidative stress, ER stress, NLRP3 inflammasome activation, pyroptosis and glucose homeostasis. This pleiotropy is also a weakness. In older patients with osteoporotic fractures who also have diabetes, chronic kidney disease, chronic inflammation or are taking systemic Txnip inhibitors, unpredictable changes in their glucose metabolism, inflammatory tone, host defence and tumour-suppressive pathways may occur [9]. Thus, local or transitory Txnip regulation may be easier than whole-body inhibition. Ex vivo pretreatment of autologous aged BMSCs, bone-targeted delivery systems, fracture-site scaffolds, or time-limited small-molecule release may be safer translational methods. System-wide glucose regulation, inflammatory cytokines, marrow niche integrity, and ectopic tissue reactions must be monitored. In short, Cao et al. have found a good metabolic route for Txnip that regulates skeletal health. To advance research on bone repair and promote more therapeutic applications in clinical practice, the following steps need to be taken: first, identify which type of skeletal cell is the main target of Txnip; second, confirm whether a change in glycolysis-TCA cycle flux has occurred; third, determine if only pathway modification has occurred; and finally, distinguish between local therapeutic effects and systemic metabolic or inflammatory side effects. To address the problems mentioned above, it will be necessary to reduce off-target effects of Txnip and enhance its accuracy as a target for aged bone regeneration. Declaration of competing interests The author(s) have no conflicts of interest relevant to this article.