ACE2 Loss May Tie Parkinson’s Disease to Bone Loss Through Shared Brain and Bone Pathways

Scientists probing why people with Parkinson’s disease so often suffer fragile, fracture-prone bones have uncovered new molecular evidence that the two conditions may be linked by a shared signaling network spanning the brain and the skeleton. A new preclinical study, published in Molecular Genetics and Genomics, reports that loss of the ACE2 protein—the same receptor famous for its role in SARS-CoV-2 infection—worsens parkinsonian symptoms in mice while simultaneously disrupting bone metabolism through parallel changes in Wnt, β-catenin, BMP, IGF-1, and inflammatory RANKL/RANK/OPG signaling pathways. The findings, generated by Tingting Liu, Yuheng Ren, Xinghua Tian, and Jianshe Wei at Henan University, add weight to an emerging concept in neuroscience and skeletal biology: the bone-brain axis, a bidirectional communication system in which skeletal hormones influence brain function and neural activity shapes bone remodeling.

The clinical backdrop to the work is well established. Epidemiological studies have repeatedly shown that patients with Parkinson’s disease face a strikingly elevated risk of osteoporosis and osteoporotic fractures, a burden that exceeds what can be explained by poor mobility, falls, or age alone. Meta-analyses cited by the researchers indicate high rates of osteoporotic fracture in Parkinson’s disease, and cross-sectional clinical work has associated biomarkers such as serum uric acid with reduced bone mineral density in affected patients. Yet the molecular mechanisms that bind neurodegeneration to bone loss have remained murky. The Henan University team set out to interrogate that relationship experimentally, asking whether a single genetic factor—absence of ACE2—could simultaneously perturb dopaminergic neuron survival in the brain and bone homeostasis in the skeleton.

To do so, the researchers used a well-characterized mouse model of parkinsonism in which the neurotoxin MPTP, or 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, selectively destroys dopamine-producing neurons in the substantia nigra. Crucially, they crossed this challenge with mice lacking a functional ACE2 gene—specifically Ace2-null males, designated Ace2−/y—because ACE2 sits at the center of the protective arm of the renin-angiotensin system, converting angiotensin II into angiotensin-(1–7), a peptide with documented anti-inflammatory and neuroprotective actions. Previous studies from other groups and from the same team had shown that ACE2 activation mitigates behavioral deficits and neuroinflammation in chemically induced Parkinson’s models, and that the ACE2/Ang-(1–7)/Mas cascade strengthens bone structure and metabolism. The new study asked what happens when this protective factor is removed entirely under parkinsonian stress.

Behavioral testing revealed a clear aggravating effect. MPTP exposure significantly worsened motor dysfunction and depression-like behaviors in the mice, and the combination of MPTP toxicity with ACE2 deficiency produced a particularly severe pathological picture in the brain. Immunohistochemistry, Western blotting, and histopathological staining showed reduced activity of dopaminergic neurons and heightened microglial activation—the inflammatory response of the brain’s resident immune cells. At the molecular level, the researchers measured elevated levels of total α-synuclein, the misfolding-prone protein that defines Parkinson’s pathology, alongside increased abundance of Caspase-3 and Bax, two canonical executioners of programmed cell death. Together, these markers indicate that ACE2 loss intensifies both the protein aggregation burden and the apoptotic pressure on vulnerable neurons.

The bone findings were equally striking, and notably they emerged in parallel rather than secondarily. In the skeletal tissue of the Ace2-deficient mice, the team documented reduced abundance of total Wnt ligands, β-catenin, bone morphogenetic proteins (BMP), and insulin-like growth factor 1 (IGF-1), along with diminished phosphorylation ratios of the downstream kinases that transmit these signals. This matters because each of these cascades is a cornerstone of bone formation: Wnt/β-catenin signaling drives osteoblast differentiation and bone accrual, BMPs are potent inducers of bone formation used clinically in spine fusion and fracture repair, and IGF-1 couples muscle and bone metabolism through mTOR-dependent pathways. The researchers are careful to note that the parallel reduction of these signaling proteins suggests potential perturbation of the cascades rather than definitive proof of pathway failure, a distinction that reflects appropriate scientific caution.

In the opposite direction, ACE2 deficiency upregulated mediators of the RANKL/RANK/OPG axis, a triad that governs osteoclast formation and bone resorption. RANKL binding to RANK on osteoclast precursors drives the differentiation of bone-resorbing cells, while OPG acts as a soluble decoy receptor that restrains the process. Dysregulation of this axis tilts bone turnover toward net loss. Intriguingly, the same axis operates in the brain, where it has been identified as a critical inflammatory signaling system in ischemic injury, and Rho GTPases downstream of these pathways modulate osteoclast differentiation directly. The coordinated shift of this inflammatory skeletal axis in both brain and bone tissue under ACE2 deficiency is one of the study’s most suggestive observations, hinting at a common pathological language spoken by the two organs.

To move from candidate pathways to gene-level targets, the team turned to transcriptomics. They mined public GEO datasets and applied weighted gene co-expression network analysis, or WGCNA, a computational method that groups genes into modules based on correlated expression patterns and identifies the hub genes most central to disease-associated modules. This analysis pinpointed ten hub genes, including DNM1, which encodes dynamin 1, a protein essential for synaptic vesicle recycling; OCRL, a phosphatidylinositol phosphate phosphatase linked to the oculocerebrorenal syndrome; and OPA1, a mitochondrial fusion protein whose mutations cause dominant optic atrophy and which has been implicated in mitochondrial parkinsonism through stem cell modeling. The dysregulation of these genes was linked to synaptic dysfunction and inflammation—two processes squarely at the heart of Parkinson’s pathophysiology.

The team then evaluated whether these hub genes could serve as diagnostic biomarkers. Using receiver operating characteristic, or ROC, analysis on public single-disease transcriptome datasets for Parkinson’s disease and osteoporosis separately, they found that the core gene signatures achieved areas under the curve ranging from 0.683 to 0.981, indicating diagnostic accuracy that spans moderate to near-perfect discrimination. Functional enrichment of the core genes pointed to involvement in synaptic signaling, MAPK signaling, and the Rap1 and Ras pathways—small GTPase cascades that regulate cell proliferation, differentiation, and cytoskeletal dynamics in both neurons and bone cells. Such dual-diagnostic performance, if replicated in human cohorts, would suggest that a shared molecular signature underlies both conditions and could be exploited clinically to identify patients at risk of combined neurodegenerative and skeletal decline.

The authors are appropriately measured in their claims. They emphasize that these are preclinical findings obtained under short-term MPTP treatment in growing young male mice, meaning that the observed bone metabolic disturbance was transient and that the results may not translate directly to elderly human patients, in whom Parkinson’s disease typically manifests and in whom bone loss is chronic and sex-dependent. They also stress that the coordinated dysregulation observed in brain and bone is consistent with a bone-brain axis pathological phenotype, but that the current experimental design cannot confirm causal bidirectional cross-talk between the tissues. Distinguishing whether ACE2 deficiency independently damages both organs, or whether pathology in one propagates to the other—perhaps through circulating osteocalcin, sympathetic nervous system output, or inflammatory mediators—will require interventional studies that manipulate one tissue and measure the other.

Even with those caveats, the study offers a compelling framework and a set of concrete targets for follow-up. Restoring ACE2 activity or mimicking its product, angiotensin-(1–7), has already shown neuroprotective effects in experimental Parkinson’s models, including reduced α-synuclein expression through the NEAT1/miR-153-3p axis, and ACE2 activation has been reported to promote hippocampal neurogenesis via Wnt/β-catenin signaling. The present results raise the possibility that such therapies could carry a skeletal benefit as well, protecting against the osteoporosis that so often compounds the disability of Parkinson’s disease. The ten hub genes, meanwhile, provide a molecular shortlist for mechanistic validation, and their diagnostic AUC values justify testing in human blood or tissue datasets. As the bone-brain axis matures from a descriptive concept into a mechanistic research program, work like this demonstrates how a single molecule, studied across two organs at once, can illuminate disease connections that medicine has long observed clinically but struggled to explain at the level of genes and signaling pathways.

Subject of Research: ACE2-dependent molecular mechanisms linking Parkinson's disease neurodegeneration and bone metabolic alterations via the bone-brain axis

Article Title: ACE2 and Parkinsonism‑related bone metabolic alterations: signaling pathways and hub gene analysis

Article References: Liu, T., Ren, Y., Tian, X., & Wei, J. (2026). ACE2 and Parkinsonism‑related bone metabolic alterations: signaling pathways and hub gene analysis. Molecular Genetics and Genomics, 301(1), Article 185. https://doi.org/10.1007/s00438-026-02511-2

Image Credits: AI Generated

DOI: 10.1007/s00438-026-02511-2

Keywords: Parkinson's disease, ACE2, bone metabolism, bone-brain axis, osteoporosis, Wnt/beta-catenin signaling, RANKL/RANK/OPG, IGF-1, hub genes, WGCNA, alpha-synuclein, MPTP mouse model