Postmenopausal osteoporosis (PMO) represents a prevalent metabolic bone disorder characterized by a decline in bone mass and an increase in fragility, which significantly heightens the likelihood of fractures (Xu et al., 2017). The etiology of PMO is multifaceted, with estrogen deficiency being the primary contributor, resulting in bone loss among postmenopausal women (Xiao et al., 2024). In cases of PMO, diminished estradiol (E2) synthesis disrupts the equilibrium of bone metabolism. This alteration affects the signaling pathways of estrogen receptors (ER), leading to a reduction in osteoblast function and an enhancement in osteoclast activity (Shih et al., 2019). The loss of estrogen instigates inflammatory responses within the bone microenvironment, culminating in accelerated bone loss and osteoporosis in nearly half of postmenopausal women (Zhang et al., 2020). Consequently, PMO emerges as a critical public health concern, underscoring the necessity for effective and safe therapeutic interventions. A thorough understanding of the underlying mechanisms contributing to PMO is imperative for the development of both preventive and therapeutic pharmacological agents.

Parkinson’s disease (PD) is a widespread neurological condition, with estrogen deficiency recognized as a contributing risk factor (Nathan et al., 2012). The absence of estrogen has been shown to exacerbate PD symptoms in postmenopausal females. Research indicates that estrogen therapy can mitigate early symptoms associated with PD in women (Parkinson Study Group POETRY Investigators, 2011).

Although PMO is classified as a metabolic bone disease and PD as a neurodegenerative disorder, both conditions are intricately linked through the commonality of estrogen deficiency. Despite their differing phenotypic presentations and pathological mechanisms, PMO and PD may share analogous molecular pathways that involve similar regulatory genes. The interplay of these shared mechanisms may significantly influence the pathogenesis and progression of PMO.

Co-expression gene analysis examines gene expression datasets derived from diverse samples with the objective of identifying genes that exhibit co-expression or regulation across various diseases and biological contexts. This analytical approach facilitates the understanding of how genes function synergistically and the regulatory frameworks they share. By scrutinizing the expression profiles of estrogen-related genes in both PMO and PD, we can identify critical genes and pathways uniquely associated with PMO. Such analyses may uncover potential interconnections between PMO and PD, thereby paving the way for novel research avenues and therapeutic strategies aimed at managing PMO (Figure 1).

In the last 10 years, a notable correlation has been identified between estrogens and the development of several health issues, particularly PMO and PD (Nathan et al., 2012). Women in the postmenopausal phase are particularly susceptible to osteoporosis owing to their diminished estrogen levels (Liu et al., 2014). It is estimated that approximately 50% of females aged over 50 will experience fractures linked to osteoporosis (Yang et al., 2022). Decreased levels of estrogen enhance the activity of osteoclasts, which in turn accelerates the process of bone resorption, resulting in a loss of bone tissue (Andrade et al., 2017).

A deficiency in estrogen leads to modifications in the expression of several cytokines, including tumor necrosis factor-alpha (TNF-α), interleukin-1 (IL-1), and interleukin-6 (IL-6). Increased concentrations of these cytokines facilitate osteoclastogenesis and intensify the process of bone resorption (Iantomasi et al., 2023).

In the current investigation, three distinct groups of DEGs were recognized, which exhibited substantial enrichment in various BP. These processes encompass cell proliferation, the differentiation of osteoblasts, modulation of cell death, regulation of SMAD protein phosphorylation restricted by pathways, cellular responses to interleukin-6, the BMP signaling pathway, signal transduction mediated by p53 class proteins, aging, and the cell cycle. Furthermore, the KEGG pathway analysis demonstrated notable enrichment of DEGs across multiple pathways, including the p53, TGF-beta, cell cycle, FoxO, cellular senescence, Rap1, cAMP, Ras, MAPK, and PI3K-Akt signaling pathways. Growing evidence indicates that estrogens could play a protective role against PD. Both epidemiological and clinical research imply that estrogens may operate as anti-apoptotic agents, antioxidants, or neurotrophic factors, thereby modulating PD and enhancing their interaction with neurotrophic factors (Shulman, 2002). Research involving animal models demonstrates that estrogens affect the synthesis, release, and metabolism of dopamine (Saunders-Pullman, 2003). Importantly, postmenopausal women exhibiting diminished bone mineral density (BMD) are at an elevated risk for developing PD, a risk that can be mitigated through the use of anti-osteoporosis medications (AOMs) (Park K. Y. et al., 2023). These observations suggest a potential link between PMO and PD, indicating that certain genes may play crucial regulatory roles. The analysis identified three groups of DEGs that were significantly enriched in biological processes, including cellular responses to reactive oxygen species, aging, apoptotic signaling pathways, and programmed cell death. These findings imply that the identified genes are intricately connected to the pathogenesis of PD. The study identified rs3796661 as a common genetic variant between OP and PD, suggesting a genetic link between the two conditions. rs3796661 is an intronic variant of the SNCA gene located on chromosome 4q22.1. Abnormal accumulation of α-synuclein is a pathological hallmark of PD. This protein normally regulates synaptic vesicle trafficking and subsequent neurotransmitter release in neurons (Wong and Krainc, 2017). The SNCA gene encoding α-synuclein was identified as the first risk locus for PD in a large-scale GWAS (Simón-Sánchez et al., 2009). In addition, studies outside the central nervous system have shown that SNCA is highly expressed in mature red blood cells, with protein levels remaining elevated in both bone marrow and circulating mature red blood cells (Nakai et al., 2007). Red blood cells derived from Snca−/− bone marrow have lower levels of reactive oxygen species (ROS) and oxidative stress (Renella et al., 2014). Increased oxidative stress has been shown to promote bone loss induced by ovariectomy, suggesting that Snca is a key regulator of bone homeostasis. Its absence results in a 40% reduction in the magnitude of estrogen deficiency-induced bone loss (Calabrese et al., 2016).

Research has demonstrated that insufficient levels of estrogen can lead to the disruption of specific microRNAs (miRNAs) that play a critical role in bone metabolism. The dysregulation of these miRNAs has a significant impact on the process of bone remodeling by modifying gene expression and influencing various signaling pathways (Iantomasi et al., 2023). For instance, miR-21-5p has been identified as a target of SKP2, which is instrumental in the regulation of osteoclast differentiation (Huang et al., 2021). In patients diagnosed with PMO, the expression of miR-146a is notably reduced, which in turn affects bone metabolism through its modulation of the Wnt/β-catenin signaling pathway (Liu et al., 2021). Furthermore, this investigation identified several miRNAs that are closely linked to hub genes, including hsa-mir-15a-5p (He et al., 2021), hsa-mir-16-5p (Wang et al., 2023), hsa-mir-25-3p (Feng et al., 2022), hsa-mir-32-5p (Wang H. J. et al., 2021), and hsa-mir-92a-3p (Zheng et al., 2024) with a particular emphasis on their relationship with SMAD7.

TFs play a key role in regulating the transcription of target genes, making their levels useful for identifying potential biomarkers. This study identified relevant TFs as regulators of DEGs associated with the pathogenesis of PMO and PD through the TF-mRNA interaction network. NFIC is a critical transcription factor in organ development and stem cell biology, enhancing the reprogramming efficiency of pluripotent stem cells induced by mineralised tissues and maintaining the stem cell niche in mineralised tissues (Lee et al., 2022). Studies have also identified NFIC as an important regulatory transcription factor in neurodegenerative processes (Rahman et al., 2019). NFYA plays a role in the regulation of neurodegenerative diseases and haematopoietic stem cells (Bungartz et al., 2012; Henderson et al., 2021). E2F1 regulates the cell populations essential for bone repair and plays a distinct role in bone formation and repair (Premnath et al., 2024). E2F1 has also been shown to alter cell viability, induce neurotoxicity and have implications in PD (Xu et al., 2005). E2F6 counteracts the pro-apoptotic activity of E2F1 (Kikuchi et al., 2007). TFAP2A affects neuronal function and survival and may play a role in the pathogenesis of neurodegenerative diseases (Hu Y. et al., 2020). TFAP2A is also involved in osteoblast activity (Chen et al., 2018). TP53 can interfere with normal angiogenesis, glycolysis and apoptosis and may play a role in the pathogenesis of osteoporosis and Parkinson’s disease (Ge et al., 2021; James et al., 2024). NRF1 has been identified as a trans-activator of human dopamine transporters, the dysfunction of which is associated with psychiatric and neuronal disorders (Zhao et al., 2019). NRF1 has been found to play a distinct role in the regulation of antioxidant enzymes during inflammation-induced oxidative stress in osteoblasts (Park et al., 2016). EGR1 inhibits the downregulation of RANKL and sclerostin gene expression, resulting in reduced levels of these proteins in osteoblasts (Miura et al., 2024). EGR1 has also been implicated in the neuroprotection of motor function and dopaminergic neurons in Parkinson’s disease (Guo et al., 2023).

Unlike TFs, microRNAs act post-transcriptionally to regulate gene expression. They are single-stranded RNAs that bind to target mRNAs, causing cleavage and a reduction in expression. These miRNAs offer several advantages as non-invasive biomarkers detectable in body fluids such as urine and saliva, making them promising for biomarker development. While hsa-miR-15a-5p has not been reported in PMO, it is upregulated in multiple myeloma, another bone disease, and is also a biomarker of ageing and cognitive impairment in the elderly (Li et al., 2020; Ye et al., 2020). hsa-miR-16-5p serves as a potential biomarker for predicting duloxetine response in major depressive disorder, suggesting its role in neurological disorders (Kim et al., 2019). Studies also suggest that hsa-miR-16-5p is associated with osteoclast activity, with its expression levels potentially reflecting the degree of bone resorption (Qin et al., 2016). hsa-miR-20a-5p, an NHL-associated miRNA within the hsa-miR-17–92 cluster, regulates B-cell development and central tolerance, but its links to PMO and PD remain unexplored (Fernandes et al., 2022). Research has shown an association between hsa-miR-25-3p and postmenopausal femoral neck BMD (Faraldi et al., 2024). hsa-miR-25-3p has been found to affect biological pathways associated with neuropsychiatric disorders (Van der Auwera et al., 2022). No studies have investigated the relationship between hsa-miR-92a-3p and PMO or PD.

SMAD7 serves a crucial inhibitory function within the TGF-β signaling pathway, categorizing it as a member of the inhibitory Smad (I-Smad) family. Its mechanism of action involves binding to the TGF-β receptor, which effectively obstructs the phosphorylation processes of Smad2 and Smad3. The role of SMAD7 extends across a variety of physiological activities, notably in regulating cellular proliferation, differentiation, and apoptosis (Yan et al., 2009). Recent studies suggest a potential correlation between SMAD7 and conditions such as PMO and PD (Yano et al., 2012; Karampetsou et al., 2022). Notably, the persistent expression of Smad7 in osteoblasts has been shown to significantly diminish both the proliferation and differentiation of these cells (Yano et al., 2012). Furthermore, bone marrow stromal cells (BMSCs) derived from Smad7 knockout (KO) mice exhibited a diminished capacity for bone formation, coupled with an enhanced propensity for adipogenesis and osteoclast differentiation (Li et al., 2014). Investigations have revealed that estrogen contributes to bone protection, in part through the modulation of TGF-β signaling, a pathway integral to the understanding of osteoporosis progression. The reduction of estrogen levels post-menopause may consequently impact the expression or functional activity of SMAD7, thereby exacerbating the progression of osteoporosis (Hioki et al., 2022). Additionally, the estrogen-related receptor β (ESRRB) has been identified as a transcriptional activator of SMAD7, serving to inhibit the phosphorylation of SMAD2/3 and its subsequent nuclear translocation (Li and Zheng, 2023). Moreover, emerging evidence suggests a potential linkage between the neurodegenerative processes underlying PD and the TGF-β signaling pathway (Karampetsou et al., 2022). Consequently, SMAD7 presents as a promising target for therapeutic interventions aimed at addressing postmenopausal osteoporosis and Parkinson’s disease.

CCNB1, or Cyclin B1, serves as a pivotal regulatory protein within the cell cycle, predominantly exerting its influence during the G2/M transition, thereby governing the process of cell mitosis (Chang et al., 2016). In osteocytes, CCNB1 collaborates with cyclin-dependent kinase 1 (CDK1) to facilitate the progression of cells into mitosis (Sun et al., 2019). A deficiency in CCNB1 within the context of PMO has been shown to diminish osteoblast proliferation and hinder the formation of new bone (Li et al., 2024). Furthermore, CCNB1 is also crucial in modulating the cell cycle in neural cells (Yan and Ziff, 1995). Consequently, enhancing the expression of CCNB1 or refining cell cycle regulation may bolster the proliferation of both osteoblasts and neuronal cells, thereby fostering the recovery of bone mass, decelerating the onset of osteoporosis, and alleviating the manifestations of PD.

The Wilms’ Tumor 1 (WT1) gene serves primarily as a tumor suppressor, with its encoded protein playing a crucial role in the regulation of cellular processes such as proliferation, differentiation, and apoptosis. In the context of PMO, the aberrant overexpression of WT1 has been associated with osteoblast apoptosis, resulting in reduced bone formation, heightened bone loss, and an exacerbation of disease progression (Wang C. et al., 2021). Notably, elevated levels of WT1 protein are found in osteoclast precursors; however, this expression significantly diminishes as osteoclastogenesis occurs. Additionally, increased expression of WT1 antisense RNA has been shown to facilitate osteoclastogenesis and enhance the differentiation stability of osteoclasts (Li et al., 2016). Furthermore, WT1 is integral to the differentiation processes of nerve cells (Sarfstein and Werner, 2006). In neuronal contexts, elevated WT1 levels correlate with the promotion of neuronal apoptosis, whereas diminished WT1 levels are associated with its suppression (Liu et al., 2001).

The gene-disease radial plot visually represents how genes are expressed or associated with different diseases. The shape of the polygon, including its peaks and valleys, shows unique gene expression patterns across various diseases. This radar plot displays the expression patterns of the genes WT1, CCNB1, and SMAD7 in relation to PMO and PD. Inference scores reveal that SMAD7 is more strongly associated with PMO and MD, whereas CCNB1 shows a closer association with BR, MBD, and DBD. SMAD7 exhibits the strongest association with PD, NM, AD, and Stroke. In relation to Memory Disorders (MD), CCNB1 is the most closely related gene.

In summary, we initially identified 11 DEGs by comparing samples from PMO, PD, and GREs. Subsequent PPI analysis allowed us to narrow this list down to 3 hub genes, which were then validated through ROC. Functional enrichment analysis indicated that these genes play a role in several important pathways. Specifically, they are involved in the regulation of the p53 signaling pathway, the TGF-beta signaling pathway, the cell cycle, and the FoxO signaling pathway.

This study presents several important limitations. One limitation of this study is the small sample size in both the training and validation sets. Future studies should use larger sample sizes to improve result reliability. Secondly, the study lacked prognostic analysis due to the GEO database’s limited patient information and prognostic data. This limitation hinders a deeper examination of patient outcomes. Additionally, while we identified hub genes with promising diagnostic potential for PMO and PD, the specific mechanisms of these genes in different datasets are still unclear.