Single-cell and experimental analyses identify mitochondrial dysfunction–related genes Hmgcs2, Nudt5, and Cpt1c in painful diabetic peripheral neuropathy

Single-cell and experimental analyses identify mitochondrial dysfunction–related genes Hmgcs2, Nudt5, and Cpt1c in painful diabetic peripheral neuropathy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Single-cell and experimental analyses identify mitochondrial dysfunction–related genes Hmgcs2, Nudt5, and Cpt1c in painful diabetic peripheral neuropathy Jiaqi Cao, Zhongjie Zhang, Yu Chen, Xiaoyi Yu, Xuan Xiang, Wei Lu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8200654/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 14 You are reading this latest preprint version Abstract Painful diabetic peripheral neuropathy (PDPN) is a multifactorial disabling complication of diabetes, yet the pathogenesis involving mitochondrial dysfunction in immune cells remains unclear. This study aimed to identify mitochondrial dysfunction-related genes (MD-RGs) in PDPN and explore their mechanisms. Based on single-cell omics analysis, we identified three key MD-RGs (Hmgcs2, Nudt5, and Cpt1c) shared across different datasets and validated them in rat PDPN models. Gene set enrichment analysis revealed their association with cell polarity, sympathetic nerve development, and neurotransmitter pathways. We further observed distinct expression patterns of these genes in fibroblasts and macrophages within the dorsal root ganglion of PDPN. Hmgcs2 and Nudt5 were significantly downregulated in fibroblasts, while Cpt1c was upregulated. Conversely, Nudt5 and Cpt1c were significantly upregulated in macrophages, with Hmgcs2 downregulated. Functional enrichment analysis revealed that fibroblasts in PDPN primarily associated with polyamine/sulfate biosynthesis, while macrophages predominantly enriched for glycerol/choline metabolism, indicating distinct metabolic functions between the two cell types. Through molecular docking and dynamic simulation analysis, we further identified chlorothiazide and hydrochlorothiazide as stable binders to HMGCS2, suggesting potential as targeted therapeutics for PDPN. These findings indicate that Hmgcs2, Nudt5, and Cpt1c are key MD-RGs in the PDPN pathogenesis and provide novel therapeutic development strategies. Biological sciences/Cell biology Health sciences/Diseases Biological sciences/Neuroscience Painful diabetic peripheral neuropathy Mitochondrial dysfunction-related genes Hmgcs2 Nudt5 Cpt1c fibroblast Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1 Introduction Painful diabetic peripheral neuropathy (PDPN), characterized by chronic neuropathic pain associated with diabetes mellitus, represents one of the most common and debilitating complications of diabetes. Clinically, PDPN typically manifests as symmetrical distal limb pain or sensory abnormalities, but may also present as brachial or lumbosacral plexopathy, or mononeuropathy, all of which significantly reduce patients’ quality of life and impose substantial economic and social burdens[ 1 , 2 ]. Epidemiological studies indicate that up to half of diabetic patients will develop peripheral neuropathy during the course of the disease, and the prevalence increases with disease duration, poor glycemic control, and aging. Despite its high prevalence and clinical significance, PDPN remains difficult to treat effectively. The pathogenesis of PDPN is multifactorial, involving complex metabolic, immune, and microvascular mechanisms. Chronic hyperglycemia, dyslipidemia, and insulin resistance contribute to metabolic stress, oxidative damage, and inflammatory activation within the peripheral nervous system. Accumulating evidence indicates that immune dysregulation plays a critical role in PDPN progression[ 3 , 4 ]. Inflammatory cell infiltration, activation of the NF-κB and receptor for advanced glycation end-products (RAGE) signaling pathways, abnormal cytokine secretion, and bone marrow–derived proinflammatory responses exacerbate neuronal and glial injury in diabetic nerves. These immune-mediated mechanisms further aggravate oxidative stress and ischemia, accelerating peripheral nerve degeneration. Mitochondrial dysfunction is another hallmark of PDPN pathogenesis[ 5 ]. Impaired oxidative phosphorylation, reduced ATP synthesis, excessive reactive oxygen species (ROS) production, and calcium dysregulation collectively disrupt neuronal energy metabolism and homeostasis[ 6 – 9 ]. In dorsal root ganglion (DRG) neurons, which are essential for sensory transmission, hyperglycemia-induced oxidative stress causes mitochondrial membrane potential loss and bioenergetic failure, resulting in axonal degeneration, Schwann cell apoptosis, and impaired nerve conduction[ 10 ]. The interaction between mitochondrial injury and immune activation establishes a pathological cycle that promotes chronic inflammation, neuronal sensitization, and persistent neuropathic pain in PDPN. Although previous studies have elucidated the contributions of metabolic stress, mitochondrial dysfunction, and immune dysregulation to PDPN, the specific molecular mechanisms linking mitochondrial impairment in immune cells to the onset and progression of PDPN remain largely unexplored. Most existing research has focused on neuronal mitochondrial damage, while the role of mitochondria-related genes (MD-RGs) in immune-mediated peripheral nerve injury is poorly characterized. Consequently, the precise regulatory mechanisms underlying immune–metabolic crosstalk in PDPN remain unclear, limiting the development of targeted interventions. Currently available pharmacological management of PDPN primarily includes pregabalin and duloxetine, which provide partial symptom relief but are limited by modest efficacy and adverse effects. Neuromodulation therapies such as spinal cord stimulation (SCS) demonstrate superior analgesic efficacy but are invasive and costly[ 11 ], whereas topical agents like lidocaine or capsaicin patches can alleviate pain but often cause local irritation[ 12 ]. These limitations underscore the urgent need to elucidate the molecular mechanisms of PDPN and identify novel therapeutic targets that can modify disease progression rather than merely alleviate symptoms. In this study, we aim to systematically elucidate the molecular mechanisms linking mitochondrial dysfunction and immune dysregulation in PDPN through the integration of single-cell omics analysis and animal model validation. We sought to identify key mitochondrial-associated genes implicated in PDPN pathogenesis and to investigate their transcriptional regulation, molecular interactions, and disease relevance. Furthermore, through molecular docking and molecular dynamics simulations, we aimed to predict potential drug candidates targeting these genes and to design animal experiments to validate the omics findings. This study strives to uncover novel mechanisms within the mitochondrial-immune regulatory network of PDPN, thereby providing innovative therapeutic targets for clinical intervention. 2 Materials and methods 2.1 Data retrieval The GSE176017 and GSE34000 datasets were obtained from the Gene Expression Omnibus (GEO) database ( https://www.ncbi.nlm.nih.gov/geo/ )[ 13 ]. In the GSE176017 dataset, the single-cell RNA sequencing (scRNA-seq) data from dorsal root ganglion (DRG) of four rats with painful diabetic peripheral neuropathy (PDPN) (PDPN group) and two normal rats group (control group) was included in this study. Some of the data in the GSE34000 dataset was mainly utilized, including DRG of three normal rats (1-, 2- and 3-week) and three streptozotocin (STZ)-injected, vehicle-treated (1-, 2- and 3-week). Besides, this study also cited 1,136 mitochondrial dysfunction-related genes (MD-RGs) downloaded from the MitoCarta3.0 database ( https://www.broadinstitute.org/mitocarta ). 2.2 scRNA-seq data quality control and dimension reduction Quality control of the scRNA-seq data from the GSE176017 dataset was performed using the R package 'Seurat' (v5.0.1)[ 14 ]. Cells expressing fewer than 200 genes were excluded, while genes detected in at least three cells were retained. The cells with greater than 200 and less than 2,500 genes per cell (nFeature RNA = 2,500) and cells with less than 20,000 nCount_RNA (count number per cell) were retained. Besides, cells with less than 10% of mitochondrial genes were also preserved. Subsequently, the 'NormalizeData' function in the R package 'Seurat' was utilized for normalization data to ensure an optimal dimensionality reduction. Next, 2,000 highly variable genes were identified by virtue of the 'FindVariableFeatures' function. Principal component analysis (PCA) was conducted on the retained cells and 2,000 hypervariable genes using the 'RunPCA' function, following data normalization with the 'ScaleData' function. 2.3 Identification and cell annotation of cell clusters Ulteriorly, the 'FindNeighbors' (k.parm = 20) was employed to construct nearest-neighbors graphs and cluster analysis of the cells was realized by 'FindClusters' functions to obtain cell clusters. By the way, if there were batch effects in the samples, the data would be integrated through the 'integrateLayers' function to eliminate the error caused by the batch effects. The 'runUMAP' function was operated to dimensionality reduction and the uniform manifold approximation and projection (UMAP) was responsible for visualizing cell clusters. Furthermore, the cell clusters were annotated to obtain cell types through the R package 'singleR' (Ver. 2.4.0) in conjunction with marker genes of rats cited from a paper[ 15 ]. The proportion of cell types in every sample was visualized in the histogram. 2.4 Differential Expression Analysis To obtain the crucial genes related to PDPN and MD-RGs, the differential expression analysis was applied according to the expression of MD-RGs in samples of GSE176017 and GSE34000 datasets. Firstly, the differentially expressed genes1 (DEGs1) betwixt PDPN and control groups in the GSE176017 dataset were identified via the 'FindMarkers' function in R package 'Seurat' (p 0.25). Additionally, their potential roles function pathways were investigated by adopting Gene Ontology (GO) and Kyoto Encyclopedia of Genes (KEGG) through the R package 'clusterProfiler' (Ver. 4.8.1)[ 16 ] (adj.p 0.50, p < 0.05). The DEGs1 intersected with DEGs2 to acquire crucial genes associated with PDPN and MD-RGs. 2.5 Functional analysis of crucial genes To clarify the role and mechanism of crucial genes in PDPN, the interactions among crucial genes and their functions were revealed. Detailedly, in the GSE34000 dataset, the correlation and functional similarities of crucial genes were defined by Spearman's correlation analysis and the Friends analysis via the R package 'GOSemSim' (Ver. 2.28.0)[ 18 ], respectively (|cor|>0.4, p < 0.05). Moreover, in order to investigate the interactions of crucial genes with other genes that kept similar functions to crucial genes, a gene-gene interaction (GGI) network was synthesized via the GeneMANIA database ( https://genemania.org/ ). 2.6 Gene set enrichment analysis (GSEA) For the sake of making a thorough inquiry into functional pathways of crucial genes, GSEA was carried out. Firstly, Spearman's coefficient analysis was engaged in estimating the correlation coefficients between crucial genes and other genes in the GSE34000 dataset, and all genes ranked in accordance with the correlation coefficients were input to the R package 'ConsensusClusterPlus' for GSEA rely on the background gene set, c5.go.bp.v7.5.symbols.gmt, which was downloaded from the Molecular Signatures Database (MSigDB, https://www.gsea-msigdb.org/ ) (p < 0.05). 2.7 Disease Ontology (DO) analysis and chromosome localization To explore the connection between diseases and crucial genes, Disease Ontology (DO) was performed on crucial genes utilizing the R package 'DOSE' (Ver:3.28.1)[ 19 ] to search for diseases associated with crucial genes. Furthermore, as chromosome localization is helpful in understanding the functions of crucial genes, the locations of three crucial genes on the chromosomes were inquired consequently on the National Center for Biotechnology Information (NCBI) website, and the results were visualized by the R package 'RCircos' (Ver:1.2.2)[ 20 ]. 2.8 Construction of a miRNA-gene-TF network Besides, the upstream molecules, miRNAs, and transcription factors (TFs) targeting crucial genes were predicted in the miRDB ( https://mirdb.org/ ) database of miRWalk ( http://mirwalk.uni-hd.de/ ) platform (confidence level = 0.95) and GRTD database ( http://gtrd20-06 . bitumen.org/) (base pair = 2,000), respectively. A miRNA-gene-TF network demonstrating regulatory patterns was synthesized. 2.9 Enrichment, pseudo-temporal trajectory, and cell communication analyses of pivotal cells Based on the GSE176017 dataset, cell types in which crucial genes were expressed differentially were defined as pivotal cells. Furthermore, the R package 'ReactomeGSA' (Ver:1.2.2)[ 20 ] was adopted to find out the biological pathways involved by pivotal cell types with the aim of uncovering the functions of each pivotal cell. The pseudo-temporal trajectory analysis of discrepant pivotal cells was implemented applying the R package 'Monocle2' (Ver:2.30.0)[ 21 ]. Additionally, the R package 'CellPhone DB' (Ver:1.5.0)[ 20 ] was executed to build the critical intercellular interaction network with the intention of analyzing the number and intensity of ligand-receptor interactions between them. 2.10 Molecular docking The drugs with high interaction scores with crucial genes in the drug-gene interaction database (DGIdb, https://dgidb.org/ ) were brought into molecular docking to explore the ligand conformations adopted within the binding sites of crucial genes and drugs. Concretely, molecular docking proceeded by applying the AutoDock (Ver:1.1.2)[ 22 ] on the basis of the crystal structures corresponding to crucial genes and the three-dimensional structures of drugs, which were obtained from the Protein Data Bank (PDB, http://www.rcsb.org/ ) and PubChem databases ( https://pubchem.ncbi.nlm.nih.gov/ ), respectively. 2.11 Molecular dynamics simulations This study utilized GROMACS v2024.4 software to perform molecular dynamics simulations, aiming to investigate the binding stability and kinetic characteristics of candidate molecules with the target protein. The system was constructed using the AMBER14SB force field to describe the protein backbone and side chains, while the small molecule ligand was parameterized using the AMBER GAFF force field, and solvated with the TIP3P water model. Under periodic boundary conditions, a cubic solvent system was created with a 1 nm distance between the solute edge and the water box boundary, and 0.15 mol/L Na + /Cl − ions were added to maintain electrical neutrality. The simulation system first underwent energy minimization via the steepest descent method to eliminate steric hindrance conflicts. To obtain reasonable molecular orientations and reduce errors in subsequent dynamic simulations, a two-stage pre-equilibration was conducted. First, in the NVT ensemble (constant number of particles, volume, and temperature), a velocity-rescale temperature control algorithm (time step of 2 fs) was used for 100 ps of equilibration to stabilize the system temperature at 300 K. The system was then transferred to the NPT ensemble (constant number of particles, pressure, and temperature), and 100 ps of pressure equilibration (1 bar) was performed via the Parrinello-Rahman pressure control algorithm to complete solvent density optimization. The production-phase simulation was carried out under isothermal-isobaric conditions (300 K, 1 bar) for 100 ns of unconstrained dynamic sampling with a time step of 2 fs, monitoring the dynamic behavior of the complex throughout the process. To quantify the binding characteristics, the root-mean-square deviation (RMSD) of the protein-ligand complex backbone atoms was calculated to assess conformational stability; the root-mean-square fluctuation (RMSF) of the protein backbone atoms was analyzed to examine the flexibility changes of residues, and the total energy fluctuations of the system were monitored to evaluate thermodynamic stability. In addition, hydrogen bond counts and occupancy between the drug and the target were statistically analyzed to quantify the binding interaction strength, and the distances between the small molecule binding site and the protein amino acid residues were analyzed to assess binding stability, interaction mechanisms, and conformational changes. 2.12 Animals experiments All animal experiments in this study were conducted and reported in accordance with the ARRIVE guidelines 2.0. A completed ARRIVE checklist is provided as a supplementary file with this manuscript. Pain Management: Given that the painful diabetic neuropathy model itself induces chronic pain, no additional analgesic was administered post-modeling, as it would directly interfere with the primary behavioral outcome (mechanical allodynia). All surgical or invasive procedures (e.g., tissue collection) were performed under terminal anesthesia. Humane and Study Endpoints: Explicit humane endpoints were defined for this study. The primary study endpoint was set at 14 weeks after dietary intervention. Animals were monitored daily. Any animal exhibiting severe distress, significant weight loss (> 20%), profound lethargy, inability to access food or water, or the development of ulcers or infections would have been euthanized immediately. No animals met these criteria prior to the scheduled endpoint. Anesthesia and Euthanasia Procedure: At the study endpoint, rats were deeply anesthetized prior to euthanasia and tissue collection. The anesthetic agent was sodium pentobarbital. It was administered via intraperitoneal injection at a dose of 50 mg/kg, using a sterile syringe. Following the loss of pedal and corneal reflexes, ensuring a surgical plane of anesthesia, euthanasia was completed by transcardial perfusion with ice-cold saline. Death was confirmed by the cessation of respiration and heartbeat. Nine clean-grade healthy 11-12-week-old spontaneous type 2 diabetic GK rats[ 23 ] with body mass (359.60 ± 24.04) g were selected for the experimental group, and nine 11-12-week-old Wistar rats with body mass (313.47 ± 14.73) g were selected for the control group (provided by Qingdao Tianxing Biotechnology Co. Ltd.). The Experimental Animal Committee of Guizhou Medical University approved the animal study (No. 2403669, which was conducted in accordance with the guidelines of the Chinese Committee for the Protection and Use of Animals. Rats have a normal diet and free access to water during feeding. 5–6 rats in a cage under pathogen-free conditions with a photoperiod 12 h, temperature maintained at 20 to 24°C, and humidity maintained at 50% to 70%. Rats were given at least 2 days to acclimatize to these conditions before being used for experiments. After 2 days of acclimatization to conventional chow, GK rats were switched to a high-fat chow (containing 67.5% conventional chow, 10% lard, 20% sucrose, and 2.5% cholesterol), and the control group was given equal amounts of conventional chow. Tail vein random blood glucose, body weight, and mechanical foot reduction threshold were measured at the same time each week in both groups (the method is as follows). Criteria for successful modeling were: random blood glucose above 16.7 mmol/L and a statistically significant decrease in the mechanical foot reduction threshold from the basal value. A significant increase in fasting glucose (> 16.7 mmol/L) and a significant decrease in the mechanical foot reduction threshold from basal values were detected at week 14, indicating successful modeling of the PDPN rat model[ 24 , 25 ]. At the end of the las pain measurement, the rats were anesthetized by intraperitoneal injection of pentobarbital 50 mg/kg. After the table stopped moving, the rats were placed face-up on the operating table. The chest was opened with scissors to expose the heart. At 0°C, 200 ml of 0.9% saline was injected into the aortic arch, and the brain of the rats was perfused until the redness of the liver disappeared. Then, the isolated L3-L5 spinal column was truncated and exposed on a frozen operating table. After removing the spinal cord tissue by clipping along the intervertebral foramina, the spinal cord dorsal root ganglion (DRG) tissue was carefully and rapidly extracted into an ice box. Efficient completion of the process within 5 minutes and subsequent preservation in liquid nitrogen is critical. 2.13 Behavioral tests- MWT measurement Rats were tested using Von Frey filaments (vFFs) to measure their MWT. After acclimatising the rats on a 2 × 2 mm metal grid for 0.5 h, vFFs were applied to the skin between the third and fourth toes of the rats to test the expected foot contraction response of the animals (with a consistent intensity for each stimulus). Starting with a 10-g filament, vFFs were applied with stimulation strengths of 10, 15, 26, 60, 100, 160, and 300 g. If the experimental vFF did not elicit a foot elevation response from the rat, stronger stimuli were applied until the expected response was observed. Each rat was measured three times and stimulus intensity values were recorded[ 26 ]. 2.14 Real-time qPCR Total RNA was extracted from dorsal root ganglion tissue using TRIzol reagent (Vazyme, Nanjing, R401) according to the procedure in the operation manual: dorsal root ganglion tissue was lysed thoroughly in 1 ml of TRIzol, and 200 uL of chloroform was added to the above lysate, which was mixed vigorously to form a milky consistency, and then allowed to stand for 5 minutes at 4°C. The tissue was then centrifuged at a low temperature (12,000 g, 45° C) to remove the RNA from the tissue. Centrifuge at low temperature (12000g, 4°C) for 15 minutes to discard the white middle layer and the red lower layer (organic layer). Add 200 uL of isopropanol and mix well for 10 minutes. Centrifuge again at low temperature (12000g, 4°C) for 15 minutes and discard the supernatant. The precipitate was washed twice with 75% ethanol in DEPC water dried thoroughly, and finally dissolved in RNase-free ddH 2 O to obtain total RNA. Total RNA concentration was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, USA) and analyzed using the RNase-free ddH 2 O method. Total RNA concentration was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, USA) and diluted to a concentration of 100 ng/uL using RNase-free ddH 2 O. Reverse transcription was performed next. Add 3 µL of 5×gDNA digester Mix, 2 µL of Total RNA, and 10 µL of RNase-free H 2 O to a 150 µL sterile, enzyme-free EP tube, gently blow to mix, and incubate at 42°C for 2 min. Add 4×Hifair® III SuperMix plus directly to the reaction tube from the previous step, stir gently to mix, and incubate at 25°C for 5 min. After that, 4×Hifair® Ⅲ SuperMix plus was added immediately into the reaction tube in the previous step, and the products were gently blown with a pipette, and hatched at 25℃ for 5 min, 55℃ for 15 min, and 85℃ for 5 min. The primers, β-actin, Nudt5, Hmgcs2, and Cpt1c, were designed and synthesized by Bioengineering (Guangzhou, China), and real-time fluorescence quantitative RT-PCR was performed on CFX96 (Bio-Rad, U.S.A.). β-actin was used as an internal control to obtain the CT values of Nudt5, Hmgcs2, and Cpt1c. 2.15 Protein immunoblotting (Western blot) The total protein of each group of cells was extracted using a Solebo whole protein extraction kit and mixed with 5×loading buffer, heated at 95°C for pre-denaturation, and 50 µg of protein was taken from 10% polyacrylamide gel for electrophoresis after protein quantification by BCA method, and then transferred to polyvinylidene difluoride (PVDF) membranes, which were closed by rapid containment solution for 90 min at room temperature, and 1:1000 diluted rabbit anti-Nudt5, Hmgcs2, and Cpt1c were added to the membrane. rabbit anti-Nudt5, Hmgcs2, and Cpt1c antibodies, and incubated overnight at 4°C. The membrane was washed three times with TBST for 10 min each time, 1:1000 dilution of horseradish peroxidase-labeled goat anti-rabbit secondary antibody was added, and TBST membrane washed three times (10min each time), and the chromogenic substrate was added for the color exposure, and finally, the development and fixation were carried out, and the relative protein streak density was quantified by the Image J image detection software quantitatively for detection and analysis. 2.16 Statistical analysis All values are expressed as mean ± SEMs. A T-test for independent samples was used for contrast between the two groups. Data were tested for normality using the Shapiro-Wilk test. Statistical analysis was performed using GraphPad Prism version 9.4.1. The R software (Ver. 4.3.1) was responsible for bioinformatics analysis. The p-value of less than 0.05 was deemed to be statistically significant unless otherwise noted. 3 Results 3.1 Construction of a Single-Cell Transcriptomic Atlas of PDPN Following quality control, we obtained expression data for 14,712 genes across 5,501 cells from six samples, including control (n = 2) and PDPN (n = 4) groups (Supplementary Figs. 1A and B). After identifying highly variable genes, performing principal component analysis, and correcting for batch effects, we identified 20 cell clusters (Figs. 1 A and B, Supplementary 1C-E). Based on reported marker genes, we annotated these clusters into nine cell types: fibroblasts, macrophages, neurons, Schwann cells, spermatogonial germ cells (SGCs), vascular endothelial cells (VECs), microglia, vascular smooth muscle cells (VSMCs), and porcine spermatogonial germ cells (pSGCs) (Figs. 1 C and D). We calculated the relative proportion of each cell type within individuals (Fig. 1 E). Notably, neurons and macrophages constituted the major cellular components across all DRG samples. Fibroblasts were predominant in 83.33% (n = 5) of samples, while microglia were relatively scarce and absent in one sample. Thus, we constructed a single-cell transcriptomic atlas of DRG tissue in PDPN. 3.2 Identification and Functional Analysis of Differentially Expressed MD-RGs Hmgcs2, Nudt5, and Cpt1c We identified 518 MD-RGs that were significantly up- or down-regulated in the PDPN group compared to the control group (p 0.25). GO pathway enrichment analysis indicated that these genes are primarily involved in mitochondrial functions, including cellular respiration, ATP synthesis coupled electron transcript, electron transfer activity, and NADH dehydrogenase activity (Fig. 2 A). KEGG pathway enrichment analysis revealed that these genes are also significantly associated with thermogenesis and aging-related degenerative diseases such as Parkinson's disease, Huntington's disease, and diabetic cardiomyopathy (Fig. 2 B). We identified three overlapping differentially expressed MD-RGs (Hmgcs2, Nudt5, and Cpt1c) across the GSE176017 and GSE34000 datasets (Fig. 2 C). 3.3 Regulatory Network, Functional Correlation, and Disease Association Analysis of the Key MD-RGs Hmgcs2, Cpt1c, and Nudt5 Hmgcs2, Cpt1c, and Nudt5 are located on chromosomes 2, 1, and 17, respectively (Fig. 3 A). We identified three miRNAs targeting Nudt5 (rno-let-7d-5p, rno-let-7e-5p, and rno-let-7i-5p) and one miRNA (rno-miR-872-3p) targeting Hmgcs2 in the miRDB database. Additionally, we discovered 26 transcription factors upstream of Hmgcs2, Cpt1c, and Nudt5, among which Hnf4a, Arnt, and Brd4 simultaneously regulate the expression of these three genes (Fig. 3 B). GSEA results revealed that Hmgcs2, Nudt5, and Cpt1c were enriched in 443, 296, and 314 functional pathways, respectively. Hmgcs2 was primarily associated with the establishment of cell polarity and macroautophagy; Nudt5 was mainly linked to carbohydrate phosphorylation and sympathetic nervous system development; while Cpt1c was predominantly related to neurotransmitter metabolic process and neuron projection arborization (Figs. 3 C-E). Hmgcs2, Cpt1c, and Nudt5 exhibited similarities in gene expression levels and functional roles. Specifically, Nudt5 gene expression showed significant negative correlations with both Hmgcs2 (r=-0.6) and Cpt1c (r=-0.14), while Hmgcs2 and Cpt1c demonstrated a significant positive correlation (r = 0.6) (Fig. 3 F). Friends analysis revealed similar gene expression patterns among Hmgcs2, Cpt1c, and Nudt5, with Nudt5 exhibiting the strongest similarity (Fig. 3 G). In the GeneMANIA database, we identified 20 genes potentially sharing similar biological functions with Hmgcs2, Cpt1c, and Nudt5 through common proteins, primarily O-acyltransferase activity and amino-acid betaine metabolic processes (Fig. 3 H). DO enrichment analysis revealed that Hmgcs2, Cpt1c, and Nudt5 are collectively associated with metabolic and organ dysfunction-related disorders such as hereditary spastic paraplegia, extrahepatic cholestasis, and biliary tract disease (Fig. 3 I). 3.4 Cell Type–Specific Expression Patterns and Functional Characteristics of Hmgcs2, Cpt1c, and Nudt5 in PDPN We compared the expression levels of Hmgcs2, Cpt1c, and Nudt5 in nine cell types between the PDPN group and the control group. Compared to the control group, Hmgcs2 and Nudt5 were significantly downregulated in fibroblasts from the PDPN group, while Cpt1c was upregulated (Fig. 4 A). In macrophages, conversely, Nudt5 and Cpt1c were significantly upregulated in the PDPN group, while Hmgcs2 was downregulated (Fig. 4 B). Functional pathway enrichment analysis indicated that fibroblasts were primarily enriched in pathways such as COX reactions, agmatine biosynthesis, and adenylate cyclase activating pathway, while macrophages were mainly associated with functions related to FGFR1c and Klotho ligand binding and activation and dermatan sulfate biosynthesis (Fig. 4 C). We also performed Cell trajectory and pseudotiming analysis on fibroblasts and macrophages, dividing their differentiation processes into three stages. Compared to the control group, both fibroblasts and macrophages from the PDPN group exhibited higher differentiation levels in stage 2 (Fig. 4 D). Additionally, cell communication analysis revealed moderate interactions between fibroblasts and macrophages, though interactions within the fibroblast population were stronger (Figs. 4 E and F). Table 1 Binding energies of nine binding modes between Hmgcs2 and Chlorothiazide and hydrochlorothiazide. Gene Drug Affinity(kcal/mol) Hmgcs2 Chlorothiazide -5.525 -5.122 -5.073 -5.056 -4.92 -4.877 -4.853 -4.82 -4.75 Hydrochlorothiazide -4.89 -4.48 -4.45 -4.05 -3.98 -3.87 -3.70 -3.68 -3.29 3.5 Drug Screening, Molecular Docking, and Dynamic Validation of HMGCS2–Ligand Interactions in PDPN Given the strong correlations between Hmgcs2, Cpt1c, Nudt5, and PDPN, we aim to identify potential drugs targeting these three genes. In the DGIdb database, we identified relatively strong binding affinities between hydrochlorothiazide (interaction score = 1.13), chlorothiazide (interaction score = 7.45), and Hmgcs2. Through molecular docking analysis, we characterized nine binding modes between Hmgcs2 and these two drugs, and depicted the stable conformation with the lowest binding energy (Table 1 and Fig. 4 G). We performed 100-ns molecular dynamics simulations on the complexes formed between HMGCS2 and the small-molecule ligands chlorothiazide and hydrochlorothiazide, respectively. The results showed that, in the HMGCS2-hydrochlorothiazide system, the RMSD value remained within the range of 0.35–0.5 nm, with the protein structure reaching dynamic equilibrium and maintaining a stable conformation between 25 and 90 ns (Fig. 5 A). RMSF analysis at the amino acid residue level indicated fluctuation ranges of 0.05–0.35 nm for individual residues, reflecting local flexibility and overall binding stability with the drug ligand (Fig. 5 B). Energy monitoring revealed no abnormal fluctuations in the total system energy, further confirming the thermodynamic stability of the complex (Fig. 5 C). Hydrogen bond interaction analysis showed that the number of hydrogen bonds formed between the drug molecule and the active site of HMGCS2 remained relatively stable throughout the simulation, mostly maintaining 1–2 bonds with occasional increases to 3–4 bonds. This suggests that the binding is maintained through stable non-covalent interactions (Fig. 5 D). Additionally, the distance between the ligand and the binding site fluctuated within the range of 0.3–0.8 nm without exhibiting a sustained unidirectional trend, indicating that the binding state between the small molecule and the protein was relatively stable within the simulation duration (Fig. 5 E). In the HMGCS2-chlorothiazide system, the RMSD value remained within the range of 0.45–0.6 nm, and the protein structure reached dynamic equilibrium and remained stable between 30–85 ns (Fig. 6 A). The fluctuation range of residues in the RMSF was 0.05–0.35 nm (Fig. 6 B). Both the energy monitoring and hydrogen bond interaction analysis showed no abnormal total system energy, and the number of hydrogen bonds was mostly 1–2 and occasionally increased to 3 (Figs. 6 C-D). The differences in the average level and fluctuation range of the distance between the ligand and the binding site reflected the differences in binding stability under different simulation systems (Fig. 6 E). In conclusion, under physiological conditions, both HMGCS2-chlorothiazide and HMGCS2-hydrochlorothiazide can form dynamically stable complexes. Their binding patterns exhibit both structural adaptability and interaction persistence, suggesting that chlorothiazide and hydrochlorothiazide hold promise as therapeutic agents targeting HMGCS2 for the treatment of PDPN. 3.7 Experimental Validation of Hmgcs2, Nudt5, and Cpt1c Expression in the PDPN Rat Model To validate the differential expression of Hmgcs2, Nudt5, and Cpt1c observed in omics data within the PDPN group, we established a PDPN animal model using Goto-Kakizaki (GK) rats fed a high-fat diet and conducted RT-PCR and Western blot experiments. After 14 weeks, successful PDPN model establishment was confirmed based on body weight, random blood glucose levels, and MWT assessment (Fig. 7 A). At both the RNA and protein levels, Hmgcs2 was significantly downregulated in the PDPN group, while Nudt5 and Cp1tc were upregulated, accordant with the conclusions from the single-cell omics analysis (Figs. 7 B and C). 4 Discussion Painful diabetic peripheral neuropathy (PDPN) is a disabling complication of diabetes mellitus characterized by chronic neuropathic pain that severely compromises quality of life[ 27 ]. Peripheral nerves depend on efficient energy metabolism, and mitochondrial integrity is essential for maintaining axonal and myelin homeostasis. Mitochondrial dysfunction, resulting from disturbances in mitochondrial number, quality, or bioenergetic capacity, leads to neuronal injury[ 28 ]. Existing evidence suggests that the pathophysiology of PDPN involves a complex interplay of inflammation, oxidative stress, and mitochondrial impairment. These processes are associated with dorsal root ganglion (DRG) hyperexcitability, calcium overload, axonal degeneration, and loss of cutaneous innervation, which collectively drive neuropathic pain. However, the mechanisms linking mitochondrial dysfunction in immune cells to PDPN remain poorly understood. In this context, we used single-cell omics analysis combined with animal model validation to identify key genes associated with mitochondrial dysfunction in PDPN, and employed molecular docking together with molecular dynamics simulations to explore potential therapeutic compounds targeting these genes. Through this integrated approach, we identified three mitochondria-related genes, Hmgcs2, Nudt5, and Cpt1c, which showed distinct expression patterns and functional characteristics, indicating their important roles in mitochondrial dysregulation, immune activation, and neuronal injury during PDPN progression. Hmgcs2 (3-hydroxy-3-methylglutaryl-CoA synthase 2) encodes a mitochondrial enzyme that catalyzes the rate-limiting step of ketogenesis, maintaining systemic energy balance and oxidative metabolism[ 29 , 30 ]. Reduced expression of Hmgcs2 disrupts mitochondrial respiration, elevates oxidative stress, and compromises neuronal energy supply[ 31 ]. Gene enrichment analysis indicated that Hmgcs2 participates in pathways related to cell polarity establishment and macroautophagy, both of which are essential for neuronal repair and maintenance of axonal integrity[ 32 ]. Notably, macroautophagy maintains neuronal homeostasis by clearing damaged organelles, and its impairment is linked to diabetic neuropathy progression[ 33 ]. Impairment of these processes interferes with axonal transport, damages synaptic architecture, and accelerates neurodegeneration. Nudt5 (nucleoside diphosphate hydrolase 5) regulates intracellular nucleotide turnover and plays a central role in ATP-dependent signaling, cell proliferation, and inflammatory activation[ 34 , 35 ]. Functional enrichment linked Nudt5 to sympathetic nervous system development, implying its contribution to hyperalgesia and autonomic imbalance in PDPN. Sympathetic dysfunction, which was a known contributor to PDPN pathogenesis, may exacerbate pain hypersensitivity and metabolic dysregulation[ 36 ]. This aligns with prior studies demonstrating sympathetic overactivity in diabetic neuropathy models[ 37 ]. Single-cell omics revealed pronounced upregulation of Nudt5 in macrophages, suggesting enhanced metabolic reprogramming and proinflammatory activity. This change may intensify oxidative stress and promote the release of cytokines that aggravate neuronal hyperexcitability. Cpt1c (carnitine palmitoyltransferase 1C) encodes a key enzyme on the outer mitochondrial membrane that mediates long-chain fatty acid transport and oxidation, coordinating lipid utilization and neuronal energy balance[ 38 ]. Gene enrichment analysis showed that Cpt1c is associated with neurotransmitter metabolic processes, implicating it in the regulation of sensory transmission and pain modulation. Neurotransmitter imbalance, particularly in sensory and autonomic neurons, has been proposed to mediate PDPN-related pain and sensory abnormalities[ 39 ]. For example, altered catecholamine metabolism in diabetic neuropathy correlates with reduced pain thresholds[ 40 ]. Further regulatory analysis indicated that Hnf4a, Arnt, and Brd4 may serve as upstream transcriptional regulators. Hnf4a (hepatocyte nuclear factor 4 alpha)is a nuclear receptor transcription factor that can control the expression of several genes linked to metabolism and is primarily engaged in the development and metabolic regulation of organs including the liver and pancreas[ 41 ]. The basic helix-loop-helix/leucine zipper (bHLH/LZ) transcription factor, Arnt (aryl hydrocarbon receptor nuclear translocator), is involved in the regulation of cell proliferation[ 42 ]. Brd4 (bromodomain containing 4), a transcription factor belonging to the bromodomain and extra-terminal domain (BET) family, controls the expression of genes linked to cell proliferation, differentiation, and apoptosis. It is also involved in gene transcription, cell cycle regulation, and chromatin remodeling[ 43 ]. These transcription factors may influence the expression of mitochondrial genes and thereby participate in PDPN pathogenesis. Our single-cell analysis reveals that both fibroblasts and macrophages play crucial roles in PDPN. In the PDPN group, macrophages exhibit upregulation of Nudt5 and Cpt1c alongside downregulation of Hmgcs2, while fibroblasts show upregulation of Cpt1c and downregulation of Hmgcs2 and Nudt5. This contrasting gene expression pattern suggests potential interactions between fibroblasts and macrophages. Previous studies have shown that fibroblasts secrete CSF1 and CCL2 to recruit macrophages, while activated macrophages release PDGF, TGF-β, and IL-6 to promote fibroblast proliferation and matrix production, thereby creating a pro-fibrotic environment that exacerbates peripheral nerve injury[ 44 ]. Additionally, macrophages were found to have the capacity to release a range of growth factors and cytokines that support neural cell differentiation and proliferation and aid in the regeneration and repair of neural tissues[ 45 ]. Fibroblasts, beyond their structural role, act as inflammatory effector cells that secrete cytokines and chemokines such as IL-6, CCL2, and CXCL12, driving leukocyte recruitment and sustaining chronic inflammation in tissues[ 46 ]. These findings indicate that metabolic reprogramming within macrophages and fibroblasts plays a key role in the pathogenesis of PDPN, tightly linking mitochondrial dysfunction with immune activation and tissue remodeling. Molecular docking and molecular dynamics simulations demonstrated that both chlorothiazide and hydrochlorothiazide could form stable binding interactions with Hmgcs2. These compounds are thiazide diuretics commonly prescribed for the treatment of edema, hypertension, and heart failure[ 47 ]. Their main pharmacological action involves inhibiting the sodium–chloride cotransporter in the renal distal tubule, thereby reducing sodium reabsorption, increasing urine output, and lowering blood pressure[ 48 ]. In this study, both drugs exhibited stable conformations within the active site of Hmgcs2, suggesting potential regulatory effects on mitochondrial energy metabolism and oxidative stress in PDPN. Beyond their diuretic actions, chlorothiazide and hydrochlorothiazide have been reported to exert additional protective effects by improving vascular function, suppressing oxidative stress, and reducing inflammatory responses, which may contribute to neural protection and improved microcirculation. These findings indicate that thiazide compounds may modulate Hmgcs2-mediated metabolic pathways and hold promise as potential therapeutic agents for PDPN. Despite support from results of animal experiments, the relatively small sample size of single-cell omics data in this research necessitates further validation of conclusions through larger-scale investigations. Furthermore, given the cross-sectional nature of the study and the absence of longitudinal sampling during animal model development, the causal relationship between Hmgcs2, Nudt5, and Cpt1c with PDPN remains unconfirmed and warrants further exploration in prospective longitudinal studies. Although we identified chlorothiazide and hydrochlorothiazide as potential therapeutic agents targeting HMGCS2 for PDPN using molecular docking and dynamic simulation methods, the efficacy of these compounds and their specific therapeutic mechanisms require further investigation. 5 Conclusion Our integrated study identified Hmgcs2, Nudt5, and Cpt1c as key mitochondrial dysfunction-related genes in the pathogenesis of PDPN through single-cell omics analysis and animal experiments, revealing their unique roles in neural signaling, sympathetic regulation, and neurotransmitter metabolism. We also identified distinct expression patterns of these genes in macrophages and fibroblasts, establishing a novel molecular framework for understanding neuropathic pain progression. Additionally, we discovered the therapeutic potential of chlorothiazide and hydrochlorothiazide targeting HMGCS2 for treating PDPN. This study provides the first definitive evidence linking these genes to PDPN, opening new translational medicine opportunities for targeted therapies and biomarker development. Declarations Acknowledgements The authors acknowledge Qingdao Tianxing Biotechnology Co. for providing the laboratory animals. Authors contributions ZJZ: Investigation, Formal analysis, Validation. JQC: Investigation, Formal analysis, Data Curation, Visualization, Writing - Original Draft. YC: Writing - Review & Editing. XY: Supervision. XX: Conceptualization. WL: Conceptualization, Methodology, Supervision, Project administration, Writing – Review & Editing. All authors have read and approved the manuscript. The authors declare that all data were generated in-house and that no paper mill was used. Funding This study was funded by National Natural Science Foundation of Guizhou Medical University (NSFC) Cultivation Program (20NSP044)、 National Natural Science Foundation of China (NSFC) Regional Fund Cultivation Program for Affiliated Hospital of Guizhou Medical University (gyfynsfc[2023]-42)、 Science and Technology Fund of Guizhou Provincial Health Commission (gzwkj2023-395) and Guizhou Province Science and Technology Plan Project (grant no.: Qianke He Foundation-ZK[2024]General 190) Data availability The datasets generated and analysed within this study are available from the Gene Expression Omnibus (GEO) repository， [ https://www.ncbi.nlm.nih.gov/geo/ ] . Under the accession numbers GSE176017 and GSE34000. The datasets generated during and analyzed during the current study are available from the corresponding author on reasonable request. Ethics approval and consent to participate The animal study was carried out in accordance with the policies of the China Animal Protection and Use Committee and was approved by the Laboratory Animal Committee of Guizhou Medical University (No. 2403669). All animal experiments in this study were conducted and reported in accordance with the ARRIVE guidelines 2.0. A completed ARRIVE checklist is provided as a supplementary file with this manuscript. Consent for publication All authors certify that they have reviewed the manuscript and approved the manuscripts’ submission in its current form. Competing interests The authors declare no competing interests. References Sloan, G. et al. The Treatment of Painful Diabetic Neuropathy. Curr. Diabetes Rev. 18 (5), e070721194556 (2022). Wang, E. J. et al. Painful Diabetic Neuropathy - Spinal Cord Stimulation, Peripheral Nerve Stimulation, Transcutaneous Electrical Nerve Stimulation, and Scrambler Therapy: A Narrative Review. Pain Physician . 25 (8), E1163–E1173 (2022). Li, W. et al. Identification of Immune Infiltration and the Potential Biomarkers in Diabetic Peripheral Neuropathy through Bioinformatics and Machine Learning Methods . Biomolecules , 13 (1). (2022). Hagen, K. M. & Ousman, S. S. Aging and the immune response in diabetic peripheral neuropathy. J. Neuroimmunol. 355 , 577574 (2021). Zhu, J. et al. Diabetic peripheral neuropathy: pathogenetic mechanisms and treatment. Front. Endocrinol. (Lausanne) . 14 , 1265372 (2023). Blagov, A. et al. Mitochondrial Dysfunction as a Factor of Energy Metabolism Disorders in Type 2 Diabetes Mellitus. Front. Biosci. (Schol Ed) . 16 (1), 5 (2024). Forbes, J. M. & Thorburn, D. R. Mitochondrial dysfunction in diabetic kidney disease. Nat. Rev. Nephrol. 14 (5), 291–312 (2018). Lin, Y. et al. Dexmedetomidine alleviates oxidative stress and mitochondrial dysfunction in diabetic peripheral neuropathy via the microRNA-34a/SIRT2/S1PR1 axis. Int. Immunopharmacol. 117 , 109910 (2023). Xu, X. et al. Ru360 Alleviates Postoperative Cognitive Dysfunction in Aged Mice by Inhibiting MCU-Mediated Mitochondrial Dysfunction. Neuropsychiatr Dis. Treat. 19 , 1531–1542 (2023). Lim, T. K. et al. Mitochondrial and bioenergetic dysfunction in trauma-induced painful peripheral neuropathy. Mol. Pain . 11 , 58 (2015). D'Souza, R. S. et al. Evidence-Based Treatment of Painful Diabetic Neuropathy: a Systematic Review. Curr. Pain Headache Rep. 26 (8), 583–594 (2022). Laklouk, M. & Baranidharan, G. Profile of the capsaicin 8% patch for the management of neuropathic pain associated with postherpetic neuralgia: safety, efficacy, and patient acceptability. Patient Prefer Adherence . 10 , 1913–1918 (2016). Yu, X. J. et al. BMP7 ameliorates intervertebral disc degeneration in type 1 diabetic rats by inhibiting pyroptosis of nucleus pulposus cells and NLRP3 inflammasome activity. Mol. Med. 29 (1), 30 (2023). Yu, L. et al. Characterization of cancer-related fibroblasts (CAF) in hepatocellular carcinoma and construction of CAF-based risk signature based on single-cell RNA-seq and bulk RNA-seq data. Front. Immunol. 13 , 1009789 (2022). Zhou, H. et al. The Development of Mechanical Allodynia in Diabetic Rats Revealed by Single-Cell RNA-Seq. Front Mol Neurosci, 15 : p. 856299. (2022). Wu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innov. (Camb) . 2 (3), 100141 (2021). Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43 (7), e47 (2015). Yu, G. Gene Ontology Semantic Similarity Analysis Using GOSemSim. Methods Mol. Biol. 2117 , 207–215 (2020). Yu, G. et al. DOSE: an R/Bioconductor package for disease ontology semantic and enrichment analysis. Bioinformatics 31 (4), 608–609 (2015). Naquin, D. et al. CIRCUS: a package for Circos display of structural genome variations from paired-end and mate-pair sequencing data. BMC Bioinform. 15 , 198 (2014). Wu, R. et al. Deciphering the age-dependent changes of pulmonary fibroblasts in mice by single-cell transcriptomics. Front. Cell. Dev. Biol. 11 , 1287133 (2023). Dong, Y., Zhao, Q. & Wang, Y. Network pharmacology-based investigation of potential targets of astragalus membranaceous-angelica sinensis compound acting on diabetic nephropathy. Sci. Rep. 11 (1), 19496 (2021). Nagao, M. et al. Selectively bred rodent models for studying the etiology of type 2 diabetes: Goto-Kakizaki rats and Oikawa-Nagao mice. Endocr. J. 70 (1), 19–30 (2023). Du, E. R. et al. Regulatory mechanisms and therapeutic potential of microglial inhibitors in neuropathic pain and morphine tolerance. J. Zhejiang Univ. Sci. B . 21 (3), 204–217 (2020). Kankowski, S., Grothe, C. & Haastert-Talini, K. Neuropathic pain: Spotlighting anatomy, experimental models, mechanisms, and therapeutic aspects. Eur. J. Neurosci. 54 (2), 4475–4496 (2021). Deuis, J. R., Dvorakova, L. S. & Vetter, I. Methods Used to Evaluate Pain Behaviors in Rodents. Front. Mol. Neurosci. 10 , 284 (2017). Elafros, M. A. et al. Towards prevention of diabetic peripheral neuropathy: clinical presentation, pathogenesis, and new treatments. Lancet Neurol. 21 (10), 922–936 (2022). Yang, K. et al. Progress in the treatment of diabetic peripheral neuropathy. Biomed. Pharmacother . 148 , 112717 (2022). Asif, S. et al. Hmgcs2-mediated ketogenesis modulates high-fat diet-induced hepatosteatosis. Mol. Metab. 61 , 101494 (2022). Peng, C. et al. Role of mitochondrial metabolic disorder and immune infiltration in diabetic cardiomyopathy: new insights from bioinformatics analysis. J. Transl Med. 21 (1), 66 (2023). Xu, Y. et al. Sodium butyrate activates HMGCS2 to promote ketone body production through SIRT5-mediated desuccinylation. Front. Med. 17 (2), 339–351 (2023). Howell, A. S. & Lew, D. J. Morphogenesis and the cell cycle. Genetics 190 (1), 51–77 (2012). Griffey, C. J. & Yamamoto, A. Macroautophagy in CNS health and disease. Nat. Rev. Neurosci. 23 (7), 411–427 (2022). Qi, H. et al. The ADP-ribose hydrolase NUDT5 is important for DNA repair. Cell. Rep. 41 (12), 111866 (2022). Xu, F. et al. N7-methylguanosine regulatory genes well represented by METTL1 define vastly different prognostic, immune and therapy landscapes in adrenocortical carcinoma. Am. J. Cancer Res. 13 (2), 538–568 (2023). Kanazawa, H. & Fukuda, K. The plasticity of cardiac sympathetic nerves and its clinical implication in cardiovascular disease. Front. Synaptic Neurosci. 14 , 960606 (2022). Jang, H. N. & Oh, T. J. Pharmacological and Nonpharmacological Treatments for Painful Diabetic Peripheral Neuropathy. Diabetes Metab. J. 47 (6), 743–756 (2023). Virmani, A. et al. The Carnitine Palmitoyl Transferase (CPT) System and Possible Relevance for Neuropsychiatric and Neurological Conditions. Mol. Neurobiol. 52 (2), 826–836 (2015). Kim, M. K. et al. The effect of clozapine on the AMPK-ACC-CPT1 pathway in the rat frontal cortex. Int. J. Neuropsychopharmacol. 15 (7), 907–917 (2012). Galiero, R. et al. Peripheral Neuropathy in Diabetes Mellitus: Pathogenetic Mechanisms and Diagnostic Options . Int. J. Mol. Sci. , 24 (4). (2023). Suzuki, S. et al. ZNF185 prevents stress fiber formation through the inhibition of RhoA in endothelial cells. Commun. Biol. 6 (1), 29 (2023). Ullah, K. et al. Targeting Endothelial HIF2alpha/ARNT Expression for Ischemic Heart Disease Therapy . Biology (Basel) , 12 (7). (2023). Chen, X. et al. BET proteins inhibitor JQ1 impairs GM-CSF-promoted peritoneal macrophage self-renewal and IL-4-induced alternative polarization. Int Immunopharmacol, 124(Pt A): p. 110942. (2023). Buechler, M. B., Fu, W. & Turley, S. J. Fibroblast-macrophage reciprocal interactions in health, fibrosis, and cancer. Immunity 54 (5), 903–915 (2021). Fu, J. et al. GABA regulates IL-1beta production in macrophages. Cell. Rep. 41 (10), 111770 (2022). Wei, K., Nguyen, H. N. & Brenner, M. B. Fibroblast pathology in inflammatory diseases . J. Clin. Invest. , 131 (20). (2021). Ernst, M. E. & Fravel, M. A. Thiazide and the Thiazide-Like Diuretics: Review of Hydrochlorothiazide, Chlorthalidone, and Indapamide. Am. J. Hypertens. 35 (7), 573–586 (2022). Borhani, N. O. Chlorothiazide and hydrochlorothiazide: a comparative study of their hypotensive, saluretic and hyperuricemic action. Ann. Intern. Med. 53 , 342–358 (1960). Additional Declarations No competing interests reported. Supplementary Files S1.pdf Fig S1 Data quality control. (A) Violin plots of gene counts in samples before quality control; violin plots of sequencing reads in samples; and violin plots of mitochondrial proportions in samples. (B) Violin plots after quality control. (C) Elbow plot of UMAP downscaling. (D) UMAP do wnscaled plot to get 20 unique clusters. (E) After removing the batch effect, the UMAP findings. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 25 Feb, 2026 Reviews received at journal 22 Feb, 2026 Reviewers agreed at journal 12 Feb, 2026 Reviewers agreed at journal 12 Feb, 2026 Reviews received at journal 12 Feb, 2026 Reviewers agreed at journal 11 Feb, 2026 Reviewers agreed at journal 11 Feb, 2026 Reviewers agreed at journal 26 Jan, 2026 Reviewers agreed at journal 14 Jan, 2026 Reviewers invited by journal 09 Jan, 2026 Editor assigned by journal 09 Jan, 2026 Editor invited by journal 07 Jan, 2026 Submission checks completed at journal 05 Jan, 2026 First submitted to journal 11 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8200654","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":572578121,"identity":"df6275f0-22de-4fb4-9888-05f0b52ca699","order_by":0,"name":"Jiaqi Cao","email":"","orcid":"","institution":"Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jiaqi","middleName":"","lastName":"Cao","suffix":""},{"id":572578122,"identity":"0a805f44-fc17-407c-93c4-4e84262c83b1","order_by":1,"name":"Zhongjie Zhang","email":"","orcid":"","institution":"Affiliated Hospital of Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhongjie","middleName":"","lastName":"Zhang","suffix":""},{"id":572578123,"identity":"fb4c9d85-3718-4acb-90ae-57d6db8e1673","order_by":2,"name":"Yu Chen","email":"","orcid":"","institution":"Affiliated Hospital of Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Chen","suffix":""},{"id":572578124,"identity":"9ff5ea94-bb48-47f8-bbf1-ce86e573591e","order_by":3,"name":"Xiaoyi Yu","email":"","orcid":"","institution":"Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyi","middleName":"","lastName":"Yu","suffix":""},{"id":572578126,"identity":"e1304038-0325-4280-adff-3623266aaa52","order_by":4,"name":"Xuan Xiang","email":"","orcid":"","institution":"Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xuan","middleName":"","lastName":"Xiang","suffix":""},{"id":572578127,"identity":"85f7df76-1cec-4578-99f1-d09f932b31a6","order_by":5,"name":"Wei Lu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsklEQVRIiWNgGAWjYBACPhCRUAHhSBClhQ2s5QzJWhjbSNIikWP44OG82mj+BuaDt3kY7PKI0WJskLjteO6MA2zJ1jwMycVEaMndJpG47VjuBgYeM2kehgOJDURo2f4jcQ5IC/83orVsY0hsqAHZwkakFp73nyUSjh3InXGYzdhyjkEyYS387GmJH3/U1OX2tzc/vPGmwo6wFig4zMDADKINiFQPBHXEKx0Fo2AUjIKRBwANEzdn0D/49QAAAABJRU5ErkJggg==","orcid":"","institution":"Affiliated Hospital of Guizhou Medical University","correspondingAuthor":true,"prefix":"","firstName":"Wei","middleName":"","lastName":"Lu","suffix":""}],"badges":[],"createdAt":"2025-11-25 08:23:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8200654/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8200654/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100163424,"identity":"220e2dd3-1d2b-428e-977f-ebc9679cd5f2","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8307373,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/cb9c677f7a4127b92f93932f.docx"},{"id":100163414,"identity":"43cdef89-574e-4a53-81e8-4f5f6e38c661","added_by":"auto","created_at":"2026-01-13 15:12:22","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7875,"visible":true,"origin":"","legend":"","description":"","filename":"98740d833c24468e82d2716a08c6db93.json","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/0f7b01265def6b32eec2341e.json"},{"id":100369146,"identity":"b43abab7-3e1a-463a-85a1-f3fb2b3ad979","added_by":"auto","created_at":"2026-01-16 07:58:43","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":411820,"visible":true,"origin":"","legend":"","description":"","filename":"S1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/7c746fb356b991c2f0e2cac6.pdf"},{"id":100163421,"identity":"32b49793-bc87-40ef-9966-af843856b6a2","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121922,"visible":true,"origin":"","legend":"","description":"","filename":"98740d833c24468e82d2716a08c6db931enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/69dfa60d8a89f1bff75f2336.xml"},{"id":100368251,"identity":"bbd95f97-f3d4-43ce-8ec5-894e58333191","added_by":"auto","created_at":"2026-01-16 07:57:46","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101990,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/a63c63cb7b32f2d4f47dde2c.png"},{"id":100367494,"identity":"aebcac64-ed42-4ab1-a9d7-265469e32a72","added_by":"auto","created_at":"2026-01-16 07:57:06","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60111,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/f920602b7ec9cf5a4332bf64.png"},{"id":100368472,"identity":"b5625e4e-a3d1-41df-b692-d25043d17b79","added_by":"auto","created_at":"2026-01-16 07:57:59","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190792,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/a59db06de393794af1c6b55d.png"},{"id":100163422,"identity":"b2fe7a05-289c-46c8-82c4-89e1bf68e084","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":131854,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/bdea52d033036e6a0848ddc0.png"},{"id":100368778,"identity":"29345861-106b-4a3d-a43e-1f925e2e1213","added_by":"auto","created_at":"2026-01-16 07:58:20","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":248284,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/6b51d5ea01d1e37ca8fcdc17.png"},{"id":100163433,"identity":"145334d8-b41a-4efb-9eae-98c06b62893b","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":209993,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/3e0e9769cabaa27d302a2ca4.png"},{"id":100163427,"identity":"cdd137cd-17a2-47a5-92dc-37e8282d0670","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":155906,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/a26b626133b81ba9507dded9.png"},{"id":100163431,"identity":"1846e637-80e8-4fdc-be55-6f4d74a9f639","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119896,"visible":true,"origin":"","legend":"","description":"","filename":"98740d833c24468e82d2716a08c6db931structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/b0d3a339f97b3c509df55d80.xml"},{"id":100163434,"identity":"413ec946-9837-4c80-a2fd-fe0d2669c32a","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":133566,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/b4f58f6c3a47eb5c2ba20ad9.html"},{"id":100163413,"identity":"43544000-7a73-49e8-95fa-ff9f80176f23","added_by":"auto","created_at":"2026-01-13 15:12:22","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":444576,"visible":true,"origin":"","legend":"\u003cp\u003escRNA-seq identified 9 cell types in DGR tissues and 2,000 differentially expressed genes between control tissues and DRG in the PDPN group. \u003cstrong\u003e(A)\u003c/strong\u003e The first 2,000 highly differentiated and normalized variable genes. \u003cstrong\u003e(B)\u003c/strong\u003e UMAP plot of cell groups identified based on highly variable gene expression. \u003cstrong\u003e(C)\u003c/strong\u003e Dot plots showing the expression of selected gens (x-axis) by cluster (y-axis), the Dot size stands for the proportion of cells in the cluster that express the gene. \u003cstrong\u003e(D)\u003c/strong\u003e tSNE plot of cell groups identified based on highly variable gene expression. \u003cstrong\u003e(E)\u003c/strong\u003e Heatmap showing identification scores for each cell type by SingleR function.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/d7879d9b787cb8a70acc1277.jpeg"},{"id":100163416,"identity":"08b68f6c-c4a9-47c5-b5bc-e174dc24c46d","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":300711,"visible":true,"origin":"","legend":"\u003cp\u003ePredictive and categorical statistics. \u003cstrong\u003e(A)\u003c/strong\u003e Histogram of gene ontology (GO) categorical statistics (multiple gene sets). x-coordinate indicates the number of genes/transcripts and y-coordinate indicates the category. The enrichment of differentially expressed genes in terms of biological processes, cellular components and molecular functions is analyzed. \u003cstrong\u003e(B)\u003c/strong\u003e KEGG pathway of differentially expressed genes in DEGs1. \u003cstrong\u003e(C) \u003c/strong\u003eDEGs1 overlapped with DEGs2 to obtain the sum of three crucial genes (Hmgcs2, Nudt5 and Cpt1c).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/de24abe0c8fcb355814a3a6b.jpeg"},{"id":100163420,"identity":"5f3aee5d-08b9-44ee-9711-ab803871e795","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1102486,"visible":true,"origin":"","legend":"\u003cp\u003eDEG analysis by mRNA-Seq. \u003cstrong\u003e(A) \u003c/strong\u003eCircos plots depicting the physical locations of Nudt5, Hmgcs2 and Cpt1c. \u003cstrong\u003e(B)\u003c/strong\u003e miRNA-gene-TF network. (C-E) GSEA of differentially characterized genes.\u003cstrong\u003e (C)\u003c/strong\u003e GSEA of Hmgcs2. \u003cstrong\u003e(D)\u003c/strong\u003e GSEA of Nudt5. \u003cstrong\u003e(E)\u003c/strong\u003e GSEA of Cpt1c. \u003cstrong\u003e(F)\u003c/strong\u003e Corrplot-related heatmap analysis of Nudt5, Hmgcs2 and Cpt1c. \u003cstrong\u003e(G)\u003c/strong\u003e Friendes analysis of Go-related genes. \u003cstrong\u003e(H) \u003c/strong\u003eConstruction of PPI network using GeneMANIA database. \u003cstrong\u003e(I)\u003c/strong\u003eDO enrichment terms.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/e8e7f1796bb93e89474165ce.jpeg"},{"id":100163418,"identity":"1af85b3e-c9b3-452b-8b8f-b3a7ad5d536f","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":558813,"visible":true,"origin":"","legend":"\u003cp\u003ePivotal cell type and function analysis and drug-related molecular docking.\u003cstrong\u003e (A-B)\u003c/strong\u003e Comparison of expression differences of 3 crucial genes in 2 crucial cell types in the control and PDPN groups.\u003cstrong\u003e (C)\u003c/strong\u003e Functional enrichment analysis of macrophages and fibroblasts using the “ReactomeGSA” software package. \u003cstrong\u003e(D)\u003c/strong\u003eCell trajectory and pseudotiming analysis of macrophages and fibroblasts. \u003cstrong\u003e(E-F)\u003c/strong\u003eSchematic representation of the strength of ligand-receptor interactions between macrophages and fibroblasts.\u003cstrong\u003e (G) \u003c/strong\u003eResults of molecular docking between Hmgcs2-chlorothiazide and Hmgcs2-hydrochlorothiazide.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/40596970bccc3751d2922b9b.jpeg"},{"id":100368270,"identity":"8e5967d6-ae5c-4e3e-a801-8c403b3ddcb4","added_by":"auto","created_at":"2026-01-16 07:57:47","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":899547,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular dynamics validation of HMGCS2-hydrochlorothiazide. \u003cstrong\u003e(A)\u003c/strong\u003e RMSD plot of protein HMGCS2. \u003cstrong\u003e(B)\u003c/strong\u003e RMSF plot of protein HMGCS2. \u003cstrong\u003e(C) \u003c/strong\u003eEnergy fluctuation plot between the small molecule drug and the protein.\u003cstrong\u003e (D) \u003c/strong\u003eHydrogen bond count plot between the small molecule drug and the protein active site. \u003cstrong\u003e(E) \u003c/strong\u003eDistance plot between the small molecule drug and the binding site.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/e50dde58a3257ad8e316f940.jpeg"},{"id":100163429,"identity":"fa5c9adc-c007-464a-ad8f-093c47b67326","added_by":"auto","created_at":"2026-01-13 15:12:23","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":824955,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular dynamics validation of HMGCS2-chlorothiazide. \u003cstrong\u003e(A)\u003c/strong\u003e RMSD plot of protein HMGCS2. \u003cstrong\u003e(B)\u003c/strong\u003eRMSF plot of protein HMGCS2. \u003cstrong\u003e(C) \u003c/strong\u003eEnergy fluctuation plot between the small molecule drug and the protein.\u003cstrong\u003e (D) \u003c/strong\u003eHydrogen bond count plot between the small molecule drug and the protein active site. \u003cstrong\u003e(E)\u003c/strong\u003eDistance plot between the small molecule drug and the binding site.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/7a375b3a62ac79fcd6249684.jpeg"},{"id":100367473,"identity":"7dffb904-7e87-4f04-b562-19ae04d7098b","added_by":"auto","created_at":"2026-01-16 07:57:05","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":587551,"visible":true,"origin":"","legend":"\u003cp\u003eGK rats were successfully modelled in the PDPN model at 14 weeks. \u003cstrong\u003e(A)\u003c/strong\u003e Blood glucose was significantly elevated in GK rats at 14 weeks. MWT test was used to evaluate peripheral neuropathy in rats, and MWT levels decreased significantly starting at week 14. Wistar rats were used as control group. Data are represented as the mean ± SEM, n = 8, ***p \u0026lt; 0.001, ****p \u0026lt; 0.0001 Compared with MWT at week 1, iterative measurements ANOVA test with Dunnett's multiple comparison test in the MWT test Quantitative assessment of Nudt5,Hmgcs2, and Cpt1c gene expression. \u003cstrong\u003e(B)\u003c/strong\u003e Expression of the three genes was detected by qRT-PCR.\u003cstrong\u003e(C) \u003c/strong\u003eThe expression of the three genes was detected by WB. The results of the two methods were consistent: Nudt5, and Cpt1c gene expression was up-regulated and Hmgcs2 gene expression was down-regulated in the PDPN group compared with the control group. Data are represented as the mean ± SEM, n=3, *****p\u0026lt;0.0001, two-sided Student t test.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/064313e28d2ae63fbd7f4054.jpeg"},{"id":100406155,"identity":"e466c153-a77f-4a8d-bb83-2b3a8bb6a087","added_by":"auto","created_at":"2026-01-16 12:45:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5962808,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/2fb89daf-0715-4d9b-87da-c7f4ef9d3fcc.pdf"},{"id":100367774,"identity":"693d0184-fbce-4f6e-be5d-7303d848227d","added_by":"auto","created_at":"2026-01-16 07:57:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":411820,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig S1 Data quality control. \u003c/strong\u003e(A) Violin plots of gene counts in samples before quality control; violin plots of sequencing reads in samples; and violin plots of mitochondrial proportions in samples. (B) Violin plots after quality control. (C) Elbow plot of UMAP downscaling. (D) UMAP do wnscaled plot to get 20 unique clusters. (E) After removing the batch effect, the UMAP findings.\u003c/p\u003e","description":"","filename":"S1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8200654/v1/6b4f77bbb64c0bea1cbee3d5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Single-cell and experimental analyses identify mitochondrial dysfunction–related genes Hmgcs2, Nudt5, and Cpt1c in painful diabetic peripheral neuropathy","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003ePainful diabetic peripheral neuropathy (PDPN), characterized by chronic neuropathic pain associated with diabetes mellitus, represents one of the most common and debilitating complications of diabetes. Clinically, PDPN typically manifests as symmetrical distal limb pain or sensory abnormalities, but may also present as brachial or lumbosacral plexopathy, or mononeuropathy, all of which significantly reduce patients\u0026rsquo; quality of life and impose substantial economic and social burdens[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Epidemiological studies indicate that up to half of diabetic patients will develop peripheral neuropathy during the course of the disease, and the prevalence increases with disease duration, poor glycemic control, and aging. Despite its high prevalence and clinical significance, PDPN remains difficult to treat effectively.\u003c/p\u003e \u003cp\u003eThe pathogenesis of PDPN is multifactorial, involving complex metabolic, immune, and microvascular mechanisms. Chronic hyperglycemia, dyslipidemia, and insulin resistance contribute to metabolic stress, oxidative damage, and inflammatory activation within the peripheral nervous system. Accumulating evidence indicates that immune dysregulation plays a critical role in PDPN progression[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Inflammatory cell infiltration, activation of the NF-κB and receptor for advanced glycation end-products (RAGE) signaling pathways, abnormal cytokine secretion, and bone marrow\u0026ndash;derived proinflammatory responses exacerbate neuronal and glial injury in diabetic nerves. These immune-mediated mechanisms further aggravate oxidative stress and ischemia, accelerating peripheral nerve degeneration.\u003c/p\u003e \u003cp\u003eMitochondrial dysfunction is another hallmark of PDPN pathogenesis[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Impaired oxidative phosphorylation, reduced ATP synthesis, excessive reactive oxygen species (ROS) production, and calcium dysregulation collectively disrupt neuronal energy metabolism and homeostasis[\u003cspan additionalcitationids=\"CR7 CR8\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. In dorsal root ganglion (DRG) neurons, which are essential for sensory transmission, hyperglycemia-induced oxidative stress causes mitochondrial membrane potential loss and bioenergetic failure, resulting in axonal degeneration, Schwann cell apoptosis, and impaired nerve conduction[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. The interaction between mitochondrial injury and immune activation establishes a pathological cycle that promotes chronic inflammation, neuronal sensitization, and persistent neuropathic pain in PDPN.\u003c/p\u003e \u003cp\u003eAlthough previous studies have elucidated the contributions of metabolic stress, mitochondrial dysfunction, and immune dysregulation to PDPN, the specific molecular mechanisms linking mitochondrial impairment in immune cells to the onset and progression of PDPN remain largely unexplored. Most existing research has focused on neuronal mitochondrial damage, while the role of mitochondria-related genes (MD-RGs) in immune-mediated peripheral nerve injury is poorly characterized. Consequently, the precise regulatory mechanisms underlying immune\u0026ndash;metabolic crosstalk in PDPN remain unclear, limiting the development of targeted interventions.\u003c/p\u003e \u003cp\u003eCurrently available pharmacological management of PDPN primarily includes pregabalin and duloxetine, which provide partial symptom relief but are limited by modest efficacy and adverse effects. Neuromodulation therapies such as spinal cord stimulation (SCS) demonstrate superior analgesic efficacy but are invasive and costly[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], whereas topical agents like lidocaine or capsaicin patches can alleviate pain but often cause local irritation[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. These limitations underscore the urgent need to elucidate the molecular mechanisms of PDPN and identify novel therapeutic targets that can modify disease progression rather than merely alleviate symptoms.\u003c/p\u003e \u003cp\u003eIn this study, we aim to systematically elucidate the molecular mechanisms linking mitochondrial dysfunction and immune dysregulation in PDPN through the integration of single-cell omics analysis and animal model validation. We sought to identify key mitochondrial-associated genes implicated in PDPN pathogenesis and to investigate their transcriptional regulation, molecular interactions, and disease relevance. Furthermore, through molecular docking and molecular dynamics simulations, we aimed to predict potential drug candidates targeting these genes and to design animal experiments to validate the omics findings. This study strives to uncover novel mechanisms within the mitochondrial-immune regulatory network of PDPN, thereby providing innovative therapeutic targets for clinical intervention.\u003c/p\u003e"},{"header":"2 Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Data retrieval\u003c/h2\u003e \u003cp\u003eThe GSE176017 and GSE34000 datasets were obtained from the Gene Expression Omnibus (GEO) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e)[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. In the GSE176017 dataset, the single-cell RNA sequencing (scRNA-seq) data from dorsal root ganglion (DRG) of four rats with painful diabetic peripheral neuropathy (PDPN) (PDPN group) and two normal rats group (control group) was included in this study. Some of the data in the GSE34000 dataset was mainly utilized, including DRG of three normal rats (1-, 2- and 3-week) and three streptozotocin (STZ)-injected, vehicle-treated (1-, 2- and 3-week). Besides, this study also cited 1,136 mitochondrial dysfunction-related genes (MD-RGs) downloaded from the MitoCarta3.0 database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.broadinstitute.org/mitocarta\u003c/span\u003e\u003cspan address=\"https://www.broadinstitute.org/mitocarta\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 scRNA-seq data quality control and dimension reduction\u003c/h2\u003e \u003cp\u003eQuality control of the scRNA-seq data from the GSE176017 dataset was performed using the R package 'Seurat' (v5.0.1)[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Cells expressing fewer than 200 genes were excluded, while genes detected in at least three cells were retained. The cells with greater than 200 and less than 2,500 genes per cell (nFeature RNA\u0026thinsp;=\u0026thinsp;2,500) and cells with less than 20,000 nCount_RNA (count number per cell) were retained. Besides, cells with less than 10% of mitochondrial genes were also preserved. Subsequently, the 'NormalizeData' function in the R package 'Seurat' was utilized for normalization data to ensure an optimal dimensionality reduction. Next, 2,000 highly variable genes were identified by virtue of the 'FindVariableFeatures' function. Principal component analysis (PCA) was conducted on the retained cells and 2,000 hypervariable genes using the 'RunPCA' function, following data normalization with the 'ScaleData' function.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Identification and cell annotation of cell clusters\u003c/h2\u003e \u003cp\u003eUlteriorly, the 'FindNeighbors' (k.parm\u0026thinsp;=\u0026thinsp;20) was employed to construct nearest-neighbors graphs and cluster analysis of the cells was realized by 'FindClusters' functions to obtain cell clusters. By the way, if there were batch effects in the samples, the data would be integrated through the 'integrateLayers' function to eliminate the error caused by the batch effects. The 'runUMAP' function was operated to dimensionality reduction and the uniform manifold approximation and projection (UMAP) was responsible for visualizing cell clusters. Furthermore, the cell clusters were annotated to obtain cell types through the R package 'singleR' (Ver. 2.4.0) in conjunction with marker genes of rats cited from a paper[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. The proportion of cell types in every sample was visualized in the histogram.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Differential Expression Analysis\u003c/h2\u003e \u003cp\u003eTo obtain the crucial genes related to PDPN and MD-RGs, the differential expression analysis was applied according to the expression of MD-RGs in samples of GSE176017 and GSE34000 datasets. Firstly, the differentially expressed genes1 (DEGs1) betwixt PDPN and control groups in the GSE176017 dataset were identified via the 'FindMarkers' function in R package 'Seurat' (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, |avg_Log2fold-change (FC)|\u0026gt;0.25). Additionally, their potential roles function pathways were investigated by adopting Gene Ontology (GO) and Kyoto Encyclopedia of Genes (KEGG) through the R package 'clusterProfiler' (Ver. 4.8.1)[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] (adj.p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Ulteriorly, the DEGs2 across PDPN and control groups in the GSE34000 dataset were searched through the R package 'limma' (Ver. 3.56.2)[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] (|Log2FC|\u0026gt;0.50, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The DEGs1 intersected with DEGs2 to acquire crucial genes associated with PDPN and MD-RGs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Functional analysis of crucial genes\u003c/h2\u003e \u003cp\u003eTo clarify the role and mechanism of crucial genes in PDPN, the interactions among crucial genes and their functions were revealed. Detailedly, in the GSE34000 dataset, the correlation and functional similarities of crucial genes were defined by Spearman's correlation analysis and the Friends analysis via the R package 'GOSemSim' (Ver. 2.28.0)[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], respectively (|cor|\u0026gt;0.4, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Moreover, in order to investigate the interactions of crucial genes with other genes that kept similar functions to crucial genes, a gene-gene interaction (GGI) network was synthesized via the GeneMANIA database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://genemania.org/\u003c/span\u003e\u003cspan address=\"https://genemania.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Gene set enrichment analysis (GSEA)\u003c/h2\u003e \u003cp\u003eFor the sake of making a thorough inquiry into functional pathways of crucial genes, GSEA was carried out. Firstly, Spearman's coefficient analysis was engaged in estimating the correlation coefficients between crucial genes and other genes in the GSE34000 dataset, and all genes ranked in accordance with the correlation coefficients were input to the R package 'ConsensusClusterPlus' for GSEA rely on the background gene set, c5.go.bp.v7.5.symbols.gmt, which was downloaded from the Molecular Signatures Database (MSigDB, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Disease Ontology (DO) analysis and chromosome localization\u003c/h2\u003e \u003cp\u003eTo explore the connection between diseases and crucial genes, Disease Ontology (DO) was performed on crucial genes utilizing the R package 'DOSE' (Ver:3.28.1)[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] to search for diseases associated with crucial genes. Furthermore, as chromosome localization is helpful in understanding the functions of crucial genes, the locations of three crucial genes on the chromosomes were inquired consequently on the National Center for Biotechnology Information (NCBI) website, and the results were visualized by the R package 'RCircos' (Ver:1.2.2)[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Construction of a miRNA-gene-TF network\u003c/h2\u003e \u003cp\u003eBesides, the upstream molecules, miRNAs, and transcription factors (TFs) targeting crucial genes were predicted in the miRDB (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mirdb.org/\u003c/span\u003e\u003cspan address=\"https://mirdb.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) database of miRWalk (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://mirwalk.uni-hd.de/\u003c/span\u003e\u003cspan address=\"http://mirwalk.uni-hd.de/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) platform (confidence level\u0026thinsp;=\u0026thinsp;0.95) and GRTD database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://gtrd20-06\u003c/span\u003e\u003cspan address=\"http://gtrd20-06\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. bitumen.org/) (base pair\u0026thinsp;=\u0026thinsp;2,000), respectively. A miRNA-gene-TF network demonstrating regulatory patterns was synthesized.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Enrichment, pseudo-temporal trajectory, and cell communication analyses of pivotal cells\u003c/h2\u003e \u003cp\u003eBased on the GSE176017 dataset, cell types in which crucial genes were expressed differentially were defined as pivotal cells. Furthermore, the R package 'ReactomeGSA' (Ver:1.2.2)[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] was adopted to find out the biological pathways involved by pivotal cell types with the aim of uncovering the functions of each pivotal cell. The pseudo-temporal trajectory analysis of discrepant pivotal cells was implemented applying the R package 'Monocle2' (Ver:2.30.0)[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Additionally, the R package 'CellPhone DB' (Ver:1.5.0)[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] was executed to build the critical intercellular interaction network with the intention of analyzing the number and intensity of ligand-receptor interactions between them.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Molecular docking\u003c/h2\u003e \u003cp\u003eThe drugs with high interaction scores with crucial genes in the drug-gene interaction database (DGIdb, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dgidb.org/\u003c/span\u003e\u003cspan address=\"https://dgidb.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were brought into molecular docking to explore the ligand conformations adopted within the binding sites of crucial genes and drugs. Concretely, molecular docking proceeded by applying the AutoDock (Ver:1.1.2)[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] on the basis of the crystal structures corresponding to crucial genes and the three-dimensional structures of drugs, which were obtained from the Protein Data Bank (PDB, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.rcsb.org/\u003c/span\u003e\u003cspan address=\"http://www.rcsb.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and PubChem databases (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubchem.ncbi.nlm.nih.gov/\u003c/span\u003e\u003cspan address=\"https://pubchem.ncbi.nlm.nih.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), respectively.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.11 Molecular dynamics simulations\u003c/h2\u003e \u003cp\u003eThis study utilized GROMACS v2024.4 software to perform molecular dynamics simulations, aiming to investigate the binding stability and kinetic characteristics of candidate molecules with the target protein. The system was constructed using the AMBER14SB force field to describe the protein backbone and side chains, while the small molecule ligand was parameterized using the AMBER GAFF force field, and solvated with the TIP3P water model. Under periodic boundary conditions, a cubic solvent system was created with a 1 nm distance between the solute edge and the water box boundary, and 0.15 mol/L Na\u003csup\u003e+\u003c/sup\u003e/Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e ions were added to maintain electrical neutrality. The simulation system first underwent energy minimization via the steepest descent method to eliminate steric hindrance conflicts. To obtain reasonable molecular orientations and reduce errors in subsequent dynamic simulations, a two-stage pre-equilibration was conducted. First, in the NVT ensemble (constant number of particles, volume, and temperature), a velocity-rescale temperature control algorithm (time step of 2 fs) was used for 100 ps of equilibration to stabilize the system temperature at 300 K. The system was then transferred to the NPT ensemble (constant number of particles, pressure, and temperature), and 100 ps of pressure equilibration (1 bar) was performed via the Parrinello-Rahman pressure control algorithm to complete solvent density optimization. The production-phase simulation was carried out under isothermal-isobaric conditions (300 K, 1 bar) for 100 ns of unconstrained dynamic sampling with a time step of 2 fs, monitoring the dynamic behavior of the complex throughout the process. To quantify the binding characteristics, the root-mean-square deviation (RMSD) of the protein-ligand complex backbone atoms was calculated to assess conformational stability; the root-mean-square fluctuation (RMSF) of the protein backbone atoms was analyzed to examine the flexibility changes of residues, and the total energy fluctuations of the system were monitored to evaluate thermodynamic stability. In addition, hydrogen bond counts and occupancy between the drug and the target were statistically analyzed to quantify the binding interaction strength, and the distances between the small molecule binding site and the protein amino acid residues were analyzed to assess binding stability, interaction mechanisms, and conformational changes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e2.12 Animals experiments\u003c/h2\u003e \u003cp\u003e All animal experiments in this study were conducted and reported in accordance with the ARRIVE guidelines 2.0. A completed ARRIVE checklist is provided as a supplementary file with this manuscript.\u003c/p\u003e \u003cp\u003ePain Management: Given that the painful diabetic neuropathy model itself induces chronic pain, no additional analgesic was administered post-modeling, as it would directly interfere with the primary behavioral outcome (mechanical allodynia). All surgical or invasive procedures (e.g., tissue collection) were performed under terminal anesthesia.\u003c/p\u003e \u003cp\u003eHumane and Study Endpoints: Explicit humane endpoints were defined for this study. The primary study endpoint was set at 14 weeks after dietary intervention. Animals were monitored daily. Any animal exhibiting severe distress, significant weight loss (\u0026gt;\u0026thinsp;20%), profound lethargy, inability to access food or water, or the development of ulcers or infections would have been euthanized immediately. No animals met these criteria prior to the scheduled endpoint.\u003c/p\u003e \u003cp\u003eAnesthesia and Euthanasia Procedure: At the study endpoint, rats were deeply anesthetized prior to euthanasia and tissue collection. The anesthetic agent was sodium pentobarbital. It was administered via intraperitoneal injection at a dose of 50 mg/kg, using a sterile syringe. Following the loss of pedal and corneal reflexes, ensuring a surgical plane of anesthesia, euthanasia was completed by transcardial perfusion with ice-cold saline. Death was confirmed by the cessation of respiration and heartbeat.\u003c/p\u003e \u003cp\u003eNine clean-grade healthy 11-12-week-old spontaneous type 2 diabetic GK rats[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] with body mass (359.60\u0026thinsp;\u0026plusmn;\u0026thinsp;24.04) g were selected for the experimental group, and nine 11-12-week-old Wistar rats with body mass (313.47\u0026thinsp;\u0026plusmn;\u0026thinsp;14.73) g were selected for the control group (provided by Qingdao Tianxing Biotechnology Co. Ltd.). The Experimental Animal Committee of Guizhou Medical University approved the animal study (No. 2403669, which was conducted in accordance with the guidelines of the Chinese Committee for the Protection and Use of Animals. Rats have a normal diet and free access to water during feeding. 5\u0026ndash;6 rats in a cage under pathogen-free conditions with a photoperiod 12 h, temperature maintained at 20 to 24\u0026deg;C, and humidity maintained at 50% to 70%. Rats were given at least 2 days to acclimatize to these conditions before being used for experiments.\u003c/p\u003e \u003cp\u003eAfter 2 days of acclimatization to conventional chow, GK rats were switched to a high-fat chow (containing 67.5% conventional chow, 10% lard, 20% sucrose, and 2.5% cholesterol), and the control group was given equal amounts of conventional chow. Tail vein random blood glucose, body weight, and mechanical foot reduction threshold were measured at the same time each week in both groups (the method is as follows). Criteria for successful modeling were: random blood glucose above 16.7 mmol/L and a statistically significant decrease in the mechanical foot reduction threshold from the basal value. A significant increase in fasting glucose (\u0026gt;\u0026thinsp;16.7 mmol/L) and a significant decrease in the mechanical foot reduction threshold from basal values were detected at week 14, indicating successful modeling of the PDPN rat model[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAt the end of the las pain measurement, the rats were anesthetized by intraperitoneal injection of pentobarbital 50 mg/kg. After the table stopped moving, the rats were placed face-up on the operating table. The chest was opened with scissors to expose the heart. At 0\u0026deg;C, 200 ml of 0.9% saline was injected into the aortic arch, and the brain of the rats was perfused until the redness of the liver disappeared. Then, the isolated L3-L5 spinal column was truncated and exposed on a frozen operating table. After removing the spinal cord tissue by clipping along the intervertebral foramina, the spinal cord dorsal root ganglion (DRG) tissue was carefully and rapidly extracted into an ice box. Efficient completion of the process within 5 minutes and subsequent preservation in liquid nitrogen is critical.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.13 Behavioral tests- MWT measurement\u003c/h2\u003e \u003cp\u003eRats were tested using Von Frey filaments (vFFs) to measure their MWT. After acclimatising the rats on a 2 \u0026times; 2 mm metal grid for 0.5 h, vFFs were applied to the skin between the third and fourth toes of the rats to test the expected foot contraction response of the animals (with a consistent intensity for each stimulus). Starting with a 10-g filament, vFFs were applied with stimulation strengths of 10, 15, 26, 60, 100, 160, and 300 g. If the experimental vFF did not elicit a foot elevation response from the rat, stronger stimuli were applied until the expected response was observed. Each rat was measured three times and stimulus intensity values were recorded[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2.14 Real-time qPCR\u003c/h2\u003e \u003cp\u003eTotal RNA was extracted from dorsal root ganglion tissue using TRIzol reagent (Vazyme, Nanjing, R401) according to the procedure in the operation manual: dorsal root ganglion tissue was lysed thoroughly in 1 ml of TRIzol, and 200 uL of chloroform was added to the above lysate, which was mixed vigorously to form a milky consistency, and then allowed to stand for 5 minutes at 4\u0026deg;C. The tissue was then centrifuged at a low temperature (12,000 g, 45\u0026deg; C) to remove the RNA from the tissue. Centrifuge at low temperature (12000g, 4\u0026deg;C) for 15 minutes to discard the white middle layer and the red lower layer (organic layer). Add 200 uL of isopropanol and mix well for 10 minutes. Centrifuge again at low temperature (12000g, 4\u0026deg;C) for 15 minutes and discard the supernatant. The precipitate was washed twice with 75% ethanol in DEPC water dried thoroughly, and finally dissolved in RNase-free ddH\u003csub\u003e2\u003c/sub\u003eO to obtain total RNA. Total RNA concentration was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, USA) and analyzed using the RNase-free ddH\u003csub\u003e2\u003c/sub\u003eO method. Total RNA concentration was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, USA) and diluted to a concentration of 100 ng/uL using RNase-free ddH\u003csub\u003e2\u003c/sub\u003eO. Reverse transcription was performed next. Add 3 \u0026micro;L of 5\u0026times;gDNA digester Mix, 2 \u0026micro;L of Total RNA, and 10 \u0026micro;L of RNase-free H\u003csub\u003e2\u003c/sub\u003eO to a 150 \u0026micro;L sterile, enzyme-free EP tube, gently blow to mix, and incubate at 42\u0026deg;C for 2 min. Add 4\u0026times;Hifair\u0026reg; III SuperMix plus directly to the reaction tube from the previous step, stir gently to mix, and incubate at 25\u0026deg;C for 5 min. After that, 4\u0026times;Hifair\u0026reg; Ⅲ SuperMix plus was added immediately into the reaction tube in the previous step, and the products were gently blown with a pipette, and hatched at 25℃ for 5 min, 55℃ for 15 min, and 85℃ for 5 min. The primers, β-actin, Nudt5, Hmgcs2, and Cpt1c, were designed and synthesized by Bioengineering (Guangzhou, China), and real-time fluorescence quantitative RT-PCR was performed on CFX96 (Bio-Rad, U.S.A.). β-actin was used as an internal control to obtain the CT values of Nudt5, Hmgcs2, and Cpt1c.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e2.15 Protein immunoblotting (Western blot)\u003c/h2\u003e \u003cp\u003eThe total protein of each group of cells was extracted using a Solebo whole protein extraction kit and mixed with 5\u0026times;loading buffer, heated at 95\u0026deg;C for pre-denaturation, and 50 \u0026micro;g of protein was taken from 10% polyacrylamide gel for electrophoresis after protein quantification by BCA method, and then transferred to polyvinylidene difluoride (PVDF) membranes, which were closed by rapid containment solution for 90 min at room temperature, and 1:1000 diluted rabbit anti-Nudt5, Hmgcs2, and Cpt1c were added to the membrane. rabbit anti-Nudt5, Hmgcs2, and Cpt1c antibodies, and incubated overnight at 4\u0026deg;C. The membrane was washed three times with TBST for 10 min each time, 1:1000 dilution of horseradish peroxidase-labeled goat anti-rabbit secondary antibody was added, and TBST membrane washed three times (10min each time), and the chromogenic substrate was added for the color exposure, and finally, the development and fixation were carried out, and the relative protein streak density was quantified by the Image J image detection software quantitatively for detection and analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e2.16 Statistical analysis\u003c/h2\u003e \u003cp\u003eAll values are expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SEMs. A T-test for independent samples was used for contrast between the two groups. Data were tested for normality using the Shapiro-Wilk test. Statistical analysis was performed using GraphPad Prism version 9.4.1. The R software (Ver. 4.3.1) was responsible for bioinformatics analysis. The p-value of less than 0.05 was deemed to be statistically significant unless otherwise noted.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Construction of a Single-Cell Transcriptomic Atlas of PDPN\u003c/h2\u003e \u003cp\u003eFollowing quality control, we obtained expression data for 14,712 genes across 5,501 cells from six samples, including control (n\u0026thinsp;=\u0026thinsp;2) and PDPN (n\u0026thinsp;=\u0026thinsp;4) groups (Supplementary Figs.\u0026nbsp;1A and B). After identifying highly variable genes, performing principal component analysis, and correcting for batch effects, we identified 20 cell clusters (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and B, Supplementary 1C-E). Based on reported marker genes, we annotated these clusters into nine cell types: fibroblasts, macrophages, neurons, Schwann cells, spermatogonial germ cells (SGCs), vascular endothelial cells (VECs), microglia, vascular smooth muscle cells (VSMCs), and porcine spermatogonial germ cells (pSGCs) (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC and D). We calculated the relative proportion of each cell type within individuals (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). Notably, neurons and macrophages constituted the major cellular components across all DRG samples. Fibroblasts were predominant in 83.33% (n\u0026thinsp;=\u0026thinsp;5) of samples, while microglia were relatively scarce and absent in one sample. Thus, we constructed a single-cell transcriptomic atlas of DRG tissue in PDPN.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e3.2 Identification and Functional Analysis of Differentially Expressed MD-RGs Hmgcs2, Nudt5, and Cpt1c\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eWe identified 518 MD-RGs that were significantly up- or down-regulated in the PDPN group compared to the control group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, |avg_Log2fold-change (FC)|\u0026gt;0.25). GO pathway enrichment analysis indicated that these genes are primarily involved in mitochondrial functions, including cellular respiration, ATP synthesis coupled electron transcript, electron transfer activity, and NADH dehydrogenase activity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). KEGG pathway enrichment analysis revealed that these genes are also significantly associated with thermogenesis and aging-related degenerative diseases such as Parkinson's disease, Huntington's disease, and diabetic cardiomyopathy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). We identified three overlapping differentially expressed MD-RGs (Hmgcs2, Nudt5, and Cpt1c) across the GSE176017 and GSE34000 datasets (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003e3.3 Regulatory Network, Functional Correlation, and Disease Association Analysis of the Key MD-RGs Hmgcs2, Cpt1c, and Nudt5\u003c/b\u003e \u003c/p\u003e \u003cp\u003eHmgcs2, Cpt1c, and Nudt5 are located on chromosomes 2, 1, and 17, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). We identified three miRNAs targeting Nudt5 (rno-let-7d-5p, rno-let-7e-5p, and rno-let-7i-5p) and one miRNA (rno-miR-872-3p) targeting Hmgcs2 in the miRDB database. Additionally, we discovered 26 transcription factors upstream of Hmgcs2, Cpt1c, and Nudt5, among which Hnf4a, Arnt, and Brd4 simultaneously regulate the expression of these three genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). GSEA results revealed that Hmgcs2, Nudt5, and Cpt1c were enriched in 443, 296, and 314 functional pathways, respectively. Hmgcs2 was primarily associated with the establishment of cell polarity and macroautophagy; Nudt5 was mainly linked to carbohydrate phosphorylation and sympathetic nervous system development; while Cpt1c was predominantly related to neurotransmitter metabolic process and neuron projection arborization (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC-E).\u003c/p\u003e \u003cp\u003eHmgcs2, Cpt1c, and Nudt5 exhibited similarities in gene expression levels and functional roles. Specifically, Nudt5 gene expression showed significant negative correlations with both Hmgcs2 (r=-0.6) and Cpt1c (r=-0.14), while Hmgcs2 and Cpt1c demonstrated a significant positive correlation (r\u0026thinsp;=\u0026thinsp;0.6) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). Friends analysis revealed similar gene expression patterns among Hmgcs2, Cpt1c, and Nudt5, with Nudt5 exhibiting the strongest similarity (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). In the GeneMANIA database, we identified 20 genes potentially sharing similar biological functions with Hmgcs2, Cpt1c, and Nudt5 through common proteins, primarily O-acyltransferase activity and amino-acid betaine metabolic processes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH). DO enrichment analysis revealed that Hmgcs2, Cpt1c, and Nudt5 are collectively associated with metabolic and organ dysfunction-related disorders such as hereditary spastic paraplegia, extrahepatic cholestasis, and biliary tract disease (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Cell Type\u0026ndash;Specific Expression Patterns and Functional Characteristics of Hmgcs2, Cpt1c, and Nudt5 in PDPN\u003c/h2\u003e \u003cp\u003eWe compared the expression levels of Hmgcs2, Cpt1c, and Nudt5 in nine cell types between the PDPN group and the control group. Compared to the control group, Hmgcs2 and Nudt5 were significantly downregulated in fibroblasts from the PDPN group, while Cpt1c was upregulated (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). In macrophages, conversely, Nudt5 and Cpt1c were significantly upregulated in the PDPN group, while Hmgcs2 was downregulated (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Functional pathway enrichment analysis indicated that fibroblasts were primarily enriched in pathways such as COX reactions, agmatine biosynthesis, and adenylate cyclase activating pathway, while macrophages were mainly associated with functions related to FGFR1c and Klotho ligand binding and activation and dermatan sulfate biosynthesis (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). We also performed Cell trajectory and pseudotiming analysis on fibroblasts and macrophages, dividing their differentiation processes into three stages. Compared to the control group, both fibroblasts and macrophages from the PDPN group exhibited higher differentiation levels in stage 2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Additionally, cell communication analysis revealed moderate interactions between fibroblasts and macrophages, though interactions within the fibroblast population were stronger (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE and F).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBinding energies of nine binding modes between Hmgcs2 and Chlorothiazide and hydrochlorothiazide.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDrug\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAffinity(kcal/mol)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"17\" rowspan=\"18\"\u003e \u003cp\u003eHmgcs2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003eChlorothiazide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.525\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.122\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.073\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-5.056\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.877\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.853\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.82\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003eHydrochlorothiazide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.89\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.48\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-4.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.98\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.87\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.70\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-3.29\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Drug Screening, Molecular Docking, and Dynamic Validation of HMGCS2\u0026ndash;Ligand Interactions in PDPN\u003c/h2\u003e \u003cp\u003eGiven the strong correlations between Hmgcs2, Cpt1c, Nudt5, and PDPN, we aim to identify potential drugs targeting these three genes. In the DGIdb database, we identified relatively strong binding affinities between hydrochlorothiazide (interaction score\u0026thinsp;=\u0026thinsp;1.13), chlorothiazide (interaction score\u0026thinsp;=\u0026thinsp;7.45), and Hmgcs2. Through molecular docking analysis, we characterized nine binding modes between Hmgcs2 and these two drugs, and depicted the stable conformation with the lowest binding energy (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG).\u003c/p\u003e \u003cp\u003eWe performed 100-ns molecular dynamics simulations on the complexes formed between HMGCS2 and the small-molecule ligands chlorothiazide and hydrochlorothiazide, respectively. The results showed that, in the HMGCS2-hydrochlorothiazide system, the RMSD value remained within the range of 0.35\u0026ndash;0.5 nm, with the protein structure reaching dynamic equilibrium and maintaining a stable conformation between 25 and 90 ns (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). RMSF analysis at the amino acid residue level indicated fluctuation ranges of 0.05\u0026ndash;0.35 nm for individual residues, reflecting local flexibility and overall binding stability with the drug ligand (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Energy monitoring revealed no abnormal fluctuations in the total system energy, further confirming the thermodynamic stability of the complex (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Hydrogen bond interaction analysis showed that the number of hydrogen bonds formed between the drug molecule and the active site of HMGCS2 remained relatively stable throughout the simulation, mostly maintaining 1\u0026ndash;2 bonds with occasional increases to 3\u0026ndash;4 bonds. This suggests that the binding is maintained through stable non-covalent interactions (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Additionally, the distance between the ligand and the binding site fluctuated within the range of 0.3\u0026ndash;0.8 nm without exhibiting a sustained unidirectional trend, indicating that the binding state between the small molecule and the protein was relatively stable within the simulation duration (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003eIn the HMGCS2-chlorothiazide system, the RMSD value remained within the range of 0.45\u0026ndash;0.6 nm, and the protein structure reached dynamic equilibrium and remained stable between 30\u0026ndash;85 ns (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). The fluctuation range of residues in the RMSF was 0.05\u0026ndash;0.35 nm (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). Both the energy monitoring and hydrogen bond interaction analysis showed no abnormal total system energy, and the number of hydrogen bonds was mostly 1\u0026ndash;2 and occasionally increased to 3 (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC-D). The differences in the average level and fluctuation range of the distance between the ligand and the binding site reflected the differences in binding stability under different simulation systems (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). In conclusion, under physiological conditions, both HMGCS2-chlorothiazide and HMGCS2-hydrochlorothiazide can form dynamically stable complexes. Their binding patterns exhibit both structural adaptability and interaction persistence, suggesting that chlorothiazide and hydrochlorothiazide hold promise as therapeutic agents targeting HMGCS2 for the treatment of PDPN.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Experimental Validation of Hmgcs2, Nudt5, and Cpt1c Expression in the PDPN Rat Model\u003c/h2\u003e \u003cp\u003eTo validate the differential expression of Hmgcs2, Nudt5, and Cpt1c observed in omics data within the PDPN group, we established a PDPN animal model using Goto-Kakizaki (GK) rats fed a high-fat diet and conducted RT-PCR and Western blot experiments. After 14 weeks, successful PDPN model establishment was confirmed based on body weight, random blood glucose levels, and MWT assessment (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). At both the RNA and protein levels, Hmgcs2 was significantly downregulated in the PDPN group, while Nudt5 and Cp1tc were upregulated, accordant with the conclusions from the single-cell omics analysis (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB and C).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003ePainful diabetic peripheral neuropathy (PDPN) is a disabling complication of diabetes mellitus characterized by chronic neuropathic pain that severely compromises quality of life[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Peripheral nerves depend on efficient energy metabolism, and mitochondrial integrity is essential for maintaining axonal and myelin homeostasis. Mitochondrial dysfunction, resulting from disturbances in mitochondrial number, quality, or bioenergetic capacity, leads to neuronal injury[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Existing evidence suggests that the pathophysiology of PDPN involves a complex interplay of inflammation, oxidative stress, and mitochondrial impairment. These processes are associated with dorsal root ganglion (DRG) hyperexcitability, calcium overload, axonal degeneration, and loss of cutaneous innervation, which collectively drive neuropathic pain. However, the mechanisms linking mitochondrial dysfunction in immune cells to PDPN remain poorly understood. In this context, we used single-cell omics analysis combined with animal model validation to identify key genes associated with mitochondrial dysfunction in PDPN, and employed molecular docking together with molecular dynamics simulations to explore potential therapeutic compounds targeting these genes. Through this integrated approach, we identified three mitochondria-related genes, Hmgcs2, Nudt5, and Cpt1c, which showed distinct expression patterns and functional characteristics, indicating their important roles in mitochondrial dysregulation, immune activation, and neuronal injury during PDPN progression.\u003c/p\u003e \u003cp\u003eHmgcs2 (3-hydroxy-3-methylglutaryl-CoA synthase 2) encodes a mitochondrial enzyme that catalyzes the rate-limiting step of ketogenesis, maintaining systemic energy balance and oxidative metabolism[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Reduced expression of Hmgcs2 disrupts mitochondrial respiration, elevates oxidative stress, and compromises neuronal energy supply[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Gene enrichment analysis indicated that Hmgcs2 participates in pathways related to cell polarity establishment and macroautophagy, both of which are essential for neuronal repair and maintenance of axonal integrity[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Notably, macroautophagy maintains neuronal homeostasis by clearing damaged organelles, and its impairment is linked to diabetic neuropathy progression[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Impairment of these processes interferes with axonal transport, damages synaptic architecture, and accelerates neurodegeneration.\u003c/p\u003e \u003cp\u003eNudt5 (nucleoside diphosphate hydrolase 5) regulates intracellular nucleotide turnover and plays a central role in ATP-dependent signaling, cell proliferation, and inflammatory activation[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Functional enrichment linked Nudt5 to sympathetic nervous system development, implying its contribution to hyperalgesia and autonomic imbalance in PDPN. Sympathetic dysfunction, which was a known contributor to PDPN pathogenesis, may exacerbate pain hypersensitivity and metabolic dysregulation[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. This aligns with prior studies demonstrating sympathetic overactivity in diabetic neuropathy models[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Single-cell omics revealed pronounced upregulation of Nudt5 in macrophages, suggesting enhanced metabolic reprogramming and proinflammatory activity. This change may intensify oxidative stress and promote the release of cytokines that aggravate neuronal hyperexcitability.\u003c/p\u003e \u003cp\u003eCpt1c (carnitine palmitoyltransferase 1C) encodes a key enzyme on the outer mitochondrial membrane that mediates long-chain fatty acid transport and oxidation, coordinating lipid utilization and neuronal energy balance[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Gene enrichment analysis showed that Cpt1c is associated with neurotransmitter metabolic processes, implicating it in the regulation of sensory transmission and pain modulation. Neurotransmitter imbalance, particularly in sensory and autonomic neurons, has been proposed to mediate PDPN-related pain and sensory abnormalities[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. For example, altered catecholamine metabolism in diabetic neuropathy correlates with reduced pain thresholds[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFurther regulatory analysis indicated that Hnf4a, Arnt, and Brd4 may serve as upstream transcriptional regulators. Hnf4a (hepatocyte nuclear factor 4 alpha)is a nuclear receptor transcription factor that can control the expression of several genes linked to metabolism and is primarily engaged in the development and metabolic regulation of organs including the liver and pancreas[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The basic helix-loop-helix/leucine zipper (bHLH/LZ) transcription factor, Arnt (aryl hydrocarbon receptor nuclear translocator), is involved in the regulation of cell proliferation[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Brd4 (bromodomain containing 4), a transcription factor belonging to the bromodomain and extra-terminal domain (BET) family, controls the expression of genes linked to cell proliferation, differentiation, and apoptosis. It is also involved in gene transcription, cell cycle regulation, and chromatin remodeling[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. These transcription factors may influence the expression of mitochondrial genes and thereby participate in PDPN pathogenesis.\u003c/p\u003e \u003cp\u003eOur single-cell analysis reveals that both fibroblasts and macrophages play crucial roles in PDPN. In the PDPN group, macrophages exhibit upregulation of Nudt5 and Cpt1c alongside downregulation of Hmgcs2, while fibroblasts show upregulation of Cpt1c and downregulation of Hmgcs2 and Nudt5. This contrasting gene expression pattern suggests potential interactions between fibroblasts and macrophages. Previous studies have shown that fibroblasts secrete CSF1 and CCL2 to recruit macrophages, while activated macrophages release PDGF, TGF-β, and IL-6 to promote fibroblast proliferation and matrix production, thereby creating a pro-fibrotic environment that exacerbates peripheral nerve injury[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Additionally, macrophages were found to have the capacity to release a range of growth factors and cytokines that support neural cell differentiation and proliferation and aid in the regeneration and repair of neural tissues[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Fibroblasts, beyond their structural role, act as inflammatory effector cells that secrete cytokines and chemokines such as IL-6, CCL2, and CXCL12, driving leukocyte recruitment and sustaining chronic inflammation in tissues[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. These findings indicate that metabolic reprogramming within macrophages and fibroblasts plays a key role in the pathogenesis of PDPN, tightly linking mitochondrial dysfunction with immune activation and tissue remodeling.\u003c/p\u003e \u003cp\u003eMolecular docking and molecular dynamics simulations demonstrated that both chlorothiazide and hydrochlorothiazide could form stable binding interactions with Hmgcs2. These compounds are thiazide diuretics commonly prescribed for the treatment of edema, hypertension, and heart failure[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. Their main pharmacological action involves inhibiting the sodium\u0026ndash;chloride cotransporter in the renal distal tubule, thereby reducing sodium reabsorption, increasing urine output, and lowering blood pressure[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. In this study, both drugs exhibited stable conformations within the active site of Hmgcs2, suggesting potential regulatory effects on mitochondrial energy metabolism and oxidative stress in PDPN. Beyond their diuretic actions, chlorothiazide and hydrochlorothiazide have been reported to exert additional protective effects by improving vascular function, suppressing oxidative stress, and reducing inflammatory responses, which may contribute to neural protection and improved microcirculation. These findings indicate that thiazide compounds may modulate Hmgcs2-mediated metabolic pathways and hold promise as potential therapeutic agents for PDPN.\u003c/p\u003e \u003cp\u003eDespite support from results of animal experiments, the relatively small sample size of single-cell omics data in this research necessitates further validation of conclusions through larger-scale investigations. Furthermore, given the cross-sectional nature of the study and the absence of longitudinal sampling during animal model development, the causal relationship between Hmgcs2, Nudt5, and Cpt1c with PDPN remains unconfirmed and warrants further exploration in prospective longitudinal studies. Although we identified chlorothiazide and hydrochlorothiazide as potential therapeutic agents targeting HMGCS2 for PDPN using molecular docking and dynamic simulation methods, the efficacy of these compounds and their specific therapeutic mechanisms require further investigation.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eOur integrated study identified Hmgcs2, Nudt5, and Cpt1c as key mitochondrial dysfunction-related genes in the pathogenesis of PDPN through single-cell omics analysis and animal experiments, revealing their unique roles in neural signaling, sympathetic regulation, and neurotransmitter metabolism. We also identified distinct expression patterns of these genes in macrophages and fibroblasts, establishing a novel molecular framework for understanding neuropathic pain progression. Additionally, we discovered the therapeutic potential of chlorothiazide and hydrochlorothiazide targeting HMGCS2 for treating PDPN. This study provides the first definitive evidence linking these genes to PDPN, opening new translational medicine opportunities for targeted therapies and biomarker development.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e The authors acknowledge Qingdao Tianxing Biotechnology Co. for providing the laboratory animals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors contributions\u003c/strong\u003e ZJZ: Investigation, Formal analysis, Validation. JQC: Investigation, Formal analysis, Data Curation, Visualization, Writing - Original Draft. YC: Writing - Review \u0026amp; Editing. XY: Supervision. XX: Conceptualization. WL: Conceptualization, Methodology, Supervision, Project administration, Writing \u0026ndash; Review \u0026amp; Editing. All authors have read and approved the manuscript. The authors declare that all data were generated in-house and that no paper mill was used.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e This study was funded by National Natural Science Foundation of Guizhou Medical University (NSFC) Cultivation Program (20NSP044)、\u0026nbsp;National Natural Science Foundation of China (NSFC) Regional Fund Cultivation Program for Affiliated Hospital of Guizhou Medical University (gyfynsfc[2023]-42)、\u0026nbsp;Science and Technology Fund of Guizhou Provincial Health Commission (gzwkj2023-395) and Guizhou Province Science and Technology Plan Project (grant no.: Qianke He Foundation-ZK[2024]General 190)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e The datasets generated and analysed within this study are available from the Gene Expression Omnibus (GEO) repository，\u0026nbsp;[ https://www.ncbi.nlm.nih.gov/geo/ ] . Under the accession numbers GSE176017 and GSE34000. The datasets generated during and analyzed during the current study are available from the corresponding author on reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e The animal study was carried out in accordance with the policies of the China Animal Protection and Use Committee and was approved by the Laboratory Animal Committee of Guizhou Medical University (No. 2403669). All animal experiments in this study were conducted and reported in accordance with the ARRIVE guidelines 2.0. A completed ARRIVE checklist is provided as a supplementary file with this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e All authors certify that they have reviewed the manuscript and approved the manuscripts\u0026rsquo; submission in its current form.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSloan, G. et al. The Treatment of Painful Diabetic Neuropathy. \u003cem\u003eCurr. Diabetes Rev.\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e (5), e070721194556 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, E. J. et al. Painful Diabetic Neuropathy - Spinal Cord Stimulation, Peripheral Nerve Stimulation, Transcutaneous Electrical Nerve Stimulation, and Scrambler Therapy: A Narrative Review. \u003cem\u003ePain Physician\u003c/em\u003e. \u003cb\u003e25\u003c/b\u003e (8), E1163\u0026ndash;E1173 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, W. et al. \u003cem\u003eIdentification of Immune Infiltration and the Potential Biomarkers in Diabetic Peripheral Neuropathy through Bioinformatics and Machine Learning Methods\u003c/em\u003e. \u003cem\u003eBiomolecules\u003c/em\u003e, \u003cb\u003e13\u003c/b\u003e(1). (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHagen, K. M. \u0026amp; Ousman, S. S. Aging and the immune response in diabetic peripheral neuropathy. \u003cem\u003eJ. Neuroimmunol.\u003c/em\u003e \u003cb\u003e355\u003c/b\u003e, 577574 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu, J. et al. Diabetic peripheral neuropathy: pathogenetic mechanisms and treatment. \u003cem\u003eFront. Endocrinol. (Lausanne)\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, 1265372 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlagov, A. et al. Mitochondrial Dysfunction as a Factor of Energy Metabolism Disorders in Type 2 Diabetes Mellitus. \u003cem\u003eFront. Biosci. (Schol Ed)\u003c/em\u003e. \u003cb\u003e16\u003c/b\u003e (1), 5 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForbes, J. M. \u0026amp; Thorburn, D. R. Mitochondrial dysfunction in diabetic kidney disease. \u003cem\u003eNat. Rev. Nephrol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (5), 291\u0026ndash;312 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, Y. et al. Dexmedetomidine alleviates oxidative stress and mitochondrial dysfunction in diabetic peripheral neuropathy via the microRNA-34a/SIRT2/S1PR1 axis. \u003cem\u003eInt. Immunopharmacol.\u003c/em\u003e \u003cb\u003e117\u003c/b\u003e, 109910 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, X. et al. Ru360 Alleviates Postoperative Cognitive Dysfunction in Aged Mice by Inhibiting MCU-Mediated Mitochondrial Dysfunction. \u003cem\u003eNeuropsychiatr Dis. Treat.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 1531\u0026ndash;1542 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLim, T. K. et al. Mitochondrial and bioenergetic dysfunction in trauma-induced painful peripheral neuropathy. \u003cem\u003eMol. Pain\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 58 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD'Souza, R. S. et al. Evidence-Based Treatment of Painful Diabetic Neuropathy: a Systematic Review. \u003cem\u003eCurr. Pain Headache Rep.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e (8), 583\u0026ndash;594 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaklouk, M. \u0026amp; Baranidharan, G. Profile of the capsaicin 8% patch for the management of neuropathic pain associated with postherpetic neuralgia: safety, efficacy, and patient acceptability. \u003cem\u003ePatient Prefer Adherence\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, 1913\u0026ndash;1918 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, X. J. et al. BMP7 ameliorates intervertebral disc degeneration in type 1 diabetic rats by inhibiting pyroptosis of nucleus pulposus cells and NLRP3 inflammasome activity. \u003cem\u003eMol. Med.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e (1), 30 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, L. et al. Characterization of cancer-related fibroblasts (CAF) in hepatocellular carcinoma and construction of CAF-based risk signature based on single-cell RNA-seq and bulk RNA-seq data. \u003cem\u003eFront. Immunol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 1009789 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, H. et al. \u003cem\u003eThe Development of Mechanical Allodynia in Diabetic Rats Revealed by Single-Cell RNA-Seq.\u003c/em\u003e Front Mol Neurosci, \u003cb\u003e15\u003c/b\u003e: p. 856299. (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. \u003cem\u003eInnov. (Camb)\u003c/em\u003e. \u003cb\u003e2\u003c/b\u003e (3), 100141 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRitchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. \u003cem\u003eNucleic Acids Res.\u003c/em\u003e \u003cb\u003e43\u003c/b\u003e (7), e47 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, G. Gene Ontology Semantic Similarity Analysis Using GOSemSim. \u003cem\u003eMethods Mol. Biol.\u003c/em\u003e \u003cb\u003e2117\u003c/b\u003e, 207\u0026ndash;215 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, G. et al. DOSE: an R/Bioconductor package for disease ontology semantic and enrichment analysis. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e (4), 608\u0026ndash;609 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaquin, D. et al. CIRCUS: a package for Circos display of structural genome variations from paired-end and mate-pair sequencing data. \u003cem\u003eBMC Bioinform.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 198 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu, R. et al. Deciphering the age-dependent changes of pulmonary fibroblasts in mice by single-cell transcriptomics. \u003cem\u003eFront. Cell. Dev. Biol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 1287133 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong, Y., Zhao, Q. \u0026amp; Wang, Y. Network pharmacology-based investigation of potential targets of astragalus membranaceous-angelica sinensis compound acting on diabetic nephropathy. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (1), 19496 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagao, M. et al. Selectively bred rodent models for studying the etiology of type 2 diabetes: Goto-Kakizaki rats and Oikawa-Nagao mice. \u003cem\u003eEndocr. J.\u003c/em\u003e \u003cb\u003e70\u003c/b\u003e (1), 19\u0026ndash;30 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDu, E. R. et al. Regulatory mechanisms and therapeutic potential of microglial inhibitors in neuropathic pain and morphine tolerance. \u003cem\u003eJ. Zhejiang Univ. Sci. B\u003c/em\u003e. \u003cb\u003e21\u003c/b\u003e (3), 204\u0026ndash;217 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKankowski, S., Grothe, C. \u0026amp; Haastert-Talini, K. Neuropathic pain: Spotlighting anatomy, experimental models, mechanisms, and therapeutic aspects. \u003cem\u003eEur. J. Neurosci.\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e (2), 4475\u0026ndash;4496 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeuis, J. R., Dvorakova, L. S. \u0026amp; Vetter, I. Methods Used to Evaluate Pain Behaviors in Rodents. \u003cem\u003eFront. Mol. Neurosci.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 284 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElafros, M. A. et al. Towards prevention of diabetic peripheral neuropathy: clinical presentation, pathogenesis, and new treatments. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (10), 922\u0026ndash;936 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, K. et al. Progress in the treatment of diabetic peripheral neuropathy. \u003cem\u003eBiomed. Pharmacother\u003c/em\u003e. \u003cb\u003e148\u003c/b\u003e, 112717 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsif, S. et al. Hmgcs2-mediated ketogenesis modulates high-fat diet-induced hepatosteatosis. \u003cem\u003eMol. Metab.\u003c/em\u003e \u003cb\u003e61\u003c/b\u003e, 101494 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng, C. et al. Role of mitochondrial metabolic disorder and immune infiltration in diabetic cardiomyopathy: new insights from bioinformatics analysis. \u003cem\u003eJ. Transl Med.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (1), 66 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, Y. et al. Sodium butyrate activates HMGCS2 to promote ketone body production through SIRT5-mediated desuccinylation. \u003cem\u003eFront. Med.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e (2), 339\u0026ndash;351 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHowell, A. S. \u0026amp; Lew, D. J. Morphogenesis and the cell cycle. \u003cem\u003eGenetics\u003c/em\u003e \u003cb\u003e190\u003c/b\u003e (1), 51\u0026ndash;77 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGriffey, C. J. \u0026amp; Yamamoto, A. Macroautophagy in CNS health and disease. \u003cem\u003eNat. Rev. Neurosci.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (7), 411\u0026ndash;427 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi, H. et al. The ADP-ribose hydrolase NUDT5 is important for DNA repair. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e (12), 111866 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, F. et al. N7-methylguanosine regulatory genes well represented by METTL1 define vastly different prognostic, immune and therapy landscapes in adrenocortical carcinoma. \u003cem\u003eAm. J. Cancer Res.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (2), 538\u0026ndash;568 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanazawa, H. \u0026amp; Fukuda, K. The plasticity of cardiac sympathetic nerves and its clinical implication in cardiovascular disease. \u003cem\u003eFront. Synaptic Neurosci.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 960606 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJang, H. N. \u0026amp; Oh, T. J. Pharmacological and Nonpharmacological Treatments for Painful Diabetic Peripheral Neuropathy. \u003cem\u003eDiabetes Metab. J.\u003c/em\u003e \u003cb\u003e47\u003c/b\u003e (6), 743\u0026ndash;756 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVirmani, A. et al. The Carnitine Palmitoyl Transferase (CPT) System and Possible Relevance for Neuropsychiatric and Neurological Conditions. \u003cem\u003eMol. Neurobiol.\u003c/em\u003e \u003cb\u003e52\u003c/b\u003e (2), 826\u0026ndash;836 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, M. K. et al. The effect of clozapine on the AMPK-ACC-CPT1 pathway in the rat frontal cortex. \u003cem\u003eInt. J. Neuropsychopharmacol.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e (7), 907\u0026ndash;917 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaliero, R. et al. \u003cem\u003ePeripheral Neuropathy in Diabetes Mellitus: Pathogenetic Mechanisms and Diagnostic Options\u003c/em\u003e. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e, \u003cb\u003e24\u003c/b\u003e(4). (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuzuki, S. et al. ZNF185 prevents stress fiber formation through the inhibition of RhoA in endothelial cells. \u003cem\u003eCommun. Biol.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e (1), 29 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUllah, K. et al. \u003cem\u003eTargeting Endothelial HIF2alpha/ARNT Expression for Ischemic Heart Disease Therapy\u003c/em\u003e. \u003cem\u003eBiology (Basel)\u003c/em\u003e, \u003cb\u003e12\u003c/b\u003e(7). (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, X. et al. \u003cem\u003eBET proteins inhibitor JQ1 impairs GM-CSF-promoted peritoneal macrophage self-renewal and IL-4-induced alternative polarization.\u003c/em\u003e Int Immunopharmacol, 124(Pt A): p. 110942. (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuechler, M. B., Fu, W. \u0026amp; Turley, S. J. Fibroblast-macrophage reciprocal interactions in health, fibrosis, and cancer. \u003cem\u003eImmunity\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e (5), 903\u0026ndash;915 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu, J. et al. GABA regulates IL-1beta production in macrophages. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e (10), 111770 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei, K., Nguyen, H. N. \u0026amp; Brenner, M. B. \u003cem\u003eFibroblast pathology in inflammatory diseases\u003c/em\u003e. \u003cem\u003eJ. Clin. Invest.\u003c/em\u003e, \u003cb\u003e131\u003c/b\u003e(20). (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErnst, M. E. \u0026amp; Fravel, M. A. Thiazide and the Thiazide-Like Diuretics: Review of Hydrochlorothiazide, Chlorthalidone, and Indapamide. \u003cem\u003eAm. J. Hypertens.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e (7), 573\u0026ndash;586 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorhani, N. O. Chlorothiazide and hydrochlorothiazide: a comparative study of their hypotensive, saluretic and hyperuricemic action. \u003cem\u003eAnn. Intern. Med.\u003c/em\u003e \u003cb\u003e53\u003c/b\u003e, 342\u0026ndash;358 (1960).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Painful diabetic peripheral neuropathy, Mitochondrial dysfunction-related genes, Hmgcs2, Nudt5, Cpt1c, fibroblast","lastPublishedDoi":"10.21203/rs.3.rs-8200654/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8200654/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePainful diabetic peripheral neuropathy (PDPN) is a multifactorial disabling complication of diabetes, yet the pathogenesis involving mitochondrial dysfunction in immune cells remains unclear. This study aimed to identify mitochondrial dysfunction-related genes (MD-RGs) in PDPN and explore their mechanisms. Based on single-cell omics analysis, we identified three key MD-RGs (Hmgcs2, Nudt5, and Cpt1c) shared across different datasets and validated them in rat PDPN models. Gene set enrichment analysis revealed their association with cell polarity, sympathetic nerve development, and neurotransmitter pathways. We further observed distinct expression patterns of these genes in fibroblasts and macrophages within the dorsal root ganglion of PDPN. Hmgcs2 and Nudt5 were significantly downregulated in fibroblasts, while Cpt1c was upregulated. Conversely, Nudt5 and Cpt1c were significantly upregulated in macrophages, with Hmgcs2 downregulated. Functional enrichment analysis revealed that fibroblasts in PDPN primarily associated with polyamine/sulfate biosynthesis, while macrophages predominantly enriched for glycerol/choline metabolism, indicating distinct metabolic functions between the two cell types. Through molecular docking and dynamic simulation analysis, we further identified chlorothiazide and hydrochlorothiazide as stable binders to HMGCS2, suggesting potential as targeted therapeutics for PDPN. These findings indicate that Hmgcs2, Nudt5, and Cpt1c are key MD-RGs in the PDPN pathogenesis and provide novel therapeutic development strategies.\u003c/p\u003e","manuscriptTitle":"Single-cell and experimental analyses identify mitochondrial dysfunction–related genes Hmgcs2, Nudt5, and Cpt1c in painful diabetic peripheral neuropathy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-13 15:12:11","doi":"10.21203/rs.3.rs-8200654/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-25T10:03:04+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-22T08:21:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"22837331785136065179803505551928285899","date":"2026-02-13T00:40:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"296758768727633703488013904001503563381","date":"2026-02-12T08:30:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-12T05:35:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"90713581447284920764250808735164758994","date":"2026-02-12T03:59:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"314561739960467830361986346490990283764","date":"2026-02-11T14:29:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"318633037527100520645266886099027002504","date":"2026-01-26T20:02:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"147107253573071622112023694004226347782","date":"2026-01-14T20:36:21+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-09T15:08:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-09T15:00:01+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-07T05:30:15+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-05T11:53:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-12-11T05:17:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9826f2d7-7e0a-4b4c-ab2a-9eaa1e66d2d4","owner":[],"postedDate":"January 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":60928745,"name":"Biological sciences/Cell biology"},{"id":60928746,"name":"Health sciences/Diseases"},{"id":60928747,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-05-15T12:24:42+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-13 15:12:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8200654","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8200654","identity":"rs-8200654","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}