N7-methylguanosine-related gene decapping scavenger enzymes as a novel biomarker regulating epithelial cell function in diabetic foot ulcers

Abstract

BACKGROUND

Chronic nonhealing wounds, such as diabetic foot ulcer (DFU), suffer from delayed healing. Identifying effective biomarkers or targets is crucial for managing these refractory wounds. While N7-methylguanosine (m7G) methylation is important in RNA modification, its connection to chronic nonhealing wounds is poorly understood.

AIM

To assess the potential m7G biomarkers in DFU and their underlying molecular mechanisms.

METHODS

Differential expression analysis and weighted gene coexpression network analysis identified key genes in DFU. Hub genes were determined through m7G-DFU intersection, and gene set enrichment analysis was conducted. Diagnostic potential of hub genes was assessed using receiver operating characteristic curves. The hub gene’s expression (decapping scavenger enzyme, DCPS) was confirmed using quantitative reverse transcription polymerase chain reaction and immunofluorescence. In vitro, normal human epidermal keratinocyte models were knocked down for DCPS, and the function was assessed through flow cytometry, western blotting, immunofluorescence, Transwell assays, and scratch assays.

RESULTS

Weighted gene coexpression network analysis and differential expression analysis revealed links between DFU datasets and methylation processes, identifying hub gene DCPS as a candidate biomarker. Notably, its diagnostic value was confirmed with a test set and receiver operating characteristic curve, achieving an area under the curve of 0.98 and 0.99. Quantitative reverse transcription polymerase chain reaction and immunofluorescence analyses showed significantly reduced expression of DCPS in the wound skin of DFU patients and streptozotocin-induced diabetic mice, indicating its role as a regulatory factor of m7G in diabetic wounds. Mechanistically, in vitro studies showed that DCPS knockdown significantly reduced cyclin-dependent kinase 6 and cyclin D1 expression, disrupted the epithelial cell cycle, inhibited cell proliferation and migration, and increased apoptosis rates.

CONCLUSION

DCPS was identified as a promising DFU biomarker and therapeutic target, regulating m7G to affect cell cycle, proliferation, and epithelial cell migration during DFU wound healing.

Key Words: Diabetic foot ulcer; N7-methylguanosine; Decapping scavenger enzyme; Biomarker; Normal human epidermal keratinocytes; Cell cycle

Core Tip: This study identified decapping scavenger enzyme (DCPS) as a potential biomarker and therapeutic target for diabetic foot ulcers. It demonstrated that DCPS regulated m7G methylation, which affected cell cycle progression, proliferation, and migration of epithelial cells, which are crucial for wound healing. DCPS expression was significantly reduced in diabetic foot ulcer patients and diabetic mouse models, and its knockdown disrupted cell function, making it a promising target for improving healing in chronic nonhealing wounds.

INTRODUCTION

Diabetes mellitus has emerged as a major global public health challenge, significantly increasing the risk of life-threatening complications. Among these, diabetic foot ulcer (DFU) represents one of the most severe chronic complications, with a prevalence affecting up to 34% of adult patients[1]. The prognosis for DFU patients is particularly concerning, demonstrating a 5-year mortality rate of approximately 30%, which escalates to > 70% in cases requiring major amputation[2]. A hallmark pathological feature of DFU is the profound impairment of wound healing processes[3]. Given these clinical consequences and the substantial socioeconomic burden imposed by DFU management, the development of effective wound healing strategies constitutes a critical research priority. Furthermore, the discovery of novel diagnostic biomarkers holds significant promise for advancing early detection and targeted treatment of this debilitating condition.

N7-methylguanosine (m7G) represents the predominant methylation modification observed in messenger RNA (mRNA), being incorporated into the 5’ cap at an early stage of transcription[4]. The m7G cap plays a pivotal role in regulating the entire mRNA life cycle, encompassing metabolism, splicing, translation, transcription, and processing, thereby influencing critical cellular processes such as gene expression and transcriptional stability[5]. Recent studies have demonstrated that expression of m7G methylation is associated with various cellular processes, including proliferation, differentiation, migration and invasion. Furthermore, inhibition of cellular methylation has been shown to impede cell proliferation and growth[6-10]. Methylation plays a crucial role in cellular function and is implicated in various metabolic disorders, including type 2 diabetes mellitus, obesity, and nonalcoholic fatty liver disease[11,12]. Several studies have proposed that methylation may contribute to diabetic complications by influencing cell cycle regulation, proliferation, and metabolic processes[13,14]. Nonetheless, the specific role of m7G methylation in the context of DFU remains inadequately understood.

The proliferation and growth of epithelial cells are critical processes in the healing of DFU and significantly contribute to overall wound healing[15]. In diabetic patients, a hyperglycemic environment can result in compromised skin integrity and inhibited cellular proliferation[16]. Prior research has established that prolonged hyperglycemia is associated with increased production of reactive oxygen species and expanded formation of advanced glycation end products, which adversely affect keratinocyte function and consequently impede the healing process of diabetic wounds[17,18]. However, the role of m7G methylation in the healing process of diabetic skin wounds remains to be elucidated. Elucidating the impact of m7G methylation on the proliferation and growth of keratinocytes within the hyperglycemic context of diabetes is important for identifying novel therapeutic targets that enhance the healing of diabetic wounds. The high-throughput screening of the Gene Expression Omnibus (GEO) database will offer a novel perspective for elucidating the role of methylation in DFU.

The decapping scavenger enzyme (DCPS), a key m7G cap-binding protein, is indispensable for the terminal step of mRNA decay. It catalyzes the hydrolysis of the 5’ mRNA cap structure into 7-methylguanosine monophosphate and a nucleoside diphosphate, thereby eliminating potential translational inhibitors, recycling nucleotide components, and maintaining mRNA degradation flux[19]. Impaired DCPS activity may cause broad dysfunction in cap-dependent processes including splicing, translation initiation, and mRNA turnover[20]. Pathologically, DCPS dysregulation is implicated in acute myeloid leukemia and spinal muscular atrophy, with emerging roles in oncogenesis. Notably, DCPS inhibition suppresses proliferation and migration in uveal melanoma[7], while genetic knockdown or pharmacological inhibition attenuates glioblastoma cell proliferation, clonogenicity, and induces apoptosis[21]. Similarly, DCPS-deficient acute myeloid leukemia cells exhibit proliferative defects, cell cycle arrest, and apoptosis[22]. These collective findings establish DCPS as a pivotal regulator of cell cycle progression, proliferation, and apoptosis, suggesting its therapeutic potential in pathological conditions such as impaired diabetic wound healing.

In this study, we used weighted gene co-expression network analysis (WGCNA) in conjunction with the limma package to identify diagnostic biomarkers for DFU by analyzing datasets sourced from the GEO database[23]. We explored the relationship between biomarkers and m7G methylation, and identified that the DCPS is a key gene associated with m7G methylation, playing a pivotal role in the progression of DFU. Surprisingly, we found using receiver operating characteristic curve analysis that DCPS had high diagnostic efficacy. Additionally, we utilized CIBERSORT to examine immune cell infiltration in DFU compared to control samples, and we explored the correlation between DCPS, as a central gene, and various immune cell populations[24]. Quantitative immunofluorescence revealed significant downregulation of DCPS in the upper dermis. To further elucidate the role of DCPS, we developed in vitro diabetes models. By constructing physiological and pathological models of normal epidermal keratinocytes (NHEKs), we evaluated the effects of DCPS on proliferation, apoptosis, and migration. We found that DCPS knockdown inhibited expression of cyclin-dependent kinase 6 (CDK6) and cyclin D1, and impaired the growth and proliferation of NHEKs. This study elucidated the potential association of methylation with diabetic skin wound healing and highlighted the role of DCPS as a regulator and predictive biomarker of m7G methylation in DFU.

MATERIALS AND METHODS

Human samples

This study analyzed the skin of diabetic and non-diabetic foot surgery patients. Tissues were snap frozen at -80 °C for quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis or embedded in paraffin.

Datasets

We sourced three raw datasets from the GEO database (https://www.ncbi.nlm.nih.gov/geo). From the GSE134431 RNA-seq dataset, which included 13 DFU and eight normal skin samples, we identified a hub gene for DFU. We assessed the model’s predictive power using independent validation datasets GSE7014, which feature transcriptional profiles from both DFU and normal skin samples (Table 1).

Table 1 Detailed information of datasets.

Dataset	Platform	Type	Count	DFU	Control
GSE134431	GPL18573	mRNA	21	13	8
GSE7014	GPL570	mRNA	36	30	6

DFU: Diabetic foot ulcer.

WGCNA

The WGCNA R package was used to construct a coexpression network using the Sangerbox platform (https://xenabrowser.net), facilitating the association of gene networks with clinical features[25]. A scale-free coexpression network was created, followed by transformation of the matrix into an adjacency matrix and a topological overlap matrix (TOM). The dissimilarity of the TOM was then calculated. Subsequently, the eigengene was determined, and a threshold of 0.25 was established to facilitate the merging of similar modules within the clustering dendrogram. Clinical information was incorporated into the modules to determine the most pertinent modules associated with DFU through Pearson correlation analysis. Core genes were chosen based on gene significance (gene significance > 0.5) and module membership (module membership > 0.8). The plum³ module, with the highest correlation coefficient of 0.8, identified 252 hub genes with significant connectivity.

Data processing and differentially expressed gene screening

The differential analysis of the expression matrix comparing control and DFU conditions was conducted using the R package limma (version 3.40.6). Differentially expressed genes (DEGs) were identified based on the criteria of an absolute log fold change > 1 and P < 0.05, as implemented in the limma package.

Gene enrichment analysis

Visualization of DCPS-related genes by gene MANIA database (http://genemania.org). For gene set enrichment analysis (http://software.broadinstitute.org/gsea/index.jsp), we divided the samples into high expression groups (≥ 50%) and low expression groups (< 50%) according to the expression level of DCPS[26]. For gene set functional enrichment analysis, we performed Kyoto Encyclopedia of Genes and Genomes analysis and Gene Ontology analysis based on Metascape (http://metascape.org) to obtain gene set enrichment results[27].

Immune cell infiltration analysis

CIBERSORT is a deconvolution algorithm specifically developed for the analysis of gene expression data, using gene expression features to quantify the relative proportions of various immune cell types[28]. We conducted an immune cell infiltration analysis with the CIBERSORT platform, using an R script from its website and the LM22 gene signature file to assess the proportions of 22 immune cell subtypes in DFU vs control samples. Differences in immune cell influx were visualized with box plots created using the ggplot2 package.

qRT-PCR array

Total RNA was extracted using TRIzol^TM reagent (Invitrogen, Carlsbad, CA, United States) to obtain a comprehensive RNA profile. The synthesis of complementary DNA was conducted using the High-Capacity complementary DNA Reverse Transcription Kit (Invitrogen, CA, United States). An iTaq^TM Universal SYBR Green Supermix (Bio-Rad, Hercules, CA, United States) was used for qRT-PCR, and the data were analyzed with CFX^TM 96 Manager Software (Bio-Rad, Hercules, CA, United States). To assess mRNA expression levels, we normalized the data to β-actin. The primer sequences utilized are shown in Table 2.

Table 2 The primers of hub genes and β-actin.

Gene	Primer	Sequence (5’-3’)
hum-DCPS	Forward	GCAGCTCCTCAACTAGGCAAG
	Reverse	GAAGCCGGAGAACGGTAAGC
mus-DCPS	Forward	GATACAGCGCCTCAACTCAAG
	Reverse	GGAGAACGGTAAGCGGACAG
hum-β-actin	Forward	CATGTACGTTGCTATCCAGGC
	Reverse	CTCCTTAATGTCACGCACGAT
mus-β-actin	Forward	GTGACGTTGACATCCGTAAAGA
	Reverse	GCCGGACTCATCGTACTCC

Hum: Human; Mus: Mouse; DCPS: Decapping scavenger enzyme.

Immunofluorescent staining and morphological staining

Tissues were fixed in paraformaldehyde for 16-24 hours at 4 °C, then 5-μm sections were prepared and deparaffinized with sodium citrate for antigen retrieval. After paraffin embedding, a blocking buffer was applied, and the sections were incubated overnight at 4 °C with DCPS-specific antibodies (67311-1-Ig, Proteintech, Chicago, IL, United States). Secondary antibodies conjugated to Alexa Fluor 488 and Alexa Fluor 555 (Thermo Fisher Scientific, Waltham, MA, United States) were used in the procedure. 4’,6-diamidino-2-phenylindole was used as a stain in the mounting medium (C1005, Beyotime, Shanghai, China). 5-ethynyl-2’-deoxyuridine (EdU) staining (K1076; APExBIO, Houston, TX, United States) and Masson trichrome staining (G1346; Solarbio, Beijing, China) were conducted. Collagen quantification in stained skin sections was done using ImageJ software.

Cell culture

NHEK cells were obtained from Procell Company (Wuhan, China). Cells (G4557; Servicebio, Wuhan, China) were cultured in minimal essential medium at 37 °C with 5% CO₂. The medium included 1% penicillin-streptomycin (WelGENE Inc., Taiwan, China) and 10% fetal bovine serum (VivaCell, Shanghai, China). Experiments used NHEK cells at passages 3-6.

Knockdown of DCPS

Transfection utilizing small interfering RNA (siRNA) was conducted using INTERFERin^® (101000028; PolyPlus Transfection, Strasbourg, France). The targeted sequence of the DCPS siRNA used in this study was 5’-GCAGTTCTCCAATGATATCTA-3’.

Western blotting

Cellular proteins were isolated using an inhibitory radio-immunoprecipitation assay buffer and a phosphatase inhibitor. They were then electrophoresed on a polyacrylamide gel and transferred to polyvinylidene difluoride membranes. The membranes were subjected to a blocking process for 1 hour with a Tris-buffered saline-Tween solution in which 5% milk was incorporated. This took place following overnight incubation at 4 °C with primary antibodies specific to CDK6 (S0B0037; Starter Bio, Hangzhou, China), cyclin D1 (S0B2322; Starter Bio, Zhejiang, China), and β-Tubulin (10094-1-AP; Proteintech, Wuhan, China). Subsequent to washing three times with Tris-buffered saline-Tween, the membranes were incubated with secondary antibodies conjugated with horseradish peroxidase (1430; Thermo Fisher Scientific, MA, United States). Antigen detection was performed using a chemiluminescence kit (WLA006, WanLeiBio, Shenyang, China), with immunoreactive bands visualized via the ChemiDoc Touch imaging system (Bio-Rad, Hercules, CA, United States) and analyzed using ImageJ software.

NHEK migration

NHEK cells were cultured in six-well plates and subjected to a high glucose treatment (30 mmol/L) for 48 hours[18]. A scratch assay was performed by making a linear wound with a 20-μL pipette tip, and cellular responses were observed over 24 hours using the Lionheart FX automated microscope (Biotek, Winooski, VT, United States). After pretreatment, cells were plated at 4 × 10⁵ cells/mL in 100 μL serum-free medium in Transwell chambers. The lower compartments contained medium with 10% fetal bovine serum. After 48 hours of incubation, the Transwell plates were removed, and the cells were fixed with 4% paraformaldehyde for 15 minutes. The chambers were stained with Coomassie Brilliant Blue for 15 minutes, rinsed with phosphate-buffered saline, and debris was removed with cotton swabs before drying at room temperature. Samples were visualized, photographed, and counted using an inverted microscope.

Flow cytometry

The Apoptosis Analysis Kit (KGA1104; Keygen BioTECH, Nanjing, China) and the Cell Cycle Analysis Kit (MRC-C-23010; Miracle, Chengdu, China) were used to assess apoptosis in our cellular models. Additionally, EdU flow cytometry (K1078; APExBIO, Taiwan, China) staining was performed.

RNA m7G dot blot

Total RNA samples were quantified and adjusted to concentrations of 200 and 400 ng/μL. After 95 °C heat denaturation for 5 minutes, 2-μL aliquots of each concentration were spotted onto nitrocellulose membranes (66485; Solarbio, Beijing, China). Membranes underwent ultraviolet crosslinking followed by blocking with 5% skimmed milk. Subsequent incubations included overnight treatment with m7G antibody (68302-1-Ig; Proteintech, Wuhan, China) at 4 °C and horseradish peroxidase-conjugated secondary antibody. Protein signals were detected using a chemiluminescence detection system, with methylene-blue-stained membranes serving as loading controls.

mRNA stability

Actinomycin D (HY-17559; MedChemExpress, Monmouth Junction, NJ, United States), a transcriptional inhibitor, was used to block de novo RNA synthesis. Following treatment, cells were collected at 4 and 8 hours post-exposure for total RNA isolation. The residual mRNA levels of CDK6 and cyclin D1 were subsequently quantified by qRT-PCR.

Statistics analysis

Statistical analysis of supplementary data was performed using GraphPad Prism 9.3. In vitro experiments had at least three replicates. Student’s t tests compared two groups, while one-way ANOVA with Tukey’s post hoc tests compared multiple groups. Significance was set at P < 0.05.

RESULTS

Identification of key modules of WGCNA

The study design is shown in Figure 1. We selected the whole transcriptome genes from GSE134431, downloaded from the GEO database (Table 1), for WGCNA analysis. β = 30 (R² = 0.86) was selected as a suitable soft threshold for the construction of the scale-free network, and used to draw a sample dendrogram (Figure 2A-C). Modules were hierarchically clustered based on the TOM matrix and similar modules on the clustering tree were merged (Figure 2D). Spearman’s correlation coefficient was performed to map the module-trait relationship and evaluate the correlation between each module and diagnosis (Figure 2E). Nineteen co-expression modules were obtained, showing the correlation of these modules with clinical information (Figure 2F). The results indicated that the correlation analysis of genes within the plum³ module demonstrated a strong association with DFU and encompassed a total of 252 genes, r = 0.52, P = 2.8 × 10^-26 (Figure 2G).

Figure 1 Research framework. GSEA: Gene set enrichment analysis; WGCNA: Weighted gene co-expression network analysis; DGEs: Differentially expressed genes; m7G: N7-methylguanosine; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; PPI: Protein-protein interaction; DCPS: Decapping scavenger enzyme; ROC: Receiver operating characteristic.

Figure 2 Identification of key modules of weighted gene coexpression network analysis. A: Sample clustering dendrogram of 21 samples of GSE134431; B and C: The scale-free fit index for various soft-thresholding powers (β) and the mean connectivity for various soft-thresholding powers; D: Dendrogram of genes clustered via the dissimilarity measure (1-topological overlap matrix); E: Heatmap of eigengene adjacency; F: Heatmap of the correlation between genes and clinical traits; G: Correlation plot between module membership and gene significance of genes included in the plum³ module, which contains 252 hub genes. DFS: Disease free survival; DFU: Diabetic foot ulcer; NA: Not available.

Gene annotation and enrichment analysis

We performed a robust analysis of DEGs using the limma software package, focusing on a comparative evaluation between the diseased cohort and the control group within our curated dataset. This rigorous statistical approach yielded a total of 2933 DEGs (P < 0.05, absolute log fold change > 1). Subsequently, by integrating these DEGs with the plum³ modules, we refined our selection to 183 DEGs-plum³, providing a focused subset for further analysis (Figure 3A and B, Supplementary Table 1).

Figure 3 Gene annotation and enrichment analysis. A: Differentially expressed genes (DEGs) analysis between diabetic foot ulcer group and control group; B: Venn diagram illustrated the intersection among DEGs and the plum³ module, and a total of 183 candidate genes were obtained for subsequent analysis. In this diagram, red denotes DEGs, and blue signifies plum³; C: Metascape bar graph for viewing top nonredundant enrichment clusters, one per cluster, using a discrete color scale to represent statistical significance; D: Network of enriched terms, colored by cluster ID, where nodes that share the same cluster ID are typically close to each other; E: Top-level Gene Ontology biological processes; F: Protein-protein interaction network and molecular complex detection components identified in the gene lists. DGEs: Differentially expressed genes; MECP2: Methyl CpG binding protein 2; GTPase: Guanosine triphosphatase; MCODE: Molecular complex detection.

To gain deeper insights into the biological significance of these genes, we embarked on a comprehensive annotation endeavor using Metascape. This involved an expansive pathway and process enrichment analysis along with protein-protein interaction enrichment analysis. These analyses enabled us to delineate the functional or pathway enrichment pertinent to the DEGs-plum³, with cluster IDs and corresponding P values systematically represented (Figure 3C). Notably, regulatory pathways related to DNA and RNA levels were widely enriched. Following this, leveraging the curated list of 183 DEGs-plum³, we constructed an intricate protein-protein interaction network and conducted Gene Ontology analysis (Figure 3D and E). Through the utilization of the STRING and BioGrid databases, we meticulously analyzed each molecular complex detection component for path and process enrichment, extracting and preserving the top three items with the most significant P values. These items served as detailed functional annotations of each component were comprehensively summarized (Figure 3F, Supplementary Table 2). This multifaceted analytical approach has not only elucidated the complex biological pathways involved, but has also offered potential targeted insights for further experimental validation.

Immune cell infiltration and correlation analysis in DCPS and DFU

Previous analyses indicated that genetic material modification regulation was crucial in DFU pathogenesis. We summarized 34 m7G methylation regulators (Supplementary Table 3) and found significant differences in DCPS, a key m7G methylation regulator, by intersecting with the differentially expressed genes-plum³ module in DFU sequencing data (Figure 4A). We then compared DCPS gene expression levels between the DFU and control groups using violin plots (Figure 4B and C). The results showed that DCPS expression was significantly reduced in the DFU group. The exceptional area under the receiver operating characteristic curve (AUC) (AUC = 0.98; AUC = 0.99) for DCPS expression in distinguishing diabetic from normal tissues (Figure 4D and E) indicates outstanding diagnostic discriminative power. This suggests DCPS as a highly robust potential biomarker for diabetes detection.

Figure 4 Immune cell infiltration and correlation analysis in decapping scavenger enzyme and diabetic foot ulcer. A: Venn diagram illustrated the intersection among differentially expressed genes, the plum³ module, and N7-methylguanosine (m7G)-related genes, identifying decapping scavenger enzyme (DCPS) as a key candidate gene within the m7G-diabetic foot ulcer (DFU) category. In this diagram, red denotes differentially expressed genes, blue signifies plum³, and yellow indicates genes from the m7G; B: Differences in DCPS gene expression between DFU and control groups within the training set; C: Differences in DCPS gene expression between DFU and control groups in the validation set GSE7014; D: Receiver operating characteristic curve for DCPS in GSE134431; E: Receiver operating characteristic curve of DCPS in GSE7014; F-H: Stacked bar chart of the immune cell, the correlation matrix of immune cell proportions and the box plot of the DCPS related immune cell proportions in data set GSE134431; I-K: Stacked bar chart of the immune cell, the correlation matrix of immune cell proportions and the box plot of the DCPS related immune cell proportions in data set GSE7014. m7G: N7-methylguanosine; DCPS: Decapping scavenger enzyme; DGEs: Differentially expressed genes; DFU: Diabetic foot ulcer; NK cell: Natural killer cell.

The pathophysiology of DFUs is intricately linked to immune-cell-mediated signaling pathways. Thus, we implemented the CIBERSORT algorithm to delineate the immune cell landscapes in the aforementioned DFU tissue sequencing datasets, with the objective of deciphering the interplay between immune regulation and infiltrating immune cells within the DFU milieu. Immune infiltration analyses provided a comprehensive profile of 22 distinct immune cell types across each sample (Figure 4F-J). Notable disparities were observed in several immune cell subsets when contrasting DFU and control skin samples. Relative to the control group, the DFU group exhibited diminished levels of CD8⁺ T cells and activated natural killer cells, whereas there was an elevation in the levels of activated mast cells, neutrophils, and M1 macrophages (Figure 4H and K). Correlation analyses revealed statistically significant positive associations between regulatory T cells and naive CD4⁺ T cells (P < 0.0001), while static mast cells showed a negative correlation with M0 macrophages (P < 0.0001). Additionally, static mast cells were positively associated with naive CD4⁺ T cells (P < 0.0001), and memory resting CD4⁺ T cells were inversely correlated with CD8⁺ T cells (P < 0.0001; Figure 4F and I). These immune shifts fostered a detrimental microenvironment that impaired growth signaling and promoted extracellular matrix degradation, ultimately impeding the proliferative phase of wound healing. These insights have suggested that aberrant expression of DCPS in DFU could be critically implicated in the modulation of diabetic wound healing processes.

Identification of DCPS-related genes and pathways

The Gene MANIA analysis elucidated a comprehensive and intricate interaction network for DCPS, spotlighting its dynamic interplay with an array of proteins, notably including the exosome component family, anaphase promoting complex subunit family, NADPH dependent diflavin oxidoreductase 1, glypican 5, and histidine triad nucleotide binding protein 1, among others (Figure 5A). This extensive network underscored the multifaceted role of DCPS within cellular processes, aligning with its functional versatility. The primary functions of DCPS, as delineated by Gene MANIA, encompassed pivotal cellular mechanisms such as the response to the nuclear-transcribed mRNA catabolic process and manifested various enzymatic activities, including exonucleolytic, nuclease, and ribonuclease activities, alongside its involvement in exonuclease activity, deadenylation-dependent decay, and the exosome (ribonuclease complex) pathways (Figure 5B).

Figure 5 Analysis of decapping scavenger enzyme related genes and functions. A and B: Coexpressed genes of decapping scavenger enzyme were analyzed by Gene MANIA; C-H: Enrichment maps from genomic gene set enrichment analysis enrichment analysis in diabetic foot ulcer. RNase: Ribonuclease; ES: Enrichment score; NP: Nominal P value.

Gene set enrichment analysis complemented these findings by revealing that DCPS was predominantly associated with critical biological processes and cellular structures, including chromatin region regulation, DNA damage site response, transcriptional extension complexes, spindle formation, tubulin assembly, and cell cycle-dependent protein regulation (Figure 5C-H). These insights have aligned with the emerging view of DCPS as a pivotal player in maintaining genomic integrity, facilitating proper cell division, and modulating transcriptional activities, thereby underscoring its essential role in cellular homeostasis and suggesting a potential pathway for its involvement in the healing process of diabetic wounds.

Validation of DCPS expression in DFU

To further elucidate the role of the hub gene DCPS within the context of skin wound healing, we systematically investigated its expression patterns leveraging data from the GEO database. More specifically, dataset GSE23006 provided an invaluable window into the temporal dynamics of DCPS expression following skin injury. Our analysis revealed a marked diminution in DCPS expression levels approximately 6 hours post-wounding, followed by a gradual re-establishment of homeostatic expression levels as healing ensued (Figure 6A). Given the sequential progression of wound healing phases, hemostasis, inflammation, proliferation, and remodeling, this temporal expression pattern underscores the potential significance of DCPS in the orchestration of the wound healing cascade. The findings suggest DCPS may facilitate cell proliferation, migration, and tissue reconstruction, thereby promoting wound healing.

Figure 6 Validation of decapping scavenger enzyme expression in skin wounds. A: Decapping scavenger enzyme (DCPS) mRNA expression decreased significantly at 6 hours post-injury and gradually recovered during wound healing; B and C: DCPS expression was reduced in skin tissues of both diabetic foot ulcer patients and diabetic mice at the RNA level, n = 3; D and E: Immunofluorescence analysis revealed reduced DCPS (red) expression in human diabetic skin tissues, n = 3 (scale bar: 400 μm); F and G: Immunofluorescence revealed reduced DCPS (red) expression in diabetic mouse skin tissues, n = 3 (scale bar: 400 μm). Statistical analyses were performed using GraphPad Prism 8. The results were expressed as mean ± SD. DFU: Diabetic foot ulcer; DCPS: Decapping scavenger enzyme; DM: Diabetes mellitus; NC: Negative control. ^aP < 0.001, and ^bP < 0.01.

To deepen our understanding of DCPS expression disparities in DFU vs nondiabetic contexts, we conducted a quantitative analysis of mRNA expression in skin tissues from both human subjects and murine models (Figure 6B and C). Our findings indicated a pronounced downregulation of DCPS in DFU-afflicted tissue as well as in the cutaneous tissues of diabetic mice. This downregulation was validated through rigorous quantitative immunofluorescence analyses, which corroborated the diminished presence of DCPS in these tissues (Figure 6D-G). DCPS expression was notably reduced in the upper skin layers of both DFU patients and diabetic mice, indicating its key role in regulating epithelial cell function. The significant changes in DCPS levels in the epithelial compartment of DFU wounds suggested an important, previously overlooked role for DCPS in the remodeling and repair of epithelial cells.

Assessment of NHEK function following DCPS knockdown in vitro

Given the significant impact of DCPS on skin epithelial cells, we first demonstrated that m7G levels in epithelial cells are suppressed under diabetic conditions using RNA dot blot assays (Figure 7). Subsequently, we knocked down DCPS with siRNA in epithelial cells and investigated its functional effects on NHEK in vitro (Figure 7). Initially, to simulate a hyperglycemic environment in endothelial cells, a high glucose concentration of 30 mmol/L was utilized. In light of the previously established association between DCPS and the cell cycle, we conducted an evaluation of the impact of DCPS on the epithelial cell cycle utilizing flow cytometry (Figure 7 and Supplementary Figure 1). Our results indicated that the G1 and S phases proceeded unimpeded under both normal physiological conditions and hyperglycemic pathological environments. The G2 phase was markedly inhibited following DCPS knockdown. During the G2 phase, cells either progress to the terminal mitotic phase or permanently exit the cell cycle[29,30]. Knockdown of DCPS induced cell cycle exit from the G2 phase, thereby significantly disrupting the ultimate trajectory of cell cycle progression (P < 0.05) (Figure 7 and Supplementary Figure 1). Subsequently, we conducted an analysis of key proteins involved in cell cycle regulation through western blot (Figure 7). This revealed that expression of CDK6 and cyclin D1 was reduced following DCPS knockdown, further elucidating the role of DCPS in the regulation of cell proliferation (P < 0.05; Figure 7). To further investigate the effects of DCPS knockdown, we examined the RNA stability of cyclin D1 and CDK6 transcripts. DCPS depletion significantly reduced RNA stability of these transcripts (P < 0.05; Figure 7). We used flow cytometric EdU assays, which demonstrated that cells treated with siDCPS (siRNA targeting DCPS) exhibited a significant reduction in proliferation, particularly under the simulated pathological conditions (P < 0.05; Figure 7 and Supplementary Figure 1). The results from the immunofluorescence staining of the EdU assay were consistent with the results obtained from the flow cytometric analysis of the EdU assays (P < 0.05; Figure 7 and Supplementary Figure 1). We examined the impact of siDCPS intervention on apoptosis. Flow cytometric analyses demonstrated that, under simulated pathological conditions, siDCPS elicited a significantly greater apoptotic response (P < 0.05; Figure 7 and Supplementary Figure 1). The outcomes of cell scratch assays and Transwell assays were utilized to assess the influence of siDCPS intervention on the migratory capacity of NHEKs. The migration of NHEKs was meticulously documented through continuous live cell imaging over 48 hours, with a black horizontal line employed to delineate the extent of wound closure (Figure 7 and Supplementary Figure 1). The quantitative analyses of the wound closure rate indicated that siDCPS significantly impaired the migratory capacity of NHEKs in physiological and pathological contexts (P < 0.05; Supplementary Figure 1). Importantly, this effect was more pronounced under simulated pathological conditions (P < 0.05; Figure 7). The results of the Transwell assay agreed with those obtained from the cell scratch assays (P < 0.05; Figure 7 and Supplementary Figure 1). Consequently, siDCPS intensified the functional impairment of NHEKs induced by elevated glucose levels, and DCPS played a significant role in regulating epithelial cell proliferation, cell cycle progression, and migratory capacity in diabetic wounds. This suggests DCPS might be a useful target for local treatment of DFU.

Figure 7 Assessment of normal epidermal keratinocyte function following decapping scavenger enzyme knockdown in vitro. Cells were treated with 30 mmol/L glucose for 48 hours before being exposed to siDCPS [small interfering RNA targeting decapping scavenger enzyme (DCPS)] for 24 hours. A: Dot blot assay showed downregulation of N7-methylguanosine under diabetic condition; B: Efficient DCPS knockdown by small interfering RNA (n = 4). Evaluate cell cycle distribution changes induced by DCPS knockdown; C and D: DCPS knockdown induced significant G2 phase arrest in cell cycle, n = 3; E-G: DCPS knockdown reduced cyclin-dependent kinase 6 and cyclin D1 protein expression, n = 3; H and I: DCPS knockdown compromises RNA stability of cyclin D1 and cyclin-dependent kinase 6 transcripts. Assess the impact of DCPS knockdown on cellular proliferation; J and K: 5-ethynyl-2’-deoxyuridine staining and statistical analyses in normal epidermal keratinocytes (NHEKs), n = 4 (scale bar: 100 μm); L and M: DCPS knockdown attenuated cellular proliferation in flow cytometry assays, n = 3; N and O: DCPS knockdown enhanced apoptosis in flow cytometry, n = 3. DCPS knockdown inhibits NHEK migration; P and Q: Transwell assays were conducted for evaluation of NHEK migration, n = 3 (scale bar: 100 μm); R-U: Cell scratch assays were observed in living cells at 0, 12, 24 and 48 hours, n = 3 (scale bar: 200 μm). The results were expressed as mean ± SD. NC: Negative control; HG: High glucose; m7G: N7-methylguanosine; siDCPS: Small interfering RNA targeting decapping scavenger enzyme; RMSD: Root mean square deviation; CDK6: Cyclin-dependent kinase 6; 7-AAD: 7-Aminoactinomycin D. ^aP < 0.001, ^bP < 0.01, and ^cP < 0.05.

DISCUSSION

The widespread use of sequencing technology has increased reliance on bioinformatics to study disease-related genes, identify biomarkers, explore signaling pathways, and find therapeutic targets, providing vital data for understanding disease mechanisms[24]. In this context, the present study uncovers a previously unrecognized association between m7G methylation and DFU, highlighting alterations in methylation levels and diminished DCPS expression in the skin of diabetic mice and patients, which may be attributed to the prolonged and chronic effects of hyperglycemia. DCPS, functioning as a correlative regulator of m7G methylation, significantly influences various biological functions of keratinocytes following knockdown, including the inhibition of cell cycle progression, reduction in cell proliferation, elevation of apoptosis rates, and impairment of migratory capabilities. In summary, this work demonstrates that DCPS is correlated with m7G methylation and affects keratinocyte proliferation, apoptosis, and migration in diabetic wound healing. Inhibiting DCPS impairs keratinocyte function, highlighting its importance as a regulatory factor and predictive biomarker for m7G methylation in DFU.

m7G methylation has been implicated in the modification of the 5’ cap of ribosomal RNA, transfer RNA, and mRNA[31-33]. In the complex environment of DFU, the alterations influenced by the m7G warrant further investigation. Consequently, we used WGCNA in conjunction with Metascape’s biological information analysis to identify aberrant methylation levels in sequencing datasets related to DFU. This approach led to the identification of DCPS as a gene exhibiting significant downregulation of m7G methylation in the DFU sequencing datasets. Further analysis revealed significantly reduced DCPS expression in the wound skin of diabetic C57 mice induced by DFU and streptozotocin, particularly in the upper dermis. This finding aligns with reports of epidermal thinning in diabetic patients compared to healthy controls[34]. Keratinocytes, the primary constituents of the epidermis, are critically implicated in the delayed wound healing observed in DFU patients. Diabetic wounds are characterized by impaired keratinocyte proliferation and migration[35,36], and 42 functional deficits exacerbated by hyperglycemia[37,38]. Type 2 diabetes mellitus adversely affects epidermal function, reducing lipid synthesis and antimicrobial peptide expression, factors that likely contribute to prolonged wound healing[39]. These findings offer supplementary evidence to support our prior bioinformatics analysis and provide preliminary validation for the hypothesis that DCPS is downregulated in the skin within a diabetic context. We propose that DCPS may serve as a critical candidate for m7G-related DFU, potentially playing a pivotal role in the healing of diabetic wounds.

DCPS functions as a regulator of m7G methylation, operating as a symmetric homodimer that undergoes dynamic cycling between closed and open conformations, which correspond to substrate binding and product release, respectively[40]. To investigate the impact of DCPS-mediated m7G on epithelial cell function within a diabetic context, we developed an in vitro pathological model utilizing NHEKs. Previous research has indicated a potential involvement of DCPS in the regulation of the cell cycle; however, the underlying mechanisms remain inadequately understood[41,42]. Our findings indicate that the knockdown of DCPS impedes the progression of the epithelial cell cycle, with a pronounced impact on the G2 phase, which serves as a critical checkpoint in the cell cycle. During this phase, there is an increased demand for intracellular protein synthesis in preparation for subsequent mitotic events[43,44]. Methylation is crucial for mRNA modification, which in turn influences the synthesis of functional proteins[45,46]. As the principal regulator of m7G methylation, DCPS plays a crucial role in the regulation of the G2 phase of the cell cycle. Its knockout results in significant inhibition of the G2 phase, ultimately forcing the cell to exit the cell cycle entirely. Knockdown of DCPS resulted in significant impairment of cell proliferation and migration, accompanied by an increase in apoptosis. Through mechanistic analyses, we identified DCPS as a key regulator mediating m7G-dependent epithelialization processes in DFU wound repair. Our findings reveal the dual clinical relevance of DCPS: It exhibits promising translational potential as a diagnostic biomarker for DFU while concurrently offering novel therapeutic targets. This discovery substantially advances our understanding of the molecular pathophysiology underlying DFU complications.

It is important to acknowledge this study’s limitations. Firstly, the clinical data utilized were sourced from a public database, which resulted in incomplete clinical information for the samples, notably the absence of comprehensive clinicopathological features within the GSE series. Secondly, the upstream regulatory mechanisms of m7G in DFU remain unclear, particularly concerning its involvement in the processes related to wound healing and re-epithelialization. Finally, while this study established DCPS as an effective biomarker associated with m7G in DFU and has validated its role through in vitro experiments, further in vivo investigations are necessary to elucidate its potential therapeutic effects on wound healing. Collectively, our findings offer novel insights into the relationship between m7G methylation and diabetic wound healing. Specifically, DCPS-mediated m7G methylation is crucial for the activity of epithelial cells during the healing process in diabetes. These insights establish a robust evidence base for the future clinical application of m7G methylation regulation in the treatment of DFU and for the identification of potential disease biomarkers.

CONCLUSION

This study identified DCPS as a promising biomarker and potential therapeutic target for DFU. As a crucial regulator of m7G methylation, DCPS significantly influences the modulation of cell cycle progression, proliferation, and migration of epithelial cells during the wound healing process associated with DFU. These findings enhance our understanding of the molecular mechanisms governing diabetic wound healing and highlight a potential therapeutic pathway for further exploration of m7G methylation levels, which could have important clinical applications in the management of DFU, especially in the context of the anticipated increase in metabolic-related lower limb diseases in the future[47].