Development of novel lysosome-related signatures and their potential target drugs based on bulk RNA-seq and scRNA-seq for diabetic foot ulcers

Tan, Longhai; Qu, Junjun; Wang, Junxia

doi:10.1186/s40246-024-00629-1

Development of novel lysosome-related signatures and their potential target drugs based on bulk RNA-seq and scRNA-seq for diabetic foot ulcers

Research
Open access
Published: 11 June 2024

Volume 18, article number 62, (2024)
Cite this article

Download PDF

You have full access to this open access article

Human Genomics Aims and scope Submit manuscript

Development of novel lysosome-related signatures and their potential target drugs based on bulk RNA-seq and scRNA-seq for diabetic foot ulcers

Download PDF

Longhai Tan¹,
Junjun Qu² &
Junxia Wang¹

446 Accesses
Explore all metrics

Abstract

Background

Diabetic foot ulcers (DFU) is the most serious complication of diabetes mellitus, which has become a global health problem due to its high morbidity and disability rates and the poor efficacy of conventional treatments. Thus, it is urgent to identify novel molecular targets to improve the prognosis and reduce disability rate in DFU patients.

Results

In the present study, bulk RNA-seq and scRNA-seq associated with DFU were downloaded from the GEO database. We identified 1393 DFU-related DEGs by differential analysis and WGCNA analysis together, and GO/KEGG analysis showed that these genes were associated with lysosomal and immune/inflammatory responses. Immediately thereafter, we identified CLU, RABGEF1 and ENPEP as DLGs for DFU using three machine learning algorithms (Randomforest, SVM-RFE and LASSO) and validated their diagnostic performance in a validation cohort independent of this study. Subsequently, we constructed a novel artificial neural network model for molecular diagnosis of DFU based on DLGs, and the diagnostic performance in the training and validation cohorts was sound. In single-cell sequencing, the heterogeneous expression of DLGs also provided favorable evidence for them to be potential diagnostic targets. In addition, the results of immune infiltration analysis showed that the abundance of mainstream immune cells, including B/T cells, was down-regulated in DFUs and significantly correlated with the expression of DLGs. Finally, we found latamoxef, parthenolide, meclofenoxate, and lomustine to be promising anti-DFU drugs by targeting DLGs.

Conclusions

CLU, RABGEF1 and ENPEP can be used as novel lysosomal molecular signatures of DFU, and by targeting them, latamoxef, parthenolide, meclofenoxate and lomustine were identified as promising anti-DFU drugs. The present study provides new perspectives for the diagnosis and treatment of DFU and for improving the prognosis of DFU patients.

A novel diabetic foot ulcer diagnostic model: identification and analysis of genes related to glutamine metabolism and immune infiltration

Article Open access 30 January 2024

Identifying potential pathogenesis and immune infiltration in diabetic foot ulcers using bioinformatics and in vitro analyses

Article Open access 01 December 2023

Three genes expressed in relation to lipid metabolism considered as potential biomarkers for the diagnosis and treatment of diabetic peripheral neuropathy

Article Open access 29 May 2023

Introduction

DFU is the most common and serious complication in diabetic patients, which is caused by the interaction of multiple risk factors [1]. With the increasing number of diabetic patients worldwide, the incidence of DFUs is also on the rise [2]. The high incidence and disability of DFUs not only lead to a decline in the quality of patients’ lives, but also cause a heavy burden on healthcare and nursing care [3]. It is estimated that between 9.1 and 26.1 million people develop diabetic foot ulcers worldwide each year, with approximately 17% of these cases eventually requiring amputation [4]. It has been shown that a DFU-related amputation occurs every 20 s in the world [5, 6], and the cost of each amputation can be more than $53,500 [7], which imposes a severe financial, physical, and psychological burden on the patient [8]. Currently, the clinical diagnosis of DFU mainly includes the diagnosis of lower limb vasculopathy and peripheral neuropathy [9, 10]. The diagnosis of lower extremity vasculopathy is based on the following criteria: (1) the diagnosis of diabetes mellitus; (2) clinical manifestations of lower extremity ischemia; (3) auxiliary examinations suggesting lower extremity vasculopathy, with an ABI of < 0.9 at rest, or an ABI of > 0.9 at rest, but with discomfort in the lower extremities during exercise, and a decrease in ABI of 15–20% after a plate exercise test, or with stenosis of blood vessels on imaging. The diagnosis of peripheral neuropathy is based on: (1) abnormal temperature sensation; (2) decreased or lost foot sensation in nylon wire examination; (3) abnormal vibration sensation; (4) loss of ankle reflexes; (5) slowing down of 2 or more items of NCV. If two or more of the above five tests are abnormal, the diagnosis is peripheral neuropathy. The clinical diagnosis of DFU mentioned above is complicated and relies heavily on the experience of clinicians, without effective molecular level diagnostic support, that is, there is a lack of effective biomarkers that can be used for clinical diagnosis of DFU. Additionally, there are still limitations in the multiple treatments for DFU, such as treating lower extremity ischemia and foot infections, surgical debridement, intravenous antibiotics, and reducing the pressure of weight-bearing on the ulcers [11,12,13]. Despite recent advances in the treatment of DFUs, a large proportion of patients with DFUs still develop chronic wounds due to irreversible processes [14]. Thus, it is urgent to identify novel biomarkers for its diagnosis and treatment.

Lysosomes are signaling pivots and degradation centers in eukaryotic cells that not only play key roles in processes such as senescence, cellular homeostasis and development; they also act as signaling centers for signal transduction, energy and amino acid sensing, and autophagy regulation [15]. Multiple aspects of these functions that converge on the lysosome are able to connect; when these pathways are unregulated, they become the basis for a wide range of human diseases [16]. It has been shown that relevant lysosomal dysfunction has been found in a variety of diseases including autoimmune, metabolic and renal disorders such as Parkinson’s disease, diabetes mellitus and lysosomal storage disease [17]. Of these, diabetes is one of the leading causes of chronic wound healing problems. When diabetics develop ulcers, they are at high risk for major complications such as infection and amputation [18]. And the DFU is one of the serious complications of diabetes that leads to chronic wound healing problems. Previous studies have shown that lysosomal abnormalities can cause diabetes mellitus, but DFU, as the most common and serious complication of diabetes mellitus, lysosomal abnormalities also have a contributing role in the development of DFU [17]. Thus, in this direction, it is necessary and clinically significant to identify lysosome-related genes that play critical regulatory roles in DFU by whole transcriptome analysis, to explore the molecular changes and functions of lysosome-related genes in the development of DFU, and to ascertain the molecular mechanisms by which lysosome-related genes contribute to the development of DFU.

In the present study, we found that Clusterin (CLU), glutamyl aminopeptidase (ENPEP) and RAB guanine nucleotide exchange factor 1 (RABGEF1) can be indicated by multiple bioinformatic and machine learning algorithms as novel hub lysosome-related genes of DFU (DLGs). Then we constructed a novel artificial neural network (ANN) diagnostic model based on the DLGs to assist in the clinical diagnosis of DFU at the molecular level, and the constructed ANN model showed good diagnostic performance in the training and validation cohorts. Following that, we explored the immune microenvironment in DFU and its relationship with DLGs through immune infiltration analysis performed by the ssGSEA algorithm. Immediately after that, we performed single-cell transcriptome analysis on DFU patients, which revealed that DLGs were also heterogeneously expressed among different cell types, which provided stronger evidence for DLGs as diagnostic target for DFU. Finally, we used molecular docking method to identify drugs that have potential therapeutic or palliative effects on DFU by targeting DLGs. In conclusion, the present study provides new perspectives for understanding the molecular mechanism of DFU and diagnosis and treatment of DFU.

Methods

Data downloading and processing

We downloaded bulk RNA-seq datasets associated with DFU from the GEO database, and datasets that matched the following criteria were included in the present study: Firstly, the dataset is expected to contain unbiased gene expression data, complete annotation information. Second, the DFU patients incorporated into the training cohort and validation cohort should be distinct and independent. Based on the mentioned above criteria, four datasets (GSE80178, GSE134431, GSE7014 and GSE68183) were included in the present study. The basic information of the datasets used in the present study is provided in Table 1. In the present study, GSE80178 and GSE134431 are used as the training cohorts, and GSE7014 and GSE68183 are used as the validation cohorts. Meanwhile, in order to remove the batch effect between datasets caused by different platforms, we used “ComBat” in the R package “sva” to remove the batch effect in the training cohort after merging GSE80178 and GSE134431 and in the validation cohort after merging GSE7014 and GSE68183, respectively [19], and we used principal component analysis (PCA) to assess the effect of de-batching effects. Additionally, the scRNA-seq dataset associated with DFU was obtained from the previous study by Theocharidis, Georgios et al. [20].

Table 1 The basic information regarding the dataset in the present study

Full size table

Differential analysis of gene expression

The expression profiles of DFUs and Ctrls were compared using the R package “limma”, and dysregulated expression genes (DEGs) were identified in both clusters using a P-value < 0.05 as a criterion.

Weighted correlation network analysis

The “WGCNA” software package was used to identify DFU-related genes in the training cohort [21]. Initially, clustering of samples is performed to detect and exclude outliers to ensure the network analysis is robust. Then constructed the network using a soft thresholding capability to highlight strong correlations and punish weak correlations. Subsequently, we transformed the neighbor-joining matrix into a topological overlap matrix (TOM). Based on the measure of variance of the TOM, and set the minimum module size to 500, and grouped genes with similar expression patterns into the same module by means of average correlation hierarchical grou**. Eventually, we assessed module characterized genes for their correlation with DFUs and identified the matches to the study objectives based on the extent of the correlation.

Protein-protein interaction analysis

We analyzed protein-protein interactions (PPI) between dysregulated lysosomal genes using the STRING database (https://string-db.org/) and Cytoscape software [22]. Dysregulated lysosomal genes were included in the PPI network, and the CytoHubba function in Cytoscape was utilized to provide a composite score for dysregulated lysosomal genes, and the top 40 genes in the composite score were ultimately selected for subsequent analysis.

Functional enrichment analysis

Aiming to ascertain which biological processes and functions the top 40 scoring genes were mainly enriched, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis on the top 40 scoring genes using the R package “clusterProfler” [23]. Pvalue < 0.05 was regarded as statistically significant.

Selection of hub lysosome-related genes of DFU

We used 3 machine algorithms to identify DLGs, namely: least absolute shrinkage and selection operator (LASSO), support vector machine recursive feature elimination (SVM-RFE), and random forest(RF). It has been shown that machine learning has been extensively applied in the biomedical sciences and is capable of efficiently and rapidly analyzing biological data and accurately identifying hub genes in gene expression profiles [Diagnostic evaluation of DLGs for DFU

We further investigated whether the selected DLGs had potential value in diagnosing DFU, and thus evaluated the performance of the DLGs. We performed receiver operating characteristic (ROC) analysis using the R software package “pROC” to derive area under roc curve (AUC) values. Specifically, we obtained expression of DLGs and disease state grou** data from DFU sets, performed ROC analysis using the “roc” function of the “pROC” software package, and derived the final AUC results using the “ci” function of “pROC”.

Construction and verification of the artificial neural network model

We constructed an ANN diagnostic model based on the transcriptome level of DLGs using the R package “neuralnet”. The seed was set to 123 after we normalized the gene expression data using the min-max normalization method.

The constructed ANN model consisted of predominantly three layers:

1) Input layer, which mainly consists of the gene expression of the three normalized DLGs.

2) Hidden layer, which mainly includes the gene expressions of the three normalized DLGs and the weights of the three DLGs, as well as the weights between the three hidden layers.

3) Output layer, which indicates the result of judging whether the sample belongs to the Ctrls or the DFUs.

Single-gene GSEA analysis

We clustered DFUs based on the median expression values of DLGs with gene set enrichment analysis (GSEA) for different subgroups to study which biofunctions and signalling pathways associated with DLGs, and considered P < 0.05 to be statistically significant.

Evaluation of immune cell infiltration and their correlation with the DLGs

Infiltration of immune cells was assessed using ssGSEA [28]. Specifically, ssGSEA was performed in R language using the R packages “GSVA” and “GSEABase” and the immunological characteristics of DFU patients were assessed using the ssGSEA algorithm, respectively. We firstly obtained the information of the genes of 28 immune cells gene sets (Table S1) from the TISIDB database (http://cis.hku.hk/TISIDB/), and then performed single-sample gene set enrichment analysis (ssGSEA) and calculated ssGSEA scores. We used the “pheatmap” R package to visualize the infiltration levels of different immune cells under different infiltration algorithms. To assess the differential infiltration abundance of different immune cells between Ctrls and DFUs, we used the Wilcox test for pairwise comparisons. Subsequently, we used the “ggplot” R package to visualize the infiltration levels of different immune cells under different infiltration algorithms. Subsequently, we used the “ggplot” R package to visualize the correlation between immune cell infiltration abundance and DLGs expression.

scRNA-seq analysis

We downloaded scRNA-seq data of DFU from the previous study by Theocharidis, Georgios et al. [20]. , and analyzed them using the R package “Seurat” [29]. Before analyzing the scRNA-seq data, we weeded out low-quality cells using the following method: cells are likely to be under stress when they have the highest percentage of mitochondrial genes of all genes. Thus, cells with more than 25% of mitochondrial genes will be filtered. Since low-mass cells or empty droplets usually contain less genes and bicellular cells contain more genes, low-mass cells were filtered using the criteria nFeature RNA < 500 and nFeature RNA > 5000. The results yielded 71,718 cells and 14,776 genes. Next, we normalized the gene expression of the cells using the “NormalizeData” function and performed PCA using the ElbowPlot function to extract the top 20 principal components (PCs), which were further analyzed using the “FindVariableFeatures” function. For unsupervised and unbiased clustering of cell subpopulations, we used the “FindNeighbors”, “FindClusters” (resolution = 0.5) and “RunUMAP” functions. We then annotated the cell types using known markers [30,Full size image

Identification and docking potential drugs that target DLGs

The previous studies have shown that the process of drug discovery begins with the identification of disease targets, and target-based drug discovery is the most common strategy for new drug development [42, 43]. The mentioned above showed that DLGs are promising as diagnostic targets for DFU, which means that we can also identify drugs that can treat or alleviate DFU by targeting DLGs. Specifically, we used the Enrichr platform (https://maayanlab.cloud/Enrichr/) for online analysis and identification. According to the DSigDB database, we identified four drugs that were able to target DLGs with p-values less than 0.05. Immediately following, we investigated the binding affinity of the selected drugs to DLGs using molecular docking techniques. The results showed that all four drugs were able to target CLU and the binding energies were relatively high, respectively, CLU-latamoxef (-7.1 kcal/mol), CLU-parthenolide (-6.0 kcal/mol), CLU-meclofenoxate (-5.1 kcal/mol) and CLU- lomustine (-4.5 kcal/mol) (Fig. 8A-D). Subsequently, latamoxef and meclofenoxate were both able to target ENPEP, respectively ENPEP-latamoxef (-8.6 kcal/mol) and ENPEP-meclofenoxate (-5.9 kcal/mol) (Fig. 8E-F). Additionally, parthenolide and lomustine were also able to target RABGEF1, respectively RABGEF1-parthenolide (-6.3 kcal/mol) and RABGEF1-lomustine (-5.1 kcal/mol) (Fig. 8G-H). Taken together, the docking results suggest that these potentially targeted drugs may influence the development of DFU by interacting with DLGs to treat or alleviate the DFU symptoms.

Discussion

DFU is a serious complication of diabetes mellitus [44], which has become a global health problem [45]. According to current statistics, approximately 80% of lower limb amputation patients have diabetic foot before surgery [46]. Although some technologies have been developed to treat DFU, the current clinical interventions and efficacy for DFU fail to reach expectations [47, 48]. The high disability and mortality rates of DFU not only impose serious financial, physical and mental burdens on patients, but also cause a heavy burden on healthcare and nursing care, thus it is urgent to identify novel molecular therapeutic targets to help improve patient prognosis and reduce disability [3]. Previous studies have shown that lysosomal abnormalities lead to various autoimmune and metabolic diseases including diabetes, but DFU, as the most common and serious complication of diabetes, lysosomal abnormalities also contribute to DFU development [17]. In addition, lysosomes as novel targets for autoimmune, metabolic and renal diseases have been frequently reported in recent years [49]. Thus, in the present study, we explored the potential of lysosomes as biomarkers for DFU, aiming to contribute to the diagnosis and treatment of DFU.

In the present study, we comprehensively analyzed DFUs based on bulk RNA-seq and scRNA-seq data by various bioinformatic and machine learning algorithms. Specifically, we first identified DEGs associated with DFU by differential analysis and WGCNA analysis, and isolated lysosome genes from DEGs using lysosome genes annotated in AmiGO2 for lysosomal genes with potential regulation in DFU. Notably, previous analyses were analyzed and identified based on transcriptome expression patterns only, without taking into account the interactions between the genes themselves, so we utilized PPI analysis to ensure the accuracy of our results on a multidimensional level. Subsequently, we performed enrichment analysis on these genes to validate their reliability, and the results showed that they were significantly associated with lysosomes, substance transport across membranes, and immune/inflammatory responses, which implies a certain degree of reliability of our results. To identify DLGs in DFUs, we applied three machine learning algorithms, and the results showed that CLU, ENPEP and RABGEF1 were indicated by all machine learning algorithms, implying their potential regulatory role in the DFU development. It’s worth mentioning that these genes have been shown to be used to treat other diseases as well, for example, CLU can serve as a marker and mediator of chemoresistance in colorectal cancer [50]. Iron deficiency in liver tissue downregulates ENPEP, promotes angiogenesis in liver tumors and is associated with poor prognosis in hepatocellular carcinoma patients [51]. Epithelial RABGEF1 deficiency promotes intestinal inflammation by dysregulating intrinsic MYD88-dependent innate signaling [52].

It is possible that DLGs may be involved mechanistically in the occurrence and development of DFUs, and thus may also be a potential diagnostic target for DFUs. Certainly, validation of the diagnostic performance of DLGs for DFU requires artificial neural network modeling [53]. As one of the main types of artificial intelligence, ANN has been widely used in clinical medicine for diagnosis and treatment by virtue of its better algorithms and accuracy [54, 55]. However, there are still some deficiencies in the current clinical diagnosis of DFU, especially the diagnosis based on the molecular level is not yet complete, and very often relies on the clinical experience of clinicians [56]. Therefore, we built an ANN model to predict whether the samples in the present study are control or DFU based on DLGs transcriptome patterns, in which more than 80% prediction accuracy was achieved for both the training and validation sets. Meanwhile, we evaluated the prediction ability of the ANN model for the training and validation sets with ROC curves, in which the AUC value of the training set was 0.876 and the AUC value of the validation set was 0.790, which suggests that there is potential for an ANN built based on the DLGs transcriptome pattern to be an independent diagnostic predictor of DFU.

Dysregulation of immune cell infiltration is also a key factor in the deterioration of DFU [41]. In molecular terms, the wound healing process occurs after the breakdown of the skin’s protective barrier. Wound healing is a regulated and highly complex biological process. It is reported that immune/inflammatory regulation is crucial in the wound healing process [57]. To investigate the role of immune/inflammatory cells in DFU wound deterioration, we performed a comprehensive assessment of the immune microenvironment of DFUs and presented different cellular components in the immune microenvironment between DFUs and Controls. Our study demonstrated the high abundance of eosinophils and neutrophils in DFUs. It has been previously shown that neutrophils are the first inflammatory leukocytes to migrate to the wound site, and that they are capable of eliminating invading pathogens and initiating inflammatory and non-inflammatory responses through various mechanisms [58], implying that the high abundance of neutrophils in DFUs may worsen the symptoms of DFUs through excessive inflammatory responses. Additionally, we found a significant abundance decrease of effector memory CD4 T cells, immature B cells, and natural killer cells in DFUs. The significant decrease in the abundance of B/T cells and natural killer cells, the mainstay of defense against exogenous pathogenic microorganisms, also increased the risk of worsening of DFU wound infections.

The technology of single-cell RNA sequencing presents itself as an advanced method for revealing the complexity and heterogeneity of RNA transcripts within a single cell, as well as the composition and function of different cell types within a tissue [59]. Relying on the advantages of single-cell RNA sequencing technology in studying tissue heterogeneity, we also utilized this technology to perform single-cell profiling of DFUs. Our study showed that DLGs are also heterogeneously expressed among different cell types, which also provides stronger evidence that lysosome-related genes can be used as diagnostic targets for DFU. In addition, we found that the percentage of macrophages in DFUs was significantly higher compared with controls. It has been previously shown that macrophages play a role in hemostasis and that pro-inflammatory macrophages infiltrate and remove dead cells and bacteria from the wound site after injury [60]. During the wound healing phase, macrophage polarization stimulates the movement and growth of fibroblasts, endothelial cells, and keratinocytes, which accelerates repair of the epidermis, dermis, and vascular system [57]. Additionally, it has been shown that decreased innate immunoregulatory function exacerbates macrophage polarization imbalance as well as decreased wound healing [61], which is also consistent with the results of the immune infiltration analyses mentioned above, suggesting that a significant down-regulation of the abundance of dominant immune cells in DFUs, including B and T cells, leads to a decrease in wound healing capacity.

DLGs have not only been validated from bulk RNA-seq and scRNA-seq, but also play a key role in regulating the immune and inflammatory states of DFU, which means that DLGs are promising targets for the diagnosis and treatment of DFU, and it is also possible to identify drugs to treat DFU by targeting DLGs. Previous studies have shown that the process of drug discovery begins with the identification of disease targets, and target-based drug discovery is the most common strategy for new drug development [42, 43]. Thus, targeting DLGs may provide a new effective therapeutic approach for DFU treatment. In the present study, we identified four potential drugs targeting DLGs using the Enrichr platform: latamoxef, parthenolide, meclofenoxate, and lomustine. The identified targeted drugs have also been demonstrated to treat immune/inflammatory diseases, including cancer, in previous study reports. For example, Latamoxef was able to largely treat febrile neutropenic patients [62]. Parthenolide has potential therapeutic effects in luminal A breast cancer [63]. Meclofenoxate is effective in the treatment of neuroleptic-induced dyskinesia [64], as well as alcoholism and dementia [65]. Lomustine has shown favorable results in the treatment of glioblastoma [66, 67]. Subsequently, we used molecular docking technique to deeply investigate the binding affinity between the mentioned above drugs and their DLGs. The results indicated that the above identified targeting drugs (latamoxef, parthenolide, meclofenoxate and lomustine) may be promising therapeutic drugs against DFU, but their therapeutic effects on DFU still need to be proved by numerous cellular and animal experiments.

Certainly, we must admit that there are limitations in the present study. First, our analyses were based only on samples from the training cohort, and although the reliability of the analyzed results was also validated in a dataset independent of the present study, they are still at an early stage and larger clinical samples are needed to corroborate them. Secondly, our work requires further validation by numerous in vivo and in vitro experiments, which will be the focus of our future research endeavors.

Conclusions

In summary, we identified DLGs for DFU using three machine learning algorithms and validated the diagnostic performance of these DLGs in an independent dataset. Meanwhile, we built a novel ANN model based on the transcriptomic patterns of DLGs that is promising for improving the clinical diagnosis of DFUs. Additionally, we investigated the correlation between the DLGs expression and the immune cell abundance in DFU, and demonstrated the heterogeneous expression of DLGs among different cells using single-cell analysis techniques. Finally, we identified promising therapeutic drugs against DFU by targeting DLGs using the Enrichr platform and molecular docking technology. These findings contribute to the understanding of the pathogenesis of DFU, help to improve the prognosis of DFU patients and reduce the disability rate, as well as provide new perspectives on the diagnosis and therapeutic strategies of DFU.

Data availability

Datasets related to this article are from public database (GSE165816, GSE134431, GSE80178, GSE7014 and GSE68183). All data generated or analyzed during this study are included in this article/Additional files, further included can be directed to the corresponding author.

Abbreviations

DFU:: Diabetic Foot Ulcers
ANN:: Artificial Neural Network
PCA:: Principal Component Analysis
DEGs:: Dysregulated Expressed Genes
TOM:: Topological Overlap Matrix
PPI:: Protein-Protein Interactions
GO:: Gene Ontology
CLU:: Clusterin
ENPEP:: Glutamyl Aminopeptidase
RABGEF1:: RAB Guanine Nucleotide Exchange Factor 1
DLGs:: Hub Lysosome-Related Genes of DFU
ROC:: Receiver Operating Characteristic
AUC:: Area Under Roc Curve
GSEA:: Gene Set Enrichment Analysis
ssGSEA:: single-sample Gene Set Enrichment Analysis
PCs:: Principal Components
SMCs:: Smooth Muscle Cells
Fibro:: Fibroblasts
VasEndo:: Vascular Endothelial Cells
DiffKera:: Differentiated Keratinocytes
BasalKera:: Basal Keratinocytes
NKT:: NK and T Cells
Macro:: Macrophages
Melano/Schwann:: Melanocytes and Schwann Cells
LymphEndo:: Lymphatic Endothelial Cells
B lymphos:: B lymphocytes
MCs:: Mast Cells
MDM:: Molecular Docking Method
SVM-RFE:: Support Vector Machine Recursive Feature Elimination
LASSO:: Least Absolute Shrinkage and Selection Operator
RF:: Random Forest
ABI:: Ankle-Brachial Index
NCV:: Nerve Conduction Velocity
AGE:: Advanced Glycation End-Product

References

Armstrong DG, Tan TW, Boulton AJM, Bus SA. Diabetic Foot Ulcers: a review. JAMA. 2023;330(1):62–75.
Article CAS PubMed Google Scholar
Walker AF, Graham S, Maple-Brown L, Egede LE, Campbell JA, Walker RJ, Wade AN, Mbanya JC, Long JA, Yajnik C, et al. Interventions to address global inequity in diabetes: international progress. Lancet. 2023;402(10397):250–64.
Article PubMed PubMed Central Google Scholar
Jones RE, Foster DS, Longaker MT. Management of chronic Wounds-2018. JAMA. 2018;320(14):1481–2.
Article PubMed Google Scholar
McDermott K, Fang M, Boulton AJM, Selvin E, Hicks CW. Etiology, epidemiology, and disparities in the Burden of Diabetic Foot Ulcers. Diabetes Care. 2023;46(1):209–21.
Article PubMed Google Scholar
Gordois A, Scuffham P, Shearer A, Oglesby A, Tobian JA. The health care costs of diabetic peripheral neuropathy in the US. Diabetes Care. 2003;26(6):1790–5.
Article PubMed Google Scholar
Sargen MR, Hoffstad O, Margolis DJ. Geographic variation in Medicare spending and mortality for diabetic patients with foot ulcers and amputations. J Diabetes Complications. 2013;27(2):128–33.
Article PubMed Google Scholar
Sen CK. Human wound and its Burden: updated 2022 Compendium of estimates. Adv Wound Care (New Rochelle). 2023;12(12):657–70.
Article PubMed Google Scholar
Zou J, Zhang W, Chen X, Su W, Yu D. Data mining reveal the association between diabetic foot ulcer and peripheral artery disease. Front Public Health. 2022;10:963426.
Article PubMed PubMed Central Google Scholar
Mishra SC, Chhatbar KC, Kashikar A, Mehndiratta A. Diabetic foot. BMJ. 2017;359:j5064.
Article PubMed PubMed Central Google Scholar
Sloan G, Selvarajah D, Tesfaye S. Pathogenesis, diagnosis and clinical management of diabetic sensorimotor peripheral neuropathy. Nat Rev Endocrinol. 2021;17(7):400–20.
Article PubMed Google Scholar
Everett E, Mathioudakis N. Update on management of diabetic foot ulcers. Ann N Y Acad Sci. 2018;1411(1):153–65.
Article PubMed PubMed Central Google Scholar
McIllhatton A, Lanting S, Lambkin D, Leigh L, Casey S, Chuter V. Reliability of recommended non-invasive chairside screening tests for diabetes-related peripheral neuropathy: a systematic review with meta-analyses. BMJ Open Diabetes Res Care 2021, 9(2).
Orlando G, Prior Y, Reeves ND, Vileikyte L. Patient and provider perspective of Smart Wearable Technology in Diabetic Foot Ulcer Prevention: a systematic review. Med (Kaunas) 2021, 57(12).
Wang X, Jiang G, Zong J, Lv D, Lu M, Qu X, Wang S. Revealing the novel ferroptosis-related therapeutic targets for diabetic foot ulcer based on the machine learning. Front Genet. 2022;13:944425.
Article CAS PubMed PubMed Central Google Scholar
Perera RM, Zoncu R. The lysosome as a Regulatory Hub. Annu Rev Cell Dev Biol. 2016;32:223–53.
Article CAS PubMed PubMed Central Google Scholar
Lamming DW, Bar-Peled L. Lysosome: the metabolic signaling hub. Traffic. 2019;20(1):27–38.
Article CAS PubMed Google Scholar
Gros F, Muller S. The role of lysosomes in metabolic and autoimmune diseases. Nat Rev Nephrol. 2023;19(6):366–83.
Article CAS PubMed Google Scholar
Greenhalgh DG. Wound healing and diabetes mellitus. Clin Plast Surg. 2003;30(1):37–45.
Article PubMed Google Scholar
Leek JT, Johnson WE, Parker HS, Jaffe AE, Storey JD. The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics. 2012;28(6):882–3.
Article CAS PubMed PubMed Central Google Scholar
Theocharidis G, Thomas BE, Sarkar D, Mumme HL, Pilcher WJR, Dwivedi B, Sandoval-Schaefer T, Sîrbulescu RF, Kafanas A, Mezghani I, et al. Single cell transcriptomic landscape of diabetic foot ulcers. Nat Commun. 2022;13(1):181.
Article CAS PubMed PubMed Central Google Scholar
Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559.
Article PubMed PubMed Central Google Scholar
Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607–13.
Article CAS PubMed Google Scholar
Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics. 2012;16(5):284–7.
Article CAS PubMed PubMed Central Google Scholar
Chen Y, Li Y, Narayan R, Subramanian A, **e X. Gene expression inference with deep learning. Bioinformatics. 2016;32(12):1832–9.
Article CAS PubMed PubMed Central Google Scholar
Noble WS. What is a support vector machine? Nat Biotechnol. 2006;24(12):1565–7.
Article CAS PubMed Google Scholar
Paul A, Mukherjee DP, Das P, Gangopadhyay A, Chintha AR, Kundu S. Improved Random Forest for classification. IEEE Trans Image Process. 2018;27(8):4012–24.
Article PubMed Google Scholar
Vasquez MM, Hu C, Roe DJ, Chen Z, Halonen M, Guerra S. Least absolute shrinkage and selection operator type methods for the identification of serum biomarkers of overweight and obesity: simulation and application. BMC Med Res Methodol. 2016;16(1):154.
Article PubMed PubMed Central Google Scholar
Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7.
Article PubMed PubMed Central Google Scholar
Satija R, Farrell JA, Gennert D, Schier AF, Regev A. Spatial reconstruction of single-cell gene expression data. Nat Biotechnol. 2015;33(5):495–502.
Article CAS PubMed PubMed Central Google Scholar
He H, Suryawanshi H, Morozov P, Gay-Mimbrera J, Del Duca E, Kim HJ, Kameyama N, Estrada Y, Der E, Krueger JG, et al. Single-cell transcriptome analysis of human skin identifies novel fibroblast subpopulation and enrichment of immune subsets in atopic dermatitis. J Allergy Clin Immunol. 2020;145(6):1615–28.
Article CAS PubMed Google Scholar
Lu Y, Liu X, Zhao J, Bie F, Liu Y, **e J, Wang P, Zhu J, **ong Y, Qin S, et al. Single-cell profiling reveals transcriptomic signatures of vascular endothelial cells in non-healing diabetic foot ulcers. Front Endocrinol (Lausanne). 2023;14:1275612.
Article PubMed Google Scholar
Reynolds G, Vegh P, Fletcher J, Poyner EFM, Stephenson E, Goh I, Botting RA, Huang N, Olabi B, Dubois A et al. Developmental cell programs are co-opted in inflammatory skin disease. Science 2021, 371(6527).
Tan JX, Finkel T. A phosphoinositide signalling pathway mediates rapid lysosomal repair. Nature. 2022;609(7928):815–21.
Article CAS PubMed PubMed Central Google Scholar
Ren H, Zhao F, Zhang Q, Huang X, Wang Z. Autophagy and skin wound healing. Burns Trauma. 2022;10:tkac003.
Article PubMed PubMed Central Google Scholar
Liu WJ, Gan Y, Huang WF, Wu HL, Zhang XQ, Zheng HJ, Liu HF. Lysosome restoration to activate podocyte autophagy: a new therapeutic strategy for diabetic kidney disease. Cell Death Dis. 2019;10(11):806.
Article PubMed PubMed Central Google Scholar
Wu Y, Quan Y, Liu Y, Liu K, Li H, Jiang Z, Zhang T, Lei H, Radek KA, Li D, et al. Hyperglycaemia inhibits REG3A expression to exacerbate TLR3-mediated skin inflammation in diabetes. Nat Commun. 2016;7:13393.
Article CAS PubMed PubMed Central Google Scholar
Shaikh-Kader A, Houreld NN, Rajendran NK, Abrahamse H. The link between advanced glycation end products and apoptosis in delayed wound healing. Cell Biochem Funct. 2019;37(6):432–42.
Article CAS PubMed Google Scholar
Guo Y, Lin C, Xu P, Wu S, Fu X, **a W, Yao M. AGEs Induced Autophagy impairs cutaneous Wound Healing via stimulating macrophage polarization to M1 in diabetes. Sci Rep. 2016;6:36416.
Article CAS PubMed PubMed Central Google Scholar
Li C, Qu L, Farragher C, Vella A, Zhou B. MicroRNA regulated macrophage activation in obesity. J Transl Int Med. 2019;7(2):46–52.
Article PubMed PubMed Central Google Scholar
Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC, Müller M. pROC: an open-source package for R and S + to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77.
Article PubMed PubMed Central Google Scholar
Zhou X, Guo Y, Yang K, Liu P, Wang J. The signaling pathways of traditional Chinese medicine in promoting diabetic wound healing. J Ethnopharmacol. 2022;282:114662.
Article CAS PubMed Google Scholar
Paananen J, Fortino V. An omics perspective on drug target discovery platforms. Brief Bioinform. 2020;21(6):1937–53.
Article CAS PubMed Google Scholar
Trajanoska K, Bhérer C, Taliun D, Zhou S, Richards JB, Mooser V. From target discovery to clinical drug development with human genetics. Nature. 2023;620(7975):737–45.
Article CAS PubMed Google Scholar
Li Y, Ju S, Li X, Li W, Zhou S, Wang G, Cai Y, Dong Z. Characterization of the microenvironment of diabetic foot ulcers and potential drug identification based on scRNA-seq. Front Endocrinol (Lausanne). 2022;13:997880.
Article PubMed Google Scholar
Li DL, Ding XY, Long J, He QL, Zeng QX, Lu N, Zou MC. Identification of autophagy-related genes in diabetic foot ulcer based on bioinformatic analysis. Int Wound J 2023.
Huang Y, Wang D, Zhang W, Yuan X, Li K, Zhang Y, Zeng M. Identification of hub genes and pathways associated with cellular senescence in diabetic foot ulcers via comprehensive transcriptome analysis. J Cell Mol Med. 2024;28(1):e18043.
Article CAS PubMed Google Scholar
Mariadoss AVA, Sivakumar AS, Lee CH, Kim SJ. Diabetes mellitus and diabetic foot ulcer: etiology, biochemical and molecular based treatment strategies via gene and nanotherapy. Biomed Pharmacother. 2022;151:113134.
Article CAS PubMed Google Scholar
Wang Z, Wei D, Li S, Tang Q, Lu G, Gu S, Lu L, Liang F, Teng J, Lin J, et al. Healing mechanism of diabetic foot ulcers using single-cell RNA-sequencing. Ann Transl Med. 2023;11(5):210.
Article CAS PubMed PubMed Central Google Scholar
Bonam SR, Wang F, Muller S. Lysosomes as a therapeutic target. Nat Rev Drug Discov. 2019;18(12):923–48.
Article CAS PubMed PubMed Central Google Scholar
Hlavca S, Chan WH, Engel RM, Abud HE. Clusterin: a marker and mediator of chemoresistance in colorectal cancer. Cancer Metastasis Rev 2024.
Wu H, Sun Y, Yang J, Gao Z, Shen H, Li M, Wang D, Tang Y. Iron deficiency downregulates ENPEP to promote angiogenesis in liver tumors. J Nutr Biochem. 2023;117:109357.
Article CAS PubMed Google Scholar
El Abbas S, Radermecker C, Bai Q, Beguin C, Schyns J, Meunier M, Pirottin D, Desmet CJ, Meuwis MA, Art T, et al. Epithelial RABGEF1 deficiency promotes intestinal inflammation by dysregulating intrinsic MYD88-dependent innate signaling. Mucosal Immunol. 2020;13(1):96–109.
Article PubMed Google Scholar
Mandair D, Reis-Filho JS, Ashworth A. Biological insights and novel biomarker discovery through deep learning approaches in breast cancer histopathology. NPJ Breast Cancer. 2023;9(1):21.
Article PubMed PubMed Central Google Scholar
Baxt WG. Application of artificial neural networks to clinical medicine. Lancet. 1995;346(8983):1135–8.
Article CAS PubMed Google Scholar
Kufel J, Bargieł-Łączek K, Kocot S, Koźlik M, Bartnikowska W, Janik M, Czogalik Ł, Dudek P, Magiera M, Lis A et al. What Is Machine Learning, Artificial Neural Networks and Deep Learning?-Examples of Practical Applications in Medicine. Diagnostics (Basel) 2023, 13(15).
Voelker R. What are Diabetic Foot Ulcers? JAMA. 2023;330(23):2314.
Article PubMed Google Scholar
Shi H, Yuan X, Yang X, Huang R, Fan W, Liu G. A novel diabetic foot ulcer diagnostic model: identification and analysis of genes related to glutamine metabolism and immune infiltration. BMC Genomics. 2024;25(1):125.
Article CAS PubMed PubMed Central Google Scholar
Wen Q, Liu D, Wang X, Zhang Y, Fang S, Qiu X, Chen Q. A systematic review of ozone therapy for treating chronically refractory wounds and ulcers. Int Wound J. 2022;19(4):853–70.
Article PubMed Google Scholar
Jovic D, Liang X, Zeng H, Lin L, Xu F, Luo Y. Single-cell RNA sequencing technologies and applications: a brief overview. Clin Transl Med. 2022;12(3):e694.
Article CAS PubMed PubMed Central Google Scholar
Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease. Nature. 2013;496(7446):445–55.
Article CAS PubMed PubMed Central Google Scholar
Bechara N, Gunton JE, Flood V, Hng TM, McGloin C. Associations between Nutrients and Foot Ulceration in Diabetes: A Systematic Review. Nutrients 2021, 13(8).
Oturai PS, Holländer NH, Hansen OP, Boas J, Bruun BG, Frimodt-Møller N, Dombernowsky P, Hansen HH. Ceftriaxone versus latamoxef in febrile neutropenic patients: empirical monotherapy in patients with solid tumours. Eur J Cancer. 1993;29a(9):1274–9.
Article CAS PubMed Google Scholar
Lapcik P, Sulc P, Janacova L, Jilkova K, Potesil D, Bouchalova P, Müller P, Bouchal P. Desmocollin-1 is associated with pro-metastatic phenotype of luminal A breast cancer cells and is modulated by parthenolide. Cell Mol Biol Lett. 2023;28(1):68.
Article CAS PubMed PubMed Central Google Scholar
Izumi K, Tominaga H, Koja T, Nomoto M, Shimizu T, Sonoda H, Imamura K, Igata A, Fukuda T. Meclofenoxate therapy in tardive dyskinesia: a preliminary report. Biol Psychiatry. 1986;21(2):151–60.
Article CAS PubMed Google Scholar
Cohen PA, Avula B, Khan I. The unapproved drug centrophenoxine (meclofenoxate) in cognitive enhancement dietary supplements. Clin Toxicol (Phila). 2022;60(10):1156–8.
Article CAS PubMed Google Scholar
Killock D. Lomustine-temozolomide combination efficacious in newly diagnosed glioblastoma. Nat Rev Clin Oncol. 2019;16(5):273.
PubMed Google Scholar
Weller M, Le Rhun E. How did lomustine become standard of care in recurrent glioblastoma? Cancer Treat Rev. 2020;87:102029.
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank the platforms and resources provided by Tian** Beichen Hospital. Also, we thank the GEO database (https://www.ncbi.nlm.nih.gov/geo/) for the open access to the data.

Funding

This work was supported by Tian** Beichen District Healthcare System Science and Technology Project (SHGY-2022008).

Author information

Authors and Affiliations

Department of Dermatology, Tian** Beichen Hospital, Tian**, 300400, China
Longhai Tan & Junxia Wang
Zhu **anyi Memorial Hospital of Tian** Medical University, Tian**, 300134, China
Junjun Qu

Authors

Longhai Tan
View author publications
You can also search for this author in PubMed Google Scholar
Junjun Qu
View author publications
You can also search for this author in PubMed Google Scholar
Junxia Wang
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

L.T. designed the project. J.Q. organized the project under supervision of L.T. L.T., J.Q., and J.W. prepared and performed all bioinformatic analysis. L.T. drafted the manuscript with input from J.Q., and J.W. All authors reviewed the manuscript.

Corresponding author

Correspondence to Longhai Tan.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article

Tan, L., Qu, J. & Wang, J. Development of novel lysosome-related signatures and their potential target drugs based on bulk RNA-seq and scRNA-seq for diabetic foot ulcers. Hum Genomics 18, 62 (2024). https://doi.org/10.1186/s40246-024-00629-1

Download citation

Received: 23 February 2024
Accepted: 27 May 2024
Published: 11 June 2024
DOI: https://doi.org/10.1186/s40246-024-00629-1

Development of novel lysosome-related signatures and their potential target drugs based on bulk RNA-seq and scRNA-seq for diabetic foot ulcers

Abstract

Background

Results

Conclusions

Similar content being viewed by others

Introduction

Methods

Data downloading and processing

Differential analysis of gene expression

Weighted correlation network analysis

Protein-protein interaction analysis

Functional enrichment analysis

Selection of hub lysosome-related genes of DFU

scRNA-seq analysis

Identification and docking potential drugs that target DLGs

Discussion

Conclusions

Data availability

Abbreviations

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher’s Note

Electronic supplementary material

Rights and permissions

About this article

Cite this article

Share this article

Keywords

Search

Navigation