In Silico Characterization of Inflammatory and Anti-Inflammatory Modulation in Diabetic Nephropathy: The Construction of a Genetic Panel

Abstract

Diabetic Nephropathy (DN) stands as a primary cause of end-stage renal disease and its etiology remains unclear. Thus, this study aims to construct a genetic panel with potential biomarkers linked to the inflammatory pathway of DN associated with the pathology’s susceptibility. Through a systematic review and meta-analysis, we selected observational studies in English, Portuguese, and Spanish, selected from the PubMed, SCOPUS, Virtual Health Library, Web of Science, and EMBASE databases. Additionally, a protein–protein interaction network was constructed to list hub genes, with differential expression analysis by microarray of kidneys with DN from the GSE30529 database to further refine results. Seventy-two articles were included, and 54 polymorphisms in 37 genes were associated with the inflammatory pathway of DN. Meta-analysis indicated a higher risk of complication associated with SNPs 59029 G/A, −511 C/T, VNTR 86 bp, −308 G/A, and −1031 T/C. Bioinformatics analyses identified differentially expressed hub genes, underscoring the scarcity of studies on CCL2 and VEGF-A genes in relation to DN. This study highlighted the intrinsic relationship between inflammatory activity in the etiology and progression of DN, enabling the effective application of precision medicine in diabetic patients for potential prognosis of the complications and contributing to cost reduction in the public health system.

1. Introduction

Diabetic Nephropathy (DN), also known as diabetic kidney disease, is considered the leading cause of end-stage renal failure. However, its etiology has not been fully elucidated [1,2]. Characterized by progressive and irreversible loss of renal function due to chronic hyperglycemia, DN results in glomerular lesions leading to a decline in glomerular filtration rate (GFR) and albuminuria [3].

Chronic hyperglycemia increases the expression of inflammatory mediators due to the injury of tubular and glomerular cells, contributing to renal damage through mesangial proliferation, podocytic/tubular damage, and leukocyte infiltration [4,5]. Hemodynamic changes found in DN, which are associated with hyperglycemia, cause glomerular infiltration, mechanical stress, and endothelial activation. Endothelial cells upregulate the expression of adhesion molecules and chemokines involved in leukocyte infiltration into the renal interstitium [6].

Chemokines enable the activation of integrins by leukocytes, resulting in the adhesion of inflammatory cells to the endothelium, followed by transmigration into various tissues [5,7]. After infiltration into inflammatory foci, leukocytes promote kidney damage through interaction and activation of glomerular and tubular cells generated by the release of chemokines, cytokines, and profibrotic factors, activating stromal renal cells that release additional chemokines and promote increased leukocyte infiltration. These mechanisms amplify the inflammatory response in a positive feedback loop that exacerbates renal damage [4,5].

Currently, the diagnosis of DN is conducted through clinical assessment and laboratory tests, such as evaluation of urinary albumin/creatinine ratio (altered values > 30 mg/g) and estimated GFR (altered values < 60 mL/min/1.73 m²). Furthermore, biopsy is the gold standard for DN diagnosis [8]. Being a multifactorial disease, not all factors that cause the disease are known. Studies indicate that DN is influenced by environmental factors such as sedentary lifestyle, hypertension, and chronic hyperglycemia [9]. However, despite lifestyle changes and glycemic control measures reducing the proportion of diabetic patients progressing to DN, these approaches are insufficient to prevent the risk of developing the disease. Epidemiological studies suggest that genetic factors play a significant role in susceptibility to DN [10].

Therefore, the objective of this study was to identify genetic polymorphisms related to the inflammatory pathway of DN through a systematic review (SR) and meta-analysis for the construction of a genetic panel. Additionally, since the identified genes operate in the same pathway and do not act individually, through in silico analysis, we aimed to identify protein–protein interaction (PPI) to classify hub genes and determine if these genes are differentially expressed genes (DEGs) using the GSE30529 dataset. The integration of these data will assist in identifying potential genetic biomarkers related to the pathogenesis and susceptibility of this silent, progressive, and irreversible disease.

2. Materials and Methods

2.1. Systematic Review

To identify the key genetic polymorphisms associated with the inflammatory process in DN, a systematic review (SR) was conducted guided by the research question: “What are the main genetic polymorphisms related to the inflammatory process in the risk of developing Diabetic Nephropathy?” The review was structured according to the PEO acronym (Population, Exposure, Outcome) (

Table 1. PEO acronym with respective descriptions.

Acronym	Description
P (Population)	Individuals with diabetes mellitus (types 1 and 2) who developed Diabetic Nephropathy as a complication.
E (Exposure	Presence of genetic polymorphisms associated with inflammation in the population.
O (Outcome)	Outcome of interest related to the development of Diabetic Nephropathy.

).

To avoid duplications of this SR, this study had its protocol registered in the International Prospective Register of Systematic Reviews (PROSPERO) on 17 February 2021 (registration number CRD42021232347). For a more comprehensive data description, we adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

The literature search took place between November 2020 and April 2021 across the PubMed, SCOPUS, Virtual Health Library (Biblioteca Virtual em Saúde—BVS), Web of Science, and EMBASE databases. Studies included in the SR were observational studies assessing humans, conducted in English, Portuguese, or Spanish, with no time restrictions. Combined terms cataloged in the Medical Subject Headings (MeSH) for ‘Diabetic Nephropathy’, ‘Genetic Polymorphism’, and ‘Inflammation’ were employed to formulate the search strategy. The search strategy was adapted for all databases (

Table 2. Search strategies developed for systematic review.

Databases	Search Strategy
Pubmed	(((‘Diabetic Nephropathies’) OR (Nephropathies, Diabetic) OR (Nephropathy, Diabetic) OR (‘Diabetic Nephropathy’) OR (‘Diabetic Kidney Disease’) OR (‘Diabetic Kidney Diseases’) OR (Kidney Disease *, Diabetic) OR (‘Diabetic Glomerulosclerosis’) OR (Glomerulosclerosis, Diabetic) OR (‘Intracapillary Glomerulosclerosis’) OR (‘Nodular Glomerulosclerosis’) OR (Glomerulosclerosis, Nodular) OR (‘Kimmelstiel-Wilson Syndrome’) OR (‘Kimmelstiel Wilson Syndrome’) OR (Syndrome, Kimmelstiel-Wilson) OR (‘Kimmelstiel-Wilson Disease’) OR (‘Kimmelstiel Wilson Disease’)) AND ((Polymorphism *, Genetic) OR (‘Genetic Polymorphism’) OR (‘Genetic Polymorphisms’) OR (‘Gene Polymorphism’) OR (‘Gene Polymorphisms’) OR (Polymorphism *, Gene) OR (Polymorphism * (Genetics)) OR (‘Genetic Susceptibility’) OR (‘Genetic Susceptibilities’) OR (Susceptibilit *, Genetic) OR (‘Genetic Predisposition’) OR (‘Genetic Predispositions’) OR (Predisposition *, Genetic)) AND ((Inflammation *) OR (‘Innate Inflammatory Response’) OR (‘Innate Inflammatory Responses’) OR (Inflammatory Response, Innate) OR (‘Inflammation Mediators’) OR (Mediators, Inflammation) OR (‘Mediators of Inflammation’)))
SCOPUS	(TITLE-ABS-KEY (“Diabetic Nephropath *”) OR TITLE-ABS-KEY (Nephropath *, Diabetic) OR TITLE-ABS-KEY (“Diabetic Kidney Disease *”) OR TITLE-ABS-KEY (Kidney Disease *, Diabetic) OR TITLE-ABS-KEY (“Diabetic Glomerulosclerosis”) OR TITLE-ABS-KEY (Glomerulosclerosis, Diabetic) OR-TITLE-ABS-KEY (“Intracapillary Glomerulosclerosis”) OR TITLE-ABS-KEY (“Nodular Glomerulosclerosis”) OR TITLE-ABS-KEY (Glomerulosclerosis, Nodular) OR TITLE-ABS-KEY (“Kimelstiel$Wilson Syndrome”) OR TITLE-ABS-KEY (Syndrome, Kimmelstiel-Wilson) OR TITLE-ABS-KEY (“Kimmelstiel$Wilson Disease”) AND TITLE-ABS-KEY (Polymorphism *, Genetic) OR TITLE-ABS-KEY (“Genetic Polymorphism *”) OR TITLE-ABS-KEY (“Gene Polymorphism *”) OR TITLE-ABS-KEY (“Polymorphism, Gene”) OR TITLE-ABS-KEY (“Polymorphism (Genetics)”) OR TITLE-ABS-KEY (“Genetic Susceptibilit *”) OR TITLE-ABS-KEY (Susceptibilit *, Genetic) OR TITLE-ABS-KEY (“Genetic Predisposition *”) OR TITLE-ABS-KEY (“Predisposition, Genetic”) AND TITLE-ABS-KEY (Inflammat *) OR TITLE-ABS-KEY (“Innate Inflammatory Response”) OR TITLE-ABS-KEY (Inflammatory Response, Innate) OR TITLE-ABS-KEY (“Inflammation Mediators”) OR TITLE-ABS-KEY (Mediators, Inflammation) OR TITLE-ABS-KEY (“Mediators of Inflammation”))
BVS	(‘Diabetic Nephropathies’) OR (Nephropath *, Diabetic) OR (‘Diabetic Nephropathy’) OR (‘Diabetic Kidney Disease’) OR (‘Diabetic Kidney Diseases’) OR (Kidney Disease *, Diabetic) OR (‘Diabetic Glomerulosclerosis’) OR (Glomerulosclerosis, Diabetic) OR (‘Intracapillary Glomerulosclerosis’) OR (‘Nodular Glomerulosclerosis’) OR (Glomerulosclerosis, Nodular) OR (‘Kimmelstiel-Wilson Syndrome’) OR (‘Kimmelstiel Wilson Syndrome’) OR (Syndrome, Kimmelstiel-Wilson) OR (‘Kimmelstiel- Wilson Disease’) OR (‘Kimmelstiel Wilson Disease’) AND (Polymorphism *, Genetic) OR (‘Genetic Polymorphism’) OR (‘Genetic Polymorphisms’) OR (‘Gene Polymorphism’) OR (‘Gene Polymorphisms’) OR (Polymorphism *, Gene) OR (Polymorphism * (Genetics)) OR (‘Genetic Susceptibility’) OR (‘Genetic Susceptibilities’) OR (Susceptibilit *, Genetic) OR (‘Genetic Predisposition’) OR (‘Genetic Predispositions’) OR (Predisposition *, Genetic) AND (Inflammat *) OR (‘Innate Inflammatory Response’) OR (Inflammatory Response, Innate) OR (‘Innate Inflammatory Responses’) OR (‘Inflammation Mediators’) OR (Mediators, Inflammation) OR (‘Mediators of Inflammation’)
Web of Science	TS = (((“Diabetic Nephropathies”) OR (Nephropathies, Diabetic) OR (Nephropathy, Diabetic) OR (“Diabetic Nephropathy”) OR (“Diabetic Kidney Disease$”) OR (Kidney Disease$, Diabetic) OR (“Diabetic Glomerulosclerosis”) OR (Glomerulosclerosis, Diabetic) OR (“Intracapillary Glomerulosclerosis”) OR (“Nodular Glomerulosclerosis”) OR (Glomerulosclerosis, Nodular) OR (“Kimmelstiel-Wilson Syndrome”) OR (“Kimmelstiel Wilson Syndrome”) OR (Syndrome, Kimmelstiel-Wilson) OR (“Kimmelstiel-Wilson Disease”) OR (“Kimmelstiel Wilson Disease”)) AND ((Polymorphism *, Genetic) OR (“Genetic Polymorphism$”) OR (“Gene Polymorphism$”) OR (Polymorphism$, Gene) OR (Polymorphism$ (Genetics)) OR (“Genetic Susceptibilit *”) OR (Susceptibilit *, Genetic) OR (“Genetic Predisposition$”) OR (Predisposition$, Genetic)) AND ((Inflammat *) OR (“Innate Inflammatory Response$”) OR (Inflammatory Response, Innate) OR (“Inflammation Mediators”) OR (Mediators, Inflammation) OR (“Mediators of Inflammation”)))
EMBASE	(‘diabetic glomerulopathy’ OR ‘diabetic glomerulosclerosis’ OR ‘diabetic intercapillary glomerulosclerosis’ OR ‘diabetic kidney disease’ OR ‘diabetic nephropathies’ OR ‘glomerulonecrosis, intercapillary’ OR ‘glomerulosclerosis, diabetic’ OR ‘glomerulosclerosis, intercapillary’ OR ‘intercapillary glomerulosclerosis’ OR ‘diabetic nephropathy’ OR ‘kimmelstiel wilson nephropathy’ OR ‘kimmelstiel wilson syndrome’ OR ‘nephropathy, diabetic’ OR ‘kidney disease’ OR ‘diabetic complication’) AND (‘genetic polymorphism’ OR ‘polymorphism (genetics)’ OR ‘polymorphism, genetic’ OR ‘genetic susceptibility’ OR ‘genetic predisposition’) AND (inflammation OR ‘acute inflammation’ OR ‘inflammation reaction’ OR ‘inflammation response’ OR ‘inflammatory condition’ OR ‘inflammatory lesion’ OR ‘inflammatory process’ OR ‘inflammatory reaction’ OR ‘inflammatory response’ OR ‘reaction, inflammation’ OR ‘response, inflammatory’ OR ‘innate immunity’ OR ‘autacoid’ OR ‘inflammation mediators’)

This term (*) promotes an amplification to results of the search. $ promotes an amplification to the search to association, for example disease and diseases.

).

The exclusion criteria were as follows: articles with a different study design, articles in languages other than English, Portuguese, or Spanish, those that did not evaluate DN, articles without a significant association and not addressing genetic polymorphisms, studies involving animals, and those focusing on genes not participating in the inflammatory pathway.

The articles identified in the search were imported into the Rayyan platform [11]. Screening was conducted by two independent reviewers (LS and KFS) and divided into two phases (I and II), where articles were evaluated based on pre-established inclusion and exclusion criteria. In Phase I, titles and abstracts were reviewed, selecting only articles addressing polymorphisms in genes related to the inflammatory pathway in DN. Phase II involved a full reading of the articles selected in Phase I. Bias risk was assessed using The Joanna Briggs Institute’s (JBI’s) critical appraisal tool, and applied for each study design. Only studies that positively answered at least 70% of the questions were considered to have a low risk of bias. Discrepancies between reviewers were discussed and resolved through consensus.

For the construction of the genetic panel, the following data were extracted: (1) authors and year of publication; (2) study design; (3) population; (4) sample size; (5) type of diabetes mellitus (DM); (6) gene and its location; (7) polymorphism(s) assessed; (8) genotypic and allelic frequencies; (9) comparison performed; (10) chi-square value; (11) odds ratio (OR)—95% confidence interval (CI); and (12) p-value. Data extraction were performed independently by the two reviewers for all included studies.

2.2. Meta-Analysis

The association of polymorphisms with DN was estimated by calculating the odds ratio (OR) with a 95% Confidence Interval (CI), using the dominant genetic model (wild type vs. heterozygote + mutant). Heterogeneity among studies was assessed using Higgins’ inconsistency test (I²). The selection of the meta-analytic model was based on the result of the heterogeneity test among studies. Therefore, the fixed-effect model (Mantel–Haenszel method) was applied when I² < 25% (low heterogeneity), assuming that discrepancies between effect estimates are merely due to chance. The random-effects model (DerSimonian–Laird method) was applied when I² > 25% (moderate or high heterogeneity).

Publication bias was assessed by a funnel plot [12], and its asymmetry was estimated using linear regression (Egger’s test) [13]. A p-value < 0.05 suggests a high likelihood of publication bias. All statistical analyses were performed using RStudio^® software (version 4.1.0).

2.3. Protein–Protein Interaction and Network Analysis

The list of genes from the constructed genetic panel was exported to the Search Tool for the Retrieval of Interacting Genes (STRING) database for the construction of the protein network. Subsequently, Cytoscape software (version 3.6.1, http://www.cytoscape.org/, accessed on 22 November 2023) was employed to visualize the protein–protein interaction (PPI) network, and the identification of the top 20 hub genes was performed using the CytoHubba tool.

2.4. Microarray Data and Data Processing

The microarray expression dataset used was GSE30529, and was obtained from the National Center for Biotechnology Information-Gene Expression Omnibus (NCBI-GEO) public database (https://www.ncbi.nlm.nih.gov/geo/, 13 December 2023). This dataset was registered on the GPL57 platform ([HG-U133A_2] Affymetrix Human Genome U133A 2.0 Array) and provides the gene expression profile of 10 samples from kidneys with DN and 12 control samples.

To identify differentially expressed genes (DEGs), two approaches were employed. The first involved a parametric statistical analysis with p-value correction using the False Discovery Rate (FDR) method proposed by Benjamini and Hochberg, with a significance level of 5%. Subsequently, a fold change of 1.2 was applied to identify proteins that were biologically regulated (up-regulated, down-regulated, and not significant). All analyses were performed using the GEO2R tool (https://www.ncbi.nlm.nih.gov/geo/info/geo2r.html, 13 December 2023) along with the GEOquery and Limma libraries of R software version 4.2.1.

3. Results

3.1. Systematic Review and Meta-Analysis

Following the initial search, 1782 articles were identified in the databases, with 387 duplicates excluded, resulting in 1395 articles for Phase I of the SR. Based on the review of titles and abstracts, all articles addressing polymorphisms in genes related to the inflammatory pathway in DN were selected. Of the initial pool, 1298 articles were excluded according to pre-established exclusion criteria. Phase II involved a thorough reading of the 97 selected articles, resulting in the exclusion of 25 articles and the inclusion of 72 in the SR (Figure 1), published between 1996 and 2020.

3.2. Methodological Quality Analysis

The 72 studies demonstrated consistency in quality assessment, each achieving a minimum of 70% positive responses on the critical appraisal tool of the Joanna Briggs Institute (JBI) (Supplementary File S1) and thus classified as low risk of bias. In case-control studies (Figure 2A), the majority responded positively to most questions, except for question number 6, as expected, given its evaluation of the presence of confounding factors. Regarding cohort studies (Figure 2B), questions 6, 8, 9, and 10 were deemed not applicable. Most articles appropriately addressed the remaining questions, except for questions 4 and 5, which focused on confounding factors and strategies to address them. In cross-sectional studies (Figure 2C), despite being in the minority, they responded positively to the questionnaire.

3.3. Individual Results

Among the included articles, 61 were case-control studies, 9 were cohort studies, and 2 were cross-sectional studies. From the data extraction, 54 polymorphisms in 37 genes associated with the inflammatory pathway of DN were identified (

Table 3. Genes and polymorphisms associated with the inflammatory pathway of DN.

Gene	Gene Location	Gene Function in the Inflammatory Pathway	Evaluated Polymorphism	Type of Polymorphism	References
ADIPOQ	3q27.3	It encodes an anti-inflammatory adipocytokine that antagonizes TNF-alpha by negatively regulating its expression in various tissues and organs and neutralizing its effects. It prevents NF-kappa-B endothelial signaling through a cAMP-dependent pathway.	rs266729 (−11377 C/G)	SNP	[14]
			rs17300539 (−11391 G/A)		[15]
			rs2241766 (45 T/G)
ALOXA5AP	13q12.3	It encodes a protein, that when associated with 5-lipoxygenase, performs the synthesis of leukotrienes (LTs). Leukotrienes are a family of lipid mediators that act as pro-inflammatory mediators.	rs3803278 (8733 T/C)	SNP	[16]
CCL2 (MCP-1)	17q12	Encodes a chemokine member of the CC subfamily involved in immunoregulatory and inflammatory processes. Displays chemotactic activity for monocytes and basophils, but not for neutrophils or eosinophils.	rs3917887 (int1del554-567)	Indel	[17]
			rs1024611 (−2518 A/G)	SNP	[18,19]
CCR5	3p21.31	Chemokine CC receptors (CCRs) predominantly recognize inflammatory CC chemokines, such as CCL3, CCL4, and RANTES. They may play a role in controlling the proliferation or differentiation of the granulocytic lineage and participate in the migration of T lymphocytes to the site of infection, acting as a chemotactic receptor.	rs1799987 (59029 G/A)	SNP	[17,20,21,22,23,24,25]
			rs333 (Delta 32)	Indel	[17,21]
Hp	16q22.2	Haptoglobin (Hp) captures hemoglobin and combines it with free plasma, enabling hepatic recycling of heme iron and preventing renal damage. It also acts as an antioxidant, antibacterial agent, and plays a role in modulating the acute-phase inflammatory response.	Hp1/Hp2	Codominant alleles	[26,27]
ICAM1	19p13.2	Encodes a cell surface glycoprotein that is particularly expressed in endothelial cells and the immune system. It binds to integrins of the CD11a/CD18 or CD11b/CD18 type.	rs5498 (K469E; 1462 A/G)	SNP	[28]
IL1α	2q14.1	Encodes a member of the interleukin-1 cytokine family. It is pleiotropic and related to various immune responses, inflammatory processes, and hematopoiesis. This cytokine is produced by monocytes and macrophages as a pro-protein, which is proteolytically processed and released in response to cellular injury, leading to apoptosis.	rs1800587 (−889 C/T)	SNP	[29]
IL1β	2q14.1	Encodes a member of the interleukin-1 cytokine family. This cytokine is produced by activated macrophages as a pro-protein, which is proteolytically processed to its active form by Caspase 1. It is an important mediator of the inflammatory response and is related to various cellular activities, including cell proliferation, differentiation, and apoptosis.	rs16944 (−511 C/T)	SNP	[30,31,32,33,34,35,36]
IL1RN	2q14.1	Encodes a member of the interleukin-1 cytokine family. This protein inhibits the activities of IL-1α and IL-1β and modulates various immune and inflammatory responses related to interleukin-1, particularly during the acute phase of infection and inflammation.	(VNTR 86 bp)	Variable number tandem repeat (VNTR)	[30,31,37]
IL4	5q31.1	Encodes a pleiotropic cytokine produced by activated T cells. IL-4, a type 2 cytokine, is considered important for tissue repair, counteracting the effects of pro-inflammatory type 1 cytokines. Additionally, it intervenes and regulates various responses in the human host, such as acute inflammation.	rs2243250 (−590 C/T)	SNP	[38,39]
IL6	7p15.3	Encodes a cytokine that acts on inflammation and the maturation of B cells. This cytokine is primarily produced in sites of acute and chronic inflammation, where it is secreted into the serum and induces a transcriptional inflammatory response through the interleukin-6 alpha receptor.	rs1800796 (−634 C/G)	SNP	[40,41,42]
			rs1800795 (−174 G/C)		[43]
			rs2069837 (A/G)		[41]
			rs1524107 (T/C)
			(−176 G/C)		[39]
IL8 (CXCL8)	4q13.3	Encodes a protein that is a member of the CXC chemokine family; one of the main mediators of the inflammatory response. IL-8 acts as a chemotactic factor, directing neutrophils to the site of infection. It also participates, along with other cytokines, in the pro-inflammatory signaling cascade.	rs4073 (−251 T/A)	SNP	[17]
			rs2227306 (+781 C/T)		[44]
IL10	1q32.1	Encodes a cytokine primarily produced by monocytes and to a lesser extent by lymphocytes. This cytokine has pleiotropic effects on immunoregulation and inflammation, and negatively regulates the expression of Th1 cytokines, MHC class II molecules, etc. It can block NF-kappa B activity and is related to the regulation of the JAK-STAT signaling pathway.	rs1800896 (−1082 G/A)	SNP	[45,46,47,48,49]
			rs1800872 (−592 C/A)		[50]
IL34	16q22.1	Encodes a cytokine that promotes the differentiation and viability of monocytes and macrophages through the colony-stimulating factor 1 receptor (CSF1R).	rs6499323 (G/A)	SNP	[51]
LTA	6p21.33	Encodes a protein member of the tumor necrosis factor family; a cytokine produced by lymphocytes. This protein also intervenes in various inflammatory, immune-stimulatory, and antiviral responses.	rs1041981 (Thr26Asn; 804 C/A)	SNP	[52]
			rs909253 (Ala252Gly; A/G)
MMP9	20q13.12	Matrix metalloproteinases (MMPs) act on pro-inflammatory mediators, regulating various aspects of inflammation, and can function as a switch in acute and chronic inflammation. Therefore, MMP9 is considered a pro-inflammatory cytokine.	rs17576 (Arg 279Gln, G/A)	SNP	[17,53]
MPO	17q22	Myeloperoxidase (MPO) is a heme protein produced during myeloid differentiation, constituting an essential component of the azurophilic granules in neutrophils. This enzyme provides essential hypohalous acids for the microbicidal activity of neutrophils.	rs2333227 (−463 G/A)	SNP	[54,55]
NFKB1	4q24	NF-κB is a transcription regulator activated by various intra- and extra-cellular stimuli, such as cytokines, free radicals, etc. When activated, it translocates to the nucleus and stimulates the expression of genes involved in various biological functions, such as cell growth and immune cell development.	rs28362491 (−94 ATTG)	Indel	[56]
NOS2	17q11.2	Encodes the protein Nitric Oxide Synthase 2, which increases the synthesis of pro-inflammatory mediators, such as IL6 and IL8.	rs2779248 (T/C)	SNP	[57]
			rs1137933 (G/A)
OPN (SPP1)	4q22.1	Encodes a cytokine involved in increasing the production of interferon-gamma and interleukin-12 and reducing interleukin-10. This cytokine is essential in the pathway leading to type I immunity.	rs11730582 (−443 C/T)	SNP	[58]
PPARγ	3p25.2	Encodes a member of the nuclear receptor subfamily of the peroxisome proliferator-activated receptor (PPAR). PPAR-γ can inhibit the expression of pro-inflammatory genes, suppressing responses mediated by NF-kappa-B.	rs1801282 (Pro12Ala; C/G)	SNP	[59]
pri-miR-125a	19q13.41	Encodes a miRNA considered a potential regulator of IL-6R. Li; Lei (2015) report that the expression level of the IL-6R protein was significantly decreased by the introduction of miR-125a mimetics.	rs12976445 (C/T)	SNP	[60]
PRKCB1	16p12.2-p12.1	Protein kinase Cs (PKCs) are a family of serine/threonine-specific protein kinases that plays an important role in B-cell activation by regulating B-cell receptor (BCR)-mediated NF-kappa-B activation.	rs3760106 (−1504 C/T)	SNP	[61]
			rs2575390 (−546 C/G)
PTX3	3q25.32	The expression of the protein produced by this gene is induced by inflammatory cytokines in response to inflammatory stimuli in various types of mesenchymal and epithelial cells. It promotes fibrocyte differentiation and is involved in the regulation of inflammation and complement activation.	rs2305619 (281 A/G)	SNP	[62,63]
RAGE (AGER)	14q32.31	Encodes a protein considered an intracellular signal transducer or pro-inflammatory peptide. It acts as a mediator of acute and chronic vascular inflammation, regulating the production/expression of TNF-alpha, oxidative stress, and endothelial dysfunction in type 2 diabetes.	rs1800624 (−374 T/A)	SNP	[64]
			rs3134940 (2184 A/G)		[65]
RANTES (CCL5)	17q12	Produces a protein chemotactic for blood monocytes, memory helper T cells, and eosinophils. Additionally, it can activate various chemokine receptors, including CCR1, CCR3, CCR4, and CCR5.	rs2280788 (−28 C/G)	SNP	[20]
SASH1	6q24.3-q25.1	Encodes a scaffold protein related to the TLR4 signaling pathway that can induce cytokine production and migration of endothelial cells in response to invading pathogens.	rs6930576 (G/A)	SNP	[66]
SEPS1 (SELENOS)	15q26.3	Encodes a transmembrane protein located in the endoplasmic reticulum. It is involved in the process of degrading misfolded proteins and may play a role in inflammation control, acting as an anti-inflammatory and antioxidant protein.	rs4975814 (G/T)	SNP	[67]
SLC12A3	16q13	Encodes an electroneutral sodium and chloride ion cotransporter. It is a receptor for the pro-inflammatory cytokine IL18 and contributes to IL18-induced cytokine production, including IFNG, IL6, IL18, and CCL2.	rs11643718 (Arg913Gln; 78 G/A)	SNP	[68]
SUMO4	6q25.1	Encodes small ubiquitin-related modifiers. The protein encoded by this gene particularly modifies IKBA, causing negative regulation of NF-kappa-B-dependent transcription of the IL12B gene.	rs237025 (M55V; c.163 G/A)	SNP	[69]
TCRBC (TRB)	7q34	Encodes T cell receptors that recognize processed foreign antigens as small peptides bound to major histocompatibility complex molecules on the surface of antigen-presenting cells.	(9.2;10.0 kb)	Not identified	[70]
TGFβ1	19q13.2	Encodes a secreted ligand of the TGF-beta protein superfamily. This protein regulates cell proliferation, differentiation, and growth, and may modulate the expression and activation of other growth factors, such as interferon-gamma and tumor necrosis factor-alpha.	rs1800470 (869 T/C)	SNP	[71,72,73,74,75,76]
			rs1800471 (915 G/C)		[74]
TLR4	9q33.1	Encodes a protein member of the Toll-like receptor (TLR) family that plays a fundamental role in pathogen recognition and activation of innate immunity. They recognize molecular patterns associated with pathogens and are involved in the production of cytokines necessary for effective immunity.	rs5030718 (14367 G/A)	SNP	[77]
TNFα	6p21.33	Encodes a multifunctional pro-inflammatory cytokine belonging to the tumor necrosis factor superfamily. This cytokine is primarily secreted by macrophages and is involved in the regulation of various biological processes, such as cell proliferation, differentiation, apoptosis, lipid metabolism, and coagulation.	rs1799964 (−1031 T/C)	SNP	[34,78,79]
			rs1800629 (−308 G/A)		[34,80,81]
			rs361525 (−238 G/A)		[34]
TSC22 (TSC22D1)	13q14.11	Encodes a leucine zipper protein expressed in many tissues and involved in the signaling of TGF-1.	(−396 A/G)	SNP	[73]
UMOD	16p12.3	Encodes a protein that can bind to immunoglobulin G, complement 1q, and tumor necrosis factor-alpha (TNF-α), signaling a role in innate immunity. It can also act as a receptor for the binding and endocytosis of cytokines (IL-1 and IL-2) and TNF.	rs4293393 (T/C)	SNP	[82]
VEGF-A	6p21.1	Encodes a protein that acts as a pro-inflammatory cytokine, increasing the permeability of endothelial cells and inducing the expression of endothelial cell adhesion molecules through its ability to act as a chemotactic agent for monocytes.	rs35569394 (−2549 D/I)	Indel	[83]

).

The genes and their respective polymorphisms most frequently mentioned in the systematic review were: CCR5 (59029 G/A), IL1B (−511 C/T), TGF-1B (869 T/C), IL10 (−1082 G/A), IL1RN (VNTR of 86 bp), and TNF (−308 G/A and −1031 T/C). The 59029 G/A polymorphism of the CCR5 gene was assessed in 8 articles in this systematic review. The meta-analysis results indicated an association between this SNP and the risk of DN (OR = 1.88; 95% CI = 1.23–2.88; and p = 0.0035) (Figure 3).

Seven studies were included in the meta-analysis that observed the SNP −511 C/T in the IL1B gene, and the results indicated an association of this SNP with the risk of developing DN (OR = 1.55; 95% CI = 1.12–2.16; and p = 0.0081) (Figure 4).

In the TGF-1B gene, the SNP 869 T/C was identified in six articles. The meta-analysis showed no association with the risk of developing DN (OR = 2.14; 95% CI = 0.95–4.82; and p = 0.0669) (Figure 5). Additionally, the SNP −1082 G/A in the IL-10 gene also showed no association with the risk of developing DN based on the meta-analysis (OR = 1.51; 95% CI = 0.79–2.87; and p = 0.2150) (Figure 6).

We identified in the IL-1RN gene a variable number tandem repeat (VNTR) polymorphism of 86 bp, which was significantly associated with the risk of developing DN (OR = 2.42; 95% CI = 1.03–5.65; and p = 0.0420) (Figure 7). Two polymorphisms in the TNF gene showed a statistically significant risk in the development of DN; namely, −308 G/A (OR = 1.98; 95% CI = 1.33–2.96; and p = 0.0008) (Figure 8) and −1031 T/C (OR = 2.73; 95% CI = 1.76–4.21; and p = < 0.0001) (Figure 9).

The publication bias analysis for the SNPs 59029 G/A, −511 C/T, 869 T/C, −1082 G/A, VNTR of 86 bp, −308 G/A, and −1031 T/C did not show publication bias according to the funnel plot (Figure 10) and Egger’s test (p = 0.8884, p = 0.8938, p = 0.0784, p = 0.7868, p = 0.6703, p = 0.8336, and p = 0.5576, respectively).

3.4. Protein Network Analysis and Identification of Hub Genes

The selected genes participate in the same pathway and do not act in isolation; therefore, we constructed a protein–protein interaction (PPI) network to identify the interactions between the proteins encoded by the selected genes. This network included 34 nodes and 261 edges, where each node represents a protein and each edge represents an interaction between proteins (Figure 11). The PPI enrichment p-value was <1.0 × 10⁻¹⁶, indicating that the protein network had significantly more interactions than expected and is biologically connected.

The analysis of hub genes revealed the top 20 genes with the highest interaction in the inflammatory pathway of DN. These genes are as follows: TNF, IL1B, IL6, CCL2, IL10, ICAM1, CXCL8, MMP9, TLR4, VEGFA, IL4, IL1A, CCL5, MPO, ADIPOQ, TGFB1, LTA, SPP1, IL1RN, and NOS2 (Figure 12 and

Table 4. Twenty genes with the highest number of interactions according to CytoHubba.

Ranking	Gene	Full Gene Name	Score
1	TNF	Tumor Necrosis Factor	54.0
2	IL1B	Interleukin 1 Beta	52.0
3	IL6	Interleukin 6	50.0
3	IL10	Interleukin 10	50.0
6	ICAM1	Intercellular Adhesion Molecule 1	48.0
6	CXCL8 (IL8)	Chemokine (C-X-C motif) ligand 8	48.0
6	MMP9	Matrix Metallopeptidase 9	48.0
6	TLR4	Toll-like receptor 4	48.0
6	VEGFA	Vascular Endothelial Growth Factor A	48.0
11	IL4	Interleukin 4	46.0
12	IL1A	Interleukin 1 Alpha	44.0
13	CCL5	Chemokine (C-C motif) ligand 5	42.0
14	MPO	Myeloperoxidase	38.0
14	ADIPOQ	Adiponectin, C1Q and Collagen Domain Containing	38.0
14	TGFB1	Transforming Growth Factor Beta 1	38.0
17	LTA	Lymphotoxin Alpha	36.0
18	SPP1	Secreted Phosphoprotein 1	34.0
18	IL1RN	Interleukin 1 Receptor Antagonist	34.0
920	NOS2	Nitric Oxide Synthase 2	32.0

).

3.5. Differential Expression Analysis—GSE30529

For the analysis of transcript data, 22,277 genes from the database GSE30529, available at the National Center for Biotechnology Information-Gene Expression Omnibus (NCBI-GEO) public database (https://www.ncbi.nlm.nih.gov/geo/, 13 December 2023), were considered. To screen for the genes with a statistical difference (p-value ≤ 0.05), a hypothesis test was performed considering the FDR, in which 4184 genes were found. Subsequently, a fold-change of 1.2 was considered to identify the DEG. Sixty-two genes were found to be down-regulated, 454 genes were up-regulated, and 3668 genes were not significant (Figure 13) (Supplementary File S2). Among the DEG (down- and up-regulated), three genes were previously identified in the systematic review of the present study: CCL5 (RANTES), CCL2 (MCP-1), and CXCL8 (IL8).

4. Discussion

Despite DN not being classified as an inflammatory disease, numerous studies suggest that molecules within the inflammatory pathway may play significant pathophysiological roles in this complication [84]. Therefore, the primary objective of this study was to identify potential molecular markers operating within the inflammatory pathway of DN. To evaluate the genes associated with the pathogenesis and susceptibility of the disease, we constructed a PPI network to identify hub genes. The analysis of microarray data confirmed that these genes exhibit differential regulation in patients with DN compared to controls. This observation underscores the potential involvement of inflammatory mechanisms in the complex pathophysiology of the disease.

In DN, proteinuria induces renal endothelial injury, triggering the synthesis and production of molecules involved in inflammatory pathways, including the expression of cytokines, chemokines, and adhesion molecules [85]. Cytokines are signaling molecules primarily active in inflammatory processes. These molecules categorize into various classes such as interleukins, tumor necrosis factors, and transforming growth factors (TGFs) [86].

Interleukins are divided into families, and the IL-1 subgroup comprises 11 members: IL-1α, IL-1β, IL-1RA, IL-18, IL-36Ra, IL-36α, IL-37, IL-36β, IL-36γ, IL-38, and IL-33. In this study, three interleukins from this family associated with pathophysiological mechanisms of DN were identified (IL-1α, IL-1β, and IL-1RA). Through meta-analysis, we identified polymorphisms in the IL1β and IL1RA genes potentially associated with increased susceptibility in the development of the complication.

IL-1β is a pro-inflammatory cytokine produced by activated macrophages, playing various roles in biological and pathological cellular functions such as proliferation, differentiation, and apoptosis. IL-1β induces the synthesis of prostaglandins, influx and activation of neutrophils, activation of T cells with cytokine production, and activation of B cells. In addition to promoting the differentiation of pro-inflammatory Th17 cells, it also plays a role in angiogenesis by inducing the production of VEGF-A synergistically with the synthesis of TNF and IL-6 (36; 31). The SNP rs16944 (−511 C/T) identified in this gene is located in the promoter region and involves a cytosine-to-thymine substitution at position −511. This polymorphism has been associated with increased gene transcription and, consequently, an exaggerated synthesis of IL-1β. Therefore, this cytokine polymorphism enhances its pro-inflammatory reactions [31,36].

According to the meta-analysis, patients carrying the T allele exhibited a high susceptibility to ND when compared to homozygotes for the C allele (OR = 1.55; 95% CI = 1.12–1.16; and p = 0.0081). This finding is consistent with studies conducted with Korean [30], Indian [31,34], German [33], Polish [35], and Iraqi [36] patients, demonstrating that individuals with type 2 diabetes mellitus (DM2) carrying this polymorphism have a higher likelihood of developing the complication. The overexpression of IL-1β can induce the proliferation of renal mesangial cells and the formation of extracellular matrix, contributing to renal damage [33,35].

The gene for interleukin-1 receptor antagonist (IL-1RA) encodes the IL-1Ra protein, which acts by inhibiting the activities of IL-1. Thus, it is an anti-inflammatory cytokine that determines the functionality of interleukins IL-1α and IL-1β (37). In our study, the variable number tandem repeat (VNTR) polymorphism corresponding to 86 base pairs in intron 2 was found to be positively associated with susceptibility to ND (OR = 2.42; 95% CI = 1.03–5.65; and p = 0.0420). Our findings align with the studies conducted by Lee et al. [30] and Achyut et al. [31], where the polymorphism was significantly associated with the complication in Indian and Korean patients. According to these studies, the polymorphism reduces the synthesis of this cytokine, and consequently, the amount of IL-1β is altered due to the modulatory activity of IL-1RA on the pro-inflammatory cytokine.

Interleukin-10 (IL-10) is an anti-inflammatory cytokine produced by B lymphocytes, Th2 cells, monocytes, and macrophages. Its anti-inflammatory activity is capable of inhibiting the synthesis of pro-inflammatory cytokines such as IL-1β, IL-1α, IL-6, tumor necrosis factor-alpha (TNF-α), and interferon-gamma (INF-γ) [87]. IL-10 is crucial for the maintenance of renal physiology and is associated with the development of renal lesions, given its predominant anatomical localization in renal mesangial cells. IL-10 promotes the deposition of mesangial immune complexes, contributing to the progression of glomerular injury in ND. The SNP −1082 G/A in the IL10 gene did not show an association with ND susceptibility. Scientific evidence suggests that this polymorphism may attenuate the anti-inflammatory protective effect mediated by the cytokine, thereby intensifying the pro-inflammatory process [88,89].

Another identified anti-inflammatory cytokine is TGF-β1, a growth factor that regulates cell proliferation, differentiation, and growth, while also playing a role in the activation and modulation of the expression of other growth factors, including INF-γ and TNF-α. Additionally, it can promote the differentiation of lineage-specific T helper 17 (Th17) cells or regulatory T cells (Treg). Low concentrations of TGF-β1, in conjunction with cytokines such as IL-6 and IL-21, induce the expression of IL-17 and IL-23 receptors, favoring the differentiation of Th17 cells. In contrast, high concentrations of the factor promote the negative regulation of IL-17 expression, favoring the development of Treg cells [76]. Our study identified two SNPs in the TGF-β1 gene: rs1800470 and rs1800471. Due to the insufficient quantity of studies for SNP rs1800471 (915 G/C), it was not possible to summarize the risk estimate through statistical analysis. However, the meta-analysis conducted for SNP rs1800470 (869 T/C) (OR = 2.14; 95% CI = 0.95–4.82; and p = 0.0669) did not demonstrate an association between the polymorphism and the risk of developing DN. Studies suggest that these polymorphisms can enhance the expression of TGF-β1 in mesangial and endothelial cells of glomeruli, favoring renal hypertrophy and the accumulation of extracellular matrix, which are factors that contribute to the development and progression of the complication [76,90].

TNF-α is a pro-inflammatory cytokine that plays a role in the regulation of cell differentiation, proliferation, and death and is involved in both innate and adaptive immune responses. Studies have demonstrated that the hyperglycemic state in patients with diabetes mellitus induces the production of TNF-α through the negative regulation of CD33 in monocytes. In renal cells, TNF-α is cytotoxic, and its elevation can trigger DN [78]. We identified three SNPs in the TNF-α gene: rs1800629 (−308 G/A), rs1799964 (−1031 T/C), and rs361525 (−238 G/A). It was only possible to conduct a meta-analysis with the SNP −308 G/A, and our statistical analysis showed that patients with the A allele have a higher susceptibility to DN (OR = 1.98; 95% CI = 1.33–2.96; and p = 0.0008). Studies report elevated levels of TNF-α in diabetic patients due to increased gene transcription [34,80,81].

The analysis of the −1031 T/C polymorphism in the TNF-α gene revealed that carriers of the C allele also had a higher risk of developing DN (OR = 2.73; 95% CI = 1.76–4.21; and p = 0.0001). Studies demonstrate that this polymorphism in the promoter region is associated with elevated plasma levels of TNF-α [78].

The most frequently mentioned gene in the literature was CCR5, a receptor for various chemokines such as CCL3, MIP-1-α, CCL4, and MIP-1-β, which stimulates the chemotaxis of T lymphocytes. Studies report that CCR5 is involved in the recruitment of monocytes and their differentiation into macrophages in the glomeruli, playing a significant role in developing glomerulosclerosis and fibrosis in DN [20,21,22,23].

In our meta-analysis, the SNP rs1799987 (59029 G/A), located in the promoter region of the CCR5 gene, showed a higher susceptibility to the development of DN (OR = 1.88; 95% CI = 1.23–2.88; and p = 0.0035). This involves a guanine-to-adenine alteration at position 59029. Studies found report that the presence of the A allele, in homozygosity or heterozygosity, is associated with the development of DN in North American [21] and Japanese DM1 patients [20,22], as well as in Indian [17,23,25] and Polish DM2 patients (OR ranging from 1.39 to 6.17) [24]. Additionally, an increase in the expression of the protein was observed in peripheral blood mononuclear cells of individuals carrying the A allele, suggesting that the polymorphism regulates the expression of the CCR5 gene (17,22–25].

Considering our objective, the pathway enrichment analysis was not performed, since transcriptome analysis was performed to validate the results previously found in the study. Thus, our in silico analyses enabled the identification of three key genes (CCL2, CCL5, and CXCL8) concomitantly found in our systematic review. These genes constitute the core, linking with other genes, and exhibit a higher differential expression in the kidneys of individuals with DN than controls (GSE30529 database) and encode pro-inflammatory cytokines. Therefore, we emphasize these two genes as potential targets that could guide further research and contribute to elucidating the pathogenesis of DN, particularly in association with inflammatory and hemodynamic processes.

The CCL2 gene encodes the homonymous protein, also known as monocyte chemoattractant protein 1 (MCP-1), and is considered one of the key molecules promoting inflammation in DN. Polymorphisms in this gene are associated with an increase in its protein expression and subsequent progressive renal failure. Inflammation in the kidney is predominantly marked by the accumulation of macrophages, which is directly related to the progression of diabetes due to increased hyperglycemia and glycated hemoglobin, favoring renal damage and albuminuria [91].

The macrophages recruited to the renal microenvironment of diabetic patients can promote renal fibrosis. These macrophages stimulate the secretion of IL-1 and TGF-β1. IL-1 can trigger an increase in the pro-inflammatory response, acting as a trigger for other cytokines. The increased expression of TGF-β1, in turn, can induce the production of extracellular matrix, contributing to renal fibrosis. Thus, CCL2/MCP-1 is associated with DN and underlying pathological mechanisms both directly and indirectly [92]. This cytokine is also capable of inducing a fibrotic response in mesangial glomerular cells. This mechanism is mediated by the activation of the NF-Κβ transcription factor in these cells, stimulating the accumulation of macrophages in this region. It is suggested that TGF-β1 stimulates the expression of CCL2/MCP-1, reinforcing the indications of a self-amplifying loop of these cytokines, and promoting the accumulation of extracellular matrix in mesangial cells [93].

CCL5 acts as a potent molecule in signalization for several immune cells, such as monocytes and T cells and its up regulation shows similar effects as CCL2. Sample biopsies of DN patients revealed overexpression of CCL5 in tubular cells, which has a direct association with the magnitude of proteinuria and cell infiltration. Moreover, serum levels of DN patients demonstrated elevated serum levels of this cytokine [94]. Urinary sediment messenger RNAs (mRNAs) of the CCL5 gene have been used as a potential biomarker of kidney disease, especially in DN, due to their significant up-expression [95].

It is well known that several factors involved in the pathogenesis of DM can lead to the activation and accumulation of immune cells in different tissues in patients with diabetes, such as hyperglycemia and the increase of reactive molecules such as [94]. The role of the CXCL8 in DN remains unclear; however, it was previously associated with podocyte damage [96]. Moreover, elevated levels of CXCL8 have been found in the early stages of DN and the level increases with the disease progression and is associated with metabolic disturbances and inflammation. Thus, urinary CXCL8 shows a strong potential as an early predictor of DN [97].

Additionally, our study demonstrates that most SNPs significantly associated with a higher susceptibility to developing DN in our meta-analysis are located in genes encoding pro-inflammatory cytokines. In this context, SNPs in genes for anti-inflammatory cytokines like TGF-β and IL-10 did not pose a risk for DN. A mechanism possibly associated with this is the relationship of the complication with sustained inflammation. Thus, through the obtained data and bioinformatics analyses, we observe an imbalance between the production of anti-inflammatory and pro-inflammatory cytokines, with the latter showing higher expression.

Our study has some limitations. Firstly, it is important to highlight that DM and its complications are complex pathologies involving the interaction of genetic and environmental factors. Some discrepancies found in the meta-analysis results may be explained by differences in sample size, individual genetic background, low statistical power, or challenges in the reproducibility of genetic studies. Additionally, it should be considered that for some SNPs, the meta-analysis was conducted with a considerably low number of studies, which may explain the found heterogeneity [98]. Furthermore, the unavailability of articles and the lack of data in potentially eligible studies led to the exclusion of some articles from our systematic review. Finally, there is a need for studies with larger sample sizes.

Nevertheless, our study identified important biomarkers for DN susceptibility. These biomarkers can assist in clinical practice in the disease. The costs of diagnosing and staging the disease are expensive for health systems [99] and the incorporation of a genetic panel can be a tool in the differential diagnosis and genetic screening of individuals susceptible to complications. In addition, the identification of genetic variants related to the risk of developing DN supports the clinical application of precision medicine, providing personalized health measures to the patient and improving the clinical success [100,101].

5. Conclusions

In conclusion, our study indicated a higher risk of DN development associated with SNPs 59029 G/A, −511 C/T, VNTR 86 bp, −308 G/A, and −1031 T/C. Moreover, the bioinformatics analyses associate the differential expression of CCL2, CCR5 and CXCL8 genes with the etiology and progression of DN due to its intrinsic relationship with inflammatory mechanisms. This study also facilitated the identification of several polymorphisms frequently associated with a higher susceptibility to the development of DN in individuals with diabetes mellitus. These polymorphisms were primarily identified in molecules of a pro-inflammatory nature, and the association of genetic variants with risk factors such as glycemic control, obesity, sedentary lifestyle, hypertension, alcohol consumption, and smoking may increase the risk for the complication. Additionally, bioinformatics analyses enabled the elucidation of differentially expressed hub genes, validating the results found and possibly confirming the central role of these genes in DN. This allows the application of our findings in experimental analyses, given the scarcity of studies on the genes CCL2 and VEGF-A identified in our systematic review, which encode cytokines with potential key roles in DN and its underlying inflammatory mechanisms. Furthermore, the results of this genetic panel allow for the effective application of precision medicine in diabetic patients for a possible prognosis of the complication, contributing to the reduction of expenses for the public health system. Further studies should be suggested at the end of the manuscript for other researchers to follow up and confirm the findings. It is important to emphasize that more studies are needed to confirm and validate our findings aiming at the practical and clinical application of the identified biomarkers.