Next Article in Journal
Serum Chemokines CCL3 and CCL7 as Complementary Diagnostic Biomarkers Across Tumor Grades in Clear Cell Renal Cell Carcinoma
Previous Article in Journal
Biophysical Modeling of Lipopolysaccharides in Gram-Negative Bacteria: From Atomic to Colloidal Models
Previous Article in Special Issue
Renal and Vascular Effects of the Allosteric Transglutaminase 2 Modulator LDN-27219 in One-Kidney DOCA–Salt Mice
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 27
Issue 5
10.3390/ijms27052491
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Identification of Common Genes Regulated by ER Stress During the Development of Diabetic Nephropathy Based on Human Transcriptome Datasets and an In Vivo Mouse Model

by
Jacques Karekezi
1,2,
Ashimwe Yves Roger
1,2,
Harry Jang
3,
Jong-Won Kim
1,2,
Seung Pil Yun
1,2,
Hye Jung Kim
1,2,
Ji Miao
3,
Sang Won Park
1,2,* and
Hwajin Kim
1,*
1
Department of Pharmacology, Institute of Medical Sciences, College of Medicine, Gyeongsang National University, Jinju 52727, Republic of Korea
2
Department of Convergence Medical Sciences, Gyeongsang National University Graduate School, Jinju 52727, Republic of Korea
3
Division of Endocrinology, Boston Children’s Hospital, Boston, MA 02115, USA
*
Authors to whom correspondence should be addressed.
Int. J. Mol. Sci. 2026, 27(5), 2491; https://doi.org/10.3390/ijms27052491
Submission received: 17 February 2026 / Revised: 4 March 2026 / Accepted: 7 March 2026 / Published: 8 March 2026
(This article belongs to the Special Issue The Latest Molecular Research on Renal Disease)
Browse Figures
Versions Notes

Abstract

Diabetic nephropathy (DN) is a serious complication in diabetic patients, leading to kidney dysfunction and ultimately end-stage renal disease. Although several pharmacological agents have been developed, treating DN remains challenging due to its complex and multifaceted pathogenesis. Endoplasmic reticulum (ER) stress plays a crucial role in DN pathology; however, the molecular mechanisms underlying reduced ER stress remain poorly understood. This study investigated the protective effects of 4-phenylbutyrate (4-PBA), an ER stress inhibitor, on DN and the related regulatory molecules through gene expression network analysis. A C57BL/6 mouse model of DN was used in combination with a high-fat diet and streptozotocin after unilateral nephrectomy and treated with 4-PBA by intraperitoneal injection for 6 weeks. The 4-PBA treatment effectively improves DN-induced renal structural and functional abnormalities by reducing albuminuria, podocyte loss, glomerular and tubular injury, and renal inflammation and cell death. These changes induced by 4-PBA were associated with decreased expression of ER stress markers and increased autophagy activities in diabetic kidneys. Importantly, 4-PBA reduced components of the complement C1q pathway, the NADPH oxidase complex, and chemokines, thereby attenuating chronic renal dysfunction. Conclusively, inhibition of ER stress is a promising pharmacological target for treating patients with DN.

1. Introduction

Diabetic nephropathy (DN) is one of the most common microvascular complications of type 1 and type 2 diabetes and the leading cause of chronic kidney disease (CKD) and end-stage renal disease (ESRD) [1,2]. DN is characterized by a decreased glomerular filtration rate (GFR), albuminuria, and elevated plasma creatinine. In the early stages of DN, podocyte loss, glomerular hypertrophy, expansion of the glomerular mesangial matrix, and thickening of the glomerular basement membrane are observed. In advanced stages, nodular glomerulosclerosis, glomerular mesangial lysis, severe inflammation, and tubulointerstitial fibrosis are observed [3,4]. Currently, intensive glycemic and blood pressure control or renin–angiotensin–aldosterone system (RAAS) inhibition are available treatments for patients with DN; however, many patients still progress to CKD, resulting in ESRD [5]. Therefore, it is urgent to understand the precise pathologic mechanisms of DN and develop new treatments that alleviate or restore kidney function.
Accumulating evidence has demonstrated that ER stress plays an essential role in the development and pathogenesis of DN. Hyperglycemia, proteinuria, advanced glycation end products (AGEs), and free fatty acids (FFA) have been shown to induce ER stress, and the downstream signaling pathways of unfolded protein response (UPR) and ER-associated degradation (ERAD) are triggered to restore protein homeostasis. However, prolonged ER stress activation results in apoptotic cell death and ultimately leads to structural and functional damage of podocytes and renal tubules [6,7]. Proximal tubular cells, in particular, are highly sensitive to ER stress due to their high protein synthesis and transport demands. Previous studies have shown that ER stress plays a crucial role in hyperglycemia-induced lipid accumulation [8], palmitate-induced apoptosis [9], and albuminuria-induced inflammasome activation [10] in renal proximal tubular cells. Therefore, persistent ER stress contributes to tubular apoptosis, inflammation, and fibrotic remodeling, thereby accelerating the DN progression [11,12].
Chemical chaperones of 4-phenylbutyrate (4-PBA) and taurine-conjugated ursodeoxycholic acid (TUDCA) have been shown to alleviate ER stress, and treatment of obese and diabetic mice with these compounds restores glucose levels and systemic insulin sensitivity [13]. These ER stress inhibitors also attenuate the pathology of DN; recent studies reported that 4-PBA blocks oxidative stress and inflammatory responses, thereby alleviating renal pathology in diabetic animal models [14,15]. Current studies are on identifying key regulatory molecules associated with ER stress in a renal cell type- and DN disease-specific manner to effectively treat patients with DN.
Network biology has emerged to integrate complex gene- and protein-interaction networks and to understand the molecular mechanisms of disease, enabling diagnosis and treatment [16]. We conducted bioinformatic analysis of bulk RNA-seq datasets relevant to human diabetic kidney disease (DKD) to identify upregulated genes, and we assessed the effects of 4-PBA in DN mouse models. Importantly, 4-PBA reduced components of the complement C1q pathway, the NADPH oxidase complex, and chemokines, thereby attenuating chronic renal dysfunction. In this study, we demonstrated that 4-PBA enhanced autophagy and reduced inflammation and oxidative stress, thereby ameliorating glomerular and tubular injury and interstitial fibrosis during the progression of DN.

2. Results

2.1. 4-PBA Improved Glomerular Filtration Function and Reduced Tubular Injury in DN

To determine the effect of 4-PBA on the renal pathology of DN, we evaluated the parameters of renal injury in control and DN mice administered with vehicle or 4-PBA (100 mg/kg) for 6 weeks (Figure 1A). DN mice treated with 4-PBA markedly reduced kidney weight and levels of plasma BUN and creatinine (Figure 1B–D). Treatment with 4-PBA also reduced urine volume, the albumin-to-creatinine ratio (UACR), and urinary NAC activity (Figure 1E–G), compared to vehicle treatment. However, body weight and levels of plasma blood glucose had no statistical differences in DN mice treated with 4-PBA, compared to DN mice treated with the vehicle (Supplementary Figure S1). Then, the expression of KIM-1 (kidney injury molecule-1) and NGAL (neutrophil gelatinase-associated lipocalin), well-known tubular injury markers [17,18], was evaluated by real-time PCR analysis. The mRNA levels of KIM-1 and NGAL were significantly upregulated in DN mice compared to controls and were reduced by 4-PBA treatment (Figure 1H,I). The results suggest that 4-PBA treatment improves glomerular filtration function and reduces tubular injury in DN.

2.2. 4-PBA Reduced Podocyte Loss and Tubular Interstitial Fibrosis in DN

We assessed the extent of histopathological changes by Periodic Acid-Schiff (PAS) staining (Figure 2A,B). DN mice showed different degrees of hyalinization, glomerular atrophy, reduced Bowman’s capsule space, and chronic glomerulonephritis; however, these glomerular injuries were significantly reduced by 4-PBA treatment (Figure 2A). The immunohistological staining of WT-1 and the mRNA levels of Nphs1 and 2 were also restored by 4-PBA treatment in DN mice (Figure 2C,E), which suggests that 4-PBA protects from podocyte loss. DN mice also showed a significant increase in tubulo-interstitial damage, including tubular dilatation, loss of brush borders, necrosis, and extracellular matrix accumulation, which was reduced by 4-PBA treatment (Figure 2B). The Picro-Sirius Red staining to assess collagen deposition and the mRNA levels of fibrogenic mediators were upregulated in DN mice, and 4-PBA treatment reduced the collagen and expression levels (Figure 2D,E), which suggests that 4-PBA protects from renal interstitial fibrosis.

2.3. 4-PBA Reduced ER Stress and Increased Autophagy in DN

The ER stress serves as a cellular adaptive response, but sustained ER stress impairs autophagy, leading to renal inflammatory and apoptotic responses in the progression of DN; thus, reducing ER stress facilitates restoration of defective autophagy and ameliorates its pathogenesis [19,20,21]. We first confirmed that 4-PBA treatment decreased the expression of ER stress markers (ATF6, CHOP, phosphorylated IRE1) in DN mice (Figure 3A). We then determined the effect of 4-PBA on autophagy by measuring the expression levels of autophagy markers (LC3B, p62, and ATG5). 4-PBA significantly increased ATG5 levels and improved autophagy flux as shown by the increased LC3B-II and decreased p62 levels in DN mice (Figure 3B). Interestingly, 4-PBA also increased the expression of PGC-1α, a cofactor of mitochondrial biogenesis (Figure 3B), which suggests that inhibition of ER stress may restore mitochondrial turnover and cellular energy metabolism [22].

2.4. 4-PBA Reduced Renal Inflammation, Apoptosis, and Oxidative Stress in DN

To determine the effect of 4-PBA on renal inflammation, we measured macrophage infiltration and cytokine expression in DN mice. Treatment with 4-PBA reduced the percentage of stained CD68-positive cells in kidney sections (Figure 4A). The expression of proinflammatory cytokines, TNF-α and IL-6, was decreased, while anti-inflammatory IL-10 was increased by 4-PBA (Figure 4B). Then, the renal apoptosis was assessed by cleavage of caspase-3 and PARP1, and 4-PBA treatment significantly reduced these apoptotic markers in DN mice (Figure 4C). Further, oxidative stress, assessed by 4-hydroxynonenal (4-HNE), a marker of lipid peroxidation, was significantly upregulated in DN mice and was significantly reduced by 4-PBA treatment (Figure 4C).

2.5. Protective Effect of 4-PBA by Restoring the Expression of C1q, Ncf4, and Ccl28 in DN Mice

To identify pathogenic pathways and regulatory genes relevant to human diabetic kidney disease (DKD), we performed an integrative RNA-seq analysis of whole kidney datasets GSE166239 [23] and GSE142025 [24]. Gene Ontology enrichment highlighted 12 top processes implicated in DKD progression, including extracellular matrix organization, inflammation, fibrosis, complement activation, and chemokine signaling (Figure 5). Cross-dataset intersection of differentially expressed genes (DEGs) yielded 76 genes consistently upregulated in human DKD, which we categorized by function (Figure 6A–C).
To place these findings at the mechanistic levels, we mapped our results onto the WikiPathways Oxidative Damage Response network (Figure 6D and Figure S2). In this pathway, reactive oxygen species (ROS) initiate mitochondrial stress and propagate signals that bifurcate into pro-survival and pro-apoptotic programs, including antioxidant gene induction (often via NF-κB) and caspase 3-mediated apoptosis. Complement activation and downstream inflammatory responses are regulated under this oxidative injury framework. We also validated the 76 candidate genes to our mouse model of DN through RT-PCR and confirmed that 20 genes were elevated in DN kidneys, and a subset was significantly suppressed by 4-PBA treatment (Figure 7). Notably, these included C1q (classical complement) [25,26], neutrophil cytosolic factor 4 (Ncf4; a component of the NOX2 NADPH oxidase complex) [27], and C-C motif chemokine ligand 28 (Ccl28; a stress-responsive, fibrogenic chemokine) [28,29]. These changes indicate engagement of complement signaling, oxidative stress generation via NOX2, and chemokine-driven inflammatory/fibrotic responses in DN progression, which are attenuated by ER stress inhibition.
Together, the computational and pathway-mapping results converge on a model in which oxidative stress (via NOX2/Ncf4) stimulates complement initiation (C1q), which in turn, amplifies inflammation and interfaces with apoptosis. Consistently, Ccl28 links oxidative stress and fibrogenic signaling. The ER stress inhibitor 4-PBA reverses expression of these regulatory genes (C1q, Ncf4, and Ccl28) (Figure 7).
Ijms 27 02491 g005
Figure 5. Human dataset analysis. (A) Gene prioritization workflow integrating differential gene expression (DGE) and pathway enrichment analysis (PEA). This schematic illustrates the computational workflow used to prioritize candidate genes from two transcriptomic datasets (GSE142025 and GSE166239). Each dataset undergoes DGE analysis to generate DGE tables, followed by gene-level comparison between datasets. Parallelly, gene set enrichment analysis (GSEA) is performed using the Reactome/MSigDB pathway database to generate PEA tables. Normalized pathway enrichment results from both datasets are then compared at the pathway level. Together, these gene- and pathway-level comparisons enable the identification and prioritization of genes and pathways that are consistently perturbed across datasets. (B) Comparison of DGE between GSE142025 and GSE166239. Scatter plot showing the relationship of log2 fold-change (log2FC) values obtained from DGE analyses of these 2 datasets. Each point represents a single gene. Red dots indicate genes that are significantly upregulated (FDR < 0.05) in both datasets. Blue dots represent genes that are significantly upregulated only in GSE142025, while green dots represent genes significantly upregulated only in GSE166239. The plot highlights shared and dataset-specific transcriptional responses across the two studies. (C) PEA comparison by Reactome; pathway enrichment overlap and normalized enrichment score comparison between GSE142025and GSE166239. ((Left) panel) Venn diagram showing the number of significantly enriched pathways identified in each dataset using GSEA based on DGE log2 fold changes and the Reactome pathway database. An FDR < 0.05 threshold was applied to define significantly enriched pathways. A total of 113 commonly enriched pathways were identified as upregulated in both datasets. ((Right) panel) A scatter plot comparing normalized enrichment scores (NES) for pathways enriched in GSE142025 and GSE166239. Each point represents a Reactome pathway. Red dots denote pathways that are significantly upregulated (FDR < 0.05) in both datasets, while purple dots highlight pathways that are significantly downregulated in both datasets. Green dot is a pathway that is significantly upregulated in GSE166239 but downregulated in GSE142025. This visualization illustrates shared and divergent pathway-level perturbations across the two transcriptomic datasets.
Figure 5. Human dataset analysis. (A) Gene prioritization workflow integrating differential gene expression (DGE) and pathway enrichment analysis (PEA). This schematic illustrates the computational workflow used to prioritize candidate genes from two transcriptomic datasets (GSE142025 and GSE166239). Each dataset undergoes DGE analysis to generate DGE tables, followed by gene-level comparison between datasets. Parallelly, gene set enrichment analysis (GSEA) is performed using the Reactome/MSigDB pathway database to generate PEA tables. Normalized pathway enrichment results from both datasets are then compared at the pathway level. Together, these gene- and pathway-level comparisons enable the identification and prioritization of genes and pathways that are consistently perturbed across datasets. (B) Comparison of DGE between GSE142025 and GSE166239. Scatter plot showing the relationship of log2 fold-change (log2FC) values obtained from DGE analyses of these 2 datasets. Each point represents a single gene. Red dots indicate genes that are significantly upregulated (FDR < 0.05) in both datasets. Blue dots represent genes that are significantly upregulated only in GSE142025, while green dots represent genes significantly upregulated only in GSE166239. The plot highlights shared and dataset-specific transcriptional responses across the two studies. (C) PEA comparison by Reactome; pathway enrichment overlap and normalized enrichment score comparison between GSE142025and GSE166239. ((Left) panel) Venn diagram showing the number of significantly enriched pathways identified in each dataset using GSEA based on DGE log2 fold changes and the Reactome pathway database. An FDR < 0.05 threshold was applied to define significantly enriched pathways. A total of 113 commonly enriched pathways were identified as upregulated in both datasets. ((Right) panel) A scatter plot comparing normalized enrichment scores (NES) for pathways enriched in GSE142025 and GSE166239. Each point represents a Reactome pathway. Red dots denote pathways that are significantly upregulated (FDR < 0.05) in both datasets, while purple dots highlight pathways that are significantly downregulated in both datasets. Green dot is a pathway that is significantly upregulated in GSE166239 but downregulated in GSE142025. This visualization illustrates shared and divergent pathway-level perturbations across the two transcriptomic datasets.
Ijms 27 02491 g005
Ijms 27 02491 g006
Figure 6. Human gene network and pathway analysis. (A) The pathways and (B) list of genes that are highly expressed in DKD patients; common DEGs were analyzed from GSE166239 and GSE142025, and 76 common DEGs were selected and categorized. (C) Tissue-specific gene regulatory network inferred from the genotype-tissue expression (GTEx) in the human kidney cortex samples. A gene regulatory network was constructed using GTEx Kidney Cortex v9 transcriptomic data and the ARACNe algorithm [30,31], applying a curated list of regulators derived from Gene Ontology annotations of transcription factors, co-factors, and signaling molecules. From the full network, a subset of 76 genes was selected to illustrate their functional roles as regulators or targets within the inferred regulatory architecture. Nodes colored in blue represent genes functioning primarily as regulators, while nodes in orange represent genes acting as targets (“passenger” roles) in the network. This visualization highlights the structure and key components driving gene regulation in kidney cortex tissue. (D) WikiPathways enrichment analysis highlighting significantly enriched biological pathways; a bubble plot displays enriched WikiPathways identified from pathway analysis. Each row represents an individual pathway, and bubbles indicate the enrichment signal strength. The x-axis denotes the enrichment signal score, while bubble size corresponds to the number of genes contributing to the enrichment. Bubble color reflects the false discovery rate (FDR), with darker hues indicating greater enrichment. Pathways are grouped based on similarity (threshold = 0.8), as indicated by the colored bar to the right. Notable enriched pathways include microglia pathogen phagocytosis, complement activation, oxidative damage response, and prostaglandin signaling. The plot summarizes key functional processes implicated by the analyzed gene sets.
Figure 6. Human gene network and pathway analysis. (A) The pathways and (B) list of genes that are highly expressed in DKD patients; common DEGs were analyzed from GSE166239 and GSE142025, and 76 common DEGs were selected and categorized. (C) Tissue-specific gene regulatory network inferred from the genotype-tissue expression (GTEx) in the human kidney cortex samples. A gene regulatory network was constructed using GTEx Kidney Cortex v9 transcriptomic data and the ARACNe algorithm [30,31], applying a curated list of regulators derived from Gene Ontology annotations of transcription factors, co-factors, and signaling molecules. From the full network, a subset of 76 genes was selected to illustrate their functional roles as regulators or targets within the inferred regulatory architecture. Nodes colored in blue represent genes functioning primarily as regulators, while nodes in orange represent genes acting as targets (“passenger” roles) in the network. This visualization highlights the structure and key components driving gene regulation in kidney cortex tissue. (D) WikiPathways enrichment analysis highlighting significantly enriched biological pathways; a bubble plot displays enriched WikiPathways identified from pathway analysis. Each row represents an individual pathway, and bubbles indicate the enrichment signal strength. The x-axis denotes the enrichment signal score, while bubble size corresponds to the number of genes contributing to the enrichment. Bubble color reflects the false discovery rate (FDR), with darker hues indicating greater enrichment. Pathways are grouped based on similarity (threshold = 0.8), as indicated by the colored bar to the right. Notable enriched pathways include microglia pathogen phagocytosis, complement activation, oxidative damage response, and prostaglandin signaling. The plot summarizes key functional processes implicated by the analyzed gene sets.
Ijms 27 02491 g006
Ijms 27 02491 g007
Figure 7. 4-PBA restored the expression of C1q, Ncf4, and Ccl28 upregulated in DN mice. Relative mRNA levels of C1q, Ncf4, and Ccl28 were determined by real-time PCR analysis. Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05 vs. control mice; and # p < 0.05 vs. DN mice.
Figure 7. 4-PBA restored the expression of C1q, Ncf4, and Ccl28 upregulated in DN mice. Relative mRNA levels of C1q, Ncf4, and Ccl28 were determined by real-time PCR analysis. Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05 vs. control mice; and # p < 0.05 vs. DN mice.
Ijms 27 02491 g007

3. Discussion

In this study, we demonstrated that 4-PBA, an ER stress inhibitor, reduced glomerular and tubular injury and improved kidney function in DN mice. Specifically, 4-PBA reduced the expression of markers of renal inflammation, apoptosis, oxidative stress, and fibrosis in DN. Further, we identified genes upregulated in human diabetic kidney, confirmed their expression, and studied the effect of 4-PBA in DN mice. The expression of C1q, Ncf4, and Ccl28 was suppressed by 4-PBA, suggesting a protective role in inhibiting ER stress through regulating these genes and associated pathways.
Upon cellular stress, unfolded proteins accumulate in the ER and activate the ER stress response, which globally reduces transcription and translation. The three major pathways are protein kinase RNA (PKR)-like ER kinase (PERK), inositol-requiring protein-1 (IRE1α), and activating transcription factor-6 (ATF6) pathways [32]. Metabolic stressors such as hyperglycemia, hyperlipidemia, cytokines, and reactive oxygen species can stimulate the ER stress response [33]. In the liver of obese mice, ER stress is activated, which suppresses insulin receptor signaling via insulin receptor substrate-1 (IRS-1) [34]. ER stress is associated with age-related renal tubular atrophy, interstitial fibrosis, and glomerulosclerosis. In particular, in aged STZ-treated mice, the expression of BiP, CHOP, p-PERK, and p-eIF2α was significantly increased compared with young controls [35]. ER stress markers are increased in humans and rats with nephrotic syndrome, as well as patients with diabetic kidney disease [36,37].
Chemical chaperones that improve ER protein folding capacity, such as 4-PBA and TUDCA, and the ER chaperone ORP150 have been used to reduce ER stress and insulin resistance [13,38]. These chemical chaperones reduce hyperglycemia-induced podocyte apoptosis and oxidative stress and restore autophagy deficiency in diabetic db/db mice [14,39]. In STZ-induced rats, 4-PBA reduces renal oxidative stress, such as NADPH oxidase activity and NF-κB activity [15]. Consistently, the current study showed that DN mice treated with 4-PBA significantly reduced podocyte loss, oxidative stress, and inflammation and also restored autophagy activity.
The hyperglycemia-induced ER stress triggers an initial protective autophagy response to remove misfolded proteins; however, in chronic stress, it switches to promoting renal apoptosis and further release of inflammatory cytokines, accelerating podocyte and tubular cell loss [19,40]. Thus, ER stress response has both protective and deleterious features, and ER stress modulators should be used cautiously for treating patients with DN. Further, as we demonstrated through a transcriptomic gene network study, key regulators of ER stress and downstream effectors, these modifiers, once developed, can provide therapeutic potential and be effective in a renal cell-specific or disease-stage-specific manner.
This study demonstrated that renal C1q mRNA expression was upregulated in DN and reduced by 4-PBA treatment. C1q, a key initiator of the classical complement pathway, is activated in glomeruli and associated with inflammation, fibrosis, and proteinuria in early-stage DN [26,41,42]. Previous histopathological studies demonstrated that renal C1q deposition correlates with glomerular lesions and severe albuminuria in patients with DN [43,44] and is also associated with poor patient survival [26]. Upstream regulators of C1q include CD33 and FGD2 (FYVE, RhoGEF, and PH domain-containing protein 2), which can be molecular targets of ER stress inhibition. We suggest C1q and its regulators as promising therapeutic targets for patients with DN.
In addition, our gene network pathway analysis through WikiPathways places Ncf4 upstream as part of the NOX2 complex that generates ROS; the C1q gene family occupies the entry point of the classical complement cascade, positioned downstream of oxidative stress and apoptosis, and is also known to the upstream of inflammatory amplification (e.g., NF κB-driven chemokine production) [45,46]. In addition, ccl28 maps to the chemokine arm that links oxidative injury to leukocyte recruitment and fibrogenic signaling [47,48]. Thus, the current study supports that ER stress inhibition attenuates oxidative stress, the complement system, and chemokine regulatory circuits, thereby delaying DN progression.
Transcriptomic analysis also implicates epigenetic modifiers, including enzymes involved in DNA methylation and histone modifications, as metabolic markers of disease onset [49,50]. Thus, reversal of these epigenetic changes may enhance the therapeutic effects of current treatments that control hyperglycemia and/or hypertension during DN progression. We are conducting an active investigation to identify epigenetic regulators that overlap with the ER stress response and DN pathology.
To date, a number of ER stress modulators have been studied, including chemical chaperones (e.g., 4-PBA and TUDCA), targeted UPR pathway inhibitors (e.g., salubrinal and BiP inducer X), incretin agonists (e.g., exenedin-4 and liraglutide), antioxidants (e.g., N-acetylcysteine and quercetin), RAAS inhibitors (e.g., valsartan and aliskiren), and other inhibitors [7,11,40,51,52]. We envision clinical benefits of ER stress inhibitors in attenuating kidney dysfunction. Although the ER stress response possesses both protective and detrimental properties and involves multiple distinct signaling cascades, the appropriate application of ER stress inhibitors along with selective modulators targeting key regulators of the ER stress identified in this study may present a promising therapeutic strategy for patients with DN.
In conclusion, inhibition of ER stress is a promising pharmacological target for treating patients with DN. Key regulators of ER stress play a crucial role in the development of DN, and human transcriptome analysis advances the identification of therapeutic targets for treating patients.

4. Materials and Methods

4.1. Experimental Animals

Wild-type C57BL/6 male mice (8-week-old) were purchased from Koatech Co. (Pyeongtaek, Republic of Korea) and maintained in the animal facility at Gyeongsang National University. All animal experiments were approved by the Institutional Board of Animal Research at Gyeongsang National University and performed according to the National Institutes of Health guidelines for laboratory animal care. Mice were housed with an alternating 12 h light/dark cycle and provided with water and standard chow ad libitum.

4.2. DN Animal Model and Treatment

We used a DN mouse model by combining unilateral nephrectomy (UNx), HFD, and STZ treatment, as previously described [53]. Briefly, mice were habituated for 1 week and subjected to UNx. After 2 days, the mice were fed with a normal chow diet or an HFD (60 kcal% fat; Research Diets, Inc., New Brunswick, NJ, USA). After 4 weeks, a single dose of STZ (100 mg/kg) was intraperitoneally injected into HFD-fed mice. After STZ injection, the mice were treated with 4-PBA (100 mg/kg, daily) or the vehicle (saline) by intraperitoneal injection for 6 weeks. The control mice were fed with a normal chow diet (NCD) and received 4-PBA or vehicle. 4-PBA was purchased from Sigma-Aldrich (St. Louis, MO, USA), and the dose of 100 mg/kg was selected based on the previous studies [39,54]. The group sizes were as follows: control (n = 6), control + 4-PBA (n = 4), DN (n = 8), and DN + 4-PBA (n = 8). Hyperglycemia was assessed by measuring fasting blood glucose level from the tail vein using an Accu-Check glucometer (Roche Diagnostics, Mannheim, Germany). After 6 weeks of 4-PBA treatment, all mice were sacrificed, and the right kidney was removed and weighed. Each half of the kidney was snap-frozen in liquid nitrogen for storage at −80 °C or fixed in 10% buffered formalin for further analysis. Blood was collected from the inferior vena cava using a heparinized syringe, then centrifuged at 3000× g for 20 min, and the supernatants were stored at −80 °C for biochemical analysis.

4.3. Biochemical Assays

Plasma creatinine was measured by a direct colorimetric Jaffe method and detected using a spectrophotometer (Shimadzu UV-1800, Tokyo, Japan), as previously described [12]. Urine samples were collected from each mouse housed in a metabolic cage (Jeungdo Bio & Plant Co., Seoul, Republic of Korea) for 15 h before sacrifice. Urine volume from each mouse was measured, and the samples were centrifuged at 2000× g for 10 min to precipitate the sediment. The supernatant was then transferred to a sterile tube for storage at −80 °C. Urine albumin and creatinine were determined by commercial assay kits from Abcam (Cambridge, MA, USA) according to the manufacturer’s instructions. Urine albumin excretion was measured as the urine albumin-to-creatinine ratio (UACR), calculated by dividing urine albumin by absolute urine creatinine. Mouse urinary NAG activity was measured by a commercial kit from Crystal Chem (Itasca, IL, USA).

4.4. Periodic Acid–Schiff (PAS) and Picro-Sirius Red Staining

The 10% formalin-fixed kidney tissues were embedded in paraffin and sectioned at 5 μm thickness. Sections were stained with PAS staining (Abcam) for histological analysis and Picro-Sirius Red staining (Abcam) for visualization of collagen deposition. All staining was performed according to standard protocols. All images in each group (n = 3) were captured using a CKX41 light microscope (Olympus, Tokyo, Japan) and analyzed using the ImageJ v.1.52a software (NIH).

4.5. Immunohistochemistry (IHC) Analysis

The 10% formalin-fixed and paraffin-embedded kidney tissue sections (5 μm thick) were prepared. Briefly, the fixed kidney sections were deparaffinized, rehydrated, and antigen-retrieved in 10 mM sodium citrate buffer (pH 6.0) for 20 min. The sections were blocked in 10% normal horse serum and incubated with a primary antibody against CD68 (Cluster of Differentiation 68, Abcam) or Wilms’ Tumor Protein 1 (WT1) from Boster Biological Technology (Pleasanton, CA, USA), overnight at 4 °C. The sections were incubated with a biotinylated secondary antibody (Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. The sections were incubated in an avidin–biotin–peroxidase complex (ABC) solution (Vector Laboratories) for 30 min, then developed with a 3,3′-diaminobenzidine (DAB) Peroxidase Substrate Kit (Vector Laboratories). Then, the sections were counterstained with Mayer’s hematoxylin and analyzed using a CKX41 light microscope (Olympus). The area of CD68-stained cells or the number of WT1-stained nuclei (equivalent to the number of podocytes) was counted from 5–10 images at 400× magnification per kidney section in each group (n = 3), using ImageJ (NIH).

4.6. Western Blot Analysis

Renal cortex tissues (20 mg) were homogenized in ice-cold radio-immunoprecipitation assay (RIPA) buffer containing protease inhibitors (Thermo Fisher Scientific, Waltham, MA, USA), sonicated, and incubated on ice for 20 min. After centrifugation, the supernatant was transferred to a clean tube, and the protein concentration was determined using a PierceTM bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific). The protein lysates were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyvinylidene difluoride (PVDF) membranes, and blocked with 5% skim milk or 3% bovine serum albumin (BSA). The membranes were incubated with primary antibodies against LC3B, p62, ATG5, caspase-3, PARP1, CHOP from Cell Signaling Technology (Danvers, MA, USA), PGC-1α from Proteintech group (Chicago, IL, USA), 4-HNE, ATF6 (Abcam), phosphorylated and total IRE1α from Novus Biologicals (Centennial, CO, USA), and β-actin (Sigma-Aldrich) in the blocking solution at 4 °C overnight. Next, the membranes were incubated with the appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (Bio-Rad, Hercules, CA, USA) at room temperature for 1 h and then visualized with the ECL substrate (Bio-Rad). The ChemiDoc XRS+ System (Bio-Rad) was used to evaluate protein band density, and relative protein levels were quantified using ImageJ software (NIH).

4.7. Quantitative Real-Time Polymerase Chain Reaction (PCR) Analysis

The total RNA was extracted from 25 mg of renal cortex with Trizol (Invitrogen, Waltham, MA, USA), and 2 μg of RNA was converted into cDNA using the RevertAid Reverse Transcription System (Thermo Fisher Scientific) according to the manufacturer’s protocol. Real-time PCR analysis was performed with a CFX Connect real-time PCR System using iQ SYBR Green Supermix (Bio-Rad). Real-time PCR analysis was performed with an initial denaturation at 94 °C for 5 min, followed by 45 cycles of 10 s at 95 °C, 10 s at 60 °C, and 30 s at 72 °C. Relative mRNA levels were normalized to those of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The primer sequences are listed in Supplementary Table S1.

4.8. Statistical Analysis

Statistical significance was determined using one-way analysis of variance (ANOVA), followed by Bonferroni’s multiple comparisons test. All statistical analyses were performed with GraphPad Prism 9 Software v.9.50 (GraphPad Software Inc., La Jolla, CA, USA). Data were expressed as the mean ± SEM. A p value < 0.05 was considered statistically significant.

4.9. RNA Seq Data Acquisition

Two publicly available diabetic nephropathy RNA-seq datasets were retrieved from the Gene Expression Omnibus (GSE166239 and GSE142025). For each dataset, corresponding Sequence Read Archive (SRA) accession numbers were identified from GEO metadata. Raw sequencing reads were downloaded as FASTQ files using the NCBI SRA. Each dataset included samples from normal kidney tissue and patients with DN.

4.10. Read Alignment and Quantification (STAR–RSEM Pipeline)

Raw FASTQ files were processed using a standardized STAR–RSEM RNA seq quantification workflow using the human reference genome (GRCh38) and corresponding GENCODE v.38 annotation (GTF). Gene-level abundance tables were used for downstream differential expression analysis.

4.11. Differential Gene Expression (DGE) Analysis

Differential gene expression (DGE) analysis was carried out separately for each dataset, contrasting normal kidney tissue with diabetic kidney biopsy samples. The limma voom framework was used, incorporating voom-based precision weights in linear modeling.

4.12. Gene Selection by Cross-Dataset Integration

Using the log2 fold change values from the differential gene expression analyses, pathway enrichment was performed independently on each dataset using GSEA with the Reactome pathway database. Enrichment scores and statistical significance were calculated by permutation testing, and pathways with an adjusted p-value (FDR) < 0.05 were considered significant. To identify pathways consistently associated with DN, we intersected the Reactome pathways that were significantly enriched across the two datasets. For each shared pathway, leading-edge genes—those contributing most strongly to enrichment—were extracted, and only the overlapping leading-edge genes present in both datasets were retained. Finally, to ensure gene-level robustness, we restricted the list to genes that were also significantly upregulated in both DGE analyses. This integrative approach identified 76 commonly upregulated genes, representing a reproducible transcriptional signature of diabetic kidney injury.

4.13. Kidney-Specific Gene Regulatory Network Reconstruction Using ARACNe

Gene regulatory network inference was performed using the ARACNe algorithm on gene expression data derived from 86 healthy Kidney Cortex samples from the GTEx v8 dataset. ARACNe estimated pairwise mutual information (MI) between transcription factors and candidate target genes using an adaptive partitioning estimator, and interactions not meeting a stringent significance threshold (p ≤ 1 × 10−6, determined through permutation-based null modeling) were removed to reduce spurious associations. To further refine direct regulatory relationships, ARACNe applied the Data Processing Inequality (DPI) with default parameters to eliminate edges likely representing indirect interactions mediated through intermediate regulators. The full inference workflow was executed across 100 independent bootstrap resamples of the expression matrix, generating a collection of bootstrap-specific networks that capture variability and sampling robustness. A consensus regulatory network was then constructed by aggregating edges reproducibly detected across bootstrap iterations, ensuring that only stable and high-confidence transcription factor–target relationships were retained for downstream analyses [30,31].

4.14. Pathway Enrichment Analysis

Pathway enrichment was performed using the MSigDB human Reactome gene sets (collection C2:CP:REACTOME) from the Molecular Signatures Database. Genes were ranked by their signed effect size (RNA-seq differential expression log2 fold change) and analyzed with the Gene Set Enrichment Analysis (GSEA) preranked algorithm. Where multiple transcripts mapped to the same gene symbol, duplicates were collapsed by retaining the entry with the largest absolute log2 fold change. Enrichment scores were computed using the weighted Kolmogorov–Smirnov statistic with permutation-based null models, and multiple testing correction was applied across all Reactome pathways using the Benjamini–Hochberg procedure. Pathways with FDR q-value < 0.05 were considered significantly enriched. Reactome pathway definitions were obtained via MSigDB (human) and trace back to the Reactome Pathway Knowledgebase [55,56,57,58].

4.15. STRING and WikiPathways

Pathway enrichment analysis was performed using a predefined list of 76 candidate genes. Functional enrichment was conducted through the STRING database (Search Tool for the Retrieval of Interacting Genes/Proteins), using its integrated pathway analysis module. The analysis was restricted to WikiPathways gene sets, and enrichment was computed using STRING’s built-in implementation of over-representation analysis based on the hypergeometric test. Pathway significance was evaluated using the false discovery rate (FDR) correction provided by STRING, and pathways with FDR-adjusted p-values < 0.05 were considered significantly enriched. All gene identifiers were mapped automatically to human (Homo sapiens) protein-coding genes using STRING’s default annotation pipeline [59,60].

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/ijms27052491/s1.

Author Contributions

Conceptualization: S.W.P. and H.K.; data curation: J.K., A.Y.R. and H.J.; formal analysis: J.K., A.Y.R. and H.J.; funding acquisition: S.W.P. and H.K.; investigation: J.K., H.J., S.W.P. and H.K.; methodology: J.K. and H.J.; project administration and supervision: S.W.P. and H.K.; writing: J.K., H.J. and H.K.; review and editing: J.-W.K., S.P.Y., H.J.K., J.M., S.W.P. and H.K. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by the Basic Science Research Program through the National Research Foundation (NRF) of South Korea, funded by the Ministry of Science, ICT, and Future Planning (2021R1A2C1010685, RS-2023-00273563, and 2021R1A5A8029490).

Institutional Review Board Statement

Protocols for animal experiments were approved by the Institutional Review Board of Animal Research at Gyeongsang National University (GNU-221123-M0168, approved on 23 November 2022), and experiments were performed according to the NIH Guidelines for Care and Use of Laboratory Animals.

Informed Consent Statement

Not applicable.

Data Availability Statement

All data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflict of interest.

Abbreviations

The following abbreviations are used in this manuscript:
4-PBA4-Phenylbutyrate
AGEsAdvanced glycation end products
α-SMAAlpha-smooth muscle actin
ATF6Activating transcription factor 6
ATG5Autophagy-related protein 5
BiPBinding immunoglobulin protein
BUNBlood urea nitrogen
C1qbComplement component 1q subcomponent b
C1qcComplement component 1q subcomponent c
Ccl28C-C motif chemokine ligand 28
CD33Cluster of differentiation 33
CD68Cluster of differentiation 68
CKDChronic kidney disease
CHOPCCAAT/enhancer-binding protein homologous protein
DEGsDifferentially expressed genes
DGEDifferential gene expression
DKDDiabetic kidney disease
DNDiabetic nephropathy
eIF2αEukaryotic initiation factor 2 alpha
EREndoplasmic reticulum
ERADEndoplasmic reticulum–associated degradation
ESRDEnd-stage renal disease
FASTQText-based format for storing biological sequence data
FDRFalse discovery rate
FFAFree fatty acids
FGD2FYVE, RhoGEF and PH domain-containing protein 2
GAPDHGlyceraldehyde 3-phosphate dehydrogenase
GEOGene Expression Omnibus
GFRGlomerular filtration rate
GOGene ontology
GSEAGene set enrichment analysis
GRCh38Genome Reference Consortium Human Build 38
IL-6Interleukin 6
IL-10Interleukin 10
IRE1αInositol-requiring enzyme 1 alpha
IRS-1Insulin receptor substrate 1
KIM-1Kidney injury molecule 1
LC3BLight chain 3B
NAGN-acetyl-β-D-glucosaminidase
Ncf4Neutrophil cytosolic factor 4
NF-κBNuclear factor kappa B
NGALNeutrophil gelatinase-associated lipocalin
NOX2NADPH oxidase 2
Nphs1Nephrin
Nphs2Podocin
ORP150Oxygen-regulated protein 150 kDa
PARP1Poly (ADP-ribose) polymerase 1
PCRPolymerase chain reaction
PERKProtein kinase RNA-like endoplasmic reticulum kinase
PGC-1αPeroxisome proliferator-activated receptor gamma coactivator 1-alpha
RAASRenin–angiotensin–aldosterone system
RNA-seqRNA sequencing
RSEMRNA-Seq by Expectation-Maximization
RT-PCRReverse transcription polymerase chain reaction
SRASequence Read Archive
STARSpliced Transcripts Alignment to a Reference
STZStreptozotocin
TNF-αTumor necrosis factor alpha
TUDCATaurine-conjugated ursodeoxycholic acid
UNxUnilateral nephrectomy
UPRUnfolded protein response
WT-1Wilms’ tumor 1

References

  1. Jefferson, J.A.; Shankland, S.J.; Pichler, R.H. Proteinuria in diabetic kidney disease: A mechanistic viewpoint. Kidney Int. 2008, 74, 22–36. [Google Scholar] [CrossRef]
  2. Liang, G.; Song, L.; Chen, Z.; Qian, Y.; Xie, J.; Zhao, L.; Lin, Q.; Zhu, G.; Tan, Y.; Li, X.; et al. Fibroblast growth factor 1 ameliorates diabetic nephropathy by an anti-inflammatory mechanism. Kidney Int. 2018, 93, 95–109. [Google Scholar] [CrossRef]
  3. Su, J.; Li, S.J.; Chen, Z.H.; Zeng, C.H.; Zhou, H.; Li, L.S.; Liu, Z.H. Evaluation of podocyte lesion in patients with diabetic nephropathy: Wilms’ tumor-1 protein used as a podocyte marker. Diabetes Res. Clin. Pract. 2010, 87, 167–175. [Google Scholar] [CrossRef]
  4. Kim, Y.; Park, C.W. New therapeutic agents in diabetic nephropathy. Korean J. Intern. Med. 2017, 32, 11–25. [Google Scholar] [CrossRef] [PubMed]
  5. Collins, A.J.; Foley, R.N.; Chavers, B.; Gilbertson, D.; Herzog, C.; Ishani, A.; Johansen, K.; Kasiske, B.L.; Kutner, N.; Liu, J.; et al. US Renal Data System 2013 Annual Data Report. Am. J. Kidney Dis. 2014, 63, A7. [Google Scholar] [CrossRef]
  6. Fan, Y.; Lee, K.; Wang, N.; He, J.C. The Role of Endoplasmic Reticulum Stress in Diabetic Nephropathy. Curr. Diab. Rep. 2017, 17, 17. [Google Scholar] [CrossRef]
  7. Ni, L.; Yuan, C.; Wu, X. Endoplasmic Reticulum Stress in Diabetic Nephrology: Regulation, Pathological Role, and Therapeutic Potential. Oxid. Med. Cell. Longev. 2021, 2021, 7277966. [Google Scholar] [CrossRef] [PubMed]
  8. Zhao, S.; Zhu, L.; Duan, H.; Liu, S.; Liu, Q.; Liu, W.; Hao, J. PI3K/Akt pathway mediates high glucose-induced lipid accumulation in human renal proximal tubular cells via spliced XBP-1. J. Cell. Biochem. 2012, 113, 3288–3298. [Google Scholar] [CrossRef] [PubMed]
  9. Lim, J.C.; Lim, S.K.; Han, H.J.; Park, S.H. Cannabinoid receptor 1 mediates palmitic acid-induced apoptosis via endoplasmic reticulum stress in human renal proximal tubular cells. J. Cell. Physiol. 2010, 225, 654–663. [Google Scholar] [CrossRef]
  10. Fang, L.; Xie, D.; Wu, X.; Cao, H.; Su, W.; Yang, J. Involvement of endoplasmic reticulum stress in albuminuria induced inflammasome activation in renal proximal tubular cells. PLoS ONE 2013, 8, e72344. [Google Scholar] [CrossRef]
  11. Zhang, R.; Bian, C.; Gao, J.; Ren, H. Endoplasmic reticulum stress in diabetic kidney disease: Adaptation and apoptosis after three UPR pathways. Apoptosis 2023, 28, 977–996. [Google Scholar] [CrossRef] [PubMed]
  12. Liu, X.; Zhang, C.; Fu, Y.; Xie, L.; Kong, Y.; Yang, X. Inflammation, Apoptosis, and Fibrosis in Diabetic Nephropathy: Molecular Crosstalk in Proximal Tubular Epithelial Cells and Therapeutic Implications. Curr. Issues Mol. Biol. 2025, 47, 885. [Google Scholar] [CrossRef] [PubMed]
  13. Ozcan, U.; Yilmaz, E.; Ozcan, L.; Furuhashi, M.; Vaillancourt, E.; Smith, R.O.; Gorgun, C.Z.; Hotamisligil, G.S. Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes. Science 2006, 313, 1137–1140. [Google Scholar] [CrossRef]
  14. Cao, A.; Wang, L.; Chen, X.; Guo, H.; Chu, S.; Zhang, X.; Peng, W. Ursodeoxycholic Acid Ameliorated Diabetic Nephropathy by Attenuating Hyperglycemia-Mediated Oxidative Stress. Biol. Pharm. Bull. 2016, 39, 1300–1308. [Google Scholar] [CrossRef]
  15. Luo, Z.F.; Feng, B.; Mu, J.; Qi, W.; Zeng, W.; Guo, Y.H.; Pang, Q.; Ye, Z.L.; Liu, L.; Yuan, F.H. Effects of 4-phenylbutyric acid on the process and development of diabetic nephropathy induced in rats by streptozotocin: Regulation of endoplasmic reticulum stress-oxidative activation. Toxicol. Appl. Pharmacol. 2010, 246, 49–57. [Google Scholar] [CrossRef] [PubMed]
  16. Barabasi, A.L.; Gulbahce, N.; Loscalzo, J. Network medicine: A network-based approach to human disease. Nat. Rev. Genet. 2011, 12, 56–68. [Google Scholar] [CrossRef]
  17. Castillo-Rodriguez, E.; Fernandez-Prado, R.; Martin-Cleary, C.; Pizarro-Sanchez, M.S.; Sanchez-Nino, M.D.; Sanz, A.B.; Fernandez-Fernandez, B.; Ortiz, A. Kidney Injury Marker 1 and Neutrophil Gelatinase-Associated Lipocalin in Chronic Kidney Disease. Nephron 2017, 136, 263–267. [Google Scholar] [CrossRef]
  18. Sabbisetti, V.S.; Waikar, S.S.; Antoine, D.J.; Smiles, A.; Wang, C.; Ravisankar, A.; Ito, K.; Sharma, S.; Ramadesikan, S.; Lee, M.; et al. Blood kidney injury molecule-1 is a biomarker of acute and chronic kidney injury and predicts progression to ESRD in type I diabetes. J. Am. Soc. Nephrol. 2014, 25, 2177–2186. [Google Scholar] [CrossRef]
  19. Ding, Y.; Choi, M.E. Autophagy in diabetic nephropathy. J. Endocrinol. 2015, 224, R15–R30. [Google Scholar] [CrossRef]
  20. Su, J.; Zhou, L.; Kong, X.; Yang, X.; Xiang, X.; Zhang, Y.; Li, X.; Sun, L. Endoplasmic reticulum is at the crossroads of autophagy, inflammation, and apoptosis signaling pathways and participates in the pathogenesis of diabetes mellitus. J. Diabetes Res. 2013, 2013, 193461. [Google Scholar] [CrossRef]
  21. Tanaka, Y.; Kume, S.; Kitada, M.; Kanasaki, K.; Uzu, T.; Maegawa, H.; Koya, D. Autophagy as a therapeutic target in diabetic nephropathy. Exp. Diabetes Res. 2012, 2012, 628978. [Google Scholar] [CrossRef]
  22. Qian, L.; Zhu, Y.; Deng, C.; Liang, Z.; Chen, J.; Chen, Y.; Wang, X.; Liu, Y.; Tian, Y.; Yang, Y. Peroxisome proliferator-activated receptor gamma coactivator-1 (PGC-1) family in physiological and pathophysiological process and diseases. Signal Transduct. Target. Ther. 2024, 9, 50. [Google Scholar] [CrossRef] [PubMed]
  23. Nordbo, O.P.; Landolt, L.; Eikrem, O.; Scherer, A.; Leh, S.; Furriol, J.; Apeland, T.; Mydel, P.; Marti, H.P. Transcriptomic analysis reveals partial epithelial-mesenchymal transition and inflammation as common pathogenic mechanisms in hypertensive nephrosclerosis and Type 2 diabetic nephropathy. Physiol. Rep. 2023, 11, e15825. [Google Scholar] [CrossRef] [PubMed]
  24. Fan, Y.; Yi, Z.; D’Agati, V.D.; Sun, Z.; Zhong, F.; Zhang, W.; Wen, J.; Zhou, T.; Li, Z.; He, L.; et al. Comparison of Kidney Transcriptomic Profiles of Early and Advanced Diabetic Nephropathy Reveals Potential New Mechanisms for Disease Progression. Diabetes 2019, 68, 2301–2314, Erratum in Diabetes 2020, 69, 797. [Google Scholar] [CrossRef] [PubMed]
  25. Huang, Y.; Xu, J.; Wu, X.; Chen, X.; Bai, X.; Zhuang, Y.; Fang, J.; Lin, X. High Expression of Complement Components in the Kidneys of Type 2 Diabetic Rats With Diabetic Nephropathy. Front. Endocrinol. 2019, 10, 459. [Google Scholar] [CrossRef]
  26. Jiao, Y.; Jiang, S.; Wang, Y.; Yu, T.; Zou, G.; Zhuo, L.; Li, W. Activation of complement C1q and C3 in glomeruli might accelerate the progression of diabetic nephropathy: Evidence from transcriptomic data and renal histopathology. J. Diabetes Investig. 2022, 13, 839–849. [Google Scholar] [CrossRef]
  27. Chen, Y.; He, F.; Wang, R.; Yao, M.; Li, Y.; Guo, D.; He, S. NCF1/2/4 Are Prognostic Biomarkers Related to the Immune Infiltration of Kidney Renal Clear Cell Carcinoma. BioMed Res. Int. 2021, 2021, 5954036. [Google Scholar] [CrossRef]
  28. Qiu, Y.; Tang, J.; Zhao, Q.; Jiang, Y.; Liu, Y.N.; Liu, W.J. From Diabetic Nephropathy to End-Stage Renal Disease: The Effect of Chemokines on the Immune System. J. Diabetes Res. 2023, 2023, 3931043. [Google Scholar] [CrossRef]
  29. Liu, W.; Zhang, J.; Zhang, D.; Zhang, L. Role of circulating inflammatory protein in the development of diabetic renal complications: Proteome-wide Mendelian randomization and colocalization analyses. Front. Endocrinol. 2024, 15, 1406442. [Google Scholar] [CrossRef]
  30. Jang, I.S.; Margolin, A.; Califano, A. hARACNe: Improving the accuracy of regulatory model reverse engineering via higher-order data processing inequality tests. Interface Focus 2013, 3, 20130011. [Google Scholar] [CrossRef]
  31. Margolin, A.A.; Nemenman, I.; Basso, K.; Wiggins, C.; Stolovitzky, G.; Dalla Favera, R.; Califano, A. ARACNE: An algorithm for the reconstruction of gene regulatory networks in a mammalian cellular context. BMC Bioinform. 2006, 7, S7. [Google Scholar] [CrossRef]
  32. Szegezdi, E.; Logue, S.E.; Gorman, A.M.; Samali, A. Mediators of endoplasmic reticulum stress-induced apoptosis. EMBO Rep. 2006, 7, 880–885. [Google Scholar] [CrossRef]
  33. Ron, D.; Walter, P. Signal integration in the endoplasmic reticulum unfolded protein response. Nat. Rev. Mol. Cell Biol. 2007, 8, 519–529. [Google Scholar] [CrossRef]
  34. Ozcan, U.; Cao, Q.; Yilmaz, E.; Lee, A.H.; Iwakoshi, N.N.; Ozdelen, E.; Tuncman, G.; Gorgun, C.; Glimcher, L.H.; Hotamisligil, G.S. Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science 2004, 306, 457–461. [Google Scholar] [CrossRef]
  35. Wu, J.; Zhang, R.; Torreggiani, M.; Ting, A.; Xiong, H.; Striker, G.E.; Vlassara, H.; Zheng, F. Induction of diabetes in aged C57B6 mice results in severe nephropathy: An association with oxidative stress, endoplasmic reticulum stress, and inflammation. Am. J. Pathol. 2010, 176, 2163–2176. [Google Scholar] [CrossRef]
  36. Inagi, R.; Nangaku, M.; Onogi, H.; Ueyama, H.; Kitao, Y.; Nakazato, K.; Ogawa, S.; Kurokawa, K.; Couser, W.G.; Miyata, T. Involvement of endoplasmic reticulum (ER) stress in podocyte injury induced by excessive protein accumulation. Kidney Int. 2005, 68, 2639–2650. [Google Scholar] [CrossRef]
  37. Lindenmeyer, M.T.; Rastaldi, M.P.; Ikehata, M.; Neusser, M.A.; Kretzler, M.; Cohen, C.D.; Schlondorff, D. Proteinuria and hyperglycemia induce endoplasmic reticulum stress. J. Am. Soc. Nephrol. 2008, 19, 2225–2236. [Google Scholar] [CrossRef] [PubMed]
  38. Ozawa, K.; Miyazaki, M.; Matsuhisa, M.; Takano, K.; Nakatani, Y.; Hatazaki, M.; Tamatani, T.; Yamagata, K.; Miyagawa, J.; Kitao, Y.; et al. The endoplasmic reticulum chaperone improves insulin resistance in type 2 diabetes. Diabetes 2005, 54, 657–663. [Google Scholar] [CrossRef]
  39. Cao, A.L.; Wang, L.; Chen, X.; Wang, Y.M.; Guo, H.J.; Chu, S.; Liu, C.; Zhang, X.M.; Peng, W. Ursodeoxycholic acid and 4-phenylbutyrate prevent endoplasmic reticulum stress-induced podocyte apoptosis in diabetic nephropathy. Lab. Investig. 2016, 96, 610–622. [Google Scholar] [CrossRef] [PubMed]
  40. Cunard, R.; Sharma, K. The endoplasmic reticulum stress response and diabetic kidney disease. Am. J. Physiol. Renal. Physiol. 2011, 300, F1054–F1061. [Google Scholar] [CrossRef] [PubMed]
  41. Momeni, A.; Nasri, H. Concurrent diabetic nephropathy and C1q nephropathy in a young male patient: The first report in literature. J. Nephropathol. 2013, 2, 201–203. [Google Scholar] [PubMed]
  42. Kotsalas, N.; Fouza, A.; Daoudaki, M. The Role of Complement Activation in Diabetic Nephropathy: Current Insights and Future Directions. J. Clin. Med. 2025, 14, 8589. [Google Scholar] [CrossRef]
  43. Bus, P.; Chua, J.S.; Klessens, C.Q.F.; Zandbergen, M.; Wolterbeek, R.; van Kooten, C.; Trouw, L.A.; Bruijn, J.A.; Baelde, H.J. Complement Activation in Patients With Diabetic Nephropathy. Kidney Int. Rep. 2018, 3, 302–313. [Google Scholar] [CrossRef]
  44. Sun, Z.J.; Li, X.Q.; Chang, D.Y.; Wang, S.X.; Liu, G.; Chen, M.; Zhao, M.H. Complement deposition on renal histopathology of patients with diabetic nephropathy. Diabetes Metab. 2019, 45, 363–368. [Google Scholar] [CrossRef]
  45. Kouser, L.; Madhukaran, S.P.; Shastri, A.; Saraon, A.; Ferluga, J.; Al-Mozaini, M.; Kishore, U. Emerging and Novel Functions of Complement Protein C1q. Front. Immunol. 2015, 6, 317. [Google Scholar] [CrossRef]
  46. Fraser, D.A.; Laust, A.K.; Nelson, E.L.; Tenner, A.J. C1q differentially modulates phagocytosis and cytokine responses during ingestion of apoptotic cells by human monocytes, macrophages, and dendritic cells. J. Immunol. 2009, 183, 6175–6185. [Google Scholar] [CrossRef]
  47. Mohan, T.; Deng, L.; Wang, B.Z. CCL28 chemokine: An anchoring point bridging innate and adaptive immunity. Int. Immunopharmacol. 2017, 51, 165–170. [Google Scholar] [CrossRef]
  48. Russo, R.C.; Ryffel, B. The Chemokine System as a Key Regulator of Pulmonary Fibrosis: Converging Pathways in Human Idiopathic Pulmonary Fibrosis (IPF) and the Bleomycin-Induced Lung Fibrosis Model in Mice. Cells 2024, 13, 2058. [Google Scholar] [CrossRef]
  49. Chai, S.; Zeng, Z.; Shang, J.; Xu, Y.; Han, X.; Ma, B.; Lv, Z.; Wang, R. Histone modification-mediated epigenetic regulation in AKI-to-CKD transition: A cell-type-specific perspective. J. Transl. Med. 2025, 23, 1346. [Google Scholar] [CrossRef] [PubMed]
  50. Kato, M.; Natarajan, R. Epigenetics and epigenomics in diabetic kidney disease and metabolic memory. Nat. Rev. Nephrol. 2019, 15, 327–345. [Google Scholar] [CrossRef] [PubMed]
  51. Cunard, R. Endoplasmic Reticulum Stress in the Diabetic Kidney, the Good, the Bad and the Ugly. J. Clin. Med. 2015, 4, 715–740. [Google Scholar] [CrossRef]
  52. Yang, M.; Liu, C.; Jiang, N.; Liu, Y.; Luo, S.; Li, C.; Zhao, H.; Han, Y.; Chen, W.; Li, L.; et al. Endoplasmic reticulum homeostasis: A potential target for diabetic nephropathy. Front. Endocrinol. 2023, 14, 1182848. [Google Scholar] [CrossRef]
  53. Kim, H.; Dusabimana, T.; Kim, S.R.; Je, J.; Jeong, K.; Kang, M.C.; Cho, K.M.; Kim, H.J.; Park, S.W. Supplementation of Abelmoschus manihot Ameliorates Diabetic Nephropathy and Hepatic Steatosis by Activating Autophagy in Mice. Nutrients 2018, 10, 1703. [Google Scholar] [CrossRef]
  54. Zhou, L.; Ding, S.; Li, Y.; Wang, L.; Chen, W.; Bo, T.; Wu, K.; Li, C.; Liu, X.; Zhao, J.; et al. Endoplasmic Reticulum Stress May Play a Pivotal Role in Lipid Metabolic Disorders in a Novel Mouse Model of Subclinical Hypothyroidism. Sci. Rep. 2016, 6, 31381. [Google Scholar] [CrossRef]
  55. Subramanian, A.; Tamayo, P.; Mootha, V.K.; Mukherjee, S.; Ebert, B.L.; Gillette, M.A.; Paulovich, A.; Pomeroy, S.L.; Golub, T.R.; Lander, E.S.; et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 2005, 102, 15545–15550. [Google Scholar] [CrossRef]
  56. Jassal, B.; Matthews, L.; Viteri, G.; Gong, C.; Lorente, P.; Fabregat, A.; Sidiropoulos, K.; Cook, J.; Gillespie, M.; Haw, R.; et al. The reactome pathway knowledgebase. Nucleic Acids Res. 2020, 48, D498–D503. [Google Scholar] [CrossRef]
  57. Liberzon, A.; Subramanian, A.; Pinchback, R.; Thorvaldsdottir, H.; Tamayo, P.; Mesirov, J.P. Molecular signatures database (MSigDB) 3.0. Bioinformatics 2011, 27, 1739–1740. [Google Scholar] [CrossRef] [PubMed]
  58. Benjamini, Y.; Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J. R. Stat. Soc. Ser. B 1995, 57, 289. [Google Scholar] [CrossRef]
  59. Martens, M.; Ammar, A.; Riutta, A.; Waagmeester, A.; Slenter, D.N.; Hanspers, K.; Miller, R.A.; Digles, D.; Lopes, E.N.; Ehrhart, F.; et al. WikiPathways: Connecting communities. Nucleic Acids Res. 2021, 49, D613–D621. [Google Scholar] [CrossRef]
  60. Szklarczyk, D.; Gable, A.L.; Nastou, K.C.; Lyon, D.; Kirsch, R.; Pyysalo, S.; Doncheva, N.T.; Legeay, M.; Fang, T.; Bork, P.; et al. The STRING database in 2021: Customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. 2021, 49, D605–D612. [Google Scholar] [CrossRef] [PubMed]
Ijms 27 02491 g001
Figure 1. 4-PBA improved glomerular filtration function and reduced tubular injury in DN. (A) The DN mouse model was generated by STZ injection (100 mg/kg) into the mice after 3 weeks of UNx and HFD feeding. After STZ, mice were administered 4-PBA (100 mg/kg) or vehicle for 6 weeks. The control mice were fed with ND and received vehicle or 4-PBA. At sacrifice, the blood, kidney, and urine samples were collected. (B) Kidney/body weight ratio, (C) plasma BUN, (D) plasma creatinine, (E) urine volume (mL/15 h), (F) urine albumin-to-creatinine ratio (UACR), and (G) urinary NAG activity were analyzed, respectively. (H,I) Relative mRNA levels of kidney injury molecule-1 (KIM-1) and neutrophil gelatinase-associated lipocalin (NGAL), specific biomarkers of renal tubular injury, were assessed by real-time PCR. Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Control (Con) or Control + 4-PBA, n = 4–6; DN or DN + 4-PBA, n = 6–7. Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice.
Figure 1. 4-PBA improved glomerular filtration function and reduced tubular injury in DN. (A) The DN mouse model was generated by STZ injection (100 mg/kg) into the mice after 3 weeks of UNx and HFD feeding. After STZ, mice were administered 4-PBA (100 mg/kg) or vehicle for 6 weeks. The control mice were fed with ND and received vehicle or 4-PBA. At sacrifice, the blood, kidney, and urine samples were collected. (B) Kidney/body weight ratio, (C) plasma BUN, (D) plasma creatinine, (E) urine volume (mL/15 h), (F) urine albumin-to-creatinine ratio (UACR), and (G) urinary NAG activity were analyzed, respectively. (H,I) Relative mRNA levels of kidney injury molecule-1 (KIM-1) and neutrophil gelatinase-associated lipocalin (NGAL), specific biomarkers of renal tubular injury, were assessed by real-time PCR. Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Control (Con) or Control + 4-PBA, n = 4–6; DN or DN + 4-PBA, n = 6–7. Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice.
Ijms 27 02491 g001
Ijms 27 02491 g002
Figure 2. 4-PBA reduced podocyte loss and tubular interstitial fibrosis in DN. Kidney sections were processed for PAS staining (glomerular area in (A) and tubular area in (B)), WT1 immunohistochemistry (C), and Picro-Sirius Red staining (D) (n = 3); representative images are shown (black arrows indicate damaged renal areas). Relative mRNA levels of Nphs-1 and Nphs-2 (specific markers for podocyte loss), as well as α-SMA and collagen I (fibrogenic markers), were assessed by real-time PCR analysis (E) (n = 3–5). Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice. Scale bar, 100 µm.
Figure 2. 4-PBA reduced podocyte loss and tubular interstitial fibrosis in DN. Kidney sections were processed for PAS staining (glomerular area in (A) and tubular area in (B)), WT1 immunohistochemistry (C), and Picro-Sirius Red staining (D) (n = 3); representative images are shown (black arrows indicate damaged renal areas). Relative mRNA levels of Nphs-1 and Nphs-2 (specific markers for podocyte loss), as well as α-SMA and collagen I (fibrogenic markers), were assessed by real-time PCR analysis (E) (n = 3–5). Relative mRNA expression was normalized to that of GAPDH (n = 3–5). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice. Scale bar, 100 µm.
Ijms 27 02491 g002
Ijms 27 02491 g003
Figure 3. 4-PBA reduced ER stress and increased autophagy in DN. Kidney tissues were lysed for Western blot analysis, and the protein expression levels of ER stress markers (A), autophagy markers (B), and β-actin (loading control) were determined. Quantitative analysis of each protein was shown (n = 3). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice.
Figure 3. 4-PBA reduced ER stress and increased autophagy in DN. Kidney tissues were lysed for Western blot analysis, and the protein expression levels of ER stress markers (A), autophagy markers (B), and β-actin (loading control) were determined. Quantitative analysis of each protein was shown (n = 3). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice.
Ijms 27 02491 g003
Ijms 27 02491 g004
Figure 4. 4-PBA reduced renal inflammation, apoptosis, and oxidative stress in DN. (A) Kidney sections were processed for immunohistochemical staining of CD68, a marker of macrophages, and representative images are shown (black arrows indicate areas infiltrated by macrophages). The CD68-positive area (%) was calculated using ImageJ v.1.52a software (n = 3). (B) Relative mRNA levels of pro-inflammatory cytokines (TNF-α, IL-6) and anti-inflammatory cytokines (IL-10) were determined by real-time PCR analysis. Relative mRNA expression was normalized to that of GAPDH (n = 3–6). (C) Kidney tissues were lysed for Western blot analysis; the extent of caspase-3 and PARP1 cleavage was determined, and oxidative stress was measured by 4-HNE, a marker of lipid peroxidation. Quantification of each protein is shown (n = 3). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice. Scale bar, 100 µm.
Figure 4. 4-PBA reduced renal inflammation, apoptosis, and oxidative stress in DN. (A) Kidney sections were processed for immunohistochemical staining of CD68, a marker of macrophages, and representative images are shown (black arrows indicate areas infiltrated by macrophages). The CD68-positive area (%) was calculated using ImageJ v.1.52a software (n = 3). (B) Relative mRNA levels of pro-inflammatory cytokines (TNF-α, IL-6) and anti-inflammatory cytokines (IL-10) were determined by real-time PCR analysis. Relative mRNA expression was normalized to that of GAPDH (n = 3–6). (C) Kidney tissues were lysed for Western blot analysis; the extent of caspase-3 and PARP1 cleavage was determined, and oxidative stress was measured by 4-HNE, a marker of lipid peroxidation. Quantification of each protein is shown (n = 3). Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis, followed by Bonferroni’s multiple comparisons. * p < 0.05, ** p < 0.01 vs. control mice; and # p < 0.05, ## p < 0.01 vs. DN mice. Scale bar, 100 µm.
Ijms 27 02491 g004
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Karekezi, J.; Yves Roger, A.; Jang, H.; Kim, J.-W.; Yun, S.P.; Kim, H.J.; Miao, J.; Park, S.W.; Kim, H. Identification of Common Genes Regulated by ER Stress During the Development of Diabetic Nephropathy Based on Human Transcriptome Datasets and an In Vivo Mouse Model. Int. J. Mol. Sci. 2026, 27, 2491. https://doi.org/10.3390/ijms27052491

AMA Style

Karekezi J, Yves Roger A, Jang H, Kim J-W, Yun SP, Kim HJ, Miao J, Park SW, Kim H. Identification of Common Genes Regulated by ER Stress During the Development of Diabetic Nephropathy Based on Human Transcriptome Datasets and an In Vivo Mouse Model. International Journal of Molecular Sciences. 2026; 27(5):2491. https://doi.org/10.3390/ijms27052491

Chicago/Turabian Style

Karekezi, Jacques, Ashimwe Yves Roger, Harry Jang, Jong-Won Kim, Seung Pil Yun, Hye Jung Kim, Ji Miao, Sang Won Park, and Hwajin Kim. 2026. "Identification of Common Genes Regulated by ER Stress During the Development of Diabetic Nephropathy Based on Human Transcriptome Datasets and an In Vivo Mouse Model" International Journal of Molecular Sciences 27, no. 5: 2491. https://doi.org/10.3390/ijms27052491

APA Style

Karekezi, J., Yves Roger, A., Jang, H., Kim, J.-W., Yun, S. P., Kim, H. J., Miao, J., Park, S. W., & Kim, H. (2026). Identification of Common Genes Regulated by ER Stress During the Development of Diabetic Nephropathy Based on Human Transcriptome Datasets and an In Vivo Mouse Model. International Journal of Molecular Sciences, 27(5), 2491. https://doi.org/10.3390/ijms27052491

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop