Skip to main content
Download PDF

Exploring the interplay between adipokine-mediated celastrol target genes and T cells in diabetic nephropathy: a mendelian randomization-based causal inference

Diabetology & Metabolic Syndrome volume 17, Article number: 89 (2025) Cite this article

Abstract

Background

Diabetic nephropathy (DN) is influenced by dysregulated adipokines, which play a key role in inflammation, immune responses, and lipid metabolism. However, the precise molecular mechanisms linking adipokine dysregulation, immune cell infiltration, and metabolic reprogramming in DN remain poorly understood. Celastrol, a bioactive lipid regulator, has been shown to mitigate renal immune-inflammatory damage by inhibiting the PI3K/Akt/NF-κB signaling pathway. Yet, its specific impact on adipokine-mediated immune responses and lipid metabolism in DN is unclear. This study aims to elucidate the interplay between adipokine-mediated target genes in DN and investigate how celastrol modulates these interactions.

Methods

Gene expression profiles of DN patients were obtained from GEO datasets (GSE30122 and GSE30528) and analyzed for differentially expressed genes (DEGs) using the limma package. Gene set variation analysis (GSVA) was conducted to assess lipid metabolism pathways, while Mendelian randomization (MR) and Pearson correlation evaluated the association between DEGs and adipokines. Immune cell infiltration was analyzed using the IOBR R package (MCP-counter and xCell methods), followed by MR analysis of DN-related immune responses. Celastrol target genes were identified using the SEA database.

Results

A total of 70 intersecting DEGs were identified. GSVA revealed that brown and beige adipocyte differentiation pathways were downregulated, while adipocyte-related pathways were upregulated in DN (p < 0.05). MR analysis demonstrated that adiponectin was negatively associated with DN (OR = 0.77, P = 0.005), whereas leptin (OR = 1.92, P = 0.016) and resistin (OR = 1.43, P < 0.001) were positively associated. Three key genes, MAGI2, FGF9, and THBS2 were linked to DN risk and T cell infiltration. THBS2 was positively correlated with T cell infiltration (OR = 0.51, P = 6.7e-06), while FGF9 (OR = -0.8, P = 2.2e-16) and MAGI2 (OR = 0.75, P = 1.3e-13) were negatively correlated. 22 celastrol target genes, including MAGI2, FGF9, and THBS2, were identified.

Conclusion

Our findings reveal that celastrol modulates DN progression through adipokine-immune crosstalk, with FGF9, MAGI2, and THBS2 emerging as key regulatory genes. These insights provide new avenues for biomarker discovery and therapeutic implications in the development of DN.

Introduction

Diabetic nephropathy (DN) is a leading cause of end-stage renal disease (ESRD) worldwide, and the prevalence of DN among adults with diabetes reaches as high as 30.9% in China [1, 2]. Blood glucose control and treatments that reduce urinary albumin do not fundamentally alter the course of DN [3]. The latest evidence-based guidelines recommend the use of sodium-glucose co-transporter 2 inhibitors, as well as angiotensin-converting enzyme inhibitors (ACEIs) / angiotensin receptor blockers (ARBs) and novel antihyperglycemic drugs. However, sodium-glucose co-transporter 2 inhibitors have not been shown to slow the progression of DN to ESRD [3]. Furthermore, due to the heterogeneity and complex mechanisms of diabetes-related kidney injury, there is still controversy regarding the current diagnosis and treatment of DN [4]. Therefore, it is urgent to explore strategies for early prevention and treatment of DN.

Lipids are the main energy source for the kidneys, and lipid metabolism disorders are closely associated with the occurrence and progression of DN [5]. Studies have indicated that the accumulation of cholesterol, phospholipids, triglycerides, fatty acids, and sphingolipids in the kidneys is related to the pathogenesis of DN [6]. Impaired fatty acid oxidation may lead to a disruption in the balance between fatty acid synthesis, uptake, and consumption, which in turn affects renal lipid metabolism and promotes the accumulation of intracellular lipids [7]. Dyslipidemia can damage the kidneys by enhancing lipid-mediated oxidative stress and inflammation, lipogenesis, ectopic lipid deposition, and changes in lipid distribution, metabolism, and β-oxidation [8]. Consequently, finding ways to alleviate DN progression by regulating lipid metabolism has become a key research focus [9].

Recent studies have shown that adipokines—bioactive peptides secreted by adipose tissue—play crucial roles in regulating lipid metabolism, insulin sensitivity, and inflammatory responses. Several adipokines, including leptin, adiponectin, and resistin, have been identified as important mediators of dyslipidemia [10]. Adiponectin is known to exert protective effects against metabolic disorders; Kawano et al. identified that adiponectin is an valuable marker for early DN detection [11], and adiponectin receptor agonists have been shown to improve renal function in type 2 diabetes (T2D) mouse models [12]. Leptin levels, commonly elevated in obesity and T2D, are associated with increased lipotoxicity and renal inflammation, which are central to DN pathogenesis. Resistin, initially recognized for its role in insulin resistance, is also implicated in lipid metabolism and inflammation. Huang et al. demonstrated that elevated levels of leptin and resistin are risk factors for DN in T2D patients with lower body mass index [13]. At the same time, another study showed that elevated plasma resistin is associated with DN risk in T2D patients [14]. While these adipokines have been widely studied, their exact causal roles in DN onset and progression remain unclear.

Celastrol—a natural quinone-methide triterpenoid compound derived from the root bark of Tripterygium wilfordii—has attracted growing attention for its role in preventing and mitigating various metabolic disorders. Our previous studies have shown that celastrol exerts anti-obesity effects by modulating interactions among systemic insulin sensitivity, adipose tissue inflammation, and skeletal muscle mitochondrial function [15, 16]. Additionally, celastrol has been shown to reduce adipokine resistin-associated matrix interactions and improve vascular smooth muscle cell migration [17]. These regulatory mechanisms function in immune responses, inflammation, and oxidative stress—all critical processes in DN pathogenesis. Recent research has demonstrated that celastrol significantly delays renal damage in DN [18]. Celastrol can lower blood glucose levels and renal function indices in db/db mice, alleviate morphological damage to kidney tissue, and induce significant anti-inflammatory effects [19]. Our recent report found that celastrol significantly enhances cell viability and reduces apoptosis in high glucose (HG)-treated HK-2 cells. Additionally, celastrol decreases pro-inflammatory cytokine levels and increases antioxidant enzyme activities, thereby mitigating fibrosis and oxidative stress induced by high glucose. These effects are primarily attributed to the inhibition of the PI3K/Akt/NF-κB signaling pathway [20]. Current evidence underscores its potential in modulating lipid metabolism, inflammation, and oxidative stress, presenting a promising therapeutic approach for managing dyslipidemia and suppressing DN progression.

Mendelian randomization (MR) is a robust epidemiological method that uses genetic variation as instrumental variables to estimate the causal effect of an exposure on an outcome, offering a feasible approach to overcome these limitations [21]. By applying MR, researchers can leverage genetic predispositions to elevated or decreased adipokine levels to elucidate their direct causal effects on DN. Recent studies suggest that celastrol acts as a leptin sensitizer, reducing body fat and leptin levels [22]. Moreover, celastrol exhibits significant regulatory effects on adiponectin and resistin levels [23]. Celastrol, a bioactive triterpenoid, has demonstrated promising renoprotective effects [20], but its role in adipokine-mediated immune modulation remains unexplored. To address this, we employed an MR approach combined with bioinformatics to elucidate adipokine-target gene interactions in DN. Specifically, this study focuses on leptin, adiponectin, and resistin, using a combined MR and bioinformatics analysis approach to investigate how genes associated with these adipokines contribute to DN susceptibility and progression. Understanding these causal relationships will not only deepen our knowledge of DN pathophysiology but may also pave the way for novel diagnostic biomarkers and therapeutic strategies targeting lipid metabolism.

Methods

Data acquisition

The expression profile datasets of DN patients GSE30122 (including 19 samples of DN and 50 normal samples) and GSE30528 (including 9 samples of DN and 13 normal samples) were downloaded from the GEO database (https://www.ncbi.nlm.nih.gov/geo). The disease samples primarily represent stages III-IV of DN, and all samples in these datasets were obtained from renal biopsies. The probes in the GEO datasets were changed to corresponding gene names using the R package “idmap3” (https://github.com/jmzeng1314/idmap3).

Identification of differentially expressed genes (DEGs) in DN

Differential expression analysis was performed on the GSE30122 and GSE30528 datasets to identify DEGs between normal and DN samples using the R package “limma”. The criteria were set as P value < 0.05 and |log2 fold change| > 1. Subsequently, the intersecting genes between GSE30122 and GSE30528 were screened through a Venn diagram drawn using “VennDiagram”.

Gene set variation analysis (GSVA) of lipid metabolism

Five GMT reference gene sets related to lipid metabolism were retrieved from MSigDB (https://www.gsea-msigdb.org/gsea/msigdb/), including: Brown_and_beige_adipocyte_differentiation, White_adipocyte_differentiation, Differentiation_.

of_white_and_brown_adipocyte, Adipocytes_diabetic_condition, and Adipocytes_.

normal_condition. Pathway enrichment scores were performed in GSE30122 using the R package “GSVA” and the differences in scoring results were statistically analyzed. Correlations between the selected intersection genes and pathway scores were also calculated. Significantly related genes were selected according to |correlation| > 0.6 and P value < 0.05.

MR analysis

We performed MR analysis to investigate the causal relationship between adipokines (adiponectin, leptin, and resistin), core genes (FGF9, MAGI2, RRM2, VCAN, CD163, THBS2), T-cell-related factors, and DN. GWAS data for these adipokines and genes were obtained from the IEU OpenGWAS database (https://gwas.mrcieu.ac.uk/) and the GWAS Catalog (https://www.ebi.ac.uk/gwas/), using database IDs: ieu-a-1 (adiponectin), ieu-a-1003 (leptin), ebi-a-GCST90012034 (resistin), eqtl-a-ENSG00000102678 (FGF9), eqtl-a-ENSG00000187391 (MAGI2), eqtl-a-ENSG00000171848 (RRM2), eqtl-a-ENSG00000038427 (VCAN), eqtl-a-ENSG00000177575 (CD163), eqtl-a-ENSG00000186340 (THBS2), GCST90001620 (HLA DR + T cell lymphocyte), and GCST90001918 (CD45 on HLA DR + T cell). For the outcome variable, GWAS data for DN (ebi-a-GCST90018832) were also retrieved. We selected instrumental variables (IVs) through correlation analysis with a P threshold of < 5e-05 and retrieved relevant data, including effect size (beta), standard error (SE), and P value for each SNP. To ensure IV independence, we used the clump_data function in the TwoSampleMR package to remove SNPs in linkage disequilibrium with parameters clump_r2 = 0.001 and clump_kb = 10,000. MR analysis was conducted using the TwoSampleMR package with several statistical methods, including Inverse Variance Weighted (IVW), Maximum Likelihood, MR-Egger, Weighted Median, and Weighted Mode, to assess the causal relationship between exposures and the outcome. To evaluate heterogeneity, we applied the mr_heterogeneity function using the IVW and MR-Egger methods and results with P > 0.05 indicated robust findings. Pleiotropy was assessed using the mr_pleiotropy_test function with the MR-Egger intercept test.

Immune landscape analysis

The immune cell infiltration analysis was conducted to assess the abundance of immune cells in the GSE30122 cohort using the R package “IOBR,” specifically employing the mcpcounter and xcell methods [24]. The immune cells with the same difference trend in the two methods were identified and subjected to Mendelian randomization analysis.

Association of adipokines-related DEGs and T cells

To further validate the association of these key DEGs and immune cells, Pearson correlation analysis was calculated for adipokines-related DEGs and T cells in GSE30122.

Relationship between core gene and Celastrol

Target genes of celastrol were predicted using SEA database (https://sea.bkslab.org/) to verify the relationship between drug and these key DEGs.

Statistical analysis

All analyses were performed using R version 4.3.2, and P < 0.05 was considered significantly different. MR analyses were performed based on the R package “TwoSampleMR”. The differential statistical analysis method is t-test, and the correlation analysis method is Pearson.

Results

Identification of DEGs and their association with lipid metabolism in DN

There were 249 and 156 DEGs in the GSE30528 (Fig. 1A) and GSE30122 (Fig. 1B), respectively. After intersecting these DEGs, 70 DEGs were acquired and shown in the Venn diagram (Fig. 1C). In the GSVA pathway enrichment score, Brown_and_beige_adipocyte_differentiation had a lower score in DN. However, the scores for Adipocytes_diabetic_condition and Adipocytes_normal_condition were higher in the DN samples (P < 0.05, Fig. 1D-E). At the same time, according to the results of Pearson correlation analysis, 6 genes (FGF9, MAGI2, RRM2, VCAN, CD163, and THBS2) were significantly related to the above three lipid metabolism pathways with |Correlation| < 0.6 and P < 0.05 (Fig. 1F).

Fig. 1
figure 1

Identification of DEGs and their association with lipid metabolism in DN

A-B. Volcano plot of different expressed genes (DEGs) between DN and normal in GSE30122 (A) and GSE30528 (B). C. Overlapping genes of DEGs in GSE30122 and GSE30528. D-E. Heatmap and barplot of GSVA scores. F. Correlation heatmap of 70 DEGs and lipid metabolism pathway

Full size image

MR analysis was performed on adipokines-related DEGs and their association with DN

The GWAS data of adipokines (ieu-a-1003, ieu-a-1, ebi-a-GCST90012034) and DN (ebi-a-GCST90018832) were retrieved from the IEU OpenGWAS (https://gwas.mrcieu.ac.uk/). All three adipokines (adiponectin, leptin, and resistin) were associated with DN (Fig. 2A). Figure 2B-D showed the SNP effects of adiponectin, leptin, and resistin on DN, analyzed using IVW, MR Egger, and Weighted Median methods. Adiponectin is negatively associated with DN risk (IVW, OR = 0.77, P = 0.005, Fig. 2B). The IVW method suggests that higher adiponectin levels may have a protective effect, but the MR Egger method did not find a significant association, indicating potential interference from horizontal pleiotropy. Leptin is positively associated with DN risk (MR Egger, OR = 1.92, P = 0.016, Fig. 2C). Resistin shows a strong association with DN risk (IVW, OR = 1.43, P < 0.001, Fig. 2D), with both the IVW and Weighted Median methods supporting that higher resistin levels may promote the progression of DN. The P values for heterogeneity and pleiotropy tests were greater than 0.05, indicating the reliability of the results (Table S1-2).

Fig. 2
figure 2

MR analysis of adipokines and DN

A. MR risk forest of adipokines and DN. B-D. MR SNP effect plots for adipokines and DN

Full size image

Furthermore, we performed Pearson correlation analysis on six genes (FGF9, MAGI2, RRM2, VCAN, CD163, and THBS2) and adipokines, which were significantly negatively correlated with adipokines (P < 0.05, Fig. 3A). To further verify the association between these six genes and DN, the above gene GWAS data (eqtl-a-ENSG00000186340, eqtl-a-ENSG00000102678, eqtl-a-ENSG00000187391, eqtl-a-ENSG00000171848, eqtl-a-ENSG00000038427, and eqtl-a-ENSG00000177575) were considered exposure factors. Using R TwoSampleMR MR analysis, MAGI2, FGF9, and THBS2 were associated with DN (Fig. 3B). SNP effect plots showed the relationship between these three genes and DN (Fig. 3C-E). MAGI2 was negatively correlated with the risk of DN (IVW, OR = 0.83, P = 0.001, Fig. 3C), FGF9 was negatively correlated with the risk of DN (IVW, OR = 0.89, P = 0.002, Fig. 3D), and THBS2 was positively correlated with the risk of DN (MR Egger, OR = 1.12, P = 0.003, Fig. 3E). The P values for both the simultaneous heterogeneity and pleiotropy tests were greater than 0.05, indicating the reliability of the results (Table S3-4).

Fig. 3
figure 3

MR analysis was performed on adipokines-related DEGs and their association with DN

(A) Association of adipokines and adipokines-related DEGs. (B) MR risk forest of adipokines-related DEGs and DN. C-E. MR SNP effect plots for adipokines-related DEGs and DN

Full size image

MR analysis of T cells and DN

The infiltration levels of immune cells in DN were further explored. Levels of T cells, monocytic lineage, and fibroblasts were significantly higher than those in normal controls in the mcpcounter method (P < 0.05, Fig. 4A), only T cells showed the same trend in the xcell method (Fig. 4B). The GWAS data of T cells (GCST90001620, GCST90001918) were downloaded from the GWAS Catalog database. T cell as the exposure factor, DN as the outcome factor, using R packages TwoSampleMR for MR analysis. Two types of T cells were related to DN (Fig. 4C). As shown in the SNP effect plot, HLA DR+ T cell%lymphocyte and DN risk were positively correlated (IVW, OR = 1.09, P = 0.032, Fig. 4D). CD45 on HLA DR+ T cells was positively correlated with the risk of DN (IVW, OR = 1.12, P = 0.020, Fig. 4E), which was consistent with the trend of T cell immune infiltration analysis. The P values for both the simultaneous heterogeneity and pleiotropy tests were greater than 0.05, indicating the reliability of the results (Table S5-6).

Fig. 4
figure 4

MR analysis of T cells and DN

A-B. Immune infiltration analysis of mcpcounter and xcell. C. MR risk forest of T cell and DN. D-E. MR SNP effect plots for T cell and DN

Full size image

Relationship between three adipokine-related genes and T cells

In addition, the identified three adipokine-related gene expression levels were validated in the GSE30122 dataset. As shown in Fig. 5A-C, THBS2 was highly expressed in DN, while FGF9 and MAGI2 were lowly expressed (P < 0.05), consistent with the trends observed in their MR analysis results. Pearson correlation analysis shows a positive correlation between THBS2 and T cells, while FGF9 and MAGI2 are negatively correlated with T cells (Fig. 5D-F).

Fig. 5
figure 5

Relationship of T cells and adipokines-related DEGs

A-C. Expression of FGF9, MAGI2, and THBS2 in GSE30122. D-F. Correlation of FGF9, MAGI2, and THBS2 with T cells

Full size image

Relationship of MAGI2, FGF9, THBS2 and Celastrol

Twenty-two target genes of celastrol, including MAGI2, FGF9 and THBS2(Fig. 6), were predicted using the SEA database (https://sea.bkslab.org/).

Fig. 6
figure 6

Target genes network of celastrol

A. Relationship network between celastrol and target genes(the green triangle is celastrol, and the orange circle is the target gene)

Full size image

Discussion

Adipokines are closely associated with the progression of DN [25]. This study investigated the potential causal role of adipokine-related genes in the pathogenesis of DN. We identified six genes significantly negatively correlated with adipokines—adiponectin, leptin, or resistin—in DN. Among them, MAGI2, FGF9, and THBS2 showed relationships with DN progression. Further immune infiltration and MR analysis revealed a strong association between T cells and DN risk, with significant correlations observed between the three identified genes and T cells.

First, pathway enrichment analysis indicated that the DN tissue had lower scores in brown and beige adipocyte differentiation. Brown adipose tissue is more energetically active than white adipose tissue in metabolic activities, with a greater number of mitochondria. Brown and beige adipocytes are not only thermogenic but also improve lipid metabolism through the secretion of specific factors [26]. Currently, activation of brown and beige cells is considered a potential target for treating obesity and T2D [26]. Our findings suggest that alterations in adipocyte differentiation may play a crucial role in DN by promoting lipid accumulation within renal tissue. Furthermore, our MR analysis revealed a link between specific adipokines—adiponectin, leptin, and resistin—and DN risk. Adiponectin is the most abundant adipokine in human plasma [27]. Previous studies have shown that low adiponectin is a major risk factor for metabolic disorders [28], and it may improve DN by inhibiting necrotic cell apoptosis [29]. Consistently, our findings indicate a negative correlation between adiponectin and DN risk. On the other hand, we found that both leptin and resistin are positively correlated with DN risk. Leptin levels are a risk factor for declining renal function [30]; leptin and resistin play crucial roles in insulin resistance-related kidney injury [31]. These findings provide further evidence for the distinct roles of adipokines in DN.

To further explore the specific mechanisms of lipid metabolism in DN, we identified 70 intersecting DEGs in the GSE30122 and GSE30528 datasets. Six genes—FGF9, MAGI2, RRM2, VCAN, CD163, and THBS2—were significantly associated with lipid metabolism pathways related to DN. Previous studies have shown that downregulation of FGF9 in primary hepatocytes promotes lipid accumulation in cells [32]; MAGI2, regulated by CLDN6, influences fatty acid synthesis metabolism [33]; inhibition of RRM2 limits the accumulation of intracellular toxic lipid peroxides [34]; VCAN is a lipid metabolism-related gene in gastric cancer [35]; CD163 is a marker for M2 macrophages, and macrophages are associated with lipid metabolism in T2D [36]; THBS2 has also been identified as a hub gene in T2D and is associated with lipid metabolism [37]. Our analysis also found a significant negative correlation between these six genes and adipokines, suggesting that as adipokine levels increase, the expression of these genes may decrease. Further MR analysis showed that FGF9 and MAGI2 are negatively correlated with DN risk, while THBS2 is positively correlated. It has been reported that FGF9 expression is significantly downregulated in high glucose-induced podocytes [38]. Currently, there is limited research on its specific role in DN. However, studies have shown that FGF9 suppresses the expression of genes involved in adipogenesis and increases the expression of genes involved in fatty acid oxidation [32]. We hypothesize that the downregulation of FGF9 may promote DN progression by facilitating lipid accumulation. Wang et al. confirmed the significant downregulation of MAGI2 in DN, although the specific mechanism remains to be studied [39]. Qiu et al. pointed out that in breast cancer, CLDN6 interacts with MAGI2 to prevent KLF5 from entering the nucleus, thereby inhibiting SREBF1 transcription and ultimately suppressing palmitic acid-induced RAS palmitoylation [33]. Our results suggest that MAGI2 may influence lipid metabolism regulation in DN through interactions with adipokines. In contrast, THBS2 is upregulated in DN and positively correlated with DN risk. Previous studies have shown a significant increase in THBS2 in T2D patients with nephropathy [40]. Mo et al. found that THBS2 knockdown inhibited apoptosis in high glucose-induced human renal proximal tubular cells [41]. Our findings provide potential mechanisms for the roles of these genes in DN, where the downregulation of MAGI2 and FGF9, as well as the upregulation of THBS2, may promote DN progression by influencing lipid metabolism.

In this study, we further investigated the role of celastrol in adipokine-mediated gene regulation. Using predictions from the SEA database, MAGI2, FGF9, and THBS2 were identified as potential target genes of celastrol. Our previous research demonstrated that celastrol significantly modulates adipokine levels, including adiponectin, leptin, and resistin, and influences key lipid metabolism pathways through the regulation of these adipokines [16]. These findings are consistent with the results of the present study. Based on the above results, we hypothesize that celastrol may regulate lipid accumulation and immune responses through its target genes, FGF9, and THBS2, promoting interactions between adipokines and their associated genes to play a critical role in the progression of DN.

In DN, dyslipidemia leads to glomerular injury and fibrosis, often accompanied by an inflammatory response [42]. Therefore, we conducted an immune infiltration analysis. Compared with normal controls, levels of T cells, monocyte lineage cells, and fibroblasts were higher in DN samples. Interestingly, GWAS data and MR analysis suggest a particular association between T cells and DN risk. T cell-mediated immunity promotes renal inflammation, fibrosis, and dysfunction, leading to the progression of chronic kidney disease [43]. Chen et al. indicated that dysregulation of immune checkpoint molecules on T cells disrupts renal homeostasis, triggering pathological inflammation and promoting DN progression [44]. Our findings did not show significant heterogeneity or multiplicity, strengthening the causal relationship between T cell infiltration and DN risk. Additionally, our data indicate a positive correlation between THBS2 expression and T cells, while FGF9 and MAGI2 exhibit a negative correlation. This is consistent with the above findings. Although there is currently rare research exploring the specific associations between these genes and T cells in DN, However, THBS2 may influence the expression of PI3K by upregulating the TGF-β signaling pathway [45]. Additionally, the TGF-β signaling pathway is closely related to vascular inflammation in diabetic nephropathy [46]. We speculate that increased expression of THBS2 in DN may promote T cell infiltration, thereby amplifying the immune response in DN. FGF9 can induce the phosphorylation of ERK and AKT, as well as the activation of CREB and Nrf2, thereby exerting antioxidant functions [38]. Wang et al. found that MAGI2 is negatively correlated with Tregs in DN [39]. Dysfunction of Tregs promotes autoimmunity and inflammation, and the imbalance between Th17 and Tregs contributes to the development of DN [47]. Decreased levels of FGF9 and MAGI2 may impair anti-inflammatory responses, exacerbating DN. The consistent trends observed in gene expression data and MR analysis support these explanations, suggesting that these genes may regulate T cell activity in DN.

However, there are a few limitations in this study that need to be addressed in future research. Firstly, the sample size of the dataset used in this study is relatively small, which may affect the reliability and generalizability of the results. Increasing the sample size in future studies could help strengthen the robustness of our findings. Secondly, this study is based on publicly available databases, which lack comprehensive patient clinical characteristics and blood sample data, including routine blood tests and adipokine levels. This limitation restricts our ability to validate the direct associations between adipokine-mediated pathways and DN progression in a clinical setting. Future studies incorporating patient-level clinical data and blood biomarkers would provide a more comprehensive understanding of these interactions. Thirdly, although our bioinformatics analyses provide valuable insights, experimental validation is necessary to confirm the accuracy and applicability of these findings in DN pathophysiology and potential therapeutic strategies.

Conclusion

In conclusion, this study provides evidence for the involvement of adipokine-related genes in DN and identifies their association with T cells. Our findings indicate that adipokine dysregulation may influence lipid metabolism pathways and lead to DN progression by promoting inflammation. Additionally, the celastrol’s target genes including MAGI2, FGF9, and THBS2 demonstrated significant correlations with T cells infiltrations. The relationship between adipokine-related genes, lipid metabolism, and T cells highlights their potential as therapeutic targets for DN. Importantly, the current study revealed the potential of celastrol as a novel therapeutic agent for DN, targeting adipokine-driven immune and metabolic dysregulation. Further research is imperatively needed to validate the therapeutic potential of targeting adipokine pathways and immune cells to mitigate DN progression.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

DN:

Diabetic nephropathy

T2D:

Type 2 diabetes

MR:

Mendelian randomization

DEGs:

Differentially expressed genes

GSVA:

Gene set variation analysis

References

  1. Yang L, Yang, et al. Lipid droplet - mitochondria coupling: A novel lipid metabolism regulatory hub in diabetic nephropathy. Front Endocrinol (Lausanne). 2022;13:1017387.

    Article  PubMed  Google Scholar 

  2. Zhou E-T. Prevalence of chronic kidney disease across levels of glycemia among adults in Pudong new area, Shanghai, China. BMC Nephrol. 2013;14:253.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Pereira C, Oliveira, et al. Metabolomics as a tool for the early diagnosis and prognosis of diabetic kidney disease. Med Res Rev. 2022;42(4):1518–44.

    Article  PubMed  Google Scholar 

  4. Zhu, Zhong, Liang, et al. Comprehensive analysis of fatty acid metabolism in diabetic nephropathy from the perspective of immune landscapes, diagnosis and precise therapy. J Inflamm Res. 2024;17:693–710.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Wang L, Wu, et al. Lipid homeostasis in diabetic kidney disease. Int J Biol Sci. 2024;20(10):3710–24.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Njeim A, Fornoni. Unraveling the crosstalk between lipids and NADPH oxidases in diabetic kidney disease. Pharmaceutics, 2023. 15(5).

  7. Kang A, Choi, et al. Defective fatty acid oxidation in renal tubular epithelial cells has a key role in kidney fibrosis development. Nat Med. 2015;21(1):37–46.

    Article  CAS  PubMed  Google Scholar 

  8. Savelieff C, Feldman. The emerging role of dyslipidemia in diabetic microvascular complications. Curr Opin Endocrinol Diabetes Obes. 2020;27(2):115–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Han D, Zhu, et al. Lipid metabolism disorder in diabetic kidney disease. Front Endocrinol (Lausanne). 2024;15:1336402.

    Article  PubMed  Google Scholar 

  10. Landecho T. Valentí, Relevance of leptin and other adipokines in Obesity-Associated cardiovascular risk. Nutrients, 2019. 11(11).

  11. Kawano A, Ishizu, et al. Molecular composition of adiponectin in urine is a useful biomarker for detecting early stage of diabetic kidney disease. J Med Invest. 2023;70(34):464–70.

    Article  PubMed  Google Scholar 

  12. Kim L, Kim, et al. The adiponectin receptor agonist adiporon ameliorates diabetic nephropathy in a model of type 2 diabetes. J Am Soc Nephrol. 2018;29(4):1108–27.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Huang P, Dong, Tao, and, Yang. The association between insulin resistance, leptin, and resistin and diabetic nephropathy in type 2 diabetes mellitus patients with different body mass indexes. Diabetes Metab Syndr Obes. 2021;14:2357–65.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Peng H, Zou, et al. Roles of plasma leptin and resistin in novel subgroups of type 2 diabetes driven by cluster analysis. Lipids Health Dis. 2022;21(1):7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Abu Bakar, Shariff T, Lee. Celastrol attenuates inflammatory responses in adipose tissues and improves skeletal muscle mitochondrial functions in high fat diet-induced obese rats via upregulation of AMPK/SIRT1 signaling pathways. Eur J Pharmacol. 2020;883:173371.

    Article  PubMed  Google Scholar 

  16. Abu Bakar M, Khalid N, Shahril. Shariff, and Karunakaran, Celastrol attenuates high-fructose diet-induced inflammation and insulin resistance via Inhibition of 11β-hydroxysteroid dehydrogenase type 1 activity in rat adipose tissues. BioFactors. 2022;48(1):111–34.

    Article  CAS  PubMed  Google Scholar 

  17. Kang, Kim, Kim, et al. Celastrol attenuates adipokine resistin-associated matrix interaction and migration of vascular smooth muscle cells. J Cell Biochem. 2013;114(2):398–408.

    Article  CAS  PubMed  Google Scholar 

  18. Nie F, Zhang, et al. Celastrol slows the progression of early diabetic nephropathy in rats via the PI3K/AKT pathway. BMC Complement Med Ther. 2020;20(1):321.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Tang W, Tang, et al. Celastrol attenuates diabetic nephropathy by upregulating SIRT1-mediated Inhibition of EZH2related wnt/β-catenin signaling. Int Immunopharmacol. 2023;122:110584.

    Article  CAS  PubMed  Google Scholar 

  20. Wang A, Bakar K, et al. Renoprotective mechanisms of Celastrol in high glucose-mediated HK-2 cell injury through Inhibition of the PI3K/Akt/NF-κB signalling pathway. Biochem Biophys Rep. 2025;41:101928.

    CAS  PubMed  PubMed Central  Google Scholar 

  21. Hemani Z. Elsworth, The MR-Base platform supports systematic causal inference across the human phenome. Elife, 2018. 7.

  22. Feng G. IL1R1 is required for Celastrol’s leptin-sensitization and antiobesity effects. Nat Med. 2019;25(4):575–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Gu S, Zhu, et al. Celastrol functions as an emerging manager of lipid metabolism: mechanism and therapeutic potential. Biomed Pharmacother. 2023;164:114981.

    Article  CAS  PubMed  Google Scholar 

  24. Meylan B, Sautès-Fridman, et al. webMCP-counter: a web interface for transcriptomics-based quantification of immune and stromal cells in heterogeneous human or murine samples. BioRxiv; 2020. p. 400754.

  25. Maximus A, Achkar. Hamid, Hasnain, and Peralta, Adipocytokines: are they the theory of everything??. Cytokine. 2020;133:155144.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Lizcano. The beige adipocyte as a therapy for metabolic diseases. Int J Mol Sci. 2019;20:5058.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Gu. Biomarkers of adiponectin: plasma protein variation and genomic DNA polymorphisms. Biomark Insights. 2009;4:123–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Heidari N, Nasri. Adiponectin and chronic kidney disease; a review on recent findings. J Nephropharmacol. 2015;4(2):63–8.

    CAS  PubMed  PubMed Central  Google Scholar 

  29. Yi, OuYang. Adiponectin improves diabetic nephropathy by inhibiting necrotic apoptosis. Arch Med Sci. 2019;15(5):1321–8.

    Article  CAS  PubMed  Google Scholar 

  30. Hanai B. Association of serum leptin levels with progression of diabetic kidney disease in patients with type 2 diabetes. Diabetes Care. 2011;34(12):2557–9.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Kang, Cha. The role of visfatin in diabetic nephropathy. Chonnam Med J. 2011;47(3):139–43.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Zhao Z, Zhang, et al. FGF9 alleviates the fatty liver phenotype by regulating hepatic lipid metabolism. Front Pharmacol. 2022;13:850128.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Jin Q, Zhang, et al. CLDN6 inhibits breast cancer growth and metastasis through SREBP1-mediated RAS palmitoylation. Cell Mol Biol Lett. 2024;29(1):112.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Tarangelo R, Kim, et al. Nucleotide biosynthesis links glutathione metabolism to ferroptosis sensitivity. Life Sci Alliance. 2022;5(4):e202101157.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Yang C. In Silico development and validation of a novel glucose and lipid metabolism-related gene signature in gastric cancer. Transl Cancer Res. 2022;11(7):1977–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Wang, Guo, Xu, et al. HIIT ameliorates inflammation and lipid metabolism by regulating macrophage polarization and mitochondrial dynamics in the liver of type 2 diabetes mellitus mice. Metabolites. 2022;13(1):14.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Chen D. ANGPTL4 May regulate the crosstalk between intervertebral disc degeneration and type 2 diabetes mellitus: A combined analysis of bioinformatics and rat models. J Inflamm Res. 2023;16:6361–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Chen, Li C, Jiang. Identification and validation of immune-related biomarkers and potential regulators and therapeutic targets for diabetic kidney disease. BMC Med Genomics. 2023;16(1):90.

    Article  Google Scholar 

  39. Wang Z, Lu Z, Mao. Unveiling diabetic nephropathy: a novel diagnostic model through single-cell sequencing and co-expression analysis. Aging. 2024;16(13):10972–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Yeh C, Chuang, et al. Differentiation of type 2 diabetes mellitus with different complications by proteomic analysis of plasma low abundance proteins. J Diabetes Metab Disord. 2015;15:24.

    Article  PubMed  Google Scholar 

  41. Mo F, Fu H, Li. MicroRNA 1228 mediates the viability of high Glucose-Cultured renal tubule cells through targeting thrombospondin 2 and PI3K/AKT signaling pathway. Kidney Blood Press Res. 2022;47(1):1–12.

    Article  CAS  PubMed  Google Scholar 

  42. Jin L, Qiao, et al. Oxidative stress and inflammation in diabetic nephropathy: role of polyphenols. Front Immunol. 2023;14:1185317.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Liu, Lv, Zhang, et al. T cells and their products in diabetic kidney disease. Front Immunol. 2023;14:1084448.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Chen, Tang Z, et al. A potential defensive role of TIM-3 on T lymphocytes in the inflammatory involvement of diabetic kidney disease. Front Immunol. 2024;15:1365226.

    Article  Google Scholar 

  45. Wang Z, Chen, Chen, Zou. Bioinformatics analysis identifies COL1A1, THBS2 and SPP1 as potential predictors of patient prognosis and immunotherapy response in gastric cancer. Biosci Rep. 2021;41(1):BSR20202564.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Elmarakby, Sullivan. Relationship between oxidative stress and inflammatory cytokines in diabetic nephropathy. Cardiovasc Ther. 2012;30(1):49–59.

    Article  CAS  PubMed  Google Scholar 

  47. Wang, Zhang, Si, et al. Dapagliflozin reverses the imbalance of T helper 17 and T regulatory cells by inhibiting SGK1 in a mouse model of diabetic kidney disease. FEBS Open Bio. 2021;11(5):1395–405.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This work was financially supported by the Fundamental Research Grant Scheme, Ministry of Higher Education, Malaysia (Ref No.: FRGS/1/2024/STG01/USM/02/1) and Shandong Province Traditional Chinese Medicine Science and Technology Project (Ref No.: M-2023158).

Author information

Authors and Affiliations

  1. Department of Pharmacy, Taishan Vocational College of Nursing, Tai’an, Shandong, 271099, China

    Xiaojuan Wang, Jing Wang & Manli Xu

  2. Bioprocess Technology Division, School of Industrial Technology, Universiti Sains Malaysia, Gelugor, Penang, 11800, Malaysia

    Xiaojuan Wang, Mohamad Hafizi Abu Bakar & Mohd Asyraf Kassim

  3. School of Materials & Mineral Resources Engineering, Universiti Sains Malaysia, Nibong Tebal, Penang, 14300, Malaysia

    Khairul Anuar Shariff

Authors
  1. Xiaojuan Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Mohamad Hafizi Abu Bakar
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mohd Asyraf Kassim
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Khairul Anuar Shariff
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Jing Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Manli Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Xiaojuan Wang: Writing - original draft, Methodology, Investigation, Formal analysis and Data curation. Mohamad Hafizi Abu Bakar: Supervision, Conceptualization, Methodology, Writing - review & editing, Validation and Funding acquisition. Mohd Asyraf Kassim: Conceptualization, Resources, Writing - review & editing. Khairul Anuar Shariff: Conceptualization, Formal analysis, Resources, Writing - review & editing.JingWang and Manli Xu were responsible for the section of the paper related to the organization, analysis, and statistics of the data. All authors reviewed the manuscript.

Corresponding author

Correspondence to Mohamad Hafizi Abu Bakar.

Ethics declarations

Ethics approval and consent to participate

The data for this study were sourced from the public GEO database, and these datasets have obtained ethical approval from GEO and are de-identified. Therefore, ethical approval is not required for this study.

Consent for publication

All authors agreed to submit the manuscript.

Competing interests

The authors declare no competing interests.

Consent to participate

Not applicable.

Additional information

Publisher’s note

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

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, X., Abu Bakar, M., Kassim, M.A. et al. Exploring the interplay between adipokine-mediated celastrol target genes and T cells in diabetic nephropathy: a mendelian randomization-based causal inference. Diabetol Metab Syndr 17, 89 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13098-025-01665-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13098-025-01665-0

Keywords

Diabetology & Metabolic Syndrome

ISSN: 1758-5996

Contact us