Introduction: IgA nephropathy (IgAN) is a leading cause of end-stage renal disease. The exact pathogenesis of IgAN is not well defined, but some genetic studies have led to a novel discovery that the (immuno)proteasome probably plays an important role in IgAN. Methods: We firstly analyzed the association of variants in the UBE2L3 region with susceptibility to IgAN in 3,495 patients and 9,101 controls, and then analyzed the association between lead variant and clinical phenotypes in 1,803 patients with regular follow-up data. The blood mRNA levels of members of the ubiquitin-proteasome system including UBE2L3 were analyzed in peripheral blood mononuclear cells from 53 patients and 28 healthy controls. The associations between UBE2L3 and the expression levels of genes involved in Gd-IgA1 production were also explored. Results: The rs131654 showed the most significant association signal in UBE2L3 region (OR: 1.10, 95% CI: 1.04–1.16, p = 2.29 × 10−3), whose genotypes were also associated with the levels of Gd-IgA1 (p = 0.04). The rs131654 was observed to exert cis-eQTL effects on UBE2L3 in various tissues and cell types, particularly in immune cell types in multiple databases. The UBE2L3, LUBAC, and proteasome subunits were highly expressed in patients compared with healthy controls. High expression levels of UBE2L3 were not only associated with higher proteinuria (r = 0.34, p = 0.01) and lower eGFR (r = −0.28, p = 0.04), but also positively correlated with the gene expression of LUBAC and other proteasome subunits. Additionally, mRNA expression levels of UBE2L3 were also positively correlated with IL-6 and RELA, but negatively correlated with the expression levels of the key enzyme in the process of glycosylation including C1GALT1 and C1GALT1C1. Conclusion: In conclusion, by combined genetic association and differed expression analysis of UBE2L3, our data support a role of genetically conferred dysregulation of the (immuno)proteasome in regulating galactose-deficient IgA1 in the development of IgAN.

IgA nephropathy (IgAN) is a leading cause of chronic kidney disease and end-stage renal disease. The exact pathogenesis of IgAN is not well defined, but some genetic studies have led to a novel discovery that the (immuno)proteasome probably plays an important role in IgAN [1, 2].

Hypothesis-free GWASs indicated that PSMB8 gene on chromosome 6 was one of the independent signals associated with IgAN [3]. The proteasome subunit beta type-8 (LMP7) is the proteasome β5i subunit whose expression is elevated in peripheral blood mononuclear cells (PBMCs) of patients with IgAN [1]. The GWAS for serum galactose-deficient IgA1 identified the top single-nucleotide polymorphism in the HECW1 gene that encodes HECT, C2, and WW domain-containing protein 1 [4]. Additionally, it was reported that HECW1 mRNA levels were negatively correlated with Gd-IgA1 levels in patients with IgAN [5].

Another important contribution by GWAS was identification of shared genetic architecture of IgAN with multiple autoimmune and inflammatory diseases, suggesting shared pathogenic pathways with IgAN. In a shared genetics study between IgAN and systemic lupus erythematosus, we previously observed variants of UBE2L3, a gene involved in ubiquitin/proteasome pathway, which were associated with susceptibility to both autoimmune diseases [6]. UBE2L3 is an E2 ubiquitin-conjugating enzyme that has a particular affinity for RING-in-between-RING E3 ligases, including HOIL-1L and HOIP, components of the linear ubiquitin chain assembly complex (LUBAC). The LUBAC (HOIL-1L, HOIP, and SHARPIN) is critical for the efficient activation of nuclear factor kappa B (NF-κB) signaling by linear ubiquitination of NEMO [7]. UBE2L3 has been further functionally validated as involved in systemic lupus erythematosus by amplifying NF-κB activation and promoting plasma cell development. It was observed that UBE2L3 was the preferred E2 conjugating enzyme for LUBAC in vivo, which limited the specificity in LUBAC-mediated NF-κB activation [7‒10]. It was also suggested that the activation of NF-κB played a key role in release of cytokines such as interleukin (IL)-6-related cytokines which mediate overproduction of Gd-IgA1 [11, 12].

These studies suggested that the ubiquitin-proteasome pathway may be involved in the development of IgAN, but there is no relevant research on the specific mechanisms. We hypothesize that in patients with IgAN, LUBAC associated with ubiquitin-conjugating enzyme E2 encoded by the UBE2L3 gene can activate IKK by ubiquitinating NEMO, which leads to the activation of the NF-κB signaling pathway, further leads to elevated levels IL-6, and ultimately increases the production of Gd-IgA1. In this study, we sought to elucidate the role of the ubiquitin-proteasome system in the development of IgAN. We firstly identified the possible independent pathogenicity variant for UBE2L3 and the association between genotypes and clinical phenotypes by genetic association analysis. Then, we confirmed whether there were abnormalities in the ubiquitin-proteasome system in IgAN and explored its involvement in the potential pathogenesis of the IgAN through differed mRNA expression analysis.

Study Participants

To test genetic associations with susceptibility to IgAN, 3,495 patients with IgAN and 9,101 healthy controls of Chinese Han ancestry were enrolled in this study. All of the patients were confirmed by renal biopsy. Patients with IgAN secondary to Henoch-Schönlein purpura, lupus nephritis, chronic liver diseases, and other immunologic disorders were excluded. Among these patients, 1,803 patients were regularly followed (PKU-IgAN cohort) from 1997 to 2020. The baseline demographic and clinical information of follow-up patients with IgAN was collected at the time of diagnosis. Written informed consent was obtained from each patient. This study complied with the Declaration of Helsinki and was approved by the medical Ethics Committee of Peking University First Hospital (IRB number: 2023-377-001).

Genotyping

Among the 3,495 patients and 9,101 healthy controls, 1,171 patients and 891 healthy individuals were genotyped with the Illumina 610-Quad BeadChip, 485 patients and 3,866 healthy individuals were genotyped with the Illumina Infinium OmniZhongHua-8 v1.3 array, 1,110 patients and 1,240 healthy individuals were genotyped with Infinium Global Screening Array-24 v1.0 BeadChip, 312 patients and 1,545 healthy individuals were genotyped with Infinium Global Screening Array-24 v2.0 BeadChip, as well as 417 patients and 1,559 healthy individuals were genotyped with the Infinium Chinese Genotyping Array-24 v1.0 BeadChip. The details for the recruitment of patients, genotyping, and genotype quality control were as we previously reported [13]. The location of the UBE2L3 gene is mapped to chromosome 22, position 21,903,736–21,978,323 (hg19, data obtained from Ensembl database). We centered our study on the variants across region 21.80 Mb–22.0 Mb on chromosome 22 spanning UBE2L3. To ensure reliability, SNPs were excluded if they showed either a call rate <95%, minor allele frequency < 0.01, or significant departure from Hardy-Weinberg equilibrium (p < 1.00 × 10−4).

Gene Expression Data

We checked correlations between genotypes and gene expressions in cis by querying the eQTLGen Consortium data [14], the QTLbase database [15], and ImmuNexUT database [16]. The results of the eQTLGen Consortium data are a meta-analysis performed on blood samples from 31,684 individuals. The QTLbase database is a database comprising genome-wide QTL summary statistics across over 95 tissue/cell types, to narrow down potential target genes. The ImmuNexUT is a dataset consisting of 28 distinct immune cell subsets from 337 patients with immune-mediated diseases and 79 healthy volunteers.

For differential gene expression analysis, we collected peripheral blood samples from 53 IgAN patients and 28 healthy donors. PBMCs were isolated from 10 mL EDTA blood by Ficoll-Paque density gradient centrifugation (GE HealthCare). Total RNA was extracted from PBMCs using TRIzol reagent (Invitrogen). Double-stranded cDNA was synthesized using an Illumina TotalPrep RNA Amplification Kit (Invitrogen). The quantification of gene expression was tested using the Affymetrix PrimeView Human Gene Expression Array.

Statistical Analysis

The association of SNPs with IgAN in the UBE2L3 region was analyzed using logistic regression models as implemented in PLINK v1.9 [17], assuming additive effects. Covariates of the first two principal components were included for adjustment as previously reported [13]. Finally, we combined single SNP association results across different batches via a standard error-weighted meta-analysis by using METAL [18]. LocusZoom [19] was used to provide regional visualization of results. The interactions were predicted by STRING database (https://cn.string-db.org/).

Baseline sociodemographic and clinical characteristics were summarized as mean ± SD or median (interquartile range) for continuous variables. Categorical variables were presented as frequencies (percent). The associations between the genotype and clinical characteristics were analyzed by using one-way ANOVA (continuous variables with normal distribution), Kruskal-Wallis test (continuous variables with nonnormal distribution), or χ2 test (categorical variables). We used the correlation analysis (Pearson correlation for normally continuous variables and Spearman correlation for nonnormally continuous variables) to analyze the associations among the expression of different genes and between gene expression levels and clinical phenotypes. If a gene has multiple probes, the mean expression of all probes of the same gene was calculated as the expression of each gene when analyzing the correlation between gene expressions. The renal survival time from the outcome event was calculated from the biopsy to the last follow-up. Follow-up time was censored if the patient was lost to follow-up. The outcome event was end-stage kidney disease (ESKD). ESKD was defined as estimated glomerular filtration rate (eGFR) less than 15 mL/min/1.73 m2, dialysis, or kidney transplantation. Survival probability without ESKD was analyzed by Kaplan-Meier curves for censored data. The Cox proportional hazards regression model was used to generate estimates of predicted risk of ESKD.

Statistical analyses were performed by SPSS Statistics version 26.0 (IBM Corporation) and GraphPad Prism 5 (GraphPad Software). A two-tailed p < 0.05 was considered statistically significant.

UBE2L3 Gene Polymorphisms Associated with Susceptibility to IgAN

A total of 128 SNPs were analyzed in the current association study. As can be seen from online supplementary Table S1 (for all online suppl. material, see https://doi.org/10.1159/000537987), among 128 SNPs investigated, 19 SNPs within the UBE2L3 gene locus were associated with susceptibility to IgAN. The rs131654[T] (p = 2.29 × 10−3, OR: 1.10, 95% CI: 1.04–1.16) showed the most significant association signal in this region (Fig. 1 a).

Fig. 1.

Meta-analysis results of the association studies of variants in the UBE2L3 region with IgAN and the cis-eQTL effect of the lead SNP. a Regional plot for IgAN association in the UBE2L3 region. b The eQTL effects within a 10-Mb region centered on rs131654 (22:21917190, hg19) in different tissue/cell types. Each row represents one tissue/cell type, and each column denotes one gene. The color from red to white indicates the levels of statistical significance expressed as –log10 (p value). c The eQTL effects of rs131654 on UBE2L3 in 28 distinct immune cell subsets.

Fig. 1.

Meta-analysis results of the association studies of variants in the UBE2L3 region with IgAN and the cis-eQTL effect of the lead SNP. a Regional plot for IgAN association in the UBE2L3 region. b The eQTL effects within a 10-Mb region centered on rs131654 (22:21917190, hg19) in different tissue/cell types. Each row represents one tissue/cell type, and each column denotes one gene. The color from red to white indicates the levels of statistical significance expressed as –log10 (p value). c The eQTL effects of rs131654 on UBE2L3 in 28 distinct immune cell subsets.

Close modal

Genotype-Phenotype Association Analysis

Genotype-phenotype association analysis was computed assuming an additive model. The clinical parameters included mean arterial pressure, eGFR calculated based on the Chronic Kidney Disease Epidemiology Collaboration formula [20], microhematuria, 24-h proteinuria, serum uric acid (UA), serum IgA levels, serum Gd-IgA1 levels, and Oxford MEST-C scores. It was observed that the genotypes of rs131654 were associated with the levels of Gd-IgA1 (GG vs. GT vs. TT) 79.65 (73.20–87.39) versus 80.65 (73.57–88.76) versus 79.38 (71.80–86.58; p = 0.04), in spite it did not survive the multiple-testing correction (Table 1).

Table 1.

The clinical characteristics of follow-up patients by rs131654 genotype

CharacteristicsGG (n = 414)GT (n = 874)TT (n = 514)p value
Age, years 35.56±12.60 35.38±12.02 35.75±11.92 0.86 
Male, n (%) 234 (56.5) 425 (48.6) 249 (48.4) 0.02 
MAP, mm Hg 94.23±11.26 94.14±11.96 94.21±12.49 0.99 
eGFR, mL/min per 1.73 m2 80.61 (55.49–106.13) 81.80 (52.94–105.62) 83.50 (55.21–107.00) 0.87 
Microhematuria, μL 34.40 (9.00–95.50) 35.00 (9.50–110.00) 25.00 (8.00–110.00) 0.76 
UTP, g/days 1.31 (0.72–2.78) 1.19 (0.62–2.30) 1.28 (0.65–2.38) 0.24 
UA, μmol/L 381.47±105.33 373.79±101.76 377.81±103.03 0.44 
IgA, g/L 3.24±1.24 3.28±1.15 3.32±1.22 0.58 
Gd-IgA1, U/mL 79.65 (73.20–87.39) 80.65 (73.57–88.76) 79.38 (71.80–86.58) 0.04 
C3, g/L 0.99±0.23 1.02±0.24 1.00±0.23 0.19 
Renal biopsy, n (%) 
 M1 165 (41.6) 317 (37.7) 201 (41.1) 0.3 
 E1 130 (32.7) 256 (30.4) 161 (32.9) 0.56 
 S1 258 (65.0) 556 (66.1) 313 (64.0) 0.73 
 T1 and T2 118 (29.8) 262 (31.1) 156 (31.9) 0.91 
 C1 and C2 234 (59.0) 491 (58.4) 292 (59.7) 0.99 
Therapy, n (%) 
 ACEI/ARB 392 (94.7) 824 (94.4) 484 (94.3) 0.97 
 Corticosteroids/cytotoxic drugs 215 (51.9) 462 (52.9) 286 (55.8) 0.46 
 Follow-up, month 51.50 (27.00–91.50) 53.00 (29.00–93.02) 52.00 (29.00–92.25) 0.71 
 ESKD, n (%) 45 (10.9) 99 (11.3) 55 (10.7) 0.93 
CharacteristicsGG (n = 414)GT (n = 874)TT (n = 514)p value
Age, years 35.56±12.60 35.38±12.02 35.75±11.92 0.86 
Male, n (%) 234 (56.5) 425 (48.6) 249 (48.4) 0.02 
MAP, mm Hg 94.23±11.26 94.14±11.96 94.21±12.49 0.99 
eGFR, mL/min per 1.73 m2 80.61 (55.49–106.13) 81.80 (52.94–105.62) 83.50 (55.21–107.00) 0.87 
Microhematuria, μL 34.40 (9.00–95.50) 35.00 (9.50–110.00) 25.00 (8.00–110.00) 0.76 
UTP, g/days 1.31 (0.72–2.78) 1.19 (0.62–2.30) 1.28 (0.65–2.38) 0.24 
UA, μmol/L 381.47±105.33 373.79±101.76 377.81±103.03 0.44 
IgA, g/L 3.24±1.24 3.28±1.15 3.32±1.22 0.58 
Gd-IgA1, U/mL 79.65 (73.20–87.39) 80.65 (73.57–88.76) 79.38 (71.80–86.58) 0.04 
C3, g/L 0.99±0.23 1.02±0.24 1.00±0.23 0.19 
Renal biopsy, n (%) 
 M1 165 (41.6) 317 (37.7) 201 (41.1) 0.3 
 E1 130 (32.7) 256 (30.4) 161 (32.9) 0.56 
 S1 258 (65.0) 556 (66.1) 313 (64.0) 0.73 
 T1 and T2 118 (29.8) 262 (31.1) 156 (31.9) 0.91 
 C1 and C2 234 (59.0) 491 (58.4) 292 (59.7) 0.99 
Therapy, n (%) 
 ACEI/ARB 392 (94.7) 824 (94.4) 484 (94.3) 0.97 
 Corticosteroids/cytotoxic drugs 215 (51.9) 462 (52.9) 286 (55.8) 0.46 
 Follow-up, month 51.50 (27.00–91.50) 53.00 (29.00–93.02) 52.00 (29.00–92.25) 0.71 
 ESKD, n (%) 45 (10.9) 99 (11.3) 55 (10.7) 0.93 

MAP, mean arterial pressure; eGFR, estimated glomerular filtration rate; UTP, urine protein; UA, uric acid; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; ESKD, end-stage kidney disease.

cis-eQTL and UBE2L3 Expression-Phenotype Analyses

The top signal, rs131654, is an intronic variant in UBE2L3, which encodes ubiquitin-conjugating enzyme E2 L3 that plays a major role in determining the formation length and connection type of ubiquitin chains. We annotated all SNPs in LD (r2 > 0.8) with rs131654 by HaploReg v4.2 [21] and found no coding variant. The eQTLGen Consortium data showed that the rs131654 was a cis-eQTL for UBE2L3 (p = 4.24 × 10−143). We then queried the QTLbase [15], within 10-Mb region centered on rs131654; we noted that this SNP exerted cis-eQTL effects on UBE2L3 in different tissues (Fig. 1b). In addition, we also found that rs131654 exerted cis-eQTL effects on UBE2L3 in diverse immune cell types by querying the ImmuNexUT database [16] (Fig. 1c). Genotypes from eQTL study [22] showed that the rs140498 risk allele, which was in strong linkage disequilibrium with rs131654 (r2 = 0.8), demonstrated a robust correlation with increased UBE2L3 expression in EBV-transformed lymphoblastoid cell lines obtained from HapMap individuals (p = 1.02 × 1−7). A strong linear relationship between alleles of rs140498 and UBE2L3 expression was similarly observed in primary human B cells (p = 4.81 × 10−6) [23].

We then assessed the expression of UBE2L3 in PBMCs from IgAN patients and healthy donors (Fig. 2a). The UBE2L3 expression in IgAN patients was significantly higher than those in healthy controls (IgAN vs. controls, for the probe with maximal mean expression 8.73 ± 0.12 vs. 8.66 ± 0.13, p = 0.02; for the mean expression of all probes of the UBE2L3 gene 6.49 ± 0.17 vs. 6.40 ± 0.14, p = 0.02). In addition, we then analyzed the differential expression of other E2 ubiquitin-binding enzymes that may be involved in the NF-κB signaling pathway (UBE2D1, UBE2D2, UBE2D3, and UBE2N) [24, 25]. Based on our gene expression profiling data, no differences were found in the expression of these genes in PBMCs from IgAN patients and healthy controls (Fig. 2a).

Fig. 2.

Differential expression of UBE2L3 and its association with clinical phenotypes and prognosis. a The differential expression of E2 ubiquitin-binding enzymes in IgAN patients and healthy controls; the probe_11754929_s_at is the one with the maximal mean expression of UBE2L3; the probe_average shows the mean expression of all probes of a specific gene. b The correlation analysis of UBE2L3 expression with proteinuria and eGFR in IgAN patients. c The Kaplan-Meier analysis demonstrated patients with higher levels of the UBE2L3 expression have a worse prognosis.

Fig. 2.

Differential expression of UBE2L3 and its association with clinical phenotypes and prognosis. a The differential expression of E2 ubiquitin-binding enzymes in IgAN patients and healthy controls; the probe_11754929_s_at is the one with the maximal mean expression of UBE2L3; the probe_average shows the mean expression of all probes of a specific gene. b The correlation analysis of UBE2L3 expression with proteinuria and eGFR in IgAN patients. c The Kaplan-Meier analysis demonstrated patients with higher levels of the UBE2L3 expression have a worse prognosis.

Close modal

Next, we analyzed the association between the expression of UBE2L3 and clinical phenotypes using correlation analysis. In patients with IgAN, we found that the elevated levels of expression of the UBE2L3 gene (probe_11754929_s_at with maximal mean expression) were correlated with higher levels of proteinuria (r = 0.34, p = 0.01) and lower levels of eGFR (r = −0.28, p = 0.04), which showed that higher levels of UBE2L3 expression in PBMCs were associated with a poorer renal function (Fig. 2b). Of the 53 patients with IgAN, we obtained follow-up information for 32 patients. We divided the 32 patients into quintiles based on the expression of the UBE2L3 gene. We found that the patients with the top 20% UBE2L3 expression levels showed a marginal significance with a higher risk of worse prognosis compared to the rest (HR: 4.16 [95% CI: 0.92–18.77], p = 0.06; log-rank test: p = 0.04, Fig. 2c).

Expression Analysis on Genes Involved in Ubiquitin-Proteasome Pathway

In addition to the high expression of the UBE2L3 gene, we found that SHARPIN (patients vs. controls 6.16 ± 0.20 vs. 6.04 ± 0.17, p = 9.00 × 10−3), RNF31 (HOIP) (6.45 ± 0.23 vs. 6.32 ± 0.18, p = 0.01), and RBCK1 (HOIL-1L) (6.23 ± 0.23 vs. 6.09 ± 0.20, p = 8.00 × 10−3) were also highly expressed in IgAN patients compared to the healthy controls (Fig. 3a), and the proteins encoded by these three genes made up a ubiquitin ligase complex called LUBAC that could activate the NF-κB signaling pathway. At the same time, the UBE2L3 gene expression was positively correlated with the SHARPIN gene expression (r = 0.77, p = 1.43 × 10−11), RNF31 gene expression (r = 0.46, p = 1.00 × 10−3), and RBCK1 gene expression (r = 0.38, p = 5.00 × 10−3) (Fig. 3b). STRING database confirmed significant interactions between UBE2L3 and components of LUBAC (Fig. 3c). Because the ubiquitin-conjugating enzyme E2 L3 protein was involved in the ubiquitin-proteasome pathway, we also checked the expression levels of key genes that could reflect whether the ubiquitin-proteasome pathway was activated. Genes related to the ubiquitin-proteasome pathway (PSMB1, PSMB2, PSMB3, PSMB4, PSMB5, PSMB6, PSMB7, PSMB8, PSMB10) were expressed at higher levels in IgAN patients than in healthy controls (Fig. 4a), which showed significant ubiquitin-proteasome pathway activation in PBMCs of IgAN patients. In addition, the elevated expression levels of the UBE2L3 gene were positively correlated with activation of the ubiquitin-proteasome pathway (Fig. 4b). The above data showed that the ubiquitin-proteasome pathway was activated in patients with IgAN. Table 2 shows the members related to the ubiquitin-proteasome pathway that may be involved in the development of IgAN.

Fig. 3.

The expression levels of the LUBAC (HOIL-1L, HOIP, and SHARPIN) and its association with UBE2L3. a The differential expression of the LUBAC (HOIL-1L, HOIP, and SHARPIN) in IgAN patients and healthy controls; these differences are statistically significant (p < 0.05). b Correlation analysis between the expression levels of the LUBAC (HOIL-1L, HOIP, and SHARPIN) and UBE2L3 in IgAN patients. c The interaction between UBE2L3, LUBAC, and NEMO. Nodes represent genes, and edges denote protein-protein interaction in STRING database.

Fig. 3.

The expression levels of the LUBAC (HOIL-1L, HOIP, and SHARPIN) and its association with UBE2L3. a The differential expression of the LUBAC (HOIL-1L, HOIP, and SHARPIN) in IgAN patients and healthy controls; these differences are statistically significant (p < 0.05). b Correlation analysis between the expression levels of the LUBAC (HOIL-1L, HOIP, and SHARPIN) and UBE2L3 in IgAN patients. c The interaction between UBE2L3, LUBAC, and NEMO. Nodes represent genes, and edges denote protein-protein interaction in STRING database.

Close modal
Fig. 4.

Expression levels of the proteasome subunits and their association with UBE2L3. a The differential expression of the proteasome subunits in IgAN patients and healthy controls; these differences are statistically significant (p < 0.05). b Correlation analysis between the expression levels of the proteasome subunits and UBE2L3 in IgAN patients.

Fig. 4.

Expression levels of the proteasome subunits and their association with UBE2L3. a The differential expression of the proteasome subunits in IgAN patients and healthy controls; these differences are statistically significant (p < 0.05). b Correlation analysis between the expression levels of the proteasome subunits and UBE2L3 in IgAN patients.

Close modal
Table 2.

Members related to the ubiquitin-proteasome pathway that may be involved in the development of IgAN

CategoryMember
Ubiquitin-conjugating enzyme E2 UBE2L3 
Ubiquitin ligase E3 LUBAC (HOIP, HOIL-1L, and SHARPIN) 
Proteasome PSMB1, PSMB2, PSMB3, PSMB4, PSMB5, PSMB6, PSMB7, PSMB8, PSMB10 
CategoryMember
Ubiquitin-conjugating enzyme E2 UBE2L3 
Ubiquitin ligase E3 LUBAC (HOIP, HOIL-1L, and SHARPIN) 
Proteasome PSMB1, PSMB2, PSMB3, PSMB4, PSMB5, PSMB6, PSMB7, PSMB8, PSMB10 

Interestingly, in patients with IgAN, the elevated expression levels of the UBE2L3 gene were positively correlated with the expression levels of the IL-6 (r = 0.37, p = 6.00 × 10−3) (Fig. 5a). And the expression levels of the UBE2L3 gene were positively associated with the RELA (NF-κB p65) (r = 0.38, p = 5.00 × 10−3) (Fig. 5b). STRING database also confirmed the interaction between LUBAC and NEMO. In addition, we found that the expression levels of the UBE2L3 gene were negatively correlated with the expression levels of the C1GALT1 gene (r = −0.49, p = 1.93 × 10−4) and C1GALT1C1 gene (Cosmc) (r = −0.42, p = 2.00 × 10−3) (Fig. 5c, d).

Fig. 5.

Associations between UBE2L3 and the expression levels of genes involved in Gd-IgA1 production. a The correlation analysis between the UBE2L3 and IL-6 expression levels in IgAN patients. b The correlation analysis between the UBE2L3 and expression levels of RELA (NF-κB p65) in IgAN patients. c The correlation analysis between the UBE2L3 and C1GALT1 expression levels in IgAN patients. d The correlation analysis between the UBE2L3 and Cosmc expression levels in IgAN patients.

Fig. 5.

Associations between UBE2L3 and the expression levels of genes involved in Gd-IgA1 production. a The correlation analysis between the UBE2L3 and IL-6 expression levels in IgAN patients. b The correlation analysis between the UBE2L3 and expression levels of RELA (NF-κB p65) in IgAN patients. c The correlation analysis between the UBE2L3 and C1GALT1 expression levels in IgAN patients. d The correlation analysis between the UBE2L3 and Cosmc expression levels in IgAN patients.

Close modal

In the present study, we identified that rs131654 in UBE2L3 gene region showed the strongest association with IgAN, whose risk allele increased UBE2L3 gene expression in multiple tissues and diverse immune cell types. The UBE2L3 gene was highly expressed in patients with IgAN and was associated with higher proteinuria and lower eGFR as well as a tendency for being associated with a higher risk of developing ESKD. Additionally, LUBAC and (immuno)proteasome subunits in PBMCs from IgAN patients were highly expressed compared to the healthy control. According to the findings, the activities of several members of the ubiquitin-proteasome system involving in ubiquitination and proteasome degradation processes, respectively, are enhanced in IgAN, leading to the abnormal function of the ubiquitin-proteasome regulatory pathway, which may be relate to the pathogenesis of IgAN.

Some genetic studies have provided clues that proteasomes may be involved in the pathogenesis of IgAN [1, 2]. Coppo et al. [1] observed an upregulation of the immunoproteasome as well as a switch from proteasome to immunoproteasome with higher catalytic efficiency in PBMCs of IgAN patients. However, the specific biological pathways involved in IgAN by the immunoproteasome remain unclear. In our study, we not only observed the upregulation of the (immuno)proteasome subunits in PBMCs from IgAN patients, but the high expression of the UBE2L3 and a positive correlation with (immuno)proteasome subunit expression. Meanwhile, we found the SHARPIN, RNF31, and RBCK1 genes encoding SHARPIN protein, HOIP protein, and HOIL-1L protein, respectively, were highly expressed in IgAN patients. SHARPIN protein, HOIP protein, and HOIL-1L protein make up the LUBAC, a ubiquitin ligase complex, which preferentially combined with UBE2L3 (an E2 ubiquitin-conjugating enzyme) to form productive E2–E3 pairs, which could activate the NF-κB pathway by binding to NEMO (NF-κB essential modulator) and ubiquitinating it [7‒9]. Combined overexpression of UBE2L3 and LUBAC led to marked upregulation of NF-κB, while dominant-negative mutant UBE2L3 or UBE2L3 silencing abolished it, which indicated that the activation of NF-κB mediated by LUBAC is highly sensitive to the expression level of UBE2L3 [7]. After consulting the STRING database, we confirmed the interaction between UBE2L3 and LUBAC as well as between LUBAC and NEMO. Interestingly, we also found that the UBE2L3 gene expression was positively correlated with the expression of the RELA (NF-κB p65) as well as IL-6, a downstream molecule of the NF-κB signaling pathway, which may further support that the NF-κB signaling pathway may be activated by the complex consisting of UBE2L3 and LUBAC in patients with IgAN. In addition, we found that the UBE2L3 expression was negatively associated with the expression levels of the C1GALT1 and CAGALT1C1 (Cosmc), which involved in galactosylation of the hinge region of the human IgA1 molecule [4, 26, 27]. Suzuki et al. [11] thought that IL-6 may reduce galactosylation of the O-glycan substrate by decreasing expression of the C1GALT1 gene, which led to an increase in the production of Gd-IgA1.

In this study, for the first time, we revealed that the aberrant (immuno)proteasome pathway may contribute to the development of IgAN by activating the NF-κB signaling pathway through the integration of genetic association analysis, differential expression analysis, and protein-protein interactions, which reveals a new immunological mechanism of IgAN and suggests a new therapeutic target. There are also some limitations to be mentioned. First, differential gene expression analyses were based on PBMCs rather than single cells, so it was not possible to confirm whether it was the aberrant expression of UBE2L3 in B or T cells that caused IgAN. However, we found a significant correlation between the rs131654 risk allele and UBE2L3 expression levels in B cells by querying other eQTL studies [22, 23]. Hence, based on our current findings and the results of several other eQTL studies, it is likely that the aberrant ubiquitin-proteasome pathway in B cells may play a role in the development of IgAN. Second, we did not experimentally confirm the interaction between UBE2L3 and LUBAC as well as between LUBAC and NEMO. But we performed bioinformatics analyses of protein interactions. We analyzed the protein-protein interactions between UBE2L3 and LUBAC as well as between LUBAC and NEMO using STRING database, which supported evidence of significant interactions. In addition, previous study has confirmed the interactions between HOIP-1L and UBE2L3 by CO-IP based on HEK293 cells [10].

In conclusion, we uncover a genetic regulator role of (immuno)proteasome in IgAN using genetic association combined with differed expression analysis. Our current study showed the role of UBE2L3-ubiquitin-proteasome in IgAN pathogenesis, which supported a role of genetically conferred dysregulation of the (immuno)proteasome in regulating galactose-deficient IgA1 in the development of IgAN. Future functional therapeutic target analysis of UBE2L3 would be warranted.

This study was approved by the Ethics Committee of Peking University First Hospital (IRB number: 2023-377-001) and was conducted in accordance with the principle of the Helsinki Declaration. Written informed consent was provided by all participants.

All the authors have nothing to disclose.

Support was provided by National Science Foundation of China (82000680, 82022010, 82131430172, 82370709, 8197061382070733, and 82070731); Beijing Natural Science Foundation (Z190023); Academy of Medical Sciences – Newton Advanced Fellowship (NAFR13\1033); King’s College London – Peking University Health Science Center Joint Institute for Medical Research (BMU2021KCL004); Fok Ying Tung Education Foundation (171030); Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (2019-I2M-5-046, 2020-JKCS-009); and National High Level Hospital Clinical Research Funding (Interdisciplinary Clinical Research Project of Peking University First Hospital, 2022CR41). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Research idea and study design: X.-J.Z., H.Z., and Y.L.; data acquisition: L.-L.X., T.G., Y.L., P.C., and S.-F.S.; data analysis/interpretation: L.-L.X., T.G., and Y.L.; statistical analysis: L.-L.X.; supervision or mentorship: X.-J.Z., H.Z., J.-C.L., L.-J.L., and S.-F.S. Each author contributed important intellectual content during manuscript drafting or revision and agreed to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate.

Additional Information

Lin-lin Xu, Ting Gan, and Yang Li contributed equally to this work.

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.

1.
Coppo
R
,
Camilla
R
,
Alfarano
A
,
Balegno
S
,
Mancuso
D
,
Peruzzi
L
, et al
.
Upregulation of the immunoproteasome in peripheral blood mononuclear cells of patients with IgA nephropathy
.
Kidney Int
.
2009
;
75
(
5
):
536
41
.
2.
Gan
T
,
Li
Y
,
Zhou
XJ
,
Zhang
H
.
Immunoproteasome in IgA nephropathy: state-of-art and future perspectives
.
Int J Biol Sci
.
2020
;
16
(
14
):
2518
26
.
3.
Kiryluk
K
,
Li
Y
,
Scolari
F
,
Sanna-Cherchi
S
,
Choi
M
,
Verbitsky
M
, et al
.
Discovery of new risk loci for IgA nephropathy implicates genes involved in immunity against intestinal pathogens
.
Nat Genet
.
2014
;
46
(
11
):
1187
96
.
4.
Kiryluk
K
,
Li
Y
,
Moldoveanu
Z
,
Suzuki
H
,
Reily
C
,
Hou
P
, et al
.
GWAS for serum galactose-deficient IgA1 implicates critical genes of the O-glycosylation pathway
.
PLoS Genet
.
2017
;
13
(
2
):
e1006609
.
5.
Liu
Y
,
Zheng
J
,
Jia
J
,
Li
H
,
Hu
S
,
Lin
Y
, et al
.
Changes E3 ubiquitin protein ligase 1 gene mRNA expression correlated with IgA1 glycosylation in patients with IgA nephropathy
.
Ren Fail
.
2019
;
41
(
1
):
370
6
.
6.
Zhou
XJ
,
Cheng
FJ
,
Zhu
L
,
Lv
JC
,
Qi
YY
,
Hou
P
, et al
.
Association of systemic lupus erythematosus susceptibility genes with IgA nephropathy in a Chinese cohort
.
Clin J Am Soc Nephrol
.
2014
;
9
(
4
):
788
97
.
7.
Lewis
MJ
,
Vyse
S
,
Shields
AM
,
Boeltz
S
,
Gordon
PA
,
Spector
TD
, et al
.
UBE2L3 polymorphism amplifies NF-κB activation and promotes plasma cell development, linking linear ubiquitination to multiple autoimmune diseases
.
Am J Hum Genet
.
2015
;
96
(
2
):
221
34
.
8.
Tokunaga
F
,
Sakata
S
,
Saeki
Y
,
Satomi
Y
,
Kirisako
T
,
Kamei
K
, et al
.
Involvement of linear polyubiquitylation of NEMO in NF-kappaB activation
.
Nat Cell Biol
.
2009
;
11
(
2
):
123
32
.
9.
Ikeda
F
,
Deribe
YL
,
Skånland
SS
,
Stieglitz
B
,
Grabbe
C
,
Franz-Wachtel
M
, et al
.
SHARPIN forms a linear ubiquitin ligase complex regulating NF-κB activity and apoptosis
.
Nature
.
2011
;
471
(
7340
):
637
41
.
10.
Fu
B
,
Li
S
,
Wang
L
,
Berman
MA
,
Dorf
ME
.
The ubiquitin conjugating enzyme UBE2L3 regulates TNFα-induced linear ubiquitination
.
Cell Res
.
2014
;
24
(
3
):
376
9
.
11.
Suzuki
H
,
Raska
M
,
Yamada
K
,
Moldoveanu
Z
,
Julian
BA
,
Wyatt
RJ
, et al
.
Cytokines alter IgA1 O-glycosylation by dysregulating C1GalT1 and ST6GalNAc-II enzymes
.
J Biol Chem
.
2014
;
289
(
8
):
5330
9
.
12.
Lin
JR
,
Wen
J
,
Zhang
H
,
Wang
L
,
Gou
FF
,
Yang
M
, et al
.
Interleukin-17 promotes the production of underglycosylated IgA1 in DAKIKI cells
.
Ren Fail
.
2018
;
40
(
1
):
60
7
.
13.
Kiryluk
K
,
Sanchez-Rodriguez
E
,
Zhou
XJ
,
Zanoni
F
,
Liu
L
,
Mladkova
N
, et al
.
Genome-wide association analyses define pathogenic signaling pathways and prioritize drug targets for IgA nephropathy
.
Nat Genet
.
2023
;
55
(
7
):
1091
105
.
14.
Võsa
U
,
Claringbould
A
,
Westra
HJ
,
Bonder
MJ
,
Deelen
P
,
Zeng
B
, et al
.
Large-scale cis- and trans-eQTL analyses identify thousands of genetic loci and polygenic scores that regulate blood gene expression
.
Nat Genet
.
2021
;
53
(
9
):
1300
10
.
15.
Zheng
Z
,
Huang
D
,
Wang
J
,
Zhao
K
,
Zhou
Y
,
Guo
Z
, et al
.
QTLbase: an integrative resource for quantitative trait loci across multiple human molecular phenotypes
.
Nucleic Acids Res
.
2020
;
48
(
D1
):
D983
91
.
16.
Ota
M
,
Nagafuchi
Y
,
Hatano
H
,
Ishigaki
K
,
Terao
C
,
Takeshima
Y
, et al
.
Dynamic landscape of immune cell-specific gene regulation in immune-mediated diseases
.
Cell
.
2021
;
184
(
11
):
3006
21.e17
.
17.
Purcell
S
,
Neale
B
,
Todd-Brown
K
,
Thomas
L
,
Ferreira
MA
,
Bender
D
, et al
.
PLINK: a tool set for whole-genome association and population-based linkage analyses
.
Am J Hum Genet
.
2007
;
81
(
3
):
559
75
.
18.
Willer
CJ
,
Li
Y
,
Abecasis
GR
.
METAL: fast and efficient meta-analysis of genomewide association scans
.
Bioinforma
.
2010
;
26
(
17
):
2190
1
.
19.
Boughton
AP
,
Welch
RP
,
Flickinger
M
,
VandeHaar
P
,
Taliun
D
,
Abecasis
GR
, et al
.
LocusZoom.js: interactive and embeddable visualization of genetic association study results
.
Bioinforma
.
2021
;
37
(
18
):
3017
8
.
20.
Levey
AS
,
Stevens
LA
,
Schmid
CH
,
Zhang
YL
,
Castro
AF
3rd
,
Feldman
HI
, et al
.
A new equation to estimate glomerular filtration rate
.
Ann Intern Med
.
2009
;
150
(
9
):
604
12
.
21.
Ward
LD
,
Kellis
M
.
HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants
.
Nucleic Acids Res
.
2012
;
40
(
Database issue
):
D930
4
.
22.
Stranger
BE
,
Nica
AC
,
Forrest
MS
,
Dimas
A
,
Bird
CP
,
Beazley
C
, et al
.
Population genomics of human gene expression
.
Nat Genet
.
2007
;
39
(
10
):
1217
24
.
23.
Fairfax
BP
,
Makino
S
,
Radhakrishnan
J
,
Plant
K
,
Leslie
S
,
Dilthey
A
, et al
.
Genetics of gene expression in primary immune cells identifies cell type-specific master regulators and roles of HLA alleles
.
Nat Genet
.
2012
;
44
(
5
):
502
10
.
24.
Baek
K
,
Krist
DT
,
Prabu
JR
,
Hill
S
,
Klügel
M
,
Neumaier
LM
, et al
.
NEDD8 nucleates a multivalent cullin-RING-UBE2D ubiquitin ligation assembly
.
Nature
.
2020
;
578
(
7795
):
461
6
.
25.
Lenoir
JJ
,
Parisien
JP
,
Horvath
CM
.
Immune regulator LGP2 targets Ubc13/UBE2N to mediate widespread interference with K63 polyubiquitination and NF-κB activation
.
Cell Rep
.
2021
;
37
(
13
):
110175
.
26.
Ju
T
,
Cummings
RD
.
A unique molecular chaperone Cosmc required for activity of the mammalian core 1 beta 3-galactosyltransferase
.
Proc Natl Acad Sci USA
.
2002
;
99
(
26
):
16613
8
.
27.
Gale
DP
,
Molyneux
K
,
Wimbury
D
,
Higgins
P
,
Levine
AP
,
Caplin
B
, et al
.
Galactosylation of IgA1 is associated with common variation in C1GALT1
.
J Am Soc Nephrol
.
2017
;
28
(
7
):
2158
66
.