ST6Gal1 is up‐regulated and associated with aberrant IgA1 glycosylation in IgA nephropathy: An integrated analysis of the transcriptome

Abstract Galactose‐deficient IgA1 (Gd‐IgA1) plays a crucial role in the development of Immunoglobulin A nephropathy (IgAN), however, the underlying pathogenic mechanisms driving Gd‐IgA1 production in B cells are not well understood. In this study, RNA‐seq analysis identified 337 down‐regulated and 405 up‐regulated genes in B cells from 17 patients with IgAN and 6 healthy controls. Among them, ST6Gal1, which was associated with IgAN in a previous genome‐wide association study (GWAS), was up‐regulated in IgAN and significantly positive correlated with elevated Gd‐IgA1. In addition, we identified increased plasma ST6Gal1 levels in 100 patients with IgAN, which were associated with higher levels of proteinuria, plasma IgA, Gd‐IgA1 levels, greater degrees of systemic complement activation including C3a, Bb, C4d, MAC and a lower proportion classified as C2 grade (crescent proportion ≥25%). Interesting, in vitro, recombinant ST6Gal1 (rST6Gal1) exposure reduced the production of Gd‐IgA1 in cultured peripheral blood mononuclear cells from IgAN patients. rST6Gal1 stimuli also increased expression of C1GALT1, which were well‐known proportional to the decrease in galactose deficiency of IgA1. In conclusions, we identified increased plasma ST6Gal1 levels and the association of ST6Gal1 with disease severity of IgAN. Additionally, rST6Gal1 administration in vitro increased expression of C1GALT1 and reduced the production of Gd‐IgA1.

RNA-seq provides more accurate data and applications in identifying differently expressed genes between pathological and control samples.
Several investigators have sought to identify specific genetic markers associated with the development and progression of IgAN through microarray profiles. 7,8 However, few studies have specifically described intracellular mechanisms in B lymphocyte associated with IgAN.
To our knowledge, this is the first study to evaluate the global mRNA expression profile in B lymphocyte of IgAN patients, which are directly involved in the disease. In this study, we identified that rST6Gal1 reduce production of Gd-IgA1 by up-regulating C1GALT1 mRNA, suggesting the potential therapeutic value in IgAN.

| Sample collection
A total of 100 patients with IgAN diagnosed in Tianjin Medical University General Hospital from July 2017 to December 2019, and 50 healthy participants were included in this study. Plasma from all participants was collected. Written informed consent was obtained from each patient and healthy participant. Clinical information and histological grading, including age, gender, 24-hour urine protein excretion, blood pressure, serum creatinine, total IgA levels, and Oxford classification M (%), E (%), S (%) and T (%), were collected at the time of renal biopsy ( Table 1).
The study was initially conducted on 47 patients with IgAN and 36 healthy participants. 17 patients (5 with low levels of Gd-IgA1, 6 moderate levels of Gd-IgA1 and 6 high levels of Gd-IgA1) with IgAN and 6 healthy participants were included in the RNA-seq experiment, 20 participants from each group were used for RNA-seq validation, and 10 participants from each group were used for the in vitro intervention experiment in PBMCs. The healthy controls were selected on the basis of their demographic characteristics, age and gender overlapped with IgAN group.

| B lymphocytes isolation
About 5 mL venous blood sample was taken into ethylenediaminetetraacetic acid (EDTA) anticoagulated tubes. Peripheral blood mononuclear cells (PBMCs) were separated by density-gradient centrifugation on Ficoll (TBD), sequential washed three times with phosphate-buffered saline (PBS) and resuspended in PBS + 1% bovine serum albumin (BSA). Peripheral B lymphocytes were isolated using CD19+ magnetic beads (Miltenyi Biotec) according to the manufacturer's instructions.

| Reverse transcription PCR (RT-PCR)
cDNA was synthesized using total RNA with revert first-strand cDNA kit according to manufacturer's protocol (Promega). Resulting C score (C0/C1/C2) 24 (24)/64 (64)/12 (12) cDNA was amplified with a 20 µL reaction mixture using SYBR Green PCR Master Mix (Roche) in an Applied Biosystem 7500 Real-Time PCR System. And the primer pairs of validated genes were listed in Table S1. The fold change between patients and controls was expressed by the 2 −△△CT method. The GAPDH gene amplification was used as a reference standard to normalize the target signal.

| Plasma ST6Gal1 detection
Plasma ST6Gal1 level was determined by a commercial enzymelinked immunosorbent assay (ELISA) kit according to the manufacturer's instruction (Abcam). Plasma samples were diluted 1:20 with diluent. At last, the absorbance was detected at 450 nm with an EL312 Bio-Kinetics microplate reader (Bio-TekInstruments).

| Assay for IgA and Gd-IgA1
IgA and Gd-IgA1 levels in plasma and in cell culture supernatant were detected using a commercial ELISA kit, as previously reported. 9 Plasma concentration of Gd-IgA1 was detected according to the manufacturer's instruction (IBL Subsequently, the colour reaction was stopped and the absorbance was measured at 450 nm.

| Plasma complement component levels
We randomly selected 40 patients with IgAN and detected complement activation products. The levels of human complement components, including C3a, Bb, C4d and C5b-9 (MAC), were determined according to the manufacturer's specifications by ELISA (Quidel).

| Peripheral blood mononuclear cells culture and treatment
Briefly, PBMCs were isolated by density gradient centrifugation and cultured in the RPMI-1640 medium supplemented with 10% foetal calf serum at 37°C in a humidified 5% CO 2 incubator in the subsequent steps. In the in vitro experiment, PBMCs were seeded into 24well plates and incubated with 0, 100 ng/mL, 200 ng/mL 500 ng/mL human recombinant ST6Gal1 (rST6Gal1, R&D Systems) for 48 hours.
The supernatants were collected for detection of IgA1 and Gd-IgA1 levels. The cells were collected to detect C1GALT1 mRNA levels.

| Statistical analysis
For continuous variables, data with a normal distribution were expressed as the mean ± SD and compared by an unpaired t test. For non-normally distributed variables, data were expressed as the median (first quartile and third quartile) and analysed by the Mann-Whitney U test. Categorical variables were summarized as proportions and were compared by a χ 2 test. A 2-tailed P-value <.05 was considered statistically significant. Statistical analysis was performed using SPSS 16.0 software.

| Identification of differentially expressed mRNAs in patients with IgAN and function analysis
We

| Correlation between dysregulated genes expression and Gd-IgA1 levels
After confirming that IgAN patients had abnormal expression genes, we next investigated whether these genes were correlated with Gd-IgA1 levels in IgAN patients. The top 10 genes whose expression was correlated with Gd-IgA1 levels (R 2 > .28) were listed in Table S2. To validate RNA-seq results, we chose the top- The expression levels of all analysed mRNAs were significantly higher in patients with IgAN, thereby confirming RNA-seq results ( Figure S3).

| Patients with IgAN had high levels of ST6Gal1
The mean ST6Gal1 level in plasma in participants with IgAN was 6196 pg/mL, significantly higher than that of healthy controls (4462 pg/mL, P < .001) (Figure 1). ST6Gal1 is widely produced by many cell types, such as hepatocytes and B lymphocytes.
Therefore, we compared the correlation between the expression of ST6Gal1 in B lymphocytes and plasma. Our results showed that there was a positive correlation between them in IgAN no matter in initial RNA-seq individuals (r = .36, P = .09, Figure S4A) or subsequent validation individuals (r = .38, P = .01, Figure S4B), suggesting that B lymphocytes may contribute to elevated plasma ST6GAL1 levels.

| Plasma ST6Gal1 levels correlated with Gd-IgA1
After the identification that patients with IgAN had increased plasma ST6Gal1 levels, we investigated whether plasma ST6Gal1 levels were correlated with Gd-IgA1. There was a strong positive correlation between plasma ST6Gal1 and Gd-IgA1 levels (r = .33, P < .001, Figure 2).

| Plasma ST6Gal1 levels correlated with severity of IgAN
We further explored the association of plasma ST6Gal1 with clinical findings and pathological lesions in patients with IgAN. We classified patients into 2 groups according to the median value of plasma ST6Gal1 levels (median value: 5825 pg/mL). We found that patients with higher ST6Gal1 levels (>5825 pg/mL) had significantly higher levels of C3a, Bb, C4d and MAC compared with those patients with lower ST6Gal1 levels (≤5825 pg/mL, Table 2). The results also showed higher levels of IgA and proteinuria in higher ST6Gal1 levels group, although the difference was not significant. As for pathological data, we found a lower proportion classified as C2 grade (crescent proportion ≥25%) in patients with higher ST6Gal1 levels.

| rST6Gal1 down-regulated Gd-IgA1 secretion from PBMCs in a dose-dependent manner
To evaluate whether ST6Gal1 served as the causal factor for in-

| rST6Gal1 up-regulated C1GALT1 mRNA expression in PBMCs from IgAN patients in a dosedependent manner
We measured the mRNA level of key glycosyltransferases,

TA B L E 2
The baseline data for patients with lower and higher ST6GAL1

CD19+ B cells are believed to be the main sources of IgA production
and may contribute to the onset and progression of IgAN. 10,11 In this study, we at first conducted a global RNA sequencing analysis in B cells from patients with IgAN and healthy controls to identify a subset of mRNA transcripts with expression levels correlated with IgAN susceptibility. Furthermore, we cultured human PBMCs to explore molecular mechanism driving Gd-IgA1 production. By combining these results, we hope to broaden the understanding of pathogenesis of IgAN.
Cell-specific transcripts are capable of identifying genes associated with clinical phenotypes. 12,13 Alterations in the functions of genes with cell restricted expression patterns are widely believed to induce specific disease manifestations. From our RNA-seq analysis, we found 337 down-regulated and 405 up-regulated genes in B lymphocytes in patients with IgAN. Pathway analysis revealed that many of these genes were related to autophagy, mTOR signalling pathway, mannose type O-glycan biosynthesis, ribosome and oxidative phosphorylation et al, which are consistent with previous studies.
However, it is difficult to decide which gene can be chosen for a further in-depth study. On the basis of the pathogenic role of Gd-IgA1 in the development of IgAN, we investigated whether these genes were correlated with Gd-IgA1 levels. The results identified the 19 genes whose expression were correlated with Gd-IgA1 levels with R 2 > .25.
Among these correlated genes, ST6Gal1 aroused our interest.  Table S3). For GSE 58539, ST6Gal1 mRNA level in IgAN was also higher than that of control, although the difference did not reach significance (P = .07, Table S3). 16 Given these evidence, ST6Gal1 appears to be an excellent candidate molecule for further investigation.
We further explored the association of ST6Gal1 on IgAN severity. We observed patients with IgAN who had higher plasma levels of

| CON CLUS ION
Based on transcriptomic data, we identified increased ST6Gal1 levels and the association of ST6Gal1 with disease severity of IgAN.
Additionally, rST6Gal1 administration could reduce the production of Gd-IgA1, which has novel therapeutic potential in the management of IgAN.

ACK N OWLED G EM ENTS
The authors thank all the study subjects for their participation. This study is supported by the National Natural Science Foundation

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

E TH I C A L A PPROVA L
All subjects provided written informed consents. The study protocol was approved by the Institutional Ethical Committee of Tianjin Medical University General Hospital.

DATA AVA I L A B I L I T Y S TAT E M E N T
Raw data used during the current study are available from the corresponding author on reasonable request for non-commercial use.