The Diagnostic and Prognostic Potential of the B-Cell Repertoire in Membranous Nephropathy

Abstract

Membranous nephropathy (MN), an autoimmune glomerular disease, is one of the most common causes of nephrotic syndrome in adults. In current clinical practice, the diagnosis is dependent on renal tissue biopsy. A new method for diagnosis and prognosis surveillance is urgently needed for patients. In the present study, we recruited 66 MN patients before any treatment and 11 healthy control (HC) and analyzed multiple aspects of the immunoglobulin heavy chain (IGH) repertoire of these samples using high-throughput sequencing. We found that the abnormalities of CDR-H3 length, hydrophobicity, somatic hypermutation (SHM), and germ line index were progressively more prominent in patients with MN, and the frequency of IGHV3-66 in post-therapy patients was significantly lower than that in pre-therapy patients. Moreover, we found that the IGHV3-38 gene was significantly related to PLA2R, which is the most commonly used biomarker. The most important discovery was that several IGHV, IGHD transcripts, CDR-H3 length, and SHM rate in pre-therapy patients had the potential to predict the therapeutic effect. Our study further demonstrated that the IGH repertoire could be a potential biomarker for prognosis prediction of MN. The landscape of circulating B-lymphocyte repertoires sheds new light on the detection and surveillance of MN.

Keywords: B-cell receptor repertoire; biomarkers; high-throughput sequencing; immunoglobulin heavy chain; membranous nephropathy.

Figure 1

The heat maps of IGH gene usage frequencies in BCR repertoire from the sample of the MN patients and HC. (A): IGHV; (B): IGHC; (C): IGHD; (D): IGHJ.

Figure 2

Six IGH transcripts were significantly different between MN patients and HC. IGHA1 (A), IGHA2 (B), IGHM (C), IGHG4 (D), IGHD (E), IGHE (F) frequency distribution in PBMC of MN patients (n = 66) and HC (n = 11) with line at the median.

Figure 3

The IGHV and IGHD gene usage between MN patients and the HC. (A) Twenty-four IGHV genes which are significantly different between MN patients and the HC (without correction for multiple comparison). (B) The frequencies of IGHV3 family and IGHV4 family have significant differences in MN patients and the HC. (C) The frequency of IGHD2-2 has significant differences in MN patients and the HC (p = 0.021).

Figure 4

The CDR-H3 length, hydrophobicity somatic hypermutation and GI of MN patients showed significant differences compared with HC. The Shannon index (A), Simpson index (B), CDR-H3 length distribution (C), the Kyte–Doolittle index of hydrophobicity (D), the rate of SHM (E), and the germline index (F) of IGH repertoire in MN patients and HC.

Figure 5

Change of the IGH repertoire between the pre-therapy and post-therapy patients. The Shannon index (A), Simpson index (B), CDR-H3 length distribution (D), the Kyte–Doolittle index of hydrophobicity (E), the rate of SHM (F) and the germline index (G) of IGH repertoire have no significant differences between pre-therapy and post-therapy, but the frequency of IGHV3-66 was significantly lower in the samples of post-therapy patients (n = 13) than paired pre-therapy samples (n = 13) (C).

Figure 6

The IGHV genes, CDR-H3 length, SHM rate, and APLA2R between the CR and non-CR patients. (A) Three IGHV genes were significantly increased in the CR group (n = 22), two of them in MN patients are significantly higher than in HC, which is represented by red. Nine IGHV genes and two IGHD genes that were significantly lower than non-CR (n = 43), and six of them in MN patients are significantly higher than in HC, which is represented by blue. (B) The APLA2R in non-CR group and CR group. (C) IGHV3-38 gene has a negative correlation with APLA2R. (D–G) CDR-H3 length, the Kyte–Doolittle index of hydrophobicity, SHM rate and germline index of IGH repertoire in CR and non-CR patients.