Circulating Testican-2 and MGT5A are Markers of Membranous Nephropathy

Abstract

Introduction: Membranous nephropathy (MN) is a common cause of nephrotic syndrome usually diagnosed using kidney biopsy.

Methods: We examined the association of 6592 plasma proteins with a diagnosis of MN in the Boston Kidney Biopsy Cohort (BKBC, n = 434), with replication of the top hits in the Nephrotic Syndrome Study Network (NEPTUNE, n = 132).

Results: In BKBC, 2 proteins, testican-2 and alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase (MGT5A), were associated with MN when compared with the reference diagnosis (normal or thin basement membrane [TBM] disease) as well as when compared with all other diagnoses among individuals who had undergone kidney biopsy for the indication of proteinuria or nephrotic syndrome. In NEPTUNE, plasma levels of both proteins, as well as glomerular expression of their cognate genes, were increased in MN compared with minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS). In receiver operating characteristic curve analyses, the addition of plasma testican-2 and MGT5A levels significantly improved discrimination of MN from other diagnoses in BKBC and NEPTUNE compared with models incorporating age, sex, race, estimated glomerular filtration rate (eGFR), and proteinuria.

Conclusion: Together, these findings motivate interest in testican-2 and MGT5A as markers and potential functional participants in MN. More work is required to understand the biological role of these proteins in the glomerular basement membrane in relation to immune complex deposition as well as to assess their performance as biomarkers alongside circulating autoantibodies in patients with MN.

Keywords: MGT5A; membranous nephropathy; testican-2.

Graphical abstract

Figure 1

Protein associations with MN in the Boston Kidney Biopsy Cohort. (a) Volcano plot of proteins associated with the diagnosis of MN versus reference group (normal and TBM). (b) Box plots of testican-2 and MGT5A levels (in RFU) for the 14 most common diagnoses in the Boston Kidney Biopsy Cohort. Analysis of variance with post hoc Holm-Sidak’s multiple comparison test. ∗∗∗∗P < 0.0001, ∗∗∗P < 0.001, ∗∗P < 0.01, ∗P < 0.05. (c) Volcano plot of proteins associated with the diagnosis of MN versus all other diagnoses if the reason for biopsy was proteinuria or nephrotic syndrome. For (a) and (c), each dot represents a protein and proteins meeting the threshold for statistical significance (P < 0.05/[6592 × 2]), above the dashed line, are labeled. AIN, acute interstitial nephritis; ANCA, antineutrophil cytoplasmic antibody vasculitis; ATN, acute tubular necrosis; DKD, diabetic kidney disease; FSGS, focal segmental glomerulosclerosis; IC GN, immune complex mediated glomerulonephritis; IgAN, IgA nephropathy; MGT5A, alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase; MN, membranous nephropathy; TBM, thin basement membrane disease; TMA, thrombotic microangiopathy.

Figure 2

Testican-2 and MGT5A correlations with eGFR, proteinuria, and PLA2R status in the Boston Kidney Biopsy Cohort. (a) Spearman correlations of testican-2 (top) and MGT5A (bottom) levels with eGFR among individuals with membranous nephropathy in the Boston Kidney Biopsy Cohort. (b) Spearman correlations of testican-2 (top) and MGT5A (bottom) levels with proteinuria among individuals with membranous nephropathy in the Boston Kidney Biopsy Cohort. (c) Testican-2 (top) and MGT5A (bottom) levels among 24 individuals with membranous nephropathy in the Boston Kidney Biopsy Cohort who had serum PLA2R antibody measurement using enzyme-linked immunosorbent assay and/or kidney biopsy PLA2R staining; PLA2R positive defined as enzyme-linked immunosorbent assay > 14 RU/ml or positive PLA2R tissue staining. Wilcoxon rank sum test. eGFR, estimated glomerular filtration rate; MGT5A, alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase; NS, nonsignificant; PLA2R, phospholipase A2 receptor.

Figure 3

Plasma testican-2 and MGT5A levels and kidney expression of cognate genes in the Nephrotic Syndrome Study Network. (a) Plasma levels of testican-2 (top) and MGT5A (bottom) among individuals with MN, MCD, and FSGS in the Nephrotic Syndrome Study Network. (b) SPOCK2 (top) and MGAT5 (bottom) RNA expression in glomeruli of individuals with MN, MCD, and FSGS in the Nephrotic Syndrome Study Network. (c) SPOCK2 (top) and MGAT5 (bottom) RNA expression in the tubulointerstitium of individuals with MN, MCD, and FSGS in the Nephrotic Syndrome Study Network. Wilcoxon rank sum test. ∗∗∗P < 0.001, ∗∗P < 0.01, ∗P < 0.05, FSGS, focal segmental glomerulosclerosis; MCD, minimal change disease; MGT5A, alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase; MN, membranous nephropathy; NS, nonsignificant.

Figure 4

Receiver operating characteristic analyses with testican-2 and alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase for diagnosis of membranous nephropathy. Receiver operating curves for (a) membranous nephropathy versus all other diagnosis in the Boston Kidney Biopsy Cohort, (b) membranous nephropathy versus all other diagnosis among individuals whose primary indication for biopsy was proteinuria or nephrotic syndrome in the Boston Kidney Biopsy Cohort, and (c) membranous nephropathy versus minimal change disease and focal segmental glomerulosclerosis in the Nephrotic Syndrome Study Network. For (a–c), the 3 models include the following: (1). age, sex, race, estimated glomerular filtration rate, proteinuria; (2)alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase and testican-2; (3)alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase, testican-2, age, sex, race, estimated glomerular filtration rate, proteinuria. C-statistics of correlated models were compared in pair-wise fashion using Delong’s test and noted in the legend. AUC, area under the receiver operating characteristic curve.