Authors’ Reply: In Silico–Based Approach to the Discovery of New Antigens in Membranous Nephropathy

Takahashi-Kobayashi and colleagues have proposed an in silico approach, using publicly available datasets, to identify additional antigens in membranous nephropathy (MN). Their strategy follows on our observation that four known target antigens in MN (PLA2R, NELL1, PCDH7, and NCAM1) were clustered in a list of top genes whose differential expression could best classify and distinguish MN at the molecular level from the other nephrotic disorders in the Nephrotic Syndrome Study Network.1 We should note that the Ju CKD Glom samples, on which Takahashi-Kobayashi et al. based their work, are part of the European Renal cDNA Biobank cohort, which we also incorporated in our analysis. Thus, the finding of five genes shared between the two studies is not entirely surprising. Our “molecular classification” of MN using a machine-learning approach was designed as an unbiased method to classify MN from other nephrotic disorders. We speculated that the process of complement-mediated podocyte injury might upregulate a limited subset of genes important in disease-initiating processes or unique mechanisms of downstream damage and/or repair. We did not set out with any a priori expectation that genes found through our approach would turn out to be target antigens. Nor do we feel there were many patients with PCHD7-, NCAM1-, or even NELL1-associated MN driving these signals. Rather, upregulation of these genes may be related to generalized transcriptional changes occurring in many patients exhibiting the lesion of MN. Why then this concentration of known antigens, and might we expect more proteins to be discovered, as Takahashi-Kobayashi et al. suggest? The first step of their pipeline identifies genes with higher expression in MN. Dysregulated expression of certain genes in podocytes may pave the way for the proteins they encode to serve as target antigens in situ. Both expression at the correct location by the podocyte and the ability to present these antigens to the immune system in a manner to break tolerance are necessary to propagate this autoimmune glomerulopathy. This could be accomplished by overexpression of a protein such as PLA2R,2 potentially causing misfolding and/or increased shedding in exosomes,3 which might facilitate antigen presentation, or by neoexpression of an antigen such as NELL1, not typically found at significant levels in podocytes. In either case, it is expected that only a small subset of individuals would be poised to respond to any one antigen, based on their HLA class II repertoire. Although we can rationalize why our approach might have demonstrated an enrichment of several target antigen genes, we would caution against relying solely on such an in silico approach to identify novel targets. Rather, a multiomics approach, combining data derived from such an algorithmic analysis of existing transcriptional data, combined with proteomic, genetic, and single-cell expression datasets, may all be needed in the discovery process. Disclosures L.H. Beck Jr. reports serving on the advisory boards for Alexion, Ionis, Novartis, and Visterra; having consultancy agreements with Alexion and Novartis; serving on the editorial boards of Kidney International Reports and Kidney Medicine; receiving research funding from Pfizer; receiving honoraria from UpToDate and Visterra; and having patents as coinventor and receiving royalties related to the US patent “Diagnostics for Membranous Nephropathy.” M. Kretzler reports receiving research funding from amfAR, Angion, AstraZeneca, Boehringer Ingelheim, Certa, Chan Zuckerberg Initiative, Chinook, Elpidera, Gilead, Goldfinch, Ionis, Jansen, JDRF, Lilly, National Institutes of Health, Novo Nordisk, Regeneron, RenalytixAI, and Travere (as principal investigator at University of Michigan); having consultancy agreements with Astellas, Boehringer Ingelheim, Certa, Novo Nordisk, Janssen, and Poxel (as an employee of University of Michigan); serving on the editorial boards of JASN, Kidney Disease, and Kidney International; and serving on the advisory board of Nephcure Kidney International. L. Mariani reports receiving honoraria from American Society of Nephrology Board Review Course and Update, Calliditas Therapeutics Advisory Board, Reata Pharmaceuticals CKD Advisory Committee, and Travere Therapeutics Advisory Board; receiving research funding from Boehringer Ingelheim and Travere Therapeutics; serving in an advisory or leadership role for Calliditas Therapeutics, Reata Pharmaceuticals, and Travere Therapeutics; and having consultancy agreements with Calliditas Therapeutics Advisory Board, Reata Pharmaceuticals CKD Advisory Committee, and Travere Therapeutics Advisory Board. The remaining author has nothing to disclose. Funding None.