Membranous nephropathy (MN) is a paradigmatic disease entity that combines severe proteinuria with a classic histopathologic and ultrastructural lesion pattern. Traditionally, MN has been divided into primary (idiopathic) and secondary forms. Among the latter, chronic infections and systemic lupus are common. As a milestone discovery, phospholipase A2 receptor (PLA2R) was identified as a major autoantigen in idiopathic MN, and subsequently, several groups reported other putative autoantigens. In this context, laser microdissection of glomeruli from kidney biopsies followed by tandem mass spectrometry was instrumental to detect aberrant protein deposits that could represent either autoantigens or markers of particular patient subsets.
H.J.A.: What do you think about the finding that some of the new putative autoantigens in MN also associate with autoimmunity?
M.H.: This is truly fascinating. The Mayo group now reported that glomerular exostosin-1/2 (EXT1/2) deposits are associated with a milder course of membranous lupus nephritis (MLN).1
H.-J.A.: Oh, really? In 2019, Sethi et al.2 first reported EXT1/2 deposits in anti-PLA2R antibody and thrombospondin type I domain containing 7A-antibody negative MN, many of which had MLN or some form of autoimmunity. However, EXT1/EXT2 autoantibodies were not detected, not even in such patients with polyclonal autoimmunity. This study clearly raised more questions than answers.3
M.H.: Exactly! Now they compared the clinical phenotype of 374 biopsy-proven cases of MLN. Around one third had glomerular EXT1/2 deposits. Patients with lupus nephritis positive for EXT1/2 were younger, had lower serum creatinine levels, and had less glomerulosclerosis and interstitial fibrosis with tubular atrophy compared with those where EXT1/2 deposits were absent.
H.-J.A.: How about differences in levels of proteinuria or kidney biopsy lesion patterns?
M.H.: EXT1/2-positive patients more often presented with proteinuria >3.5 g/d compared with EXT1/2-negative patients. Among the EXT1/2-positive biopsies, 75.4% were classified as a pure class V lupus nephritis, whereas the remaining 24.6% had class V combined with proliferative lupus nephritis.
H.-J.A.: You think testing for EXT1/2 could be useful in clinical terms?
M.H.: Maybe for predicting prognosis. Follow-up data were available in a subset of 160 patients. EXT1/2-negative patients developed ESKD faster and more frequently than EXT1/2-positive patients.
H.-J.A.: Well, these patients were better from the start as you said, so probably, EXT1/2 specifically affects the filtration barrier in a way that nephrotic syndrome appears earlier, leading to an earlier kidney biopsy and treatment compared with those without EXT1/2 in MLN. Look, EXT1/2 is involved in the regulation of the glomerular endothelial glycocalyx that also contributes to the filtration barrier. The glycocalyx plays an essential role in shear-induced endothelial activation and mechanosensation. It contains heparan sulfate chains, and exopstosin family members regulate heparane sulfate chain assembly.4 Mutations in EXT family members can cause hereditary multiple exostoses syndrome, usually resulting in benign osteocartilagenous masses but rarely also in hereditary steroid-resistant nephrotic syndrome.5 But, how this all links to an aberrant deposition of EXT1/2 protein in a subset of patients with MLN still puzzles me.
M.H.: Indeed, mice with a heterozygous loss of EXT1/2 function develop endothelial dysfunction for their disorganized endothelial glycocalyx and impaired nitric oxide production. Obviously, EXT1/2 is essential for glycocalyx homeostasis and endothelial function.6
H.-J.A.: I do not think that these observations tell us anything about the aberrant deposition of EXT1/2 in patients with lupus. For now, this is all speculation. As no antibodies against EXT1/2 were detected, this seems to resemble more a protein storage problem similar to amyloidosis-related nephrotic syndrome, but here, the deposits are restricted to the glomeruli only. Maybe these patients with lupus have autoantibodies against one of the EXT1/2-degrading enzymes or genetic variants that affect EXT1/2 degradation promoting, such as EXT1/2 deposits. Or a defect in one of the necessary chaperones. If this would be the case, there might be specific molecular targets to attenuate EXT1/2 deposition.
M.H.: Yes, there are many such possibilities, but for now, this study provides an entry into further research of this topic. We learned that primary MN is more heterogeneous than previously thought; this now seems to apply also for secondary MN, at least for lupus membranous nephritis.
H.-J.A.: Absolutely, and this is extremely exciting from a conceptual perspective and fits very well to the general trend that questions the use of morphology-based diagnostic categories of kidney disease. Genotyping patients teaches us that glomerular diseases are mostly rare diseases that present with diverse unspecific histopathologic lesion patterns. The prospect of defining kidney diseases by their cause and pathomechanisms rather than by how they look under the microscope may really overcome major hurdles for progress in nephrology. I started medicine at a time when diseases were still named after the person who first described them, which was completely unuseful. But naming diseases and basing treatments on unspecific lesions seen under the microscope rather than after the underlying pathomechanism only produces confusion and malpractice, “FSGS” being a classic example.7 I am not sure where EXT1/2-positive and -negative MLN will take us to a more personalized management of MN, but it fuels the same idea of an unexpected pathophysiologic diversity of traditional disease entities that deserve reclassification.
M.H.: Well, for the moment all of these new antigens and markers for MN did not yet translate into anything clinically relevant in terms of patient management, apart from the use of PLA2R autoantibodies as biomarkers for diagnosis and response to treatment. As this does not apply for EXT1/2-positive MLN, for now this discovery does not affect the management of MN.
H.-J.A.: You are right, but this study is another great example of how using novel techniques such as laser microdissection and mass spectrometry can help to revise our current conceptual framework of kidney disease, as once did immunostaining and as we are now witnessing with genetics. Let us remain curious of what else we can learn about EXT1/2-positive MLN in the future. The study of Ravindran et al.1 will certainly stimulate numerous research activities to unravel the unknowns of this fascinating observation.
H.-J. Anders reports personal fees from AstraZeneca, Bayer, Boehringer, GSK, Inositec, Janssen, and NOXXON, outside the submitted work. The remaining author has nothing to disclose.
H.-J. Anders was supported by Deutsche Forschungsgemeinschaft grant AN372/24-1. M. Hilhorst is supported by Nierstichting grant DKF19OK007.
1. Ravindran A, Casal Moura M, Fervenza FC, Nasr S, Alexander M, Fidler M, et al.: Exostosin-positive and exostosin-negative represent 2 different phenotypes of membranous lupus nephritis. J Am Soc Nephrol 32: 695–706, 2021
2. Sethi S, Madden BJ, Debiec H, Charlesworth MC, Gross L, Ravindran A, et al.: Exostosin 1/exostosin 2-associated membranous nephropathy. J Am Soc Nephrol 30: 1123–1136, 201931061139
3. Anders HJ: Nephropathic autoantigens in the spectrum of lupus nephritis. Nat Rev Nephrol 15: 595–596, 201931197262
4. Bishop JR, Schuksz M, Esko JD: Heparan sulphate proteoglycans fine-tune mammalian physiology. Nature 446: 1030–1037, 200717460664
5. Roberts IS, Gleadle JM: Familial nephropathy and multiple exostoses with exostosin-1 (EXT1) gene mutation. J Am Soc Nephrol 19: 450–453, 200818216313
6. Mooij HL, Cabrales P, Bernelot Moens SJ, Xu D, Udayappan SD, Tsai AG, et al.: Loss of function in heparan sulfate elongation genes EXT1 and EXT 2 results in improved nitric oxide bioavailability and endothelial function. J Am Heart Assoc 3: e001274, 201425468659
7. Kopp JB, Anders HJ, Susztak K, Podestà MA, Remuzzi G, Hildebrandt F, et al.: Podocytopathies. Nat Rev Dis Primers 6: 68, 202032792490