Membranous nephropathy: Clearer pathology and mechanisms identify potential strategies for treatment

Abstract

Primary membranous nephropathy (PMN) is one of the common causes of adult-onset nephrotic syndrome and is characterized by autoantibodies against podocyte antigens causing in situ immune complex deposition. Much of our understanding of the disease mechanisms underpinning this kidney-limited autoimmune disease originally came from studies of Heymann nephritis, a rat model of PMN, where autoantibodies against megalin produced a similar disease phenotype though megalin is not implicated in human disease. In PMN, the major target antigen was identified to be M-type phospholipase A2 receptor 1 (PLA2R) in 2009. Further utilization of mass spectrometry on immunoprecipitated glomerular extracts and laser micro dissected glomeruli has allowed the rapid discovery of other antigens (thrombospondin type-1 domain-containing protein 7A, neural epidermal growth factor-like 1 protein, semaphorin 3B, protocadherin 7, high temperature requirement A serine peptidase 1, netrin G1) targeted by autoantibodies in PMN. Despite these major advances in our understanding of the pathophysiology of PMN, treatments remain non-specific, often ineffective, or toxic. In this review, we summarize our current understanding of the immune mechanisms driving PMN from animal models and clinical studies, and the implications on the development of future targeted therapeutic strategies.

Keywords: Heymann nephritis; autoantibody; immunology; membranous nephropathy; podocyte; treatment.

Figure 1

Mechanism of podocyte injury in Heymann nephritis (A) and primary membranous nephropathy (B) Created with BioRender.com. GBM, glomerular basement membrane; ROS, reactive oxygen species; cPLA2, cytosolic phospholipase A2; PGE2, prostaglandin E2; TxA2, thromboxane A2; COX, cyclo-oxygenase; AA, arachidonic acid; ER, endoplasmic reticulum; RAP, receptor associated protein; Treg, regulatory T cell; Th1, CD4⁺ T helper-1 cell; Tfh, CD4⁺ T follicular helper cell; FH, fumarate hydratase; WT-1, Wilms tumor-1; PLA2R, M-type phospholipase A2 receptor 1; C3aR, C3a receptor; C5aR, C5a receptor; MBL, mannan-binding lectin; MASP, mannan-binding lectin serine protease; ZO-1, zonula occludens-1; THSD7A, thrombospondin type-1 domain-containing protein 7A; NELL1, neural epidermal growth factor-like 1 protein; Sema3B, semaphorin 3B; PCDH7, protocadherin 7; HTRA1, high temperature requirement A serine peptidase 1; NTNG1, netrin G1; Th2, CD4⁺ T helper-2 cell; Th17, CD4⁺ T helper-17 cell; IFN-γ, interferon-γ; *cognate antigen for anti-NELL1, anti-Sema3B, anti-PCDH7, anti-serine protease HTRA1, and anti-NTNG1 autoantibodies respectively. (A): 1. In Heymann nephritis, anti-Fx1A antibodies bind primarily to megalin on podocytes and activate the classical complement pathway (via the IgG2b subclass). 2. Anti-RAP antibodies are also detected and may represent epitope spreading due to the association between megalin and RAP, which assists the transport of megalin from ER to cell surface. 3. Complement activation may be potentially exacerbated by binding of anti-Fx1A antibodies to complement regulatory proteins Crry and CD59. 4. Sublethal C5b-9 injury causes intracellular calcium influx, activating cPLA2, which hydrolyzes membrane phospholipids of the podocyte, ER and nuclear envelope. This causes ER stress, generation of ROS, disruption of the slit diaphragm protein nephrin, and release of AA with subsequent COX-mediated generation of prostanoids (PGE2, TxA2) that increase glomerular filtration pressure, exacerbating proteinuria. Glomerular IgG and complement deposition attract glomerular infiltrates of CD8⁺ T cells, Th1 cells, and macrophages, potentially via the IgG Fc receptor and anaphylatoxins C3a/C5a. 5. Autoantibody production in Heymann nephritis is at least in part dependent on Qa-1-expressing Tfh cells, which are inhibited by CD8⁺ Tregs. (B): 1. In primary membranous nephropathy (PMN), autoantibodies are predominantly directed against PLA2R, and 2. less commonly against THSD7A, 3. NELL1, PCDH7, serine protease HTRA1 and NTNG1, except in pediatric PMN where Sema3B is the most common podocyte antigen targeted by autoantibodies. 4. Increased B cells and Tfh cells are observed, which interact in the germinal center of lymph nodes to induce differentiation of B cells into high-affinity antibody-producing plasma cells. Autoantibodies in PMN are primarily of the IgG4 subclass, potentially due to Th2-mediated IL-4 production. Loss of tolerance to podocyte antigens such as PLA2R may be due to loss of thymus-derived podocyte antigen-specific Tregs. 5. Glycosylated anti-PLA2R IgG4 bind the MBL/MASP-1/2 complex to activate the lectin complement pathway. 6. This causes activation of C3aR and C5aR as well as deposition of C5b-9, which activates cathepsin L- and an aspartic protease-mediated proteolysis of the actin cytoskeleton protein synaptopodin and slit diaphragm protein NEPH1 respectively. Anti-PLA2R antibody-containing serum is also associated with reduced FH, impaired autophagy and reduced adhesion to the GBM, though it is unclear whether these effects are mediated via complement activation. 7. Impairment in autophagy results in internalization of nephrin and the actin cytoskeleton protein α-actinin-4. 8. Reduced FH leads to intracellular fumarate accumulation, which is associated with increased generation of ROS, reduced slit diaphragm protein ZO-1 and reduced transcription of WT-1, which normally activates the expression of podocyte proteins nephrin and podocalyxin. The mechanisms of podocyte injury of other autoantibodies associated with PMN remain incompletely understood.

Figure 2

Overview of the complement system. Created with BioRender.com. MBL, mannan-binding lectin; MASP, mannan-binding lectin serine protease; C4BP, C4b-binding protein; DAF, decay-accelerating factor; MCP, membrane cofactor protein; CR1, complement receptor 1; MAC, membrane attack complex; C3aR, C3a receptor; C5aR, C5a receptor; PMN, primary membranous nephropathy; PLA2R, M-type phospholipase A2 receptor 1; THSD7A, thrombospondin type-1 domain-containing protein 7A. The complement system consists of 3 pathways: classical pathway, lectin pathway and alternative pathway, which can be conceptually understood using 4A’s: attachment, activation, amplification, and attack (steps of complement activation in bold text below). In the classical pathway, complement C1q attaches to the Fc receptor of antigen-bound antibodies (IgG1, IgG3 and IgM in humans, and IgG2a and IgG2b in rodents). The lectin pathway is similar to the classical pathway where lectins (such as ficolins, MBLs or collectins) attach to carbohydrates on pathogens, activating MASP-1 and MASP-2 (analogous to C1r and C1s of the classical pathway), which cleave C4 and C2 to form C3 convertase (C4b2b). In the alternative pathway, small amounts of C3 are spontaneously activated (“C3 tickover”) due to a labile thioester bond, forming a C3b fragment that binds to factor B (B), which in turn is cleaved by factor D (D) to form the alternative pathway C3 convertase (C3bBb) that is stabilised by properdin (P), forming C3bBbP. Regardless of the pathway of activation, C3 convertase cleaves C3 into C3a and C3b, which amplifies alternative pathway activation whereby further factor B binds to C3b forming more C3 convertase. Apart from amplifying C3 convertase formation, C3b also acts as an opsonin and binds to C3 convertase to form C5 convertase (C4b2b3b and C3bBbC3bP), which cleaves C5 into C5a and C5b. C3a and C5a act as potent anaphylatoxins via their respective receptors (C3aR and C5aR). Complement attack is initiated by C5b sequentially binding C6, C7, C8, and C9 to form the MAC (C5b-9), which causes transmembrane pores that cause osmotic cell lysis or sublethal cell injury, activating internal cellular processes. The complement system is tightly regulated by proteins present in plasma (fluid-phase) and on cell surfaces including the podocyte (solid-phase). Fluid-phase regulators include C1q inhibitor (binds C1r/C1s and MASP-1/MASP-2 to prevent classical and lectin pathway activation), C4BP (binds C4b to prevent formation of the classical/lectin pathway C3 convertase), factor H (binds C3b to prevent formation of the alternative pathway C3 convertase), factor I (cofactor for factor H and MCP, which both inhibit C3 convertase formation), vitronectin (binds the C5b-7 complex to prevent MAC formation) and clusterin (binds C7, C8, C9 to prevent MAC formation). Solid-phase regulators include DAF (dissociates C2b from C4b and Bb from C3b to inactivate C3 convertase), MCP (binds C3b and C4b to prevent C3 convertase formation), CR1 (binds C3b and C4b to prevent C3 convertase formation) and CD59 (binds C8 and C9 to prevent MAC formation).

Figure 3

Potential immunological targets for novel therapeutics in primary membranous nephropathy. Created with BioRender.com. BAFF-R, B-cell activating factor receptor; CAR, chimeric antigen receptor; CAAR, chimeric autoantibody receptor; PLA2R, M-type phospholipase A2 receptor 1; Treg, regulatory T cell, In primary membranous nephropathy, autoreactive B cells produce autoantibodies targeting podocyte antigens (represented by PLA2R in this figure but also include thrombospondin type-1 domain-containing protein 7A, neural epidermal growth factor-like 1 protein, semaphorin 3B, protocadherin 7, high temperature requirement A serine peptidase 1, and netrin G1). These autoreactive B cells originate from the bone marrow, mature from immature B cells to mature B cells, become activated upon interaction with CD4⁺ T helper cells (via cytokines and CD40 ligation) and undergo isotype switching. Activated B cells subsequently migrate to the germinal center of secondary lymphoid organs where they interact with CD4⁺ T follicular helper cells, resulting in B cell proliferation and somatic hypermutation, ultimately differentiating into plasmablast and long-lived plasma cells which secrete high-affinity autoantibodies. Immature, mature, activated and germinal center B cells express CD19 and CD20, which can be targeted by 2^nd and 3^rd generation anti-CD20 monoclonal antibodies (eg. ofatumumab and obinutuzumab), which may be superior to rituximab, or CD19 CAR T cells. However, in cases of primary membranous nephropathy resistant to 1^st generation anti-CD20 monoclonal antibodies (eg. rituximab) that respond to 2^nd or 3^rd generation anti-CD20 monoclonal antibodies, this may be due to changes in the CD20 antigen that may be restricted to autoreactive B cells. CD19 and CD20 are also not expressed on plasmablasts or long-lived plasma cells which continue to secrete autoantibodies despite B cell depletion. Monoclonal antibodies targeting BAFF (eg. belimumab), preventing binding to the BAFF receptor, target both B cells and plasmablasts but not long-lived plasma cells. Proteasome inhibitors have been shown to be effective in depleting plasmablasts and plasma cells. Anti-CD38 monoclonal antibodies (eg. daratumumab) may also be effective against plasma cells, plasmablasts and B cells. However, proteasome inhibitors and anti-CD38 monoclonal antibodies are associated significant toxicity and short-duration therapy to deplete plasma cells followed by adjunctive therapy (eg. B cell depletion or T cell co-stimulatory blockade to prevent B cell activation) may be better tolerated. With the discovery of disease antigens underpinning primary membranous nephropathy, CAR T cells or CAAR T cells targeting autoantibodies (eg. anti-PLA2R) on autoreactive B cells, plasmablasts and plasma cells are conceivable. Lastly, harnessing the capacity of CD4⁺ regulatory T cells to induce immune tolerance at sites of inflammation independent of antigen specificity (infectious tolerance) is attractive. Such strategies may involve CAR regulatory T cells directed to the kidney (eg. targeting PLA2R on the podocyte) or low-dose IL-2, which preferentially expands regulatory T cells not effector T cells.