CD8⁺ cells and glomerular crescent formation: outside-in as well as inside-out

Crescentic glomerulonephritis, a complication of severe immune glomerular injury, is the pathological correlate of rapidly progressive glomerulonephritis, mediated by both humoral and cellular effectors. In the current issue of the JCI, Chen et al. have implicated Bowman’s capsule in functionally isolating potentially immune effectors, specifically antigen-specific CD8⁺ T lymphocytes, from podocytes. They suggest that, in crescentic glomerulonephritis, immune-mediated glomerular endothelial injury results in inside-out injury to the glomerulus, with subsequent leukocyte migration through a weakened or ruptured Bowman’s capsule, resulting in outside-in injury. Effector T cells then recognize nephritogenic antigens presented by podocytes or other cells within the urinary space, enhancing injury and crescent formation.

The most severe and rapidly progressive forms of glomerulonephritis are characterized by the presence of glomerular crescents outside the glomerular capillary tuft within the urinary space. These lesions consist of fibrin, proliferating epithelial cells, macrophages, and CD4⁺ and CD8⁺ T lymphocytes. Without treatment or resolution of the underlying cause, the crescents first become fibrocellular, then fibrous. Subsequently, the glomerular tuft becomes obsolescent, with no chance of repair or recovery. While crescentic lesions are occasionally seen in nonimmune pathologies, glomerular crescents are typically a complication of several types of autoimmune glomerulonephritis. Crescent formation indicates severe and intense nephritogenic immunity, and people with crescentic glomerulonephritis are often prescribed significant immunosuppression to limit local inflammation and modify the underlying pathological immune response (1). Glomerular crescents are most frequently found in the small-vessel vasculitides, Goodpasture’s disease, and antineutrophil cytoplasmic antibody–associated (ANCA-associated) vasculitis, in which the autoantigens involved are either intrinsic to the kidney or initially present on neutrophils, then deposited in the kidney (2–4). Crescent formation can also complicate some cases of immune complex glomerulonephritis, including lupus nephritis, IgA nephropathy, and postinfectious glomerulonephritis. In addition to the formation of glomerular crescents and areas of segmental necrosis within the glomerular tuft, the most severe lesions also feature rupture of Bowman’s capsule (5). This basement membrane is lined by parietal epithelial cells and normally contains the glomerular tuft and the urinary space (Figure 1, left panel). In a variety of inflammatory states, renal mononuclear phagocytes and T cells can be found adjacent to, but outside, Bowman’s capsule, and in human and experimental crescentic glomerulonephritis, more severe lesions with more leukocytes are observed when Bowman’s capsule has ruptured (6).

Figure 1

Schematic diagram summarizing inside-out and outside-in elements of glomerular crescent formation. Left panel: The basic structure of a healthy glomerulus consisting of the Bowman’s capsule lined with parietal epithelial cells and the glomerular filtration barrier composed of endothelial cells, the glomerular basement membrane (GBM), and podocytes. Podocytes line the GBM with foot processes that wrap around the glomerular capillaries. Middle panel: Examples of some of the inside-out mechanisms that can cause endocapillary damage and early crescent formation following glomerulonephritis. Inset, left to right, neutrophils bind to autoantibodies in the GBM, as in anti-GBM glomerulonephritis. Antibodies bind to antigens lodged in the glomerulus with subsequent macrophage recruitment. ANCA activates neutrophils and mediates their recruitment to the glomerulus. CD8⁺ T cells can recognize antigenic peptides, potentially presented by endothelial cells. Right panel: Chen and colleagues describe the outside-in mechanism in which the antigen-specific CD8⁺ T cells, along with macrophages, located outside the glomerulus migrate into the urinary space following a breach in Bowman’s capsule. Within the urinary space, CD8⁺ cells recognize antigens presented on podocytes, exacerbating injury and crescent formation.

The function of the glomerulus as a high-flow, high-pressure filter means that it is vulnerable to the full range of mediators that induce pathological inflammation (7). Humoral immunity has long been associated with glomerulonephritis. Despite the importance of cellular effectors in other autoimmune diseases, the role of effector T cells in severe autoimmune glomerulonephritis is less well defined, but effector CD4⁺ and CD8⁺ cells have been shown to participate in glomerular injury (8, 9). In humans, intrarenal CD8⁺ cells correlate with outcomes and CD8⁺ cell molecular signatures denote a poorer prognosis (4, 10, 11). Furthermore, in both autoimmune and nonautoimmune models of glomerulonephritis, depletion of CD8⁺ T cells limits histological and functional injury and transfer of CD8⁺ T cells can induce injury (12–14).

In this issue of the JCI, Chen et al. shed light on the role of CD8⁺ cells in crescentic glomerulonephritis. They provide evidence for the ingress of periglomerular CD8⁺ T cells through localized breaches in Bowman’s capsule, to subsequently participate in glomerular crescent formation and markedly amplify glomerular injury (15). Mice expressing the model antigen EGFP on podocytes received EGFP-specific CD8⁺ cells from T cell receptor (TCR) transgenic mice, evocatively named just EGFP death inducing (Jedi) mice. This model system required a separate and additional immunological trigger, as even when activated, as are T cells in autoimmune renal disease, usually via systemic exposure to autoantigen, these cells do not recognize EGFP on podocytes within normal glomeruli, potentially due to a lack of physical colocation. However, when injury to the elements of the glomerular filtration barrier is induced by the injection of antirenal basement membrane globulin (also known as nephrotoxic serum), the Jedi effector T cells are highly pathogenic. These CD8⁺ cells recognize their cognate antigen on podocytes, resulting in podocyte loss, epithelial and local CD8⁺ infiltration and proliferation with enhanced crescent formation, and functional kidney injury. The study also examined 2D biopsy sections from human crescentic glomerulonephritis and showed that CD8⁺ cells were more commonly found in glomeruli in which Bowman’s capsule was ruptured. The authors present compelling correlative evidence, though the relation between Bowman’s capsule rupture and crescent formation remains an association. Although challenging, further studies using in vivo dual photon microscopy may be able to directly visualize leukocytes eroding or breaking through Bowman’s capsule.

How does one reconcile the prominence of crescent formation in endocapillary focused forms of rapidly progressive glomerulonephritis with the studies of Chen et al. that used a model podocyte antigen? Most intrarenal antigens currently known to be involved in human crescentic glomerulonephritis are located in or around the glomerular endothelium. The prototypic podocyte antigen in glomerulonephritis is the phospholipase A2 receptor 1 (PLA2R1), which causes the majority of cases of membranous nephropathy, a disease only uncommonly complicated by glomerular crescent formation (16). Experimentally, podocytes not only express MHC class I (MHC-I) and MHC-II, but can act to present antigen to effector T cells. On the other hand, in ANCA-associated vasculitis, which is commonly associated with crescent formation, key autoantigens, such as myeloperoxidase (MPO), are found not only on neutrophils, in serum, and around the endothelium, but also subepithelially (3, 4, 17, 18). Thus, the capacity of podocytes to present antigen and the presence of autoantigens in and around podocytes after endothelial injury link endocapillary injury to podocytes and justify Chen et al.’s use of a podocyte antigen in their studies (15).

The answer to the question of how to incorporate the findings of Chen et al. into an integrated model of crescent formation may come from examining previous experimental studies. While several experimental systems have used cells from TCR transgenic mice in experimental glomerulonephritis, arguably the most instructive have been two studies examining effector responses to ovalbumin (OVA) as a model antigen, using CD4⁺ OVA–specific OT-II cells and/or CD8⁺ OVA–specific OT-I cells (19, 20). In the first study, expressing OVA specifically on podocytes and transferring OVA-specific CD4⁺ and CD8⁺ cells resulted in significant proteinuria and OVA and dendritic cell–mediated accumulation of leukocytes in periglomerular and tubulointerstitial areas (20). Initially, there was little glomerular injury aside from signs of activated parietal cells, but at a later stage, after transfer of cells on several occasions, focal and segmental glomerulosclerosis was present and OVA-expressing cells were observed near Bowman’s capsule. In the second system, OVA was planted intravascularly on and around the endothelium (19). Transfer of either Th1 or Th17 OT-II cells induced significant endocapillary glomerular injury, with early crescent formation after Th1 cell transfer, likely via intravascular antigen recognition (21). The studies of Chen et al. might be considered a composite of the Summers and the Heymann studies (19, 20). The initial injury in this dual stimulus model of nephrotoxic serum nephritis is induced by an antirenal basement membrane globulin preparation, which in its early days, models endocapillary inflammation mediated by in situ immune complex formation (22). This initial injury allows leakage of plasma proteins and glomerular antigens into the urinary space. In more severe forms of immune endocapillary injury, fibrin and other proteins induce leukocyte chemotaxis and accumulation, podocyte bridging and dedifferentiation, and parietal cell and renal progenitor cell activation, often resulting in the formation of early glomerular crescents (23–25). Combined with these inflammatory events in the urinary space, T cells and macrophages located outside the glomerulus then migrate through breaks in Bowman’s capsule, where antigen-specific cells in the Jedi mouse model mount a significant immune attack and promote crescent formation (Figure 1 summarizes these concepts).

With the studies of Chen et al., there is now more evidence for a two-step process, inside-out, then outside-in, in the development of crescentic glomerulonephritis, the pathological correlate of the feared clinical presentation of rapidly progressive glomerulonephritis. The studies further support a role of cell-mediated delayed-type hypersensitivity–like effector responses and suggest that further focus on selectively inhibiting cell-mediated immunity in these severe and often difficult to treat diseases is warranted.

ARK’s work in the field of immune renal disease is supported by grants from the National Health and Medical Research Council of Australia.

Address correspondence to: A. Richard Kitching, Centre for Inflammatory Diseases, Monash University Department of Medicine, Monash Medical Centre, 246 Clayton Rd., Clayton, Victoria 3168, Australia. Phone: 61.3.8572.2590; Email: richard.kitching@monash.edu.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest. 2018;128(8):3231–3233. https://doi.org/10.1172/JCI122045.

See the related article at Bowman’s capsule provides a protective niche for podocytes from cytotoxic CD8⁺ T cells.

1. Couser WG. Glomerulonephritis. Lancet. 1999;353(9163):1509–1515.
   View this article via: PubMed CrossRef Google Scholar
2. Hudson BG, Tryggvason K, Sundaramoorthy M, Neilson EG. Alport’s syndrome, Goodpasture’s syndrome, and type IV collagen. N Engl J Med. 2003;348(25):2543–2556.
   View this article via: PubMed CrossRef Google Scholar
3. Falk RJ, Terrell RS, Charles LA, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Natl Acad Sci U S A. 1990;87(11):4115–4119.
   View this article via: PubMed CrossRef Google Scholar
4. O’Sullivan KM, et al. Renal participation of myeloperoxidase in antineutrophil cytoplasmic antibody (ANCA)-associated glomerulonephritis. Kidney Int. 2015;88(5):1030–1046.
   View this article via: PubMed CrossRef Google Scholar
5. Rastaldi MP, Ferrario F, Tunesi S, Yang L, D’Amico G. Intraglomerular and interstitial leukocyte infiltration, adhesion molecules, and interleukin-1 alpha expression in 15 cases of antineutrophil cytoplasmic autoantibody-associated renal vasculitis. Am J Kidney Dis. 1996;27(1):48–57.
   View this article via: PubMed CrossRef Google Scholar
6. Lan HY, Nikolic-Paterson DJ, Atkins RC. Involvement of activated periglomerular leukocytes in the rupture of Bowman’s capsule and glomerular crescent progression in experimental glomerulonephritis. Lab Invest. 1992;67(6):743–751.
   View this article via: PubMed Google Scholar
7. Kitching AR, Hutton HL. The players: cells involved in glomerular disease. Clin J Am Soc Nephrol. 2016;11(9):1664–1674.
   View this article via: PubMed CrossRef Google Scholar
8. Ooi JD, et al. The immunodominant myeloperoxidase T-cell epitope induces local cell-mediated injury in antimyeloperoxidase glomerulonephritis. Proc Natl Acad Sci U S A. 2012;109(39):E2615–E2624.
   View this article via: PubMed CrossRef Google Scholar
9. Rennke HG, Klein PS, Sandstrom DJ, Mendrick DL. Cell-mediated immune injury in the kidney: acute nephritis induced in the rat by azobenzenearsonate. Kidney Int. 1994;45(4):1044–1056.
   View this article via: PubMed CrossRef Google Scholar
10. McKinney EF, et al. A CD8⁺ T cell transcription signature predicts prognosis in autoimmune disease. Nat Med. 2010;16(5):586–591.
    View this article via: PubMed CrossRef Google Scholar
11. Couzi L, et al. Predominance of CD8⁺ T lymphocytes among periglomerular infiltrating cells and link to the prognosis of class III and class IV lupus nephritis. Arthritis Rheum. 2007;56(7):2362–2370.
    View this article via: PubMed CrossRef Google Scholar
12. Chang J, et al. CD8⁺ T cells effect glomerular injury in experimental anti-myeloperoxidase GN. J Am Soc Nephrol. 2017;28(1):47–55.
    View this article via: PubMed CrossRef Google Scholar
13. Reynolds J, Norgan VA, Bhambra U, Smith J, Cook HT, Pusey CD. Anti-CD8 monoclonal antibody therapy is effective in the prevention and treatment of experimental autoimmune glomerulonephritis. J Am Soc Nephrol. 2002;13(2):359–369.
    View this article via: PubMed Google Scholar
14. Kawasaki K, Yaoita E, Yamamoto T, Kihara I. Depletion of CD8 positive cells in nephrotoxic serum nephritis of WKY rats. Kidney Int. 1992;41(6):1517–1526.
    View this article via: PubMed CrossRef Google Scholar
15. Chen A, et al. Bowman’s capsule provides a protective niche for podocytes from cytotoxic CD8⁺ T cells. J Clin Invest. 2018;128(8):3413–3424.
    View this article via: JCI CrossRef Google Scholar
16. Beck LH Jr, et al. M-type phospholipase A2 receptor as target antigen in idiopathic membranous nephropathy. N Engl J Med. 2009;361(1):11–21.
    View this article via: PubMed CrossRef Google Scholar
17. Goldwich A, et al. Podocytes are nonhematopoietic professional antigen-presenting cells. J Am Soc Nephrol. 2013;24(6):906–916.
    View this article via: PubMed CrossRef Google Scholar
18. Coers W, et al. Podocyte expression of MHC class I and II and intercellular adhesion molecule-1 (ICAM-1) in experimental pauci-immune crescentic glomerulonephritis. Clin Exp Immunol. 1994;98(2):279–286.
    View this article via: PubMed Google Scholar
19. Summers SA, et al. Th1 and Th17 cells induce proliferative glomerulonephritis. J Am Soc Nephrol. 2009;20(12):2518–2524.
    View this article via: PubMed CrossRef Google Scholar
20. Heymann F, et al. Kidney dendritic cell activation is required for progression of renal disease in a mouse model of glomerular injury. J Clin Invest. 2009;119(5):1286–1297.
    View this article via: JCI PubMed CrossRef Google Scholar
21. Westhorpe CLV, et al. Effector CD4⁺ T cells recognize intravascular antigen presented by patrolling monocytes. Nat Commun. 2018;9(1):747.
    View this article via: PubMed CrossRef Google Scholar
22. Odobasic D, Ghali JR, O’Sullivan KM, Holdsworth SR, Kitching AR. Glomerulonephritis induced by heterologous anti-GBM globulin as a planted foreign antigen. Curr Protoc Immunol. 2014;106:15.26.1–15.2620.
    View this article via: PubMed Google Scholar
23. Singh SK, Jeansson M, Quaggin SE. New insights into the pathogenesis of cellular crescents. Curr Opin Nephrol Hypertens. 2011;20(3):258–262.
    View this article via: PubMed CrossRef Google Scholar
24. Tipping PG, Dowling JP, Holdsworth SR. Glomerular procoagulant activity in human proliferative glomerulonephritis. J Clin Invest. 1988;81(1):119–125.
    View this article via: JCI PubMed CrossRef Google Scholar
25. Holdsworth SR, Thomson NM, Glasgow EF, Atkins RC. The effect of defibrination on macrophage participation in rabbit nephrotoxic nephritis: studies using glomerular culture and electronmicroscopy. Clin Exp Immunol. 1979;37(1):38–43.
    View this article via: PubMed Google Scholar