commentary
alternative enzymatic pathways such as ACE2.1 This monocarboxypeptidase is abundantly expressed on the proximal tubule brush border, and its enzymatic activity cleaves the C-terminal amino acid from Ang I. ACE2 might, therefore, provide an alternative pathway for Ang I degradation.1 ACE2 also converts Ang II to Ang1-7 which may bind to its own receptor and act as an endogenous inhibitor of Ang II type 1 receptor signaling.1 ACE2 might, therefore, have a major impact on renal RAS activity. Lastly, liver-specific AGT KO in the study by Matsusaka et al. had a modest beneficial effect on glomerular injury in Nep25 mice as assessed by glomerular nephrin expression.7 In contrast, tubulointerstitial injury was exacerbated in these mice. The authors suggest that this tubulointerstitial damage may be the result of hypotension and tubular ischemia in the liver AGT KO mice. An alternative explanation, however, is that AGT synthesis by the kidney contributed to the more severe tubular dilation and increased tubulointerstitial cell proliferation in this group. Unfortunately, more definitive evidence of glomerular and tubulointerstitial injury could not be assessed in the study by Matsusaka et al.7 because of the short duration of the experimental protocol. In future studies, a more comprehensive histologic evaluation of the renal phenotype may provide additional useful information on the relative impact of liver- and kidneyderived AGT on the disease process. In summary, the findings of Matsusaka et al.7 demonstrate an important role for liver-derived AGT in renal Ang II generation using a model with a severe defect in glomerular permselectivity. In this model, renal Ang II generation was dependent on AGT synthesis by the liver despite upregulation of AGT mRNA in the kidney. An important caveat to these findings is that the large defect in glomerular permselectivity in the Nep25 model likely promoted high concentrations of both AGT and renin in the tubular fluid. Given prominent expression of ACE on the proximal tubule brush border, all the components necessary for robust Ang II Kidney International (2014) 85
generation were present in the tubular lumen in this model. As a result, the contribution of renal AGT synthesis to renal Ang II generation in the kidney was minimized and probably difficult to detect. It will be critical to test the generalizability of the findings of Matsusaka et al.7 in other animal models. DISCLOSURE
The authors declared no competing interests. ACKNOWLEDGMENTS
RFS receives salary support from the National Institutes of Health (R01-DK087707, RO1-DK094987, and P30-DK096493) and the Veterans Administration (BX000791). DIO receives salary support from the National Institutes of Health (T32-DK007731). REFERENCES 1.
2.
Kobori H, Nangaku M, Navar LG et al. The intrarenal renin-angiotensin system: from physiology to the pathobiology of hypertension and kidney disease. Pharmacol Rev 2007; 59: 251–287. Navar LG, Harrison-Bernard LM, Imig JD. Renal actions of angiotensin II at AT1 receptor blockers. In: Epstein M, Brunner HR (eds)
3.
4.
5.
6.
7.
8.
9.
10.
Angiotensin II Receptor Antagonists. Hanley & Belfus: Philadelphia, 2000, pp 189–214. Douglas JG, Hopfer U. Novel aspect of angiotensin receptors and signal transduction in the kidney. Annu Rev Physiol 1994; 56: 649–669. Leyssac PP. A micropuncture study of glomerular filtration and tubular reabsorption of endogenous renin in the rat. Renal Physiol 1978; 1: 181–188. Matsusaka T, Niimura F, Shimizu A et al. Liver angiotensinogen is the primary source of renal angiotensin II. J Am Soc Nephrol 2012; 23: 1181–1189. Pohl M, Kaminski H, Castrop H et al. Intrarenal renin angiotensin system revisited: role of megalin-dependent endocytosis along the proximal nephron. J Biol Chem 2010; 285: 41935–41946. Matsusaka T, Niimura F, Pastan I et al. Podocyte injury enhances filtration of liverderived angiotensinogen and renal angiotensin II generation. Kidney Int 2014; 85: 1068–1077. Gonzalez-Villalobos RA, Janjoulia T, Fletcher NK et al. The absence of intrarenal ACE protects against hypertension. J Clin Invest 2013; 123: 2011–2023. Zhou Y, Boron WF. Role of endogenously secreted angiotensin II in the CO2-induced stimulation of HCO3 reabsorption by renal proximal tubules. Am J Physiol Renal Physiol 2008; 294: F245–F252. Rohrwasser A, Morgan T, Dillon HF et al. Elements of a paracrine tubular reninangiotensin system along the entire nephron. Hypertension 1999; 34: 1265–1274.
see basic research on page 1078
The heterogeneous mononuclear phagocyte system of the kidney Kevin J. Woollard1 and Charles D. Pusey1 The kidney has a diverse repertoire of cells that make up the mononuclear phagocyte system (MPS). Wu et al. identify a population of CD8aa þ CD11c þ MHC-II þ blood precursors that display dendritic cell-like characteristics in the glomeruli. These cells show increased recruitment in a rat anti-glomerular basement membrane glomerulonephritis model and were able to attenuate disease. This study highlights the importance of the MPS in kidney disease and the need to better understand it to develop immunotherapeutics translatable to the renal patient. Kidney International (2014) 85, 1011–1014. doi:10.1038/ki.2013.448
1
Department of Medicine, Imperial College London, London, UK Correspondence: Charles D. Pusey, Renal and Vascular Inflammation Section, Department of Medicine, Imperial College London, Hammersmith Campus, Du Cane Road, London W12 0NN, UK. E-mail: [email protected]
The mononuclear phagocyte system (MPS) was established in 1968 by van Furth and Cohn and can be grouped into myeloid precursors in the bone marrow, circulating blood monocytes, tissue macrophages, and dendritic cells (DCs).1,2 1011
commentary
Macrophages are classically defined as tissue-resident phagocytic cells that contribute to steady-state tissue homeostasis by performing functions such as clearing apoptotic material and producing growth factors. During infection, they can also perform antimicrobial effector functions such as phagocytosis and production of toxic metabolites, chemokines, and cytokines. DCs are largely defined by the specialized functions of antigen presentation and regulation of immune effector cells.2 Bone marrow-derived monocytes are classically thought to differentiate into tissue macrophages, and some populations of DCs. However, recent data show that, like macrophages (M1 and M2) and DCs, monocytes have heterogeneous phenotypes in mammals, which have multiple sub-populations and display independent effector functions. Moreover, populations of monocytes may not differentiate into macrophage/DC tissue effector cells but can independently contribute to immunity.3 Importantly, some populations of macrophages/DCs are not derived from bone marrow progenitors but spatially occupy organs and tissue, including the kidney, from embryonic origins and can proliferate in situ. This, along with studies demonstrating blood monocyte maturation or conversion, has increased the complexity of studying the MPS.2,4 Flow cytometry studies in murine, swine, and human samples have described in detail the vast range of independent (but not necessarily exclusive) and overlapping receptors and proteins that are expressed by cells of the MPS.1–4 Although not described here in detail, these include CD43, CX3CR1, F4/80, CD11b, Ly6C (GR1), CD14, CD16, CD8, CCR2, CD68, CD103, CD206, CD11c, and MHC-II, to name but a few. Kidney biology is no exception to this repertoire of MPS cell phenotypes (Figure 1). This may account for why many investigators, whether originating from monocyte/macrophage or DC disciplines, may have unknowingly studied the same resident kidney MPS 1012
Inflamed glomerulus Parietal epithelium Bowman’s space Podocytes Basement membrane
Capillary lumen
Fenestrated endothelium Mesangium
Interstitial tissue
Crescent
+ Dendritic-like GIL CD8αα cell cells
CD8αα+ CD3– cells
M1-like macrophage
Ly6Chigh Ly6Clow Mesangial ‘surveying’ ‘inflammatory’ cell monocytes monocytes
M2-like macrophage
Renal cortex blood vessel
Figure 1 | Heterogeneous MPS cells of the kidney. This simplified schematic represents some of the cells that make up the mononuclear phagocyte system (MPS) in the kidney during glomerulonephritis (GN). Sub-populations of monocytes and macrophages—in particular, inflammatory M1-like macrophages or Ly6Chigh monocytes—can be recruited both into the kidney interstitial tissue and to the glomerular endothelium. Unique subsets of M2like macrophages and ‘surveying’ Ly6Clow monocytes that can crawl along the endothelial interface within the renal cortex in the steady state may also play a role in GN. Multiple subpopulations of DCs in specific compartments of the kidney have been identified as having effector functions in autoimmune and inflammatory GN. Wu et al. identify a unique precursor CD8aa þ CD11c þ MHC-II þ CD3– DC-like cell that circulates in peripheral blood but can enter the glomeruli (glomeruli-infiltrating leukocyte (GIL) CD8aa þ ) in the presence of GN and can modulate disease. Further work is needed to understand their complex recruitment, retention, trafficking, and fate in GN, and, importantly, whether these cell populations are involved in human GN.
cell, while often overlooking data from other fields. Further, the term ‘resident’ requires some thought; whether this cell is tissue resident from recruited precursors, proliferation, and/or trafficking is an important distinction. Currently there is no consensus on organ-specific definitions of resident cells or the characteristics of residency, either in the steady state or under inflammatory responses. Important knowledge of the kidney MPS began in the 1980s with studies in normal animal kidneys that identified dendritic-like cells within the renal interstitium. These cells were designated antigen-presenting DCs on the basis of
MHC-II expression, or resident macrophages because of F4/80 expression. Further, MHC-II-positive globular macrophages were identified in the periarterial connective tissue, glomerulus, and subcapsular regions.5 These early studies could not predict the variety of additional functions now attributed to the kidney MPS or the fact that functional plasticity might exist depending on cues provided to the kidney MPS from the environment. This is important, as emerging data show that the tissue specificity and hence environment of the MPS is an important factor governing phenotype.6 Moreover, the kidney MPS at steady state changes dramatically Kidney International (2014) 85
commentary
during inflammation, in which resident and recruited cells exhibit great heterogeneity and dynamic phenotypes.4 In the mouse, subsets of plasmacytoid and conventional DCs have been found in healthy kidneys. A subset of conventional DCs of myeloid origin with a phenotype of CD4 þ CD11c þ CX3CR1 þ is dominant and may function in an immune surveillance network. Notably, none of these kidney DCs are present within glomeruli, although they may border the glomeruli. Moreover, kidney interstitial tissues, but not glomeruli, may contain a small number of CD8 þ lymphoid DCs.5 CD11c þ cells (which classically have been characterized as DCs but may also be activated monocytes/macrophages) have been found in interstitial or periglomerular areas in inflammatory glomerulonephritis (GN), and depletion of these cells attenuated disease in murine models.5 Studies have suggested that DCs play a role in both attenuation and promotion of autoimmunity in murine lupus nephritis.5 Several macrophage populations have also been described in kidney compartments both from monocyte precursors and from embryonic resident-like populations.4,5 Regarding monocytes, sub-populations have been described in the steady state, and increased recruitment of inflammatorylike monocytes has been reported in various models of kidney inflammation.5 Regardless of possible overlapping nomenclatures, these studies have clearly identified populations of the MPS in the kidney. Given the importance of the MPS in tissue hemostasis and immunity, the study of its functional characteristics specific to kidney biology in health and disease is of vital importance. Wu et al.7 (this issue) now complement the understanding of the kidney MPS by identifying a peripheral blood mononuclear cell (PBMC), CD8aa þ CD11c þ MHC-II þ CD3–, that specifically infiltrated inflamed glomeruli in a rat model of anti-glomerular basement membrane (anti-GBM) GN. This study suggests that these PBMC CD8aa þ cells are precursors to the glomeruli-infiltrating leukocyte (GIL), Kidney International (2014) 85
a DC-like CD8aa þ CD11b/c þ MHC þ OX62– tissue effector cell in the glomeruli of rats that was previously identified by this group.8 Early infiltration into glomeruli of GN-resistant Lewis rats may be anti-inflammatory, and contribute to their resistance as compared with GN-susceptible Wistar– Kyoto rats. Most intriguingly, the transfer of these PBMC CD8aa þ cells from Lewis rats into GN-prone Wistar– Kyoto rats at the early stages of disease led to significantly attenuated GN. Therefore, this emerging DC-like population that presumably derives from bone marrow precursors may be involved in the termination of an existing autoimmune inflammatory GN. This new observation along with existing studies of MPS phenotypes in GN raises important questions that may help elucidate the need for a diverse repertoire of cells. What is the origin and fate of these cells? What are the effector functions of these cells, and what are the mechanisms of recruitment, retention, and function? Is this specific to kidney compartments, in particular the glomeruli? Finally, if one can identify novel mechanisms and homology, could this be translated to new treatments for human GN? Wu et al.7 show that this CD8aa þ subset is capable of inducing T-cell apoptosis, and this may be the functional mechanism of attenuated GN. Indeed, their rat anti-GBM GN model is a well-characterized model involving an autoreactive T-cell mechanism. Their results suggest that the CD8aa þ GILs might terminate glomerular inflammation in vivo prior to fibrosis. Along with autoreactive T cells in their model, anti-GBM antibodies against diverse GBM autoantigens are produced at later stages of disease. Although these autoantibodies may not be involved in initiation of GN, their role in anti-GBM GN pathogenesis remains to be elucidated. In this study, a low level of GBM-bound IgG antibodies, without significant inflammation, accompanied this inhibition of T cellmediated glomerular injury by adoptive transfer of CD8aa þ CD3– cells. The origin and role of these antibodies
remain to be established, although there is a clear role for anti-GBM antibodies in other similar models of autoimmune GN. A paradoxical consequence of GIL CD8aa þ cell infiltration in GN-resistant and -prone rats, described in this study, may depend on the timing of recruitment and infiltration. As the authors show, recruitment at early stages of inflammation leads to healing or recovery, whereas infiltration after inflammation leads to fibrosis, a pathological end point. It is, however, important to note that this mechanism is one among many others that contribute to GN susceptibility, which include functions of T-regulatory cells and Fc receptors.9 Importantly, in order to assess the potential of this identified subtype of DC-like cells as an immunotherapeutic strategy for earlystage GN, further work would be needed. In particular, the human counterparts to these rat CD8aa þ cells need to be identified, as human DCs do not express CD8aa. Also, what is the mechanism of expansion and recruitment of these cells, and what is their fate once in the glomeruli? Finally, what is the functional correlate of these DC-like populations; are they are specific to autoimmunity, or is their expansion linked directly to immunization? Other recently shown functional characteristics of the kidney MPS include specific cortical and medullary DCs that play specialized roles in antigen processing and presentation to T-helper cells, and innate immune effector functions dependent on CX3CR1.10 An elegant study by Devi et al., using intravital microscopy, showed that the glomeruli environment in GN is important for myeloidcell recruitment, including CX3CR1 þ monocytes. The retention, surveillance, and overall dwell time of these cells described a new independent paradigm for leukocyte recruitment associated with GN in the mouse.11 Overall, it is clear that the kidney MPS is made up of diverse sub-populations of cells (Figure 1). These may have independent effector functions, and time will tell which are important for human GN. What is clear is that 1013
commentary
research in the compartmental recruitment, retention, fate, and function of these cells is vital to understanding kidney biology in both health and disease.
5.
6.
7.
DISCLOSURE
The authors declared no competing interests. 8.
REFERENCES 1.
2.
3.
4.
van Furth R, Cohn ZA. The origin and kinetics of mononuclear phagocytes. J Exp Med 1968; 128: 415–435. Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease. Nature 2013; 496: 445–455. Jakubzick C, Gautier EL, Gibbings SL et al. Minimal differentiation of classical monocytes as they survey steady-state tissues and transport antigen to lymph nodes. Immunity 2013; 39: 599–610. Merad M, Sathe P, Helft J et al. The dendritic cell lineage: ontogeny and function of dendritic cells and their subsets in the steady
9.
10.
11.
state and the inflamed setting. Annu Rev Immunol 2013; 31: 563–604. Nelson PJ, Rees AJ, Griffin MD et al. The renal mononuclear phagocytic system. J Am Soc Nephrol 2012; 23: 194–203. Miller JC, Brown BD, Shay T et al. Deciphering the transcriptional network of the dendritic cell lineage. Nat Immunol 2012; 13: 888–899. Wu J, Zhou C, Robertson J et al. Peripheral blood CD8aa þ CD11c þ MHC-II þ CD3– cells attenuate autoimmune glomerulonephritis in rats. Kidney Int 2014; 85: 1078–1090. Wu J, Zhou C, Robertson J et al. Identification of a bone marrow-derived CD8aa þ dendritic cell-like population in inflamed autoimmune target tissue with capability of inducing T cell apoptosis. J Leukoc Biol 2010; 88: 849–861. Tarzi RM, Cook HT, Pusey CD. Crescentic glomerulonephritis: new aspects of pathogenesis. Semin Nephrol 2011; 31: 361–368. Hochheiser K, Heuser C, Krause TA et al. Exclusive CX3CR1 dependence of kidney DCs impacts glomerulonephritis progression. J Clin Invest 2013; 123: 4242–4254. Devi S, Li A, Westhorpe CL et al. Multiphoton imaging reveals a new leukocyte recruitment paradigm in the glomerulus. Nat Med 2013; 19: 107–112.
see basic research on page 1091
The Ebf1 knockout mouse and glomerular maturation Kai M. Schmidt-Ott1,2,3 Mice deficient in the transcription factor early B-cell factor 1 (Ebf1) lack mature B lymphocytes but have additional phenotypes suggesting functions outside the hematopoietic system. Fretz et al. report that these mice also exhibit quantitative and qualitative developmental renal defects and develop progressive podocyte foot process effacement. The findings not only suggest that Ebf1 may be pivotal to the transcriptional podocyte network, but also illustrate the importance of distinguishing cell-autonomous and non-autonomous inputs to podocyte maturation and integrity. Kidney International (2014) 85, 1014–1016. doi:10.1038/ki.2013.472
1 Max Delbru¨ck Center for Molecular Medicine, Berlin, Germany; 2Experimental and Clinical Research Center, Charite´–Universita¨tsmedizin Berlin, Berlin, Germany and 3Department of Nephrology, Campus Mitte, Charite´– Universita¨tsmedizin Berlin, Berlin, Germany Correspondence: Kai M. Schmidt-Ott, Max Delbru¨ck Center for Molecular Medicine, RobertRo¨ssle-Strasse 10, 13125, Berlin, Germany. E-mail: [email protected]
1014
Embryonic renal development and glomerular maturation are intimately linked. During the earliest phases, nephron progenitor cells condense, convert to primitive nephron epithelia, and form patterned early nephron structures called S-shaped bodies. These S-shaped bodies contain a layer of primitive podocytes that initially form a columnar epithelial layer and a flat cell layer that will consti-
tute Bowman’s capsule.1 Next, a capillary enters the glomerular cleft and contacts the podocytes to form primitive capillaryloop-stage glomeruli. Thereafter, podocytes flatten and form foot processes. This process roughly coincides with the onset of urine production. As urinary filtration increases, additional signs of glomerular maturation can be observed. The capillary loops expand and form a progressively complex network covered by intricate interdigitating podocyte foot processes. The expansion leads to an increasing filtration surface area and a morphological transformation into adult-type mature glomeruli (Figure 1). Several transcription factors regulate early steps of podocyte differentiation, including Tcf21/Pod1, Lmx1b, Mafb, and Foxc2.1 In contrast, less is known about the transcriptional mechanisms that coordinate the later phases of glomerular maturation and the maintenance of glomerular integrity. Fretz et al.2 (this issue) now propose a novel role for the early B-cell factor 1 (Ebf1) transcription factor in regulating late glomerular maturation and the maintenance of filtration barrier integrity. Ebf1 belongs to a family of transcription factors containing an atypical zinc-finger DNAbinding domain and a non–basic helix– loop–helix dimerization domain. Ebf1 is necessary for the development of mature B lymphocytes, which was first shown almost two decades ago by the absence of these cells in Ebf1 knockout (Ebf1 / ) mice.3 Ebf1 / mice lack Ebf1 in all cells, and thus the presence of other defects comes as no surprise. For instance, Ebf1 / mice exhibit excess apoptosis in the brain striatum.4 They also have a defect in facial branchiomotor neuronal migration.5 In addition, the mice are smaller than their littermates, and many die before age 4 weeks for unclear reasons. Fretz and colleagues initially studied the role of Ebf1 in osteoblast function in Ebf1 / mice.6 They observed elevated serum levels of osteocalcin, a non-collagenous protein formed by osteoblasts. Although the finding suggested an increased osteocalcin synthesis in Ebf1 / osteoblasts, the authors paradoxically observed reduced rates of osteocalcin synthesis. Since osteocalcin is cleared by the kidneys,7 Kidney International (2014) 85
alternative enzymatic pathways such as ACE2.1 This monocarboxypeptidase is abundantly expressed on the proximal tubule brush border, and its enzymatic activity cleaves the C-terminal amino acid from Ang I. ACE2 might, therefore, provide an alternative pathway for Ang I degradation.1 ACE2 also converts Ang II to Ang1-7 which may bind to its own receptor and act as an endogenous inhibitor of Ang II type 1 receptor signaling.1 ACE2 might, therefore, have a major impact on renal RAS activity. Lastly, liver-specific AGT KO in the study by Matsusaka et al. had a modest beneficial effect on glomerular injury in Nep25 mice as assessed by glomerular nephrin expression.7 In contrast, tubulointerstitial injury was exacerbated in these mice. The authors suggest that this tubulointerstitial damage may be the result of hypotension and tubular ischemia in the liver AGT KO mice. An alternative explanation, however, is that AGT synthesis by the kidney contributed to the more severe tubular dilation and increased tubulointerstitial cell proliferation in this group. Unfortunately, more definitive evidence of glomerular and tubulointerstitial injury could not be assessed in the study by Matsusaka et al.7 because of the short duration of the experimental protocol. In future studies, a more comprehensive histologic evaluation of the renal phenotype may provide additional useful information on the relative impact of liver- and kidneyderived AGT on the disease process. In summary, the findings of Matsusaka et al.7 demonstrate an important role for liver-derived AGT in renal Ang II generation using a model with a severe defect in glomerular permselectivity. In this model, renal Ang II generation was dependent on AGT synthesis by the liver despite upregulation of AGT mRNA in the kidney. An important caveat to these findings is that the large defect in glomerular permselectivity in the Nep25 model likely promoted high concentrations of both AGT and renin in the tubular fluid. Given prominent expression of ACE on the proximal tubule brush border, all the components necessary for robust Ang II Kidney International (2014) 85
generation were present in the tubular lumen in this model. As a result, the contribution of renal AGT synthesis to renal Ang II generation in the kidney was minimized and probably difficult to detect. It will be critical to test the generalizability of the findings of Matsusaka et al.7 in other animal models. DISCLOSURE
The authors declared no competing interests. ACKNOWLEDGMENTS
RFS receives salary support from the National Institutes of Health (R01-DK087707, RO1-DK094987, and P30-DK096493) and the Veterans Administration (BX000791). DIO receives salary support from the National Institutes of Health (T32-DK007731). REFERENCES 1.
2.
Kobori H, Nangaku M, Navar LG et al. The intrarenal renin-angiotensin system: from physiology to the pathobiology of hypertension and kidney disease. Pharmacol Rev 2007; 59: 251–287. Navar LG, Harrison-Bernard LM, Imig JD. Renal actions of angiotensin II at AT1 receptor blockers. In: Epstein M, Brunner HR (eds)
3.
4.
5.
6.
7.
8.
9.
10.
Angiotensin II Receptor Antagonists. Hanley & Belfus: Philadelphia, 2000, pp 189–214. Douglas JG, Hopfer U. Novel aspect of angiotensin receptors and signal transduction in the kidney. Annu Rev Physiol 1994; 56: 649–669. Leyssac PP. A micropuncture study of glomerular filtration and tubular reabsorption of endogenous renin in the rat. Renal Physiol 1978; 1: 181–188. Matsusaka T, Niimura F, Shimizu A et al. Liver angiotensinogen is the primary source of renal angiotensin II. J Am Soc Nephrol 2012; 23: 1181–1189. Pohl M, Kaminski H, Castrop H et al. Intrarenal renin angiotensin system revisited: role of megalin-dependent endocytosis along the proximal nephron. J Biol Chem 2010; 285: 41935–41946. Matsusaka T, Niimura F, Pastan I et al. Podocyte injury enhances filtration of liverderived angiotensinogen and renal angiotensin II generation. Kidney Int 2014; 85: 1068–1077. Gonzalez-Villalobos RA, Janjoulia T, Fletcher NK et al. The absence of intrarenal ACE protects against hypertension. J Clin Invest 2013; 123: 2011–2023. Zhou Y, Boron WF. Role of endogenously secreted angiotensin II in the CO2-induced stimulation of HCO3 reabsorption by renal proximal tubules. Am J Physiol Renal Physiol 2008; 294: F245–F252. Rohrwasser A, Morgan T, Dillon HF et al. Elements of a paracrine tubular reninangiotensin system along the entire nephron. Hypertension 1999; 34: 1265–1274.
see basic research on page 1078
The heterogeneous mononuclear phagocyte system of the kidney Kevin J. Woollard1 and Charles D. Pusey1 The kidney has a diverse repertoire of cells that make up the mononuclear phagocyte system (MPS). Wu et al. identify a population of CD8aa þ CD11c þ MHC-II þ blood precursors that display dendritic cell-like characteristics in the glomeruli. These cells show increased recruitment in a rat anti-glomerular basement membrane glomerulonephritis model and were able to attenuate disease. This study highlights the importance of the MPS in kidney disease and the need to better understand it to develop immunotherapeutics translatable to the renal patient. Kidney International (2014) 85, 1011–1014. doi:10.1038/ki.2013.448
1
Department of Medicine, Imperial College London, London, UK Correspondence: Charles D. Pusey, Renal and Vascular Inflammation Section, Department of Medicine, Imperial College London, Hammersmith Campus, Du Cane Road, London W12 0NN, UK. E-mail: [email protected]
The mononuclear phagocyte system (MPS) was established in 1968 by van Furth and Cohn and can be grouped into myeloid precursors in the bone marrow, circulating blood monocytes, tissue macrophages, and dendritic cells (DCs).1,2 1011
commentary
Macrophages are classically defined as tissue-resident phagocytic cells that contribute to steady-state tissue homeostasis by performing functions such as clearing apoptotic material and producing growth factors. During infection, they can also perform antimicrobial effector functions such as phagocytosis and production of toxic metabolites, chemokines, and cytokines. DCs are largely defined by the specialized functions of antigen presentation and regulation of immune effector cells.2 Bone marrow-derived monocytes are classically thought to differentiate into tissue macrophages, and some populations of DCs. However, recent data show that, like macrophages (M1 and M2) and DCs, monocytes have heterogeneous phenotypes in mammals, which have multiple sub-populations and display independent effector functions. Moreover, populations of monocytes may not differentiate into macrophage/DC tissue effector cells but can independently contribute to immunity.3 Importantly, some populations of macrophages/DCs are not derived from bone marrow progenitors but spatially occupy organs and tissue, including the kidney, from embryonic origins and can proliferate in situ. This, along with studies demonstrating blood monocyte maturation or conversion, has increased the complexity of studying the MPS.2,4 Flow cytometry studies in murine, swine, and human samples have described in detail the vast range of independent (but not necessarily exclusive) and overlapping receptors and proteins that are expressed by cells of the MPS.1–4 Although not described here in detail, these include CD43, CX3CR1, F4/80, CD11b, Ly6C (GR1), CD14, CD16, CD8, CCR2, CD68, CD103, CD206, CD11c, and MHC-II, to name but a few. Kidney biology is no exception to this repertoire of MPS cell phenotypes (Figure 1). This may account for why many investigators, whether originating from monocyte/macrophage or DC disciplines, may have unknowingly studied the same resident kidney MPS 1012
Inflamed glomerulus Parietal epithelium Bowman’s space Podocytes Basement membrane
Capillary lumen
Fenestrated endothelium Mesangium
Interstitial tissue
Crescent
+ Dendritic-like GIL CD8αα cell cells
CD8αα+ CD3– cells
M1-like macrophage
Ly6Chigh Ly6Clow Mesangial ‘surveying’ ‘inflammatory’ cell monocytes monocytes
M2-like macrophage
Renal cortex blood vessel
Figure 1 | Heterogeneous MPS cells of the kidney. This simplified schematic represents some of the cells that make up the mononuclear phagocyte system (MPS) in the kidney during glomerulonephritis (GN). Sub-populations of monocytes and macrophages—in particular, inflammatory M1-like macrophages or Ly6Chigh monocytes—can be recruited both into the kidney interstitial tissue and to the glomerular endothelium. Unique subsets of M2like macrophages and ‘surveying’ Ly6Clow monocytes that can crawl along the endothelial interface within the renal cortex in the steady state may also play a role in GN. Multiple subpopulations of DCs in specific compartments of the kidney have been identified as having effector functions in autoimmune and inflammatory GN. Wu et al. identify a unique precursor CD8aa þ CD11c þ MHC-II þ CD3– DC-like cell that circulates in peripheral blood but can enter the glomeruli (glomeruli-infiltrating leukocyte (GIL) CD8aa þ ) in the presence of GN and can modulate disease. Further work is needed to understand their complex recruitment, retention, trafficking, and fate in GN, and, importantly, whether these cell populations are involved in human GN.
cell, while often overlooking data from other fields. Further, the term ‘resident’ requires some thought; whether this cell is tissue resident from recruited precursors, proliferation, and/or trafficking is an important distinction. Currently there is no consensus on organ-specific definitions of resident cells or the characteristics of residency, either in the steady state or under inflammatory responses. Important knowledge of the kidney MPS began in the 1980s with studies in normal animal kidneys that identified dendritic-like cells within the renal interstitium. These cells were designated antigen-presenting DCs on the basis of
MHC-II expression, or resident macrophages because of F4/80 expression. Further, MHC-II-positive globular macrophages were identified in the periarterial connective tissue, glomerulus, and subcapsular regions.5 These early studies could not predict the variety of additional functions now attributed to the kidney MPS or the fact that functional plasticity might exist depending on cues provided to the kidney MPS from the environment. This is important, as emerging data show that the tissue specificity and hence environment of the MPS is an important factor governing phenotype.6 Moreover, the kidney MPS at steady state changes dramatically Kidney International (2014) 85
commentary
during inflammation, in which resident and recruited cells exhibit great heterogeneity and dynamic phenotypes.4 In the mouse, subsets of plasmacytoid and conventional DCs have been found in healthy kidneys. A subset of conventional DCs of myeloid origin with a phenotype of CD4 þ CD11c þ CX3CR1 þ is dominant and may function in an immune surveillance network. Notably, none of these kidney DCs are present within glomeruli, although they may border the glomeruli. Moreover, kidney interstitial tissues, but not glomeruli, may contain a small number of CD8 þ lymphoid DCs.5 CD11c þ cells (which classically have been characterized as DCs but may also be activated monocytes/macrophages) have been found in interstitial or periglomerular areas in inflammatory glomerulonephritis (GN), and depletion of these cells attenuated disease in murine models.5 Studies have suggested that DCs play a role in both attenuation and promotion of autoimmunity in murine lupus nephritis.5 Several macrophage populations have also been described in kidney compartments both from monocyte precursors and from embryonic resident-like populations.4,5 Regarding monocytes, sub-populations have been described in the steady state, and increased recruitment of inflammatorylike monocytes has been reported in various models of kidney inflammation.5 Regardless of possible overlapping nomenclatures, these studies have clearly identified populations of the MPS in the kidney. Given the importance of the MPS in tissue hemostasis and immunity, the study of its functional characteristics specific to kidney biology in health and disease is of vital importance. Wu et al.7 (this issue) now complement the understanding of the kidney MPS by identifying a peripheral blood mononuclear cell (PBMC), CD8aa þ CD11c þ MHC-II þ CD3–, that specifically infiltrated inflamed glomeruli in a rat model of anti-glomerular basement membrane (anti-GBM) GN. This study suggests that these PBMC CD8aa þ cells are precursors to the glomeruli-infiltrating leukocyte (GIL), Kidney International (2014) 85
a DC-like CD8aa þ CD11b/c þ MHC þ OX62– tissue effector cell in the glomeruli of rats that was previously identified by this group.8 Early infiltration into glomeruli of GN-resistant Lewis rats may be anti-inflammatory, and contribute to their resistance as compared with GN-susceptible Wistar– Kyoto rats. Most intriguingly, the transfer of these PBMC CD8aa þ cells from Lewis rats into GN-prone Wistar– Kyoto rats at the early stages of disease led to significantly attenuated GN. Therefore, this emerging DC-like population that presumably derives from bone marrow precursors may be involved in the termination of an existing autoimmune inflammatory GN. This new observation along with existing studies of MPS phenotypes in GN raises important questions that may help elucidate the need for a diverse repertoire of cells. What is the origin and fate of these cells? What are the effector functions of these cells, and what are the mechanisms of recruitment, retention, and function? Is this specific to kidney compartments, in particular the glomeruli? Finally, if one can identify novel mechanisms and homology, could this be translated to new treatments for human GN? Wu et al.7 show that this CD8aa þ subset is capable of inducing T-cell apoptosis, and this may be the functional mechanism of attenuated GN. Indeed, their rat anti-GBM GN model is a well-characterized model involving an autoreactive T-cell mechanism. Their results suggest that the CD8aa þ GILs might terminate glomerular inflammation in vivo prior to fibrosis. Along with autoreactive T cells in their model, anti-GBM antibodies against diverse GBM autoantigens are produced at later stages of disease. Although these autoantibodies may not be involved in initiation of GN, their role in anti-GBM GN pathogenesis remains to be elucidated. In this study, a low level of GBM-bound IgG antibodies, without significant inflammation, accompanied this inhibition of T cellmediated glomerular injury by adoptive transfer of CD8aa þ CD3– cells. The origin and role of these antibodies
remain to be established, although there is a clear role for anti-GBM antibodies in other similar models of autoimmune GN. A paradoxical consequence of GIL CD8aa þ cell infiltration in GN-resistant and -prone rats, described in this study, may depend on the timing of recruitment and infiltration. As the authors show, recruitment at early stages of inflammation leads to healing or recovery, whereas infiltration after inflammation leads to fibrosis, a pathological end point. It is, however, important to note that this mechanism is one among many others that contribute to GN susceptibility, which include functions of T-regulatory cells and Fc receptors.9 Importantly, in order to assess the potential of this identified subtype of DC-like cells as an immunotherapeutic strategy for earlystage GN, further work would be needed. In particular, the human counterparts to these rat CD8aa þ cells need to be identified, as human DCs do not express CD8aa. Also, what is the mechanism of expansion and recruitment of these cells, and what is their fate once in the glomeruli? Finally, what is the functional correlate of these DC-like populations; are they are specific to autoimmunity, or is their expansion linked directly to immunization? Other recently shown functional characteristics of the kidney MPS include specific cortical and medullary DCs that play specialized roles in antigen processing and presentation to T-helper cells, and innate immune effector functions dependent on CX3CR1.10 An elegant study by Devi et al., using intravital microscopy, showed that the glomeruli environment in GN is important for myeloidcell recruitment, including CX3CR1 þ monocytes. The retention, surveillance, and overall dwell time of these cells described a new independent paradigm for leukocyte recruitment associated with GN in the mouse.11 Overall, it is clear that the kidney MPS is made up of diverse sub-populations of cells (Figure 1). These may have independent effector functions, and time will tell which are important for human GN. What is clear is that 1013
commentary
research in the compartmental recruitment, retention, fate, and function of these cells is vital to understanding kidney biology in both health and disease.
5.
6.
7.
DISCLOSURE
The authors declared no competing interests. 8.
REFERENCES 1.
2.
3.
4.
van Furth R, Cohn ZA. The origin and kinetics of mononuclear phagocytes. J Exp Med 1968; 128: 415–435. Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease. Nature 2013; 496: 445–455. Jakubzick C, Gautier EL, Gibbings SL et al. Minimal differentiation of classical monocytes as they survey steady-state tissues and transport antigen to lymph nodes. Immunity 2013; 39: 599–610. Merad M, Sathe P, Helft J et al. The dendritic cell lineage: ontogeny and function of dendritic cells and their subsets in the steady
9.
10.
11.
state and the inflamed setting. Annu Rev Immunol 2013; 31: 563–604. Nelson PJ, Rees AJ, Griffin MD et al. The renal mononuclear phagocytic system. J Am Soc Nephrol 2012; 23: 194–203. Miller JC, Brown BD, Shay T et al. Deciphering the transcriptional network of the dendritic cell lineage. Nat Immunol 2012; 13: 888–899. Wu J, Zhou C, Robertson J et al. Peripheral blood CD8aa þ CD11c þ MHC-II þ CD3– cells attenuate autoimmune glomerulonephritis in rats. Kidney Int 2014; 85: 1078–1090. Wu J, Zhou C, Robertson J et al. Identification of a bone marrow-derived CD8aa þ dendritic cell-like population in inflamed autoimmune target tissue with capability of inducing T cell apoptosis. J Leukoc Biol 2010; 88: 849–861. Tarzi RM, Cook HT, Pusey CD. Crescentic glomerulonephritis: new aspects of pathogenesis. Semin Nephrol 2011; 31: 361–368. Hochheiser K, Heuser C, Krause TA et al. Exclusive CX3CR1 dependence of kidney DCs impacts glomerulonephritis progression. J Clin Invest 2013; 123: 4242–4254. Devi S, Li A, Westhorpe CL et al. Multiphoton imaging reveals a new leukocyte recruitment paradigm in the glomerulus. Nat Med 2013; 19: 107–112.
see basic research on page 1091
The Ebf1 knockout mouse and glomerular maturation Kai M. Schmidt-Ott1,2,3 Mice deficient in the transcription factor early B-cell factor 1 (Ebf1) lack mature B lymphocytes but have additional phenotypes suggesting functions outside the hematopoietic system. Fretz et al. report that these mice also exhibit quantitative and qualitative developmental renal defects and develop progressive podocyte foot process effacement. The findings not only suggest that Ebf1 may be pivotal to the transcriptional podocyte network, but also illustrate the importance of distinguishing cell-autonomous and non-autonomous inputs to podocyte maturation and integrity. Kidney International (2014) 85, 1014–1016. doi:10.1038/ki.2013.472
1 Max Delbru¨ck Center for Molecular Medicine, Berlin, Germany; 2Experimental and Clinical Research Center, Charite´–Universita¨tsmedizin Berlin, Berlin, Germany and 3Department of Nephrology, Campus Mitte, Charite´– Universita¨tsmedizin Berlin, Berlin, Germany Correspondence: Kai M. Schmidt-Ott, Max Delbru¨ck Center for Molecular Medicine, RobertRo¨ssle-Strasse 10, 13125, Berlin, Germany. E-mail: [email protected]
1014
Embryonic renal development and glomerular maturation are intimately linked. During the earliest phases, nephron progenitor cells condense, convert to primitive nephron epithelia, and form patterned early nephron structures called S-shaped bodies. These S-shaped bodies contain a layer of primitive podocytes that initially form a columnar epithelial layer and a flat cell layer that will consti-
tute Bowman’s capsule.1 Next, a capillary enters the glomerular cleft and contacts the podocytes to form primitive capillaryloop-stage glomeruli. Thereafter, podocytes flatten and form foot processes. This process roughly coincides with the onset of urine production. As urinary filtration increases, additional signs of glomerular maturation can be observed. The capillary loops expand and form a progressively complex network covered by intricate interdigitating podocyte foot processes. The expansion leads to an increasing filtration surface area and a morphological transformation into adult-type mature glomeruli (Figure 1). Several transcription factors regulate early steps of podocyte differentiation, including Tcf21/Pod1, Lmx1b, Mafb, and Foxc2.1 In contrast, less is known about the transcriptional mechanisms that coordinate the later phases of glomerular maturation and the maintenance of glomerular integrity. Fretz et al.2 (this issue) now propose a novel role for the early B-cell factor 1 (Ebf1) transcription factor in regulating late glomerular maturation and the maintenance of filtration barrier integrity. Ebf1 belongs to a family of transcription factors containing an atypical zinc-finger DNAbinding domain and a non–basic helix– loop–helix dimerization domain. Ebf1 is necessary for the development of mature B lymphocytes, which was first shown almost two decades ago by the absence of these cells in Ebf1 knockout (Ebf1 / ) mice.3 Ebf1 / mice lack Ebf1 in all cells, and thus the presence of other defects comes as no surprise. For instance, Ebf1 / mice exhibit excess apoptosis in the brain striatum.4 They also have a defect in facial branchiomotor neuronal migration.5 In addition, the mice are smaller than their littermates, and many die before age 4 weeks for unclear reasons. Fretz and colleagues initially studied the role of Ebf1 in osteoblast function in Ebf1 / mice.6 They observed elevated serum levels of osteocalcin, a non-collagenous protein formed by osteoblasts. Although the finding suggested an increased osteocalcin synthesis in Ebf1 / osteoblasts, the authors paradoxically observed reduced rates of osteocalcin synthesis. Since osteocalcin is cleared by the kidneys,7 Kidney International (2014) 85
