BRIEF REVIEW www.jasn.org

The Renal Mononuclear Phagocytic System

† ‡ Peter J. Nelson,* Andrew J. Rees, Matthew D. Griffin, Jeremy Hughes,¶ Christian Kurts,§ and Jeremy Duffield*

*University of Washington, Seattle, Washington; †Medical University of Vienna, Vienna, Austria; ‡National University of Ireland, Galway, Ireland; ¶University of Edinburgh, Edinburgh, United Kingdom; and §University Clinic of Bonn, Bonn, Germany

ABSTRACT The renal mononuclear phagocytic system, conventionally composed of macro- resident Mø because of F4/80 expres- phages (Mø) and dendritic cells (DCs), plays a central role in health and disease of sion.8–10 In addition, rare MHC class II+ the kidney. Overlapping definitions of renal DCs and Mø, stemming from historically globular Mø were identified in the glo- separate research tracks and the lack of experimental tools to specifically study the merulus11,12 and subcapsular and periar- roles of these cells in vivo, have generated confusion and controversy, however, terial connective tissue.13,14 Subsequently, regarding their immunologic function in the kidney. This brief review provides an however, multicolor flow cytometry re- appraisal of the current state of knowledge of the renal mononuclear phagocytic vealed substantial overlap and heteroge- system interpreted from the perspective of immunologic function. Physical charac- neity in expression of historically defined teristics, ontogeny, and known functions of the main subsets of renal mononuclear Mø (CD11b, F4/80, and CD68) and DC as they relate to homeostasis, surveillance against injury and infection, (CD11c, MHC II, and CD80/86) markers and immune-mediated inflammatory injury and repair within the kidney are de- by interstitial rMoPh.15–19 Along with scribed. Gaps and inconsistencies in current knowledge are used to create a road- similar descriptions in normal human map of key questions to be answered in future research. kidney,20–22 these studies established the predominant rMoPh at steady state as J Am Soc Nephrol 23: 194–203, 2012. doi: 10.1681/ASN.2011070680 networked, dendritiform, interstitial cells that coexpress multiple markers previously thought to segregate Mø The mononuclear phagocytic system in additional and at times, overlapping from DCs. Hence, many investigators, nonlymphoid organs such as the kidney functional properties that extend be- whether originating from Mø or DC dis- is composed of diverse subsets of bone yond these classic paradigms. Further- ciplines, were unknowingly studying the marrow–derived (Mø) more, similar to mononuclear phagocytes same predominant rMoPh, while often and dendritic cells (DCs).1 Mø are de- in skin, lung, and gut, the predominant, neglecting knowledge from the other dis- fined as tissue-resident phagocytic cells resident renal mononuclear phagocytes cipline. As a result, the use of markers as that contribute to steady-state tissue ho- (rMoPh) simultaneously express markers surrogates for Mø or DC functions within meostasis by clearing apoptotic material traditionally associated with either Mø or the renal mononuclear phagocytic system and producing growth factors. During DCs.4–7 Indeed, as discussed below, many (Table 1) is increasingly questioned.22 infection, they perform antimicrobial ef- research groups have characterized these This debate has also been heightened by fector functions such as phagocytosis rMoPh for over three decades as either lack of specificity of reagents commonly and production of toxic metabolites Mø or DCs, despite growing evidence of used to induce loss of function of DCs or and proinflammatory .1,2 phenotypic and functional overlap, Mø (Table 2). DCs, however, are defined primarily by creating a barrier to cohesive progress in the specialized functions of pre- the field. Published online ahead of print. Publication date sentation and regulation of immune Knowledge of rMoPh began with available at www.jasn.org. effector cells. Classically, DCs collect groundbreaking studies in normal kid- antigenic material in tissues and then neys of mice and rats that identified Correspondence: Dr. Peter J. Nelson, Division of Nephrology and Kidney Research Institute, Health migrate to lymph nodes for the purpose abundant dendritiform cells within the Sciences Center, Room BB1259, 1959 NE Pacific of presenting antigen to naïve T cells.1,3 renal interstitium. These cells were des- Street, Seattle, WA 98195. Email: [email protected]

As summarized in Table 1, however, ignated antigen-presenting DCs on the Copyright © 2012 by the American Society of Mø and DCs within the kidney exhibit basisofMHCclassIIexpression8,9 or Nephrology

194 ISSN : 1046-6673/2302-194 J Am Soc Nephrol 23: 194–203, 2012 www.jasn.org BRIEF REVIEW

Table 1. Distinct and shared functions of macrophages and dendritic cells Transgene reporters of cx3cr1 (CX3CR1) in the kidney and csf1r (CD115) fate map over 90% of Mø rMoPh within the kidney at steady classic functions state.17,18,33 The large majority of + phagocytosis of tissue debris and pathogens CX3CR1 cells is dendritiform rMoPh, mediation of tissue injury and disease progression which form a contiguous network a activated phenotypes exhibit distinct polarized functions throughout the entire renal interstitium contemporary additions of mice, lining the microvascular, peritub- phagocytosis to remodel matrix ular, and periglomerular spaces.17,33,34 support for nephrogenesis and cell fate decisions These predominant rMoPh are definable paracrine role in angiogenesis and epithelial regeneration fi + + fi by a marker pro le: CD11b MHCII polarized function can be brogenic + +/– +/– – potent source of chemokines and cytokinesa CX3CR1 CD11c F4/80 CD103 + mediate anti-inflammatory/immunosuppressive effects through innate responsesa (Figure 1, CD11b rMoPh), which is DCs similar to that of interstitial CD11b-like classic functions DC and tissue macrophages described in phagocytosis of antigen in kidney other nonlymphoid tissues.6,7,16–18 migration to renal lymph nodes Among this CD11b+ rMoPh subset, the to T reported frequency of expression of contemporary additions CD11c (either with or without F4/80 fi antigen presentation within the kidney to restimulate or modify in ltrating T lymphocytes coexpression) has varied between 40% mediate anti-inflammatory/immunosuppressive effects through innate responses and and 90%,15–17,35 indicating that addi- modulation of T lymphocytes (peripheral tolerance)a activated phenotypes exhibit distinct polarized functionsa tional phenotypic characterization of potent source of chemokines and cytokinesa these cells is yet required. In addition, aOverlapping functions. F4/80 is reportedly expressed preferen- tially by medullary rMoPh at steady state.10,17 This ambiguity resulting from con- phenotypes.30–32 In light of these shifting The second definable rMoPh subset, ventional nomenclature, now recognized paradigms, we reappraise here the cur- representing #5% of steady-state in- and intensely debated for the mono- rent knowledge of the renal mononuclear terstitial rMoPh, exhibits the surface 5 2 + – nuclear phagocytic system in general, phagocytic system. We focus on identi- phenotype CD11b MHCII CX3CR1 2 dictates that functional properties are fying gaps and inconsistencies in our CD11c+F4/80 CD103+ (Figure 1, CD103+ required to properly characterize the re- understanding of the ontogeny and func- rMoPh) and is, therefore, not visible in 1 18 nal mononuclear phagocytic system. tion of rMoPh in surveillance, tolerance, CX3CR1-GFP reporter mice. The func- Studies on whether the predominant in- and tissue cytoprotection during steady tion of this second subset in the kidney terstitial rMoPh isolated from kidneys state, as well as , tissue injury, is unknown at present, but similar cells perform like splenic DCs by presenting and repair associated with inflammation. exist in skin, lung, and intestine where antigen and activating naïve T lympho- The emerging role for plasticity within they perform specialized functions, in- cytes or like peritoneal Mø by killing the renal mononuclear phagocytic sys- cluding cross-presentation and induction phagocytosed microbes showed the for- tem is also discussed. Notably, knowl- of regulatory T cells.1,4,6,18,36 Plasmacytoid mer function.15 This finding supports edge of these topics has come primarily DCs, which differ ontogenically, func- other reports suggesting migratory and by studying the mouse as a model organ- tionally, and phenotypically from CD11b+ phagocytic properties more typical of ism, and translation to humans, which rMoPh and CD103+ rMoPh, are difficult DCs.16,23 However, these early studies has been limited,20,21 remains an impor- to detect in normal mouse and human could not anticipate the multitude of tant goal. kidneys and may be absent.20,21 CD11b+ additional Mø and DC functions now rMoPh within normal glomeruli are assigned to the renal mononuclear phago- scarce, more globular-shaped, and cytic system24–30 or the fact that functional ONTOGENY OF MONONUCLEAR seem to lack CD11c and F4/80.15,19,37 plasticity—rather than dichotomy— PHAGOCYTES OF THE KIDNEY The rMoPh of normal human kidneys might exist depending on cues provided IN STEADY STATE show generally similar anatomic distribu- to rMoPh from the environment.22,29,30 tion and population heterogeneity,20,21 Moreover, the renal mononuclear phago- During the past 10 years, studies of specific but not all murine markers (F4/80 and cytic system at steady state changes dra- transgene reporter mice and multicolor CX3CR1) can be used to classify human matically during inflammation, in which flow cytometry of mouse kidney cell prep- rMoPh.Totheextentthatithasbeen resident and recruited rMoPh exhibit arations reveal many characteristics of characterized to date, the phenotype of great heterogeneity and dynamic resident rMoPh during normal health. the predominant human rMoPh displays

J Am Soc Nephrol 23: 194–203, 2012 Renal Mononuclear Phagocytic System 195 BRIEF REVIEW www.jasn.org

Table 2. Features and pitfalls of commonly used strategies to induce loss of function of rMoPh CD11c-diphtheria toxin receptor mouse102 conventional use ablation of CD11c+ rMoPh in vivo to study loss of DC function within the kidney (Table 1) degree of fidelity CD11c is expressed by DCs systemically, not just within the kidney CD11c is expressed by the majority of CD11b+ rMoPh, which may also perform functions associated with Mø CD11c is expressed by other cells such as activated T lymphocytes, alveolar Mø, and splenic marginal zone Mø pitfalls experimental outcomes are confounded by loss of function of other cells that express CD11c, both within and outside the kidney, including rMoPh that may function as Mø (Table 1); as a result, DC-dependent and -independent functions of rMoPh may not be separable, and products from rapid death of CD11c+ cells may also influence immune responses CD11b-diphtheria toxin receptor mouse103 conventional use ablation of CD11b+ rMoPh in vivo to study loss of Mø function within the kidney (Table 1) degree of fidelity CD11b is expressed by Mø systemically, not just within the kidney CD11b is expressed by the majority of CD11c+ rMoPh, which may also perform functions associated with DCs CD11b is expressed by other cells such as , B lymphocytes, and pitfalls experimental outcomes are confounded by the loss of other cells that express CD11b, both within and outside the kidney, including rMoPh that may function as DCs (Table 1); as a result, Mø-dependent and -independent functions of rMoPh may not be separable, and products from rapid death of CD11b+ cells may also influence inflammatory responses Clondronate liposomes104 conventional use ablation of phagocytic rMoPh, often to study loss of phagocytic function within the kidney degree of fidelity all mononuclear phagocytic cells are ablated systemically rMoPh are broadly phagocytic does not discriminate between phagocytic rMoPh performing antigen-dependent versus -independent functions pitfalls experimental outcomes are confounded by the loss of all phagocytic cells, both within and outside the kidney; as a result, DC- and Mø- dependent functions of rMoPh may not be separable, and products from rapid death of phagocytic cells may also influence immune and inflammatory responses the following marker profile: CD11b+ all rMoPh represent the progeny of MDP when injected, expands circulating CD68+BDCA-1+CD14+DC-SIGN+ remains unclear, and recent studies of monocytes.45 Whether the density of – CD11c+MHC II+CD207 .6,20,21 mouse and human hematopoeisis raise rMoPh in normal kidneys also increases Studies of the ontogeny of resident the possibility that some rMoPh origi- with CSF-1 injection is unclear, but in rMoPh in mice show that both major nate from lymphoid progenitors.41 The vivo blockade of CSF-1/CD115 interac- subsets derive from a common hemato- development of the renal mononuclear tions results in depletion of most resident poietic /Mø/DC precursor phagocytic system commences before rMoPh.43 Systemic administration of (MDP), either through a common DC nephrogenesis,42 but whether vestigial, Flt3L, the ligand for CD135, increases precursor, which gives rise only to pre- intrarenal progenitor cell niches con- the density of rMoPh,46,47 but whether DCs and DCs, or through circulating tribute to renewal of rMoPh populations Flt3L is expressed within the kidney is monocytes (Figure 1).1,18,38 Two bone in postnatal life is unknown.1,18 unknown. Nevertheless, isolated kidney marrow–derived monocyte subsets, dis- The renal-derived poietins and other stromal cells secrete other trophic factors tinguished by high or low expression of renal-specific factors that induce MDP for MDPs,48 exemplifying the probable the surface protein Ly6C in mouse and lineages to rMoPh are relatively unclear. role of the renal parenchyma to differen- expression patterns of CD14 and CD16 With the exception of CD103+ rMoPh, tiate the renal mononuclear phagocytic in humans,1 may each contribute to the activation of both the Fms-like tyrosine system.49 resident rMoPh pool at steady state (Fig- kinase 3 receptor (CD135) and CD115 Few studies have addressed cellular ure 1).18,39 However, circulating mono- are required for the development and turnover within the renal mononuclear cytes play a larger precursor role for the maintenance of the vast majority of phagocytic system.50 Bromodeoxyuridine additional rMoPh populations that arise rMoPh (Figure 1).18,43 Colony-stimulating (BrdU) labeling and parabiosis experi- during inflammatory infiltration of the factor 1 (CSF-1), the ligand for CD115, is ments in mice show that 10% of the pre- kidney (discussed below).35,40 Whether expressed by the renal parenchyma44 and dominant CD11b+ rMoPh in recipient

196 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 194–203, 2012 www.jasn.org BRIEF REVIEW

to diverse environmental cues that in- form their subsequent functions. In the normal kidney, rMoPh are responsive to a range of stimuli associated with tissue perturbations such as pathogen- associated molecular patterns, damage- associated molecular patterns (DAMPs), other cell and matrix debris, immuno- globulin, complement components, ox- ygen tension, adenosine, cytokines, chemokines, and cell–cell contact.24–29 Nonetheless, the specific protective roles of each surveillance mechanism and the possibilities that surveillance mechanisms differ by anatomic loca- tion or rMoPh subpopulation have not been extensively investigated to date. Additional studies are needed to increase understanding of how rMoPh respond to cues with varying character- istics and within different renal com- partments. Peripheral tolerance to self- within the kidney that may escape is also likely to be orchestrated + Figure 1. Ontogeny of the renal mononuclear phagocytic system of the mouse at steady by rMoPh. It is known that CD11c state and during inflammation. At steady state, bone marrow–resident MDPs give rise to rMoPh at steady state sample self-antigens common precursors (CDPs) and pre-DCs. Pre-DCs circulate to the healthy derived from tubular filtrate, renal cells, kidney to generate CD103+ rMoPh and a proportion of the resident CD11b+ rMoPh. MDPs or renal matrix.16,53,54 These cells trans- 2 also give rise to circulating Ly-6C+ monocytes and Ly-6C monocytes, which act as ad- port and present antigens to cognate T ditional precursors for resident CD11b+ rMoPh. During renal inflammation, large numbers lymphocytes in renal lymph nodes, sup- of Ly6C+ (inflammatory) monocytes are released from the bone marrow in CCR2/CCL2- plementing additional rMoPh-independent dependent fashion and enter the injured kidney along with a proportion of circulating lymphatic transport and presentation of 2 fi + Ly6C monocytes. In ltrating Ly-6C monocytes serve as precursors for rMoPh with the small filtered antigen, and promote tol- fl characteristics of classically activated (M1) Mø and in ammatory DCs. Alternatively ac- erance to such antigens through mech- tivated (M2) Mø arise from M1 Mø through reprogramming (plasticity). Additional work anisms such as cross-tolerance.53,55 The is required to determine whether M2 Mø or other inflammatory rMoPh develop directly 2 from infiltrating Ly6C monocytes. The growth factor axes, Flt3L/CD135 and CSF-1/CD115, intrarenal sampling, transport, and pre- important in the development of rMoPh are also shown in the figure. sentation of antigen occur through in- terstitial rMoPh, and no studies to date have shown the ability of intraglomeru- mice becomes BrdU-positive within 12 complete turnover of the renal mononu- lar rMoPh to educate T lymphocytes hours. However, it requires 30 days for clear phagocytic system is slow at steady through antigen presentation, whether rMoPh from parabiotic donors to com- state. to tolerance or immunity. Interestingly, pletely disappear after mice are sepa- rMoPh isolated from normal kidneys on rated. Of interest, turnover rates for the basis of CD11c expression induce CD103+ rMoPh are approximately two FUNCTIONS OF rMoPh IN the expansion of regulatory T lympho- times as fast.18 In contrast to these exper- SURVEILLANCE, TOLERANCE, cytes in allogeneic mixed leukocyte re- imental approaches, bone marrow AND TISSUE CYTOPROTECTION actions,46 suggesting the possibility that transplant studies in mice suggest that re- rMoPh propagate regulatory T lympho- population of rMoPh by MDP lineages Surveillance, the anatomic and func- cytes in the periphery at steady state in requires at least 8 weeks and possibly sev- tional readiness to respond to environ- vivo. Aside from hypoxemia,56–58 how- eral months—variability that may be mental cues, is well described for the ever, the cues within the kidney that caused by irradiation techniques and renal mononuclear phagocytic system, may influence rMoPh to enhance or strain of mice.19,35,51,52 In any respect, and rMoPh are often considered sentinels break peripheral tolerance are not well these studies together suggest that of the renal , responding understood.54,55,59

J Am Soc Nephrol 23: 194–203, 2012 Renal Mononuclear Phagocytic System 197 BRIEF REVIEW www.jasn.org

In addition to maintaining tolerance diverse functional phenotypes (Figure With prolonged innate inflammation, to self-antigens, mounting evidence sug- 1).35,40,69 Bone marrow release of Ly-6C+ likely in concert with DAMPs,77–79 these gests that rMoPh at steady state also monocytes and trafficking of these cells CD11c+ rMoPh mature and now stimu- protect or lessen damage to the renal to the injured kidney is triggered by late Th1 effectors.80,81 Studies in models parenchyma after diverse insults through CCR2 ligation.40,72 In glomerular of ureteral obstruction suggest that innate,anti-inflammatory responses.19,60–62 inflammation, a similar influx and ex- this pattern may also influence intra- This finding may reflect the innate phase pansion of rMoPh, functioning primarily renal Th17 responses orchestrated by response by tolerogenic rMoPh utilizing as Mø, is well described.73,74 Whether in- resident interstitial rMoPh.69 Addi- counter-regulators such as single Ig re- traglomerular rMoPh present at the ini- tional research is clearly needed to ceptor-related protein,63 IFN regulatory tiation of injury also migrate to renal define the full spectrum of immune stim- factor 4,64 adenosine 2A receptor,65 hem- lymph nodes for antigen presentation is ulatory properties of the major rMoPh eoxygenase-1,56,62 and Fc g-receptor.66 unknown, but glomerular infiltration by subpopulations as well as any role for Secretion of IL-10 by resident rMoPh rMoPhs with the phenotype of conven- intraglomerular rMoPh in adaptive im- is emerging as one of the key effectors tional and plasmacytoid DCs has been mune responses. of cytoprotection,61,66 raising the pos- described in human lupus nephritis.21,75 In addition to orchestrating other sibility of therapeutically augmenting In any respect, the degree to which in- leukocyte populations that may induce this secretion by rMoPh to enhance cy- flammatory monocyte-derived rMoPh injury, rMoPh may also directly damage toprotection.30,67 Additional research perform the repertoire of Mø and DC the renal parenchyma. As described is needed to determine whether this functions (Figure 1 and Table 1) within above, resident rMoPh elaborate damag- function of the renal mononuclear glomerular or tubulointerstitial com- ing innate cytokines.30,40,70 In addition, phagocytic system represents an active partmentsisnotfullyunderstood.In recruited inflammatory monocytes35,40 response by rMoPh to silence insults the following sections, we summarize that differentiate into inflammatory that should not be perceived as injurious current knowledge of the role of the renal rMoPh also produce oxygen radicals, to the kidney, such as periods of brief mononuclear phagocytic system in active hydrogen peroxide, nitric oxide, IL-1, ischemia or products from normal cellu- renal immunity and inflammation and TNF-a, and other effectors of tissue in- lar turnover.68 their resolution. jury.82–86 The vast majority of reports on this acute phase of injury, whether fo- cused on glomerular or tubulointersti- FROM STEADY STATE TO FUNCTIONS OF rMoPh IN tial compartments, assign responsibility IMMUNITY, TISSUE INJURY, AND for tissue damage to resident or re- TISSUE REPAIR cruited rMoPh polarized to a classically Unlike rMoPh functioning at steady activated (M1, proinflammatory) Mø state, rMoPh during inflammation ex- Therenal mononuclear phagocyticsystem phenotype.27,30 In contrast, later fi- hibithighlycomplex and dynamic respon- plays pivotal roles in proinflammatory brotic phases of injury are orchestrated ses that can determine renal outcomes innate and adaptive immune responses by resident or recruited rMoPh polar- after injury.30–32 Studies of mouse mod- initiated within the kidney. rMoPh resi- ized to an alternatively activated (M2, els of acute kidney injury induced by dent at steady state are induced to rapidly anti-inflammatory and wound healing) endotoxemia, ischemia reperfusion, produce proinflammatory innate cyto- Mø phenotype.29,30,87 acute urinary obstruction, and pyelone- kines (TNF, IL-6, and IL-1) and chemo- In contrast to their role in tissue phritis show that resident rMoPh in the kines (CCL2, CCL5, CXCL10, and injury, rMoPh are also associated with interstitium increase in size, upregulate CXCL2) depending on environmental repair of damaged renal parenchyma MHC and costimulatory proteins, and cues.40,70,74,76 As a result, Ly6C+ mono- during resolution of renal inflamma- secrete inflammatory mediators within cytes and other leukocytes are recruited tion. Seminal studies using clodronate- hours after initiating injury.15,16,69,70 In into the kidney, providing a secondary containing liposomes (Table 2) to ablate some studies, an unknown fraction of and major source of innate media- rMoPh show that recruited glomerular resident rMoPh vacate the interstitium tors.35,40,74 This innate phase, evident rMoPh are required for the spontaneous within 24–48 hours coincident with the during either glomerular or tubulointer- repair that occurs after glomerular injury accumulation of rMoPh in draining re- stitial inflammation, shapes subsequent in the Thy1.1 model of mesangioprolif- nal lymph nodes.16,71 During the same adaptive immune responses by rMoPh, erative GN.88 Ablation and adoptive interval, large numbers of Ly-6C+ (in- although this progression has been more transfer studies conducted later simi- flammatory) monocytes as well as clearly described for interstitial compared larly showed that recruited rMoPh ac- 2 some Ly6C monocytes infiltrate the with glomerular rMoPh. For example, in tively promote repair of injured renal kidney and along with remaining resi- models of GN, resident interstitial CD11c+ tubules.67,87,89–91 These reparative dent rMoPh, undergo activation and rMoPh suppress T helper type 1 (Th1) rMoPh clear DAMPs and other cell and differentiation to progeny rMoPh with lymphocytes recruited early after injury.60 matrix debris, stimulate proliferation of

198 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 194–203, 2012 www.jasn.org BRIEF REVIEW surviving cells through elaboration of wnt ligands, and promote angiogene- sis.92 This biphasic response by re- cruited rMoPh to first injure and then repair renal parenchyma may reflect a process of reprogramming. After injury, resident and recruited rMoPh subpopu- lations exhibit distinct expression of proinflammatory, anti-inflammatory, profibrotic, and reparative factors35,40 suggestive of the presence of heteroge- neous populations of rMoPh fulfilling Figure 2. Proposed model for plasticity of rMoPh during inflammation and repair. During disparate functions. Although func- inflammation, rMoPh are derived from both resident and infiltrating cells. The release of tional heterogeneity is undoubtedly DAMPs provides a proinflammatory stimulus and activates rMoPh to produce injurious present among rMoPh at any given mediators such as TNFa. In the absence of ongoing injury, such rMoPh subsequently un- time after an episode of renal injury, in- dergo an intrinsic phenotypic shift with upregulation of anti-inflammatory and reparative dividual rMoPh can switch from inflam- mediators such as hemeoxygenase-1 and wnt ligands. An anti-inflammatory and reparative matory to reparative phenotypes within phenotype may also be induced by uptake of apoptotic cells or exposure to IL-10 and other the renal interstitium,30,90 a process re- mediators. rMoPh and their precursors are exposed to renal-derived poietins (e.g.,CSF-1 ferred to as plasticity (discussed below). and colony-stimulating factor), which can influence their differ- Regardless of their origin, additional entiation and functional polarization. studies are needed to determine whether reparative rMoPh are retained after kid- neys heal to comprise a long-lasting component of the renal mononuclear Table 3. Roadmap of key topics for future research on the renal mononuclear system at steady state. phagocytic system General identify markers that indicate specific functions of rMoPh PLASTICITY OF FUNCTION OF delineate the functions of glomerular rMoPh from tubulointerstitial rMoPh rMoPh DURING RENAL INJURY correlate murine with human rMoPh subsets and functions AND REPAIR Ontogeny define the precursors and factors that differentiate rMoPh subsets fi As the recent study by Lee et al.90 illus- de ne the developmental heterogeneity of rMoPh along the nephron Surveillance trates, the ability of rMoPh to change define the repertoire of surveillance mechanisms used by rMoPh functions over time may be an impor- define the extent of differential surveillance by rMoPh along the nephron tant general property of the renal mono- Tolerance nuclear phagocytic system (Figure 2). identify the cues that enhance or break peripheral tolerance by rMoPh Classic examples of rMoPh plasticity in- identify the subsets of rMoPh involved in peripheral tolerance clude the maturation of renal DCs to edu- Tissue cytoprotection cate T lymphocytes and the polarization identify the cues that enhance or diminish cytoprotection by rMoPh ofrenalMøtoM1orM2functional identify the subsets of rMoPh involved in cytoprotection phenotypes in response to environ- From steady state to inflammation mental cues.30,93 More recent studies delineate the migratory patterns and chemotactic factors for resident and recruited rMoPh fl show that bone marrow-derived Mø or identify the subsets of rMoPh that promote versus retard in ammation after injury Immunity rMoPh display biphasic expression of fi fl de ne the mechanisms of antigen capture and presentation along the nephron proin ammatory factors followed by define the role of resident versus recruited rMoPh in programming T lymphocytes fl anti-in ammatory and reparative factors Tissue injury in response to challenge with lipo- define the cytotoxic contribution of resident versus recruited rMoPh polysaccharide or ischemic injury, re- define the fibrogenic and fibrolytic contribution of resident versus recruited rMoPh spectively.62,92,94,95 Within the kidney, Tissue repair specific paracrine factors released from define the overlapping programs by rMoPh to support nephrogenesis and tissue repair renal parenchymal cells, such as IL-10, define the interaction of rMoPh with tissue-resident progenitors/stem cells DAMPs, or apoptotic bodies, initiate or Plasticity augment this phenotypic reprogramming determine the differences in plasticity of resident versus recruited rMoPh fi of rMoPh.90,96,97 de ne the molecular mechanisms of plasticity of rMoPh during injury and repair

J Am Soc Nephrol 23: 194–203, 2012 Renal Mononuclear Phagocytic System 199 BRIEF REVIEW www.jasn.org

Table 4. Key conclusions At steady state, subsets of rMoPh exist that coexpress markers previously thought to segregate Mø and DCs. These cells exhibit functional properties considered typical of both DCs and Mø. The renal mononuclear phagocytic system performs important homeostatic and sentinel functions in the steady-state kidney and maintains against self- and innocuous antigens. The renal mononuclear phagocytic system changes dramatically during injury to the kidney, with resident and recruited rMoPh exhibiting functional heterogeneity and plasticity in defense against microbes, mediation of parenchymal injury, and promotion of tissue repair.

The plasticity of rMoPh may also especially in the last 5 years. As this brief by Science Foundation Ireland Grants SFI PI represent inducible reactivation of de- review highlights, however, rMoPh may 06/IN.1/B652 and SFI SRC 09/SRC/B1794 and velopmental programsnormally activein fulfill definitions and functions both of Health Research Board of Ireland Grant HRA differentiating precursors. For example, DCs and Mø, hampering definitive clas- HSR/2010/63. J.H. is supported by United pre-DCs are the immediate precursors sification. Indeed, parallel streams of lit- Kingdom Medical Research Council Grants for CD103+ rMoPh; however, pre-DCs erature have been created that do not G0700330 and G0801235 and the Genzyme can also be induced to repopulate provide a fully integrated body of Renal Innovation Program. C.K. is supported CD11b+ rMoPh.18 The former progeny knowledge to this point. As a roadmap by Deutsche Forschungsgemeinschaft Grants functions solely as DCs, although this for future research, we highlight in Table Ku1063/5, KFO228, and SFBTR57, the EU findinghasnotbeenformallyshown 3someofthespecific areas of investiga- consortium INTRICATE, and the German within the kidney, whereas the latter tion that may yield valuable new in- National Academic Foundation. J.S.D. is sup- progeny clearly harbors subsets per- sights. Engaging in this roadmap will ported by NIH Grants DK73299, DK84077, forming functions as Mø.18 The balance require acknowledging that the com- and DK8739, a Genzyme Research in Progress of poietins produced within the kidney plexity of the renal mononuclear phago- Award, the Nephcure Foundation, and a for rMoPhs, particularly during inflam- cytic system, as with the mononuclear Sponsored Research Agreement with Regulus mation or its resolution, may be a key phagocytic system in general,5 does not Therapeutics. driving force for this altered lineage fit well with simple, binary naming con- commitment.44 Intriguingly, although ventions. Specifically, historic divisions nostudieshavebeendoneonrMoPh that have separated renal investigators DISCLOSURES per se, emerging evidence suggests that into Mø and DC camps are falling, al- None. intrinsic molecular programs of plastic- lowing for some key unifying conclu- 98 ity governed by microRNA (miRNA) sionstobemade(Table4)andpaving REFERENCES determine the commitment and transi- the way for novel, joint approaches to tion of rMoPh to function as DCs or Mø. better understand the role of the renal 1. Geissmann F, Manz MG, Jung S, Sieweke In addition, studies of bone marrow- mononuclear phagocytic system in MH, Merad M, Ley K: Development of derived Mø suggest that epigenetic chro- health and disease of the kidney. monocytes, macrophages, and dendritic – matin modification by histone-modifying cells. Science 327: 656 661, 2010 99,100 2. Gordon S: The macrophage. Bioessays 17: enzymes, and gene-silencing 977–986, 1995 101 polycomb group proteins may be key ACKNOWLEDGMENTS 3. Banchereau J, Steinman RM: Dendritic cells regulators of Mø activation status. Future and the control of immunity. Nature 392: research on plasticity within the renal The authors acknowledge that many more 245–252, 1998 mononuclear phagocyte system may excellentoriginalresearch studies of relevance 4. Heath WR, Carbone FR: Dendritic cell sub- sets in primary and secondary re- help to develop a unifying nomenclature to the understanding of the renal mono- sponses at body surfaces. Nat Immunol 10: that represents the function of rMoPh. nuclear phagocytic system have been pub- 1237–1244, 2009 lished than could be cited in this brief review 5. Geissmann F, Gordon S, Hume DA, Mowat article. AM, Randolph GJ: Unravelling mononuclear A ROADMAP FOR FUTURE P.J.N. is supported by US National In- phagocyte heterogeneity. Nat Rev Immunol 10: 453–460, 2010 PROGRESS stitutes of Health (NIH) Grants DK83375, 6. Guilliams M, Henri S, Tamoutounour S, DK83375S1, and DK83912 and a Sponsored Ardouin L, Schwartz-Cornil I, Dalod M, Significant progress in understanding Research Agreement with Genzyme. A.J.R. is Malissen B: From skin dendritic cells to the renal mononuclear phagocytic sys- supported by the European Union (EU) FP7 asimplified classification of human and tem has been achieved over the past Marie Curie Program MACRORIEN and mouse dendritic cell subsets. Eur J Immunol 40: 2089–2094, 2010 three decades. Many typical DC and TranSVIR, the EU FP7 Health Program 7. Hume DA: Macrophages as APC and the Mø-associated functions of the major INTRICATE, and the Austrian Vienna Science dendritic cell myth. JImmunol181: 5829– rMoPh subsets have been described, and Technology Fund. M.D.G. is supported 5835, 2008

200 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 194–203, 2012 www.jasn.org BRIEF REVIEW

8. Hart DN, Fabre JW: Major histocompatibil- normal and pathological conditions. Kidney Heath WR: Cross-presentation of viral and ity complex antigens in rat kidney, ureter, Int 71: 1001–1008, 2007 self antigens by skin-derived CD103+ den- and bladder. Localization with monoclonal 21. Segerer S, Heller F, Lindenmeyer MT, dritic cells. Nat Immunol 10: 488–495, 2009 and demonstration of Ia-positive Schmid H, Cohen CD, Draganovici D, 37. Tesch GH, Maifert S, Schwarting A, Rollins dendritic cells. Transplantation 31: 318–325, Mandelbaum J, Nelson PJ, Gröne HJ, Gröne BJ, Kelley VR: Monocyte chemoattractant 1981 EF, Figel AM, Nössner E, Schlöndorff D: protein 1-dependent leukocytic infiltrates 9. Hart DN, Fuggle SV, Williams KA, Fabre JW, Compartment specific expression of den- are responsible for in Ting A, Morris PJ: Localization of HLA-ABC dritic cell markers in human glomerulone- MRL-Fas(lpr) mice. JExpMed190: 1813– and DR antigens in human kidney. Trans- phritis. Kidney Int 74: 37–46, 2008 1824, 1999 plantation 31: 428–433, 1981 22. Ferenbach D, Hughes J: Macrophages and 38.FoggDK,SibonC,MiledC,JungS, 10. Hume DA, Gordon S: Mononuclear phago- dendritic cells: what is the difference? Kid- Aucouturier P, Littman DR, Cumano A, cyte system of the mouse defined by immu- ney Int 74: 5–7, 2008 Geissmann F: A clonogenic bone marrow nohistochemical localization of antigen F4/80. 23. Austyn JM, Hankins DF, Larsen CP, Morris progenitor specific for macrophages and Identification of resident macrophages in re- PJ, Rao AS, Roake JA: Isolation and char- dendritic cells. Science 311: 83–87, 2006 nal medullary and cortical interstitium and acterization of dendritic cells from mouse 39. Geissmann F, Jung S, Littman DR: Blood the juxtaglomerular complex. J Exp Med heart and kidney. JImmunol152: 2401– monocytes consist of two principal subsets 157: 1704–1709, 1983 2410, 1994 with distinct migratory properties. Immunity 11. Schreiner GF, Kiely JM, Cotran RS, Unanue 24. Teteris SA, Engel DR, Kurts C: Homeostatic 19: 71–82, 2003 ER: Characterization of resident glomerular and pathogenic role of renal dendritic cells. 40. Li L, Huang L, Sung SS, Vergis AL, Rosin DL, cells in the rat expressing Ia determinants Kidney Int 80: 139–145, 2011 Rose CE Jr, Lobo PI, Okusa MD: The che- and manifesting genetically restricted in- 25. Panzer U, Kurts C: T cell cross-talk with kid- mokine receptors CCR2 and CX3CR1 me- teractions with lymphocytes. J Clin Invest ney dendritic cells in glomerulonephritis. diate monocyte/macrophage trafficking in 68: 920–931, 1981 J Mol Med (Berl) 88: 19–26, 2010 kidney ischemia-reperfusion injury. Kidney 12. Schreiner GF, Cotran RS: Localization of an 26. John R, Nelson PJ: Dendritic cells in the Int 74: 1526–1537, 2008 Ia-bearing glomerular cell in the mesan- kidney. JAmSocNephrol18: 2628–2635, 41. Dorshkind K: Not a split decision for human gium. JCellBiol94: 483–488, 1982 2007 hematopoiesis. Nat Immunol 11: 569–570, 13. Kaissling B, Le Hir M: Characterization and 27. Duffield JS: Macrophages and immuno- 2010 distribution of interstitial cell types in the logic inflammation of the kidney. Semin 42. Rae F, Woods K, Sasmono T, Campanale N, renal cortex of rats. Kidney Int 45: 709–720, Nephrol 30: 234–254, 2010 Taylor D, Ovchinnikov DA, Grimmond SM, 1994 28. Rees AJ: Monocyte and macrophage bi- Hume DA, Ricardo SD, Little MH: Charac- 14. Kaissling B, Hegyi I, Loffing J, Le Hir M: ology: an overview. Semin Nephrol 30: terisation and trophic functions of murine Morphology of interstitial cells in the healthy 216–233, 2010 embryonic macrophages based upon the kidney. Anat Embryol (Berl) 193: 303–318, 29. Vernon MA, Mylonas KJ, Hughes J: Macro- use of a Csf1r-EGFP transgene reporter. 1996 phages and renal fibrosis. Semin Nephrol Dev Biol 308: 232 –246, 2007 15. Krüger T, Benke D, Eitner F, Lang A, Wirtz 30: 302–317, 2010 43. MacDonald KPA, Palmer JS, Cronau S, M, Hamilton-Williams EE, Engel D, Giese B, 30. Wang Y, Harris DCH: Macrophages in renal Seppanen E, Olver S, Raffelt NC, Kuns R, Müller-Newen G, Floege J, Kurts C: Iden- disease. JAmSocNephrol22: 21–27, 2011 Pettit AR, Clouston A, Wainwright B, tification and functional characterization 31. Rodríguez-Iturbe B, Pons H, Herrera-Acosta Branstetter D, Smith J, Paxton RJ, Cerretti of dendritic cells in the healthy murine J, Johnson RJ: Role of immunocompetent DP, Bonham L, Hill GR, Hume DA: An anti- kidney and in experimental glomerulo- cells in nonimmune renal diseases. Kidney body against the colony-stimulating factor 1 nephritis. J Am Soc Nephrol 15: 613–621, Int 59: 1626–1640, 2001 receptor depletes the resident subset of 2004 32. Swaminathan S, Griffin MD: First responders: monocytes and tissue- and tumor-associated 16. Dong X, Swaminathan S, Bachman LA, understanding monocyte-lineage trafficin macrophages but does not inhibit inflam- Croatt AJ, Nath KA, Griffin MD: Antigen the acutely injured kidney. Kidney Int 74: mation. Blood 116: 3955–3963, 2010 presentation by dendritic cells in renal 1509–1511, 2008 44. Jang MH, Herber DM, Jiang X, Nandi S, Dai lymph nodes is linked to systemic and local 33. Sasmono RT, Oceandy D, Pollard JW, Tong XM, Zeller G, Stanley ER, Kelley VR: Distinct injury to the kidney. Kidney Int 68: 1096– W, Pavli P, Wainwright BJ, Ostrowski MC, in vivo roles of colony-stimulating factor-1 1108, 2005 Himes SR, Hume DA: A macrophage col- isoforms in renal inflammation. JImmunol 17. Soos TJ, Sims TN, Barisoni L, Lin K, Littman ony-stimulating factor receptor-green fluo- 177: 4055–4063, 2006 DR, Dustin ML, Nelson PJ: CX3CR1+ in- rescent protein transgene is expressed 45. Misawa E, Sakurai T, Yamada M, Tamura Y, terstitial dendritic cells form a contiguous throughout the mononuclear phagocyte Motoyoshi K: Administration of macro- network throughout the entire kidney. Kid- system of the mouse. Blood 101: 1155–1163, phage colony-stimulating factor mobilized ney Int 70: 591–596, 2006 2003 both CD11b+CD11c+ cells and NK1.1+ 18. Ginhoux F, Liu K, Helft J, Bogunovic M, 34. Kaissling B, Le Hir M: The renal cortical in- cells into peripheral blood. Int Immuno- Greter M, Hashimoto D, Price J, Yin N, terstitium: morphological and functional pharmacol 4: 791–803, 2004 Bromberg J, Lira SA, Stanley ER, Nussenzweig aspects. Histochem Cell Biol 130: 247–262, 46. Coates PT, Duncan FJ, Colvin BL, Wang Z, M, Merad M: The origin and development 2008 Zahorchak AF, Shufesky WJ, Morelli AE, of nonlymphoid tissue CD103+ DCs. JExp 35. Lin SL, Castaño AP, Nowlin BT, Lupher ML Thomson AW: In vivo-mobilized kidney Med 206: 3115–3130, 2009 Jr, Duffield JS: Bone marrow Ly6Chigh dendritic cells are functionally immature, 19. Tadagavadi RK, Reeves WB: Renal dendritic monocytes are selectively recruited to in- subvert alloreactive T-cell responses, and cells ameliorate nephrotoxic acute kidney jured kidney and differentiate into func- prolong organ allograft survival. Trans- injury. J Am Soc Nephrol 21: 53–63, 2010 tionally distinct populations. JImmunol plantation 77: 1080–1089, 2004 20. Woltman AM, de Fijter JW, Zuidwijk K, Vlug 183: 6733–6743, 2009 47. Morelli AE, Coates PT, Shufesky WJ, AG, Bajema IM, van der Kooij SW, van Ham 36. Bedoui S, Whitney PG, Waithman J, Eidsmo Barratt-Boyes SM, Fung JJ, Demetris AJ, V, van Kooten C: Quantification of dendritic L, Wakim L, Caminschi I, Allan RS, Wojtasiak Thomson AW: Growth factor-induced mo- cell subsets in human renal tissue under M, Shortman K, Carbone FR, Brooks AG, bilization of dendritic cells in kidney and

J Am Soc Nephrol 23: 194–203, 2012 Renal Mononuclear Phagocytic System 201 BRIEF REVIEW www.jasn.org

liver of rhesus macaques: implications for HJ, Kurts C: Renal dendritic cells stimulate 72. Serbina NV, Pamer EG: Monocyte emigra- transplantation. Transplantation 83: 656– IL-10 production and attenuate nephrotoxic tion from bone marrow during bacterial 662, 2007 nephritis. JAmSocNephrol19: 527–537, infection requires signals mediated by 48. Huang Y, Johnston P, Zhang B, Zakari A, 2008 chemokine receptor CCR2. Nat Immunol 7: Chowdhry T, Smith RR, Marbán E, Rabb H, 61. Tadagavadi RK, Reeves WB: Endogenous 311–317, 2006 Womer KL: Kidney-derived stromal cells IL-10 attenuates cisplatin nephrotoxicity: 73. Erwig LP, Rees AJ: Macrophage activation modulate dendritic and T cell responses. role of dendritic cells. J Immunol 185: 4904– and programming and its role for macro- JAmSocNephrol20: 831–841, 2009 4911, 2010 phage function in glomerular inflamma- 49. Svensson M, Kaye PM: Stromal-cell regula- 62. Ferenbach DA, Nkejabega NC, McKay J, tion. Kidney Blood Press Res 22: 21–25, tion of dendritic-cell differentiation and Choudhary AK, Vernon MA, Beesley MF, 1999 function. Trends Immunol 27: 580–587, 2006 Clay S, Conway BC, Marson LP, Kluth DC, 74. Holdsworth SR, Tipping PG: Leukocytes in 50. Helft J, Ginhoux F, Bogunovic M, Merad M: Hughes J: The induction of macrophage glomerular injury. Semin Immunopathol 29: Origin and functional heterogeneity of non- hemeoxygenase-1 is protective during 355–374, 2007 lymphoid tissue dendritic cells in mice. Im- acute kidney injury in aging mice. Kidney Int 75. Bagavant H, Deshmukh US, Wang H, Ly T, munol Rev 234: 55–75, 2010 79: 966–976, 2011 Fu SM: Role for nephritogenic T cells 51. McKenzie JL, Beard ME, Hart DN: De- 63. Lech M, Avila-Ferrufino A, Allam R, Segerer in lupus glomerulonephritis: progression pletion of donor kidney dendritic cells pro- S, Khandoga A, Krombach F, Garlanda C, to renal failure is accompanied by T cell longs graft survival. Transplant Proc 16: Mantovani A, Anders HJ: Resident den- activation and expansion in regional 948–951, 1984 dritic cells prevent postischemic acute renal lymph nodes. J Immunol 177: 8258–8265, 52. Stein-Oakley AN, Jablonski P, Kraft N, failure by help of single Ig IL-1 receptor- 2006 Biguzas M, Howard BO, Marshall VC, related protein. J Immunol 183: 4109– 76. Tittel AP, Heuser C, Ohliger C, Knolle PA, Thomson NM: Differential irradiation effects 4118, 2009 Engel DR, Kurts C: Kidney dendritic cells on rat interstitial dendritic cells. Transplant 64. Lassen S, Lech M, Römmele C, Mittruecker induce innate immunity against bacterial Proc 23: 632–634, 1991 HW, Mak TW, Anders HJ: Ischemia re- pyelonephritis. JAmSocNephrol22: 53. Lukacs-Kornek V, Burgdorf S, Diehl L, perfusion induces IFN regulatory factor 4 in 1435–1441, 2011 Specht S, Kornek M, Kurts C: The kidney- renal dendritic cells, which suppresses pos- 77. Schaefer L, Babelova A, Kiss E, Hausser HJ, renal lymph node-system contributes to tischemic inflammation and prevents acute Baliova M, Krzyzankova M, Marsche G, cross-tolerance against innocuous circulat- renal failure. J Immunol 185: 1976–1983, Young MF, Mihalik D, Götte M, Malle E, ing antigen. J Immunol 180: 706–715, 2008 2010 Schaefer RM, Gröne HJ: The matrix com- 54. Macconi D, Chiabrando C, Schiarea S, 65. Day YJ, Huang L, Ye H, Linden J, Okusa MD: ponent biglycan is proinflammatory and Aiello S, Cassis L, Gagliardini E, Noris M, Renal ischemia-reperfusion injury and aden- signals through Toll-like receptors 4 and 2 in Buelli S, Zoja C, Corna D, Mele C, Fanelli R, osine 2A receptor-mediated tissue protec- macrophages. JClinInvest115: 2223– Remuzzi G, Benigni A: Proteasomal pro- tion: role of macrophages. Am J Physiol 2233, 2005 cessing of albumin by renal dendritic cells Renal Physiol 288: F722–F731, 2005 78. Lang A, Benke D, Eitner F, Engel D, Ehrlich generates antigenic peptides. JAmSoc 66. Castano AP, Lin SL, Surowy T, Nowlin BT, S, Breloer M, Hamilton-Williams E, Specht Nephrol 20: 123–130, 2009 Turlapati SA, Patel T, Singh A, Li S, Lupher S, Hoerauf A, Floege J, von Bonin A, Kurts 55. EdgttonKL,KausmanJY,LiM,O’Sullivan K, ML Jr, Duffield JS: Serum amyloid P inhibits C: Heat shock protein 60 is released in Lo C, Hutchinson P, Yagita H, Holdsworth fibrosis through Fc gamma R-dependent immune-mediated glomerulonephritis and SR, Kitching AR: Intrarenal antigens activate monocyte-macrophage regulation in vivo. aggravates disease: in vivo evidence for an CD4+ cells via co-stimulatory signals from Sci Transl Med 1: 5ra13, 2009 immunologic danger signal. JAmSoc dendritic cells. J Am Soc Nephrol 19: 515– 67. Cao Q, Wang Y, Zheng D, Sun Y, Wang Y, Nephrol 16: 383–391, 2005 526, 2008 Lee VWS, Zheng G, Tan TK, Ince J, 79. Wu H, Ma J, Wang P, Corpuz TM, 56. Kotsch K, Martins PN, Klemz R, Janssen U, Alexander SI, Harris DCH: IL-10/TGF-beta- Panchapakesan U, Wyburn KR, Chadban SJ: Gerstmayer B, Dernier A, Reutzel-Selke A, modified macrophages induce regulatory HMGB1 contributes to kidney ischemia re- Kuckelkorn U, Tullius SG, Volk HD: Heme T cells and protect against adriamycin ne- perfusion injury. J Am Soc Nephrol 21: oxygenase-1 ameliorates ischemia/reperfusion phrosis. J Am Soc Nephrol 21: 933–942, 1878–1890, 2010 injury by targeting dendritic cell maturation 2010 80. Heymann F, Meyer-Schwesinger C, Hamilton- and migration. Antioxid Redox Signal 9: 2049– 68. Manicassamy S, Pulendran B: Dendritic cell Williams EE, Hammerich L, Panzer U, Kaden 2063, 2007 control of tolerogenic responses. Immunol S, Quaggin SE, Floege J, Gröne HJ, Kurts 57. Cho WY, Choi HM, Lee SY, Kim MG, Kim Rev 241: 206–227, 2011 C: Kidney dendritic cell activation is re- HK, Jo SK: The role of Tregs and CD11c(+) 69. Dong X, Bachman LA, Miller MN, Nath KA, quired for progression of renal disease in a macrophages/dendritic cells in ischemic Griffin MD: Dendritic cells facilitate accu- mouse model of glomerular injury. JClin preconditioning of the kidney. Kidney Int mulation of IL-17 T cells in the kidney fol- Invest 119: 1286–1297, 2009 78: 981–992, 2010 lowing acute renal obstruction. Kidney Int 81. Hochheiser K, Engel DR, Hammerich L, 58. George JF, Braun A, Brusko TM, Joseph R, 74: 1294–1309, 2008 Heymann F, Knolle PA, Panzer U, Kurts C: Bolisetty S, Wasserfall CH, Atkinson MA, 70. Dong X, Swaminathan S, Bachman LA, Croatt Kidney dendritic cells become pathogenic Agarwal A, Kapturczak MH: Suppression by AJ, Nath KA, Griffin MD: Resident dendritic during crescentic glomerulonephritis with CD4+CD25+ regulatory T cells is depen- cells are the predominant TNF-secreting proteinuria. JAmSocNephrol22: 306–316, dent on expression of heme oxygenase-1 in cell in early renal ischemia-reperfusion in- 2011 antigen-presenting cells. Am J Pathol 173: jury. Kidney Int 71: 619– 628, 2007 82. Chavele KM, Martinez-Pomares L, Domin J, 154–160, 2008 71. Roake JA, Rao AS, Morris PJ, Larsen CP, Pemberton S, Haslam SM, Dell A, Cook HT, 59. Rees A: Cross dendritic cells anger T cells Hankins DF, Austyn JM: Dendritic cell loss Pusey CD, Gordon S, Salama AD: Mannose after kidney injury. JAmSocNephrol20: 3– from nonlymphoid tissues after systemic receptor interacts with Fc receptors and is 5, 2009 administration of lipopolysaccharide, tu- critical for the development of crescentic 60. Scholz J, Lukacs-Kornek V, Engel DR, mor necrosis factor, and interleukin 1. JExp glomerulonephritis in mice. J Clin Invest Specht S, Kiss E, Eitner F, Floege J, Groene Med 181: 2237–2247, 1995 120: 1469–1478, 2010

202 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 194–203, 2012 www.jasn.org BRIEF REVIEW

83. Wang Y, Wang Y, Cao Q, Zheng G, Lee VW, macrophage phenotypes contribute to 99. De Santa F, Totaro MG, Prosperini E, Zheng D, Li X, Tan TK, Harris DC: By homing kidney injury and repair. J Am Soc Nephrol Notarbartolo S, Testa G, Natoli G: The his- to the kidney, activated macrophages po- 22: 317–326, 2011 tone H3 lysine-27 demethylase Jmjd3 links tently exacerbate renal injury. Am J Pathol 91. Menke J, Iwata Y, Rabacal WA, Basu R, inflammation to inhibition of polycomb- 172: 1491–1499, 2008 Yeung YG, Humphreys BD, Wada T, mediated gene silencing. Cell 130: 1083– 84. DuffieldJS,ErwigLP,WeiX,LiewFY,Rees Schwarting A, Stanley ER, Kelley VR: CSF-1 1094, 2007 AJ, Savill JS: Activated macrophages direct signals directly to renal tubular epithelial 100. Satoh T, Takeuchi O, Vandenbon A, Yasuda apoptosis and suppress mitosis of me- cells to mediate repair in mice. J Clin Invest K, Tanaka Y, Kumagai Y, Miyake T, sangial cells. J Immunol 164: 2110–2119, 119: 2330–2342, 2009 Matsushita K, Okazaki T, Saitoh T, Honma K, 2000 92. Lin SL, Li B, Rao S, Yeo EJ, Hudson TE, Matsuyama T, Yui K, Tsujimura T, Standley 85. Jansen A, Cook T, Taylor GM, Largen P, Nowlin BT, Pei H, Chen L, Zheng JJ, Carroll DM, Nakanishi K, Nakai K, Akira S: The Riveros-Moreno V, Moncada S, Cattell V: TJ, Pollard JW, McMahon AP, Lang RA, Jmjd3-Irf4 axis regulates M2 macrophage Induction of nitric oxide synthase in rat im- Duffield JS: Macrophage Wnt7b is critical polarization and host responses against hel- mune complex glomerulonephritis. Kidney for kidney repair and regeneration. Proc minth infection. Nat Immunol 11: 936–944, Int 45: 1215–1219, 1994 Natl Acad Sci USA 107: 4194–4199, 2010 2010 86. Tipping PG, Leong TW, Holdsworth SR: 93. Williams TM, Little MH, Ricardo SD: Mac- 101. Sienerth AR, Scheuermann C, Galmiche A, Tumor necrosis factor production by glo- rophages in renal development, injury, and Rapp UR, Becker M: Polycomb group pro- merular macrophages in anti-glomerular repair. Semin Nephrol 30: 255–267, 2010 tein Bmi1 negatively regulates IL-10 expres- basement membrane glomerulonephritis in 94. Lucas M, Stuart LM, Savill J, Lacy-Hulbert A: sion in activated macrophages. Immunol Cell rabbits. Lab Invest 65: 272–279, 1991 Apoptotic cells and innate immune stimuli Biol 89: 812–816, 2011 87. Cao Q, Wang C, Zheng D, Wang Y, Lee combine to regulate macrophage cyto- 102. JungUnutmaz S, Wong D, Sano P, De los VWS, Wang YM, Zheng G, Tan TK, Yu D, kine secretion. J Immunol 171: 2610–2615, Santos G, Sparwasser K, Wu T, Vuthoori S, Alexander SI, Harris DCH, Wang Y: IL-25 2003 Ko S, Zavala K, Pamer F, Littman EG DR, induces M2 macrophages and reduces re- 95. Vijayan V, Baumgart-Vogt E, Naidu S, Qian G, Lang RA: In vivo depletion of CD11c+ nal injury in proteinuric kidney disease. Immenschuh S: Bruton’s tyrosine kinase is re- dendritic cells abrogates priming of CD8+ J Am Soc Nephrol 22: 1229–1239, 2011 quired for TLR-dependent heme oxygenase-1 T cells by exogenous cell-associated anti- 88. Westerhuis R, van Straaten SC, van gene activation via Nrf2 in macrophages. J gens. Immunity 17: 211–220, 2002 Dixhoorn MG, van Rooijen N, Verhagen NA, Immunol 187: 817–827, 2011 103. Duffield JS, Forbes SJ, Constandinou CM, Dijkstra CD, de Heer E, Daha MR: Distinc- 96. Wang Y, Tay YC, Harris DC: Proximal tubule Clay S, Partolina M, Vuthoori S, Wu S, Lang tive roles of and monocytes in cells stimulated by lipopolysaccharide in- R, Iredale JP: Selective depletion of mac- anti-thy-1 nephritis. Am J Pathol 156: 303– hibit macrophage activation. Kidney Int 66: rophages reveals distinct, opposing roles 310, 2000 655–662, 2004 during liver injury and repair. J Clin Invest 89. Vinuesa E, Hotter G, Jung M, Herrero- 97. Savill J, Dransfield I, Gregory C, Haslett C: A 115: 56–65, 2005 Fresneda I, Torras J, Sola A: Macrophage blast from the past: clearance of apoptotic 104. van Rooijen N, van Nieuwmegen R: Elimina- involvement in the kidney repair phase af- cells regulates immune responses. Nat Rev tion of phagocytic cells in the spleen after in- ter ischaemia/reperfusion injury. JPathol Immunol 2: 965–975, 2002 travenous injection of liposome-encapsulated 214: 104–113, 2008 98. Bi Y, Liu G, Yang R: MicroRNAs: novel reg- dichloromethylene diphosphonate. An 90. Lee S, Huen S, Nishio H, Nishio S, Lee HK, ulators during the immune response. JCell enzyme-histochemical study. Cell Tissue Res Choi BS, Ruhrberg C, Cantley LG: Distinct Physiol 218: 467–472, 2009 238: 355–358, 1984

J Am Soc Nephrol 23: 194–203, 2012 Renal Mononuclear Phagocytic System 203