The MIF Receptor CD74 in Diabetic Podocyte Injury

JASN Express. Published on October 8, 2008 as doi: 10.1681/ASN.2008020194

BASIC RESEARCH www.jasn.org

Maria Dolores Sanchez-Nin˜o,* Ana Belen Sanz,* Pekka Ihalmo,† Markus Lassila,‡ ʈ ʈ Harry Holthofer,§ Sergio Mezzano, Claudio Aros, Per-Henrik Groop,† Moin A. Saleem,¶ Peter W. Mathieson,¶ Robert Langham,** Matthias Kretzler,†† Viji Nair,†† Kevin V. Lemley,‡‡ ʈʈ Robert G. Nelson,§§ Eero Mervaala, Deborah Mattinzoli,¶¶ Maria Pia Rastaldi,¶¶ Marta Ruiz-Ortega,* Jose Luis Martin-Ventura,* Jesus Egido,* and Alberto Ortiz*

*Fundacion Jimenez Diaz, Universidad Autonoma de Madrid, Fundacio´n Renal In˜igo Alvarez de Toledo, Madrid, Spain; †Folkha¨lsan Institute of Genetics, Folkha¨lsan Research Center, University of Helsinki, and Division of Nephrology, Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland; ‡Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, Finland; §Center for BioAnalytical Sciences, ʈ Dublin City University, Ireland; Division of Nephrology, Universidad Austral, Valdivia, Chile; ¶Academic and Children’s Renal Unit, University of Bristol, Bristol, United Kingdom; **Department of Medicine, St. Vincent’s Hospital, University of Melbourne, Victoria, Australia; ††Division of Nephrology, University of Michigan, Ann Arbor, Michigan; ‡‡Division of Nephrology, Childrens Hospital Los Angeles, Los Angeles, California; §§Phoenix Epidemiology and Clinical Research Branch, National Institute of Diabetes and Digestive and Kidney Diseases, ʈʈ Phoenix, Arizona; Institute of Biomedicine/Pharmacology, University of Helsinki, Finland; and ¶¶Fondazione D’Amico per la Ricerca sulle Malattie Renali, Milan, Italy

ABSTRACT Although metabolic derangement plays a central role in diabetic nephropathy, a better understanding of secondary mediators of injury may lead to new therapeutic strategies. Expression of macrophage migration inhibitory factor (MIF) is increased in experimental diabetic nephropathy, and increased tubulointerstitial mRNA expression of its receptor, CD74, has been observed in human diabetic nephropathy. Whether CD74 transduces MIF signals in podocytes, however, is unknown. Here, we found glomerular and tubulointerstitial CD74 mRNA expression to be increased in Pima Indians with type 2 diabetes and diabetic nephropathy. Immunohistochemistry confirmed the increased glomerular and tubular expression of CD74 in clinical and experimental diabetic nephropathy and localized glomerular CD74 to podocytes. In cultured human podocytes, CD74 was expressed at the cell surface, was upregulated by high concentrations of glucose and TNF-␣, and was activated by MIF, leading to phosphorylation of extracellular signal–regulated kinase 1/2 and p38. High glucose also induced CD74 expression in a human proximal tubule cell line (HK2). In addition, MIF induced the expression of the inflammatory mediators TRAIL and monocyte chemoattractant protein 1 in podocytes and HK2 cells in a p38-dependent manner. These data suggest that CD74 acts as a receptor for MIF in podocytes and may play a role in the pathogenesis of diabetic nephropathy.

J Am Soc Nephrol ●●: –, 2009. doi: 10.1681/ASN.2008020194

Diabetic nephropathy (DN) is one of the major diabetic nephropathy and induces tubular cell apo- complications of diabetes and the most common ptosis.3 A better understanding of secondary medi- cause of ESRD. Hyperglycemia activates secondary mediators that lead to DN. In this regard, inflam- Received February 15, 2008. Accepted July 16, 2008. matory cytokines may be critical in the develop- Published online ahead of print. Publication date available at ment of microvascular diabetic complications and www.jasn.org. 1 nephropathy. Monocyte chemoattractant protein Correspondence: Dr. Alberto Ortiz, Unidad de Dia´lisis, Fundacioń 1 (MCP-1) is considered a diagnostic marker and Jimeńez Dıáz,Avda Reyes Cato´licos 2, 28040 Madrid, Spain. Phone: therapeutic target in DN.2 TNF-related apoptosis ϩ34-915-504940; Fax: ϩ34-915-442636; E-mail: [email protected] inducing ligand (TRAIL) expression is increased in Copyright ᮊ ●●●● by the American Society of Nephrology

J Am Soc Nephrol ●●: –, 2009 ISSN : 1046-6673/●●00- 1 BASIC RESEARCH www.jasn.org

Figure 1. Increased CD74 expression in human DN biopsies. (A) Transcriptomic analysis of CD74 mRNA in glomerular and tubulointerstitial compartments of living kidney donors (control, n ϭ 6) and Pima Indians with diabetes (DN, n ϭ 20) and MCD (n ϭ 4). Data are means Ϯ SD; #P Ͻ 0.0001 versus control. (B) Semiquan- tification of CD74 immunoreactivity within the glomerular and tubular compartments in the kidney biopsies from control subjects (n ϭ 5) and patients with DN (n ϭ 5) and MCD (n ϭ 4). **P ϭ 0.01 versus control. (C) Representative image of a control and a DN sample. A marked increase in CD74 staining in DN is visible in both glomerular and tubular cells. (D) Co-local- ization of CD74 and synaptopodin in podocytes in a glomerulus from DN biopsy. ators of injury may lead to the design of new therapeutic strat- MIF.14 CD74 antigen (invariant polypeptide of MHC, HLA-DR egies. ␥) is a type II transmembrane protein that plays a critical role in Macrophage migration inhibitory factor (MIF) is a widely class II MHC antigen processing.15 In addition, CD74 may trans- expressed pleiotropic cytokine, exhibiting a broad range of im- duce MIF-induced activation of the extracellular signal–regulated mune and inflammatory activities.4 MIF has been implicated in kinase 1/2 (ERK1/2) mitogen-activated protein kinase (MAPK) renal injury. A marked increase in MIF was noted in glomeruli cascade and promotes cell proliferation and prostaglandin E2 pro- and tubules in proliferative forms of human glomerulonephritis, duction in leukocytes and fibroblasts.14 On the basis of the obser- where de novo MIF expression was localized to glomerular endo- vation that CD74 expression was increased in transcriptome stud- thelial and mesangial cells.5 Cultured tubular and mesangial cells ies of human DN, we investigated the role of CD74 in transducing produce MIF in response to inflammatory stimuli.6,7 Enhanced MIF signals in human podocytes. CD74 was found to be increased podocyte expression of a MIF transgene induces podocyte injury in podocytes and tubular cells from humans and animals with in mice, independent of leukocyte infiltration, suggesting auto- DN, and cultured podocytes expressed CD74 that mediated MIF crine effects of MIF on podocytes.8 Urine MIF concentration is activation of MAPK as shown by knockdown experiments. Fur- significantly increased in proliferative forms of glomerulonephri- thermore, MIF promoted MCP-1 and TRAIL expression in tis and correlates with the degree of renal injury.9 podocytes and tubular cells. Evidence has also linked MIF to DN. MIF expression is increased in experimental DN.10 Serum MIF concentration are elevated in individuals with type 2 diabetes11 and also locally at RESULTS sites injured by diabetes in patients with proliferative diabetic retinopathy.12 Identification of CD74 as an Overexpressed Gene in Although MIF was identified as a soluble, T cell–derived Human DN factor in 1966,13 the nature of its membrane receptor was un- Increased renal interstitial CD74 mRNA expression (a 1.92- known until 2003, when CD74 was shown to be a receptor for fold increase over living kidney donor controls; P ϭ 0.0015)

2 Journal of the American Society of Nephrology J Am Soc Nephrol ●●: –, 2009 www.jasn.org BASIC RESEARCH had been noted by transcriptomic analysis of human DN biopsy with the fact that CD74 may undergo regulated intramem- tubulointerstitium.3,16 We now report similar findings in the renal brane proteolysis and migrate to the nucleus.18 Of more rele- tubulointerstitium of a different cohort of DN patients: Pima In- vance to the present research, we also observed CD74 at the cell dians with DN, on whom CD74 mRNA was increased 1.95-fold membrane. Staining for nephrin confirmed the differentiated over living kidney donor control subjects (PϽ0.0001; Figure 1A). state of podocytes (Figure 3E). Flow cytometry confirmed the In addition, a 1.64-fold increase in CD74 mRNA was noted in the presence of CD74 at the cell surface (Figure 3F), supporting its glomeruli from Pima Indians with diabetes (P Ͻ 0.0001; Figure possible role as a receptor. 1A). By contrast, changes in minimal-change disease (MCD) were Ͻ1.5-fold in both compartments (Figure 1A). Actions of MIF in Cultured Podocytes Are Mediated In an independent second group of patients with DN, im- by Engagement of CD74 munohistochemistry demonstrated increased expression of We investigated the sensitivity of podocytes to the CD74 ligand CD74 protein in both the glomerular and tubular compart- MIF. MIF was previously linked to glomerular injury and the ments (Figure 1, B and C). By contrast, no significant changes complications of diabetes,10–12 but the MIF receptor(s) in the in protein expression were observed in MCD (Figure 1B). kidney had not been characterized. Phosphorylation of ERK1/2 Confocal microscopy co-localized CD74 and synaptopodin, was recently found to be a CD74-mediated MIF effect in extrare- indicating that podocytes contribute to glomerular CD74 ex- nal cells.14 In podocytes, MIF induced phosphorylation of pression in DN (Figure 1D). ERK1/2 and p38 MAPK in a time- and dosage-dependent manner (Figure 4). As previously reported for CD74-mediated responses Increased CD74 Expression in Experimental DN in extrarenal cells, ERK1/2 phosphorylation in response to MIF In a chronic rat model of DN induced by a single injection of persisted for up to 24 h (Figure 4A). We observed similar results streptozotocin (STZ) and characterized by the development for p38 MAPK (Figure 4, C and D). of albuminuria at 7 mo of follow-up (1071 Ϯ 247 versus 390 Ϯ 70 ␮g/24 h; P Ͻ 0.02; Supplemental Table 1), CD74 was increased in whole kidney by West- ern blot (Figure 2, A and B). Immuno- histochemistry localized the increased CD74 expression to both glomeruli and tubules of DN rats (Figure 2C). In glomeruli, CD74 was localized to podocytes (Figure 2C). Quantification of immunohistochemistry preparations showed a significant increase in glomerular CD74 staining (Figure 2D).

Expression of CD74 in Cultured Human Podocytes We found differentiated cultured human podocytes to express CD74 mRNA and protein (Figure 3). Either a high- glucose medium or the inflammatory cytokine TNF-␣ increased CD74 mRNA (Figure 3, A and B) and protein expression (Figure 3, C and D). We observed no changes (Ͻ7% variation at 24 or 48 h) in CD74 expression with mannitol, used as an osmolarity control. The delayed increase in protein is consistent with findings in other cell types stimulated with IFN-␥.17 Confocal micros- Figure 2. Expression of CD74 protein in experimental DN. (A) Western blot of whole kidney. copy showed that the bulk of CD74 lo- Representative image. (B) Western blot quantification expressed as percentage increase over calizes to the perinuclear region (Figure control. Data are means Ϯ SD of 10 rats per group. *P Ͻ 0.05 versus control. (C) Immuno- 3E). This is consistent with findings in histochemistry, representative images of control kidney, DN, and a detail of DN in which other cell types.18 We also found CD74 podocyte staining for CD74 is observed (arrows). (D) Quantification of glomerular CD74 in the nucleus. This is also consistent expression by immunohistochemistry. **P Ͻ 0.01 versus control. Magnification, ϫ200.

J Am Soc Nephrol ●●: –, 2009 CD74 and Podocytes 3 BASIC RESEARCH www.jasn.org

MIF Increases TRAIL and MCP-1 Expression in Podocytes and Tubular Cells MCP-1 and TRAIL are mediators of renal injury in diabetes.2,3 MIF increased TRAIL mRNA in podocytes in a time-dependent manner (Figure 6A). Inhibitors of ERK1/2 and p38 prevented the upregu- lation of TRAIL and MCP-1 expression (Figure 6, B and C), suggesting the need for the combined participation of both pathways. High glucose also upregulated CD74 expression in cultured HK2 human proximal tubular cells (Figure 7, A and B), and MIF also increased TRAIL and MCP-1 expression in these cells (Figure 7, C through E). We previously observed de novo TRAIL expression in diabetic glomeruli.3 We now report that podocytes are sites of TRAIL expression in human DN (Figure 8A) and that, in addition to tubular cells,3 TRAIL kills podocytes cultured in a diabetic milieu (Figure 8B).

DISCUSSION

MIF is a pleiotropic cytokine that has been implicated in the development of renal injury5–8,10; however, to date, the receptor mediating MIF actions on kidney cells had not been characterized. We now show (1) that the expression of CD74, a receptor for MIF, is increased in human DN as well as in experimental Figure 3. Expression of CD74 by human cultured podocytes. (A) High glucose DN; (2) that podocytes and tubular epi- increases CD74 mRNA. Real-time reverse transcription–PCR (RT-PCR). Data are thelial cells express CD74 both in culture Ϯ Ͻ means SD of three independent experiments. **P 0.01 versus control 48 h. (B) and in vivo; and (3) that CD74 mediates TNF increases CD74 mRNA. Real-time RT-PCR. Data are means Ϯ SD of three signal transduction initiated by MIF in independent experiments. **P Ͻ 0.01 versus control. (C) High glucose increases CD74 protein expression. Representative Western blot and quantification. Data are means Ϯ these cells, increasing the expression of SD of three independent experiments. **P Ͻ 0.01 versus control 48 h. (D) TNF TRAIL and MCP-1. These results iden- increases CD74 protein. Data are means Ϯ SD of three independent experiments. tify CD74 as a potential therapeutic tar- **P Ͻ 0.01 versus control 48 h. (E) Confocal microscopy localized CD74 to the get in DN. perinuclear region and the cell membrane. Nephrin staining used the same secondary We confirmed increased CD74 ex- antibody and confirmed the differentiated status of podocytes. (F) Flow cytometry of pression in human DN at the mRNA nonpermeabilized cells showed that CD74 is expressed in the cell surface. and protein levels in two different co- horts of patients. Findings in the rat model corroborated those in humans. To confirm that MIF actions on podocytes were mediated Although it is conceivable that CD74 plays a role in other forms by CD74, we knocked down CD74 by specific small interfering of renal disease in which MIF has been found to be increased, RNA (siRNA; Figure 5A). As a result, podocyte ERK1/2 and data from patients with MCD suggest that it is not upregulated p38 phosphorylation responses to MIF were lost (Figure 5, in all forms of glomerular injury. In this regard, high glucose B and C). levels could contribute to the increased podocyte and tubular

4 Journal of the American Society of Nephrology J Am Soc Nephrol ●●: –, 2009 www.jasn.org BASIC RESEARCH

jury, including DN.5,9,10 Despite this, the cellular targets of MIF in glomerular injury had not been studied in detail and the receptor mediating MIF actions had not been identified. We now show that the expression of the MIF receptor CD74 is increased in epithelial tubular and glomerular cells in DN, thereby placing cytokine and receptor in the same compartments. The contribution of MIF to renal injury could involve pro- moting inflammatory mediator expression and modulation of apoptosis and cell proliferation.4 In this regard, there is in vivo evidence of a direct injurious role for MIF on glomerular cells: Transgenic expression of MIF in podocytes led to proteinuria, azotemia, and podocyte injury in the absence of glomerular hyper- cellularity, suggesting that podocytes are direct targets for MIF.8 MIF activated ERK1/2 and p38 MAPK in podocytes. ERK1/2 and p38 MAPK have been implicated in the pro- gression of various glomerulopathies.20 They are phosphorylated in podocytes from patients with DN.21 MAPK pathways mediate different signaling events either alone or in concert with other pathways.22 p38 MAPK play a critical role in the regulation of proinflamma- tory cytokine production, cell survival and apoptosis, and the stability of the cytoskeleton.23–26 Phosphorylation of p38 MAPK is increased in injured podocytes, and p38 MAPK inhibition sup- pressed podocyte injury in experimental 20 Figure 4. MIF activates MAPK in cultured podocytes. (A through D) MIF stimulation nephrotic syndrome. Increased glo- results in ERK1/2 (A and B) and p38 (C and D) phosphorylation. Time course (A, **P Ͻ 0.01 merular ERK1/2 phosphorylation was versus control *P Ͻ 0.05 versus control; C, **P Ͻ 0.01 versus control *P Ͻ 0.05 versus also observed in these models.20 The control) and dose-response at 30 min (B, *P Ͻ 0.05 versus control; D, **P Ͻ 0.01 versus ERK cascade transmits signals from control). Representative Western blots and quantification (means Ϯ SD) of three inde- many extracellular agents to regulate pendent experiments. cellular processes through actions on more than 160 substrates.20 ERK1/2 and p38 activation were required for MIF- cell CD74 expression in patients with diabetes; however, in- induced induction of TRAIL and MCP-1 in podocytes and creased glomerular CD74 may not be specific from DN. In fact, tubular cells. TRAIL is a lethal cytokine whose expression in- preliminary studies showed that CD74 mRNA expression is creases in tubular cells and appears de novo in podocytes in increased in hypertensive nephropathy (P.I. et al., unpublished DN.3 TRAIL induces tubular cell apoptosis in a diabetic milieu observation). In this regard, podocyte CD74 was also upregulated and is also lethal for podocytes. by inflammatory cytokines, such as TNF-␣, that are present in a In summary, we have identified a novel role of CD74 in number of glomerular pathologies, including DN.19 glomerular injury. As a receptor for MIF, CD74 could contrib- MIF was previously implicated in renal injury. Increased ute to podocyte injury in susceptible patients with diabetes by glomerular and tubular MIF expression as well as urinary ex- activating MAPK cascades and increasing TRAIL and MCP-1 cretion was noted in human and experimental glomerular in- expression.

J Am Soc Nephrol ●●: –, 2009 CD74 and Podocytes 5 BASIC RESEARCH www.jasn.org

ray analysis following the manufacturer’s protocol has been described. We initially obtained image files through Affymetrix Gene- Chip software (MAS5).16 Subsequently, we performed robust multichip analysis using RMAexpress (http://statwww.berkeley.edu/ users/bolstad/RMAExpress). Starting from the normalized robust multichip analysis, the Significance Analysis of Microarrays (SAM; version 1.21, http://www-stat.stanford. edu/ϳtibs/SAM/) software was applied using a false discovery rate of 1% to identify genes that were significantly differentially regulated between the analyzed groups.3,16 We evaluated changes in expression of Ն1.5- fold for significance. On the basis of this analysis, we found CD74 to be significantly upregulated in the glomerular and interstitial compartments of patients with DN.

Cell Culture and Reagents Human podocytes are an immortalized cell line, as described previously,27 transfected with a temperature-sensitive SV40 gene con- struct and a gene encoding the catalytic do- main of human telomerase.28 At a permissive temperature of 33°C, the cells remain in an undifferentiated proliferative state, whereas raising the temperature to 37°C results in growth arrest and differentiation to the pa- rental podocyte phenotype. Undifferentiated podocyte cultures were maintained at 33°C Figure 5. Downregulation of CD74 prevents MIF actions on podocytes. siRNA down- in RPMI 1640 medium with penicillin; strep- regulation of CD74 protein (Western blot; A) prevents ERK1/2 (B) and p38 (C) phosphorylation induced by 50 ng/ml MIF for 30 min. Data are means Ϯ SD of three independent tomycin; insulin, transferrin, and selenite; experiments. (B) *P Ͻ 0.05 versus control; **P Ͻ 0.01 versus MIF alone. (C) *P Ͻ 0.05 and 10% FCS. Once cells had reached 70 to versus control; **P Ͻ 0.01 versus MIF alone. 80% confluence, they were cultured at 37°C for at least 14 d before use, when full differ- CONCISE METHODS entiation had taken place. For experiments, cells were cultured in serum-free medium 24 h before the addition of the stimuli and Transcriptomic Analysis throughout the experiment. For high-glucose experiment, glucose Human renal biopsies were collected in a multicenter study, the Eu- was added in the medium to reach a final concentration of 700 mg/dl ropean Renal cDNA Bank (ERCB).3,16 The protocol was approved by versus control medium with 200 mg/dl glucose. The same amount of the local ethical committees. We obtained informed consent from mannitol was added as an osmolarity control. TNF-␣ (30 ng/ml) was patients according to local guidelines, and we processed samples ac- from Immugenex (Los Angeles, CA), and TRAIL (100 ng/ml) was cording to the ERCB protocol. For oligonucleotide array–based gene from Alexis (La¨ufelfingen, Switzerland).3 Culture of HK2 human expression profiling, we included a total of 30 kidney biopsies from proximal tubular epithelial cells (ATCC, Rockville, MD) and the as- individual patients: Pretransplantation kidney biopsies from living sessment of apoptosis as hypodiploid cells by flow cytometry have donors used as control (n ϭ 6) and protocol biopsies from Pima been described.3,29 PD98059 and SB203580 (Stressgen Bioreagent, Indians with histologic diagnosis of DN (n ϭ 20) and from patients Ann Arbor, MI) were dissolved in DMSO. with MCD as a proteinuric nephropathy control (n ϭ 4; Supplemen- tal Table 2). Animal Model We manually microdissected tubulointerstitial and glomerular We studied two groups of ten 10-wk-old Wistar Kyoto rats (Criffa, compartments from cortical tissue segments. We isolated total RNA Barcelona, Spain). We induced diabetes by a single intraperitoneal using a commercially available isolation protocol. For probe labeling, injection of STZ (Sigma, St. Louis, MO), 50 mg/kg in 0.01 M citrate we used a modification of the Eberwein protocol. Affymetrix microar- buffer (pH 4.5).30 Control rats received the STZ vehicle. We killed rats

6 Journal of the American Society of Nephrology J Am Soc Nephrol ●●: –, 2009 www.jasn.org BASIC RESEARCH

Figure 6. MIF increases TRAIL and MCP-1 expression in podocytes. (A) Time course of TRAIL mRNA expression in podocytes treated with 10 ng/ml MIF. Real-time RT-PCR. Data are means Ϯ SD of three independent experiments. *P Ͻ 0.05 versus control. (B and C) In cells stimulated with 10 ng/ml MIF for 24 h, pretreatment with inhibitors of ERK1/2 (20 ␮M PD98059) and p38 MAPK (5 ␮M SB203580) for 1 h prevented the induction of TRAIL mRNA (B, *P Ͻ 0.05 versus control, **P Ͻ 0.01 versus MIF alone) and MCP-1 mRNA (C, *P Ͻ 0.05 versus control, **P Ͻ 0.01 versus MIF alone). Data are means Ϯ SD of three independent experiments. at 7 mo after induction of diabetes.31 All studies were performed in uria was measured by ELISA (Celltrend, Luckenwalde, Germany; accordance with the European Union normative. We administered Supplemental Table 1). insulin (1 to 4 IU subcutaneously; Insulatard NPH; Novo Nordisk, Bagsvaerd, Denmark) weekly so as to prevent death but not with the Immunofluorescence and Immunohistochemistry Staining aim of totally correcting hyperglycemia. We initiated insulin admin- We performed immunohistochemistry for CD74 in human kidney istration 7 d after STZ, having checked that the animals had glycemia biopsies on formalin-fixed paraffin sections (4 ␮m) using Histostain Ͼ400 mg/dl (Glucocard; Menarini, Barcelona, Spain). We measured SP Kit (Broad Spectrum, AEC, 95–9943; Zymed Labs, San Francisco, systolic BP monthly in conscious, restrained rats by the tail-cuff CA). Antigen was retrieved by microwave heating (10 min in citric sphygmomanometer (NARCO, Biosystems, Austin, TX). Albumin- acid, 10 mM [pH 6.0]). Endogenous peroxidase activity was

Figure 7. CD74 expression and MIF actions in cultured tubular cells. (A) High glucose increases CD74 mRNA expression at 24 h. Real-time RT-PCR. Data are means Ϯ SD of three independent experiments. #P Ͻ 0.0001 versus control. (B) High glucose increases CD74 protein expression at 24 h. Represen- tative Western blot and quantification. Data are means Ϯ SD of three independent experiments. #P Ͻ 0.0001 versus low-glucose control. (C) Exposure to 10 ng/ml MIF for 24 h increases TRAIL expression in tubular cells. Representative Western blot and quantification. Data are means Ϯ SD of three independent experiments. *P Ͻ 0.05 versus control. (D and E) Cells were pretreated with inhibitors of ERK1/2 (20 ␮M PD98059) and p38 MAPK (5 ␮M SB203580) for 1 h and stimulated with 10 ng/ml MIF for 24 h: Effect on TRAIL mRNA (D, *P Ͻ 0.05 versus control, **P Ͻ 0.01 versus MIF alone) and MCP-1 mRNA expression (E, *P Ͻ 0.05 versus control, **P Ͻ 0.01 versus MIF alone). Data are means Ϯ SD of three independent experiments.

J Am Soc Nephrol ●●: –, 2009 CD74 and Podocytes 7 BASIC RESEARCH www.jasn.org

Figure 8. TRAIL is expressed by podocytes in diabetic kidney injury and is lethal for cultured podocytes. (A) Representative TRAIL immunohistochemistry of a human glomerulus from a patient with DN. The arrow points a stained podocyte. De novo TRAIL expression in glomeruli was found in 13 of 17 biopsies from patients with DN by our group, whereas no glomerular staining for TRAIL was observed in control glomeruli.3 (B) Culture for 24 h in the presence of 100 ng/ml TRAIL increases the apoptosis rate in podocytes cultured in a high-glucose milieu. *P Ͻ 0.05 versus low glucose; **P Ͻ 0.01 versus high glucose. Data are means Ϯ SD of three independent experiments. quenched with 3%. Sections were blocked with serum, incubated with MA); blocked with 5% skim milk in PBS/0.5% vol/vol Tween 20 for anti-CD74 (rabbit polyclonal, sc-20082, 1:30; Santa Cruz Biotechnol- 1 h; washed with PBS/Tween; and incubated with goat polyclonal ogy, Santa Cruz, CA) overnight at 4°C, and stained with biotinylated anti-CD74 (1:500), mouse monoclonal anti–p-ERK (1:500), rabbit secondary antibody. For the semiquantification of staining intensity, polyclonal anti-ERK1/2 (1:2000), mouse monoclonal anti-p-p38 (1: we visually assessed 10 random views (ϫ200 magnification) in each 500), or goat polyclonal anti-p38 (1:2500; all from Santa Cruz Bio- slide as 0 to 3 (0, no staining; 1, moderate staining; 2, strong staining; technology), and mouse monoclonal anti-TRAIL (1:1000, Pharmin- 3, very strong staining) in a blinded manner. We calculated the mean gen). Antibodies were diluted in 5% milk PBS/Tween. Blots were value for all of the views in one slide. CD74 expression was expressed washed with PBS/Tween and subsequently incubated with appropri- as percentage change over control. The nonaffected parts of tumor ate horseradish peroxidase–conjugated secondary antibody (1:2000; nephrectomies served as a control group (n ϭ 5), and we studied Amersham, Aylesbury, UK). After washing, the blots were developed patients with DN (n ϭ 5) and MCD (n ϭ 4; Supplemental Table 2). with the chemiluminescence method (ECL; Amersham). Blots were TRAIL immunohistochemistry has been described.3 then probed with mouse monoclonal anti–␣-tubulin antibody For immunofluorescence, the unfixed renal tissue was embedded (1:2000; Sigma) and levels of expression were corrected for minor in Tissue-Tek (Societa`Italiana Chimici, Rome, Italy), snap-frozen in differences in loading. a mixture of isopentane and dry-ice, and stored at Ϫ80°C. Subse- quently, 5-␮m sections were fixed in cold acetone and stained with Real-Time Reverse Transcription–PCR rabbit anti-CD74 (Santa Cruz Biotechnology) followed by AlexaFluor RNA was isolated by Trizol (Invitrogen, Paisley, UK).31 One micro- 488 goat anti-rabbit IgG (Molecular Probes, Invitrogen, Milan, Italy). gram of RNA was reverse-transcribed with High Capacity cDNA Ar- After washing, the procedure was repeated with the second primary chive Kit (Applied Biosystems, Foster City, CA). Real-time PCR reac- antibody (mouse anti-synaptopodin; Progen, Frankfurt, Germany), tions were performed on an ABI Prism 7500 sequence detection PCR followed by AlexaFluor 546 goat anti-mouse (Molecular Probes). Sec- system (Applied Biosystems) according to the manufacturer’s proto- tions were mounted with antifading mounting medium (Vectashield; col using the ⌬⌬Ct method.31 Expression levels are given as ratios to Vector Laboratories, Burlingame, CA). Specificity of labeling was dem- glyceraldehyde-3-phosphate dehydrogenase. Predeveloped primer onstrated by the lack of staining after substituting proper control IgG and probe assays were obtained for human glyceraldehyde-3-phosphate (rabbit primary antibody isotype control and mouse primary antibody dehydrogenase, MCP-1, TRAIL, and CD74 from Applied Biosystems. isotype control, both from Zymed, Invitrogen) for the primary antibody. In rats immunohistochemistry was carried out in paraffin-embed- Flow Cytometry Analysis of Cell Surface CD74 ded tissue sections 5 ␮m thick.31 The primary antibody was goat poly- Expression clonal anti-CD74 (1:60, Santa Cruz Biotechnology). Sections were Cells were detached with 2 mM EDTA and 5 ϫ 105 cells were incu- counterstained with Carazzi’s hematoxylin. Negative controls in- bated for 30 min at 4°C with 8 ␮g/ml rabbit anti-CD74 antibody cluded incubation with a nonspecific Ig of the same isotype as the (Santa Cruz Biotechnology) or control IgG followed by a 30-min 4°C primary antibody. The percentage of positive glomerular surface area incubation with 1:100 FITC secondary antibody (Pharmingen, San was quantified in 20 glomeruli per rat. Diego, CA).32 Mean cell fluorescence was calculated using Cell Quest Software (Becton Dickinson, Franklin Lakes, NJ). Western Blot Tissue and cell samples were homogenized in lysis buffer31; then sep- Confocal Microscopy arated by 10 or 12% SDS-PAGE under reducing conditions and trans- Cells plated onto Labtek slides were fixed in 4% paraformaldehyde ferred to polyvinylidene difluoride membranes (Millipore, Bedford, and permeabilized in 0.2% Triton X-100 in PBS for 10 min each. After

8 Journal of the American Society of Nephrology J Am Soc Nephrol ●●: –, 2009 www.jasn.org BASIC RESEARCH washing in PBS, cells were incubated overnight at 4°C with rabbit A: The death ligand TRAIL in diabetic nephropathy. J Am Soc Nephrol polyclonal anti-CD74 antibody (1:50; Santa Cruz Biotechnology) or 19: 904–914, 2008 rabbit polyclonal anti-nephrin (1:100),33 followed by incubation with 4. Morand EF, Leech M, Bernhagen J: MIF: A new cytokine link between rheumatoid arthritis and atherosclerosis. Nat Rev Drug Discov 5: 399– FITC secondary antibody (1:200; Sigma). Cell nuclei were counter- 410, 2006 stained with propidium iodide. After washing, cells were mounted in 5. Lan HY, Yang N, Nikolic-Paterson DJ, Yu XQ, Mu W, Isbel NM, Metz CN, 70% glycerol in PBS and analyzed with a DM-IRB confocal micro- Bucala R, Atkins RC: Expression of macrophage migration inhibitory scope (Leica DM, Bannockburn, IL).31 factor in human glomerulonephritis. Kidney Int 57: 499–509, 2000 6. Rice EK, Nikolic-Paterson DJ, Hill PA, Metz CN, Bucala R, Atkins RC, Tesch GH: Interferon-gamma induces macrophage migration inhibi- Transfection of siRNA tory factor synthesis and secretion by tubular epithelial cells. Nephrol- Cells were grown in six-well plates (Costar, Cambridge, MA) and ogy (Carlton) 8: 156–161, 2003 transfected with a mixture of 25 nmol/ml siRNA (Santa Cruz), 7. Rice EK, Tesch GH, Cao Z, Cooper ME, Metz CN, Bucala R, Atkins RC, Opti-MEM I Reduced Serum Medium and Lipofectamine 2000 (In- Nikolic-Paterson DJ: Induction of MIF synthesis and secretion by vitrogen). After 18 h, cells were washed and cultured for 48 h in com- tubular epithelial cells: A novel action of angiotensin II. Kidney Int 63: 1265–1275, 2003 plete medium and serum-depleted for 24 h before addition of 8. Sasaki S, Nishihira J, Ishibashi T, Yamasaki Y, Obikane K, Echigoya M, MIF. This time point was selected from a time course of decreasing Sado Y, Ninomiya Y, Kobayashi K: Transgene of MIF induces podocyte CD74 protein expression in response to siRNA. A negative control injury and progressive mesangial sclerosis in the mouse kidney. Kid- scrambled siRNA provided by the manufacturer did not reduce CD74 ney Int 65: 469–481, 2004 protein. 9. Brown FG, Nikolic-Paterson DJ, Hill PA, Isbel NM, Dowling J, Metz CM, Atkins RC: Urine macrophage migration inhibitory factor reflects the severity of renal injury in human glomerulonephritis. JAmSoc Statistical Analysis Nephrol 13[Suppl 1]: S7–S13, 2002 Data are given as means Ϯ SD. Mann-Whitney, two-sided t test or 10. Chow F, Ozols E, Nikolic-Paterson DJ, Atkins RC, Tesch GH: Macro- one-way ANOVA was applied to indicate significantly different mean phages in mouse type 2 diabetic nephropathy: Correlation with dia- values in comparison with the control group. P Ͻ 0.05 was considered betic state and progressive renal injury. Kidney Int 65: 116–128, 2004 statistically significant. 11. Herder C, Kolb H, Koenig W, Haastert B, Muller-Scholze S, Rathmann W, Holle R, Thorand B, Wichmann HE: Association of systemic concentrations of macrophage migration inhibitory factor with impaired glucose tolerance and type 2 diabetes: Results from the Cooperative ACKNOWLEDGMENTS Health Research in the Region of Augsburg, Survey 4 (KORA S4). Diabetes Care 29: 368–371, 2006 12. Tashimo A, Mitamura Y, Nagai S, Nakamura Y, Ohtsuka K, Mizue Y, This study was supported by grant FIS 06/0046, SAF03/884, LSHB- Nishihira J: Aqueous levels of macrophage migration inhibitory factor CT-2004-6 (ADDNET), LSHB-CT-2007-036644 (DIALOK), So- and monocyte chemotactic protein-1 in patients with diabetic retinop- ciedad Espanõla de Nefrología, ISCIII-RETIC REDinREN/RD06/ athy. Diabet Med 21: 1292–1297, 2004 0016, and Comunidad de Madrid/FRACM/S-BIO0283/2006. Salary 13. David JR: Delayed hypersensitivity in vitro: its mediation by cell-free substances formed by lymphoid cell-antigen interaction. Proc Natl support was provided by FIS to A.B.S., MEC to M.D.S.N., Programa Acad Sci U S A 56: 72–77, 1966 Intensificacioń Actividad Investigadora (ISCIII/Agencia Laín- 14. Leng L, Metz CN, Fang Y, Xu J, Donnelly S, Baugh J, Delohery T, Chen Entralgo/CM) to A.O., and the Finnish Kidney Foundation and the Y, Mitchell RA, Bucala R: MIF signal transduction initiated by binding Kyllikki and Uolevi Lehikoinen Foundation for P.I. Fondecyt to CD74. J Exp Med 197: 1467–1476, 2003 1080083 to S.M. This research was supported, in part, by the Intra- 15. Becker-Herman S, Arie G, Medvedovsky H, Kerem A, Shachar I: CD74 mural Research Program of the National Institute of Diabetes and is a member of the regulated intramembrane proteolysis-processed protein family. Mol Biol Cell 16: 5061–5069, 2005 Digestive and Kidney Diseases. 16. Schmid H, Boucherot A, Yasuda Y, Henger A, Brunner B, Eichinger F, Anneli von Behr is acknowledged for technical assistance and Mar Nitsche A, Kiss E, Bleich M, Grone HJ, Nelson PJ, Schlondorff D, Gonzalez García-Parrenõ and Alberto Puime for technical help. Cohen CD, Kretzler M: Modular activation of nuclear factor-kappaB transcriptional programs in human diabetic nephropathy. Diabetes 55: 2993–3003, 2006 17. Houghton AN, Thomson TM, Gross D, Oettgen HF, Old LJ: Surface DISCLOSURES antigens of melanoma and melanocytes: Specificity of induction of Ia None. antigens by human gamma-interferon. J Exp Med 160: 255–269, 1984 18. Leng L, Bucala R: Insight into the biology of macrophage migration inhibitory factor (MIF) revealed by the cloning of its cell surface receptor. Cell Res 16: 162–168, 2006 REFERENCES 19. Navarro JF, Mora-Fernandez C: The role of TNF-alpha in diabetic nephropathy: Pathogenic and therapeutic implications. Cytokine 1. Mora C, Navarro JF: Inflammation and diabetic nephropathy. Curr Growth Factor Rev 17: 441–450, 2006 Diab Rep 6: 463–468, 2006 20. Koshikawa M, Mukoyama M, Mori K, Suganami T, Sawai K, Yoshioka T, 2. Tesch GH: MCP-1/CCL2: A new diagnostic marker and therapeutic Nagae T, Yokoi H, Kawachi H, Shimizu F, Sugawara A, Nakao K: Role target for progressive renal injury in diabetic nephropathy. Am J of p38 mitogen-activated protein kinase activation in podocyte injury Physiol Renal Physiol 294: F697–F701, 2008 and proteinuria in experimental nephrotic syndrome. J Am Soc Neph- 3. Lorz C, Benito-Martin A, Boucherot A, Ucero AC, Rastaldi MP, Henger rol 16: 2690–2701, 2005 A, Armelloni S, Santamaria B, Berthier CC, Kretzler M, Egido J, Ortiz 21. Sakai N, Wada T, Furuichi K, Iwata Y, Yoshimoto K, Kitagawa K,

J Am Soc Nephrol ●●: –, 2009 CD74 and Podocytes 9 BASIC RESEARCH www.jasn.org

Kokubo S, Kobayashi M, Hara A, Yamahana J, Okumura T, Takasawa Mathieson PW, Saleem MA: Nephrin is critical for the action of insulin K, Takeda S, Yoshimura M, Kida H, Yokoyama H: Involvement of on human glomerular podocytes. Diabetes 56: 1127–1135, 2007 extracellular signal-regulated kinase and p38 in human diabetic ne- 29. Justo P, Sanz A, Lorz C, Gomez-Garre D, Mezzano S, Egido J, Ortiz A: phropathy. Am J Kidney Dis 45: 54–65, 2005 Expression of Smac/Diablo in tubular epithelial cells and during acute 22. Roux PP, Blenis J: ERK and p38 MAPK-activated protein kinases: A renal failure. Kidney Int Suppl S52–S56, 2003 family of protein kinases with diverse biological functions. Microbiol 30. Ortiz A, Ziyadeh FN, Neilson EG: Expression of apoptosis-regula- Mol Biol Rev 68: 320–344, 2004 tory genes in renal proximal tubular epithelial cells exposed to high 23. Adams JL, Badger AM, Kumar S, Lee JC: p38 MAP kinase: Molecular ambient glucose and in diabetic kidneys. J Investig Med 45: 50–56, target for the inhibition of pro-inflammatory cytokines. Prog Med 1997 Chem 38: 1–60, 2001 31. Sanz AB, Justo P, Sanchez-Nino MD, Blanco-Colio LM, Winkles JA, 24. Kimura C, Zhao QL, Kondo T, Amatsu M, Fujiwara Y: Mechanism of Kreztler M, Jakubowski A, Blanco J, Egido J, Ruiz-Ortega M, Ortiz A: UV-induced apoptosis in human leukemia cells: roles of Ca2ϩ/ The cytokine TWEAK modulates renal tubulointerstitial inflammation. Mg(2ϩ)-dependent endonuclease, caspase-3, and stress-activated J Am Soc Nephrol 19: 695–703, 2008 protein kinases. Exp Cell Res 239: 411–422, 1998 32. Lorz C, Ortiz A, Justo P, Gonzalez-Cuadrado S, Duque N, Gomez- 25. Guay J, Lambert H, Gingras-Breton G, Lavoie JN, Huot J, Landry J: Guerrero C, Egido J: Proapoptotic Fas ligand is expressed by normal Regulation of actin filament dynamics by p38 map kinase-mediated kidney tubular epithelium and injured glomeruli. J Am Soc Nephrol phosphorylation of heat shock protein 27. J Cell Sci 110: 357–368, 1997 11: 1266–1277, 2000 26. Schiffer M, Bitzer M, Roberts IS, Kopp JB, ten Dijke P, Mundel P, 33. Ihalmo P, Rinta-Valkama J, Mai P, Astrom E, Palmen T, Pham TT, Floss Bottinger EP: Apoptosis in podocytes induced by TGF-beta and T, Holthofer H: Molecular cloning and characterization of an endog- Smad7. J Clin Invest 108: 807–816, 2001 enous antisense transcript of Nphs1. Genomics 83: 1134–1140, 2004 27. Saleem MA, O’Hare MJ, Reiser J, Coward RJ, Inward CD, Farren T, Xing CY, Ni L, Mathieson PW, Mundel P: A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression. J Am Soc Nephrol 13: 630–638, 2002 Supplemental information for this article is available online at http://www. 28. Coward RJ, Welsh GI, Koziell A, Hussain S, Lennon R, Ni L, Tavare JM, jasn.org/.

10 Journal of the American Society of Nephrology J Am Soc Nephrol ●●: –, 2009