BASIC RESEARCH www.jasn.org

Blocking on Leukocytes Attenuates Kidney Ischemia-Reperfusion Injury

Gang Jee Ko,*† Hye Ryoun Jang,* Yanfei Huang,* Karl L. Womer,* Manchang Liu,* Elizabeth Higbee,* Zuoxiang Xiao,‡ Hideo Yagita,§ Lorraine Racusen,‡ Abdel Rahim A. Hamad,‡ and Hamid Rabb*

*Division of Nephrology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland; †Department of Medicine, Korea University College of Medicine, Seoul, Korea; ‡Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland; and §Department of Immunology, Juntendo University School of Medicine, Tokyo, Japan

ABSTRACT Inflammation contributes to the pathogenesis of ischemic acute kidney injury (AKI), and T cells mediate the early phase of ischemia-reperfusion injury (IRI). The Fas/Fas ligand (FasL) pathway modulates the balance of subsets in the peripheral circulation as well as multiple inflammatory responses, suggesting that FasL may mediate ischemic AKI. Here, we induced bilateral renal IRI in mice bearing a loss-of-function mutation of FasL (the gld mutation) and in wild-type mice. Compared with wild-type mice, serum creatinine was lower in gld mice (1.4 Ϯ 0.9 mg/dl versus 2.6 Ϯ 0.4) at 24 hours after IRI (P Ͻ 0.05). In addition, gld mice had fewer TNF-␣-producing T lymphocytes in the kidneys and renal lymph nodes. Furthermore, pharmacologic blockade of FasL protected the kidneys of wild-type mice from IRI. Analysis of bone marrow chimeric mice suggested that the pathogenic effect of FasL involves leukocytes; reconstitution of wild-type mice with gld splenocytes attenuated IRI. In contrast, reconstitution of gld mice with wild-type splenocytes enhanced IRI. These data demonstrate that FasL, particularly on leukocytes, mediates ischemic AKI.

J Am Soc Nephrol 22: 732–742, 2011. doi: 10.1681/ASN.2010010121

Renal ischemia/reperfusion injury (IRI) is a leading inflammatory cells,9,10 and induction of cause of acute kidney injury (AKI) in both allografts production.11 Previous studies have demonstrated and native kidneys.1 Despite advances in medical diverse results regarding FasL effects on various in- care, the mortality and morbidity associated with flammatory processes,12–15 which may be related to AKI after IRI still remains high,2 and specific ther- the pleiotropic effects of FasL and expressing cell apy is not available. Therefore, many studies have types.16,17 Besides lymphocytes, FasL is frequently focused on elucidating the pathophysiology of renal IRI with the goal of identifying new therapeutic tar- gets. Received January 27, 2010. Accepted December 14, 2010. Inflammation has been demonstrated to con- Published online ahead of print. Publication date available at tribute to the pathogenesis of renal IRI, with a role www.jasn.org. for leukocytes, adhesion molecules, and cyto- Present address: Dr. Hye Ryoun Jang, Nephrology Division, De- kines.3,4 Fas ligand (FasL) is a member of the TNF partment of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea. family that induces apoptosis by cross-linking its 5 Correspondence: Dr. Hamid Rabb, Johns Hopkins University . Fas is known to be involved in various School of Medicine, Ross Building, Room 965, 720 Rutland Ave- forms of renal injury such as IRI6 and tubular injury nue, Baltimore, MD 21205. Phone: 410-502-1555; Fax: 410-614- in glomerulonephritis.7,8 FasL is also believed to 5129; E-mail: [email protected]. have a role in inflammation through recruitment of Copyright © 2011 by the American Society of Nephrology

732 ISSN : 1046-6673/2204-732 J Am Soc Nephrol 22: 732–742, 2011 www.jasn.org BASIC RESEARCH

expressed on epithelial and endothelial cells18 and is critical for 1B). When renal tubular injury scores were compared, gld mice regulating T cell homeostasis.19 A double negative (DN) T cell had reduced tubular damage. There were significantly more (CD3ϩCD4ϪCD8Ϫ) population that is normally rare in the damaged tubules in the outer medulla of WT than gld mice lymphoid tissues gradually accumulates in lymph nodes and (58.6 Ϯ 3.3 versus 38.2 Ϯ 9.4% of damaged tubules, P Ͻ 0.05; spleens of mice bearing spontaneous loss-of-function point Figure 2). Immunohistochemical staining with caspase 3 on mutation of FasL (referred to as the gld mutation). In spite of day 3 after IRI showed decreased apoptosis in the tubules of gld accumulation of DN T cells and activated CD4ϩ and CD8ϩ T than in WT mice (0.0037 Ϯ 0.0008 versus 0.0018 Ϯ 0.0007% of cells in mutant mice, FasL deficiency confers protection rather the total area, P Ͻ 0.05; Supplemental Text 1). than exacerbation of organ-specific T cell–mediated autoim- mune diseases.20,21 T lymphocytes are important mediators of FasL Deficiency Reduced TNF-␣ Production in T Cells IRI of kidney and other organs, and FasL could have a role in after IRI in Kidney and Renal Lymph Node IRI through modulation of T cells homeostasis. Although var- When baseline characteristics of different leukocytes were ious effects of FasL on IRI have been reported,22–25 little is compared between WT and gld mice, there was significantly known about the role of FasL in renal IRI, particularly from the higher frequency of CD3ϩCD4ϪCD8Ϫ double negative (DN) angle of leukocytes. T cells of spleens and blood in gld than in WT mice (Supple- In this report, we demonstrated using FasL-deficient (gld) mental Text 2). However, no significant difference of their fre- mice and FasL-blocking mAb that inhibition of FasL signifi- quency in the kidneys was observed between the two geno- cantly attenuated kidney IRI in mice. FasL deficiency induced types, although the kidney normally contains a large decreased inflammatory cytokine production by T lympho- population of DN T cells. In postischemic kidneys at day 3 after cytes isolated from postischemic kidneys both in gld mice and IRI, the NK cell population was reduced in gld as compared wild-type (WT) mice treated with FasL-blocking antibody. At- with that in WT mice (WT versus gld: 13.1 Ϯ 1.6 versus 9.0 Ϯ tenuation of IRI as assessed by functional and histologic pa- 0.8% population; P Ͻ 0.05), whereas percentages of other leu- rameters in WT chimeric mice reconstituted with gld spleno- kocyte subsets did not change (Table 1). Because FasL is known cytes as compared with WT mice reconstituted with WT to modulate T cells homeostasis, we measured production of splenocytes demonstrated that the pathogenic effect was me- proinflammatory by T cells from ischemic kidneys of diated predominantly by FasL of leukocyte origin. WT and gld mice using intracellular cytokine staining. TNF-␣ production in T cells from kidneys was slightly lower in gld than in WT mice 1 day after IRI (WT versus gld; 11.5 Ϯ 1.8 RESULTS versus 6.9 Ϯ 1.1% population; P ϭ 0.06), but became signifi- cantly different on day 3 after IRI (WT versus gld; 15.5 Ϯ 4.0 FasL Deficiency Attenuated Renal Injury after IRI versus 3.4 Ϯ 0.7% population; P Ͻ 0.05). T cells isolated from Renal dysfunction was evaluated by measuring serum creati- renal lymph nodes in gld also showed lower production of nine on 3 consecutive days after IRI (Figure 1A). The gld mice TNF-␣ compared with those in WT (WT versus gld; 53.7 Ϯ 1.4 demonstrated lower serum creatinine compared with wild- versus 42.8 Ϯ 2.6 for TCR␤ϩ cells and 57.3 Ϯ 4.1 versus 45.9 Ϯ type (WT) control group after IRI. (WT versus gld: day 1, 2.6 Ϯ 1.3 for TCR␤ϩCD4ϩ cells % population at day 3 after IRI; P Ͻ 0.4 mg/dl versus 1.4 Ϯ 0.3; day 2, 3.3 Ϯ 0.2 versus 1.4 Ϯ 0.3; day 0.05). IFN-␥ production by T cells from gld mice was also 3, 2.7 Ϯ 0.3 versus 1.3 Ϯ 0.3; P Ͻ 0.05). FasL deficiency also lower on day 1 after IRI (WT versus gld; 23.0 Ϯ 0.7 versus 8.6 Ϯ improved survival after IRI. On day 3 after IRI, the mortality 1.6 for TCR␤ϩ cells and 22.4 Ϯ 1.9 versus 9.6 Ϯ 1.8 for was higher in WT than in gld (40 versus 0%, P ϭ 0.02; Figure TCR␤ϩCD4ϩ cells % population at day 1 after IRI; P Ͻ 0.05), but it was not significantly different be- tween groups on day 3 after IRI (Figure 3, Supplemental Text 3).

FasL-Blocking Antibody Attenuated Renal Injury after IRI Increase in serum creatinine after IRI was attenuated in mice treated with FasL- blocking monoclonal antibody (Ab) com- pared with isotype control hamster IgG- treated group (control versus FasL- blocking Ab: day 1, 2.4 Ϯ 0.3 mg/dl versus Ϯ Ϯ Ϯ Figure 1. Improved renal function and survival after renal IRI in FasL deficient (gld) mice 1.7 0.2; day 2, 3.3 0.2 versus 2.1 0.3; Ϯ Ϯ Ͻ compared to wild type (WT). Serum creatinine in gld mice was lower compared with WT day 3, 2.5 0.4 versus 1.4 0.3; P 0.05; after bilateral renal pedicle clamping (n ϭ 11 for each group, panel A), and survival in gld Figure 4A). Histologic assessment revealed mice was significantly higher at 3 days after IRI (B). *P Ͻ 0.05 compared with WT. reduced tubular damage in the medulla of

J Am Soc Nephrol 22: 732–742, 2011 Fas Ligand of Leukocyte in Kidney IRI 733 BASIC RESEARCH www.jasn.org

mice treated with FasL-blocking Ab as compared with isotype control group (con- trol versus FasL-blocking Ab; 52.8 Ϯ 5.6 versus 39.6 Ϯ 4.1% of damaged tubules, P Ͻ 0.05; Figure 4, B and C).

FasL-Blocking Antibody Reduced Inflammatory Cytokine Production in T Lymphocytes after IRI We also analyzed the effects of FasL-blocking Ab treatment on cytokine production in kid- ney-infiltrating leukocytes. TNF-␣ and IFN-␥ production from total renal T lymphocytes at day 3 after IRI was markedly reduced in mice treated with FasL-blocking Ab compared with mice in the control group (control versus FasL- blocking Ab: 20.7 Ϯ 2.5 versus 5.9 Ϯ 0.9 for TNF-␣ and 17.2 Ϯ 2.8 versus 8.6 Ϯ 1.1 for IFN-␥ % population; P Ͻ 0.05). Cy- tokine production from TCR␤ϩCD4ϩ T lymphocytes at 3 days after IRI was also markedly reduced in kidneys of FasL- blocking Ab–treated mice than in the control group (Figure 5).

FasL Deficiency in Splenocytes Conferred Renal Protection after IRI To determine the source of FasL effect on renal IRI, we generated three groups of par- tial bone marrow chimeric mice developed Figure 2. Reduced tubular damage of post ischemic kidneys in FasL deficient (gld) by adoptive transfer of donor splenocytes mice compared to wild type (WT). On day 3 after IRI (D3), postischemic kidneys of gld into sublethally irradiated hosts. Partial mice showed decreased tubular damage compared with WT in medulla. Arrows chimerism was verified with the population indicate necrotic tubules and arrowheads indicate necrotic debris. Magnification: of CD3ϩCD4ϪCD8Ϫ double negative (DN) ϫ200. *P Ͻ 0.05 compared with WT. T cells in spleens, with the fact that the pro-

Table 1. Comparison of populations of T cells, B cells, NKT cells, NK cells, and macrophages in the postischemic kidneys between groups (day 3 after IRI) Wild Type gld Cell counts/ml 3.5 Ϯ 0.9 ϫ 105 4.1 Ϯ 1.7 ϫ 105 % population ϩ ϩ ϩ CD4 T cells (CD4 TCR␤ ) among TCR␤ cells 47.2 Ϯ 2.8 48.4 Ϯ 1.9 ϩ ϩ ϩ ϩ ϩ activated CD4 T cells(CD69 CD4 TCR␤ ) among CD4 TCR␤ cells 19.3 Ϯ 0.8 21.2 Ϯ 3.7 ϩ ϩ ϩ CD8 T cells (CD8 TCR␤ ) among TCR␤ cells 27.6 Ϯ 1.5 33.7 Ϯ 2.3 ϩ ϩ ϩ ϩ ϩ activated CD8 T cells (CD69 CD8 TCR␤ ) among CD8 TCR␤ cells 18.3 Ϯ 3.5 22.6 Ϯ 5.5 Ϫ Ϫ ϩ ϩ double negative T cells (CD4 CD8 TCR␤ ) among TCR␤ cells 22.8 Ϯ 2.4 14.0 Ϯ 5.6 ϩ TCR␥␦ cells (TCR␥␦ ) among gated KMNC 1.73 Ϯ 0.3 1.6 Ϯ 0.3 ϩ B cells (CD19 ) among gated KMNC 36.7 Ϯ 10.9 22.3 Ϯ 7.2 ϩ ϩ ϩ differentiated B cells (MHCII CD19 ) among CD19 cells 3.8 Ϯ 1.5 3.3 Ϯ 0.6 ϩ ϩ ϩ ϩ ϩ ϩ IgM CD5 B cells (IgM CD5 CD19 ) among CD19 cells 2.9 Ϯ 0.2 1.9 Ϯ 0.3 ϩ Ϫ NK cells (NK1.1 TCR␤ ) among gated KMNC 13.1 Ϯ 1.6 9.0 Ϯ 0.8a ϩ ϩ NKT cells (NK1.1 TCR␤ ) among gated KMNC 4.6 Ϯ 1.0 3.4 Ϯ 0.5 ϩ ϩ ϩ ϩ ϩ ϩ CD1d NKT cells (CD1d NK1.1 TCR␤ ) among NK1.1 TCR␤ cells 23.8 Ϯ 3.7 24.8 Ϯ 1.5 macrophage (F4/80highCD11clow) among gated KMNC 2.9 Ϯ 0.5 1.7 Ϯ 0.5 Except for NK cells, which were lower in gld mice, there were no significant differences between groups. KMNC, kidney mononuclear cells. aP Ͻ 0.05 compared to WT.

734 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 732–742, 2011 www.jasn.org BASIC RESEARCH

mice showed reduced tubular damage in the medulla as compared with WT into WT control (WT into WT versus WT into gld: 37.0 Ϯ 5.3 versus 16.7 Ϯ 4.0% of damaged tubules; P Ͻ 0.05; Figure 7, B and C).

FasL Deficiency Effect on Myeloperoxidase Activity in Postischemic Kidneys Myeloperoxidase (MPO) activity was measured in kidney extracts after IRI to assess the effect of FasL deficiency on neutrophil (also macrophages) infiltra- tion. MPO activity, which was increased after IRI in WT, however, was not signif- icantly different between gld and WT mice at day 3 after IRI (IRI WT versus IRI gld: 10.2 Ϯ 0.4 versus 7.9 Ϯ 2.4 pg/mg of kidney extracts; P ϭ 0.07; Figure 8A). Among chimeric mice, WT mice recon- stituted with WT splenocytes showed sig- nificantly higher MPO activity compared with other groups (WT into WT versus WT into gld versus gld into WT: 7.1 Ϯ 2.1 versus 4.4 Ϯ 2.0 versus 2.5 Ϯ 1.3 pg/mg of kidney extracts; P Ͻ 0.05; Figure 8B). MPO level was significantly lower in mice Figure 3. TNF-␣ and IFN-␥ production from T cells in kidneys and renal lymph nodes treated with FasL-blocking Ab as com- after IRI was decreased in FasL deficient (gld) mice compared to wild type (WT). TNF-␣ pared with mice in the isotype control ϩ ϩ production in T cells, particularly in TCR␤ CD4 T cells from postischemic kidneys and group (isotype versus FasL-blocking Ab: renal lymph nodes (LN), at 3 days after IRI was lower in gld mice. *P Ͻ 0.05 compared 4.9 Ϯ 0.7 versus 2.1 Ϯ 0.6 pg/mg of kid- with WT. ney extracts; P Ͻ 0.05; Figure 8C).

portion of DN T cells in spleens was significantly different be- FasL Deficiency Reduced Inflammatory Cytokine tween wild-type (WT) and gld mice. As shown in Figure 6, in gld Expression in Postischemic Kidneys mice reconstituted with WT splenocytes (WT into gld), the DN Renal cytokines and chemokines were measured with T cell population in spleen was decreased to the same level as in samples extracted from postischemic kidneys of WT and gld mice WT mice reconstituted with WT splenocytes (WT into WT), at day 3 after IRI. IFN-␥ production in kidneys was lower in gld whereas the DN T cell population significantly increased in than in WT mice (WT versus gld: 17.5 Ϯ 2.5 versus 11.9 Ϯ 0.9 WT mice reconstituted with gld splenocytes (gld into WT). pg/mg of kidney extracts; P Ͻ 0.05; Figure 9). Among chimeric Proportions of DN T cell population in spleens of WT into mice, MCP-1 production in kidneys of gld into WT group was WT, WT into gld, and gld into WT were 3.3 Ϯ 0.8 versus 1.7 Ϯ lower than that in the other two groups (WT into WT versus WT 0.2 versus 26.2 Ϯ 6.5% population, respectively (P Ͻ 0.05). into gld versus gld into WT: 4927.2 Ϯ 2917.9 versus 3554.4 Ϯ Percentages of DN T cells in kidneys of different groups were 1578.2 versus 736.2 Ϯ 171.7 pg/mg; P Ͻ 0.05; Figure 9). similar. After IRI, WT into gld mice showed no differences in serum creatinine compared with WT into WT. By contrast, in gld into WT chimera, serum creatinine was lower at both day 2 DISCUSSION and day 3 as compared with WT into WT after IRI (WT into WT versus gld into WT: day 2, 1.5 Ϯ 0.2 mg/dl versus 0.7 Ϯ 0.1; In this study, we demonstrated that FasL deficiency conferred day 3, 1.5 Ϯ 0.4 versus 0.6 Ϯ 0.2; P Ͻ 0.05; Figure 7A). Upon protection against AKI after IRI, and reduced production of histologic assessment, gld into WT chimera showed less tubu- inflammatory cytokines in T cells. Furthermore, our results lar damage in both the cortex and medulla than WT into WT implicated FasL on leukocytes rather than on nonhemopoietic chimera (WT into WT versus gld into WT: cortex, 7.6 Ϯ 2.7 tissues in kidney IRI. versus1.0 Ϯ 0.4; medulla, 37.0 Ϯ 5.3 versus 17.1 Ϯ 4.3% of Fas/FasL interaction is known to induce cell death through damaged tubules; P Ͻ 0.05; Figure 7, B and C). WT into gld activation of the caspase signaling cascade, leading to cleavage

J Am Soc Nephrol 22: 732–742, 2011 Fas Ligand of Leukocyte in Kidney IRI 735 BASIC RESEARCH www.jasn.org

of chromatin and loss of structural integ- rity.26 FasL, whose expression is normally highly restricted, is frequently expressed on different cell types in the kidney including mesangial cells, tubular epithelial cells, fi- broblasts, endothelial cells, and leuko- cytes.18 Increased expression of Fas and FasL has been documented in various renal diseases and this pathway is believed to have a role in promoting renal injury through in- duction of apoptosis in glomerular and tubu- lar cells.27 However, manipulation of FasL ex- pression has showed inconsistent results in renal injury,18 suggesting diverse roles for this molecule. FasL is a major regulator of clonal expansion and peripheral homeo- stasis of T cells.19,26 Thus, it was expected that FasL deficiency would lead to T cell– mediated autoimmune diseases and in- crease of activated T cells. However, despite persistence of autoreactive T cells, FasL de- ficiency prevents autoimmune diabetes.21 Moreover, FasL cross-linking leads to re- cruitment of neutrophils and increased in- flammation,12 and FasL deficiency attenu- ates inflammatory cell infiltration and cytokine secretion.11,22 However, overex- pression of FasL in dendritic cells decreased allergen-specific T cells and airway inflam- mation.15,25 Studies on the effects of FasL in IRI have also produced conflicting re- sults.22–25 FasL expression was increased in infiltrating neutrophils and T lymphocytes after renal IRI;28 however, little is known about their role in AKI after IRI. We hypothesized that FasL would exert an important role in renal IRI through mediat- ing inflammation. In our current study, FasL deficiency attenuated renal injury after IRI and improved survival. Renal protective ef- fect from IRI in FasL-deficient mice (gld mice) was accompanied by reduced pro- duction of inflammatory cytokines by T cells isolated at different time points from kidneys of gld mice after IRI, which likely contributed to the attenuation of IRI. Pre- vious studies have shown that TNF-␣ pro- duction synergistically enhance Fas-FasL Figure 4. Improved kidney function and tissue damage after IRI with FasL blocking Ab expression and cellular apoptosis.29 FasL is compared to isotype control Ab. Serum creatinine was lower in FasL-blocking Ab– also important in mediating the cytotoxic treated group after IRI compared with mice treated with isotype control Ab (n ϭ 7 for 30,31 each group, panel A). On day 3 after IRI, post ischemic kidneys of FasL-blocking function of T cells, and FasL deficiency Ab–treated group showed decreased tubular damage compared with mice treated confers beneficial effects on T cell–medi- 32 with isotype control Ab in medulla (panels B and C). Arrows indicate necrotic tubules ated corneal rejection. We found that and arrow-heads indicate necrotic debris. Magnification: ϫ200. #P Ͻ 0.05 compared level of IFN-␥ in postischemic kidneys of with isotype control Ab group. gld mice was lower than that in WT mice,

736 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 732–742, 2011 www.jasn.org BASIC RESEARCH

generated three different types of partial chi- meric mice. Chimerism was verified by studying splenic T cells. The proportion of DN T cells in blood and spleens of adult gld mice is usually significantly higher than that in WT mice. The frequency of DN T cell pop- ulation in splenocytes of recipients was de- pendent on the genotype of donors. WT mice reconstituted with gld splenocytes showed more DN T cells in spleens than gld mice re- constituted with WT splenocytes. Impor- tantly, renal protection was observed in WT mice that received gld splenocytes (gld into WT), as indicated by lower serum creatinine at day 2 and day 3 after IRI, and less tubular damage both in the cortex and in the medulla than WT mice reconstituted with WT splenocytes (WT into WT). In contrast, se- rum creatinine after IRI in gld mice reconsti- tuted with WT splenocytes (WT into gld) was not different from that of WT into WT, al- though tubular damage in medulla was milder in WT into gld. This implies that FasL on leukocytes rather than on nonhemopoi- etic cells is the major mediator of the patho- genic effect of FasL seen during kidney IRI. The increased proportion of CD3ϩCD4Ϫ CD8Ϫ DN T lymphocytes in blood and spleens of gld mice has been previously described.21 35 Figure 5. Decreased TNF-␣ and IFN-␥ production by kidney T cells during IRI in mice Consistent with our previous study, our cur- treated with FasL blocking Ab compared to isotype control Ab. TNF-␣ and IFN-␥ rent data demonstrated that normal kidney production in T cells from postischemic kidneys at 3 days after IRI was lower in contained a large number of DN T cells. The FasL-blocking Ab–treated mice compared with isotype control Ab—treated group. proportion of DN T cells in kidneys, however, #P Ͻ 0.05 compared with isotype control Ab group. was not different between WT and gld mice. Similarly, the gld mutation does not change the which could also help explain attenuation of tissue injury in gld frequency of intraepithelial DN T cell population.36 Although an in- mice. Decreased production of the TNF-␣ and IFN-␥ inflamma- creasing body of evidence revealed the effect of DN T cells in immune tory cytokines by T lymphocytes isolated from postischemic kid- response,37,38 their role in kidney diseases is not well known and a neys was also seen after treatment of WT mice with FasL-blocking fertile topic for future studies. Ab. Thus, both genetically and pharmacologically, loss of FasL In summary, FasL deficiency conferred renal protection after function is associated with reduced production of inflammatory IRI, likely through reduced production of inflammatory cyto- cytokines by the kidney-infiltrating T cells. kines by T lymphocytes. The importance of FasL in ischemic AKI FasL deficiency could also lead to reduced neutrophil infiltra- was confirmed by complementary genetic and pharmacologic ap- tion. This possibility was assessed indirectly by measuring MPO proaches using gld mutant mice and FasL-blocking mAb, respec- activity in postischemic kidneys. FasL deficiency resulted in a tively. Blockade of FasL on leukocytes may represent a novel ther- trend toward lower MPO level after IRI as compared with WT. In apeutic strategy for preventing ischemic AKI. addition, the MPO level was significantly decreased by FasL blockade. Trafficking of NK cells into the kidney after IRI was different between WT and gld, but similar in mice treated with CONCISE METHODS FasL-blocking Ab compared with mice treated with control Ab (data were not shown). The possible role for NK cells in mediating Animal and Experimental Protocols FasL effect on renal IRI is yet to be fully determined. Wild-type (C57BL/6J) and FasL-deficient (gld) mice on the same Recent studies have demonstrated Fas and FasL expression af- background (B6Smn.C3-Faslgld/J) were purchased from the Jackson ter IRI,33 and the effect of this interaction was dependent on the Laboratory (n ϭ 11 for each group). All mice were 8-week-old males cell types.17,34 To determine the pathogenic source of FasL, we and were housed in a specific pathogen-free barrier facility. The Johns

J Am Soc Nephrol 22: 732–742, 2011 Fas Ligand of Leukocyte in Kidney IRI 737 BASIC RESEARCH www.jasn.org

ered formalin followed by paraffin embedding, and then renal sections were stained with hema- toxylin and eosin (H&E). Renal tubular damage was scored in a blinded fashion by a renal pathol- ogist.

Caspase 3 Immunohistochemistry Cleaved p17 caspase-3 is the active fragment of pro-caspase-3, which targets the key modulator of the apoptotic pathway and therefore represents a good marker for apoptosis. The staining was per- formed on formalin-fixed kidney sections (4 ␮m), after deparaffinization and rehydration. Cleaved caspase-3 (17/19 kDa) was stained using rabbit an- ti-mouse polyclonal Ab (Cell Signaling Technol- ogy, Inc., Danvers, MA), following the manufac- turer’s protocol, with biotinylated secondary Ab to rabbit IgG (Jackson ImmunoResearch, West Grove, PA). Ten randomly chosen, nonoverlap- ping fields from the same section of renal cortex and outer medulla were captured (ϫ200, Olym- pus BX51, Japan). The area of positive staining (dark brown color) was calculated with a com- puter program, Image-Pro Plus (Media Cybernet- ics, Silver Spring, MD), and it was expressed as the proportion of stained area over the entire field.

Flow Cytometric Analysis of ϩ Ϫ Ϫ Figure 6. Increased numbers of double negative T cells (CD3 CD4 CD8 ) in spleen Kidney-Infiltrating Mononuclear Cells of wild type recipients reconstituted with gld splenocytes compared to wild type (KMNCs) splenocytes (WT into WT) or gld recipient reconstituted with wild type splenocytes (WT KMNCs were isolated according to a method pre- into gld). Splenic DN T cells of gld recipient reconstituted with wild-type splenocytes viously described.39 Briefly, decapsulated kidneys (WT into gld) decreased to the same level with wild-type mice reconstituted with were immersed in RPMI buffer (Mediatech, Ma- wild-type splenocytes (WT into WT), whereas wild-type recipient reconstituted with gld nassas, VA) containing 5% fetal bovine serum splenocytes (gld into WT) showed significant increase of DN T cells in spleens. †P Ͻ (FBS) and disrupted mechanically using Stom- Ͻ 0.05 compared with WT into WT; ‡P 0.05 compared with WT into gld. acher 80 Biomaster (Seward, UK). Samples were Hopkins University Animal Care and Use Committee approved all strained, washed, and resuspended in 36% Percoll (Amersham Pharma- studies. Mice were anesthetized with an intraperitoneal injection of cia Biotech, Piscataway, NJ) followed by gentle overlaying onto 72% Per- sodium pentobarbital (75 mg/kg). After abdominal midline incision, coll. After centrifugation at 1000g for 30 minutes at room temperature, both renal pedicles were bluntly dissected and clamped with a micro- KMNCs were collected from the Percoll interface, washed twice, and vascular clamp (Roboz Surgical Instrument, Gaithersburg, MD) for counted on a hemocytometer using trypan blue exclusion. Staining for 30 minutes. During the procedure, mice were kept well hydrated with surface markers of KMNCs was then performed as described previ- warm sterile saline at a constant temperature (37°C). After the clamps ously.39 Isolated KMNCs were preincubated with anti-CD16/CD32 Fc were removed, the wounds were sutured and the mice allowed to receptor–blocking antibody for 10 minutes to minimize nonspecific an- recover with free access to chow and water. Cohorts of mice were tibody binding. Cells were then incubated with anti-mouse anti-CD1d, sacrificed on day 1 and day 3 after surgery. CD3, CD4, CD5, CD8, CD19, CD69, IgM, TCR-␤, TCR-␥␦, and NK1.1 (all from BD Biosciences) for 25 minutes at 4°C, washed with FACS Assessment of Renal Function buffer, and fixed 1% paraformaldehyde solution. Four-color immuno- Blood samples were obtained from the tail vein before and at 24, 48, fluorescence staining was analyzed using a FACSCalibur instrument (BD and 72 hours after renal IRI. Serum creatinine (mg/dl) was measured Biosciences) and FCS Express V3 (De Novo Software, Los Angeles). Each with a Cobas Mira plus autoanalyzer (Roche Diagnostics Corp., Indi- assay included at least 10,000 gated events. anapolis, IN).

Tissue Histologic Analysis Intracellular Cytokine Staining Mice were sacrificed at 72 hours after renal IRI. All kidneys were har- One million KMNCs were suspended in RPMI 1640 medium with vested after full exsanguination. Tissue samples were fixed with 10% buff- 10% FBS, L-glutamine, and penicillin/streptomycin and incubated

738 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 732–742, 2011 www.jasn.org BASIC RESEARCH

Figure 8. Reduced myeloperoxidase activity in post ischemic kid- neys in mice treated with FasL blocking Ab compared to isotype control Ab. Myeloperoxidase, which was increased at 3 days after IRI in WT compared with nonoperated WT mice, was not different in postischemic kidneys of gld compared with WT, but showed a trend to be lower (P ϭ 0.07; panel A). In the partial chimeric mice, WT mice reconstituted with WT splenocytes showed higher MPO activity compared with other groups after IRI (B). MPO level after IRI was significantly lower in FasL-blocking Ab–treated mice compared with Figure 7. Wild type mice reconstituted with gld splenocytes isotype control Ab–treated mice (C). WT and gld: kidneys from nonop- (gld into WT) show significant functional and histologic renal erated mice; IRI WT and IRI gld: postischemic kidneys at 3 days after IRI. protection after IRI compared to wild type reconstituted with *P Ͻ 0.05 compared with WT, †P Ͻ 0.05 compared with WT into WT, wild type splenocytes (WT into WT). Serum creatinine was and #P Ͻ 0.05 compared with isotype control Ab group. significantly lower in wild-type mice reconstituted with gld splenocytes (gld into WT) at day 2 and day 3 after IRI compared with wild-type receiving wild-type splenocytes (WT into WT). at 37°C in the presence of phorbol myristate acetate (5 ng/ml), Serum creatinine of gld mice receiving wild-type splenocytes ionomycin (500 ng/ml; both Sigma-Aldrich), and monensin (BD (WT into gld) was not different from other groups (n ϭ 8 for Biosciences). After a 5-hour culture, cells were washed and stained for each group; panel A). In histologic examination (B), tubular ␤ damage was significantly lower both in cortex and in medulla surface markers using anti–CD4-PerCP and anti–TCR- -allophycocya- of gld into WT compared with WT into WT. WT into gld showed nin Abs. These cells were then permeabilized in Cytofix/Cytoperm solu- reduced tubular damage in medulla compared with WT into tion (BD Biosciences) for 20 minutes and washed twice with perm/wash WT (Figure 7C). Arrows indicate necrotic tubules and arrow- buffer. Cells were then incubated with anti–TNF-␣-FITC and anti– heads indicate necrotic debris. Magnification: ϫ200. †P Ͻ 0.05 IFN-␥-PE Abs for 20 minutes, washed with perm/wash buffer, and compared with WT into WT. analyzed with a FACSCalibur instrument (BD Biosciences).

J Am Soc Nephrol 22: 732–742, 2011 Fas Ligand of Leukocyte in Kidney IRI 739 BASIC RESEARCH www.jasn.org

recipients (800 cGy). Three types of partial chi- mera were developed. The first group was WT recipients receiving WT splenocytes and the sec- ond was WT recipients receiving splenocytes from gld mice, whereas the third group was gld mice receiving WT splenocytes (n ϭ 8 for each group). Bilateral renal pedicle clamping was performed in recipient mice 2 days after adop- tive transfer. Functional and histologic parame- ters were assessed.

Measurement of Myeloperoxidase Activity by ELISA Myeloperoxidase level in postischemic kidneys was measured in whole-kidney protein extracts in each experimental group at day 3 after IRI by ELISA (Hycult biotech, The Netherlands) ac- cording to the manufacturer’s recommended protocol.

Bioplex Protein Array A panel of cytokines was measured in whole kid- ney protein extracts using the Bioplex Protein Array system (Bio-rad, Hercules, CA), which is a multiplexed, particle-based, flow cytometric as- say that utilizes anticytokine monoclonal anti- bodies linked to microspheres incorporating distinct properties of two fluorescence dyes. Our assay was designed to detect and quantify IL-4, IL-6, IL-10, IFN-␥, and monocyte chemot- actic protein-1 (MCP-1). Each cytokine value was normalized by dividing the raw cytokine Figure 9. Decreased IFN-␥ expression in post-ischemic kidneys of FasL deficient (gld) concentration (pg/ml) by the kidney protein mice compared to wild type. Level of IFN-␥ in postischemic kidneys was lower in gld concentration (mg/ml) measured by the Brad- mice compared with that in WT at day 3 after IRI. In partial chimeric mice, level of ford assay. MCP-1 in postischemic kidneys of gld into WT was significantly lower compared with Ͻ Ͻ those in other groups. *P 0.05 compared with WT, †P 0.05 compared with WT into Statistical Analysis WT, and ‡P Ͻ 0.05 compared with WT into gld. All data were expressed as mean Ϯ SEM). Group means were compared using Mann-Whitney In Vivo FasL Blockade analysis and ANOVA followed by Newman-Keuls post hoc analysis in Neutralizing anti-FasL monoclonal IgG (MFL4) is previously de- GraphPad Prism version 4 (GraphPad Software, LA Jolla, CA). Survival scribed.40 Anti-FasL MFL4 antibody (Ab) or control hamster IgG rate was compared using Kaplan-Meier method in GraphPad Prism ver- (n ϭ 7 for each group) were injected intraperitoneally (500 ␮g per sion 4. Statistical significance was determined as a P value Ͻ0.05. mouse). Bilateral renal pedicle clamping was performed 24 hours af- ter Ab administration, and booster injection at half dose was admin- istered into the peritoneal cavity immediately after reperfusion during ACKNOWLEDGMENTS the procedure. Functional and histologic parameters were assessed. G. J. Ko was supported by the Basic Science Research Program Comparison of Renal Function after IRI in Mice through the National Research Foundation of Korea (NRF) funded by Receiving Sublethal Irradiation and Adoptive Transfer the Ministry of Education, Science, and Technology (E00029). H.R. of Splenocytes was supported by the US National Institutes of Health, National Kid- To determine the relative importance of leukocytes or organ resident ney Foundation and by Mr. Rogelio Miro of Panama. cells on renal protection of FasL, we prepared partial bone marrow chimeric mice by adoptive transfer of splenocytes after sublethal irra- diation. Single-cell suspension prepared from spleens of donors was DISCLOSURES injected intravenously (2.5 ϫ 107 cells) into sublethally irradiated None.

740 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 732–742, 2011 www.jasn.org BASIC RESEARCH

REFERENCES 20. Itoh N, Imagawa A, Hanafusa T, Waguri M, Yamamoto K, Iwahashi H, Moriwaki M, Nakajima H, Miyagawa J, Namba M, Makino S, Nagata S, Kono N, Matsuzawa Y: Requirement of Fas for the development of 1. Perico N, Cattaneo D, Sayegh MH, Remuzzi G: Delayed graft function autoimmune diabetes in nonobese diabetic mice. J Exp Med 186: in kidney transplantation. Lancet 364: 1814–1827, 2004 613–618, 1997 2. Star RA: Treatment of acute renal failure. Kidney Int 54: 1817–1831, 21. Mohamood AS, Guler ML, Xiao Z, Zheng D, Hess A, Wang Y, Yagita H, 1998 Schneck JP, Hamad AR: Protection from autoimmune diabetes and 3. Thakar CV, Zahedi K, Revelo MP, Wang Z, Burnham CE, Barone S, T-cell lymphoproliferation induced by FasL mutation are differentially Bevans S, Lentsch AB, Rabb H, Soleimani M: Identification of throm- regulated and can be uncoupled pharmacologically. Am J Pathol 171: bospondin 1 (TSP-1) as a novel mediator of cell injury in kidney 97–106, 2007 ischemia. J Clin Invest 115: 3451–3459, 2005 22. Shiraishi H, Toyozaki T, Tsukamoto Y, Saito T, Masuda Y, Hiroshima K, 4. Bonventre JV, Weinberg JM: Recent advances in the pathophysiology of Ohwada H, Kobayashi N, Hiroe M: Antibody binding to fas ligand ischemic acute renal failure. J Am Soc Nephrol 14: 2199–2210, 2003 attenuates inflammatory cell infiltration and cytokine secretion, lead- 5. Brunner T, Mogil RJ, LaFace D, Yoo NJ, Mahboubi A, Echeverri F, ing to reduction of myocardial infarct areas and reperfusion injury. Lab Martin SJ, Force WR, Lynch DH, Ware CF, et al: Cell-autonomous Fas Invest 82: 1121–1129, 2002 (CD95)/Fas-ligand interaction mediates activation-induced apoptosis 23. Nogae S, Miyazaki M, Kobayashi N, Saito T, Abe K, Saito H, Nakane in T-cell hybridomas. Nature 373: 441–444, 1995 PK, Nakanishi Y, Koji T: Induction of apoptosis in ischemia-reperfusion 6. Hamar P, Song E, Kokeny G, Chen A, Ouyang N, Lieberman J: Small model of mouse kidney: Possible involvement of Fas. JAmSoc interfering RNA targeting Fas protects mice against renal ischemia- Nephrol 9: 620–631, 1998 reperfusion injury. Proc Natl Acad SciUSA101: 14883–14888, 2004 24. Tekin D, Xi L, Kukreja RC: Genetic deletion of fas receptors or Fas 7. Erkan E, Garcia CD, Patterson LT, Mishra J, Mitsnefes MM, Kaskel FJ, ligands does not reduce infarct size after acute global ischemia- Devarajan P: Induction of renal tubular cell apoptosis in focal segmen- reperfusion in isolated mouse heart. Cell Biochem Biophys 44: 111– tal glomerulosclerosis: Roles of proteinuria and Fas-dependent path- 117, 2006 ways. J Am Soc Nephrol 16: 398–407, 2005 25. Yang J, Jones SP, Suhara T, Greer JJ, Ware PD, Nguyen NP, Perlman 8. Del Rio M, Imam A, DeLeon M, Gomez G, Mishra J, Ma Q, Parikh S, H, Nelson DP, Lefer DJ, Walsh K: Endothelial cell overexpression of fas Devarajan P: The death domain of kidney ankyrin interacts with Fas ligand attenuates ischemia-reperfusion injury in the heart. J Biol Chem and promotes Fas-mediated cell death in renal epithelia. JAmSoc 278: 15185–15191, 2003 Nephrol 15: 41–51, 2004 26. Krueger A, Fas SC, Baumann S, Krammer PH: The role of CD95 in the 9. Dupont PJ, Warrens AN: Fas ligand exerts its pro-inflammatory effects regulation of peripheral T-cell apoptosis. Immunol Rev 193: 58–69, via neutrophil recruitment but not activation. Immunology 120: 133– 2003 139, 2007 27. Gonzalez-Cuadrado S, Lorz C, Garcia del Moral R, O’Valle F, Alonso C, 10. Cruise MW, Melief HM, Lukens J, Soguero C, Hahn YS: Increased Fas Ramiro F, Ortiz-Gonzalez A, Egido J, Ortiz A: Agonistic anti-Fas anti- ligand expression of CD4ϩ T cells by HCV core induces T cell- bodies induce glomerular cell apoptosis in mice in vivo. Kidney Int 51: dependent hepatic inflammation. J Leukoc Biol 78: 412–425, 2005 1739–1746, 1997 11. Park DR, Thomsen AR, Frevert CW, Pham U, Skerrett SJ, Kiener PA, 28. Miyazawa S, Watanabe H, Miyaji C, Hotta O, Abo T: Leukocyte accu- Liles WC: Fas (CD95) induces proinflammatory cytokine responses by mulation and changes in extra-renal organs during renal ischemia monocytes and monocyte-derived macrophages. J Immunol reperfusion in mice. J Lab Clin Med 139: 269–278, 2002 170: 6209–6216, 2003 29. Sayers TJ, Brooks AD, Lee JK, Fenton RG, Komschlies KL, Wigginton 12. Gregory MS, Koh S, Huang E, Saff RR, Marshak-Rothstein A, Mukai S, JM, Winkler-Pickett R, Wiltrout RH: Molecular mechanisms of immune- Ksander BR: A novel treatment for ocular tumors using membrane mediated lysis of murine renal cancer: Differential contributions of FasL vesicles to activate innate immunity and terminate immune priv- perforin-dependent versus Fas-mediated pathways in lysis by NK and ilege. Invest Ophthalmol Vis Sci 46: 2495–2502, 2005 T cells. J Immunol 161: 3957–3965, 1998 13. Dekel B, Bocher WO, Marcus H, Yussim A, Reisner Y: Acute cellular 30. Nagata S, Golstein P: The Fas death factor. Science 267: 1449–1456, rejection of human renal tissue by adoptive transfer of allogeneic human 1995 peripheral blood mononuclear cells into chimeric rats: Sequential 31. Legge KL, Braciale TJ: Lymph node dendritic cells control CD8ϩ T cell expression of cytokines, chemokines and cytolytic effector molecules, responses through regulated FasL expression. Immunity 23: 649–659, and their regulation by CTLA-4-Ig. Int Immunol 11: 1673–1683, 1999 2005 14. Modiano JF, Sun J, Lang J, Vacano G, Patterson D, Chan D, Franzusoff 32. Hegde S, Beauregard C, Mayhew E, Niederkorn JY: CD4(ϩ) T-cell- A, Gianani R, Meech SJ, Duke R, Bellgrau D: Fas ligand-dependent mediated mechanisms of corneal allograft rejection: Role of Fas- suppression of autoimmunity via recruitment and subsequent termi- induced apoptosis. Transplantation 79: 23–31, 2005 nation of activated T cells. Clin Immunol 112: 54–65, 2004 33. Nakajima H, Mizuta N, Fujiwara I, Sakaguchi K, Ogata H, Magae J, 15. Chuang YH, Suen JL, Chiang BL: Fas-ligand-expressing adenovirus- Yagita H, Koji T: Blockade of the Fas/Fas ligand interaction suppresses transfected dendritic cells decrease allergen-specific T cells and air- hepatocyte apoptosis in ischemia-reperfusion rat liver. Apoptosis 13: way inflammation in a murine model of asthma. J Mol Med 84: 1013–1021, 2008 595–603, 2006 34. Pinkoski MJ, Brunner T, Green DR, Lin T: Fas and Fas ligand in gut and 16. Boonstra JG, van der Woude FJ, Wever PC, Laterveer JC, Daha MR, liver. Am J Physiol Gastrointest Liver Physiol 278: G354–G366, 2000 van Kooten C: Expression and function of Fas (CD95) on human renal 35. Ascon DB, Ascon M, Satpute S, Lopez-Briones S, Racusen L, Colvin tubular epithelial cells. J Am Soc Nephrol 8: 1517–1524, 1997 RB, Soloski MJ, Rabb H: Normal mouse kidneys contain activated and 17. An S, Hishikawa Y, Koji T: Induction of cell death in rat small intestine CD3ϩCD4- CD8- double-negative T lymphocytes with a distinct TCR by ischemia reperfusion: Differential roles of Fas/Fas ligand and Bcl- repertoire. J Leukoc Biol 84: 1400–1409, 2008 2/Bax systems depending upon cell types. Histochem Cell Biol 123: 36. Mohamood AS, Bargatze D, Xiao Z, Jie C, Yagita H, Ruben D, Watson J, 249–261, 2005 Chakravarti S, Schneck JP, Hamad AR: Fas-mediated apoptosis regulates 18. Ortiz A, Lorz C, Egido J: The Fas ligand/Fas system in renal injury. the composition of peripheral alphabeta T cell repertoire by constitutively Nephrol Dial Transplant 14: 1831–1834, 1999 purging out double negative T cells. PLoS One 3: e3465, 2008 19. Hamad AR, Schneck JP: -induced T cell death is regulated by 37. Hamad AR, Mohamood AS, Trujillo CJ, Huang CT, Yuan E, Schneck CD4 expression. Int Rev Immunol 20: 535–546, 2001 JP: B220ϩ double-negative T cells suppress polyclonal T cell activa-

J Am Soc Nephrol 22: 732–742, 2011 Fas Ligand of Leukocyte in Kidney IRI 741 BASIC RESEARCH www.jasn.org

tion by a Fas-independent mechanism that involves inhibition of IL-2 40. Kayagaki N, Yamaguchi N, Nagao F, Matsuo S, Maeda H, Ok- production. J Immunol 171: 2421–2426, 2003 umura K, Yagita H: Polymorphism of murine Fas ligand that affects 38. Zhang ZX, Yang L, Young KJ, DuTemple B, Zhang L: Identification of the biological activity. Proc Natl Acad SciUSA94: 3914–3919, a previously unknown antigen-specific regulatory T cell and its mech- 1997 anism of suppression. Nat Med 6: 782–789, 2000 39. Ascon DB, Lopez-Briones S, Liu M, Ascon M, Savransky V, Colvin RB, Soloski MJ, Rabb H: Phenotypic and functional characterization of kidney-infiltrating lymphocytes in renal ischemia reperfusion injury. Supplemental information for this article is available online at http://www. J Immunol 177: 3380–3387, 2006 jasn.org/.

742 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 732–742, 2011