PEPITEM/ 15 Axis Inhibits T Lymphocyte Infiltration and Glomerulonephritis in a Mouse Model of Systemic Lupus Erythematosus This information is current as of September 29, 2021. Hidehito Matsubara, Yoshitaka Shimizu, Masaaki Arai, Akira Yamagata, Seigo Ito, Toshihiko Imakiire, Masashi Tsunoda, Hiroo Kumagai and Naoki Oshima J Immunol published online 13 March 2020 http://www.jimmunol.org/content/early/2020/03/12/jimmun Downloaded from ol.1900213 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 29, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2020 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published March 13, 2020, doi:10.4049/jimmunol.1900213 The Journal of Immunology

PEPITEM/Cadherin 15 Axis Inhibits T Lymphocyte Infiltration and Glomerulonephritis in a Mouse Model of Systemic Lupus Erythematosus

Hidehito Matsubara,* Yoshitaka Shimizu,† Masaaki Arai,‡ Akira Yamagata,* Seigo Ito,* Toshihiko Imakiire,* Masashi Tsunoda,x Hiroo Kumagai,* and Naoki Oshima*

Control of lymphocyte infiltration in kidney is a potential therapeutic strategy for lupus nephritis, considering that control of lymphocyte migration by sphingosine 1 phosphate has been implicated in inflammation-related pathology. The peptide inhibitor of the transendo- thelial migration (PEPITEM)/cadherin (CDH) 15 axis was recently reported to promote sphingosine 1 phosphate secretion. In this study, we investigated whether CDH15 is expressed in the kidney of MRL/lprmiceandwhetherlymphocyteinfiltrationissuppressedbyex- Downloaded from ogenously administered PEPITEM. Mice (18 wk old) were randomized into 4-wk treatment groups that received PEPITEM or PBS encapsulated in dipalmitoylphosphatidylcholine liposomes. Enlargement of the kidney, spleen, and axillary lymph nodes was suppressed by PEPITEM treatment, which also blocked infiltration of double-negative T lymphocytes into the kidney and glomerular IgG/C3 de- position, reduced proteinuria, and increased podocyte density. Immunohistochemical analysis revealed that the PEPITEM receptor CDH15 was expressed on vascular endothelial cells of glomeruli and kidney arterioles, skin, and peritoneum in lupus mice at 22 wk of age but not in 4-wk-old mice. These results suggest that PEPITEM inhibits lymphocyte migration and infiltration into the kidney, http://www.jimmunol.org/ thereby preserving the kidney structure and reducing proteinuria. Thus, PEPITEM administration may be considered as a potential therapeutic tool for systemic lupus erythematosus. The Journal of Immunology, 2020, 204: 000–000.

he pathogenesis of systemic lupus erythematosus (SLE), a phosphate (S1P) (7, 8), a lipid mediator derived from sphingolipids. chronic autoimmune disease, is not well understood; how- Tissue and blood concentrations of S1P are important for lymphocyte T ever, various targeted therapeutic agents are being developed infiltration into tissues (9–11), which plays an important role in local (1). It is known that nephritis manifests in MRL/MpJJmsSlc-lpr/lpr inflammation (12, 13). PEPITEM controls lymphocyte rolling, (MRL/lpr) mice with age and is accompanied by T lymphocyte adhesion, and migration under inflammatory conditions (7) and by guest on September 29, 2021 infiltration into the interstitium, vasculitis, and immune complex promotes S1P synthesis in vascular endothelial cells through deposition (2). cadherin (CDH) 15, which is expressed as a receptor on inflam- Lupus nephritis is characterized by infiltration of inflammatory matory vascular endothelial cells. Secreted S1P acts on S1P re- cells into the kidney and formation of immune complexes followed ceptors of surrounding T lymphocytes and blocks the interaction by the release of soluble inflammatory cytokines associated with between lymphocyte function-associated Ag-1 and intracellular tissue injury. The local adaptive immune response further amplifies adhesion molecule-1, thereby inhibiting T lymphocyte extrava- inflammation (3), which exacerbates kidney function. Therefore, sation. The S1P receptor and PEPITEM are being investigated controlling the migration of T lymphocytes may serve as a ther- for development of novel therapeutic approaches for autoim- apeutic approach for SLE (4–6). mune diseases. Fingolimod (FTY720), a drug targeting the S1P Peptide inhibitor of transendothelial migration (PEPITEM) is an receptor, has been used for the treatment of multiple sclerosis endogenous peptide that promotes the synthesis of sphingosine 1 (14, 15), whereas PEPITEM is currently being investigated for the treatment of rheumatoid arthritis. *Department of Nephrology and Endocrinology, National Defense Medical College, However, the precise role of PEPITEM in the inflammatory Tokorozawa, Saitama 359-8513, Japan; †Department of Applied Biochemistry, Tokai pathology of SLE remains unclear, and the use of PEPITEM for University, Hiratsuka, Kanagawa 259-1207, Japan; ‡Department of Biochemistry, x SLE treatment has not yet been reported. Therefore, in this study, National Defense Medical College, Tokorozawa, Saitama 359-8513, Japan; and De- partment of Preventive Medicine and Public Health, National Defense Medical Col- we investigated whether exogenously administered PEPITEM can lege, Tokorozawa, Saitama 359-8513, Japan suppress lymphocyte infiltration into various organs of MRL/lpr ORCIDs: 0000-0001-9517-7746 (H.M.); 0000-0002-3496-1198 (M.A.); 0000-0003- mice. We also performed histological and biochemical analyses to 2892-1859 (M.T.). determine whether suppressing lymphocyte infiltration can pre- Received for publication February 25, 2019. Accepted for publication January 26, vent the development of nephritis. 2020. Address correspondence and reprint requests to Dr. Hidehito Matsubara, Department Materials and Methods of Nephrology and Endocrinology, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama 359-8513, Japan. E-mail address: [email protected] Reagents Abbreviations used in this article: A/G, albumin/globulin; CDH, cadherin; DAB, The peptide and Abs used in this study are listed in Table I. 3,39-diaminobenzidine tetrahydrochloride; DN, double-negative; DPPC, dipalmitoyl- phosphatidylcholine; PEPITEM, peptide inhibitor of the transendothelial migration; Animals PTC, peritubular capillary; SLE, systemic lupus erythematosus; a-SMA, a–smooth muscle actin; S1P, sphingosine 1 phosphate; WT, Wilms tumor. Animal breeding and experimental procedures were performed in accor- dance with the animal experiment ethical guidelines of the National Defense Copyright Ó 2020 by The American Association of Immunologists, Inc. 0022-1767/20/$37.50 Medical College. Female MRL/lpr mice (18 wk old; SLC, Tokyo, Japan)

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1900213 2 PEPITEM/CADHERIN 15 AXIS INHIBITS GLOMERULONEPHRITIS IN SLE were s.c. injected twice weekly with PEPITEM contained in dipalmi- and blood urea nitrogen (enzymatic method) levels and albumin/globulin toylphosphatidylcholine (DPPC) liposomes at 400 mg/dose (n = 12), as (A/G) ratio were measured (Oriental Yeast, Tokyo, Japan). Anti- previously described (16–18). Mice in the vehicle control group were dsDNA Ab was measured by REVIS anti-dsDNA Mouse ELISA Kit injected with only DPPC liposomes (n = 12). At 22 wk of age, mice (Shibayagi). were deeply anesthetized and euthanized, and blood was collected by cardiac puncture. The kidneys, spleen, and axillary lymph nodes were Immunohistochemistry and immunofluorescence analysis removed for histological and biochemical analyses. PEPITEM (amino acid sequence SVTEQGAELSNEER) was purified by postsynthesis Paraffin-embedded sections were stained with H&E and periodic acid– HPLC (SCRUM, Tokyo, Japan). The peptide (purity $95%) was dis- Schiff. The sections were heat-treated with citrate or Tris-EDTA buffer and solved in PBS and combined with DPPC liposomes prior to injection. then reacted with the primary Ab followed by the appropriate secondary Liposomes were readily accumulated in the reticuloendothelial system Ab. Immunoreactivity was visualized with 3,39-diaminobenzidine tetra- (kidney, spleen, etc.) (19), and were therefore used as a drug delivery hydrochloride (DAB) or by fluorescence detection. Frozen-fresh kidney system to target reactive peptides with a short half-life to the kidney. sections were also labeled with the primary Ab and corresponding sec- DPPC liposome was provided by Dr. Shimizu (Tokai University, Kanagawa, ondary Ab and were reacted with DAB. For detection of FITC-labeled Japan). PEPITEM (0.02 mg/ml), the sections were washed lightly and refixed with 4% paraformaldehyde at room temperature for 10 min. Thereafter, Lymphocyte analysis HRP-conjugated anti-FITC Ab (HPI, Burlington, MA) and anti-rabbit IgG-HRP (Nichirei, Tokyo, Japan) were allowed to react and then de- The excised kidney was sliced with a scalpel blade, passed through a veloped with DAB. Fresh-frozen skin and peritoneum were fixed with mesh sieve, treated with collagenase at 37˚C for 30 min, and passed ice-cold acetone for 10 min and then a mixed primary Ab was reacted through a 70-mm mesh to obtain dissociated kidney mononuclear cells 3 7 overnight at 4˚C. The secondary Ab was reacted with goat anti-rabbit (20). Cell density was adjusted to 1 10 /ml; after adding Fc blocker, Alexa Fluor 594 (Invitrogen, Carlsbad, CA) and goat anti-rat Alexa Fluor the cells were labeled with primary Abs (21) and then sorted by flow

488 (Invitrogen) for 1 h at room temperature. Specimens were visualized Downloaded from cytometry on a BD FACSCanto II instrument (BD Biosciences, Franklin with a Pulse-SIM BZ-X700 microscope (Keyence, Osaka, Japan). IgG Lakes, NJ). and C3 were scored on a relative scale of 0–3 for at least 20 glomeruli Urine and blood sample collection per mouse. CD3 and Ki67 immunoreactivity was scored for at least 40 glomeruli per mouse. Sections were cut out in the longitudinal direction Urinary albumin level was measured with the REVIS Albumin Assay Kit at 3- and 9-mm thickness, and staining for Wilms tumor (WT) 1 was (Shibayagi, Gunma, Japan). Urinary creatinine (enzymatic method), se- performed. The complete cortex was photographed with a 40-fold visual rum total (Biuret method), albumin (Bromocresol Green method), field, and the number of WT1-stained podocyte nuclei per glomerulus http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. PEPITEM suppresses organ enlargement and T lymphocyte infiltration in the kidney. (a) Comparison of organ weights in the PEPITEM and vehicle groups (n = 12 each). Kidney, spleen, and lymph node weights were lower in the PEPITEM than in the vehicle group. (b) Absolute number of WBCs among kidney mononuclear cells (n = 5). WBC count was lower in the PEPITEM group (c and d) Percentage of Th cells (CD3+CD4+CD82), cytotoxic T lymphocytes (CTL; CD3+CD42CD8+), and DN T cells (CD3+CD42CD82) in kidney-mononuclear cells, as detected by flow cytometry (n = 5). There were no differences in the Th cell and CTL fractions between the groups, but the DN T cell population was smaller in the PEPITEM than in the vehicle group. Values represent mean 6 SD. *p , 0.05 versus vehicle. The Journal of Immunology 3

Table I. Peptides and Abs used

Peptides and Abs Host Manufacturer N-terminal FITC-PEPITEM Scrum CD3e Goat Santa Cruz Biotechnology, Santa Cruz, CA F4/80 Rabbit Santa Cruz Biotechnology Ki-67 Rat DAKO, Agilent Technology, Santa Clara, CA CDH15 Rabbit Cusabio Biotech, Newark, DE CD45 Rat, monoclonal 30F-11; BD Biosciences Pharmingen, San Diego, CA CD3e Rat, monoclonal 17A2; BioLegend, San Diego, CA CD4 Rat, monoclonal RM4-5; BioLegend CD8 Rat, monoclonal 53-6.7; BioLegend Nephrin Guinea pig PROGEN Biotechnik GmbH, Heidelberg, Germany WT1 Rabbit Santa Cruz Biotechnology CD31 Rat, monoclonal ER-MP12; Acris, Herford, Germany FITC-IgG Goat MP Biomedicals, Santa Ana, CA FITC-C3 Goat MP Biomedicals HRP–a-SMA Mouse, monoclonal 1A4; DAKO HRP–b-actin Goat Santa Cruz Biotechnology Downloaded from was determined. The number of podocytes was calculated using the Isolation of glomeruli following equation and compared with a previously described method (22): podocyte number = average glomerular volume/average volume per Glomeruli were isolated by a previously reported method (24) with slight podocyte. modifications. After refluxing the Dynabeads M-450 Tosylactivated (Dynal, Nephrin fluorescence was calculated and compared by semiquantification Oslo, Norway) from the heart, the kidney cortex was harvested. After incubation with the following equation (23): specific fluorescence site = fluorescence at 37˚C for 30 min, the cells were incubated with collagenase. The collagenase- m intensity lining area (lining the capillary loops)/total glomerular area. treated tissue was passed through a 100- m cell strainer, washed with PBS, and Images were analyzed with ImageJ v. 1.4.3.67 software (National In- centrifuged (1600 rpm, 4˚C, 5 min). Glomeruli containing Dynabeads were http://www.jimmunol.org/ stitutes of Health, Bethesda, MD). collected with a magnetic particle concentrator (Dynal) and washed with PBS. All the steps were performed at 4˚C, except the collagenase treatment step. Immunoelectron microscopy Electrophoresis and Western blotting Excised kidneys were fixed overnight at 4˚C in 4% paraformaldehyde and frozen with liquid nitrogen. After incubation with the primary Ab, speci- After isolating the glomeruli from the kidneys excised from two mice, ultrasonic mens were washed and incubated with gold colloid (particle size = 10 disruption was performed in radioimmunoprecipitation assay buffer containing nm)–conjugated secondary Ab (Cytodiagnostics, Burlington, ON, Canada) protease inhibitor. Equal amounts of protein lysate were denatured with sample and examined by electron microscopy (JEM-1400 Plus; Japan Electron buffer, separated by 10% SDS-PAGE, and transferred to a polyvinylidene Optics Laboratory, Tokyo, Japan). difluoride membrane. Protein bands were visualized with a ChemiDoc XPS+ by guest on September 29, 2021

FIGURE 2. PEPITEM reduces albuminuria. (a) Comparison of urine protein fractions of MRL/+ mice and vehicle and PEPITEM-treated MRL/lpr mice by electrophoresis. Distinct were excreted in the urine from the vehicle group and MRL/+ mice. (b) Comparison of albuminuria based on the albumin creatinine ratio. The ratio was smaller in the PEPITEM than in the vehicle group. Graph shows the median and interquartile range. *p , 0.05 versus vehicle. 4 PEPITEM/CADHERIN 15 AXIS INHIBITS GLOMERULONEPHRITIS IN SLE

Table II. Laboratory test results were considered statistically significant. Data were analyzed using JMP v. 8.0.2 software (SAS Institute, Cary, NC). Vehicle (n = 12) PEPITEM (n = 12) BUN (mmol/l) 9.97 6 0.42 9.32 6 0.51 Results 6 6 TP (g/l) 69.2 2.4 67.6 1.9 PEPITEM prevents organ enlargement and reduces the Alb (g/l) 30.8 6 0.6 32.2 6 1.1 A/G 0.86 6 0.05 1.11 6 0.05a number of WBCs in kidney mononuclear cells in MRL/lpr mice a Anti-dsDNA Ab (U/ml) 1696 6 211 1345 6 138 In MRL/lpr mice, lymphocytes infiltrate into multiple organs, ap value , 0.05 (PEPITEM versus vehicle). resulting in their swelling with age. We found that overall body weight as well as kidney, spleen, and axillary lymph node weights system (Bio-Rad Laboratories, Hercules, CA), and signal intensity was in PEPITEM-treated mice were lower than those in the vehicle- quantified using the Image-Lab software (Bio-Rad Laboratories). Urine samples treated ones (Fig. 1a). were denatured in sample buffer and resolved by 12% SDS-PAGE using a TGX To examine the effect of PEPITEM on cellular infiltration into Stain-Free Fast Cast Acrylamide Kit (Bio-Rad Laboratories). Protein bands were visualized by irradiation with UV light. Each lane contained 5 mlofurine. the kidney and determine the fraction of lymphocytes, we per- formed flow cytometry analysis of kidney mononuclear cells. Statistical analysis Kidney WBC count was lower in the PEPITEM group than that in All measurements are shown as mean 6 SD. Differences between groups the control group (Fig. 1b). Although there was no difference in were evaluated with the Wilcoxon rank sum test; differences with p , 0.05 the ratio of cluster of differentiation (CD) 4+ (Th cells) to CD8+ Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 3. PEPITEM suppresses cell invasion. (a) H&E and periodic acid–Schiff staining of the kidney tissue. Perivascular lymphocyte infiltration was reduced in the PEPITEM group, and changes in glomerular cell proliferation were less apparent in the PEPITEM group. (b and c) Quantification of CD3+ cells in glomeruli. Fewer CD3+ cells were observed in the PEPITEM group. (b and d) Quantification of F4/80+ cells in the interstitium. Fewer F4/80+ cells were observed in the PEPITEM group. (b and e) Quantification of Ki67+ cells in glomeruli. Fewer Ki67+ cells were observed in the PEPITEM group. Scale bar, 40 mm. Values represent mean 6 SD. *p , 0.05 versus vehicle. GCS, glomerular cross-section; HPF, high-power field. The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/

FIGURE 4. PEPITEM inhibits IgG and C3 deposition in glomeruli. (a) IgG and C3 deposition in frozen-fresh kidney sections was detected using FITC- conjugated anti-IgG and -C3 Abs, and immunoreactivity was scored. (b) IgG and C3 deposition in glomeruli was reduced in the PEPITEM group. Scale bar, 40 mm. Values represent mean 6 SD (n = 6). *p , 0.05 versus vehicle. by guest on September 29, 2021 cytotoxic T lymphocytes between the groups, the ratio of double- significantly lower in PEPITEM-treated mice than that in control negative (DN; CD3+CD42CD82) T lymphocytes was lower in mice (Fig. 4a, 4b). the PEPITEM group (Fig. 1c, 1d). These results suggested that PEPITEM has an indirect protective effect on podocytes PEPITEM suppresses lymphocyte migration and infiltration into various organs (Table I). The number of cells expressing the podocyte marker WT1 (Fig. 5a, 5b) and the glomerular slit diaphragm marker nephrin (Fig. 5c, 5d) PEPITEM treatment alters urine protein concentrations was higher in the glomeruli of PEPITEM-treated mice than in the We examined whether PEPITEM administration could suppress the glomeruli of control mice. progression of nephritis. Urine protein electrophoresis revealed that CDH15 expression in vascular endothelial cells of the kidney, PEPITEM reduced the urine concentration of various proteins skin, and peritoneum in MRL/lpr mice (mainly albumin) relative to the levels observed in MRL/lpr mice (Fig. 2a). The albumin/creatinine ratio was lower in the PEPITEM Expression of CDH15 was examined in the kidney, skin, and than in the control group after treatment for 4 wk (Fig. 2b); however, peritoneum, which are the diseased organs in SLE. CDH15 was not there were no difference in serum urea nitrogen or albumin levels detected in the vascular endothelium of the kidneys in 4-wk-old (Table II). Furthermore, the A/G ratio was elevated in the PEPITEM mice but was present in the vascular smooth muscles (Fig. 6a, group, suggesting that serum Ig was decreased by PEPITEM treat- 6b). In contrast, CDH15 expression was observed not only in ment (Table II). Anti-ds DNA Ab was decreased in the PEPITEM smooth muscle cells, but also on vascular endothelial cells in the treatment group compared with the control group (Table II). peritubular capillaries (PTCs) and glomerular capillaries at 22 wk (Fig. 6c, 6d). Immunoelectron microscopy analysis showed that PEPITEM improves nephritis in MRL/lpr mice CDH15 was localized on the surface of vascular endothelial cells Histological analysis revealed that as lymphocyte infiltration into of glomeruli (Fig. 6e) and in the vascular lumen (Fig. 6f). CDH15 the interstitium decreased (Fig. 3a), immunopositivity for CD3+ expression in pooled glomeruli isolated from two MRL/+ or MRL/ (a marker of T cells) in the glomerulus also decreased in the lpr mice at 22 wk of age was confirmed by Western blotting PEPITEM group (Fig. 3b, 3c). Similarly, expression of interstitial (Fig. 6g). No cells were found to express a–smooth muscle actin F4/80+ (a marker of monocytes and macrophages) cells (Fig. 3b, 3d) (a-SMA) in glomeruli of 22-wk-old MRL/+ and MRL/lpr mice by and the number of cells positive for the proliferation marker Ki67 in immunohistochemistry (Fig. 6h), suggesting that CDH15 detected in + the glomeruli were also reducedinthesemice(Fig.3b,3e). isolated glomeruli by Western blotting was not derived from a-SMA vascular smooth muscle cells. We also detected cells double-positive IgG and C3 deposition in glomeruli is alleviated by PEPITEM for the vascular endothelial cell surface marker CD31 (Fig. 7a, 7d) We assessed the extent of IgG and C3 deposition in the glomerulus and CDH15 (Fig. 7b, 7e) in the skin and peritoneum of 22-wk-old by immunofluorescence labeling and found that deposition was mice by immunofluorescence labeling (Fig. 7c, 7f). 6 PEPITEM/CADHERIN 15 AXIS INHIBITS GLOMERULONEPHRITIS IN SLE Downloaded from http://www.jimmunol.org/

FIGURE 5. PEPITEM protects against podocytes. (a and b) Immunohistochemical analysis of the number of WT1-positive cells in glomeruli. There were more WT1+ cells in the PEPITEM group. (c and d) Relative fluorescence intensity of nephrin in glomeruli. The fluorescence was higher in the PEPITEM group. Scale bars, 40 mm(a and c). Values represent mean 6 SD (n = 6). *p , 0.05 versus vehicle. by guest on September 29, 2021 PEPITEM binds to glomerular and kidney artery endothelium lower level in healthy elderly people and in patients with rheu- and vascular smooth muscle matoid arthritis and type 1 diabetes than in healthy humans (7). Immunohistochemical staining using labeled PEPITEM revealed The relationship between chronic inflammation and adipo- immunopositivity only in vascular smooth muscle cells of 4-wk- has recently been reported (26, 27). Although PEPITEM old mice, whereas glomeruli and PTCs were negative (Fig. 8a, released from B cells is produced by stimulation of adiponectin, 8b). No positivity was observed in cells stained with the FITC- adiponectin receptors on B cells are decreased owing to chronic labeled scrambled peptide (containing the same amino acids as inflammatory pathology and aging, thereby decreasing blood PEPITEM but randomly arranged) used as a negative control PEPITEM concentration. As SLE is characterized by chronic in- (Fig. 8c). In contrast, positive cells were detected in the vas- flammatory pathology, we speculated that endogenous PEPITEM cular smooth muscle and vascular endothelial cells of arterioles is downregulated in this disease. In this study, we showed that (Fig. 8d, 8e) as well as in the vascular smooth muscle cells and inflammatory cell infiltration into kidney, spleen, and axillary vascular endothelial cells of the interlobular artery (Fig. 8f) in lymph nodes was reduced by PEPITEM treatment. The flow 22-wk-old mice (Fig. 9). cytometry analysis showed that, in addition to the total number of T cells, the total number of DN T cells and its ratio to total lymphocytes was decreased in the kidneys. Furthermore, a Discussion + The four major findings of this study are as follows. 1) The weights decrease in the number of F4/80 cells in the renal interstitium and of the kidney, spleen, and axillary lymph nodes in MRL/lpr mice amelioration of deposition of IgG and C3 in the glomeruli upon were reduced by PEPITEM treatment relative to those in the ve- PEPITEM treatment were observed by immunohistochemistry. hicle group; 2) PEPITEM diminished the total number of WBCs in This is likely because persistent B cell and macrophage activation the kidney as well as the proportion of DN T cells; 3) glomerular was suppressed along with the decrease in T cells (28, 29). Au- injury and vasculitis were milder and albumin excretion was lower toimmune B cells often require T cell help so that decreased ac- in the PEPITEM group; and 4) CDH15, the PEPITEM receptor, tivation of T cells can result in less activation of B cells. This is was expressed on vascular endothelial cells of the kidney, skin, and true in the MRL/lpr model in which T cells are required for dis- peritoneum. ease to occur. Stimulation with costimulatory molecules such as Proteome analysis of macrophages/monocytes collected from CD40, CD80, and CD86 from T cells, clonal expansion of B cells ascites of patients with epithelial ovarian cancer revealed the by follicular T cells, Ab isotype switch, and plasma cell differ- presence of peptides of the sequence SVTEQGAELSNEER de- entiation have been reported (30, 31). Thus, it appears that the rived from 14-3-3 z protein (25). It was previously reported that a decrease in T cells affected the interaction with B cells, resulting peptide with a similar sequence, cleaved from the 14-3-3 z-d in a decrease in B cells and immune complex deposition in the protein and secreted from B cells into blood, was present at a kidney. Podocytes were indirectly protected by the reduction in The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 6. CDH15 is expressed in the renal vascular endothelium as a function of disease progression. (a and b) Glomerulus and kidney arterioles in a 4-wk-old MRL/lpr mouse. Vascular smooth muscle of arterioles and interlobular arteries were positive for CDH15 (arrow), whereas glomeruli, PTCs, and vascular endothelial cells were negative. (c and d) Glomerulus and kidney blood vessels in a 22-wk-old MRL/lpr mouse. Immunopositivity was observed in vascular smooth muscle (arrow), vascular endothelial (arrowhead), and glomerular cells. (e and f) Localization of CDH15 in endothelial cells of glomerular capillary and interlobular artery, as determined by immunoelectron microscopy. CDH15 was expressed on the luminal side in vascular endothelial cells (size of gold particles = 10 nm). (g) CDH15 was detected by Western blotting of pooled lysates of glomeruli isolated from two MRL/+ mice (n = 2) and two age-matched MRL/lpr mice (n = 2). (h)Noa-SMA–positive cells were detected in glomeruli of MRL/+ or MRL/lpr mice by immunohistochemistry. Scale bars, 40 mmin(a)–(d), 200 nm in (e) and (f), and 100 mmin(h). Arrowheads and arrows indicate endothelium and smooth muscle cells, respectively; glomeruli are circled. immune complex deposition and infiltration of cells into the kidney, CDH15 is involved in muscle development and regeneration and which may have suppressed proteinuria. is mainly expressed in vascular smooth muscle cells and satellite The physiological significance of DN T cells is unknown (32, cells (39). In contrast, few studies have reported CDH15 expres- 33). It was reported that such cells constitute ∼20% of kidney sion in vascular endothelial cells (40). Our immunohistochemical mononuclear cells in C57BL/6J wild-type mice (34). It was also analysis revealed that CDH15 was not present on vascular endo- demonstrated that DN T cells in mice with ischemia reperfusion thelial cells in the kidneys of 4-wk-old mice but was detected in injury showed a protective response to early changes (35). Few the vascular endothelium and arterioles as the disease progressed. studies have investigated the role of kidney DN T cells in SLE, Our observation that CDH15 is expressed by vascular endothelial although they have been shown to express IL-17 and contribute to cells of the luminal surface of glomerular capillaries, which has pathophysiology in SLE patients (36–38). MRL/lpr mice are not been previously demonstrated, is consistent with its role as a characterized by infiltration of DN T cells into multiple organs; in this receptor for PEPITEM (7). CDH15 expression on vascular endo- study, we demonstrated that this was abrogated by PEPITEM treat- thelial cells is accelerated by inflammation and is presumed to be a ment (Fig. 1c, 1d). However, we did not examine whether DN T cells defense mechanism for maintaining homeostasis. Our proposed reside in or are recruited to the kidneys. Additional studies are needed scheme of the relation between PEPITEM (ligand) and CDH15 to clarify the role of DN T cells in the pathophysiology of SLE. (receptor) in the vascular endothelium of glomeruli is shown in 8 PEPITEM/CADHERIN 15 AXIS INHIBITS GLOMERULONEPHRITIS IN SLE

FIGURE 7. CDH15 is expressed in the vascular endothelium of skin and peritoneum in 22-wk-old MRL/lpr mice. (a–c) CDH15 expression in the skin of a 22-wk-old MRL/lpr mouse. Double immunofluorescence labeling revealed colocalization (yellow) of the endothelial marker CD31 (green) and CDH15 (red). (d–f) CDH15 expression in the peritoneum of a 22-wk-old MRL/lpr mouse. CD31 (green) and CDH15 (red) were colocalized (yellow). Scale bar,

40 mm. Arrows and arrowheads indicate CDH15 expressed on the endothelium and on a smooth muscle cell, respectively. Downloaded from

Fig. 9. PEPITEM in the blood may modulate S1P secretion currently used for long-term treatment of SLE; however, controlling by endothelial cells throughCDH15expressedonglomerular inflammatory cell infiltration by administration of endogenous pep- capillary and vascular endothelial cells. S1P may then act on tides could be safer and more effective in preventing infection and T lymphocytes expressing S1P receptor and thereby suppress ensuring treatment compliance.

T lymphocyte migration. Nevertheless, this study has limitations. It has been reported that http://www.jimmunol.org/ When the S1P signal is received as a second messenger via S1P the amount of endogenous PEPITEM is decreased in chronic in- receptor, T lymphocytes, including DN T cells, participate in flammation; however, this was not measured in our MRL/lpr mice, various physiological processes, including phenotype determination and the dose-dependent effects of lymphocyte migration should be and recirculation (41). Fingolimod, an S1P receptor modulator, considered in determination of the optimal blood concentration of inhibits lymphocyte migration from capillaries into the mesangial PEPITEM. Because a short half-life is a frequent obstacle in area in MRL/lpr mice (4–6). Similar inhibition of lymphocyte peptide treatment, liposomes were used as a delivery vehicle in this migration and improvement of clinical severity was observed study. Further research on the pharmacokinetics of PEPITEM in experimental autoimmune encephalomyelitis mice, a model delivered by liposome is necessary in the future. In addition, we for multiple sclerosis (14, 15). Thus, strategies that target the did not measure the S1P concentration in MRL/lpr mice. S1P is by guest on September 29, 2021 PEPITEM/CDH15 axis and the control of lymphocyte migration/ produced by various cells (e.g., RBC, platelets, and macrophages); infiltration by S1P may be effective for the treatment of auto- in the kidneys of MRL/lpr mice, S1P expression was confirmed in immune diseases. infiltrating, smooth muscle, and other cells (data not shown). Thus, In the current study, we demonstrated that CDH15 is expressed it would be difficult to examine in vivo changes in S1P from on vascular endothelial cells not only in the kidney, but also in the endothelial cells because of PEPITEM administration. skin and peritoneum of MRL/lpr mice. PEPITEM is expressed to In summary, exogenously administered PEPITEM inhibited suppress inflammation in skin lesions with altered CDH15 ex- T lymphocyte infiltration into multiple organs in MRL/lpr mice, pression and serositis in SLE. Multiple immunosuppressants are thereby contributing to the protection of podocytes and reduction in

FIGURE 8. PEPITEM binds to the renal vascular endothelium. (a and b) Immunohistochemical analysis of 4-wk-old MRL/lpr mouse kidney with FITC- labeled PEPITEM. Vascular smooth muscle of arterioles and interlobular arteries was positive for PEPITEM (arrow). (c) Color development of FITC- labeled scrambled peptide as a negative control. No immunopositivity was observed. (d) Renal cortical area of a 22-wk-old mouse in a low-power field. (e) Enlarged view of (d). Immunopositivity (dark brown) was detected in vascular smooth muscle cells (arrow) and vascular endothelial cells of arterioles (arrowhead). (f) Interlobular artery with vasculitis. Immunopositivity (dark brown) was detected in vascular smooth muscle cells (arrow) and vascular endothelial cells (arrowhead). Scale bars, 40 mm. Arrowheads and arrows indicate endothelium and smooth muscle cells, respectively; glomeruli are circled. The Journal of Immunology 9

FIGURE 9. Proposed mechanism under- lying the blocking of T lymphocyte migra- tion and reduced albuminuria after PEPITEM treatment. Exogenously administered PEPITEM in blood binds to CDH15 (receptor) expressed on vascular endothelial cells, which secrete S1P that binds to T lymphocytes expressing the S1P receptor. Consequently, migration of T lymphocytes out of glomerular blood ves- sels is suppressed. Downloaded from

proteinuria. Moreover, CDH15 expression on vascular endothelial 1 causes tissue retention by inhibiting the entry of peripheral tissue T lymphocytes into afferent lymphatics. Nat. Immunol. 9: 42–53. cells increased with disease progression. These results indicate that 10. Matloubian, M., C. G. Lo, G. Cinamon, M. J. Lesneski, Y. Xu, V. Brinkmann, http://www.jimmunol.org/ the PEPITEM/CDH15 axis is involved in local homeostasis in M. L. Allende, R. L. Proia, and J. G. Cyster. 2004. Lymphocyte egress from inflammatory pathology. Thus, control of inflammation via ex- thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature 427: 355–360. ogenous administration of PEPITEM may be an effective treat- 11. Sanna, M. G., J. Liao, E. Jo, C. Alfonso, M. Y. Ahn, M. S. Peterson, B. Webb, ment for diseases such as SLE. In the future, the mechanisms S. Lefebvre, J. Chun, N. Gray, and H. Rosen. 2004. Sphingosine 1-phosphate underlying S1P changes via the PEPITEM/CDH15 axis should be (S1P) receptor subtypes S1P1 and S1P3, respectively, regulate lymphocyte recirculation and heart rate. J. Biol. Chem. 279: 13839–13848. elucidated. 12. Lai, W. Q., A. W. Irwan, H. H. Goh, H. S. Howe, D. T. Yu, R. Valle-On˜ate, I. B. McInnes, A. J. Melendez, and B. P. Leung. 2008. Anti-inflammatory effects of sphingosine kinase modulation in inflammatory arthritis. J. Immunol. 181: Acknowledgments 8010–8017. We thank Dr. Eishin Yaoita (Kidney Research Center, Niigata University 13. Reinshagen, M. 2015. [Regulation of T-cell trafficking in chronic inflammatory by guest on September 29, 2021 Graduate School of Medical and Dental Sciences, Niigata, Japan) for tech- disease--far more complex!]. Z. Gastroenterol. 53: 1108. 14. Chun, J., and H. P. Hartung. 2010. Mechanism of action of oral fingolimod nical advice on the isolation of glomeruli, Dr. Kazuo Yamakami (National (FTY720) in multiple sclerosis. Clin. Neuropharmacol. 33: 91–101. Defense Medical College, Saitama, Japan) for advice regarding peptide han- 15. Candido, K., H. Soufi, M. Bandyopadhyay, and S. Dasgupta. 2016. Therapeutic dling, and the staff in the department for their assistance. impact of sphingosine 1-phosphate receptor signaling in multiple sclerosis. Mini Rev. Med. Chem. 16: 547–554. 16. Yamakami, K., H. Tsumori, Y. Sakurai, Y. Shimizu, K. Nagatoshi, and Disclosures K. Sonomoto. 2013. Sustainable inhibition efficacy of liposome-encapsulated The PEPITEM used in this study has been licensed for commercial use nisin on insoluble glucan-biofilm synthesis by Streptococcus mutans. Pharm. Biol. 51: 267–270. by the University of Birmingham (WO2013/104928). The authors have 17. Varypataki, E. M., K. van der Maaden, J. Bouwstra, F. Ossendorp, and no financial conflicts of interest. W. Jiskoot. 2015. Cationic liposomes loaded with a synthetic long peptide and poly(I:C): a defined adjuvanted vaccine for induction of antigen-specific T cell cytotoxicity. AAPS J. 17: 216–226. References 18. Shimizu, Y., H. Takagi, T. Nakayama, K. Yamakami, T. Tadakuma, N. Yokoyama, and N. Kojima. 2007. Intraperitoneal immunization with 1. Parikh, S. V., and B. H. Rovin. 2016. Current and emerging therapies for lupus oligomannose-coated liposome-entrapped soluble leishmanial antigen induces nephritis. J. Am. Soc. Nephrol. 27: 2929–2939. antigen-specific T-helper type immune response in BALB/c mice through uptake 2. Rottman, J. B., and C. R. Willis. 2010. Mouse models of systemic lupus by peritoneal macrophages. Parasite Immunol. 29: 229–239. erythematosus reveal a complex pathogenesis. Vet. Pathol. 47: 664–676. 19. Sercombe, L., T. Veerati, F. Moheimani, S. Y. Wu, A. K. Sood, and S. Hua. 2015. 3. Davidson, A. 2016. What is damaging the kidney in lupus nephritis? Nat. Rev. Advances and challenges of liposome assisted drug delivery. Front. Pharmacol. Rheumatol. 12: 143–153. 6: 286. 4. Mike, E. V., and C. Putterman. 2017. Amelioration of neuropsychiatric systemic 20. Martina, M. N., S. Bandapalle, H. Rabb, and A. R. Hamad. 2014. Isolation of lupus erythematosus by fingolimod-mediated sphingosine-1 phosphate receptor double negative ab T cells from the kidney. J. Vis. Exp. 87: 51192. modulation. J. Immunol. 198: 127.25. 21. Takechi, H., T. Oda, O. Hotta, K. Yamamoto, N. Oshima, T. Matsunobu, 5. Okazaki, H., D. Hirata, T. Kamimura, H. Sato, M. Iwamoto, T. Yoshio, A. Shiotani, H. Nagura, H. Shimazaki, S. Tamai, et al. 2013. Clinical and im- J. Masuyama, A. Fujimura, E. Kobayashi, S. Kano, and S. Minota. 2002. Effects munological implications of increase in CD208+ dendritic cells in tonsils of of FTY720 in MRL-lpr/lpr mice: therapeutic potential in systemic lupus eryth- patients with immunoglobulin A nephropathy. Nephrol. Dial. Transplant. 28: ematosus. J. Rheumatol. 29: 707–716. 3004–3013. 6. Wenderfer, S. E., S. M. Stepkowski, and M. C. Braun. 2008. Increased survival 22. Sanden, S. K., J. E. Wiggins, M. Goyal, L. K. Riggs, and R. C. Wiggins. 2003. Evaluation of a thick and thin section method for estimation of podocyte number, and reduced renal injury in MRL/lpr mice treated with a novel sphingosine-1- glomerular volume, and glomerular volume per podocyte in rat kidney with phosphate receptor agonist. Kidney Int. 74: 1319–1326. Wilms’ tumor-1 protein used as a podocyte nuclear marker. J. Am. Soc. Nephrol. 7. Chimen, M., H. M. McGettrick, B. Apta, S. J. Kuravi, C. M. Yates, A. Kennedy, 14: 2484–2493. A. Odedra, M. Alassiri, M. Harrison, A. Martin, et al. 2015. Homeostatic reg- 23. Zhang, S. Y., M. Kamal, K. Dahan, A. Pawlak, V. Ory, D. Desvaux, V. Audard, ulation of T cell trafficking by a B cell-derived peptide is impaired in autoim- M. Candelier, F. BenMohamed, M. Matignon, et al. 2010. c-mip impairs mune and chronic inflammatory disease. Nat. Med. 21: 467–475. podocyte proximal signaling and induces heavy proteinuria. [Published erratum 8. Saba, J. D. 2015. A B cell-dependent mechanism restrains T cell transendothelial appears in 2010 Sci. Signal. 3: er7.] Sci. Signal. 3: ra39. migration. Nat. Med. 21: 424–426. 24. Takemoto, M., N. Asker, H. Gerhardt, A. Lundkvist, B. R. Johansson, Y. Saito, 9. Ledgerwood, L. G., G. Lal, N. Zhang, A. Garin, S. J. Esses, F. Ginhoux, M. Merad, and C. Betsholtz. 2002. A new method for large scale isolation of kidney glo- H. Peche, S. A. Lira, Y. Ding, et al. 2008. The sphingosine 1-phosphate receptor meruli from mice. Am. J. Pathol. 161: 799–805. 10 PEPITEM/CADHERIN 15 AXIS INHIBITS GLOMERULONEPHRITIS IN SLE

25. Kobayashi, R., M. Deavers, R. Patenia, T. Rice-Stitt, J. Halbe, S. Gallardo, and 34. Ascon, D. B., M. Ascon, S. Satpute, S. Lopez-Briones, L. Racusen, R. B. Colvin, R. S. Freedman. 2009. 14-3-3 zeta protein secreted by tumor associated M. J. Soloski, and H. Rabb. 2008. Normal mouse kidneys contain activated and monocytes/macrophages from ascites of epithelial ovarian cancer patients. CD3+CD4- CD8- double-negative T lymphocytes with a distinct TCR repertoire. Cancer Immunol. Immunother. 58: 247–258. J. Leukoc. Biol. 84: 1400–1409. 26. Pang, T. T. L., M. Chimen, E. Goble, N. Dixon, A. Benbow, S. E. Eldershaw, 35. Martina, M. N., S. Noel, A. Saxena, S. Bandapalle, R. Majithia, C. Jie, D. Thompson, S. C. Gough, and P. Narendran. 2013. Inhibition of islet immu- L. J. Arend, M. E. Allaf, H. Rabb, and A. R. Hamad. 2016. Double-negative ab noreactivity by adiponectin is attenuated in human type 1 diabetes. J. Clin. T cells are early responders to AKI and are found in human kidney. J. Am. Soc. Endocrinol. Metab. 98: E418–E428. Nephrol. 27: 1113–1123. 27. Kadowaki, T., T. Yamauchi, N. Kubota, K. Hara, K. Ueki, and K. Tobe. 2006. 36. Crispı´n, J. C., M. Oukka, G. Bayliss, R. A. Cohen, C. A. Van Beek, Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the I. E. Stillman, V. C. Kyttaris, Y. T. Juang, and G. C. Tsokos. 2008. Expanded metabolic syndrome. J. Clin. Invest. 116: 1784–1792. double negative T cells in patients with systemic lupus erythematosus produce 28. Swain, S. L., and R. W. Dutton. 1987. Consequences of the direct interaction of IL-17 and infiltrate the kidneys. J. Immunol. 181: 8761–8766. helper T cells with B cells presenting antigen. Immunol. Rev. 99: 263–280. 37. Crispı´n, J. C., and G. C. Tsokos. 2009. Human TCR-ab+ CD4- CD8- T cells can 29. Mosser, D. M., and J. P. Edwards. 2008. Exploring the full spectrum of mac- derive from CD8+ T cells and display an inflammatory effector phenotype. rophage activation. [Published erratum appears in 2010 Nat. Rev. Immunol. 10: 460.] Nat. Rev. Immunol. 8: 958–969. J. Immunol. 183: 4675–4681. 30. Petersone, L., N. M. Edner, V. Ovcinnikovs, F. Heuts, E. M. Ross, E. Ntavli, 38. Krebs, C. F., T. Schmidt, J.-H. Riedel, and U. Panzer. 2017. T helper type 17 C. J. Wang, and L. S. K. Walker. 2018. T cell/B cell collaboration and auto- cells in immune-mediated glomerular disease. Nat. Rev. Nephrol. 13: 647–659. immunity: an intimate relationship. Front. Immunol. 9: 1941. 39. Irintchev, A., M. Zeschnigk, A. Starzinski-Powitz, and A. Wernig. 1994. Ex- 31. Linterman, M. A., R. J. Rigby, R. K. Wong, D. Yu, R. Brink, J. L. Cannons, pression pattern of M-cadherin in normal, denervated, and regenerating mouse P. L. Schwartzberg, M. C. Cook, G. D. Walters, and C. G. Vinuesa. 2009. Fol- muscles. Dev. Dyn. 199: 326–337. licular helper T cells are required for systemic autoimmunity. J. Exp. Med. 206: 40. Cifuentes-Diaz, C., M. Nicolet, H. Alameddine, D. Goudou, M. Dehaupas, 561–576. F. Rieger, and R. M. Me`ge. 1995. M-cadherin localization in developing adult 32. Martina, M. N., S. Noel, A. Saxena, H. Rabb, and A. R. Hamad. 2015. Double and regenerating mouse skeletal muscle: possible involvement in secondary negative (DN) ab T cells: misperception and overdue recognition. Immunol. Cell myogenesis. Mech. Dev. 50: 85–97. Biol. 93: 305–310. 41. Cinamon, G., M. A. Zachariah, O. M. Lam, F. W. Foss, Jr., and J. G. Cyster. Downloaded from 33. Turner, J. E., M. Becker, H. W. Mittru¨cker, and U. Panzer. 2018. Tissue- 2008. Follicular shuttling of marginal zone B cells facilitates antigen transport. resident lymphocytes in the kidney. J. Am. Soc. Nephrol. 29: 389–399. Nat. Immunol. 9: 54–62. http://www.jimmunol.org/ by guest on September 29, 2021