Interstitium of Obstructive Nephropathy Anti-Inflammatory

Genetically Modified Bone Marrow-Derived Vehicle Cells Site Specifically Deliver an Anti-Inflammatory Cytokine to Inflamed Interstitium of Obstructive Nephropathy This information is current as of September 27, 2021. Hiroko Yamagishi, Takashi Yokoo, Toshiyuki Imasawa, Tetsuya Mitarai, Tetsuya Kawamura and Yasunori Utsunomiya J Immunol 2001; 166:609-616; ; doi: 10.4049/jimmunol.166.1.609 Downloaded from http://www.jimmunol.org/content/166/1/609

References This article cites 35 articles, 7 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/166/1/609.full#ref-list-1

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

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 © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Genetically Modified Bone Marrow-Derived Vehicle Cells Site Specifically Deliver an Anti-Inflammatory Cytokine to Inflamed Interstitium of Obstructive Nephropathy1

Hiroko Yamagishi,* Takashi Yokoo,*† Toshiyuki Imasawa,* Tetsuya Mitarai,‡ Tetsuya Kawamura,* and Yasunori Utsunomiya2*

In this study, we used genetically modiﬁed bone marrow-derived CD11b؉CD18؉ vehicle cells to deliver IL-1 receptor antagonist (IL-1ra) for treatment of inﬂamed renal interstitium in an animal model of unilateral ureteral obstruction (UUO). Vehicle cells that expressed the ICAM-1 ligands, CD11b and CD18, were obtained from bone marrow cells of DBA/2j mice and adenovirally transduced with the IL-1ra gene or glucocerebrosidase (GC) gene ex vivo. In kidneys treated to develop UUO, levels of ICAM-1,

IL-1␤, and IL-1R expression increased within 3 days compared with contralateral untreated kidneys in the same mice. Similarly, Downloaded from the macrophage inﬁltration in the cortical interstitium increased after 3 days in UUO kidneys, but not untreated kidneys. After UUO developed, DBA/2j mice were injected i.v. with either IL-1ra؉ vehicle cells (IL-1ra-treated mice) or GC؉ vehicle cells GC-treated mice) at 24 h after UUO. Six days after the injection of these vehicle cells, marked increase of CD11b؉ IL-1ra؉ vehicle) cells was observed in the ICAM-1-positive interstitium of UUO kidneys from IL-1ra-treated mice. In contrast, no CD11b؉ IL-1ra؉ cells appeared in ICAM-1-negative contralateral kidneys from these mice. Furthermore, the inﬁltration of macrophages (p <

/in the interstitium of UUO kidneys http://www.jimmunol.org (0.005 ؍ expression of ICAM-1 (p < 0.005), and presence of ␣-smooth muscle actin (p ,(0.001 were significantly decreased in IL-1ra-treated mice compared with GC-treated mice. These findings suggest that IL-1 may contribute to the development of renal interstitial injury and that our method can deliver a functioning gene encoding an antiin- flammatory cytokine gene specifically at that site by interacting with local adhesion molecules. The Journal of Immunology, 2001, 166: 609–616.

ubulointerstitial injury is the final common pathway for As a modulator of IL-1 activity, the IL-1 receptor antagonist progressive renal disease of several types. Although cy- (IL-1ra)3 can suppress IL-1 activity, as evident by the ability of tokines, infiltrating cells, and adhesion molecules may all this receptor to block experimental glomerulonephritis (10, 12, T by guest on September 27, 2021 be involved in the pathogenesis of this interstitial injury (1–4), the 13). Based on this concept, we previously established a novel sys- underlying mechanisms are not fully understood. Moreover, no tem for using bone marrow-derived cells as vehicles for site-spe- effective therapy currently exists for the related disease, interstitial cific delivery of an IL-1ra gene into inflamed glomeruli. This pro- fibrosis. cedure, which suppressed local IL-1 action (14, 15), successfully IL-1 is an important proinflammatory cytokine with a wide prevented the progression of glomerular injury evoked by Ab to range of effects, including activation of endothelial cells, stimula- the glomerular basement membrane (GBM). tion of tissue infiltration by neutrophils and macrophages, and in- With that background, we initiated this two-part study. For the duction of other mediators of inflammation such as TNF-␣, IL-8, first part, we examined the time course of macrophage infiltration ICAM-1, and NO (5, 6). As an example of its potential for damage, as well as ICAM-1, IL-1 mRNA, and IL-1R expression in mice a pathological role has been identified for IL-1 in experimental and treated to develop a unilateral ureteral obstruction (UUO) in the human glomerulonephritis (7–10). Glomerular as well as tubular cortical interstitium. This model is a well-established archetype of epithelial cells may synthesize and release IL-1 (8, 10, 11), yet no renal interstitial injury (2–4). For the second study, we use genet- proof exists to implicate IL-1 as a cause of tubulointerstitial injury. ically modified bone marrow-derived CD11bϩ CD18ϩ vehicle cells to deliver IL-1ra to inflamed interstitium of UUO kidneys as a new therapeutic approach for controlling tubulointerstitial injury.

*Department of Internal Medicine, Division of Nephrology and Hypertension, and †Department of Gene Therapy, Institute of DNA Medicine, Jikei University School of Materials and Methods Medicine, Tokyo, Japan; and ‡Department of Internal Medicine, Saitama Medical Animals School, Kawagoe, Saitama, Japan Thirty DBA/2j female mice were purchased from Nippon Crea (Tokyo, Received for publication June 9, 2000. Accepted for publication October 6, 2000. Japan). All animals used in this study were maintained in our animal fa- The costs of publication of this article were defrayed in part by the payment of page cility on standard laboratory chow. charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a grant from the Study Group on IgA Nephropathy (to H.Y.) and a grant from the Ministry of Education (Japan) (to Y.U.). 2 Address correspondence and reprint requests to Dr. Yasunori Utsunomiya, Depart- 3 Abbreviations used in this paper: IL-1ra, IL-1 receptor antagonist; GBM, glomerular ment of Internal Medicine, Division of Nephrology and Hypertension, Jikei Univer- basement membrane; GC, glucocerebrosidase; hpf, high power ﬁeld; IL-1Rt1, IL-1R sity School of Medicine, 3-25-8 Nishi-shinbashi, Minato-ku, Tokyo, 105-8461 Japan. type I; ␣-SMA, ␣-smooth muscle actin; UUO, unilateral ureteral obstruction.

FIGURE 1. Expression of ICAM-1 in renal cortexes 3 days after treatment to cause to UUO. Original magniﬁcation, ϫ400. ICAM-1 was clearly observed in cortical tubular epithelial cells, interstitia, and vessels in UUO kidneys (A), whereas contralateral untreated kidneys had no obvious ICAM-1 staining (B). Downloaded from http://www.jimmunol.org/

Unilateral ureteral obstruction ␮g/ml amphotericin B. Cells were seeded onto unprocessed 10-cm dishes at a concentration of 1 ϫ 107 cells/dish and cultured in a humidified atmosphere of At 8 wk of age, 20 mice were anesthetized by the i.p. injection of pento- 5% CO2 for 1 wk. These vehicle cells were verified as expressing CD11b and barbital, and their right ureters were ligated and cut down as described (16) CD18, both of which are ligand of ICAM-1 by FACS (14). to cause UUO. Five of these mice with UUO kidneys were sacrificed for histological examination and RT-PCR analyses at posttreatment days 3, 5, 7, and 14. Recombinant adenovirus preparation and in vivo injection of IL-1ra Establishment of vehicle cells Replication-defective recombinant adenoviruses carrying IL-1ra by guest on September 27, 2021 Bone marrow-derived CD11bϩ and CD18ϩ vehicle cells were established (AxCAmIL-1RA) were purchased from Riken DNA Bank (Ibaraki, Japan), as previously described (14). Briefly, bone marrow cells were harvested and adenoviruses carrying glucocerebrosidase (GC) cDNA (AxACGC) from the femur and tibia of the 7- to 8-wk-old DBA/2j mice and suspended (17) were kindly provided by Dr. T. Ohashi (Department of Gene Therapy, in DMEM (Life Technologies, Grand Island, NY) supplemented with 10% Jikei University School of Medicine, Tokyo, Japan). Both were under the heat-inactivated FBS, 20% heat-inactivated horse serum, 20% L-929-con- control of a CMV enhancer chicken ␤-actin hybrid promoter (18). Recom- ditioned medium, 100 U/ml penicillin G, 100 ␮g/ml streptomycin, and 0.25 binant viruses were propagated and isolated from 293 host cells. Bone

FIGURE 2. Immunohistochemical detection of F4/80-positive cells in the interstitia of UUO kidneys at day 3. Original magniﬁcation, ϫ400. F4/80- positive cells were recruited into the interstitia of a UUO kidney (A). In contrast, few F4/80-positive cells were found in the periglomerular space or interstitia of an unobstructed kidney from the same mouse (B). The Journal of Immunology 611

FIGURE 3. Time course of interstitial ICAM-1 expression and F4/80-positive cell inﬁltration in UUO-treated mice. In UUO kidneys (left), interstitial ICAM-1 expression peaked at day 5. The number of F4/80-positive macrophages in the interstitium of UUO kidneys increased by day 3 and peaked at day 7. However, these histological changes were not found in contralateral kidneys (right). Symbols are: (F, E), ICAM-1 expression; (f, Ⅺ), F4/80-positive cells.

marrow-derived vehicle cells, as described above, were cultivated with For indirect immunofluorescence, the sections were stained to identify Downloaded from 10% FBS, 20% horse serum, and 20% L-929-conditioned medium for 6 IL-1raϩ cells, IL-1R, and ICAM-1 expression in the tubulointerstitium. To days, and then AxCAmIL-1RA or AxCAGC was infected at a multiplicity prevent nonspecific binding of avidin and biotin, an avidin/biotin-blocking of infection ϭ 200. kit (Vector Laboratories) was used, after which sections were incubated 6 Mice with UUO were injected through the tail vein with 5 ϫ 10 of with either rat anti-murine IL-1ra Ab (dilution 1/100; Genzyme-Techne, ϭ either IL-1ra-infected vehicle cells (IL-1ra-treated group, n 5), or GC- Cambridge, MA), rat anti-murine ICAM-1 mAb (YN1/1.7.4), or rabbit ϭ infected vehicle cells (GC-treated group as a control, n 5) at 24 h after anti-mouse IL-1R type 1 (IL-1Rt1) Ab (dilution 1/50; Santa Cruz Biotech- UUO treatment. Six days after the injection of these vehicle cells, sera from nology, Santa Cruz, CA) overnight at 4°C, followed by rinsing in PBS. http://www.jimmunol.org/ individual mice were collected, and all mice were sacrificed to obtain kid- Incubation for 60 min at 37°C followed with biotinylated mouse anti-rat ney tissues for histological examination. IgG (␬)-chain mAb (MARK-1) for IL-1ra and ICAM-1 staining or biotin- Immunohistochemistry ylated sheep anti-rabbit IgG Ab (American Quolex) for IL-1Rt1 staining. After rinsing in PBS, the sections were incubated with either FITC-labeled The detection of infiltrating F4/80-positive macrophages, infused Mac-1ϩ avidin (dilution 1/50; Becton Dickinson, San Jose, CA) for ICAM-1 and (CD11b) vehicle cells, and ICAM-1 or ␣-smooth muscle actin (␣-SMA) IL-1Rt1 staining, or rhodamine-labeled streptavidin (dilution 1/50; Amer- expression in the cortical interstitium relied on immunohistochemistry ican Quolex) for IL-1ra staining for 60 min at 37°C. Negative controls were based on the avidin-biotin-peroxidase method, as described (19). Kidney performed by replacing the first-step Ab by incubation buffer only or by specimens were embedded in OTC compound (Miles Scientific, Naper- isotype-matched Abs. The sections were again rinsed in PBS, then Ϫ ville, IL) and quickly frozen in dry ice and acetone at 70°C. Cryostat mounted in p-phenylenediamine (Sigma, St. Louis, MO)-PBS-glycerine, ␮ by guest on September 27, 2021 sections (3 m) were rinsed in PBS (10 mmol/L sodium phosphate, pH 7.2, and observed under a Zeiss Axiophot fluorescence photomicroscope. 0.9% saline) three times for 15 min each. First, the endogenous biotins in the sections were blocked as directed in an avidin/biotin-blocking kit (Vec- tor Laboratories, Burlingame, CA). Next, the sections were incubated with the primary Abs overnight at 4°C. Primary Abs used in this study were rat Morphometric analysis anti-mouse F4/80 mAb for macrophage staining (20murine ICAM-1 mAb Interstitial F4/80-positive cells. The number of F4/80-positive cells in the (21) (YN1/1.7.4; kindly provided by Toray Medical Company, Tokyo, Ja- cortical interstitium was counted using an eyepiece graticule. Twenty mi- pan), rat anti-murine Mac-1 mAb (M1-70.15.11.5.HL; American Type ␣ croscopic fields were counted, and the average number of F4/80-positive Culture Collection, Manassas, VA), or mouse anti- -SMA mAb (dilution 2 1:50; Dako, Glostrup, Denmark). Then the sections were incubated with cells within a 1-mm cortical area was calculated. Only cells with clearly either biotinylated mouse anti-rat IgG (␬)-chain mAb (MARK-1; Zymed identifiable nuclei were counted. Laboratories, San Francisco, CA) for F4/80, ICAM-1 and Mac-1 staining Interstitial IL-1ra-positive cells. The number of IL-1ra-positive cells in or biotinylated rabbit anti-mouse IgG (␬) Ab (American Quolex, La the cortical interstitium was counted under a high power field (hpf) Mirada, CA) for ␣-SMA staining. After a 60-min incubation at room tem- (ϫ1000) fluorescence microscope. The value shown for each group rep- perature, incubation proceeded with avidin-biotin-peroxidase complex resents an average number of IL-1ra-positive cells obtained from a series (Vector ABC Elite staining kit; Vector Laboratories, Burlingame, CA). The of randomly selected fields (more than 10) in each section. peroxidase was developed with a diaminobenzidine substrate solution (Per- Interstitial ICAM-1 expression. The degree of ICAM-1 expression in the oxidase Substrate kit; Vector Laboratories). Both kits were used according to the manufacturer’s instructions. Finally, the sections were counter- cortical interstitium was determined as a percentage of cortical ICAM-1- stained with methyl green. positive interstitial area over the total cortical interstitial area among more than 20 fields in each section of the kidney. Each field was scored from 0 to 4: 0 ϭ no changes; 1 ϭ changes affecting Ͻ25% of total interstitial area; 2 ϭ changes affecting 25% to 50% of the total interstitial area; 3 ϭ changes affecting 50% to 75% of the total interstitial area; 4 ϭ changes affecting Ͼ75% of the total interstitial area. A mean value was calculated to repre- sent the degree of these changes for each section. Interstitial ␣-SMA expression. Interstitial immunostaining for ␣-SMA was quantified by the pointing counting method (22) using an eyepiece graticule of 1 cm2 with 10 equidistant lines at a final magnification of ϫ400. The results were expressed as the percentage of ␣-SMA-positive area in the cortical interstitium calculated according to the following formula and av- FIGURE 4. RT-PCR analysis of IL-1␤ mRNA in renal cortexes from eraged among more than 20 fields in each section. The percentage of mice at day 3 after UUO treatment. Levels of IL-1␤ mRNA in UUO kid- ␣-SMA-positive area ϭ (number of grid intersections with ␣-SMA-posi- neys (U) from five individual mice were markedly higher than in their tive staining in the interstitium)/(total number of grid intersections in the contralateral untreated kidneys (C). M, DNA markers. Intensity ratio of interstitium) ϫ 100. Interstitium was defined as that portion of the cortex IL-1␤ mRNA/GAPDH: 1366 Ϯ 460 in UUO kidneys vs 318 Ϯ 302 in excluding glomeruli, tubules, arteries, arterioles, and veins. Fields of me- contralateral untreated kidneys, p Ͻ 0.005. dulla were not included in this study. 612 SITE-SPECIFIC GENE DELIVERY INTO INFLAMED INTERSTITIUM

FIGURE 5. Expression of IL-1Rt1 receptor in the renal cortex 3 days after UUO treatment. Original magnification, ϫ1000. IL-1Rt1 expression was up-regulated in both the apical and basolateral membranes of tubules from UUO kidneys (A), whereas IL- 1Rt1 receptor was weakly expressed at the apical membranes of tubules and vascular endothelial cells in contralateral untreated kidneys (B). Downloaded from http://www.jimmunol.org/ RT-PCR and preparation of the mouse IL-1 cDNA probe neys, whereas contralateral untreated kidneys contained no appar- The renal cortex was isolated from individual mice in all groups stored at ent ICAM-1 (Fig. 1, A and B). In addition, an abundance of F4/ 4°C and used for mRNA extraction with a QuickPrep Micro mRNA Pu- 80-positive macrophages was present in the interstitium of UUO rification Kit (Amersham Pharmacia Biotech AB, Uppsala, Sweden). kidneys (Fig. 2). As shown in Fig. 3, interstitial expression of First-stranded cDNA was synthesized using avian myeloblastosis virus ICAM-1 in UUO kidneys peaked at day 5, and ICAM-1 expression reverse transcriptase (Roche, Germany). Five microliters of each cDNA significantly increased throughout the observation period (mean were amplified to a final volume of 20 ␮l PCR mixture containing 10 mM ␮ score: day 3, 0.5 Ϯ 0.1; day 5, 1.8 Ϯ 0.2; day 7, 1.9 Ϯ 0.4; day 14, Tris-HCl, 50 mM KCl, 1.5 mM MgCl2, 200 M each of dNTP, 10 pmol of oligonucleotide primers for either IL-1␤ or GAPDH, and 0.5 U of Taq 1.9 Ϯ 0.4, p Ͻ 0.0001). The number of F4/80-positive macro-

DNA polymerase (TaKaRa Shuzo, Kyoto, Japan). PCR was conducted for phages in the interstitium of UUO kidneys clearly correlated with by guest on September 27, 2021 29 cycles for IL-1␤ or 23 cycles for GAPDH with a thermal cycler (PC- ICAM-1 expression and significantly increased with time starting 701; ASTEC, Fukuoka, Japan), as follows: 94°C for 30 s, 58°C for 30 s, ␤ at 3 days, peaking at 7 days, and remaining stable until 14 days and 72°C for 30 s. The sequence of primers is as follows: IL-1 : (sense) 2 5Ј-TTGAAGAAGAGCCCATCCTC-3Ј, (antisense) 5Ј-GAGGTGCTG after UUO developed (day 3, 59.3 Ϯ 13/mm ; day 5, 162.4 Ϯ GATGTACCAGTT-3Ј, and GAPDH: (sense) 5Ј-AAGGTCATCCATGA 16.5/mm2; day 7, 244.9 Ϯ 17.4/mm2; day 14, 236.5 Ϯ 25/mm2, CAACTT-3Ј, (antisense) 5Ј-CAGTGTAGCCCAGGATGCC-3Ј, respec- p Ͻ 0.0001 in Fig. 3). No such changes were found in the con- ␤ tively. The predicted sizes of PCR products for IL-1 and GAPDH are 411 tralateral untreated kidneys throughout the observation period. and 348 bp, respectively. For semiquantitative analysis, each PCR product was visualized with Thus, ICAM-1 expression may recruit macrophages into the inter- ethidium bromide after agarose gel electrophoresis. The intensity of fluo- stitium of UUO kidneys. rescence for each PCR product was calculated with National Institutes of Health image 1.58 software, and normalized by that of GAPDH mRNA. Up-regulation of IL-1␤ and IL-1 receptor expression in DNA markers (100-bp DNA ladder) were purchased from New England UUO kidneys Biolabs (Beverly, MA). Next we used RT-PCR to examine IL-1␤ expression in the renal Analysis of serum creatinine cortexes of mice with UUO at posttreatment day 3 (Fig. 4). UUO Concentrations of serum creatinine were measured using the VISION anal- kidneys showed a significant 4-fold increase of IL-1␤ mRNA com- ysis kit (Abbott Laboratories, North Chicago, IL), which is based on the pared with contralateral untreated kidneys. Immunofluorescence Jaffe reaction. revealed IL-1Rt1 receptor weakly expressed at the apical mem- Statistical analysis branes of tubules and vascular endothelial cells in untreated kidneys, whereas IL-1R expression was up-regulated in both the lu- Ϯ Results are expressed as means SD. Statistical analysis was performed minal and basolateral membranes of tubules from UUO kidneys using the two-sample t test to compare data in different groups and repeated ␤ measure ANOVA to compare scores of ICAM-1 expression and numbers (Fig. 5). This heightened expression of IL-1R and also of IL-1 of F4/80-positive macrophages. A p value of less than 0.05 was considered mRNA endured in UUO kidneys up to day 14. statistically significant. CD11bϩ vehicle cells deliver IL-1ra into inflamed interstitia of Results UUO kidneys Macrophage infiltration into kidneys of ICAM-1-expressing mice We previously established bone marrow-derived vehicle cells that with UUO express the ICAM-1 ligands, CD11b and CD18, and adenovirally We first investigated ICAM-1 expression and the infiltration of transduced these cells with a gene encoding IL-1ra (IL-1raϩ ve- macrophages into kidney sections of test mice. At day 3 after hicle cells) (14, 15). Western blot analysis confirmed that these UUO, ICAM-1 was clearly observed by immunostaining of corti- vehicle cells synthesized and secreted IL-1ra protein (15). To ex- cal tubular epithelial cells, interstitia, and vessels of treated kid- amine the function of these cells in the UUO system, CD11bϩ The Journal of Immunology 613

ently, infused CD11bϩ IL-1raϩ vehicle cells site speciﬁcally delivered IL-1ra into the inﬂamed interstitium by interaction with ICAM-1.

Attenuation of interstitial inflammation and fibrosis by IL-1raϩ vehicle cells To investigate the physiological effects of IL-1raϩ vehicle cells, we next measured serum creatinine levels and assessed renal his- tology. Six days after the injection of either IL-1raϩ vehicle cells or GCϩ vehicle cells, serum creatinine levels of both recipient groups were similar: IL-1ra-treated mice, 0.48 Ϯ 0.21 mg/dl vs GC-treated mice, 0.58 Ϯ 0.22 mg/dl ( p Ͼ 0.05) (data not illus- trated). After histological analysis (Fig. 7A), interstitial ICAM-1 staining of UUO kidneys was significantly lower in IL-1ra-treated mice than in GC-treated mice (mean scores, respectively: 1.5 Ϯ 0.3 vs 2.1 Ϯ 0.1, p Ͻ 0.005). To the contrary, the interstitium of contralateral kidneys from IL-1ra-treated mice and GC-treated mice had only faint immunostaining for ICAM-1 (mean score:

0.2 Ϯ 0.1 and 0.3 Ϯ 0.1, respectively). In addition, IL-1ra-treated Downloaded from mice had a markedly decreased number of macrophages in the interstitium of their UUO kidneys compared with GC-treated mice (191.1 Ϯ 17.9/mm2 vs 343.4 Ϯ 8.13/mm2, p Ͻ 0.001 in Fig. 7B). However, similar numbers of interstitial macrophages were observed in contralateral kidneys of IL-1ra- and GC-treated mice 2 2

(41.5 Ϯ 11.5/mm vs 25.5 Ϯ 8.70/mm , respectively). http://www.jimmunol.org/ Because ␣-SMA may be an important factor in the interstitial fibrosis (23, 24), we next injected IL-1raϩ vehicle cells to assess the effect on interstitial expression of ␣-SMA in UUO kidneys. As Fig. 8A illustrates, intense and widespread ␣-SMA immunoreac- tivity developed in the interstitium of UUO kidneys from GC- treated mice. However, IL-1ra-treated mice had significantly less interstitial ␣-SMA expression in UUO kidneys (Fig. 8B). Fig. 9 shows that the percentage of ␣-SMA-positive area in total interstitial area of each UUO kidney was significantly decreased in by guest on September 27, 2021 IL-1ra-treated mice compared with GC-treated mice (27.4 Ϯ 2.7 and 36.3 Ϯ 4.4, respectively, p ϭ 0.005). In contrast, no interstitial ␣-SMA was detectable in the untreated kidneys of either IL-1ra- treated mice or GC-treated mice (percentage of ␣-SMA-positive area in total interstitial area: 1.7 Ϯ 1.1 and 2 Ϯ 0.6, respectively). IL-1raϩ vehicle cells may then suppress both interstitial ICAM-1 expression and the macrophage infiltration, thereby inhibiting ␣-SMA expression in UUO kidneys. FIGURE 6. Recruitment of IL-1raϩ vehicle cells into the interstitium of UUO kidneys 6 days after the injection of vehicle cells. A, IL-1raϩ cells Discussion (arrows). Rhodamin staining: original magnification, ϫ1000. B, Marked ϩ In this study, we noted distinct increases of ICAM-1 molecule increase of IL-1ra cells was observed in the interstitium of UUO kidneys expression in the interstitium, tubular epithelial cells, and vessels from IL-1ra-treated mice (IL-1ra transfectant). In contrast, few IL-1raϩ of the renal cortex after the induction of UUO in mice (Fig. 1). vehicle cells appeared in the interstitium of UUO kidneys from GC-treated p Ͻ 0.0001 vs contralateral. Interstitial ICAM-1 was expressed within 3 days after treatment to ,ء .(mice (GC transfectant induce UUO, with a peak response by the fifth day that stayed high up to the fourteenth day. Subsequently, macrophages were re- IL-1raϩ vehicle cells were injected i.v. into mice with UUO. Six cruited into the interstitium in these UUO kidneys (Figs. 2 and 3). days later, CD11bϩ cells were clearly visible in the interstitium of In addition, interstitial expression of ICAM-1 was most intense in UUO kidneys expressing ICAM-1, but not ICAM-1-negative un- areas of marked macrophage infiltration. Numerous studies have treated kidneys (data not shown). Immunohistochemical study indicated that ICAM-1 plays a major role in immune injury to the (Fig. 6, A and B) revealed significant numbers of vehicle cells kidney (25, 26). In particular, Ricardo et al. (27) demonstrated that producing IL-1ra within the interstitium of UUO kidneys from ICAM-1 on the renal cortex is important for recruiting macrophage IL-1ra-treated mice, whereas few IL-1raϩ vehicle cells occupied in into the interstitium of obstructed kidneys, supporting our obser- the interstitium of UUO kidneys from GC-treated mice (the num- vations. Furthermore, Cheng et al. (28) reported that treatment ber of IL-1raϩ vehicle cells per hpf: 7.2 Ϯ 0.8 vs 0.8 Ϯ 0.4, with ICAM-1 antisense oligonucleotides alleviated the infiltration respectively, p Ͻ 0.0005). Parts of cortical tubules in UUO kid- of inflammatory cells and accumulation of the extracellular matrix neys were also stained with IL-1ra. In contrast, IL-1raϩ vehicle in the tubulointerstitium of mice with renal UUO. cells were not observed in contralateral kidneys from either IL- The stimulus that induces ICAM-1 to appear tubular epithelial 1ra-treated mice or GC-treated mice (the number of IL-1raϩ ve- cells and interstitial cells in the renal cortex after ureteral ligation hicle cells per hpf: 0.6 Ϯ 0.4 vs 0.4 Ϯ 0.2, respectively). Appar- remains unclear. However, proinflammatory cytokines such as 614 SITE-SPECIFIC GENE DELIVERY INTO INFLAMED INTERSTITIUM

FIGURE 7. Effects of IL-1ra gene delivery on interstitial ICAM-1 expression and macrophage infiltration in renal cortexes. Six days after the i.v. injection of either GCϩ vehicle cells (GC-treated mice; GC transfectant) or IL-1raϩ vehicle cells (IL-1ra-treated mice; IL-1ra transfectant), the expression of ICAM-1 (A) and macrophage infiltration (B) in the interstitium of UUO kidneys decreased significantly in IL-1ra- ,ء .treated mice compared with GC-treated mice p Ͻ 0.0001 vs contralateral.

IL-1, TNF-␣, and IFN-␥ are considered inducers of ICAM-1 ex- ICAM-1, they promote the cells’ migration to sites of ICAM-1 pression on glomerular endothelial cells, mesangial cells, and renal expression (14). Using this system, we delivered IL-1ra into tubular epithelial cells (29, 30). Recently, we and others found that ICAM-1-expressing inflamed glomeruli in rabbits with anti-GBM blocking the action of IL-1 by the administration of IL-1ra pre- glomerulonephritis and blocked the progression of glomerular in- Downloaded from vented renal injury in a rat model of anti-GBM glomerulonephritis jury by local suppression of IL-1 action (15). In our preliminary (7–12, 15). Thus, our attention focused on the role of IL-1 in the examinations, infused vehicle cells were recruited into the inter- development of tubulointerstitial injury after the onset of stitium of UUO kidneys at 3 days and remained stable until 14 renal UUO. days after UUO developed. Next, we used these vehicle cells ex- Our former PCR analysis showed increased levels of IL-1␤ pressing CD11b and CD18 to deliver IL-1ra into inflamed renal mRNA in kidneys with induced UUO, suggesting that IL-1 may be interstitium through site-specific interaction with ICAM-1 in an http://www.jimmunol.org/ synthesized in the renal cortex. Although we could not identify the attempt to attenuate interstitial injury in mice induced to develop origin of IL-1␤ in acetone-fixed tissues from UUO kidneys, tubu- UUO. Although interstitial ICAM-1 was expressed by the third lar IL-1␤ expression in glomerulonephritic humans and animals day after treatment to incite UUO in our study, others demon- has been demonstrated with PLP-fixed tissues or in situ hybrid- strated increased interstitial ICAM-1 expression in the renal cortex ization (8, 10, 13). In the present study, the expression of IL-1Rt1 by 12 h after the inception of UUO (28, 32). Therefore, our mice receptor in the tubules was up-regulated in the UUO kidneys. were injected i.v. with syngeneic bone marrow-derived vehicle These data suggested that IL-1 may enhance ICAM-1 expression cells at 24 h after UUO induction. Immunohistochemical studies on tubular epithelial cells and interstitial cells via both autocrine later confirmed that infused CD11bϩ cells producing IL-1ra were and paracrine modes. recruited into the cortical interstitium of UUO kidneys. Moreover, by guest on September 27, 2021 Currently, no clinical therapy effectively halts the interstitial fi- CD11bϩ cell recruitment correlated with ICAM-1 expression. In brosis of progressive renal diseases. However, local delivery of contrast, parts of cortical tubules in UUO kidneys were stained anti-inflammatory cytokines using gene transfer may provide such with IL-1ra, suggesting that intracellular form of IL-1ra might be a tool. To adapt that technology, we established a novel ex vivo expressed in tubular epithelial cells as in keratinocytes and intes- gene delivery system that uses bone marrow-derived vehicle cells tinal epithelial cells (33). Although IL-1ra is produced by activated bearing CD11b and CD18; because these molecules are ligands of neutrophils and monocytes (31), few cells producing IL-1ra were

FIGURE 8. Expression of ␣-SMA in UUO kidneys from GC-treated mice and IL- 1ra-treated mice. Original magniﬁcation, ϫ400. Six days after the i.v. injection of either GCϩ vehicle cells (GC-treated mice) or IL-1raϩ vehicle cells (IL-1ra-treated mice), the kidney sections were examined for the cortical ␣-SMA expression by immunohistochemistry, as described in Materials and Meth- ods. A, UUO kidneys from GC-treated mice; B, UUO kidneys from IL-1ra-treated mice. The Journal of Immunology 615

Systemic administration of IL-1ra has been used as a therapeutic agent for several inflammatory disorders, including glomerulonephritis (10–12). However, the continuous infusion of high-dose IL-1ra required to ubiquitously block IL-1 action affects normal body functions. Consequently, the local suppression of IL-1 action at specific sites is a more effective mode of therapy for humans. Additionally, our system has an advantage over previous gene delivery methods (28, 36) in the site-specific delivery of a functioning gene that encodes anti-inflammatory cytokine directly into inflamed interstitium through the interaction of adhesion molecules. In conclusion, this study suggests that IL-1 may contribute to the development of tubulointerstitial injury, and that the delivery of IL-1ra gene by CD11bϩ vehicle cells can attenuate interstitial inflammation and fibrosis in obstructive nephropathy. Our novel gene delivery system offers a promising strategy for the treatment of some progressive kidney diseases.

FIGURE 9. Effects of IL-1ra gene delivery on interstitial ␣-SMA ex- Acknowledgments

pression in the cortex of UUO kidneys. Six days after the i.v. injection of Downloaded from ϩ ϩ We thank Drs. I. Saito and Y. Kanegae (Institute of Medical Science, either GC vehicle cells (GC-treated mice; GC transfectant) or IL-1ra University of Tokyo, Tokyo, Japan) for providing us the adenoviral cas- vehicle cells (IL-1ra-treated mice; IL-1ra transfectant), the extent of inter- sette cosmid, and K. Fujita (Department of Pediatrics, Showa University), stitial ␣-SMA expression in UUO kidneys decreased significantly in IL- H. Hasegawa, N. Tsuboi, and T. Ohashi for valuable instructions. We also p Ͻ 0.0001 vs ,ء .1ra-treated mice compared with that in GC-treated mice thank Drs. M. Iwashima (Institute of Molecular Medicine and Genes, Med- contralateral. ical College of Georgia), T. Hosoya, and P. Minick for critical editing of the manuscript, and T. Murata for histological analyses. http://www.jimmunol.org/ recruited into the same animals, contralateral untreated kidneys, References 1. Kuncio, G. S., E. G. Neilson, and T. Haverty. 1991. Mechanisms of tubulointer- and the kidneys from GC-treated mice. These data indicated that stitial fibrosis. Kidney Int. 39:550. ϩ CD11b vehicle cells can site specifically deliver IL-1ra into in- 2. Klahr, S., and J. J. Morrissey. 1998. The role of growth factors, cytokines and flamed interstitium through the interaction of adhesion molecules. vasoactive compounds in obstructive nephropathy. Semin. Nephrol. 18:622. 3. Diamond, J. R., S. D. Ricardo, and S. Klahr. 1998. Mechanisms of interstitial In addition, IL-1ra secreted by vehicle cells appeared to suppress fibrosis in obstructive nephropathy. Semin. Nephrol. 18:594. IL-1 action and subsequently decrease tubulointerstitial inflamma- 4. Klahr, S. 1998. Obstructive nephropathy. Kidney Int. 54:286. tion by inhibiting ICAM-1 expression and macrophage infiltration 5. Dinarello, C. A. 1991. Interleukin-1 and interleukin-1 antagonism. Blood 77: 1627. into UUO kidneys. In fact, degrees of tubular expression of IL- 6. Atkins, R. C., D. J. Nikolic-Paterson, and H. Y. Lan. 1996. Tubulointerstitial by guest on September 27, 2021 1Rt1 receptor did not differ between IL-1ra-treated group and GC- injury in glomerulonephritis. Nephrology (Suppl. 1):S2. 7. Boswell, J. M., M. A. Yui, D. W. Burt, and V. E. Kelley. 1988. Increased tumor treated group, and supernatants from IL-1ra-infected vehicle cells necrosis factor and IL-1␤ gene expression in the kidneys of mice with lupus ␤ suppress IL-1 -induced PGE2 production in NIH3T3 cells (un- nephritis. J. Immunol. 141:3050. published data), suggesting that IL-1ra secreted from IL-1ra-in- 8. Jenkins, D. A. S., D. R. Wojtacha, P. Swan, S. Fleming, and A. D. Cumming. 1994. Intrarenal localization of interleukin-1␤ mRNA in crescentic glomerulo- fected vehicle cells may inhibit the bioaction of IL-1. Although the nephritis. Nephrol. Dial. Transplant. 9:1228. fate of these infused vehicle cells is not fully understood, it seems 9. Atkins, R. C. 1995. Interleukin-1 in crescentic glomerulonephritis. Kidney Int. that the majority stay in the spleen and liver without attaching to 48:576. 10. Tesch, G. H., N. Yang, H. Yu, H. Y. Lan, R. Foti, S. J. Chadban, R. C. Atkins, adhesion molecules and cells, but those cells that escape to the and D. J. Nikolic-Paterson. 1997. Intrinsic renal cells are the major source of circulation later accumulate at their destined site when adhesion interleukin-1␤ synthesis in normal and diseased rat kidney. Nephrol. Dial. Trans- plant. 12:1109. molecules are expressed (14). 11. Noronha, I. L., C. Kruger, K. Andrassy, E. Ritz, and R. Waldherr. 1993. In situ The present study also demonstrated that IL-1ra suppressed the production of TNF-␣, IL-1␤ and IL-2R in ANCA-positive glomerulonephritis. expression of ␣-SMA in interstitial myofibroblasts, an event that Kidney Int. 43:682. 12. Tang, W. W., L. Feng, J. L. Vannice, and C. B. Wilson. 1994. Interleukin-1 may influence the process of interstitial fibrosis (24, 25). IL-ra may receptor antagonist ameliorates experimental anti-glomerular basement mem- use any of several possible pathways to accomplish the results brane antibody-associated glomerulonephritis. J. Clin. Invest. 93:273. witnessed in this study. First, IL-ra may blockade the action of 13. Nikolic-Paterson, D. J., H. Y. Lan, P. A. Hill, J. L. Vannice, and R. C. Atkins. 1994. Suppression of experimental glomerulonephritis by the interleukin-1 re- IL-1 on interstitial fibroblasts. IL-1 production is generally thought ceptor antagonist: inhibition of intercellular adhesion molecule-1 expression. to stem from tubular epithelial cells and macrophages infiltrating J. Am. Soc. Nephrol. 4:1695. 14. Yokoo, T., Y. Utsunomiya, T. Ohashi, T. Imasawa, T. Kogure, Y. Futagawa, UUO kidneys. However, Rubbia-Brandt et al. (34) demonstrated T. Kawamura, Y. Eto, and T. Hosoya. 1998. Inflamed site-specific gene delivery that s.c. administration of IL-1␣ into rat connective tissue did not using bone marrow-derived CD11bϩCD18ϩ vehicle cells in mice. Hum. Gene increase the animals’ content of ␣-SMA-positive myofibroblasts. Ther. 9:1731. 15. Yokoo, T., T. Ohashi, Y. Utsunomiya, H. Kojima, T. Imasawa, T. Kogure, Yet, Nikolic-Paterson et al. (13) found that the suppression of ex- Y. Hisada, Y. Eto, T. Kawamura, and T. Hosoya. 1999. Prophylaxis of antibody- perimental glomerulonephritis by IL-1ra was mediated by inhibi- induced acute glomerulonephritis with genetically modified bone marrow-derived tion of ICAM-1 expression. In our study, interstitial ICAM-1 ex- vehicle cells. Hum. Gene Ther. 10:2673. 16. Sakai, T., T. Kawamura, and T. Shirasawa. 1997. Mizoribine improves renal pression was decreased in UUO kidneys from IL-1ra-treated mice. tubulointerstitial fibrosis in unilateral ureteral obstruction (UUO)-treated rat by Furthermore, the IL-1ra reduced macrophage infiltration and might inhibiting the infiltration of macrophages and the expression of ␣-smooth muscle actin. J. Urol. 158:2316. have suppressed their function, which would enhance the pheno- 17. Miyake, S., M. Makimura, Y. Kanegae, S. Harada, Y. Sato, K. Takamori, type change of interstitial fibroblasts into myofibroblasts (16, 35). C. Tokuda, and I. Saito. 1996. Efficient generation of recombinant adenoviruses Therefore, IL-1ra treatment may attenuate interstitial fibrosis in using adenovirus DNA-terminal protein complex and a cosmid bearing the full- length virus genome. Proc. Natl. Acad. Sci. USA 93:1320. obstructed nephropathy by inhibition of ICAM-1 expression and of 18. Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for high- macrophage infiltration. expression transfectants with a novel eukaryotic vector. Gene 108:193. 616 SITE-SPECIFIC GENE DELIVERY INTO INFLAMED INTERSTITIUM

19. Utsunomiya, Y., K. Omura, T. Yokoo, T. Imasawa, T. Kawamura, A. Abe, 28. Cheng, Q.-L., X.-M. Chen, F. Li, H.-L. Lin, Y. Z. Ye, and B. Fu. 2000. Effects K. Hirano, T. Mitarai, N. Maruyama, and O. Sakai. 1996. Macrophage-colony of ICAM-1 antisense oligonucleotide on the tubulointerstitium in mice with uni- stimulating factor (M-CSF) enhances proteinuria and recruitment of macrophages lateral ureteral obstruction. Kidney Int. 57:183. into the glomerulus in experimental murine nephritis. Clin. Exp. Immunol. 106: 29. Cotran, R. S., and J. S. Pober. 1989. Effects of cytokines on vascular endothelium: 286. their role in vascular and immune injury. Kidney Int. 35:969. 20. Hume, D., A. Robinson, G. Mcpherson, and S. Gordon. 1983. The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of 30. Jonjic, N., P. Jilek, S. Bernascon, G. Peri, I. Martin-Padura, S. Cenzuales, antigen. J. Exp. Med. 158:1522. E. Dejana, and A. Mantovani. 1992. Molecules involved in the adhesion and 21. Wuthrich, R. P., A. M. Jevnikar, F. Takei, L. H. Glimcher, and V. E. Kelley. cytotoxicity of activated monocytes on endothelial cells. J. Immunol. 148:2080. 1990. Intercellular adhesion molecule-1 (ICAM-1) expression is up-regulated in 31. Floege, J., and A. J. Rees. 1997. Growth factors and cytokines. In Immunologic autoimmune murine lupus nephritis. Am. J. Pathol. 136:441. renal diseases. E. G. Neilson and W. G. Couser, eds. Lippincott-Raven, Phila- 22. Kaneto, H., J. Morrissey, R. McCracken, A. Reyes, and S. Klahr. 1994. Enalapril delphia, p. 417. reduces collagen type IV synthesis and expression of the interstitium in the ob- 32. Diamond, J. R., D. Kees-Folts, G. Ding, J. E. Frye, and N. C. Restrepo. 1994. structed rat kidney. Kidney Int. 45:1637. ␤ 23. Diamond, J. R., H. van Goor, G. Ding, and E. Engelmyer. 1995. Myofibroblasts Macrophages, monocyte chemoattractant peptide-1, and TGF- 1 in experimental in experimental hydronephrosis. Am. J. Pathol. 146:121. hydronephrosis. Am. J. Physiol. 266:F926. 24. Kliem, V., R. J. Johnson, C. E. Alpers, A. Yoshimura, W. G. Couser, K. M. Koch, 33. Dinarello, C. A. 1996. Biologic basis for interleukin-1 in disease. Blood 87:2095. and J. Floege. 1996. Mechanism involved in the pathogenesis of tubulointerstitial 34. Rubbia-Brandt, L., A. P. Sappino, and G. Gabbiani. 1991. Locally applied GM- fibrosis in the 5/6-nephrectomized rats. Kidney Int. 49:666. CSF induces the accumulation of ␣-smooth muscle actin containing myofibro- 25. Brady, H. R. 1994. Leukocyte adhesion molecules and kidney diseases. Kidney blasts. Virchows Arch. B Cell Pathol. Incl. Mol. Pathol. 60:73. Int. 45:1285. 26. Dal Canton, A. 1995. Adhesion molecules in renal disease. Kidney Int. 48:1687. 35. Diamond, J. R. 1995. Macrophages and progressive renal disease in experimental 27. Ricardo, S. D., M. E. Levinson, M. R. DeJoseph, and, J. R. Diamond. 1996. hydronephrosis. Am. J. Kidney Dis. 26:133. Expression of adhesion molecules in rat renal cortex during experimental hydro- 36. Kelly, V., and K. Moore. 1997. Application of a gene transfer strategy to identify nephrosis. Kidney Int. 50:2002. molecules that incite autoimmune kidney injury. Exp. Nephrol. 5:144. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021