IL-10 Suppresses Chemokines, Inflammation, and Fibrosis ina Model of Chronic Renal Disease

Wei Mu,* Xiaosen Ouyang,* Anupam Agarwal,† Li Zhang,* David A. Long,*Pedro E. Cruz,‡ Carlos A. Roncal,* Olena Y. Glushakova,* Vince A. Chiodo,§

Mark A. Atkinson,� William W. Hauswirth,§ Terry R. Flotte,‡ Bernardo Rodriguez-Iturbe,¶

and Richard J. Johnson**Division of Nephrology, ‡Department of Pediatrics, Powell Gene Therapy Center, and Genetics Institute, §Departmentsof Ophthalmology and Molecular Genetics and Microbiology, and �Department of Pathology, University of Florida,Gainesville, Florida; †Division of Nephrology, University of Alabama, Birmingham, Alabama; and ¶Division ofNephrology, University of Maracaibo, Maracaibo, Venezuela

IL-10 is a pluripotent cytokine that plays a pivotal role in the regulation of immune and inflammatory responses. Whereasshort-term administration of IL-10 has shown benefit in acute glomerulonephritis, no studies have addressed the potentialbenefits of IL-10 in chronic renal disease. Chronically elevated blood levels of IL-10 in rats were achieved by administrationof a recombinant adeno-associated virus serotype 1 IL-10 (rAAV1–IL-10) vector. Control rats were given a similar dose ofrAAV1-GFP. Four weeks after injection, IL-10 levels in serum were measured by ELISA, and chronic renal disease was inducedby a 5/6 nephrectomy (n � 6 in each group). Eight weeks later, rats were killed and renal tissue was obtained for RNA, protein,and immunohistochemical analysis. Serum levels of IL-10 were 12-fold greater in the rAAV1–IL-10 group by 4 wk afterrAAV1–IL-10 administration (345 � 169 versus 28 � 15 pg/ml; P � 0.001), and levels were maintained throughout theexperiment. rAAV1–IL-10 treatment resulted in less proteinuria (P < 0.05), lower serum creatinine (P < 0.05), and highercreatinine clearances (P < 0.01) compared with rAAV1-GFP–treated rats. Renal interstitial infiltration was significantlyattenuated by rAAV1–IL-10 administration as assessed by numbers of CD4�, CD8�, monocyte-macrophages (ED-1�) anddendritic (OX-62�) cells (P < 0.05), and this correlated with reductions in the renal expression of monocyte (renal monocytechemoattractant protein-1 mRNA and protein) and T cell (RANTES mRNA) chemokines. rAAV1–IL-10 administrationdecreased mRNA levels of IFN-� and IL-2 in the kidney. The reduction in inflammatory cells was associated with a significantreduction in glomerulosclerosis and interstitial fibrosis. It is concluded that IL-10 blocks inflammation and improves renalfunction in this model of chronic renal disease. The feasibility of long-term overexpression of a gene using the AAV serotype1 vector system in a model of renal disease is also demonstrated.

J Am Soc Nephrol 16: 3651–3660, 2005. doi: 10.1681/ASN.2005030297

I L-10 is a pluripotent cytokine produced by many activatedimmune cell types, including T-helper (Th2) cells, B cells,macrophages, monocytes, and keratinocytes (1). IL-10 has

been shown to suppress the production of proinflammatorymediators and to downregulate co-stimulatory molecules thatare critical for activation of T cells (2–4), to inhibit macrophageactivation (1–3), and to inhibit vascular smooth muscle cellactivation in vitro and in vivo (5,6).

IL-10 has been reported to attenuate inflammation in a variety

of animal models, including experimental vascular injury (7,8),autoimmune disease (9), and hepatic ischemia/reperfusion (10).IL-10 has also been reported to reduce inflammation and mesan-gial cell proliferation in acute glomerulonephritis induced withanti–Thy 1 antibody (11) and to suppress glomerulosclerosis for-mation in the FGS/Kist, a rat that develops spontaneous focalsegmental glomerulosclerosis (12). To date, there have been nostudies on the potential benefits of long-term IL-10 therapy inmodels of chronic renal disease.

In this study, we examined the effect of IL-10 in a model ofchronic renal disease induced by 5/6 nephrectomy. Systemic ex-pression of IL-10 was achieved using an adeno-associated virusserotype 1–IL-10 (rAAV1–IL-10) vector. We report a beneficialeffect of IL-10 in the remnant kidney model, which is partlymediated by a reduction in the local inflammatory response.

Materials and MethodsRat IL-10 Expression Vector Construction

A reverse transcriptase–PCR product that contained the rat IL-10cDNA was provided by P. Cheepsunthorn and L. Watkins (University

Received March 17, 2005. Accepted September 14, 2005.

Published online ahead of print. Publication date available at www.jasn.org.

W.M. and X.O. contributed equally to this work.

W.W.H. and the University of Florida could be entitled to patent royalties forinventions related to this work, and both own equity in a company that maycommercialize some of the technology described herein.

Address correspondence to: Dr. Wei Mu, Division of Nephrology, University ofFlorida, P.O. Box 100224, Gainesville, FL 32610. Phone: 352-392-3677; Fax: 352-392-5465; E-mail muw@medicine.ufl.edu

Copyright © 2005 by the American Society of Nephrology ISSN: 1046-6673/1612-3651

of Colorado at Boulder, Boulder, CO). The rIL-10 cDNA was clonedinto a TA cloning vector (pCR-XL-TOPO; Invitrogen Corp., Carlsbad,CA) and sequenced by the UF-ICBR sequencing facility. Only a singlesilent base variation was found in the rIL-10 cDNA. The rIL-10–AAV1(pTR2-CB–rIL-10) packageable vector plasmid was made by subcloningrIL-10 into pTR2-CB plasmid (AAV1-ITR–containing vector). The ex-pression of rIL-10 is driven by the hybrid cytomegalovirus enhancer/chicken � actin promoter/hybrid intron cassette (13).

Recombinant AAV Virus ProductionrAAV production was performed by co-transfection of the rAAV1-

ITR vector construct and a combined Ad/AAV helper construct. Indi-vidual helper constructs all express the Ad5 E2a, E4, and VA genes; theAAV Rep gene; and the specific AAV Cap gene desired for pseudotyp-ing (most with AAV1). Transfections were performed by calcium phos-phate co-precipitation in an Ad-E1a– and E1b-expressing permissivehuman cell line (HEK293). rAAV was purified from cells by iodixanoldensity gradient centrifugation. DNA dot blot assay was performed toquantify the titer of rAAV (14).

AnimalsThe study was approved by the Animal Care Committee at the

University of Florida (Gainesville, FL). Male Sprague-Dawley rats (n �

20; Charles River Laboratories, Inc., Wilmington, MA) that weighedbetween 150 and 175 g received an intramuscular injection into thecaudal muscle of the pelvic limb. These injections used 100 �l of salinethat contained 1 � 1010 infectious units of either rAAV1-GFP (n � 10)or rAAV1–IL-10 (n � 10) per rat. Four weeks after injection, IL-10 levelsin serum were measured by ELISA (see below). Two rats from therAAV1–IL-10 group did not achieve serum IL-10 levels �100 pg/mland therefore were excluded from the experiment. The remaining ratsunderwent the remnant kidney procedure.

Rats underwent a right subcapsular nephrectomy with surgical re-section of the upper and lower thirds of the left kidney (n � 20), andtissues were kept for later analysis. To document equivalent reductionin renal mass, the resected kidney was weighed, and the remnantkidney weight was calculated as the weight of the resected wholekidney minus the resected sections of the other kidney. Only rats witha remnant kidney/body weight of 1 to 2% were used, resulting in ninerats in the rAAV1-GFP groups and six rats in the rAAV1–IL-10. Threerats in the rAAV1-GFP group died during the experiment; thus, sixanimals per group were left at the end of the study.

Serum samples were collected at 4, 6, 8, 10, and 12 wk for serum IL-10measurement. Urine samples were collected at 2, 4, 6, and 8 wk afterremnant kidney surgery for urine protein measurement. Eight weeksafter nephrectomy, rats were killed for RNA, protein, and immunohis-tochemical analysis.

Proteinuria and Renal FunctionUrinary protein excretion (24 h) was measured using the Bio-Rad DC

Protein Assay Kit II (Bio-Rad, Hercules, CA), and blood urea nitrogenand creatinine were determined by Alfa Wassermann’s VetACE bio-chemistry machine (Alfa Wassermann, West Caldwell, NJ); urine cre-atinine was measured using the Sigma Creatinine kit (Sigma, St. Louis,MO). Creatinine clearance (Ccr) rates were calculated as U � V/[P],where U and P denote urinary and serum creatinine concentrations,respectively, and V represents ml urine/min.

Renal Pathologic and Histologic StudiesRenal tissues were fixed in Methyl Carnoy’s fixative, and 4-�m

paraffin sections were stained with periodic acid-Schiff and hematox-

ylin. Immunohistochemistry was performed using the following affin-ity-purified primary antibodies: mAb against CD4, CD8, ED1, andOX62 (BD Pharmingen, San Diego, CA). Horseradish peroxidase–con-jugated goat anti-mouse IgG antibodies (Rockland Immunochemicals,Inc., Gilbertsville, PA) were used as secondary antibodies.

For immunohistochemistry, renal tissues were fixed in 4% (wt/vol)buffered paraformaldehyde and frozen at �70°C. Cryosections (4 �m)were incubated with first antibody as described above overnight at 4°C.Sections were washed in PBS, inactivated with endogenous peroxidasein 0.3% H2O2 in methanol, labeled with second antibody as describedabove followed by mouse peroxidase anti-peroxidase, and developedwith DAB substrate kit (Vector Laboratories, Burlingame, CA) to pro-duce a brown color.

For quantification of immunohistochemistry staining, stained sectionswere imaged using a Axioplan 2 imaging microscope (Carl Zeiss Inc.,Thornwood, NY), CR5 digitized color camera, and image-analyzed usingZeiss AutoMeasure software (Axiovision 4.1). The kidney cortex area wasevaluated by an investigator without previous knowledge of the experi-mental group of the animals from which the tissue was taken. Singleimage frames (700 � 550 �m) were captured at �400 magnification, and 20frames per sample were used to count the number of CD4�, CD8�, ED-1�,and OX62� cells (expressed as cell number/[700 � 550 �m]).

Glomerulosclerosis and tubulointerstitial fibrosis scores were alsomeasured in all renal biopsies. For glomerulosclerosis, the percentageof glomeruli that exhibited focal or global glomerulosclerosis was de-termined by evaluation of all glomeruli present in the biopsy. Glomer-ulosclerosis was defined as segmental increases in the glomerular ma-trix, segmental collapse, obliteration of capillary lumina, andaccumulation of hyaline, often with synechial attachment to Bowman’scapsule (15). Twenty glomeruli per sample were calculated for glomer-ulosclerosis scores. Tubulointerstitial injury was defined as inflamma-tory cell infiltration, tubular dilation and/or atrophy, or interstitialfibrosis. Injuries were graded semiquantitatively by a blinded observer,who examined at least 20 cortical fields (magnification �200) in peri-odic acid-Schiff–stained biopsies. Only cortical tubules were includedin the following scoring system (16): 0 � changes �10% of the cortex;1� � changes in up to 20% of the cortex; 2� � changes in up to 40%of the cortex; 3� � changes in up to 60% of the cortex; 4� � changes inup to 80% of the cortex; 5� � changes in �80% of the cortex sections. Eachscore per biopsy was based on 20 frames (representing 700 � 550 �m2

tissue areas). Interstitial fibrosis was also assessed by using an Axioplan 2microscope with Zeiss’s AutoMeasure software to quantify the positivestaining areas of interstitial collagen I and collagen III immunostaining on20 frames per sample tissue area (700 � 550 �m)2

RNA Isolation, Reverse Transcription, and Real-Time PCRTotal RNA was isolated using the SV Total RNA Isolation kit (Pro-

mega, Madison, WI) according to the manufacturer’s protocol. TheRNA was eluted with 50 �l of RNase-free water. All RNA was quan-tified by spectrophotometer, and the optical density 260/280 nm ratioswere determined. Reverse transcription was performed in a one-stepprotocol using the iScript cDNA Synthesis Kit (Bio-Rad) according tothe manufacturer’s protocols. Reactions were incubated at 25°C for 5min, and 42°C for 30 min, 85°C for 5 min and cooled at 4°C in aThermocycler (Eppendorf, Hamburg, Germany). Primers (Table 1)were designed by Genetool software (BioTools Inc., Edmonton, Can-ada), and oligonucleotides were synthesized by Sigma Genosys (Sigma-Genosys Ltd., Woodlands, TX).

Real time PCR analyses were performed using the Opticon PCRmachine (MJ Research, Waltham, MA). The SYBR Green master mix kit(BIO-RAD) was used for all reactions with real-time PCR. Briefly, PCRwas performed as: 94°C for 2 min followed by 40 cycles of denaturation,

3652 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 3651–3660, 2005

annealing, and extension at 94°C for 15 s, 64°C for 30 s, 72°C for 45 s,respectively, and final extension at 72°C for 10 min. PCR reaction foreach sample was done in duplicate for all of the product and for theglyceraldehyde-3-phospate dehydrogenase control. Ratios for eachproduct/glyceraldehyde-3-phospate dehydrogenase mRNA were cal-culated for each sample and expressed as the mean � SD.

Measurement of Cytokine and Chemokine by ELISAAt each time point mentioned above for serum and tissue prepara-

tion, IL-10 and monocyte chemoattractant protein-1 (MCP-1) were mea-sured by ELISA (Rat IL-10 BD OptEIA ELISA Set and Rat MCP-1 BDOptEIA ELISA Set; BD Biosciences Pharmingen, San Diego, CA).Briefly, wells of polystyrene microtiter plates (Polysorp F96; Nunc,Glostrup, Denmark) were coated with Capture antibody of IL-10 orMCP-1 in carbonate bicarbonate buffer (pH 9.6) overnight at 4.0°C. Thenext day, the well was washed three times with PBS that contained0.5% Tween 20 (PBST) and blocked with PBS that contained 5% BSA for1 h at room temperature. After three washes, 100 �l of standard and ratserum samples were added into wells. After 2 h of incubation at roomtemperature and five washes, 100 �l of detection antibody and enzymereagent mixture were added to each well and the plate was incubatedfor 1 h at room temperature. Unbound detection antibody and enzymemixture was removed by seven consecutive washings with PBST. Atotal of 100 �l of substrate solution was added to each well, and theplate was incubated in dark at room temperature for 30 min. Thereaction was stopped by addition of 50 �l of stop solution, and absor-bance was measured at 405 nm with a microplate scanning spectropho-tometer (Powerwave 200; BIO-TEK Instruments, Winooski, VT)

Statistical AnalysesAll data are presented as mean � SD. Differences in the various

parameters between groups were evaluated by single-factor ANOVA.Differences in parameters at each time point after remnant kidneysurgery were compared by paired t test. Correlation between differentfactors was evaluated by Pearson correlation, Kendall �-b, and Spear-man � correlation. Significance was defined as P � 0.05.

ResultsHigh-Circulating IL-10 Levels in Rats Given an Injection ofrAAV1–IL-10

To confirm overexpression of IL-10 and to evaluate the serumconcentration, we collected serum samples at 4, 6, 8, 10, and 12

wk after injection and assayed. IL-10 levels were elevated in allrats that received rAAV1–IL-10, whereas IL-10 remained low inrats that received rAAV1-GFP as control. Levels of IL-10 were12-fold higher at 4 wk and an average of 10-fold higher thancontrols at the 8- and 12-wk time points (Figure 1).

Preservation of Renal Function in rAAV1–IL-10 ratsThe rAAV1–IL-10–treated animals maintained better renal

function compared with the rAAV1-GFP group with signifi-cantly less proteinuria (P � 0.05), lower mean serum creatinines(P � 0.05), and higher Ccr rates (P � 0.01). Serum blood ureanitrogen levels in rAAV1–IL-10 rats were also lower than thatobserved in rAAV1-GFP rats, but the difference was NS (Table2). The proteinuria in both groups gradually increased afterremnant kidney surgery, but rAAV1–IL-10–treated rats hadless urine protein excretion beginning 4 wk after surgery com-pared with rAAV1-GFP control rats (Figure 2).

Table 1. Real-time PCR primers used in the studya

Gene Symbol Gene Description AccessionNumber Forward Primer Sequence (5V–3V) Reverse Primer Sequence (5V–3V)

IL-2 IL-2 NM_053836 TGCAGCGTGTGTTGGATTTGAC TTGCTGGCTCATCATCGAATTGIFN-� IFN-� NM_008337 AACCAGGCCATCAGCAACAACA ACCGACTCCTTTTCCGCTTCCTMCP-1 Chemokine (C-C motif)

ligand 2NM_031530 AGCCCAGAAACCAGCCAACTC GCCGACTCATTGGGATCATCTT

RANTES Chemokine (C-C motif)ligand 5

NM_031116 CCCGTACCTGCCTCCCCATAT GGCGGTTCCTTCGAGTGACAA

TGF-� TGF-�1 NM_021578 ACCGGGTGGCAGGCGAGAG CGGGACAGCAATGGGGGTTCTCOL I Procollagen, type I, �2 NM_053356 GGCTTCAAAGGCATTCGAGGAC GCTCCGACACGCCCTCTCTCCOL III Collagen, type III, �1 NM_032085 GGTGCGGGTGCTGATGGAG GGGCCATGTTCACCTCTCTCACGAPDH Glyceraldehyde-3-

phosphatedehydrogenase

M32599 CTTCACCACCATGGAGAAGGC GGCATGGACTGTGGTCATGAG

aMCP-1, monocyte chemoattractant protein-1,

Figure 1. Serum IL-10. Recombinant adeno-associated virus se-rotype 1 IL-10 (rAAV1-IL-10 rats; f) have increased serumIL-10 levels (pg/ml as measured by ELISA) compared withrAAV1-GFP control rats (F) beginning at 4 wk after rAAV1injection, and levels were maintained until the rats were killed.Shown are mean � SD; six rats are included in each group.***P � 0.001.

J Am Soc Nephrol 16: 3651–3660, 2005 IL-10 Improves Chronic Renal Disease 3653

IL-10 Suppresses Renal Inflammatory Cell Infiltration inRemnant Kidney

Rats that had remnant kidneys and were treated withrAAV1-GFP displayed renal inflammation, as noted by infiltra-tion of macrophages (ED-1 positive), dendritic cells (OX62 pos-itive), and CD4� and CD8� lymphocytes in the interstitial areasof the cortex (Figure 3). rAAV1–IL-10–treated rats showedsignificantly less infiltration of each of these cell types. The levelof serum IL-10 protein in individual animals correlated inverselywith the number of macrophage (ED-1� as cell marker) cells (r �

0.63), dendritic cells (OX62� as cell marker; r � 0.63), CD4� cells(r � 0.61), and CD8� cells (r � 0.69) present (Figure 4).

IL-10 Suppresses Chemokine and Cytokine Expression inRemnant Kidney

To better understand potential mechanisms by which IL-10reduces renal inflammation, we examined the effect of IL-10 onthe expression of the chemokine monocyte chemoattractantprotein-1 (MCP-1 or CCL2) and RANTES (or CCL5), which areimportant chemokines that stimulate monocyte and T cell che-motaxis. We also examined mRNA expression of the cytokineIFN-� which has been reported to augment macrophage-medi-ated renal injury (17), and mRNA expression of the cytokineIL-2 which has an important role in the activation of T cells (18).Quantitative real-time PCR of whole kidney samples obtained

at killing demonstrated a reduction in MCP-1, RANTESmRNA, and IFN-� and IL-2 mRNA expression in rAAV1–IL-10–treated rats compared with rAAV1-GFP rats (Figure 5).Kidney tissue lysates were also assayed for MCP-1 (measuredby ELISA) and confirmed a lower expression of MCP-1 proteinin rAAV1–IL-10–treated rats (Figure 6). The level of MCP-1protein in individual animals correlated closely with the num-ber of macrophage (ED-1�) cells (r � 0.63), dendritic cells (r �

0.47), CD4� cells (r � 0.73), and CD8� cells (r � 0.66) present(Figure 7). Similarly, the level of RANTES mRNA correlatedwith the number of CD4� (r � 0.68), CD8� (0.66), and OX62�

dendritic cells (r � 0.58; P � 0.05 for all comparisons; Figure 7).Serum IL-10 levels also correlated inversely with both renaltissue MCP-1 levels (r � �0.64) and renal RANTES mRNA (r �

�0.73; P � 0.05; Figure 8).

Effect of IL-10 on Renal Fibrosis in Rats withRemnant Kidneys

Both groups developed progressive renal disease with focalglomerular scarring and interstitial fibrosis. We found that thedifferences in glomerulosclerosis and interstitial injury scoresbetween groups were significant (Figure 9; Table 3). Moreover,IL-10–treated rats showed an approximately 50 to 60% reduc-tion in mRNA for TGF-�, collagen I, and collagen III comparedwith control rats as evaluated by real-time PCR (Figure 5), andthis was associated with an approximately 50% reduction incollagen staining (Table 3).

DiscussionIL-10 is widely accepted as an anti-inflammatory cytokine (1)

and might have beneficial effects on some inflammatory-re-lated diseases (7–10). In kidney disease, short-term expressionof IL-10 has been reported to improve glomerulonephritis (11)and glomerulosclerosis (12). We hypothesized that long-termexpression of IL-10 by rAAV1 might provide a novel means fortreating chronic progressive kidney disease. To study this, weselected the remnant kidney model as it is considered to be aclassic model of progressive renal disease and is characterizedby renal inflammation and the slow development of glomeru-losclerosis and tubulointerstitial fibrosis (19,20) that resembleshuman disease.

Chronic systemic overexpression of IL-10 was achieved byusing rAAV gene delivery. This gene delivery system has beenused by others to suppress the development of type 1 diabetes(21–24). It has superiority over many gene delivery approaches

Table 2. Changes of renal function in rAAV1-IL-10 treated group versus rAAV1-GFP control group 8 wk after RKsurgerya

Pre-RK (2 Wk; n � 12) rAAV1-GFP (n � 6) rAAV1–IL-10 (n � 6) P

Serum creatinine (mg/dl) 0.13 � 0.05 1.2 � 0.4 0.9 � 0.1 0.042b

Serum BUN (mg/dl) 6.4 � 1.9 56 � 21 44 � 6 0.153Creatinine clearance rate (ml/min) 1.59 � 0.21 0.9 � 0.3 1.3 � 0.3 0.047b

arAAV1–IL-10, recombinant adeno-associated virus serotype 1 IL-10; RK, remnant kidney; BUN, blood urea nitrogen. Dataare expressed as average � SD.

bP � 0.05.

Figure 2. Proteinuria. A time-dependent increase of proteinuria(mg/16 h) in both groups is observed after remnant kidneysurgery. The progression of proteinuria is significantly inhib-ited by rAAV1–IL-10 treatment 6 wk after surgery comparedwith rAAV1-GFP control rats. Shown are mean � SD; 6 rats areincluded in each group. *P � 0.05.

3654 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 3651–3660, 2005

in that elevated levels of the cytokine of choice can be achievedfor the lifetime of the animals (25). Most studies suggest thatelevated blood levels of the gene of interest are not achieveduntil 2 to 4 wk after gene delivery. We therefore waited 4 wk toensure increased blood levels of IL-10 before initiation of theremnant kidney disease model. At this time point, circulatingIL-10 levels were 12-fold higher compared with control rats,and these levels were maintained for all 8 wk after induction ofthe remnant kidney model. The levels of IL-10 achieved arewell within the range of its known biologic effects on culturedcells (26). We also observed no toxicity in our model, and therewas no difference in weight or behavior between the IL-10–and GFP-overexpressing rats. We thus could address accuratelythe effect of chronic IL-10 on the inflammatory cell response inthis model.

The primary observation was that IL-10 treatment improvedthe remnant kidney model, as assessed functionally by betterCcr and lower proteinuria and histologically by less glomeru-losclerosis and interstitial fibrosis. The improvement was sub-stantial and resulted in a 30% improvement in renal functionand a 50% or greater effect on renal scarring.

The potential mechanism by which IL-10 is renoprotective islikely complex, as IL-10 has multiple actions. Its primary effectis to inhibit activation and effector function of T cells, mono-cytes, and macrophages. IL-10 binds to its receptor (IL-10R),where it activates several signaling pathways depending on thecell type, including stat 3–dependent and –independent path-ways (27), and NF-�B and AP-1 (28,29). A major target of IL-10is the monocyte, where it blocks its activation and proliferation(30–34). IL-10 potently inhibits production of cytokines, such asIL-1 and TNF, and chemokines, such as MCP-1 and RANTES inactivated monocytes/macrophages (33–37). IL-10 also inhibitsdendritic cell function (38–40) and T cell activation, the latter inpart by its effect on macrophages (41–44). IL-10R1 expressionhas also been observed on nonhemopoietic cells (45), raising thepossibility that some of the actions of IL-10 could be mediatedby direct effects on resident renal cell populations.

Because IL-10 acts primarily on monocytes and other im-mune cell populations, we examined the effect of IL-10 on themonocyte, dendritic cell, and T cell infiltration in our model.Macrophages are known to be present in the remnant kidneymodel, where they correlate with fibrosis (46). T cells are alsoknown to be present in human and experimental chronic renaldisease, although their exact role in the fibrogenic responseremains debated (47–50).

Indeed, we confirmed that there was an infiltration of mono-cyte/macrophages, T cells, and dendritic cells in our model. Itis interesting that IL-10 was associated with an approximately50% reduction in this inflammatory response. The possibilitythat this could be one of the renoprotective mechanisms forIL-10 in this model is supported by a similar study in whichtreatment of the remnant kidney model with mycophenolatemofetil also reduced macrophage and T cell infiltration withfunctional and histologic improvement (16).

We further identified a potential mechanism by which IL-10can block the local inflammatory response. Thus, one of themost important chemokines for macrophages in the kidney is

Figure 3. Suppression of inflammatory cells infiltration inrAAV1–IL-10 treated rats. Immunohistochemical stainingshows CD4� (A and B) and CD8� (C and D) lymphocytes,macrophages (indicated by ED-1� cells; E and F), and dendriticcells (OX62� cells; G and H). rAAV1-GFP controls (A, C, E, andG) had significantly more inflammatory cells (brown) in theinterstitial area in cortex of kidney than the rAAV1–IL-10group. The graph summarizes the cell counts in the two groups.The cell number is from image frames that are 700 � 550 �m onhistologic slides using a �40 objective. Data are expressed asmean � SD; 5 rats are included in each group. *P � 0.05; **P �0.01; ***P � 0.001. Magnification �400.

J Am Soc Nephrol 16: 3651–3660, 2005 IL-10 Improves Chronic Renal Disease 3655

MCP-1, and MCP-1 is known to be expressed in the remnantkidney model, where it correlates with the monocyte infiltra-tion (20,51,52). Inhibition of MCP-1 has also been found toresult in less tubulointerstitial inflammation and fibrosis in

several disease models (7,53,54). IL-10 is known to potentlysuppress MCP-1 in activated macrophages (33); hence, this is alikely mechanism to account for the reduction in monocyteinfiltration observed in the IL-10–overexpressing rats. Al-though MCP-1 is also produced by tubular cells, we were notable to show an effect of IL-10 to suppress MCP-1 in proximaltubular (NRK52E) cells that express the IL-10R (data notshown). Thus, it is likely that the effect of IL-10 is via its

Figure 4. Serum IL-10 level correlates with number of infiltrating inflammatory cells. Serum IL-10 levels correlated inversely withthe number of macrophage (ED-1�) cells (r � 0.63), dendritic cells (r � 0.63), CD4� cells (r � 0.61), and CD8� cells (r � 0.69)present (P � 0.05 for all). Œ, rAAV1-GAP control group; � rAAV1-IL-10 treatment group.

Figure 5. Cytokine, chemokine, and fibrosis mediators inrAAV1-IL-10 treatment versus rAAV1-GFP control by real-timePCR. The graph shows the fold decrease of mRNA expressionof various cytokine, chemokine, and fibrosis mediators byrAAV1–IL-10–treated animals compared with the rAAV1-GFPcontrol animals by real-time PCR. rAAV1–IL-10 treatment sig-nificantly reduced mRNA expression of inflammatory cytokineIFN-� and IL-2, chemokine monocyte chemoattractant-1(MCP-1) and RANTES, TGF-�, and collagen I (COL I) andcollagen III (COLL III); *P � 0.05; **P � 0.01. Six rats areincluded in each group.

Figure 6. Kidney tissue MCP-1 levels. rAAV1–IL-10 rats havedecreased renal MCP-1 levels compared with rAAV1-GFP con-trol rats by ELISA (corrected for kidney protein concentration);P � 0.028. Six rats are included in each group.

3656 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 3651–3660, 2005

suppression of MCP-1 in the inflammatory cell population andthat this slowed the recruitment of monocyte/macrophagesinto the renal interstitium. It is also possible that some of theprotective effects were due to inhibition of RANTES and the Tcell infiltration; however, the role of these latter cells in chronicrenal disease is still unclear.

Consistent with this hypothesis, we found that serum IL-10levels correlated inversely with both renal tissue MCP-1 levelsand renal RANTES mRNA (Figure 8), and the protein level of

MCP-1 protein correlated closely with the number of macro-phage, dendritic cells, and T cells. Furthermore, the mRNAlevel of RANTES correlated closely with the number of den-dritic cells and T cells but not with macrophage number (Figure7), which likely reflects that RANTES works primarily on the Tcell population.

The reduction in renal inflammation was associated withsignificantly less renal fibrosis, as reflected by decreased TGF-�mRNA and decreased collagen mRNA and protein (Figures 5

Figure 7. Correlation of MCP-1 protein and RANTES with inflammatory cell infiltration. The level of renal MCP-1 protein inindividual animals correlated closely with the number of macrophages (ED-1� cells; r � 0.63), dendritic cells (r � 0.47), CD4� cells(r � 0.73), and CD8� cells (r � 0.66) present. Similarly, the level of RANTES mRNA correlated with the number of CD4� (r � 0.68),CD8� (0.66), and OX62� dendritic cells (r � 0.58; P � 0.05 for all comparisons). �, rAAV1-IL-10 group; Œ rAAV1-GFP group.

J Am Soc Nephrol 16: 3651–3660, 2005 IL-10 Improves Chronic Renal Disease 3657

and 9; Table 3). The reduction in inflammation and fibrosis wasalso associated with less proteinuria and better preserved renalfunction. The lower proteinuria likely results from the lesssevere glomerular injury, which could also potentially result inless tubulointerstitial injury and inflammation because protein-uria is known to activate tubular cells.

Whereas the injection of AAV–IL-10 resulted in high circu-lating IL-10 levels, IL-10 mRNA levels that were assessed byreal-time PCR were actually decreased in the kidneys of IL-10–overexpressing rats (data not shown). This is likely the conse-quence of the reduction in immune cell infiltration (because

these cells are major sources of IL-10). This suggests that theeffect that we observed was due to the high circulating levels ofIL-10, which could have both systemic and intrarenal effects.

In conclusion, IL-10 treatment slowed renal progression in ananimal model of chronic renal disease. It is interesting that theeffect may be mediated by suppression of chemokines andreduction in infiltrating inflammatory cells, resulting in a lessvigorous fibrotic response. The possibility that IL-10 may havedirect suppressive effects on fibrosis also remains possible.Further studies are indicated to elucidate the protective mech-anisms by which IL-10 acts in chronic renal disease.

Figure 8. Serum IL-10 level correlates with kidney tissue MCP-1 protein and kidney RANTES mRNA. Serum IL-10 levels correlatedinversely with both renal tissue MCP-1 levels (r � �0.64) and renal RANTES mRNA (r � �0.73; P � 0.05). �, rAAV1-IL-10 group;Œ, rAAV1-GFP group.

Figure 9. Renal histology. Light microscopy shows glomeruli and tubular interstitial areas from rAAV1-GFP rats (A) andrAAV1–IL-10–treated rats (B). Glomerulosclerosis and interstitial tubular injury can been seen in both groups, but rAAV1–IL-10–treated rats have less glomerulosclerosis and less tubular injury as data showed in Table 3. Magnification, �400, periodicacid-Schiff.

Table 3. Evidence for improved tubular interstitial fibrosis in rAAV1–IL-10–treated ratsa

rAAV1-GFP (n � 6) rAAV1–IL-10 (n � 6) rAAV1–IL-10/rAAV2-GFP P

Glomerulosclerosis score (%) 31.2 � 14.8 10.8 � 1.8 0.35 0.018b

Tubular interstitial injury score 1.9 � 0.8 1.1 � 0.4 0.57 0.026b

COL I (%) 5.4 � 2.3 2.3 � 0.1 0.43 0.009c

COL III (%) 10.5 � 2.4 4.3 � 1.9 0.41 0.001c

aData are expressed as average � SD.bP � 0.05.cP � 0.01.

3658 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 3651–3660, 2005

AcknowledgmentsThe study was supported by National Institutes of Health grants

DK-52121 and HL-68607 and a George O’Brien Center grant (DK-64322)and Gatorade Funds and JDRF.

References1. Goldman M, Velu T: Interleukin-10 and its implications for

immunopathology. Adv Nephrol Necker Hosp 24: 79–90,1995

2. de Waal Malefyt R, Haanen J, Spits H, Roncarolo MG, teVelde A, Figdor C, Johnson K, Kastelein R, Yssel H, deVries JE: Interleukin 10 (IL-10) and viral IL-10 stronglyreduce antigen-specific human T cell proliferation by di-minishing the antigen-presenting capacity of monocytesvia downregulation of class II major histocompatibilitycomplex expression. J Exp Med 174: 915–924, 1991

3. Willems F, Marchant A, Delville JP, Gerard C, Delvaux A,Velu T, de Boer M, Goldman M: Interleukin-10 inhibits B7and intercellular adhesion molecule-1 expression on hu-man monocytes. Eur J Immunol 24: 1007–1009, 1994

4. Petit-Bertron AF, Pedron T, Gross U, Coppee JY, SansonettiPJ, Cavaillon JM, Adib-Conquy M: Adherence modifies theregulation of gene expression induced by interleukin-10.Cytokine 29: 1–12, 2005

5. Selzman CH, McIntyre RC Jr, Shames BD, Whitehill TA,Banerjee A, Harken AH: Interleukin-10 inhibits humanvascular smooth muscle proliferation. J Mol Cell Cardiol 30:889–896, 1998

6. Mazighi M, Pelle A, Gonzalez W, Mtairag el M, PhilippeM, Henin D, Michel JB, Feldman LJ: IL-10 inhibits vascularsmooth muscle cell activation in vitro and in vivo. Am JPhysiol Heart Circ Physiol 287: H866–H871, 2004

7. Zimmerman MA, Reznikov LL, Raeburn CD, Selzman CH:Interleukin-10 attenuates the response to vascular injury.J Surg Res 121: 206–213, 2004

8. Feldman LJ, Aguirre L, Ziol M, Bridou JP, Nevo N, MichelJB, Steg PG: Interleukin-10 inhibits intimal hyperplasiaafter angioplasty or stent implantation in hypercholester-olemic rabbits. Circulation 101: 908–916, 2000

9. Kok MR, Yamano S, Lodde BM, Wang J, Couwenhoven RI,Yakar S, Voutetakis A, Leroith D, Schmidt M, Afione S,Pillemer SR, Tsutsui MT, Tak PP, Chiorini JA, Baum BJ:Local adeno-associated virus-mediated interleukin 10 genetransfer has disease-modifying effects in a murine model ofSjogren’s syndrome. Hum Gene Ther 14: 1605–1618, 2003

10. Yoshidome H, Kato A, Edwards MJ, Lentsch AB: Interleu-kin-10 suppresses hepatic ischemia/reperfusion injury inmice: implications of a central role for nuclear factor kap-paB. Hepatology 30: 203–208, 1999

11. Kitching AR, Katerelos M, Mudge SJ, Tipping PG, PowerDA, Holdsworth SR: Interleukin-10 inhibits experimentalmesangial proliferative glomerulonephritis. Clin Exp Im-munol 128: 36–43, 2002

12. Choi YK, Kim YJ, Park HS, Choi K, Paik SG, Lee YI, ParkJG: Suppression of glomerulosclerosis by adenovirus-mediated IL-10 expression in the kidney. Gene Ther 10:559–568, 2003

13. Song S, Embury J, Laipis PJ, Berns KI, Crawford JM, FlotteTR: Stable therapeutic serum levels of human alpha-1 an-titrypsin (AAT) after portal vein injection of recombinant

adeno-associated virus (rAAV) vectors. Gene Ther 8: 1299–1306, 2001

14. Zolotukhin S, Potter M, Zolotukhin I, Sakai Y, Loiler S,Fraites TJ Jr, Chiodo VA, Phillipsberg T, Muzyczka N,Hauswirth WW, Flotte TR, Byrne BJ, Snyder RO: Produc-tion and purification of serotype 1, 2, and 5 recombinantadeno-associated viral vectors. Methods 28: 158–167, 2002

15. Kang DH, Yu ES, Yoon KI, Johnson R: The impact ofgender on progression of renal disease: Potential role ofestrogen-mediated vascular endothelial growth factor reg-ulation and vascular protection. Am J Pathol 164: 679–688,2004

16. Romero F, Rodriguez-Iturbe B, Parra G, Gonzalez L,Herrera-Acosta J, Tapia E: Mycophenolate mofetil preventsthe progressive renal failure induced by 5/6 renal ablationin rats. Kidney Int 55: 945–955, 1999

17. Ikezumi Y, Atkins RC, Nikolic-Paterson DJ: Interferon-gamma augments acute macrophage-mediated renal in-jury via a glucocorticoid-sensitive mechanism. J Am SocNephrol 14: 888–898, 2003

18. Gerez L, Madar L, Shkolnik T, Kristal B, Arad G, Reshef A,Steinberger A, Ketzinel M, Sayar D, Shasha S, et al.: Regu-lation of interleukin-2 and interferon-gamma gene expres-sion in renal failure. Kidney Int 40: 266–272, 1991

19. Floege J, Alpers CE, Burns MW, Pritzl P, Gordon K, CouserWG, Johnson RJ: Glomerular cells, extracellular matrix ac-cumulation, and the development of glomerulosclerosis inthe remnant kidney model. Lab Invest 66: 485–497, 1992

20. Schiller B, Moran J: Focal glomerulosclerosis in the rem-nant kidney model—An inflammatory disease mediatedby cytokines. Nephrol Dial Transplant 12: 430–437, 1997

21. Zhang YC, Pileggi A, Agarwal A, Molano RD, Powers M,Brusko T, Wasserfall C, Goudy K, Zahr E, Poggioli R,Scott-Jorgensen M, Campbell-Thompson M, Crawford JM,Nick H, Flotte T, Ellis TM, Ricordi C, Inverardi L, AtkinsonMA: Adeno-associated virus-mediated IL-10 gene therapyinhibits diabetes recurrence in syngeneic islet cell trans-plantation of NOD mice. Diabetes 52: 708–716, 2003

22. Goudy KS, Burkhardt BR, Wasserfall C, Song S, Campbell-Thompson ML, Brusko T, Powers MA, Clare-Salzler MJ,Sobel ES, Ellis TM, Flotte TR, Atkinson MA: Systemicoverexpression of IL-10 induces CD4�CD25� cell popu-lations in vivo and ameliorates type 1 diabetes in nonobesediabetic mice in a dose-dependent fashion. J Immunol 171:2270–2278, 2003

23. Goudy K, Song S, Wasserfall C, Zhang YC, Kapturczak M,Muir A, Powers M, Scott-Jorgensen M, Campbell-Thomp-son M, Crawford JM, Ellis TM, Flotte TR, Atkinson MA:Adeno-associated virus vector-mediated IL-10 gene deliv-ery prevents type 1 diabetes in NOD mice. Proc Natl AcadSci U S A 98: 13913–13918, 2001

24. You S, Chen C, Lee WH, Brusko T, Atkinson M, Liu CP:Presence of diabetes-inhibiting, glutamic acid decarboxy-lase-specific, IL-10-dependent, regulatory T cells in naivenonobese diabetic mice. J Immunol 173: 6777–6785, 2004

25. Conlon TJ, Flotte TR: Recombinant adeno-associated virusvectors for gene therapy. Expert Opin Biol Ther 4: 1093–1101, 2004

26. van der Leij J, van den Berg A, Blokzijl T, Harms G, vanGoor H, Zwiers P, van Weeghel R, Poppema S, Visser L:Dimeric galectin-1 induces IL-10 production in T-lympho-

J Am Soc Nephrol 16: 3651–3660, 2005 IL-10 Improves Chronic Renal Disease 3659

cytes: An important tool in the regulation of the immuneresponse. J Pathol 204: 511–518, 2004

27. O’Farrell A-M, Liu Y, Moore KW, Mui AL-F: IL-10 inhibitsmacrophage activation and proliferation by distinct signal-ing mechanisms: Evidence for stat3- dependent and -inde-pendent pathways. EMBO J 17: 1006–1018, 1998

28. Wang P, Wu P, Siegel MI, EganRW, Billah MM: Interleukin(IL)-10 inhibits nuclear factor kappa B (NF kappa B) acti-vation in human monocytes. IL-10 and IL-4 suppress cy-tokine synthesis by different mechanisms. J Biol Chem 270:9558–9563, 1995

29. Hurme M, Henttinen T, Karppelin M, Varkila K, Mati-kainen S: Effect of interleukin-10 on NF-kappaB and AP-1activities in interleukin-2 dependent CD8 T lymphoblasts.Immunol Lett 42: 129–133, 1994

30. Bogdan C, Vodovotz Y, Nathan C: Macrophage deactiva-tion by interleukin 10. J Exp Med 174: 1549–1555, 1991

31. Ralph P, Nakoinz I, Sampson-Johannes A, Fong S, Lowe D,Min H-Y, Lin L: IL-10, T lymphocyte inhibitor of humanblood cell production of IL-1 and tumor necrosis factor.J Immunol 148: 808–814, 1992

32. Murphy EE, Terres G, Macatonia SE, Hsieh C-S, Mattson J,Lanier L, Wysocka M, Trinchieri G, Murphy K, O’Garra A:B7 and interleukin-12 cooperate for proliferation and IFNproduction by mouse Th1 clones that are unresponsive toB7 costimulation. J Exp Med 180: 223–231, 1994

33. de Waal Malefyt R, Abrams J, Bennett B, Figdor C, de VriesJ: IL-10 inhibits cytokine synthesis by human monocytes:an autoregulatory role of IL-10 produced by monocytes. JExp Med 174: 1209–1220, 1991

34. Fiorentino DF, Zlotnik A, Mosmann TR, Howard MH,O’Garra A: IL-10 inhibits cytokine production by activatedmacrophages. J Immunol 147: 3815–3822, 1991

35. DeWaal Malefyt R, Figdor CG, Huijbens R, Mohan PS,Bennett B, Culpepper J, Dang W, Zurawski G, de Vries JE:Effects of IL-13 on phenotype, cytokine production, andcytotoxic function of human monocytes. Comparison withIL-4 and modulation by IFN-gamma or IL-10. J Immunol151: 6370–6381, 1993

36. D’Andrea A, Aste AM, Valiante NM, Ma X, Kubin M,Trinchieri G: Interleukin 10 (IL-10) inhibits human lym-phocyte interferon gamma-production by suppressing nat-ural killer cell stimulatory factor/IL-12 synthesis in acces-sory cells. J Exp Med 178: 1041–1048, 1993

37. Gruber MF, Williams CC, Gerrard TL: Macrophage-colony-stimulating factor expression by anti-CD45 stimulated humanmonocytes is transcriptionally up-regulated by IL-1 beta andinhibited by IL-4 and IL-10. J Immunol 152: 1354–1361, 1994

38. Allavena P, Piemonti L, Longoni D, Bernasconi S, Stoppac-ciaro A, Ruco L, Mantovani A: IL-10 prevents the differenti-ation of monocytes to dendritic cells but promotes their mat-uration to macrophages. Eur J Immunol 28: 359–369, 1998

39. Kalinski P, Schuitemaker JH, Hilkens CM, Kapsenberg ML:Prostaglandin E2 induces the final maturation of IL-12-deficient CD1aCCD83C dendritic cells: The levels of IL-12are determined during the final dendritic cell maturationand are resistant to further modulation. J Immunol 161:2804–2809, 1998

40. Fortsch D, Rollinghoff M, Stenger S: IL-10 converts humandendritic cells into macrophage-like cells with increasedantibacterial activity against virulent mycobacterium tu-berculosis. J Immunol 165: 978–987, 2000

41. DeWaal Malefyt R, Haanen J, Spits H, Roncarolo M-G, teVelde A, Figdor C, Johnson K, Kastelein R, Yssel H, de VriesJE: IL-10 and viral IL-10 strongly reduce antigen-specific hu-man T cell proliferation by diminishing the antigen-presenting capacity of monocytes via downregulation of classII MHC expression. J Exp Med 174: 915–924, 1991

42. Fiorentino DF, Zlotnik A, Vieira P, Mosmann TR, HowardM, Moore KW, O’Garra A: IL-10 acts on the antigen pre-senting cell to inhibit cytokine production by Th1 cells.J Immunol 146: 3444–3451, 1991

43. Ding L, Shevach EM: IL-10 inhibits mitogen-induced T cellproliferation by selectively inhibiting macrophage co-stimulatory function. J Immunol 148: 3133–3139, 1992

44. Hsu D-H, Moore KW, Spits H: Differential effects of inter-leukin-4 and -10 on interleukin-2-induced interferon-syn-thesis and lymphokine-activated killer activity. Int Immu-nol 4: 563–569, 1992

45. Moore KW, de Waal Malefyt R, Coffman RL, O’Garra A:Interleukin-10 and the interleukin-10 receptor. Annu RevImmunol 19: 683–765, 2001

46. Floege J, Burns MW, Alpers CE, Yoshimura A, Pritzl P,Gordon K, Seifert RA, Bowen-Pope DF, Couser WG, John-son RJ: Glomerular cell proliferation and PDGF expressionprecede glomerulosclerosis in the remnant kidney model.Kidney Int 42: 297–309, 1992

47. Kitching AR, Tipping PG, Huang XR, Mutch DA, Holds-worth SR: Interleukin-4 and interleukin-10 attenuate estab-lished crescentic glomerulonephritis in mice. Kidney Int 52:52–59, 1997

48. Kitching AR, Tipping PG, Timoshanko JR, Holdsworth SR:Endogenous interleukin-10 regulates Th1 responses thatinduce crescentic glomerulonephritis. Kidney Int 57: 518–525, 2000

49. Tipping PG, Kitching AR, Huang XR, Mutch DA, Holds-worth SR: Immune modulation with interleukin-4 and in-terleukin-10 prevents crescent formation and glomerularinjury in experimental glomerulonephritis. Eur J Immunol27: 530–537, 1997

50. Johnson RJ, Herrera-Acosta J, Schreiner GF, Rodriguez-Iturbe B: Role of immunocompetent cells in nonimmunerenal diseases. Kidney Int 59: 1626–1640, 2001

51. Taal MW, Zandi-Nejad K, Weening B, Shahsafaei A, Kato S, LeeKW, Ziai F, Jiang T, Brenner BM, MacKenzie HS: Proinflamma-tory gene expression and macrophage recruitment in the ratremnant kidney. Kidney Int 58: 1664–1676, 2000

52. Ota T, Tamura M, Osajima A, Doi Y, Kudo H, Anai H,Miyazaki M, Nishino T, Nakashima Y: Expression ofmonocyte chemoattractant protein-1 in proximal tubularepithelial cells in a rat model of progressive kidney failure.J Lab Clin Med 140: 43–51, 2002

53. Yoshioka T, Okada T, Maeda Y, Ikeda U, Shimpo M, No-moto T, Takeuchi K, Nonaka-Sarukawa M, Ito T, Taka-hashi M, Matsushita T, Mizukami H, Hanazono Y, KumeA, Ookawara S, Kawano M, Ishibashi S, Shimada K,Ozawa K: Adeno-associated virus vector-mediated inter-leukin-10 gene transfer inhibits atherosclerosis in apoli-poprotein E-deficient mice. Gene Ther 11: 1772–1779, 2004

54. Tsuruta S, Nakamuta M, Enjoji M, Kotoh K, Hiasa K,Egashira K, Nawata H: Anti-monocyte chemoattractantprotein-1 gene therapy prevents dimethylnitrosamine-in-duced hepatic fibrosis in rats. Int J Mol Med 14: 837–842,2004

3660 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 3651–3660, 2005