ARTICLE IN PRESS

0940-2993/$ - se

doi:10.1016/j.et

�CorrespondE-mail addr

Experimental and Toxicologic Pathology 59 (2007) 53–59

www.elsevier.de/etp

Early induction of osteoactivin expression in rat renal tubular epithelial

cells after unilateral ureteral obstruction

Atsushi Nakamura�, Aiko Ishii, Chieri Ohata, Toshi Komurasaki

Medicinal Research Laboratories, Taisho Pharmaceutical Co., LTD., 1-403 Yoshino-Cho, Kita-Ku, Saitama-Shi,

Saitama 331-9530, Japan

Received 7 February 2007; accepted 29 March 2007

Abstract

In this study, we examined the expression of osteoactivin in the rat kidney after unilateral ureteral obstruction. MaleWistar rats were sacrificed at 6 h, and on days 1, 2, 3 and 7 after the obstruction. The renal tubular lumens graduallydilated, and marked interstitial fibrosis was confirmed histologically on day 3 after the obstruction. The expressions ofosteoactivin and collagen type III were examined by quantitative real-time RT-PCR. An 8-fold increase in osteoactivinmRNA expression as compared with that in the sham-operated group was observed at 6 h after the obstruction,whereas no elevation of collagen type III mRNA expression was observed at this early stage. Furthermore, semi-quantitative RT-PCR was performed to identify upregulation of expression of matrix metalloproteinase-13 mRNArelative to that in the sham-operated control, and normalized to the expression level of b-actin. Intense osteoactivinexpression localized in the dilated tubular epithelium and interstitial fibroblasts in the obstructive kidney was detectedby immunohistochemistry and by in situ hybridization. These results suggested that the early-phase upregulation ofosteoactivin expression in the tubular epithelium in response to renal injury might play a role in triggering renalinterstitial fibrosis via activation of matrix metalloproteinase expression and collagen remodeling in rats.r 2007 Elsevier GmbH. All rights reserved.

Keywords: Osteoactivin; Renal interstitial fibrosis; Unilateral ureteral obstruction; In situ hybridization; Immunohistochemistry;

MMP-13; Collagen type III

Introduction

Osteoactivin/human GPNMB/human HGFIN/mouseDC-HIL is a type I transmembrane glycoprotein(Weterman et al., 1995; Safadi et al., 2001; Shikanoet al., 2001). Osteoactivin has been shown to be stronglyexpressed in the bones of osteopetrotic mutant rats (op/op) and to play an important role in the regulation ofosteoblast differentiation (Selium et al., 2003; Owenet al., 2003). Furthermore, the expression of osteoactivin

e front matter r 2007 Elsevier GmbH. All rights reserved.

p.2007.03.005

ing author. Fax: +8148 652 7254.

ess: a.nakamura@po.rd.taisho.co.jp (A. Nakamura).

in human cancers may suggest a role of this protein intumor cell invasion and metastasis via the inductionof matrix metalloproteinase (MMP) expression (Richet al., 2003).

Abnormal accumulation of extracelluar matrix(ECM) components, particularly of collagen type III,in the tubular interstitial area, has been shown to beassociated with impaired renal function (Teppo et al.,2003). The collagen mass reflects the dynamic balancebetween its synthesis and degradation (Laurent, 1987).This collagen turnover is regulated by the expression ofMMPs and their endogenous inhibitors, the tissueinhibitors of metalloproteinases (TIMPs, Norman and

ARTICLE IN PRESSA. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–5954

Fine, 1999). These genes expressed in the early stages ofthe proximal tubular degeneration-interstitial fibrosissequence possibly play an important role in renal failure.

Renal interstitial fibrosis is known as a step in thefinal common pathway to end-stage renal failure (Nath,1992; Nangaku, 2004). The rat model of unilateralureteral obstruction (UUO) is a well-characterizedmodel of renal interstitial fibrosis, which involvescellular proliferation and extracellular matrix remodel-ing. In the current study, we identified an osteoactivinthat has recognized to be dominantly expressed in a ratmodel of aging-induced chronic renal failure, using arepeated polymerase chain reaction subtraction method.In addition, we evaluated the expression and localiza-tion of osteoactivin and its related genes in the rat UUOkidney by molecular pathological techniques includingin situ hybridization (ISH) and immunohistochemistry.

Materials and methods

Induction of renal interstitial fibrosis in rats

Male Wistar rats were obtained from Charles RiverJapan (Yokohama, Japan) and acclimated until use. Theanimals were maintained under 12 h light/12 h darkcycles, at a constant temperature of 2373 1C andrelative humidity of 50720%. Animal care and useconfirmed to the guidelines of the Institutional AnimalCare and Use Committee of Taisho Pharmaceutical Co.,LTD. Thirty-six rats, all 8 weeks of age, were prepared.UUO in the animals was induced surgically underNembutal (Dainippon Sumitomo Pharma, Osaka,Japan) anesthesia by ligation of the left ureter withsterilized 5–0 silk Matsuda Sutures (Tokyo, Japan).Sham-operated animals were used as controls. The ratswere then sacrificed under anesthesia and exsanguinatedat 6 h, and 1, 2, 3, or 7 days (n ¼ 6 for each group) afterthe induction of UUO, and the animal kidneys weresubjected to the following analyses.

Detection of gene expression by RT-PCR or real-

time RT-PCR

In the sample preparation for the determination ofgene expression by molecular biological analysis, the

Table 1. List of designed primers for RT-PCR or real-time RT-PC

Gene Accession no Forward primer

MMP-13 XM_343345 50- GATGGGCCTTCTGGTCTT

Osteoactivin NM_133298 50- GGTAGCAGATGTCCCAA

Collagen IIIa1 XM_343563 50- GACAGATGCTGGTGCTG

b-Actin NM_031144 50-GTCAGGTCATCACTATCG

resected kidney was treated with RNALater (Ambion,Austin, TX, USA) overnight at 4 1C to prevent RNAdegradation by endogenous RNase. RNA purificationwas carried out using TRIzol Reagent (Invitrogen, SanDiego, CA, USA) according to the manufacturer’sinstructions. Then, after the cDNA was synthesized witholigo-dT primer using the SuperScript PreamplificaitonSystem (Invitrogen), it was purified using the PCRPurification Kit (Qiagen, Valencia, CA, USA), anddissolved in 10mM Tris-1mM EDTA (TE) buffer at10mg/ml. The gene product of rat MMP-13 was amplifiedby RT-PCR using relevant forward and reverse primersunder the following PCR conditions: denaturation at95 1C for 15min, followed by 36 amplification cyclesconsisting of 10 s at 95 1C, 20 s at 50 1C, and 20 s at 72 1C.The amplified gene product was separated by 2.5%agarose gel electrophoresis and identified by SYBRGreen staining (Qiagen). Quantitative real-timeRT-PCR was performed with DNA Engine Opticon 2(Bio-Rad Laboratories, Hercules, CA, USA) using theQuantiTect SYBR Green RT-PCR Kit (Qiagen) inaccordance with the same protocol as that used for theRT-PCR. The annealing temperatures were as follows:52 1C for osteoactivin, 52 1C for collagen type III, and61 1C for b-actin. The primer sequences are shown inTable 1. b-actin was used as the internal control.

ISH

For the detection of osteoactivin RNA expression inhistological sections of the kidney, the left kidney wasremoved and longitudinally divided; then, one half wasimmediately fixed in neutralized 10% formalin buffersolution (Wako, Osaka, Japan) for 2 or 3 days and thenembedded in paraffin. A technique for ISH analysis ofparaffin sections using a digoxigenin (DIG)-labeledRNA probe has been reported previously (Ishii et al.,2004). For the preparation of a RNA probe ofosteoactivin, a 464-bp PCR product was amplified withthe relevant forward and reverse primers using EX Taqpolymerase (Takara, Otsu, Japan); the primer sequenceswere as follows: forward, 50-TGTTCCCCAGATGC-CAGAAGG-30, and reverse, 50-GAAGGTTTCATCA-GACGAGTTCCG-30 (NIH accession #: NM_133298).It was then ligated into the pTBlue vector (Novagen,Madison, WI, USA). The T7 promoter containing the

R

Reverse primer

CT-30 50- GGAAACATCAGGGCTCCAG-30

TCC -30 50- CCATTGAAGGCTCTCCTCAC -30

AGA -30 50- GCCTGATCCATGTAGGCAAT -30

GCAATG-30 50-AGAGGTCTTTACGGATGTCAACG-30

ARTICLE IN PRESSA. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–59 55

osteoactivin PCR product was PCR-amplified with theforward or reverse primers and the T7 promoter adaptorprimer. Dig-labeled RNA probes were synthesized byin vitro transcription using the T7 Polymerase Tran-scription System (Roche).

Immunohistochemistry

The anti-osteoactivin antibody was used for this study(goat IgG, 10 mg/ml, R&D Systems, Minneapolis, USA).For the detection of osteoactivin, the antigen retrievalprocedure used was as follows; 3 mm-thick paraffinsections were boiled in 0.01M citrate buffer (pH 6.0)in a microwave oven for 18min. Immunostaining wasperformed with the Ventana NX Automated Immuno-histochemistry System (Ventana Japan) according to themanufacturer’s instructions; briefly, incubation withanti-osteoactivin antibody as the primary antibody for30min at 37 1C, was followed by incubation with biotin-labeled secondary antibodies (rabbit anti-goat IgG,10 mg/ml, Vector Laboratories, Burlingame, CA, USA)and streptoavidin-labeled horseradish peroxidase (HRP,Ventana). The reactivity was detected with diamino-benzidine (Ventana), and the nuclei by hematoxylinstaining.

Results

Quantitative detection of osteoactivin and collagen

type III

Onset of renal interstitial fibrosis in the UUO kidneywas followed by rapid deterioration, including rapidaccumulation of the ECM proteins. The time-course ofchanges in the mRNA expression levels of osteoactivinand collagen type III in the UUO kidney was measured

Fig. 1. Expression of osteoactivin and collagen type III mRNA as e

(B), Collagen type III. The expression level of each target gene was

samples, the expression level of each gene in the sham kidney was se

from three experiments.

by quantitative real-time RT-PCR. The results revealedan 8-fold increase in the gene expression of osteoactivinas compared with that in the sham-operated kidney at6 h after the induction of UUO, that was preserved untilday 3 after the UUO induction (Fig. 1(A)). However, noincrease in expression of the collagen type III gene couldbe confirmed at this time-point. Expression of thecollagen type III mRNA increased gradually andreached a plateau of 3-fold as compared with that inthe sham-operated kidney on day 3 after the inductionof UUO (Fig. 1(B)). No histological evidence of fibrosiscould be detected in the 6-h UUO kidney (data notshown). The expression of both osteoactivin mRNA andcollagen type III mRNA decreased by day 7 after theinduction of UUO (day-7 UUO kidney) with progres-sion of the renal failure.

Localization of osteoactivin mRNA and protein

ISH analysis was conducted to investigate thedistribution of osteoactivin mRNA in the UUO kidneys.In the 6-h UUO kidney, intense signals of osteoactivinexpression were detected in the renal tubular epithelialcells in the inner to outer stripe of the outer medulla(Fig. 2(B)), whereas only faint signals were observedin the corresponding regions in the sham kidney(Fig. 2(A)). While the osteoactivin signals persisted inthe tubular localization, they became progressively moremarked in the region of the renal interstitial fibrosisfrom day 1 to day 3 after the induction of UUO(Figs. 2(C) and (D)). No signals were detected using thesense probe in the day-3 UUO or the sham-operatedkidneys (data not shown).

The localization of the osteoactivin protein wasexamined by immunohistochemistry using a commercialpolyclonal antibody. As shown in Figs. 3(C) and (D),the osteoactivin protein was detected using the antibody

xamined by quantitative real-time RT-PCR. (A), osteoactivin;

expressed as a ratio to that of b-actin. For comparison among

t as 1.0. The graphs are presented by plotting the means7SD

ARTICLE IN PRESS

Fig. 2. Localization of osteoactivin mRNA as examined by in situ hybridization. (A) sham kidney; (B) 6-h; (C) day-1; (D) day-3

UUO kidney. In the ISH examination, positive signals were originally visualized in blue by NBT/BCIP and alkaline phosphatase.

Original magnification; X100.

A. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–5956

in both the tubular epithelium and the interstitial regionin the day-1 and day-3 UUO kidneys (Figs. 3(A) and(B)), but not in the 6-h UUO or the sham-operatedkidneys (Figs. 3(A) and (B)). Negative staining bynormal goat IgG was confirmed in all the specimens(data not shown).

Semi-quantitative detection of MMP-13

A small amount of MMP-13 mRNA has beendetected in the normal rat kidney (less than 0.6% ofthe osteoactivin mRNA expression level), therefore,semi-quantitative detection of MMP-13 was performedby RT-PCR amplification over 36 cycles. Whereas anarrow band reflective of MMP-13 expression wasobserved in the 6-h UUO kidney, and a clear band inthe day-1 and day-3 UUO kidneys, no band wasobserved in the sham-operated kidney. The expression

level of b-actin remained unchanged in all the samples(Fig. 4).

Discussion

Upregulation of osteoactivin expression in the renalinterstitial fibrosis of rat UUO was confirmed in this study.Osteoactivin expression has also been shown to beupregulated in a rat hepatic cirrhosis model generated byfeeding the rats with a choline-deficient, L-amino acid-defined diet and the protein was suggested to play animportant role in the onset of hepatic fibrosis andprogression of hepatocellular carcinoma (Onaga et al.,2003). Thus, osteoactivin may be involved in a commonpathway in the pathogenesis of fibrosis. The finding ofupregulation of osteoactivin expression in renal interstitialfibrosis raised the following questions; when is the geneexpression induced, where are the sites of localization of

ARTICLE IN PRESS

Fig. 3. Osteoactivin protein expression as examined by immunohistochemistry. (A) sham kidney; (B) 6-h; (C) day-1; (D) day-3 UUO

kidney. Positive signals were originally visualized in brown. Original magnification; X100.

Fig. 4. MMP-13 mRNA expression as examined by semi-quantitative RT-PCR. After PCR for MMP-13 (top panel; 36 cycles) and

b-actin (bottom panel; 25 cycles) as described in Materials and Methods, a 10-ml aliquot of each reaction mixture was eletrophoresed

through a 2.5% agarose gel and then stained with SYBR Green. Two different mRNA samples of sham and each UUO kidney were

represented.

A. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–59 57

ARTICLE IN PRESSA. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–5958

the RNA and protein, and how does the gene functionduring the onset and aggravation of renal interstitialfibrosis in the rat model of UUO. To answer thesequestions, we performed the molecular and histologicalexaminations described in this study.

UUO is well recognized as an experimental animalmodel of typical renal interstitial fibrosis in which theprocess of onset and aggravation renal interstitialfibrosis can be easily evaluated (Klahr and Morrissey,2002; Bascands and Schanstra, 2005). In this study,marked upregulation of collagen type III expression, amarker of renal interstitial fibrosis was also confirmed inthe day-3 UUO kidney by quantitative real-timeRT-PCR. Whereas osteoactivin RNA expression wasfound to be induced early in the time-course, that is, at6 h after the induction of UUO, and persist untilday 3 after the induction of UUO, no up-regulationof collagen type III RNA was observed at 6 h afterthe induction of UUO. Thus, induction of osteoactivinexpression prior to the accumulation of the extra-cellular matrix (ECM) was demonstrated in this study.

Various genes, including MMPs and TIMPs might beinvolved in the progression of fibrosis remodeling (Eddyand Giachelli, 1995; Eddy, 1996). Since in vitro over-expression of osteoactivin controlled by the cytomegalo-virus promoter was demonstrated to induce the expressionof MMPs in NIH3T3 fibroblasts (Ogawa et al., 2005), apossible role of osteoactivin in the induction of MMPs infibrosis remodeling was also suggested. In this study, theexpression of MMP-13 in the rat kidney was detected at6h after the induction of UUO, enhanced on day 1 afterthe induction of UUO, and persisted until day 3 after theinduction of UUO. Expression of MMP-13 as well asosteoactivin was found at all stages, from the onset offibrosis to aggravation of fibrosis to the stage ofremodeling. In rats, MMP-13 is an important enzymethat catalyzes the hydrolysis of substrates, includingcollagen types I–III. The involvement of MMP-13 in thepathogenesis of cancer, arthritis and alcoholic hepaticfibrosis has been reported (Kanayama, 2001; Fosang et al.,1996; Yan et al., 2005), whereas MMP-2 and MMP-9 inhuman. Induction of MMP-13 via that of osteoactivinexpression may be a possible common mechanism under-lying the development of fibrosis in various tissues,including the kidney.

An important role of tubular epithelial cells during theonset and aggravation of renal interstitial fibrosis has beenreported (Abbbate et al., 2002; Ng et al., 1998). ISHstaining in this study clarified that renal tubular epithelialcells expressed osteoactivin de novo in the early stage, thatis, at 6 h, after the induction of UUO. The osteoactivinprotein possesses a signal sequence including consisting ofMESLCGVLVFLLLAAGL in its N-terminal end (NIHaccession: NP_579832). Western-blot analysis using ananti-V5 tag antibody of stable transformant humanembryonic kidney 293 (HEK293) cells expressing rat

osteoactivin containing the V5-tag confirmed the presenceof a 65-kDa protein in the culture supernatant as well ashomogenate of the transformant cells (unpublished data).This osteoactivin produced and secreted by the renalepithelial cells after renal injury might act locally to induceMMP activation as early as at 6h after the induction ofUUO. Furthermore, expression of the osteoactivin proteinin the renal interstitial region in addition to its expressionin the tubular epithelial cells, as confirmed by immuno-histochemistry, may suggest the role of the protein in theaggravation of the fibrosis.

Osteoactivin has a motif sequence of polycystickidney disease 1 (PKD1), first identified in polycystickidney disease, and contains an immunoglobulin-likefold. A homologous sequence to that of microbialcollagenase was recognized at positions 279–315 in thePKD1 motif, ISYKWNFGNDTGLFVSNNHTLNH-TYVLNGTGFNFNLTV by bioinformatics analysisusing NCBI Blast ProDom. However, gelatin zymogra-phy electrophoresis to confirm collagenase activity incommercial recombinant osteoactivin (R&D systems,catalog #2330-AC) revealed no clear band (unpublisheddata), suggesting that the osteoactivin protein itself maynot exhibit collagenase activity.

In a preliminary experiment using the commercialrecombinant protein, osteoactivin (R&D systems) didnot induce the production of MMP-1, -2, -3, -9,MT-MMP-1, -2, or the TIMP-1 and -2 protein inHEK293 cells, as examined using the respective specificantibodies. Therefore, a different response to osteoactivinof different cells, including epithelial and fibroblast cells, isspeculated. The mechanisms underlying MMP geneexpression and activation by upregulation of osteoactivinmight involve signal transduction systems mediated by apotential osteoactivin receptor. However, the existence ofosteoactivin receptors has not yet been demonstrated.

In conclusion, osteoactivin expression is increased inrenal interstitial fibrosis occurring after UUO. In thetubular interstitial region of the injured kidney, thede novo local production of osteoactivin induces theupregulation of MMP-13 and collagen type III. Theseresults were confirmed by ISH and molecular biologicalanalyses, including RT-PCR and real-time RT-PCR.The distribution pattern of osteoactivin suggests that itmay play an important role in the development ofinterstitial fibrosis via inducing MMP-13 production bymediating signaling between the tubular epithelial cellsand interstitial fibroblasts. Thus, osteoactivin may be apossible candidate molecule for the development oftherapeutic agents for chronic renal failure.

Acknowledgements

The authors thank Ms Yukiko Sakai for technicalassistance of animal care and RNA purification.

ARTICLE IN PRESSA. Nakamura et al. / Experimental and Toxicologic Pathology 59 (2007) 53–59 59

References

Abbbate M, Zoja C, Rottoli D, et al. Proximal tubular cells

promote fibrogenesis by TGF-beta 1-mediated induction of

peritubular myofibroblasts. Kidney Int 2002;61:2066–77.

Bascands LJ, Schanstra PJ. Obstructive nephropathy: insights

from genetically engineered animals. Kidney Int 2005(68):

925–37.

Eddy EA. Molecular insights into renal interstitial fibrosis.

J Am Soc Nephrol 1996;7:2495–508.

Eddy AA, Giachelli CM. Renal expression of genes that

promote interstitial inflammation and fibrosis in rats with

protein-overload proteinuria. Kidney Int 1995;47:1546–57.

Fosang JA, Last K, Knauper V, et al. Degradation of cartilage

aggrecan by collagenease-3 (MMP-13). FEBS Lett 1996;

380:17–20.

Ishii A, Nakamura M, Nakamura A, et al. Localization of

calcitonin receptor mRNA in rat kidney: an in situ hybridiza-

tion study. Acta Hisochem Cytochem 2004;37:259–65.

Kanayama H. Matrix metalloproteinases and bladder cancer.

J Med Invest 2001;48:31–43.

Klahr S, Morrissey J. Obstructive nephropathy and renal

fibrosis. Am J Physiol Renal Physiol 2002;287:F861–75.

Laurent GJ. Dynamic state of collagen degradation in vivo

and their possible role in regulation of collagen mass. Am

J Kidney Dis 1987;252:C1–9.

Nangaku M. Mechanisms of tubulointerstitial injury in the

kidney: final common pathways to end-stage renal failure.

Internal Med 2004;43(1):9–17.

Nath KA. Tubulointerstitial changes as a major determinant

in the progression of renal damage. Am J Kidney Dis 1992;

20(1):1–17.

Ng YY, Huang TP, Yang WC, et al. Tubular epithelial-

myofibroblast transdifferentiation in progressive tubuloin-

terstitial fibrosis in 5/6 nephrectomized rats. Kidney Int

1998;54:864–76.

Norman JT, Fine LG. Progressive renal disease: fibroblasts,

extracellular matrix, and integrins. Exp Nephrol 1999;7:

167–77.

Ogawa T, Nikaw T, Furochi H, et al. Osteoactivin upregulates

expression of MMP-3 and MMP-9 in fibroblasts infiltrated

into denervated skeletal muscle in mice. Am J Physiol Cell

Physiol 2005;289:C697–707.

Onaga M, Ido A, Hasuike S, et al. Osteoactivin expressed

during cirrhosis development in rats fed a choline-deficient,

L-amino acid-defined diet, accelerates motility of hepatoma

cells. J Hepatol 2003;39:779–85.

Owen TA, Smock SL, Prakash S, et al. Identification and

characterization of the genes encoding human and mouse

osteoactivin. Crit Rev Eukaryot Gene Expr 2003;13:

205–20.

Rich JN, Shi Q, Hjelmeland M, et al. bone-related genes

expressed in advanced malignancies induce invasion and

metastasis in a genetically defined human cancer model.

J Biol Chem 2003;278:15951–7.

Safadi FF, Xu J, Smock SL, et al. Cloning and characteriza-

tion of osteoactivin, a novel cDNA expressed in osteo-

blasts. J Cell Biochem 2001;84:12–26.

Selium AA, Abdelmagid SM, Kanaan RA, et al. Anti-

osteoactivin antibody inhibits osteoblast differentiation

and function in vitro. Crit Rev Eukaryot Gene Expr

2003;13:265–75.

Shikano S, Bonkobara M, Zukas PK, et al. Molecular clo-

ning of dendritic cell-associated transmembrane protein,

DC-HIL, that promotes RGD-dependent adhesion of

endothelial cells through recognition of heparin sulphate

proteoglycans. J Biol Chem 2001;276:8125–34.

Teppo AM, Tornroth T, Honkanen E, et al. Urinary amino-

terminal propeptide of type III procollagen (PIIINP) as a

marker of interstitial fibrosis in renal transplant recipients.

Transplantation 2003;27(12):2113–9.

Weterman MA, Ajubin N, van Dinter IM, et al. Nmb, a novel

gene, is expressed in low-metastatic human melanoma cell

lines and xenografts. Int J Cancer 1995;60:73–81.

Yan S, Ghen GM, Yu CH, et al. Expression pattern of

matrix metalloproteinases-13 in a rat model of alcoholic

liver fibrosis. Hpeatobiliary Pancreast Dis Int 2005;4(4):

569–72.