increased expression of tgf-beta1 but not of its receptors contributes to human obstructive...

Kidney International, Vol. 56 (1999), pp. 2137–2146

Increased expression of TGF-b1 but not of its receptorscontributes to human obstructive nephropathy

HIROYUKI KANETO, HARUO OHTANI, ATSUSHI FUKUZAKI, SHIGETO ISHIDOYA, ATSUSHI TAKEDA,YUKIHIKO OGATA, HIROSHI NAGURA, and SEIICHI ORIKASA

Department of Urology and Pathology, Tohoku University School of Medicine, Sendai, Japan

Increased expression of TGF-b1 but not of its receptors con- obstructed kidney (OBK) at three days after unilateraltributes to human obstructive nephropathy. ureteral obstruction (UUO), and that the level continued

Background. Previous studies have revealed an increased to be elevated until the 14th day of UUO [1]. Otherexpression of transforming growth factor-b1 (TGF-b1) andinvestigators showed an increase in TGF-b1 mRNA asdeposition of extracellular matrix in the kidney of animals withearly as 12 hours of UUO [3]. Thus, an increase in TGF-b1ureteral obstruction. However, these relationships have not

been elucidated in the hydronephrotic kidney of humans. contributes to tubulointerstitial fibrosis in both acute andMethods. We analyzed the tissue expression of extracellular chronic phases of obstructive nephropathy. Palmer et al

matrix proteins, TGF-b1, and its receptors in the human kidney demonstrated that the urinary TGF-b1 concentration inwith ureteral obstruction by immunohistochemistry and re-the renal pelvis of children with congenital hydronephro-verse transcription-polymerase chain reaction (RT-PCR). Ob-sis was significantly higher than that in the bladder urinestructed kidneys (OBKs) were obtained from patients with

ureteral tumors. A kidney specimen from patients with a renal of the same group or that in the control group [6]. Thus,tumor was used as control (CNKs). it is conceivable that in humans, TGF-b1 contributes to

Results. The interstitial volume was significantly increased renal fibrosis in the setting of acute or chronic ureteralin OBKs in comparison with CNKs. OBKs showed increasedobstruction by similar mechanisms to those described indeposition of collagen types I and IV and fibronectin in thethe experimental animal models.renal interstitium. RT-PCR revealed overexpression of colla-

gen a1(IV) mRNA and fibronectin mRNA in OBKs. OBKs Although several receptors for TGF-b have been iden-showed a significantly increased mRNA expression of TGF-b1 tified, an interaction between type I and type II receptorsin comparison with CNKs. The immunoreactivity for TGF-b1

has an important role in signal transduction through theincreased markedly in the interstitium of OBKs. There was aserine/threonine kinase domain [7, 8]. However, theresignificant correlation between the TGF-b1 mRNA level and

the interstitial volume. However, there was no significant dif- are few reports describing the expression of TGF-b re-ference between OBKs and CNKs in the relative mRNA level ceptors in renal injury. Tamaki et al found mRNA up-nor in immunoreactivity for TGF-b receptors. regulation of both TGF-b and type II receptors in ani-

Conclusions. These data suggest that TGF-b1 may contributemals with adriamycin-induced nephropathy, a model ofto the interstitial fibrosis found in the human kidney with ureteralglomerulosclerosis and interstitial fibrosis [9]. In thisobstruction, mainly because of an increase in the expression of

this cytokine without significant changes to its receptors. model, up-regulation of both the ligand (TGF-b1) and itsreceptor may have a role in interstitial fibrosis. Recently,Sutaria et al demonstrated up-regulation of TGF-b type

Animal experimental models have elucidated the I and type II receptors in both obstructed and contralat-mechanisms of tubulointerstitial injury in the kidney with eral kidneys of rats with UUO [10].ureteral obstruction. Transforming growth factor-b1 This study was therefore designed to analyze the ex-(TGF-b1) is a major cytokine involved in this injury pression of extracellular matrix (ECM) proteins and[1–5]. These authors first demonstrated that TGF-b1 TGF-b1 and to investigate whether receptors for TGF-bmRNA levels are up-regulated in the renal cortex of the have a role in obstructive nephropathy in the human

kidney with ureteral obstruction.Key words: tubulointerstitial disease, cytokine, unilateral ureteral ob-struction, fibrosis, renal injury, hydronephrosis.

METHODSReceived for publication May 12, 1999 Materials and tissue preparationand in revised form July 20, 1999Accepted for publication August 3, 1999 The obstructed kidneys (OBKs) were obtained from

nine patients who underwent nephroureterectomy be- 1999 by the International Society of Nephrology

2137

Kaneto et al: TGF-b1 in human obstructive nephropathy2138

cause of transitional cell carcinoma of the ureter (5 males The primary antibodies used are listed in Table 1.TGF-b1 antibody (Ab96), TGF-b receptor type I anti-and 4 females, mean age 68.4 years, range 50 to 81 years).

Ultrasonography and/or intravenous urography revealed body (VPN), and TGF-b receptor type II antibody (DRL)recognize LAPs of TGF-b1 and intracellular domains ofmild to severe degrees of hydronephrosis. However, we

could not confirm the onset of ureteral obstruction be- TbRI and TbRII, respectively. The properties of theseantibodies have been described previously [13–15]. Thecause all patients had shown hydronephrosis at the time

of their visit to the hospital. The control kidneys (CNKs), second antibodies were peroxidase-conjugated sheepantimouse IgG, F(ab9)2 and donkey antirabbit IgG,which did not show hydronephrosis, were obtained from

eight patients who underwent nephrectomy because of F(ab9)2 (Amersham, Buckinghamshire, UK). Rabbit im-munoglobulin or mouse IgG1 (Dako, Tokyo, Japan) wasrenal cell carcinoma (6 males and 2 females, mean age

52.8 years, range 27 to 69 years). We obtained tissue used as a negative control.from morphologically normal areas of these CNKs. For

Characterization and quantitation of mRNAhistological studies, a surgical specimen of each kidneyby RT-PCRwas fixed in 4% paraformaldehyde (PFA) for two to

four hours at 48C, washed in phosphate-buffered saline Total RNA was extracted from the renal cortex byRNAzol B (Biotecx Laboratories, Houston, TX, USA)(PBS), immersed in 10% and 20% sucrose/PBS each for

four hours, embedded in OCT compound, and frozen in and dissolved in TE buffer [10 mm Tris, 1 mm ethylenedi-aminetetraacetic acid (EDTA)]. cDNA was synthesizedliquid nitrogen. PFA-fixed, paraffin-embedded sections

were also prepared for conventional histological exami- from 8 mg of each total RNA using oligo (dT) 15 primer.The mixture of RNA and oligo dT was incubated for 10nation. For RNA extraction, the renal cortex was dis-

sected, immediately frozen, and stored at 2708C until use. minutes at 658C and chilled on ice. The reverse tran-scriptase (RT) buffer, 5 mm MgCl2, 20 U of RNase inhibi-

Morphometric analysis of interstitial volume tor, 1 mm dNTP and 15 U of AMV RT (Promega, Madi-son, WI, USA) were added, incubated at 428C for 60A standard point-counting method was used to quanti-

tate the interstitial volume on the stained sections, as minutes, heated to 958C for three minutes, and chilledon ice. Three microliters of each cDNA were used fordescribed previously [11, 12]. We used a grid containing

36 sampling points engraved on an eyepiece and counted polymerase chain reaction (PCR). To each cDNA wasapplied 47 ml of reaction mixture containing PCR bufferthe points falling on the renal interstitium, including

tubular basement membrane, directly in the microscopic (50 mm KCl, 10 mm Tris-HCl, and 1.5 mm MgCl2), 200mm of each dNTP, and 50 pmol of oligonucleotide prim-field under 3200 magnification in consecutive nonover-

lapping fields (over 50 fields). We determined the per- ers for TGF-b1, TbRI, TbRII, and collagen a1(IV), 25pmol of fibronectin primer, or 20 pmol of glyceralde-centage of the total points/36 3 the number of counted

fields as the volume fraction (Vv). hyde-3-phosphate dehydrogenase (GAPDH) primer.The mixture was first incubated in a thermal cycler at

Immunohistochemical study 728C, and then 1.25 U of Taq polymerase (PharmaciaBiotech, Milwaukee, WI, USA) was added to each tube.Paraffin sections (3 mm in thickness) were deparaffin-

ized and used for immunostaining of fibronectin and PCR was carried out with a Perkin Elmer Cetus DNAThermal Cycler using the following programs (optimalcollagen types I and IV. The deparaffinized sections were

treated in 0.05 m Tris-HCl, pH 7.6, and digested with cycles were developed previously for each primer): 948C(1 min), 608C (1 min), and 728C (1 min) for TGF-b10.1% trypsin in 0.1% CaCl2/0.05 m Tris-HCl, pH 7.6, for

30 minutes at room temperature. Frozen sections (4 mm (30 cycles), collagen a1(IV) (26 cycles), fibronectin (28cycles), and GAPDH (26 cycles); 948C (1 min), 658C (1in thickness) were used for immunostaining of latency-

associated peptide (LAP) for TGF-b1, and TGF-b type I min), and 728C (1 min) for TbRI (34 cycles) and TbRII(32 cycles). The primers were designed according to the(TbRI) and type II (TbRII) receptors. The sections were

washed in PBS and immersed in 10% sheep or donkey human cDNA sequences for TGF-b1, TbRI, TbRII, col-lagen a1(IV), fibronectin, and GAPDH (Table 2). Fibro-normal serum to block nonspecific binding sites for 30

minutes at room temperature and were then incubated nectin primers were designed to flank the EDA regionof the human cDNA. The resultant sizes of PCR productsin the primary antibody overnight at 48C. The sections

were immersed in 0.03% H2O2 in methanol to block were 496 bp and 226 bp of EDA positive (EDA1) andEDA negative (EDA2) variants, respectively. PCRendogenous peroxidase activity and were then incubated

in the secondary antibody (1:100 dilution) for one hour products were sequenced using a dGTP BigDye Termi-nator Cycle Sequencing kit and an ABI PRISM 310at room temperature. The sections were then washed in

PBS, followed by incubation with 0.005% H2O2/0.03% Genetic Analyzer (PE Applied Biosystems, Foster City,CA, USA) to confirm the identity of PCR product asdiaminobenzidine (DAB) in 0.05 m Tris-HCl, pH 7.6,

and were counterstained with 4% methylgreen. target cDNA. After amplification, 15 ml of each PCR

Kaneto et al: TGF-b1 in human obstructive nephropathy 2139

Table 1. Antibodies used in the present study

Antibody Clone Final dilution Source

Fibronectin Polyclonal 1:1000 Dako, Tokyo, JapanCollagen type I Polyclonal 1:200 Chemicon, Temecula, USACollagen type IV CIV22 1:100 Dako, Glostrup, DenmarkTGF-b1 (Ab96) Polyclonal 1:200 The Cancer Institutea

TGF-b receptor type I (VPN) Polyclonal 1:50 The Cancer Institutea

TGF-b receptor type II (DRL) Polyclonal 1:10 The Cancer Institutea

a These antibodies were kindly provided by Dr. K. Miyazono, Department of Biochemistry, The Cancer Institute, Tokyo, Japan

Table 2. Primer sequences

Gene Primers Size Reference

TGF-b1 (antisense) 59-T TGCAGTGTGT TAT CCGTGC TGT C 187 bp [16](sense) 59-CAGAAATACAGCAACAAT T CC TGG

TbRI (antisense) 59-GACAGTGCGGT TGTGGCAGAT 354 bp [14](sense) 59-TGT T T C TGCCACC T C TGTAC

TbRII (antisense) 59-T C T CAAAC TGC T C TGAAGTGT 598 bp [17](sense) 59-ACC T CCAT C TGTGAGAAGCC

Collagen a1(IV) (antisense) 59-GGGTGTGT TAGT TACGCAAAT 434 bp [18](sense) 59-GC TGTGGAT CGGC TAC T C T T

Fibronectin (antisense) 59-T T C T T T CAT TGGT CCGGT C T T 496 bp [19](sense) 59-GCAGCCCACAGTGGAGTATGT 226 bp

GAPDH (antisense) 59-T CCACCACCC TGT TGC TGTA 527 bp [20](sense) 59-T CC TGCACCACCAAC TGC T T

product were electrophoresed through 2.0% agarose gel volume. The interstitial volume of OBKs was signifi-cantly greater than that of CNKs (40 6 19.4% vs. 16 6with ethidium bromide (0.5 mg/ml). The gel was photo-

graphed with Polaroid type 665 negative film over ultra- 2.3%, P , 0.01). There was a correlation between thedegree of hydronephrosis assessed by ultrasonographyviolet light. The negative film was developed, and the

bands on the film were scanned by densitometry (Molec- and the interstitial volume in the OBK group. However,there was no significant correlation between the intersti-ular Analyst software; Bio-Rad, Richmond, CA, USA).

The ratio of optical density of each target mRNA against tial volume and the ages of the patients in either theOBK or CNK groups.that of GAPDH mRNA was determined as a relative

mRNA level.Immunostaining for ECM proteins

Statistical analysis Figure 2 demonstrates the immunohistochemical ap-pearance of ECM proteins. Fibronectin was weaklyData are shown as mean 6 sd and were analyzed

by unpaired Student’s t-test. The relationship between stained and confined to the peritubular space in the renalcortex of CNKs (Fig. 2A). OBKs showed a remarkableTGF-b1 mRNA levels and the interstitial volume was

examined by linear regression analysis. increase in fibronectin in the widened interstitial space,but not in the tubular basement membrane (Fig. 2B).Collagen type I (as with fibronectin) was increased in

RESULTSthe interstitium of OBKs (Fig. 2D) in comparison with

Morphology of renal cortex that of CNKs (Fig. 2C). Collagen type IV was confined totubular basement membranes and peritubular capillariesFigure 1A shows the renal cortex of a CNK (40-year-

old male) stained with periodic acid-Schiff. The peritubu- and was not found in the interstitial space in CNKs (Fig.2E). In contrast, OBKs showed an increased depositionlar space was narrow, and there were few interstitial

cells. In contrast to this, the OBK (70-year-old male) of collagen type IV in both the interstitium and thetubular basement membrane (Fig. 2F).showed an increased interstitial space with deposition of

ECM and a marked atrophy of cortical tubules (Fig. 1B).mRNA expression for collagen a1(IV) and fibronectin

Interstitial volume of renal cortex We examined gene expression of collagen type IVas a membranous ECM and that of fibronectin as anThe range of interstitial volume was 13.1 to 19.1% in

CNKs and 14 to 68% in OBKs. Eight of nine OBKs had interstitial ECM. The representative PCR products ofcollagen a1(IV), fibronectin, and GAPDH are shown inover 20%, and five of these had over 40% interstitial


Fig. 1. Morphology of the renal cortex of control kidney (CNK; 40-year-old male; a) and obstructed kidney (OBK; 70-year-old male; b; PASstaining; original magnification 3200).

Figure 3. The GAPDH level was approximately the same not consistent with the region of increased deposition ofin every case. OBKs showed higher collagen a1(IV) ECM.mRNA levels than CNKs. Both EDA1 and EDA2 vari-

mRNA expression for TGF-b1, TbRI, and TbRIIants of fibronectin mRNA were increased in the OBKsas compared with CNKs, although the increase in expres- We first amplified cDNA synthesized with or withoutsion was slight in two cases of OBKs. The ratio of EDA1 RT from the total RNA of CNK to confirm the specificityto EDA2 mRNA level was not different between CNKs of PCR primers. The PCR product of the predicted sizeand OBKs. was obtained by using cDNA synthesized with RT,

whereas no bands were seen in the PCR product fromImmunostaining for TGF-b1, TbRI, and TbRII

cDNA synthesized without RT. Figure 5 demonstratesAb96 recognizes TGF-b1 LAP either with or without representative PCR products of TGF-b1, TbRI, TbRII,

mature TGF-b1 and LAP bound to latent TGF-b–binding and GAPDH obtained from OBKs (2 cases) and CNKsprotein [13]. Figure 4 a and b demonstrate the immuno- (2 cases). GAPDH mRNA levels were not differentstaining of TGF-b in the renal cortex. Immunoreactivity among these four cases. The amount of TGF-b1 PCRfor TGF-b was observed in the interstitial space in both product was slightly greater in OBKs than in CNKs.CNKs and OBKs. In CNKs, TGF-b was seen around OBKs showed approximately the same mRNA levels ofthe peritubular capillaries, but not in the renal tubular TbRI and TbRII as CNKs. We evaluated the relativecells or in glomeruli (Fig. 4a). OBKs showed intense mRNA levels by calculating the ratio of TGF-b1, TbRI,immunoreactivity for TGF-b in the increased interstitial

or TbRII mRNA against GAPDH mRNA (Fig. 6). Thespace, especially around interstitial cells, but not in renal

relative amount of TGF-b1 mRNA in OBKs was signifi-tubules (Fig. 4b). OBKs with 14% interstitial volume

cantly greater than that in CNKs (0.93 6 0.138 vs. 0.69 6showed the same immunoreactivity for TGF-b as CNKs.0.186, P , 0.05). However, the mRNA levels of TbRIThere was no detectable difference in the pattern ofand TbRII were not significantly different betweenTGF-b1 distribution among OBKs.OBKs and CNKs.VPN staining produced positive signals for TbRI in

collecting ducts and distal tubules and showed no quanti-Correlation between interstitial volume and TGF-b1tative differences between CNK (Fig. 4c) and OBK (Fig.mRNA level of obstructed kidneys4d). DRL staining for TbRII exhibited the same pattern

Based on the assumption that the interstitial volumeof distribution as TbRI. In CNKs, TbRII was seen inof the renal cortex in OBKs may reflect the degree ofcollecting ducts and distal tubules, whereas in some spec-fibrogenesis in the interstitium, we analyzed the relation-imens, TbRII was present weakly beneath the basementship between interstitial volume (Vv) and the ratio ofmembrane of proximal tubules (Fig. 4e). In OBKs, theOBK TGF-b1 mRNA (Fig. 7), revealing a positive corre-pattern was approximately the same as in CNKs (Fig. 4f).lation (r 5 0.799, P , 0.01, Pearson). This positive corre-No immunoreactivity for TbRI or TbRII was observed inlation was confirmed with the addition of data frominterstitial cells or glomeruli either in CNKs or OBKs.

The localization of TbRI- and TbRII-positive cells was CNKs (data not shown).


Fig. 2. Immunostaining for extracellular matrix (ECM) proteins in the renal cortex of CNK (a, c, and e) and OBK (b, d, and f ). (a and b)Fibronectin. (c and d) Collagen type I. (e and f) Collagen type IV. Original magnification 3400.


Fig. 3. Representative polymerase chain re-action (PCR) products of collagen a1(IV),fibronectin, and GAPDH in CNKs (1–4) andOBKs (5–8). Fibronectin shows two mRNAvariants of EDA1 (496 bp) and EDA2 (226bp). M, DNA size marker.

DISCUSSION TGF-b up-regulates both fibronectin isoforms or thatEDA2 mRNA is regulated by other cytokines in theWe have demonstrated that the renal cortex of OBKskidneys of humans with obstructive nephropathy.shows variable degrees of increase in the relative volume

The mechanisms underlying the progression of tubulo-of the cortical interstitium from values in the normalinterstitial injuries have been studied by many investiga-range up to 68%, whereas the interstitial volume of nor-tors. TGF-b is the major cytokine promoting the deposi-mal CNKs was 13 to 19%. The variance of the increasetion of ECM during fibrogenesis in renal injury [reviewedof the interstitial volume of OBKs may be due to thein 26]. However, there have been few reports concerningdegree and duration of ureteral obstruction becauseTGF-b in the onset of kidney disease in humans [27, 28].there was a correlation between the degree of hydrone-We found an increase in the expression of TGF-b1phrosis assessed by ultrasonography and the interstitialmRNA and protein in human kidneys with ureteral ob-volume. In the experimental model of UUO, it has beenstruction. A positive correlation between TGF-b1 mRNAshown that interstitial volume increases in OBKs of ratsexpression and the interstitial volume suggests that sus-as early as three days after complete ureteral ligationtained TGF-b1 up-regulation results in severe renal fi-[12] and that in the chronic phase, the interstitial volumebrosis in OBKs in humans. The expansion of the intersti-in OBKs increased by 40% from a value of 2.8% in thetial space was due mainly to the accumulation of ECMkidney of normal rabbits [4].proteins in the interstitium. Thus, these data suggest thatThis immunohistochemical study revealed that theTGF-b1 contributes to interstitial fibrosis in the chronicexpansion of the cortical interstitium of OBKs was partlyphase of human obstructive nephropathy.due to the increased deposition of both interstitial and

There is still controversy concerning the cellular sourcesmembranous ECM. Fibronectin and collagen type I wereof TGF-b1 in tubulointerstitial fibrosis during UUO.confined to the cortical interstitium in both CNKs andMacrophages are potential sources of TGF-b1 in OBKs.OBKs, whereas collagen type IV increased in both theIt has been shown that monocytes/macrophages infiltrateinterstitium and the basement membrane of OBKs.the renal interstitium of OBKs as early as four hours ofmRNA expression for collagen type IV and fibronectinUUO [29]. Diamond et al demonstrated that intracellu-was up-regulated in OBKs as compared with CNKs. Fi-lar TGF-b1 was positive in the monocytes/macrophagesbronectin mRNA precursor is alternatively spliced tothat had infiltrated the renal interstitium of OBKs, andyield several isoforms [21]. It has been shown that TGF-bthat there was a significant correlation between the num-preferentially up-regulates the EDA-containing isoformber of infiltrating mononuclear cells and the TGF-b1(both mRNA and protein) in cultured human fibroblastsmRNA level [3]. Other investigators found immunoreac-[22, 23]. Viedt, Burger, and Hansch demonstrated thattivity for TGF-b1 in interstitial mononuclear cells in thehuman tubular epithelial cells exhibit a preferential in-OBKs 24 hours after the commencement of UUO [5].crease in EDA1 mRNA after stimulation by TGF-b [24,They also detected TGF-b1 in the juxtaglomerular appa-25]. In the experimental UUO model, OBKs overex-ratus, peritubular capillaries, and tubular cells. Yama-pressed EDIIIA (the rat equivalent to human EDA)-moto et al demonstrated that in an anti-mesangial modelpositive mRNA but not the ED-IIIA-negative variant

[1]. However, in this study, both EDA1 and EDA2 of serum nephritis, TGF-b–positive interstitial mononu-clear cells exhibit immunoreactivity for fibroblastic/myo-variants increased in OBKs as compared with CNKs.

The ratio of EDA1 to EDA2 mRNA level was not fibroblastic cells but not for monocytes/macrophages[30]. They also showed an increase in TGF-b transcriptsdifferent between CNKs and OBKs. Thus, it may be that


Fig. 4. Immunostaining for TGF-b1 (a and b), TbRI (c and d ) and TbRII (e and f ) in the renal cortex of CNK (a, c, and e) and OBK (b, d,and f; original magnification 3400).


Fig. 5. Representative PCR products ofTGF-b1, TbRI, and TbRII in OBKs (1, 2)and CNKs (3, 4). M, DNA size marker.

Fig. 6. Relative amounts of TGF-b1, TbRI,and TbRII mRNA in OBKs and CNKs. Thevertical axis expresses mRNA as a ratio ofGAPDH mRNA (N 5 8 for CNKs and N 59 for OBKs). #P , 0.05 vs. CNK.

al reported that TGF-b1 was positive both in tubularepithelial cells and in interstitial cells in adriamycin ne-phropathy [9]. They suggested that only a part of thepopulation of infiltrating macrophages is the source ofTGF-b1 because the number of ED-1–positive cells ex-ceeded that of TGF-b1–positive cells. Thus, both resi-dent renal tubular cells and interstitial cells, includinginfiltrating mononuclear cells, may be responsible forTGF-b production in experimental model.

Transforming growth factor-b is synthesized in a pre-cursor form and is cleaved to active TGF-b by a LAP.Niemir et al found overexpression of LAP but a lowamount of active TGF-b1 in the interstitium of humanIgA glomerulonephritic kidneys and suggested that con-version to active TGF-b from latent TGF-b was limitedFig. 7. Correlation between the relative amount of TGF-b1 mRNA

and the interstitial volume (Vv) of OBKs. There is a significant correla- in this model [27]. In this investigation, we used an anti-tion. body that recognizes LAP with or without active

TGF-b1. LAP was positive mainly in the interstitium,but not in the interstitial mononuclear cells or in renaltubular cells, in both CNKs and OBKs. There was noin both glomeruli and the renal interstitium of the af-difference in the pattern of TGF-b1 distribution amongfected kidney. Renal tubular cells may be responsibleOBKs. Because latent TGF-b is produced in responsefor production of TGF-b1 because an up-regulation ofto up-regulation of TGF-b transcription in the OBK, itTGF-b1 mRNA was observed in renal cortical tubules

rather than glomerular cells of OBKs [1]. Tamaki et is likely that latent TGF-b is produced and promptly


4. Sharma AK, Mauer SM, Kim Y, Michael AF: Interstitial fibrosissecreted into the interstitium, where it is subsequentlyin obstructive nephropathy. Kidney Int 44:774–788, 1993

converted to the active form [31]. 5. Pimentel JL, Sundell CL, Wang S, Kopp JB, Montero A, Marti-nez-Maldonado M: Role of angiotensin II in the expression andWright et al previously showed an up-regulation ofregulation of transforming growth factor-b in obstructive nephrop-TGF-b in OBKs but not in the contralateral kidneys ofathy. Kidney Int 48:1233–1246, 1995

rats after 3 and 14 days of UUO [32]. The same group 6. Palmer LS, Maizels M, Kaplan WE, Firlit CF, Cheng EY:Urine levels of transforming growth factor-b1 in children withdemonstrated an increase in mRNA and protein expres-ureteropelvic junction obstruction. Urology 50:769–773, 1997sions for TbRI and TbRII in both obstructed and contra-

7. Okadome T, Yamashita H, Franzen P, Moren A, Heldin CH,lateral kidneys [10]. They suggested that TGF-b has a Miyazono K: Distinct roles of the intracellular domains of trans-

forming growth factor-b type I and type II receptors in signalpossible role in fibrosis of OBKs and in hypertrophy oftransduction. J Biol Chem 269:30753–30756, 1994the contralateral kidneys through the up-regulation of

8. Wrana JL, Attisano L, Wieser R, Ventura F, Massague J: Mech-TGF-b receptor. Horvath et al showed intense immuno- anism of activation of the TGF-b receptor. Nature 370:341–347,

1994reactivity for TbRI and TbRII in the proximal and distal9. Tamaki K, Okuda S, Ando T, Iwamoto T, Nakayama N, Fujishimatubules in normal human kidney, whereas in the allograft

M: TGF-b1 in glomerulosclerosis and interstitial fibrosis of adria-rejection, the immunoreactivity decreased in these cells mycin nephropathy. Kidney Int 45:525–536, 1994

10. Sutaria PM, Ohebshalom M, McCaffrey TA, Vaughan ED Jr,[33]. We showed that TbRI and TbRII are localized inFelsen D: Transforming growth factor-b receptor types I and IIthe collecting ducts and distal tubules but not in the are expressed in renal tubules and are increased after chronic

interstitial cells or glomeruli, in both CNKs and OBKs. unilateral ureteral obstruction. Life Sci 62:1965–1972, 199811. Moller JC, Skriver E, Olsen S, Maunsbach AB: UltrastructuralHowever, we found no significant quantitative changes

analysis of human proximal tubules and cortical interstitium inin the protein and mRNA levels of these receptors in chronic renal disease (hydronephrosis). Virchows Arch 402:209–OBKs. Because the localization of these cells was not 237, 1984

12. Kaneto H, Morrissey J, MacCracken R, Reyes AA, Klahr S:necessarily consistent with the region of increased depo-Enalapril reduces collagen type IV synthesis and expansion ofsition of ECM, TGF-b receptors expressed in these renal the interstitium in the obstructed kidney of the rat. Kidney Int

tubular cells may not contribute to the interstitial fibrosis. 45:1637–1647, 199413. Olofsson A, Miyazono K, Kanzaki T, Colosetti P, EngstromIt is possible that in the chronic phase of renal fibrosis,

U, Heldin CH: Transforming growth factor-b1, -b2, and -b3 se-the immunoreactivity for receptors is too low to be de- creted by a human glioblastoma cell line. Identification of smalltected by immunohistochemistry. The difference in ex- and different forms of large latent complexes. J Biol Chem

267:19482–19488, 1992pression of ligand and receptors reported by different14. Franzen P, ten Dijke P, Ichijo H, Yamashita H, Schulz P,investigators may be due to the immunoreactivity of the Heldin CH, Miyazono K: Cloning of a TGF b type I receptor

antibodies used or the different types of inflammation that forms a heteromeric complex with the TGF b type II receptor.Cell 75:681–692, 1993encountered.

15. Taketazu F, Kato M, Gobl A, Ichijo H, ten Dijke P, Itoh J,We conclude that TGF-b1 is involved in the interstitial Kyogoku M, Ronnelid J, Miyazono K, Heldin CH, Funa K:

Enhanced expression of transforming growth factor-bs and trans-deposition of ECM proteins in human hydronephroticforming growth factor-b type II receptor in the synovial tissues ofkidneys caused by ureteral obstruction, via up-regulationpatients with rheumatoid arthritis. Lab Invest 70:620–630, 1994

of TGF-b1 expression, without affecting the expression 16. Derynck R, Jarrett JA, Chen EY, Eaton DH, Bell JR, AssoianRK, Roberts AB, Sporn MB, Goeddel DV: Human transformingof TGF-b receptors.growth factor-b complementary DNA sequence and expression innormal and transformed cells. Nature 316:701–705, 1985

ACKNOWLEDGMENTS 17. Lin HY, Wang XF, Ng-Eaton E, Weinberg RA, Lodish HF:Expression cloning of the TGF-b type II receptor, a functionalWe thank Kohei Miyazono, M.D., and Michiyasu Kato, M.D., De-transmembrane serine/threonine kinase. Cell 68:775–785, 1992partment of Biochemistry, The Cancer Institute, Tokyo, Japan, for

18. Brinker JM, Gudas LJ, Loidl HR, Wang SY, Rosenbloom J,providing the antibodies for TGF-b and its receptors.Kefalides NA, Myers JC: Restricted homology between humanalpha 1 type IV and other procollagen chains. Proc Natl Acad SciReprint requests to Hiroyuki Kaneto, M.D., Department of Urology,USA 82:3649–3653, 1985Tohoku University School of Medicine, 1-1, Seiryo-machi, Aoba-ku,

19. Kornblihtt AR, Vibe-Pedersen K, Baralle FE: Isolation andSendai, 980-8574, Japan.characterization of cDNA clones for human and bovine fibronec-E-mail: [email protected] tins. Proc Natl Acad Sci USA 80:3218–3222, 1983

20. Ercolani L, Florence B, Denaro M, Alexander M: Isolationand complete sequence of a functional human glyceraldehyde-3-REFERENCESphosphate dehydrogenase gene. J Biol Chem 263:15335–15341,

1. Kaneto H, Morrissey J, Klahr S: Increased expression of TGF-b1 1988mRNA in the obstructed kidney of rats with unilateral ureteral 21. Norton PA, Hynes RO: Alternative splicing of chicken fibronectinligation. Kidney Int 44:313–321, 1993 in embryos and in normal and transformed cells. Mol Cell Biol

2. Ishidoya S, Morrissey J, McCracken R, Klahr S: Angiotensin 7:4297–4307, 1987II receptor antagonist ameliorates renal tubulointerstitial fibrosis 22. Blaza E, Borsi L, Allemanni G, Zardi L: Transforming growthcaused by unilateral ureteral obstruction. Kidney Int 47:1285–1294, factor b regulates the levels of different fibronectin isoforms in1995 normal human cultured fibroblasts. FEBS Lett 228:42–44, 1988

3. Diamond JR, Kees-Folts D, Ding G, Frye JE, Restrepo NC: 23. Borsi L, Castellani P, Risso AM, Leprini A, Zardi L: Trans-Macrophages, monocyte chemoattractant peptide-1, and TGF-b1 forming growth factor-b regulates the splicing pattern of fibronec-in experimental hydronephrosis. Am J Physiol 266:F926–F933, tin messenger RNA precursor. FEBS Lett 261:175–178, 1990

24. Viedt C, Burger A, Hansch GM: Fibronectin synthesis in tubular1994


epithelial cells: Up-regulation of the EDA splice variant by trans- 29. Schreiner GF, Harris KPG, Purkerson ML, Klahr S: Immuno-logical aspects of acute ureteral obstruction: Immune cell infiltrateforming growth factor b. Kidney Int 48:1810–1817, 1995in the kidney. Kidney Int 34:487–493, 198825. Burger A, Wagner C, Viedt C, Reis B, Hug F, Hansch GM:

30. Yamamoto T, Noble NA, Miller DE, Boder WA: SustainedFibronectin synthesis by human tubular epithelial cells in culture:expression of TGF-b1 underlies development of progressive kid-Effects of PDGF and TGF-b on synthesis and splicing. Kidney Intney fibrosis. Kidney Int 45:916–927, 199454:407–415, 1998

31. Munger JS, Harpel JG, Gleizes PE, Mazzieri R, Nunes I, Rifkin26. Kuncio GS, Neilson EG, Haverty T: Mechanisms of tubulointer-DB: Latent transforming growth factor-b: Structural features andstitial fibrosis. Kidney Int 39:550–556, 1991mechanisms of activation. Kidney Int 51:1376–1382, 199727. Niemir ZI, Stein H, Noronha IL, Kruger C, Andrassy K, Ritz 32. Wright EJ, McCaffrey TA, Robertson AP, Vaughan ED Jr,

E, Waldherr R: PDGF and TGF-b contribute to the natural Felsen D: Chronic unilateral ureteral obstruction is associatedcourse of human IgA glomerulonephritis. Kidney Int 48:1530–1541, with interstitial fibrosis and tubular expression of transforming1995 growth factor-b. Lab Invest 74:528–537, 1996

28. Yoshioka K, Takemura T, Murakami K, Okada M, Hino S, Miya- 33. Horvath LZ, Friess H, Schilling M, Borisch B, Deflorin J,moto H, Maki S: Transforming growth factor-b protein and mRNA Gold LI, Korc M, Buchler MW: Altered expression of trans-in glomeruli in normal and diseased human kidneys. Lab Invest forming growth factor-bs in chronic renal rejection. Kidney Int

50:489–498, 199668:154–163, 1993

increased expression of tgf-beta1 but not of its receptors contributes to human obstructive...

Documents