Tt

Y

RnbmtiiC

CTlcvea

©3

T

ISCHEMIA/REPERFUSION INJURYRenal

he C-Terminal Peptide of Connective Tissue Growth Factor Blockshe Full Molecule Binding to Tubular Epithelial Cell

. Shi, W. Wang, Z. Tu, L. Zhang, J. Qiu, Q. Li, F. Ye, S. Li, H. Bu, and Y. Li

ABSTRACT

Objectives. Interstitial fibrosis is a critical pathologic change in chronic allograft nephrop-athy. The cytokine connective tissue growth factor (CTGF, also CCN2) plays an importantrole in epithelial-mesenchymal transformation (EMT) of tubular epithelial cells to renalinterstitial fibrosis. The hexadeca-peptide within the C-terminal of CTGF (named P2)contains the unique binding domain of CTGF to its potential receptor, integrin �v�3. Thisstudy examined whether P2 bound preferentially to the receptor and served as an inhibitorof CTGF.Methods. All studies used an established rat kidney tubular epithelial cell line NRK-52E.Chemically synthesized P2 was purified, and some of it labeled with FITC. The affinityof CTGF or P2 to NRK-52E cells was examined by a solid-phase cell adhesion assay.Competitive binding between P2 and CTGF to NRK-52E cells was examined with flowcytometric analysis.Results. Both P2 and CTGF bound to the NRK-52E cells, mediating cell adhesion.When the cells were incubated in the mixture of P2 and CTGF, P2 bound to the cellspreferentially. Furthermore, when cells were preincubated with excessive CTGF, it becamedifficult for subsequent P2 binding to occur.Conclusions. P2 and CTGF seemed to bind to cell membranes at the same binding

domain. P2 competitively blocked CTGF binding, acting as a CTGF inhibitor.

ifs

mU

PCsSC

TW

ENAL INTERSTITIAL fibrosis is one of the mostcommon pathological changes in most end-stage re-

al diseases.1 Chronic allograft nephropathy (CAN) hasecome the main cause of graft lost. The typical pathologicorphologic changes in the end-stage allograft include

ubular atrophy and interstitial scarring.2 Many factors arenvolved in CAN; however, the disordered, progressiventerstitial fibrotic lesion may offer a major explanation forAN.Connective tissue growth factor (CTGF, also known as

CN2) plays an important role in pathological fibrosis.3

he overexpression of CTGF is essential to the fibroticesion.4 It has been revealed that in renal tubular epithelialells (TECs), express upregulated CTGF both in vivo and initro,5,6 which suggests that CTGF plays a crucial role in thepithelial-mesenchymal transformation (EMT) of TEC,

nd might be a potential target to treat renal fibrosis.7 C

2006 by Elsevier Inc. All rights reserved.60 Park Avenue South, New York, NY 10010-1710

ransplantation Proceedings, 38, 2187–2189 (2006)

The interaction between CTGF and its receptors on TECs unknown. CTGF has been considered to be a novel ligandor integrins. It may functionally engage different integrinubtypes to elicit specific biological effects. Gao et al

From the Key Laboratory of Transplant Engineering and Im-unology, Ministry of Health, West China Hospital, Sichuanniversity, Chengdu, P. R. China.Supported in part by a grant from the National Basic Research

rogram of China (No. 2003CB515504) and Program forhangjiang Scholars and Innovative Research Team in Univer-ity, Ministry of Education, and in part by the National Naturalcientific Foundation (No. 30571761), and by the Foundation ofhina Medical Board of New York Inc. (CMB, Inc.).Address reprint requests to Hong Bu, Key Laboratory of

ransplant Engineering and Immunology, Ministry of Health,est China Hospital, Sichuan University, Chengdu, 610041 P. R.

hina. E-mail: hongbu@hotmail.com

0041-1345/06/$–see front matterdoi:10.1016/j.transproceed.2006.06.093

2187

rCriTCb

hbIp

MC

RowcC

C

RCiphrpPptpcsttps

S

TwNi

tbAfmgCaacB4v

s.

C

TtDFhc

cw1Apflt

RP

AbwwttaewaCs

FNtrpT

2188 SHI, WANG, TU ET AL

eported that a motif containing 16 residues in module 4 ofTGF is a unique, essential binding domain for its potential

eceptor, integrin �v�3, on hepatic stellate cells.8 Both thentegrin �v and �3 subunits have also been detected onECs,9,10 which strongly suggests that blocking binding ofTGF to integrin �v�3 may partially interrupt CTGFiological effects on NRK-52E cell transdifferentation.In this study, we showed that the hexadeca-peptide

omologous to the binding motif in CTGF competitivelyound to NRK-52E cells, thus acting as a CTGF inhibitor.t may offer a promising strategy to block the CTGF signalathway.

ATERIALS AND METHODSell Culture

at renal tubular epithelial line NRK-52E cells were originally,btained from the American Type Culture Collection. The cellsere cultured in Dulbecco’s modified Eagle’s medium (DMEM)ontaining 10% fetal bovine serum in saturated humidity and 5%O2. They were passaged about every 2 days.

TGF and Peptide

ecombinant human CTGF was a product of BioVision (USA).TGF protein used in this study was a lower molecular weight

soform, containing the 98 amino acid residues of the C-terminalortion of the full-length CTGF protein, which exerts fulleparin binding, cell adhesion, and mitogenic activity. ThishCTGF is highly conserved between rat and human. Theeptide IRTPKISKPIKFELSG, according to Gao et al named2, was synthesized using a CS036 peptide synthesizer. The finalroduct was dissolved in 50% aqueous acetonitrile and lyophilizedo yield a white solid. The crude peptide was examined by reversehase high performance liquid chromatography for purity. Theorrect molecular weight was confirmed by electrospray masspectrometry using a Kratos Axima CFR plus V2.3.2a mass spec-rometer.11 The major peak was purified to �95% purity. Some ofhe peptide was labeled with FITC linked to the N-terminal asreviously described.12 The synthetic P2 was diluted in dimethylulfoxide and stored at �70°C prior to further use.

olid Phase Adhesion Assay

o test the binding ability of P2 and CTGF to the NRK-52E cells,e used a solid phase adhesion assay.13 After trypsinization,RK-52E cells were washed with PBS three times and resuspended

n serum-free DMEM.CTGF protein or P2, diluted to 5 �mol/L, was used at 50 �L/well

o precoat 96-well plates (4°C for 16 hours). Then, the wells werelocked for 1 hour with PBS containing 1% bovine serum albumin.cell suspension 2.5 � 105/well added to the microplates was cultured

or 2 hours to attach. Culture medium was discarded by inverting theicroplate and blotting onto paper towels. Wells washed four times

ently with PBS were prewarmed to 37°C. Two hundred microliters ofyQUANT GR dye/cell lysis buffer (CyQUANT cell proliferationssay kit. Molecular Probes, USA) was added to each well followed by5-minute incubation in the dark at room temperature. Adherent

ells were quantitated by measuring the fluorescence intensity using aio-Tek Fluorescence Micro-plate Reader FLx800 at an excitation of85 nm and an emission of 530 nm. All results were expressed as mean

alues � SEM. One-way analysis of variance was used to determine g

tatistical differences between groups using SPSS10.0 software. P �05 was considered significant.

ompetitive Binding Assay

o determine whether CTGF and P2 bound to the same domain inhe cell membrane, 1 � 106 cells suspended in 1 mL serum freeMEM with .5 �mol/L CTGF were cultured for 1 hour. AfterITC-labeled P2 was added, the cells were incubated for anotherour. The fluorescence density of the cells was analyzed by flowytometry.

In another experiment, five groups of NRK-52E cells (1 � 106

ells/mL) resuspended in 1 mL serum-free medium were incubatedith 5 �mol/L FITC-labeled P2. Various dosages of CTGF (0, 5 �0�4, 5 � 10�2, 5 or 500 �mol/L) were added to the microtubes.fter a 1-hour incubation at room temperature on a rockerlatform and three washes with PBS, all samples were analyzed byow cytometry. The fluorescence density of the samples reflectedhe quantity of bound FITC-P2.14

ESULTS2 Bind to the Cells

fter incubation in 96-well plates for 2 hours, cells attachedut could not stick to the surface tightly, whereas, fourashes with PBS made almost all the cells detach. When theells were precoated with CTGF or P2, cells adhered more

ightly to the surface. A large quantity of cells adhered tohe well bottom with fluorescence intensities of 646 � 74nd 612 � 39, respectively. There was no significant differ-nce between the two groups (P � .05). However, in theells without precoating much fewer cells adhered, namely,fluorescence intensity of 127 � 21. (Compared to the

TGF or to the P2 group, P � .05; Figure 1). These datahowed that both CTGF and P2 acted as adhesive mole-

ig 1. Solid phase binding assay of CTGF and P16 toRK-52E cells. After 2-hour incubation and four washings in

he wells without precoat (blank), almost all of the cells wereemoved. Significant amounts of cells remained in the wellsreincubated with either CTGF or P2. (*P � .05 vs blank group.here was no significant difference between CTGF and P2

roup, P � .05).

cf

F

IcwFtwtt

D

IpthCc

PcaNC

flmaoabc

tlCfitppi

eb

fioppfiv

R

K

m5

fr

b(

ggt

ts

gr

(bs

tt5

on

fa

ic

fifrB

an

C TERMINAL PEPTIDE TISSUE GROWTH FACTOR 2189

ules mediating attachment of NRK-52E cells to the sur-ace.

low Cytometric Analyses

n the groups with FITC-P2 but without CTGF, fluores-ence could be detected. When the cells were precoatedith an excessive quantity of CTGF, the fluorescence ofITC-labeled P2 on cells was weak or could not be de-

ected. Surprisingly, in a P2 and CTGF coexistence system,ith the increased concentrations of CTGF in the medium

here was no fluorescence density change compared withhe group containing only P2.

ISCUSSION

t has been speculated that blockage of the CTGF signalathway may offer a promising approach to delay or inverthe fibrosis process.7 In this study, we synthesized P2, theexadeca-peptide homologous to the binding motif ofTGF,8 to test whether it bound to renal tubular epithelial

ells, thus blocking subsequent CTGF binding.A solid-phase adhesion assay was used to confirm whether

2 possessed the same characteristics as CTGF to mediate theell adhesion.13 Our results showed that synthetic P2 acted asn adhesive molecule to mediate solid phase attachment ofRK-52E cells. When the wells were precoated with P2 orTGF, the cells bound to the surface more tightly.The fluorescence of FITC-labeled P2 was detected by

ow cytometry, which showed that P2 had bound to the cellembrane. However, when the cells were precoated with

n excessive quantity of CTGF, the fluorescence was weakr could not be detected. This result suggested that CTGFnd P2 bind to the same membrane receptor. When CTGFound to the cells, the receptors were preoccupied and P2ould no longer bind to the cells.

Furthermore, when CTGF and P2 were both added tohe medium, the increased concentration of CTGF hadittle effect on P2 binding, even though the concentration ofTGF was equal to or much higher than that of P2. Thisnding suggested that the much smaller molecule P2 boundo the cell membrane more easily. The data showed com-etitive binding between P16 and CTGF, and that P16ossessed much stronger affinity and was capable of inhib-

ting CTGF binding to NRK-52E cells.Since CTGF possesses the ability to induce transdiffer-

ntation of NRK-52E cells, blockage of the interaction

etween CTGF and NRK-52E cells may interfere with the 2

brosis in the chronic allograft injury. P2, which is homol-gous to the binding motif of CTGF, bound to the cellsreferentially, inhibiting CTGF binding. This may offer aromising approach to delay or reverse chronic interstitialbrosis. We plan to explore further studies focusing on inivo effects of P2.

EFERENCES

1. Nath KA: The tubulointerstitium in progressive renal disease.idney Int 54:992, 19982. Inkinen KA, Soots AP, Krogerus LA, et al: Fibrosis andatrix metalloproteinases in rat renal allografts. Transpl Int 18:

06, 20053. Qi W, Twigg S, Chen X, et al: Integrated actions of trans-

orming growth factor-betal and connective tissue growth factor inenal fibrosis. Am J Physiol Renal Physiol 288:F800, 2005

4. Leask A: Transcriptional profiling of the scleroderma fibro-last reveals a potential role for connective tissue growth factorCTGF) in pathological fibrosis. Keio J Med 53:74, 2004

5. Kobayashi T, Inoue T, Okada H, et al: Connective tissuerowth factor mediates the profibrotic effects of transformingrowth factor-beta produced by tubular epithelial cells in responseo high glucose. Clin Exp Nephrol 9:114, 2005

6. Yokoi H, Mukoyama M, Sugawara A, et al: Role of connec-ive tissue growth factor in fibronectin expression and tubulointer-titial fibrosis. Am J Physiol Renal Physiol 282:F933, 2002

7. Okada H, Kikuta T, Kobayashi T, et al: Connective tissuerowth factor expressed in tubular epithelium plays a pivotal role inenal fibrogenesis. J Am Soc Nephrol 16:133, 2005

8. Gao R, Brigstock DR: Connective tissue growth factorCCN2) induces adhesion of rat activated hepatic stellate cells byinding of its C-terminal domain to integrin �v�3 and heparanulfate proteoglycan. J Biol Chem 279:8848, 2004

9. Roy-Chaudhury P, Hillis G, McDonald S, et al: Importance ofhe tubulointerstitium in human glomerulonephritis. II. Distribu-ion of integrin chains beta 1, alpha 1 to 6 and alpha V. Kidney Int2:103, 199710. Londono I, Bamri-Ezzine S, Gingras D, et al: Redistribution

f integrins in tubular epithelial cells during diabetic glycogenephrosis. Nephron Exp Nephrol 98:e22, 200411. Dikmans AJ, Morr M, Zander N, et al: A new compact disc

ormat of high density array synthesis applied to peptide nucleiccids and in situ MALDI analysis. Mol Divers 8:197, 2004

12. Kelly KA, Reynolds F, Weissleder R, et al: Fluoresceinsothiocyanate-hapten immunoassay for determination of peptide-ell interactions. Anal Biochem 330:181, 2004

13. Yakovlev S, Zhang L, Ugarova T, et al: Interaction ofbrin(ogen) with leukocyte receptor alpha M beta 2 (Mac-1):urther characterization and identification of a novel bindingegion within the central domain of the fibrinogen gamma-module.iochemistry 44:617, 200514. Siiman O, Burshteyn A, Concepcion O, et al: Competitive

ntibody binding to soluble CD16B antigen and CD16B antigen oneutrophils in whole blood by flow cytometry. Cytometry 44:30,

001