BBRC Biochemical and Biophysical Research Communications 295 (2002) 1014–1019 www.academicpress.com

Upregulation of monocyte chemoattractant protein 1 and effects of transforming growth factor-b 1 in Peyronie’s disease Ching-Shwun Lin,* Guiting Lin, Zhong Wang, Suzan A. Maddah, and Tom F. Lue Knuppe Molecular Urology Laboratory, Department of Urology, School of Medicine, University of California, San Francisco, CA 94143-1695, USA Received 14 June 2002

Abstract Peyronie’s disease (PD) is characterized by fibrosis in the tunica albuginea (TA) of the penis, which becomes bent as a result. We have previously shown that transforming growth factor-b 1 (TGF-b1) is upregulated in the TA of patients with PD and can initiate PD-like lesions in rat models. In this study we isolated three types of fibroblasts: P cells from the lesions of PD patients, C cells from the normal-appearing areas of the TA of the same patients, and N cells from the TA of patients without PD. We examined these cells for the expression of two fibrogenic cytokines, connective tissue growth factor (CTGF), and Monocyte Chemoattractant Protein 1 (MCP-1). We found that CTGF was expressed at similar levels in P, C, and N cells, whereas MCP-1 was significantly more expressed in P cells than in C cells and more in C cells than in N cells. Higher MCP-1 expression was also found in the lesions than in normal TA. Treatment with TGF-b1-induced higher expression of MCP-1 but not CTGF in all three types of cells, with C cells being the most responsive. Based on these observations, we propose that MCP-1 could be a novel therapeutic target in PD. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Tunica albugenia; Penis; TGF; MCP-1; CTGF; Peyronie’s disease

The tunica albuginea (TA) encloses the corpora cavernosa of the penis and is a bi-layered structure composed mainly of collagen bundles and elastic fibers [1]. The TA’s elasticity permits the increase in girth and length of the penis and its tensile strength provides the rigidity for the erect penis. However, this compliance is compromised in a condition called Peyronie’s disease (PD) that afflicts 1–3% of men and is characterized by a fibrotic lesion (plaque) in the TA, resulting in penile curvature during erection and, in severe cases, inability to maintain erection [2,3]. Etiological factors of PD include vitamin E deficiency, the use of b-blockers, autoimmune response, and genetic disorder. However, the most plausible cause is believed to be repetitive injuries to the penis during intercourse followed by hyperactive wound-healing that results in excess extracellular matrix (ECM) production and disorganized ECM deposition in the TA, thereby forming scar-like tissues [4]. Similar to other tissue repair processes, the woundhealing process in the TA involves transforming growth *

Corresponding author. Fax: +1-415-353-9586. E-mail address: [email protected] (C.-S. Lin).

factor-b (TGF-b), which exists in three forms in mammals: TGF-b1, -2, and -3 with TGF-b1 being most strongly implicated in fibrosis. In a previous study of 30 patients with PD and 6 patients with normal-appearing TA, we detected TGF-b1 expression in the TA of 26 PD patients (86%) and one non-PD patient (17%). The sole non-PD patient with TGF-b1 expression had twice undergone surgery for the revision of his penile prosthesis and had surgery-induced fibrosis in the TA. In an experiment with rats, incision in the TA followed by suture repair produced inflammatory changes similar to those seen in the acute phase of PD but not the overt changes in the chronic phase [5]. It also resulted in a slight and transient up-regulation of TGF-b1 protein expression. In other experiments with rats, intra-tunical injection of TGF-b1 [6] or its analog, cytomedulin [7] produced PDlike histologic alterations such as chronic cellular infiltration, focal and diffuse elastosis, thickening, disorganization, and clumping of the collagen bundles. While TGF-b1 is clearly an initiation factor in the wound-healing process leading to fibrosis, the importance of other factors that serve as intermediates has only begun to be appreciated. For example, connective

0006-291X/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 0 0 0 6 - 2 9 1 X ( 0 2 ) 0 0 7 6 5 - 9

C.-S. Lin et al. / Biochemical and Biophysical Research Communications 295 (2002) 1014–1019

tissue growth factor (CTGF), a cytokine of the CCN family, has recently been shown to be involved in many fibrotic diseases and a target for TGF-b1-initiated gene upregulation [8–10]. It has also been shown that monocyte chemoattractant protein-1 (MCP-1), a small protein of the C–C chemokine superfamily, is upregulated in fibrosis of the skin [11], lung [12], liver [13], and kidney [14]. Upregulation of MCP-1 mRNA expression by TGF-b1 has also been observed in cultured skin fibroblasts [15]. In the present study we first examined CTGF and MCP-1 expression in fibroblasts that we isolated from the TA of patients with or without PD. We then examined whether treatment of these cells with TGF-b1 had an effect on CTGF and MCP-1 expression. Materials and methods Establishment of cell culture. All procedures regarding the collection and use of human tissues were approved by the UCSF Committee on Human Research. The sources of tissues for cell culture were (1) the fibrotic plaques of patients with PD undergoing venous grafting surgery [16] to correct the penile curvature, (2) the adjacent normal-appearing TA of the same patients, and (3) the normal TA of patients undergoing penile prosthesis implantation. The surgically removed tissues were washed three times in sterile PBS (phosphate-buffered saline) and cut into 2–3 mm3 segments. The segments were placed evenly onto a 100-mm cell culture dish (Falcon-Becton Dickinson Labware, Franklin Lakes, NJ) inside a cell culture hood. Approximately 10 min later, 10 ml of Dulbecco’s modified Eagle’s medium (DMEM) containing penicillin (100 U/ml), streptomycin (100 lg/ml), and 10% FBS was carefully pipetted into the dish. The dish was then kept undisturbed in a humidified 37 °C incubator with 5% CO2 . Five days later, tissue segments that have detached from the dish were removed, and the medium was replaced with fresh medium. Another 5 days later, all tissue segments were removed and the medium was again replaced with fresh medium. When small islands of cells were noticeable, they were trypsinized and transferred to a fresh culture dish. Expansion of each cell strain was continued with change of medium every 3 days and passages (trypsinization and seeding) approximately every 10 days. All cells used in the following experiments were from passages 4 to 10. RNA preparation. RNAs of cavernous tissues and cell cultures were isolated by the Tri-Reagent RNA extraction method (Molecular Research Center, Cincinnati, OH). Despite the manufacturer’s claim, RNAs prepared by this method usually are contaminated by trace amount of DNA, as assessed by polymerase chain reaction (PCR) with b-actin primers (data not shown). To remove the contaminating DNA, each RNA sample (20–50 lg) was treated with 10 units of RNase-free DNase I (Roche Molecular Biochemicals, Pleasanton, CA) at 37 °C for 30 min. The RNA was then purified by phenol/chloroform extraction and ethanol precipitation. Quantity and purity of RNAs were mea-

1015

sured by spectrophotometry with UV adsorption at wavelengths 260 and 280. Their integrity was visualized by the sharpness of the 28s and 18s ribosomal RNA bands in agarose gels. Oligonucleotide primers. Primer pairs for RT-PCR analysis of CTGF, MCP-1, and b-actin genes are listed in Table 1. RT-PCR analysis. Reverse transcription-polymerase chain reaction (RT-PCR) was performed in an RT step and a PCR step, as described previously [17]. Briefly, 2.5 lg of each RNA was reverse transcribed in a reaction volume of 20 ll. This RT product was then diluted 5–100fold with TE buffer (10 mM Tris–HCl, pH 8, 1 mM EDTA). One ll of each dilution was used in a 10 ll PCR to identify the optimal input within the linear amplification range. PCR was performed in DNA Engine thermocycler (MJ Research, Watertown, MA) under calculated temperature control. The cycling program was set for 35 cycles of 94 °C, 10 s; 55 °C, 10 s; 72 °C, 10 s, followed by one cycle of 72 °C, 5 min. The PCR products were electrophoresed in 1.5% agarose gels in the presence of ethidium bromide, visualized by UV fluorescence, and recorded by the ChemiImager 4000 System (Alpha Innotech, San Leandro, CA). Quantification of the PCR products was determined by the densitometry program of the ChemiImager 4000 System. Quantification of MCP-1 secretion. Cells were seeded at 5
104 cells per well in one ml of DMEM with 10% FBS in 12-well culture plates. In 48–72 h, the cells reached 80% confluence and were then treated with 40 ng/ml of TGF-b1 in DMEM with 0.1% BSA. The culture media were harvested at 0, 1, 8, and 24 h and assayed with an MCP-1 ELISA kit (R&D Systems, Minneapolis, MN). All assays were performed in triplicate in each experiment and all data presented under Results are the average of three independent experiments. Immunohistochemical staining. Freshly dissected human TA samples were fixed for 4 h in 2% formaldehyde, 0.002% picric acid in 0.1 M phosphate buffer, pH 8.0. Following a 24-h incubation in cold 30% sucrose, 0.1 M phosphate buffer, pH 8.0, tissues were embedded in O.C.T. Compound (Sakura Finetek USA, Torrance, CA), frozen in liquid nitrogen, and storedat )70 °C until use. Tissue sections were cut at 8 lM, mounted on charged slides (Superfrost Plus, Fisher Scientific, Pittsburgh, PA), and air dried. Endogenous peroxidase activity was eliminated by incubating sections for 10 min in 3% hydrogen peroxide/ methanol. Sections were incubated 30 min with 3% horse serum in PBS + 0.3% Triton X-100 followed by MCP-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA; 1:500 dilution) overnight. After rinses, an avidin–biotin–enzyme complex (ABC) staining kit (Vectastain, Vector Laboratories, Burlingame, CA) was used, with diaminobenzidine as chromogen. Sections were counterstained with hematoxylin (bluish staining of all cell nuclei), while positive (antigen expressing) cells were stained with DAB (brown color). Statistic analysis. Data were analyzed with the software of SSP10.0 for Windows 98. Student’s t test was used to determine significance.

Results Establishment of cell cultures. For clarity, we designate cells cultured from the plaque of PD patients as P

Table 1 Oligonucleotide primers Gene

Primer name

Sequence

Size of PCR product

b-Actin

b-Actin-s b-Actin-a

TCTACAATGAGCTGCGTGTG AATGTCACGCACGATTTCCC

368 bp

MCP-1

MCP-1s MCP-1a

GAGATCTGTGCTGACCCCAA GACCCTCAAACATCCCAGG

341 bp

CTGF

CTGF-s CTGF-a

GAAGGGCAAAAAGTGCATCC GACAGTTGTAATGGCAGGCA

235 bp

1016

C.-S. Lin et al. / Biochemical and Biophysical Research Communications 295 (2002) 1014–1019

cells (P for plaque), those from adjacent normal-appearing TA of the same patients as C cells (C for corner, as the tissue was removed from the corners of the Hshaped incision in the venous grafting procedure [16]), and those from the TA of non-PD patients as N cells (N for normal). P1 and C1 cells were from patient number 1; P2 and C2 from patient number 2; and so forth. All P, C, and N cells had characteristics of fibroblasts (data not shown). Expression of MCP-1 and CTGF. A total of four sets of P, C, and N cells were subjected to RT-PCR analysis for the expression of MCP-1 and CTGF mRNA. The results shown in Fig. 1 indicate that P cells expressed more MCP-1 than C cells ðP < 0:01Þ and C cells more than N cells ðP < 0:01Þ. In contrast, differences in

Fig. 1. RT-PCR analysis of MCP-1 and CTGF expression in N, P, and C cells. RNAs of N, P, and C cells were subjected to RT-PCR with primer pair MCP-1s and MCP-1a, primer pair CTGF-s and CTGF-a, and primer pair b-actin-s and b-actin-a (Table 1). The reaction products were electrophoresed in a 1.5% agarose gel with ethidium bromide and photographed, as shown in A. The photograph was analyzed by densitometry and the resulting values were used to derive the chart shown in B. As an example, the value of 0.94 for MCP-1 under the N column is the average of the ratios between the MCP-1 bands and the b-actin bands for N1, N2, N3, and N4 cells (their MCP-1 versus bactin ratios are 1.00, 0.60, 1.18, and 0.98, respectively).

CTGF expression between P and C cells and between C and N cells were not significant ðP > 0:05Þ. Detection of MCP-1 in the TA of PD patients. Immunohistochemical analysis of the TA from four PD and four non-PD patients showed that MCP-1 was readily detectable in all four PD tissues but was either much less expressed or not detectable in all four non-PD tissues. A representative photograph comparing plaque and normal TA is shown in Fig. 2. Effects of TGF-b1 on MCP-1 and CTGF expression. We tested the effects of TGF-b1 on MCP-1 and CTGF expression in N1, P1, and C1 cells. At either 8 or 24 h after TGF-b1 treatment, N1, P1, and C1 cells all expressed significantly more MCP-1 than their respective untreated controls (Fig. 3). In particular, upregulation of MCP-1 expression by TGF-b1 was more pronounced in C cells than in either P or N cells. On the other hand,

Fig. 2. Localization of MCP-1 in the TA of PD (P) and non-PD (N) patients. Under identical experimental conditions, the P tissue sample exhibits higher basal-level MCP-1 expression in all parts of the section than the N tissue sample. In addition, it exhibits locally intense expression of MCP-1 (arrows).

C.-S. Lin et al. / Biochemical and Biophysical Research Communications 295 (2002) 1014–1019

1017

Fig. 4. Quantification of secreted MCP-1 from PBS- and TGF-b1treated N, P, and C cells. Cells were treated with PBS or 40 ng/ml of TGF-b1 for 0, 1, 8, or 24 h. Their culture media were then subjected to ELISA for MCP-1. The data shown are the averages of three independent experiments, in each of which all assays were performed in triplicate. It is noteworthy that the largest difference between PBStreated and TGF-b1-treated cells occurred with the C cells at 24 h.

Discussion

Fig. 3. RT-PCR analysis of MCP-1 and CTGF expression in controls and TGF-b1-treated N, P, and C cells. N, P, and C cells were untreated (controls) or treated with TGF-b1 for 8 or 24 h. Their RNAs were then isolated and subjected to RT-PCR with primer pair MCP-1s and MCP-1a, primer pair CTGF-s and CTGF-a, and primer pair b-actin-s and b-actin-a (Table 1). The reaction products were electrophoresed in a 1.5% agarose gel with ethidium bromide and photographed, as shown in A. The photograph was analyzed by densitometry and the resulting values were used to derive the chart shown in B. Because of the apparent similarity in CTGF expression among all cell samples, only the MCP-1 values are shown in B. As an example, the value of 0.579 for control under the N column is the ratio between the MCP-1 band and the b-actin band for untreated N cells.

CTGF expression was not significantly affected by TGFb1 treatment in any of these cells. TGF-b1 induces MCP-1 secretion. The culture media of TGF-b1-treated and control cells were subjected to ELISA analysis for secreted MCP-1. The results were largely in agreement with those of the RT-PCR analysis, that is, significantly higher levels of MCP-1 expression were seen in all TGF-b1-treated cells and the highest degree of upregulation in C cells (Fig. 4).

While PD is a well-established fibrotic disease, we believe that the present study is the first to examine its pathogenesis with a specific emphasis on fibrogenic factors other than TGF-b1. It is known that TGF-b1 exerts its fibrogenic effects through the regulation of certain downstream genes [18]. Therefore, it has been proposed that these downstream genes might be more tissue-specific and therefore better therapeutic targets as there are concerns over the possible risks of the long-term suppression of TGF-b1 function [19]. Consequently, several downstream fibrogenic factors, including CTGF and MCP-1, have been actively pursued in recent years as possible targets for therapy. In particular, CTGF has been shown to be a promising therapeutic target for scleroderma [8] and its gene promoter is responsive to TGF-b1 stimulation in dermal fibroblasts [9]. However, it appears that CTGF would not be a suitable target for treatment of PD as we found in the present study that there was no significant difference in CTGF expression between PD and non-PD cells or between TGF-b1-treated and -untreated TA cells. MCP-1 has also been identified as a fibrogenic factor in a wide variety of fibrotic diseases [11]. Recently, using microarray analysis, Magee et al. [20] identified MCP-1 as one of several genes upregulated in the TA of PD patients. However, unlike CTGF, MCP-1 is not firmly associated with TGF-b1, as there has been only one report that showed MCP-1 mRNA was upregulated by

1018

C.-S. Lin et al. / Biochemical and Biophysical Research Communications 295 (2002) 1014–1019

TGF-b1 [15]. In the present study, we showed that MCP-1 was not only differentially expressed in TA cells of PD and non-PD patients but also was highly responsive to TGF-b1 stimulation. In particular, the upregulation of MCP-1 by TGF-b1 was more pronounced in C cells than in P or N cells, suggesting that PD patients are perhaps more susceptible to the fibrogenic effects of TGF-b1. This observation and the fact that both TGF-b1 and MCP-1 are capable of inducing their own production [11,21] suggest that early intervention might be necessary in the management of PD so as to prevent the full-blown MCP-1 production and its consequential hyperactive fibrogenesis. Current medical management of PD includes the administration of oral systemic agents such as Potaba, tamoxifen, acetyl-L -carnitine, colchicine, and vitamin E [3]. Tamoxifen was shown to downregulate MCP-1 expression in cultured coronary artery smooth muscle cells [22], and vitamin E to suppress the induction of MCP-1 following liver injury [23]. Whether Potaba, acetyl-L carnitine, or colchicine has an effect on MCP-1 expression is not known. Several additional pharmaceutical agents have been shown to exhibit inhibitory effects on MCP-1 expression and therefore may offer treatment benefits to PD patients. Examples are given below. Wogonin, an antioxidant and anti-inflammatory herb from Scutellaria baicalensis, was shown to inhibit phorbol ester-induced MCP-1 expression in human umbilical vein endothelial cells and therefore may prevent atherosclerosis [24]. Magnolol, a potent antioxidant from Magnolia officinalis, was found to attenuate MCP1 expression after balloon injury of the aorta and may offer some protection against postangioplasty restenosis [25]. Thiazolidinedione, which is used to improve the insulin resistance of individuals with diabetes mellitus [26] and rolipram, a phosphoddiesterase-4 inhibitor [27] were also shown to inhibit MCP-1 expression in lung and nerve disease models, respectively. While these compounds have the potential to become treatment alternatives for PD, we are particularly interested in another compound, pentoxifylline, as a promising MCP-1 modulator and PD medication. Pentoxifylline has been used to treat radiation-induced fibrosis [28,29] and may have beneficial effects on hepatic [30,31] and peritoneal fibrosis [32]. Pentoxifylline has also been shown to reduce MCP-1 expression in peripheral blood mononuclear [33] and pulmonary epithelial cells [34], and in experimental mesangial proliferative glomerulonephritis [35]. In our clinics at UCSF, we have prescribed pentoxifylline to PD patients with satisfactory results and we have observed its antiproliferative effects on PD cell cultures (manuscript in preparation). In conclusion, the present study has identified MCP-1 as a marker and a potential therapeutic target in PD. It also strengthened the link between TGF-b1 and MCP-1.

Acknowledgments This work was supported in part by grants from the California Urology Foundation (to C.-S. Lin), Mr. Arthur Rock, and the Rock Foundation, and NIH (2R01-DK-45370, to T. F. Lue).

References [1] G. Brock, G.L. Hsu, L. Nunes, B. von Heyden, T.F. Lue, J. Urol. 157 (1) (1997) 276–281. [2] G.H. Lischer, A. Nehra, Curr. Opin. Urol. 11 (6) (2001) 631–636. [3] Gholami, S.S., Gonzalez-Cadavid, N.F., Lin, C.-S., Rajfer, J., Lue, T.F., 2002. J. Urol. (in press). [4] C.J. Devine Jr., K.D. Somers, S.G. Jordan, S.M. Schlossberg, J. Urol. 157 (1) (1997) 285–290. [5] A.I. El-Sakka, C.A. Selph, T.S. Yen, R. Dahiya, T.F. Lue, J. Urol. 159 (5) (1998) 1700–1707. [6] T.J. Bivalacqua, E.K. Diner, T.E. Novak, Y. Vohra, S.C. Sikka, H.C. Champion, P.J. Kadowitz, W.J. Hellstrom, J. Urol. 163 (6) (2000) 1992–1998. [7] A.I. El-Sakka, H.M. Hassoba, R.M. Chui, R.S. Bhatnagar, R. Dahiya, T.F. Lue, J. Urol. 158 (6) (1997) 2284–2290. [8] C.P. Denton, D.J. Abraham, Curr. Opin. Rheumatol. 13 (6) (2001) 505–511. [9] A. Leask, S. Sa, A. Holmes, X. Shiwen, C.M. Black, D.J. Abraham, Mol. Pathol. 54 (3) (2001) 180–183. [10] A. Leask, A. Holmes, D.J. Abraham, Curr. Rheumatol. Rep. 4 (2) (2002) 136–142. [11] T. Yamamoto, B. Eckes, T. Krieg, Eur. J. Immunol. 31 (10) (2001) 2936–2941. [12] M. Hasegawa, S. Sato, K. Takehara, Clin. Exp. Immunol. 117 (1) (1999) 159–165. [13] F. Marra, R. DeFranco, C. Grappone, S. Milani, S. Pastacaldi, M. Pinzani, R.G. Romanelli, G. Laffi, P. Gentilini, Am. J. Pathol. 152 (2) (1998) 423–430. [14] C.M. Lloyd, A.W. Minto, M.E. Dorf, A. Proudfoot, T.N. Wells, D.J. Salant, J.C. Gutierrez-Ramos, J. Exp. Med. 185 (7) (1997) 1371–1380. [15] J. Slavin, E. Unemori, T.K. Hunt, E. Amento, Growth Factors 12 (2) (1995) 151–157. [16] T.F. Lue, A.I. El-Sakka, J. Urol. 160 (6 Pt 1) (1998) 2047–2049. [17] C.S. Lin, H.C. Ho, S. Gholami, K.C. Chen, A. Jad, T.F. Lue, Biochem. Biophys. Res. Commun. 285 (2) (2001) 565–569. [18] M. Gharaee-Kermani, S.H. Phan, Curr. Pharm. Des. 7 (11) (2001) 1083–1103. [19] T.J. Franklin, Int. J. Biochem. Cell Biol. 29 (1) (1997) 79–89. [20] T.R. Magee, A. Qian, J. Rajfer, F.C. Sander, L.A. Levine, N.F. Gonzalez-Cadavid, Urology 59 (3) (2002) 451–457. [21] E. Van Obberghen-Schilling, N.S. Roche, K.C. Flanders, M.B. Sporn, A.B. Roberts, J. Biol. Chem. 263 (16) (1988) 7741–7746. [22] E. Seli, B. Selam, G. Mor, U.A. Kayisli, T. Pehlivan, A. Arici, Menopause 8 (4) (2001) 296–301. [23] F. Marra, R. DeFranco, C. Grappone, M. Parola, S. Milani, G. Leonarduzzi, S. Pastacaldi, U.O. Wenzel, M. Pinzani, M.U. Dianzani, G. Laffi, P. Gentilini, J. Investig. Med. 47 (1) (1999) 66– 75. [24] Y.L. Chang, J.J. Shen, B.S. Wung, J.J. Cheng, D.L. Wang, Mol. Pharmacol. 60 (3) (2001) 507–513. [25] Y.L. Chen, K.F. Lin, M.S. Shiao, Y.T. Chen, C.Y. Hong, S.J. Lin, Basic Res. Cardiol. 96 (4) (2001) 353–363. [26] A. Momoi, K. Murao, H. Imachi, T. Ishida, W.M. Cao, M. Sato, J. Takahara, Chest 120 (4) (2001) 1293–1300. [27] N. Abbas, L.P. Zou, S.H. Pelidou, B. Winblad, J. Zhu, Autoimmunity 32 (2) (2000) 93–99. [28] S. Delanian, Br. J. Radiol. 71 (848) (1998) 892–894.

C.-S. Lin et al. / Biochemical and Biophysical Research Communications 295 (2002) 1014–1019 [29] M. Fischer, J. Wohlrab, W. Marsch, Eur. J. Dermatol. 11 (1) (2001) 38–40. [30] M.M. de Souza, L.M. Silva, A.A. Barbosa Jr., I.R. de Oliveira, R. Parana, Z.A. Andrade, Braz. J. Med. Biol. Res. 33 (11) (2000) 1329–1334. [31] J. Wu, M.A. Zern, J. Gastroenterol. 35 (9) (2000) 665–672.

1019

[32] C.C. Fang, C.J. Yen, Y.M. Chen, R.S. Shyu, T.J. Tsai, P.H. Lee, B.S. Hsieh, Kidney Int. 57 (6) (2000) 2626–2633. [33] T. Krakauer, J. Leukoc. Biol. 66 (1) (1999) 158–164. [34] T. Krakauer, Immunopharmacology 46 (3) (2000) 253–261. [35] Y.M. Chen, C.T. Chien, M.I. Hu-Tsai, K.D. Wu, C.C. Tsai, M.S. Wu, T.J. Tsai, Kidney Int. 56 (3) (1999) 932–943.