- Email: [email protected]
EXPRESSION OF HEPATOCYTE GROWTH FACTOR AND ITS RECEPTOR C-MET IN ACQUIRED RENAL CYSTIC DISEASE ASSOCIATED WITH RENAL CELL CARCINOMA
0022-5347/04/1716-2166/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 171, 2166 –2170, June 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000124263.51906.b9
EXPRESSION OF HEPATOCYTE GROWTH FACTOR AND ITS RECEPTOR C-MET IN ACQUIRED RENAL CYSTIC DISEASE ASSOCIATED WITH RENAL CELL CARCINOMA RYUICHIRO KONDA,* HIROSHI SATO, FUMIHIKO HATAFUKU, TATSURU NOZAWA, NAOMASA IORITANI AND TOMOAKI FUJIOKA From the Department of Urology, Iwate Medical University School of Medicine, Morioka, Department of Urology, Sendai Social Insurance Hospital, Sendai and Department of Nephrology, Endocrinology and Hypertension, Tohoku University School of Medicine, Sendai, Japan
ABSTRACT
Purpose: The development of renal cell carcinoma (RCC) is a critical problem in acquired renal cystic disease (ARCD) in patients with end stage renal disease, particularly males on chronic hemodialysis. We examined the expression of hepatcyte growth factor (HGF) and its receptor c-met in ARCD associated with RCC to elucidate the role of the HGF/c-met pathway in renal cyst development and subsequent tumor transformation. Materials and Methods: Immunohistochemical expression of HGF, c-met and Bcl-2 was examined in 15 normal tissue samples of kidneys obtained from nephrectomy for localized RCC and 19 with ARCD associated with RCC from 16 patients on dialysis. The expression of HGF mRNA was examined by reverse transcriptase-polymerase chain reaction. Results: In normal kidneys immunostaining for HGF was scarcely detected in renal tubular cells. Bcl-2 and c-met immunostaining was observed mainly in distal tubular cells. In nontumor areas of kidneys with ARCD associated RCC the expression of HGF and c-met was up-regulated in tubular and the cyst epithelial cells. Hyperplastic cysts with multilayer epithelium and micropapillary projection were the predominant cysts stained with HGF and c-met. Immunostaining for HGF and c-met was also detected in RCC regions. The Bcl-2 immunostaining pattern was similar to that of HGF and c-met in nontumor and RCC regions. On reverse transcriptasepolymerase chain reaction HGF mRNA expression was up-regulated in nontumor areas and RCC regions in ARCD. Conclusions: These results suggest that up-regulation of HGF and its receptor c-met may be involved in renal cyst formation and subsequent tumor transformation in patients with end stage renal disease. Increased Bcl-2 expression may promote this process through the inhibition of apoptosis. KEY WORDS: kidney; carcinoma, renal cell; kidney failure; hepatocyte growth factor; cysts
A high prevalence of renal cell carcinoma (RCC) is the most serious complication in patients with end stage renal failure (ESRF).1, 2 In recent reports ultrasound detected RCC in 3.8% of patients with end stage renal disease and pathological examination identified RCC in 4.2% of those who underwent ipsilateral native nephrectomy at transplantation, which represents a 100-fold increase in prevalence compared with the general population.3, 4 The prevalence and severity of acquired renal cystic disease (ARCD) increase with the duration of dialysis. ARCD occurs in 20% of patients on dialysis for 1 to 3 years compared with greater than 90% on dialysis for 5 to 10 years, while the rate approaches 100% after 10 years.2 Lesions of ARCD are suggested to predispose to RCC development in dialysis and transplant cases, and the prevalence of RCC also increases with dialysis duration.1, 2, 5 Although the prevalence of ARCD is only slightly higher in men (male-to-female ratio 2.9:1), ARCD associated RCC is found predominantly in men (male-tofemale ratio 7:1 compared with 2:1 in the general population).1 Cystic formation and subsequent tumor transformation are thought to be a continuous process with evolving
phenotypic expression from a simple cyst and epithelial hyperplasia to tumor eventually.1 Multiple factors may be attributed to this continuous process, including tubular cell hypertrophy and hyperplasia secondary to the loss of functioning renal tissue, uremic metabolites, growth factors and sex hormones.1, 2, 6 However, little information is available on factors that have a principal role in renal cystic development and subsequent tumor transformation in patients with ESRF. Hepatocyte growth factor (HGF) is a multifunctional cytokine that exerts renotropic action in various types of renal damage through modulating renal regeneration and compensatory renal growth (tubular cell proliferation and hypertrophy), and inhibiting apoptosis.7, 8 A decrease in the renal expression of HGF accelerates the progression of renal damage and the supplementation of exogenous of HGF has preventive and therapeutic effects in cases of acute and chronic renal failure, and fibrosis.7, 8 In contrast, mice with over expression of this growth factor have renal failure through tubular cystic dilatation and progressive glumerulosclerosis.9 Furthermore, diverse tumorigenesis with aberrant morphogenesis has been noted in HGF transgenic mice.10 The over expression of HGF and its receptor c-met has been observed in renal cell carcinoma and it has been suggested that it is involved in the onset and progression of renal cell carcinomas.11–13 Since increased serum HGF is found in patients with
Accepted for publication January 2, 2004. Study received human ethics review committee approval. * Correspondence: Department of Urology, Iwate Medical University School of Medicine, 19-1 Uchimaru, Morioka, 020-8505, Japan (telephone: ⫹81-19-651-5111; FAX: ⫹81-19-623-1527; e-mail: [email protected]). 2166
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
chronic renal failure and higher serum HGF is associated with a longer duration of dialysis,14 we speculated that in those with ESRF a decreased functional nephron mass induces prolonged stimulation of tubular cell proliferation and hypertrophy by up-regulating HGF and its receptor c-met, resulting in renal cyst formation and subsequent tumor transformation. We performed this study to elucidate our hypothesis that the HGF/c-met pathway has a critical role in the development of acquired cysts and neoplasms in patients with end stage renal disease. MATERIALS AND METHODS
Specimens. A total of 19 renal specimens were obtained from 14 male and 2 female patients 29 to 73 years old (average age 53.6) on dialysis who underwent unilateral (13) or bilateral (3) nephrectomy due to ARCD associated with RCC from 2001 to 2002 (table 1). ARCD was defined as at least 3 cysts per kidney on ultrasonography and computerized tomography. Dialysis duration was 2 to 28 years (average 14.3) and longer than 10 years in all except 1 case. No cases had distant metastasis on computerized tomography, magnetic resonance imaging and bone scintigraphy before surgery. All patients were tumor-free 12 to 26 months after surgery. The control group consisted of 9 males and 6 females 23 to 73 years old (average age 55) with localized conventional RCC of the clear cell type. Normal renal tissue samples were obtained from 15 nephrectomized kidneys from the control subjects. For histological and immunohistochemical examinations renal tissue samples were fixed in 10% buffered formalin, embedded in paraffin, and cut into 3 m sections. Sections were stained with hematoxylin and eosin, periodic acid-Schiff and azan-Mallory triple stain for routine histological examination. For RNA extraction tissue samples were immersed in RNA stabilization reagent and stored at 4C until use. On histological examination the predominant type of RCC in ARCD associated RCC cases was the papillary subtype (12 of 19), while the conventional subtype that is predominant in sporadic RCC was found in only 7 (clear cell in 2 and granular cell in 5) (table 1). Immunohistochemistry. Immunostaining of HGF, c-met and Bcl-2 was performed using antihuman HGF ␣ polyclonal antibodies that recognize activated HGF (IBL, Fujioka, Gunma, Japan), antihuman c-met polyclonal antibodies and mouse antihuman Bcl-2 monoclonal antibody. Sections (3 m) from paraffin embedded specimens were deparaffinized in xylene, rehydrated in graded ethanol and immersed in 100% methanol with 0.3% hydrogen peroxide to block endogenous peroxidase. For HGF and c-met immunostaining the sections were treated with pepsin for 15 minutes at 25C. For Bcl-2 immunostaining the sections were immersed in a solution for antigen retrieval and autoclaved for 15 minutes. After treating with 10% normal goat serum the sections were incubated with primary antibodies overnight at 4C. After washing with 0.05 mol/l tris-hydrochloric acid buffered saline containing 0.1% Tween 20 the sections were incubated with peroxidase conjugated goat antirabbit or antimouse Ig (Envision⫹, DakoCytomation, Kyoto, Japan) for 30 minutes at room temperature. Peroxidase activity was detected by incu-
TABLE 1. Characteristics of 16 patients enrolled in this study Features No. males (%) No. females (%) Mean age at operation ⫾ SD Mean dialysis duration ⫾ SD (yrs) No. histological type (%): Conventional* Papillary * Clear cell in 2 and granular cell in 5 patients.
14 (88) 2 (13) 53.6 ⫾ 13.1 14.3 ⫾ 6.5 19 7 (37) 12 (63)
2167
bation in 3,3⬘-diaminobenzidine teterahydrochloride solution. Sections were then counterstained with hematoxylin. Negative control slides stained with Ig fraction from normal rabbit serum or normal mouse IgG instead of primary antibody showed no immunoreactivity. Analysis of immunostaining was performed according to a previous report.15 Staining intensity was semiquantitatively scored as negative, weak, moderate and strong. The distribution of immunostaining in tubular, RCC and cyst epithelial cells was graded as negative, focal (10% or less), regional (11% to 50%) or diffuse (greater than 50%). The staining patterns diffuse and weak, focal to diffuse and moderate, and focal and strong were considered positive (⫹), while regional or diffuse and strong was considered intense (⫹⫹). In the acquired cyst the staining pattern within individual cysts was scored and the percent of cysts showing negative, positive or intense epithelium staining was calculated by the formula, percent of positively (negatively or intensely) stained cysts ⫽ number of positively (negatively or intensely) stained cysts/total number of cyst ⫻ 100 (%). Staining results were assessed independently by a urologist and a nephrologist blinded to clinical and pathological findings. In ARCD cases the results of immunohistochemical examination in proximal and distal tubules are not shown separately because identifying these 2 tubules was difficult on routine histological examination. Reverse transcriptase-polymerase chain reaction (RT-PCR). RNA was extracted from 3 normal kidneys, and from the nontumor and RCC regions of 2 kidneys with ARCD associated RCC using an RNeasy Mini Kit (Qiagen, Tokyo, Japan). The RT reaction was performed using oligo(deoxythymidine)12-18, SuperScript II (Gibco BRL, Gaithersburg, Maryland), at 42C for 50 minutes. PCR was then performed at 96C for 2 minutes and 35 cycles (at 94C for 30 seconds, 64C for 30 seconds and 72C for 60 seconds), followed by 72C for 4 minutes using the corresponding primers and Taq Gold DNA polymerase (Applied Biosystems, Tokyo, Japan). PCR products were electrophoresed in 2% agarose gel for 30 minutes, stained with ethidium bromide and examined under ultraviolet light. The sequences of primers used for HGF (human HGF cDNA) for 1071 to 1090 were 5⬘-GAGCATGACATGACTCCTGA-3⬘ and for 1335 to 1354 they were 5⬘-GGTTCCCAGAAGATATGACG-3⬘. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the internal control. Statistical analysis. Data are presented as the mean ⫾ SD. Correlations between the intensity of HGF and c-met staining, and the duration of dialysis was examined by the Spearman test. Statistical analysis of the results was performed by Fisher’s PLSD test with p ⬍0.05 considered significant. StatView for Macintosh (SAS Institute, Inc., Cary, North Carolina) was used for statistical analysis. RESULTS
Histological findings. The average number of acquired cysts found in 19 ARCD cases was 26.3 ⫾ 18.1 (range 10 to 72). We observed about 500 acquired cysts in which to examine the immunoreactive expression of HGF, c-met and Bcl-2. Acquired cysts were classified into 3 subtypes according to cyst epithelium morphology, namely flat, cuboidal and hyperplastic, as previously reported.6 In 19 ARCD cases acquired cysts lined by cuboidal epithelium predominated, accounting for about 40%. Flat and hyperplastic epithelial cysts shared similar ratios. Hyperplastic cysts had multilayer epithelium and micropapillary projection. In some of these cysts papillary growth of tumor into the cystic lumen was observed. Immunohistochemical staining for HGF. In 15 normal kidneys immunopositive HGF was detected in the endothelial cells of arteries and arterioles but no or faint staining was found in tubular cells (table 2 and fig. 1, a). In ARCD kidneys immunostaining for HGF was noted in tubular cells (table 2
2168
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
TABLE 2. Immunohistochemical staining for HGF, c-met and Bcl-2 in 15 normal kidneys and 19 with ARCD associated with RCC
Normal kidney tubule: Proximal Distal ARCD ⫹ RCC: Nondilated tubule Dilated tubule Papillary RCC Conventional RCC Overall
No. HGF (%)
No. Kidneys
⫺
15 15
15 (100) 15 (100)
19 19 12 7 19
0 0 0 0 0
⫹ 0 0 7 (37) 4 (21) 2 (17) 2 (29) 4 (21)
No. c-met (%) ⫹⫹ 0 0 12 (63) 15 (79) 10 (83) 5 (71) 15 (79)
⫺
⫹
11 (73) 0
4 (27) 15 (100)
0 0 0 0 0
10 9 4 3 7
(53) (47) (33) (43) (37)
No. Bcl-2 (%) ⫹⫹ 0 0 9 (47) 10 (53) 8 (67) 4 (57) 12 (63)
⫺
⫹
15 (100) 0
0 15 (100)
0 0
12 11 9 4 13
6 (32) 7 (37) 3 (25) 2 (29) 5 (26)
1 (5) 1 (5) 0 1 (14) 1 (5)
(63) (58) (75) (57) (68)
⫹⫹
TABLE 3. Negative, positive and intense immunostaining for HGF, c-met and Bcl-2 in flat, cuboidal and hyperplastic epithelial cells lining cysts Cyst Epithelium HGF: Flat Cuboidal Hyperplastic c-met: Flat Cuboidal Hyperplastic Bcl-2: Flat Cuboidal Hyperplastic
FIG. 1. Immunohistochemical staining for HGF (a to c), c-met (d and e) and Bcl-2 (f). HGF staining (brown areas) in normal kidney shows immunoreactivity only in endothelial cells of arterioles and not in tubular cells (a). HGF staining in kidney with ARCD associated RCC (b and c). Increased HGF staining was noted in tubular cells and cystic epithelial cells in nontumor area (b) and in tumor cells in RCC region (c). RCC projected into hyperplastic cyst lumen. c-met staining in normal kidney shows immunoreactivity in distal tubular cells (d), while in kidney with ARCD associated RCC there was increased c-met staining in cystic epithelial cells. Bcl-2 staining in kidney with ARCD associated RCC showed positive staining in tubular and cystic epithelial cells (f).
and fig. 1, b). Intense staining of these cells was observed in 12 of 19 ARCD cases (63%). Dilated tubules were scattered in ARCD cases and these cells also showed immunostaining for HGF with intense staining observed in 15 of 19 ARCD cases (79%) (table 2 and fig. 1, b). HGF staining was detected in all RCC regions with intense staining noted in 10 of 12 papillary (83%) and 5 of 7 conventional (71%) RCC cases, all of the granular type (table 2 and fig. 1, c). Immunostaining for HGF was detected in all hyperplastic cysts and in 99.5% of cuboidal epithelial cysts (table 3). Intense staining for HGF in cystic epithelial cells was demonstrated in 88.3% of cuboidal and 94.8% of hyperplastic cysts. In contrast, 20.7% of flat epithelial cysts showed negative staining for HGF and intense staining was detected in 12.7% of flat cysts. Immunoreactive expression of HGF was significantly higher in cuboidal and hyperplastic cysts than
%⫺
%⫹
% ⫹⫹
20.7 0.5 0
66.7 11.2 5.2
12.7 88.3 94.8
35.9 2.2 0
53.4 24.9 19.2
10.7 72.9 80.8
56 6.6 3.3
40.8 36.6 37
3.2 56.8 59.8
in flat cysts (p ⬍0.0001). To evaluate the correlation between the intensity of HGF staining and the duration of dialysis we calculated the percent of intense HGF stained cysts (number of intensely stained cysts/total number of cysts ⫻ 100). No significant correlation was found between the 2 (Spearman test p ⬎0.1). Immunohistochemical staining of c-met. In normal kidneys distal tubular cells showed immunostaining for c-met in all samples (table 2 and fig. 1, d), while proximal tubules also stained positive in 4 of the 15 samples (27%) (table 2). In ARCD cases c-met immunostaining was noted in tubular cells. Intense staining in nondilated and dilated tubular cells was observed in 9 of 19 (47%) and 10 of 19 (53%) ARCD cases, respectively (table 2). All RCC regions showed immunostaining for c-met (table 2). Intense staining was detected in 8 of 12 papillary (67%) and 4 of 7 conventional (57%) RCC cases, all of the granular type. In cyst epithelium immunostaining for c-met was observed in 64.1% of flat, 97.8% of cuboidal and 100% of hyperplastic cysts (table 3 and fig. 1, e). Intense staining for c-met in cyst epithelial cells was noted in 10.7% of flat, 72.9% of cuboidal and 80.8% of hyperplastic cysts. Immunoreactive expression of c-met was significantly higher in cuboidal and hyperplastic cysts than in flat cysts (p ⬍0.0001). We calculated the percent of intense c-met stained cysts using a formula similar to that for HGF and analyzed the correlation with the duration of dialysis. There was no significant correlation between the percent of intense c-met stained cysts and the duration of dialysis (p ⬎0.1). Immunostaining for Bcl-2. In normal kidneys distal tubular cells showed immunostaining for Bcl-2 (table 2). No or faint staining was noted in proximal tubular cells. Immunostaining for Bcl-2 was detected in tubular cells of all ARCD cases except 1 (table 2 and fig. 1, f). Intense staining was observed in nondilated and dilated tubular cells in 6 of 19 (32%) and 7 of 19 (37%) ARCD cases, respectively. Immunostaining for Bcl-2 was noted in all RCC regions except 1 (conventional RCC, clear cell type) (table 2). Intense staining was noted in 3 of 12 papillary (25%) and 2 of 7 conventional (29%) RCC cases, all of the granular type. Negative staining for Bcl-2 was observed in 56% of flat,
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
6.6% of cuboidal and 3.3% of hyperplastic cysts (table 3). In contrast, 3.2% of flat, 56.8% of cuboidal and 59.8% of hyperplastic cysts showed intense staining for Bcl-2. Immunoreactive expression of Bcl-2 was significantly higher in cuboidal and hyperplastic cysts than in flat cysts (p ⬍0.0001). RT-PCR analysis. HGF mRNA expression was examined in the nontumor and RCC regions of 2 ARCD associated RCC samples and in 3 normal kidney samples by RT-PCR. PCR products of HGF and GAPDH were recognized at 284 and 226 bp, respectively. HGF mRNA was detected in nontumor and RCC regions of ARCD associated RCC cases (fig. 2). In contrast, no or low expression of HGF mRNA was detected in normal kidney samples (fig. 2). DISCUSSION
Histological examination revealed 3 types of cysts lined by flat, cuboidal and hyperplastic epithelium in ARCD. Hyperplasia of tubular cells is considered to be the most important cause of cyst development and subsequent tumor transformation. Increased proliferative activity expressed as a percent of proliferating cell nuclear antigen positive cells has been demonstrated in hyperplastic cysts in ARCD.6 In this study immunohistochemical examinations revealed increased expression of HGF and its receptor c-met in tubular cells and cystic epithelial cells in ARCD cases. In cystic epithelial cells immunostaining for these proteins was significantly higher in cysts lined with cuboidal and hyperplastic epithelium than in those lined with flat epithelium. Bcl-2, an anti-apoptotic protein, was also expressed predominantly in cysts lined with cuboidal and hyperplastic epithelium. RCC grew into the cystic lumens of some hyperplastic cysts. Previous studies have shown that most acquired cysts are derived from the proximal tubules.1, 2 In another study hyperplastic cysts in ARCD cases stained positive with the lectin Tetragonolobus purpureas (proximal tubule marker) but not with antibodies to epithelial membrane antigen (marker of distal nephron, distal convoluted tubule and collecting duct).6 In contrast, flat and cuboidal cyst epithelium stained with one of these tubular markers.6 The epithelium of hyperplastic cysts have higher proliferative activity than other types of cysts.6 In autosomal dominant polycystic kidney disease the HGF concentration in cyst fluids was higher in proximal than in distal cysts.16 In the current study cysts lined with hyperplastic and cuboidal epithelium showed higher staining for HGF, c-met and Bcl-2 compared with flat epithelial cysts, although we did not distinguish whether the cysts originated from proximal or distal tubules. These observations suggest that not only hyperplastic, but also cuboidal epithelial cysts have a higher potential for proliferation and tumorigenicity than flat epithelial cyst. Cysts lined with cuboidal epithelium
2169
but not those with flat epithelium may mainly evolve to hyperplastic cysts. This study revealed that HGF mRNA and protein as well as c-met protein were up-regulated in nontumor and RCC regions in ARCD associated RCC. Tubular cells and cystic epithelial cells were the predominant HGF expressing cells in nontumor areas of ARCD. In the kidney stromal cells, such as macrophages, endothelial cells and mesangial cells, ordinarily express HGF mRNA and protein.7, 8 Tubular cell expression of HGF protein or mRNA has been noted in obstructed kidneys after ureteral obstruction and in rejecting kidneys after renal transplantation.17, 18 We speculate that HGF produced locally exhibits its biological effects via paracrine and autocrine related mechanisms, although we did not identify the cells expressing HGF mRNA in ARCD. Expression of HGF and c-met mRNA and/or protein in sporadic RCC has been reported and the HGF/c-met pathway is thought to be involved in the onset and progression of RCC.11–13 The current study revealed over expression of HGF mRNA and protein as well as c-met protein in ARCD associated RCC. The HGF/c-met pathway may contribute not only to tubular hyperplasia and cyst formation, but also to tumor transformation and RCC development. Moreover, immunostaining for Bcl-2 was noted in most RCC regions. A previous study has shown immunostaining for Bcl-2 in 70% of RCC cases, suggesting that Bcl-2 over expression has a role in tumorigenesis.19 In ARCD Bcl-2 over expression on tubular, cystic epithelial and RCC cells may promote the action of the HGF/c-met pathway for the development of acquired cysts and tumor, and the growth of associated RCC by inhibiting programmed cell death. As previously reported, ARCD associated RCC is found predominantly in males, and the prevalence of ARCD and associated RCC increases with the duration of dialysis.1–5 In this study 14 of the 16 patients (88%) were males. The duration of dialysis was 2 to 28 years (average 15.9) and it was more than 10 years in all patients except 1, who was on dialysis for 2 years. Concerning male predominance, sexual hormones are one of the candidates for promoting the development of acquired cysts and neoplasms.1, 2 Further study is needed to elucidate the contribution of sexual hormones to the process of cyst formation and tumor transformation in patients on dialysis. The duration of chronic renal failure or dialysis is thought to be the principal factor in the development of acquired cysts.1, 3, 18 Since serum HGF increases with the progression of renal failure and it is higher in patients on dialysis for a longer duration,14 prolonged stimulation of renal tubular proliferation by endogenous HGF may have a critical role in the continuous process of cyst formation and subsequent tumor transformation in patients with ESRF. However, we did not find any correlation between the intensity of HGF and c-met staining, and the duration of dialysis. CONCLUSIONS
Our observations suggest that the HGF/c-met pathway has a critical role in the progression of acquired cyst formation and subsequent tumor transformation in patients with ESRF. Over expressed Bcl-2 may promote this process by inhibiting apoptosis. Further study is necessary to elucidate the reason why ARCD and associated RCC are predominant in males. GAPDH primers were obtained from Applied Biosystems Japan, Tokyo, Japan. REFERENCES
FIG. 2. RT-PCR results for HGF in normal kidney and kidney with ARCD associated RCC. PCR product for HGF and GAPDH were electrophoretically separated by 2% agarose gel. Bands show HGF mRNA at 284 bp and GAPDH mRNA at 226 bp. HGF mRNA expression was observed in nontumor and RCC regions in ARCD associated RCC but not in normal kidney.
1. Truong, L. D., Krishnan, B., Cao, J. T., Barrios, R. and Suki, W. N.: Renal neoplasm in acquired cystic kidney disease. Am J Kidney Dis, 26: 1, 1995 2. Ishikawa, I.: Acquired cystic disease: mechanisms and manifestations. Semin Nephrol, 11: 671, 1991
2170
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
3. Gulanikar, A. C., Daily, P. P., Kilambi, N. K., Hamrick-Turner, J. E. and Butkus, D. E.: Prospective pretransplant ultrasound screening in 206 patients for acquired renal cysts and renal cell carcinoma. Transplantation, 66: 1669, 1998 4. Denton, M. D., Magee, C. C., Ovuworie, C., Mauiyyedi, S., Pascual, M., Colvin, R. B. et al: Prevalence of renal cell carcinoma in patients with ESRD pre-transplantation: a pathologic analysis. Kidney Int, 61: 2201, 2002 5. Hoshida, Y., Nakanishi, H., Shin, M., Satoh, T., Hanai, J. and Aozasa, K.: Renal neoplasias in patients receiving dialysis and renal transplantation: clinico-pathological features and p53 gene mutations. Transplantation, 68: 385, 1999 6. Nadasdy, T., Laszik, Z., Lajoie, G., Blick, K. E., Wheeler, D. E. and Silva, F. G.: Proliferative activity of cyst epithelium in human renal cystic diseases. J Am Soc Nephrol, 5: 1462, 1995 7. Vargas, G. A., Hoeflich, A. and Jehle, P. M.: Hepatocyte growth factor in renal failure: promise and reality. Kidney Int, 57: 1426, 2000 8. Matsumoto, K. and Nakamura, T.: Hepatocyte growth factor: renotropic role and potential therapeutics for renal diseases. Kidney Int, 59: 2023, 2001 9. Takayama, H., LaRochelle, W. J., Sabnis, S. G., Otsuka, T. and Merlino, G.: Renal tubular hyperplasia, polycystic disease, and glomerulosclerosis in transgenic mice overexpressing hepatocyte growth factor/scatter factor. Lab Invest, 77: 131, 1997 10. Takayama, H., LaRochelle, W. J., Sharp, R., Otsuka, T., Kriebel, P., Anver, M. et al: Diverse tumorigenesis associated with aberrant development in mice overexpressing hepatocyte growth factor/scatter factor. Proc Natl Acad Sci USA, 94: 701, 1997
11. Natali, P. G., Prat, M., Nicotra, M. R., Bigotti, A., Olivero, M., Comoglio, P. M. et al: Overexpression of the met/HGF receptor in renal cell carcinomas. Int J Cancer, 69: 212, 1996 12. Pisters, L. L., El-Naggar, A. K., Luo, W., Malpica, A. and Lin, S. H.: C-met proto-oncogene expression in benign and malignant human renal tissues. J Urol, 158: 724, 1997 13. Horie, S., Aruga, S., Kawamata, H., Okui, N., Kakizoe, T. and Kitamura, T.: Biological role of HGF/MET pathway in renal cell carcinoma. J Urol, 161: 990, 1999 14. Sugimura, K., Kim, T., Goto, T., Kasai, S., Takemoto, Y., Matsuda, J. et al: Serum hepatocyte growth factor levels in patients with chronic renal failure. Nephron, 70: 324, 1995 15. Miyata, Y., Koga, S., Kanda, S., Nishikido, M., Hayashi, T. and Kanetake, H.: Expression of cyclooxygenase-2 in renal cell carcinoma: correlation with tumor cell proliferation, apoptosis, angiogenesis, expression of matrix metalloproteinase-2, and survival. Clin Cancer Res, 9: 1741, 2003 16. Horie, S., Higashihara, E., Nutahara, K., Mikami, Y., Okubo, A., Kano, M. et al: Mediation of renal cyst formation by hepatocyte growth factor. Lancet, 344: 789, 1994 17. Mizuno, S., Matsumoto, K. and Nakamura, T.: Hepatocyte growth factor suppresses interstitial fibrosis in a mouse model of obstructive nephropathy. Kidney Int, 59: 1304, 2001 18. Yamaguchi, K., Nalesnik, M. A. and Michalopoulos, G. K.: Expression of HGF mRNA in human rejecting kidney as evidenced by in situ hybridization. Urol Res, 24: 349, 1996 19. Huang, A., Fone, P. D., Gandour-Edwards, R., Devere White, R. W. and Low, R. K.: Immunohistochemical analysis of BCL-2 protein expression in renal cell carcinoma. J Urol, 162: 610, 1999
Vol. 171, 2166 –2170, June 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000124263.51906.b9
EXPRESSION OF HEPATOCYTE GROWTH FACTOR AND ITS RECEPTOR C-MET IN ACQUIRED RENAL CYSTIC DISEASE ASSOCIATED WITH RENAL CELL CARCINOMA RYUICHIRO KONDA,* HIROSHI SATO, FUMIHIKO HATAFUKU, TATSURU NOZAWA, NAOMASA IORITANI AND TOMOAKI FUJIOKA From the Department of Urology, Iwate Medical University School of Medicine, Morioka, Department of Urology, Sendai Social Insurance Hospital, Sendai and Department of Nephrology, Endocrinology and Hypertension, Tohoku University School of Medicine, Sendai, Japan
ABSTRACT
Purpose: The development of renal cell carcinoma (RCC) is a critical problem in acquired renal cystic disease (ARCD) in patients with end stage renal disease, particularly males on chronic hemodialysis. We examined the expression of hepatcyte growth factor (HGF) and its receptor c-met in ARCD associated with RCC to elucidate the role of the HGF/c-met pathway in renal cyst development and subsequent tumor transformation. Materials and Methods: Immunohistochemical expression of HGF, c-met and Bcl-2 was examined in 15 normal tissue samples of kidneys obtained from nephrectomy for localized RCC and 19 with ARCD associated with RCC from 16 patients on dialysis. The expression of HGF mRNA was examined by reverse transcriptase-polymerase chain reaction. Results: In normal kidneys immunostaining for HGF was scarcely detected in renal tubular cells. Bcl-2 and c-met immunostaining was observed mainly in distal tubular cells. In nontumor areas of kidneys with ARCD associated RCC the expression of HGF and c-met was up-regulated in tubular and the cyst epithelial cells. Hyperplastic cysts with multilayer epithelium and micropapillary projection were the predominant cysts stained with HGF and c-met. Immunostaining for HGF and c-met was also detected in RCC regions. The Bcl-2 immunostaining pattern was similar to that of HGF and c-met in nontumor and RCC regions. On reverse transcriptasepolymerase chain reaction HGF mRNA expression was up-regulated in nontumor areas and RCC regions in ARCD. Conclusions: These results suggest that up-regulation of HGF and its receptor c-met may be involved in renal cyst formation and subsequent tumor transformation in patients with end stage renal disease. Increased Bcl-2 expression may promote this process through the inhibition of apoptosis. KEY WORDS: kidney; carcinoma, renal cell; kidney failure; hepatocyte growth factor; cysts
A high prevalence of renal cell carcinoma (RCC) is the most serious complication in patients with end stage renal failure (ESRF).1, 2 In recent reports ultrasound detected RCC in 3.8% of patients with end stage renal disease and pathological examination identified RCC in 4.2% of those who underwent ipsilateral native nephrectomy at transplantation, which represents a 100-fold increase in prevalence compared with the general population.3, 4 The prevalence and severity of acquired renal cystic disease (ARCD) increase with the duration of dialysis. ARCD occurs in 20% of patients on dialysis for 1 to 3 years compared with greater than 90% on dialysis for 5 to 10 years, while the rate approaches 100% after 10 years.2 Lesions of ARCD are suggested to predispose to RCC development in dialysis and transplant cases, and the prevalence of RCC also increases with dialysis duration.1, 2, 5 Although the prevalence of ARCD is only slightly higher in men (male-to-female ratio 2.9:1), ARCD associated RCC is found predominantly in men (male-tofemale ratio 7:1 compared with 2:1 in the general population).1 Cystic formation and subsequent tumor transformation are thought to be a continuous process with evolving
phenotypic expression from a simple cyst and epithelial hyperplasia to tumor eventually.1 Multiple factors may be attributed to this continuous process, including tubular cell hypertrophy and hyperplasia secondary to the loss of functioning renal tissue, uremic metabolites, growth factors and sex hormones.1, 2, 6 However, little information is available on factors that have a principal role in renal cystic development and subsequent tumor transformation in patients with ESRF. Hepatocyte growth factor (HGF) is a multifunctional cytokine that exerts renotropic action in various types of renal damage through modulating renal regeneration and compensatory renal growth (tubular cell proliferation and hypertrophy), and inhibiting apoptosis.7, 8 A decrease in the renal expression of HGF accelerates the progression of renal damage and the supplementation of exogenous of HGF has preventive and therapeutic effects in cases of acute and chronic renal failure, and fibrosis.7, 8 In contrast, mice with over expression of this growth factor have renal failure through tubular cystic dilatation and progressive glumerulosclerosis.9 Furthermore, diverse tumorigenesis with aberrant morphogenesis has been noted in HGF transgenic mice.10 The over expression of HGF and its receptor c-met has been observed in renal cell carcinoma and it has been suggested that it is involved in the onset and progression of renal cell carcinomas.11–13 Since increased serum HGF is found in patients with
Accepted for publication January 2, 2004. Study received human ethics review committee approval. * Correspondence: Department of Urology, Iwate Medical University School of Medicine, 19-1 Uchimaru, Morioka, 020-8505, Japan (telephone: ⫹81-19-651-5111; FAX: ⫹81-19-623-1527; e-mail: [email protected]). 2166
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
chronic renal failure and higher serum HGF is associated with a longer duration of dialysis,14 we speculated that in those with ESRF a decreased functional nephron mass induces prolonged stimulation of tubular cell proliferation and hypertrophy by up-regulating HGF and its receptor c-met, resulting in renal cyst formation and subsequent tumor transformation. We performed this study to elucidate our hypothesis that the HGF/c-met pathway has a critical role in the development of acquired cysts and neoplasms in patients with end stage renal disease. MATERIALS AND METHODS
Specimens. A total of 19 renal specimens were obtained from 14 male and 2 female patients 29 to 73 years old (average age 53.6) on dialysis who underwent unilateral (13) or bilateral (3) nephrectomy due to ARCD associated with RCC from 2001 to 2002 (table 1). ARCD was defined as at least 3 cysts per kidney on ultrasonography and computerized tomography. Dialysis duration was 2 to 28 years (average 14.3) and longer than 10 years in all except 1 case. No cases had distant metastasis on computerized tomography, magnetic resonance imaging and bone scintigraphy before surgery. All patients were tumor-free 12 to 26 months after surgery. The control group consisted of 9 males and 6 females 23 to 73 years old (average age 55) with localized conventional RCC of the clear cell type. Normal renal tissue samples were obtained from 15 nephrectomized kidneys from the control subjects. For histological and immunohistochemical examinations renal tissue samples were fixed in 10% buffered formalin, embedded in paraffin, and cut into 3 m sections. Sections were stained with hematoxylin and eosin, periodic acid-Schiff and azan-Mallory triple stain for routine histological examination. For RNA extraction tissue samples were immersed in RNA stabilization reagent and stored at 4C until use. On histological examination the predominant type of RCC in ARCD associated RCC cases was the papillary subtype (12 of 19), while the conventional subtype that is predominant in sporadic RCC was found in only 7 (clear cell in 2 and granular cell in 5) (table 1). Immunohistochemistry. Immunostaining of HGF, c-met and Bcl-2 was performed using antihuman HGF ␣ polyclonal antibodies that recognize activated HGF (IBL, Fujioka, Gunma, Japan), antihuman c-met polyclonal antibodies and mouse antihuman Bcl-2 monoclonal antibody. Sections (3 m) from paraffin embedded specimens were deparaffinized in xylene, rehydrated in graded ethanol and immersed in 100% methanol with 0.3% hydrogen peroxide to block endogenous peroxidase. For HGF and c-met immunostaining the sections were treated with pepsin for 15 minutes at 25C. For Bcl-2 immunostaining the sections were immersed in a solution for antigen retrieval and autoclaved for 15 minutes. After treating with 10% normal goat serum the sections were incubated with primary antibodies overnight at 4C. After washing with 0.05 mol/l tris-hydrochloric acid buffered saline containing 0.1% Tween 20 the sections were incubated with peroxidase conjugated goat antirabbit or antimouse Ig (Envision⫹, DakoCytomation, Kyoto, Japan) for 30 minutes at room temperature. Peroxidase activity was detected by incu-
TABLE 1. Characteristics of 16 patients enrolled in this study Features No. males (%) No. females (%) Mean age at operation ⫾ SD Mean dialysis duration ⫾ SD (yrs) No. histological type (%): Conventional* Papillary * Clear cell in 2 and granular cell in 5 patients.
14 (88) 2 (13) 53.6 ⫾ 13.1 14.3 ⫾ 6.5 19 7 (37) 12 (63)
2167
bation in 3,3⬘-diaminobenzidine teterahydrochloride solution. Sections were then counterstained with hematoxylin. Negative control slides stained with Ig fraction from normal rabbit serum or normal mouse IgG instead of primary antibody showed no immunoreactivity. Analysis of immunostaining was performed according to a previous report.15 Staining intensity was semiquantitatively scored as negative, weak, moderate and strong. The distribution of immunostaining in tubular, RCC and cyst epithelial cells was graded as negative, focal (10% or less), regional (11% to 50%) or diffuse (greater than 50%). The staining patterns diffuse and weak, focal to diffuse and moderate, and focal and strong were considered positive (⫹), while regional or diffuse and strong was considered intense (⫹⫹). In the acquired cyst the staining pattern within individual cysts was scored and the percent of cysts showing negative, positive or intense epithelium staining was calculated by the formula, percent of positively (negatively or intensely) stained cysts ⫽ number of positively (negatively or intensely) stained cysts/total number of cyst ⫻ 100 (%). Staining results were assessed independently by a urologist and a nephrologist blinded to clinical and pathological findings. In ARCD cases the results of immunohistochemical examination in proximal and distal tubules are not shown separately because identifying these 2 tubules was difficult on routine histological examination. Reverse transcriptase-polymerase chain reaction (RT-PCR). RNA was extracted from 3 normal kidneys, and from the nontumor and RCC regions of 2 kidneys with ARCD associated RCC using an RNeasy Mini Kit (Qiagen, Tokyo, Japan). The RT reaction was performed using oligo(deoxythymidine)12-18, SuperScript II (Gibco BRL, Gaithersburg, Maryland), at 42C for 50 minutes. PCR was then performed at 96C for 2 minutes and 35 cycles (at 94C for 30 seconds, 64C for 30 seconds and 72C for 60 seconds), followed by 72C for 4 minutes using the corresponding primers and Taq Gold DNA polymerase (Applied Biosystems, Tokyo, Japan). PCR products were electrophoresed in 2% agarose gel for 30 minutes, stained with ethidium bromide and examined under ultraviolet light. The sequences of primers used for HGF (human HGF cDNA) for 1071 to 1090 were 5⬘-GAGCATGACATGACTCCTGA-3⬘ and for 1335 to 1354 they were 5⬘-GGTTCCCAGAAGATATGACG-3⬘. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the internal control. Statistical analysis. Data are presented as the mean ⫾ SD. Correlations between the intensity of HGF and c-met staining, and the duration of dialysis was examined by the Spearman test. Statistical analysis of the results was performed by Fisher’s PLSD test with p ⬍0.05 considered significant. StatView for Macintosh (SAS Institute, Inc., Cary, North Carolina) was used for statistical analysis. RESULTS
Histological findings. The average number of acquired cysts found in 19 ARCD cases was 26.3 ⫾ 18.1 (range 10 to 72). We observed about 500 acquired cysts in which to examine the immunoreactive expression of HGF, c-met and Bcl-2. Acquired cysts were classified into 3 subtypes according to cyst epithelium morphology, namely flat, cuboidal and hyperplastic, as previously reported.6 In 19 ARCD cases acquired cysts lined by cuboidal epithelium predominated, accounting for about 40%. Flat and hyperplastic epithelial cysts shared similar ratios. Hyperplastic cysts had multilayer epithelium and micropapillary projection. In some of these cysts papillary growth of tumor into the cystic lumen was observed. Immunohistochemical staining for HGF. In 15 normal kidneys immunopositive HGF was detected in the endothelial cells of arteries and arterioles but no or faint staining was found in tubular cells (table 2 and fig. 1, a). In ARCD kidneys immunostaining for HGF was noted in tubular cells (table 2
2168
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
TABLE 2. Immunohistochemical staining for HGF, c-met and Bcl-2 in 15 normal kidneys and 19 with ARCD associated with RCC
Normal kidney tubule: Proximal Distal ARCD ⫹ RCC: Nondilated tubule Dilated tubule Papillary RCC Conventional RCC Overall
No. HGF (%)
No. Kidneys
⫺
15 15
15 (100) 15 (100)
19 19 12 7 19
0 0 0 0 0
⫹ 0 0 7 (37) 4 (21) 2 (17) 2 (29) 4 (21)
No. c-met (%) ⫹⫹ 0 0 12 (63) 15 (79) 10 (83) 5 (71) 15 (79)
⫺
⫹
11 (73) 0
4 (27) 15 (100)
0 0 0 0 0
10 9 4 3 7
(53) (47) (33) (43) (37)
No. Bcl-2 (%) ⫹⫹ 0 0 9 (47) 10 (53) 8 (67) 4 (57) 12 (63)
⫺
⫹
15 (100) 0
0 15 (100)
0 0
12 11 9 4 13
6 (32) 7 (37) 3 (25) 2 (29) 5 (26)
1 (5) 1 (5) 0 1 (14) 1 (5)
(63) (58) (75) (57) (68)
⫹⫹
TABLE 3. Negative, positive and intense immunostaining for HGF, c-met and Bcl-2 in flat, cuboidal and hyperplastic epithelial cells lining cysts Cyst Epithelium HGF: Flat Cuboidal Hyperplastic c-met: Flat Cuboidal Hyperplastic Bcl-2: Flat Cuboidal Hyperplastic
FIG. 1. Immunohistochemical staining for HGF (a to c), c-met (d and e) and Bcl-2 (f). HGF staining (brown areas) in normal kidney shows immunoreactivity only in endothelial cells of arterioles and not in tubular cells (a). HGF staining in kidney with ARCD associated RCC (b and c). Increased HGF staining was noted in tubular cells and cystic epithelial cells in nontumor area (b) and in tumor cells in RCC region (c). RCC projected into hyperplastic cyst lumen. c-met staining in normal kidney shows immunoreactivity in distal tubular cells (d), while in kidney with ARCD associated RCC there was increased c-met staining in cystic epithelial cells. Bcl-2 staining in kidney with ARCD associated RCC showed positive staining in tubular and cystic epithelial cells (f).
and fig. 1, b). Intense staining of these cells was observed in 12 of 19 ARCD cases (63%). Dilated tubules were scattered in ARCD cases and these cells also showed immunostaining for HGF with intense staining observed in 15 of 19 ARCD cases (79%) (table 2 and fig. 1, b). HGF staining was detected in all RCC regions with intense staining noted in 10 of 12 papillary (83%) and 5 of 7 conventional (71%) RCC cases, all of the granular type (table 2 and fig. 1, c). Immunostaining for HGF was detected in all hyperplastic cysts and in 99.5% of cuboidal epithelial cysts (table 3). Intense staining for HGF in cystic epithelial cells was demonstrated in 88.3% of cuboidal and 94.8% of hyperplastic cysts. In contrast, 20.7% of flat epithelial cysts showed negative staining for HGF and intense staining was detected in 12.7% of flat cysts. Immunoreactive expression of HGF was significantly higher in cuboidal and hyperplastic cysts than
%⫺
%⫹
% ⫹⫹
20.7 0.5 0
66.7 11.2 5.2
12.7 88.3 94.8
35.9 2.2 0
53.4 24.9 19.2
10.7 72.9 80.8
56 6.6 3.3
40.8 36.6 37
3.2 56.8 59.8
in flat cysts (p ⬍0.0001). To evaluate the correlation between the intensity of HGF staining and the duration of dialysis we calculated the percent of intense HGF stained cysts (number of intensely stained cysts/total number of cysts ⫻ 100). No significant correlation was found between the 2 (Spearman test p ⬎0.1). Immunohistochemical staining of c-met. In normal kidneys distal tubular cells showed immunostaining for c-met in all samples (table 2 and fig. 1, d), while proximal tubules also stained positive in 4 of the 15 samples (27%) (table 2). In ARCD cases c-met immunostaining was noted in tubular cells. Intense staining in nondilated and dilated tubular cells was observed in 9 of 19 (47%) and 10 of 19 (53%) ARCD cases, respectively (table 2). All RCC regions showed immunostaining for c-met (table 2). Intense staining was detected in 8 of 12 papillary (67%) and 4 of 7 conventional (57%) RCC cases, all of the granular type. In cyst epithelium immunostaining for c-met was observed in 64.1% of flat, 97.8% of cuboidal and 100% of hyperplastic cysts (table 3 and fig. 1, e). Intense staining for c-met in cyst epithelial cells was noted in 10.7% of flat, 72.9% of cuboidal and 80.8% of hyperplastic cysts. Immunoreactive expression of c-met was significantly higher in cuboidal and hyperplastic cysts than in flat cysts (p ⬍0.0001). We calculated the percent of intense c-met stained cysts using a formula similar to that for HGF and analyzed the correlation with the duration of dialysis. There was no significant correlation between the percent of intense c-met stained cysts and the duration of dialysis (p ⬎0.1). Immunostaining for Bcl-2. In normal kidneys distal tubular cells showed immunostaining for Bcl-2 (table 2). No or faint staining was noted in proximal tubular cells. Immunostaining for Bcl-2 was detected in tubular cells of all ARCD cases except 1 (table 2 and fig. 1, f). Intense staining was observed in nondilated and dilated tubular cells in 6 of 19 (32%) and 7 of 19 (37%) ARCD cases, respectively. Immunostaining for Bcl-2 was noted in all RCC regions except 1 (conventional RCC, clear cell type) (table 2). Intense staining was noted in 3 of 12 papillary (25%) and 2 of 7 conventional (29%) RCC cases, all of the granular type. Negative staining for Bcl-2 was observed in 56% of flat,
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
6.6% of cuboidal and 3.3% of hyperplastic cysts (table 3). In contrast, 3.2% of flat, 56.8% of cuboidal and 59.8% of hyperplastic cysts showed intense staining for Bcl-2. Immunoreactive expression of Bcl-2 was significantly higher in cuboidal and hyperplastic cysts than in flat cysts (p ⬍0.0001). RT-PCR analysis. HGF mRNA expression was examined in the nontumor and RCC regions of 2 ARCD associated RCC samples and in 3 normal kidney samples by RT-PCR. PCR products of HGF and GAPDH were recognized at 284 and 226 bp, respectively. HGF mRNA was detected in nontumor and RCC regions of ARCD associated RCC cases (fig. 2). In contrast, no or low expression of HGF mRNA was detected in normal kidney samples (fig. 2). DISCUSSION
Histological examination revealed 3 types of cysts lined by flat, cuboidal and hyperplastic epithelium in ARCD. Hyperplasia of tubular cells is considered to be the most important cause of cyst development and subsequent tumor transformation. Increased proliferative activity expressed as a percent of proliferating cell nuclear antigen positive cells has been demonstrated in hyperplastic cysts in ARCD.6 In this study immunohistochemical examinations revealed increased expression of HGF and its receptor c-met in tubular cells and cystic epithelial cells in ARCD cases. In cystic epithelial cells immunostaining for these proteins was significantly higher in cysts lined with cuboidal and hyperplastic epithelium than in those lined with flat epithelium. Bcl-2, an anti-apoptotic protein, was also expressed predominantly in cysts lined with cuboidal and hyperplastic epithelium. RCC grew into the cystic lumens of some hyperplastic cysts. Previous studies have shown that most acquired cysts are derived from the proximal tubules.1, 2 In another study hyperplastic cysts in ARCD cases stained positive with the lectin Tetragonolobus purpureas (proximal tubule marker) but not with antibodies to epithelial membrane antigen (marker of distal nephron, distal convoluted tubule and collecting duct).6 In contrast, flat and cuboidal cyst epithelium stained with one of these tubular markers.6 The epithelium of hyperplastic cysts have higher proliferative activity than other types of cysts.6 In autosomal dominant polycystic kidney disease the HGF concentration in cyst fluids was higher in proximal than in distal cysts.16 In the current study cysts lined with hyperplastic and cuboidal epithelium showed higher staining for HGF, c-met and Bcl-2 compared with flat epithelial cysts, although we did not distinguish whether the cysts originated from proximal or distal tubules. These observations suggest that not only hyperplastic, but also cuboidal epithelial cysts have a higher potential for proliferation and tumorigenicity than flat epithelial cyst. Cysts lined with cuboidal epithelium
2169
but not those with flat epithelium may mainly evolve to hyperplastic cysts. This study revealed that HGF mRNA and protein as well as c-met protein were up-regulated in nontumor and RCC regions in ARCD associated RCC. Tubular cells and cystic epithelial cells were the predominant HGF expressing cells in nontumor areas of ARCD. In the kidney stromal cells, such as macrophages, endothelial cells and mesangial cells, ordinarily express HGF mRNA and protein.7, 8 Tubular cell expression of HGF protein or mRNA has been noted in obstructed kidneys after ureteral obstruction and in rejecting kidneys after renal transplantation.17, 18 We speculate that HGF produced locally exhibits its biological effects via paracrine and autocrine related mechanisms, although we did not identify the cells expressing HGF mRNA in ARCD. Expression of HGF and c-met mRNA and/or protein in sporadic RCC has been reported and the HGF/c-met pathway is thought to be involved in the onset and progression of RCC.11–13 The current study revealed over expression of HGF mRNA and protein as well as c-met protein in ARCD associated RCC. The HGF/c-met pathway may contribute not only to tubular hyperplasia and cyst formation, but also to tumor transformation and RCC development. Moreover, immunostaining for Bcl-2 was noted in most RCC regions. A previous study has shown immunostaining for Bcl-2 in 70% of RCC cases, suggesting that Bcl-2 over expression has a role in tumorigenesis.19 In ARCD Bcl-2 over expression on tubular, cystic epithelial and RCC cells may promote the action of the HGF/c-met pathway for the development of acquired cysts and tumor, and the growth of associated RCC by inhibiting programmed cell death. As previously reported, ARCD associated RCC is found predominantly in males, and the prevalence of ARCD and associated RCC increases with the duration of dialysis.1–5 In this study 14 of the 16 patients (88%) were males. The duration of dialysis was 2 to 28 years (average 15.9) and it was more than 10 years in all patients except 1, who was on dialysis for 2 years. Concerning male predominance, sexual hormones are one of the candidates for promoting the development of acquired cysts and neoplasms.1, 2 Further study is needed to elucidate the contribution of sexual hormones to the process of cyst formation and tumor transformation in patients on dialysis. The duration of chronic renal failure or dialysis is thought to be the principal factor in the development of acquired cysts.1, 3, 18 Since serum HGF increases with the progression of renal failure and it is higher in patients on dialysis for a longer duration,14 prolonged stimulation of renal tubular proliferation by endogenous HGF may have a critical role in the continuous process of cyst formation and subsequent tumor transformation in patients with ESRF. However, we did not find any correlation between the intensity of HGF and c-met staining, and the duration of dialysis. CONCLUSIONS
Our observations suggest that the HGF/c-met pathway has a critical role in the progression of acquired cyst formation and subsequent tumor transformation in patients with ESRF. Over expressed Bcl-2 may promote this process by inhibiting apoptosis. Further study is necessary to elucidate the reason why ARCD and associated RCC are predominant in males. GAPDH primers were obtained from Applied Biosystems Japan, Tokyo, Japan. REFERENCES
FIG. 2. RT-PCR results for HGF in normal kidney and kidney with ARCD associated RCC. PCR product for HGF and GAPDH were electrophoretically separated by 2% agarose gel. Bands show HGF mRNA at 284 bp and GAPDH mRNA at 226 bp. HGF mRNA expression was observed in nontumor and RCC regions in ARCD associated RCC but not in normal kidney.
1. Truong, L. D., Krishnan, B., Cao, J. T., Barrios, R. and Suki, W. N.: Renal neoplasm in acquired cystic kidney disease. Am J Kidney Dis, 26: 1, 1995 2. Ishikawa, I.: Acquired cystic disease: mechanisms and manifestations. Semin Nephrol, 11: 671, 1991
2170
HEPATOCYTE GROWTH FACTOR AND C-MET IN ACQUIRED RENAL CYSTIC DISEASE
3. Gulanikar, A. C., Daily, P. P., Kilambi, N. K., Hamrick-Turner, J. E. and Butkus, D. E.: Prospective pretransplant ultrasound screening in 206 patients for acquired renal cysts and renal cell carcinoma. Transplantation, 66: 1669, 1998 4. Denton, M. D., Magee, C. C., Ovuworie, C., Mauiyyedi, S., Pascual, M., Colvin, R. B. et al: Prevalence of renal cell carcinoma in patients with ESRD pre-transplantation: a pathologic analysis. Kidney Int, 61: 2201, 2002 5. Hoshida, Y., Nakanishi, H., Shin, M., Satoh, T., Hanai, J. and Aozasa, K.: Renal neoplasias in patients receiving dialysis and renal transplantation: clinico-pathological features and p53 gene mutations. Transplantation, 68: 385, 1999 6. Nadasdy, T., Laszik, Z., Lajoie, G., Blick, K. E., Wheeler, D. E. and Silva, F. G.: Proliferative activity of cyst epithelium in human renal cystic diseases. J Am Soc Nephrol, 5: 1462, 1995 7. Vargas, G. A., Hoeflich, A. and Jehle, P. M.: Hepatocyte growth factor in renal failure: promise and reality. Kidney Int, 57: 1426, 2000 8. Matsumoto, K. and Nakamura, T.: Hepatocyte growth factor: renotropic role and potential therapeutics for renal diseases. Kidney Int, 59: 2023, 2001 9. Takayama, H., LaRochelle, W. J., Sabnis, S. G., Otsuka, T. and Merlino, G.: Renal tubular hyperplasia, polycystic disease, and glomerulosclerosis in transgenic mice overexpressing hepatocyte growth factor/scatter factor. Lab Invest, 77: 131, 1997 10. Takayama, H., LaRochelle, W. J., Sharp, R., Otsuka, T., Kriebel, P., Anver, M. et al: Diverse tumorigenesis associated with aberrant development in mice overexpressing hepatocyte growth factor/scatter factor. Proc Natl Acad Sci USA, 94: 701, 1997
11. Natali, P. G., Prat, M., Nicotra, M. R., Bigotti, A., Olivero, M., Comoglio, P. M. et al: Overexpression of the met/HGF receptor in renal cell carcinomas. Int J Cancer, 69: 212, 1996 12. Pisters, L. L., El-Naggar, A. K., Luo, W., Malpica, A. and Lin, S. H.: C-met proto-oncogene expression in benign and malignant human renal tissues. J Urol, 158: 724, 1997 13. Horie, S., Aruga, S., Kawamata, H., Okui, N., Kakizoe, T. and Kitamura, T.: Biological role of HGF/MET pathway in renal cell carcinoma. J Urol, 161: 990, 1999 14. Sugimura, K., Kim, T., Goto, T., Kasai, S., Takemoto, Y., Matsuda, J. et al: Serum hepatocyte growth factor levels in patients with chronic renal failure. Nephron, 70: 324, 1995 15. Miyata, Y., Koga, S., Kanda, S., Nishikido, M., Hayashi, T. and Kanetake, H.: Expression of cyclooxygenase-2 in renal cell carcinoma: correlation with tumor cell proliferation, apoptosis, angiogenesis, expression of matrix metalloproteinase-2, and survival. Clin Cancer Res, 9: 1741, 2003 16. Horie, S., Higashihara, E., Nutahara, K., Mikami, Y., Okubo, A., Kano, M. et al: Mediation of renal cyst formation by hepatocyte growth factor. Lancet, 344: 789, 1994 17. Mizuno, S., Matsumoto, K. and Nakamura, T.: Hepatocyte growth factor suppresses interstitial fibrosis in a mouse model of obstructive nephropathy. Kidney Int, 59: 1304, 2001 18. Yamaguchi, K., Nalesnik, M. A. and Michalopoulos, G. K.: Expression of HGF mRNA in human rejecting kidney as evidenced by in situ hybridization. Urol Res, 24: 349, 1996 19. Huang, A., Fone, P. D., Gandour-Edwards, R., Devere White, R. W. and Low, R. K.: Immunohistochemical analysis of BCL-2 protein expression in renal cell carcinoma. J Urol, 162: 610, 1999