SEARCH

SEARCH BY CITATION

Keywords:

  • bcl-2;
  • competitive RT–PCR;
  • Fas;
  • immunohistochemistry;
  • renal cell carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Background:  In order to characterize the alteration of apoptotic regulatory molecule expression during tumor progression of renal cell carcinoma (RCC), we compared the expression between tumor and normal tissues, and evaluated the relationship of the expression in tumors with pathological and clinical characteristics.

Methods:  Competitive reverse transcription–polymerase chain reaction (RT–PCR) and immunohistochemistry (IHC) allowed the determination of Fas and bcl-2 mRNA and protein expression in surgically resected tumor and normal tissue of 50 RCC.

Results:  The mRNA expression of Fas and bcl-2 in RCC was significantly reduced compared to that in normal tissues. An IHC analysis was supportive of the RT–PCR results. In terms of relationships with pathological and clinical characteristics, the mRNA and protein expression of Fas in high-stage or high-grade tumors was significantly higher than that in low-stage or low-grade tumors. Moreover, a statistically poor prognosis was observed in tumor cases expressing a high amount of mRNA. In bcl-2 analysis, the mRNA and protein expression was significantly reduced in clear cell tumors compared to chromophobe cell tumors.

Conclusion:  It is suggested that the reduced expression of Fas and bcl-2 in RCC compared with the expression in normal kidney is a prominent alteration of apoptotic regulatory molecules. The alteration of the up-regulated Fas expression might be characterized during the tumor progression stage. It is also suggested that the effect of alteration of bcl-2 expression might be minimal during the tumor progression stage because of the reduced expression in tumors of the clear cell type, which is the most dominant cell type in RCC.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Fas and bcl-2 are major apoptotic regulators in some kinds of malignancies, and Fas-mediated apoptosis have been observed in different kinds of malignancies.1–3 However, the Fas/Fas ligand (FasL) system has not been characterized well in renal cell carcinoma (RCC). Fas-mediated apoptosis is closely related to cancer cell apoptosis led by immunotherapy. One of the processes by which cytotoxic T lymphocytes (CTL) kill target cells is the induction of apoptosis via the Fas/FasL system, in which CTL recognize antigens, are activated, up-regulate FasL, and then promote ligation between FasL and Fas on the target cells, which induces apoptosis.4 On the other hand, bcl-2 is involved in the selection of long-living cells and in rescuing them from apoptosis. Vaux et al. first reported that bcl-2 expression conferred prolonged cell survival and possibly contributed to tumorigenesis.5 Multiple studies have since verified an association between bcl-2 overexpression and many carcinomas, including carcinomas in urological areas such as the bladder6 and prostate,7 and its overexpression is thought to be a obstacle for cancer treatment.

In the present study, we focused on the investigation of these two apoptotic regulators (Fas and bcl-2) which have opposite functions. We analysed the mRNA expression by competitive reverse transcription–polymerase chain reaction (RT–PCR), and the protein expression by immunohistochemistry (IHC) in order to characterize the alteration of the expression of these molecules between tumor and normal kidney, and during tumor progression of RCC.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Tissue sample collection

Fifty RCC, surgically resected between 1997 and 2001 at our institution, were used. Case backgrounds are summarized in Table 1 according to the International Union Against Cancer (UICC) standards (sixth edition). Tumor and normal (parenchyma) samples were collected after informed consent, and the study has been approved by the constituted ethics committee of our institution. All samples were divided into two parts. One part was used for the competitive RT–PCR, and the other was used for the immuno- and hematoxylin–eosin (HE)-staining. The competitive RT–PCR samples were snap-frozen in liquid nitrogen immediately after resection and stored at −80°C until use, and the immuno- and HE-staining samples were fixed in formalin and embedded in paraffin. In order to exclude the influence of the expression of these two molecules in CTL, it was confirmed that there was no obvious CTL invasion in the tumor tissue by observation of the HE slides in each experiment.

Table 1.  Patient and tumor characteristics
CharacteristicsNumber of cases
Mean age, years (range): 65.4 (35–81)
Sex
 Male/female32/18
Tumor site
 Left29
 Right21
 Bilateral 0
Clinical result
 Died due to renal cell carcinoma/other13/4
Metastasis at first diagnosis or recurrent metastasis17
Tumor stage
 pT1a15
 pT1b11
 pT2 8
 pT3a12
 pT3b 2
 pT4 2
Tumor grade
 110
 222
 3–418
Cell type
 Clear cell24
 Granular cell 8
 Chromophobe cell11
 Spindle cell 1
 Cyst-associated 1
 Papillary 5

Reverse transcription–polymerase chain reaction for Fas and bcl-2

Reverse transcription using the SuperScript preamplification system (Life Technologies, Gaithersburg, MD, USA) was performed according to manufacturer's instruction. β-Actin mRNA served as an internal control to ensure that an exact amount of high-integrity total RNA was reverse transcribed to produce cDNA. The sequence of primers and thermal conditions in PCR are listed in Table 2.

Table 2.  The sequence of primers used for RT–PCR and mutagenesis of RT–PCR products, and the thermal conditions for PCR
  1. PCR, polymerase chain reaction; RT–PCR, reverse transcription–polymerase chain reaction.

APrimer used for RT–PCRSize of target productSize of competitor product
Fas F5′AGGAAAGCTAGGGACTGCAC 3′180 bp170 bp
Fas R5′GCACTTGGTATTCTGGGTCC 3′  
bcl-2 F5′GCTTTGCCACGGTGGTGGAG 3′210 bp200 bp
bcl-2 R5′CAAAGGCATCCCAGCCTCCG 3′  
β-actin F5′AGCATCCCCCAAAGTTCACA 3′191 bp183 bp
β-actin R5′AAGCAATGCTATCACCTCCC 3′  
BPrimer used for PCR mutagenesis  
mFas F5′-CTGTGTACTCCTTCCCTTCTCGCAGTCTGGTTCATCCCCA-3′  
mFas R5′-AGAAGGGAAGGAGTACACAG-3′  
mbcl-2 F5′-GTCCACCAGGGGCGACATCTGCTCTCCACACACATGACCC-3′  
mbcl-2 R5′-AGATGTCGCCCCTGGTGGAC-3′  
mβ-actin F5′-GCATTGTTACAGGAAGTCCCCTAAAAGCCACCCCA-3′  
mβ-actin R5′-AGAAGTGGGGTGGCTTTTAG-3′  
Fasbcl-2β-actin
Initial heating95°C, 3 min95°C, 3 min95°C, 3 min
Denaturation95°C, 30 s95°C, 30 s95°C, 30 s
Annealing55°C, 30 s58°C, 30 s55°C, 30 s
Extension72°C, 1 min72°C, 1 min72°C, 1 min

Preparation of competitor plasmid by mutagenesis of reverse transcription–polymerase chain reaction products

Reverse transcription–polymerase chain reaction products were subcloned into pGEM T Vector (Promega), and inserted into Escherichia coli cells. After the extraction of a sufficient amount of plasmid, PCR was performed using the plasmid as a template and primers as listed in Table 2. PCR products were inserted into E. coli cells, and the competitor plasmid was constituted by self-recombination of these PCR products. A sufficient amount of competitor was prepared after E. coli was cultured and competitor plasmid was extracted. The schematic explanation of procedures is summarized in Figure 1.

image

Figure 1. The schematic explanation of preparation of competitor plasmid by mutagenesis of reverse transcription–polymerase chain reaction (RT–PCR) products is drawn. By using competitor plasmid as template and the same primers as used in RT–PCR, 8–10 bp short length PCR products compared to those of RT–PCR can be obtained by PCR reaction.

Download figure to PowerPoint

Quantitation of products

The thermal conditions in competitive PCR were the same as in RT–PCR. As the forward primer was fluorescently labeled with Cy-5, competitive PCR products were detected with an automated DNA sequencer. After competitive PCR, the reaction mixture was electrophoresed on a 6% polyacrylamide and 6 mol/L urea gel (HydroLink LongRanger; Amersham Pharmacia Biotech, Little Chalfont, UK) on an AlFred DNA Sequencer system (Amersham Pharmacia Biotech). Usually, five kinds of dilution series of the competitor were mixed with PCR reaction mixture. The peaks were analysed with Alle Links software (Amersham Pharmacia Biotech). Results were plotted as log (competitor amounts [per tube]) versus log (competitor peak/target gene peak area). The initial amount of target gene mRNA was defined as the point on the plot where log (competitor/target gene) was 0. The mRNA levels were expressed as their ratio to β-actin mRNA to standardize the RNA amounts in the initial samples.

Immunohistochemistry for Fas and bcl-2

All immunohistochemical analyses were routinely processed with a streptavidin–biotin method. The Histofine Kit (Nichirei, Tokyo, Japan) was used for this study with minor modification. Briefly, tissue sections were cut at 5 µm and mounted on silane-coated slides. After the sections were deparaffinized, antigen retrieval was performed by the heat treatment of deparaffinized sections in 10 mmol/L citrate buffer, pH 6.0, at 100°C for 15 min. They were incubated for 30 min in 3% H2O2 solution to block endogenous peroxidases. Nonspecific binding was then blocked by a 10% non-immune goat serum. Next, the primary antibody was applied and left overnight at 4°C. The second biotinylated antibody was then applied for 1 h at room temperature, and peroxidase labeled streptavidin was applied for 1 h at room temperature. The slides were finally developed with 3,3′-diaminobenzidine (DAB) as a substrate, and lightly counterstained with Mayer's hematoxylin. Negative controls were prepared by omission of the primary antibodies. More precisely, the following antibodies were used.

Fas (Ab-1): a polyclonal rabbit antihuman Fas antibody (Oncogene Research Products, Cambridge, MA, USA; dilution of 1: 40).

Clone 124: a primary monoclonal mouse antihuman bcl-2 protein antibody (DAKO, Glostrup, Denmark; diluted at 1:50).

Statistical analysis

The differences in the case distribution with regard to mRNA expression positivity and immunoreactivity among the groups were analysed by the χ2-test. Comparisons of the mRNA amounts among the groups were analysed by the Student's t-test. Survival curves were made using Kaplan–Meier analysis, and comparisons of the survival rate between two groups were analysed by the log–rank test. A multivariate Cox proportional hazards regression model was constructed to assess the independent prognostic factors.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Comparison of mRNA expression by competitive reverse transcription–polymerase chain reaction analysis between tumor and normal tissues

The incidence of expression of Fas was 31/50 (62%) in tumor and 43/50 (86%) in normal tissues, and of bcl-2 was 34/50 (68%) in tumor and 43/50 (86%) in normal tissues. In cases of mRNA negative for the first time, the RT–PCR procedure was repeated twice more in order to avoid a false negative due to technical errors. If three RT–PCR showed negative expression, we defined the amount of mRNA as equal to 0. The mRNA amount of Fas and bcl-2 in tumors was significantly reduced compared to normal tissues (Fig. 2).

image

Figure 2. Comparison of Fas and bcl-2 mRNA amounts between tumor and normal tissues. The data of all cases are plotted. The horizontal bars indicate the mean ratios of Fas or bcl-2/β-actin mRNA of the groups.

Download figure to PowerPoint

Comparison between pathological parameters of tumors and mRNA expression by competitive reverse transcription–polymerase chain reaction analysis

We defined pT1 and 2 as low-stage and G1 and 2 as low-grade tumors, and defined pT3a, 3b, and 4 as high-stage and G3 and 4 as high-grade tumors. The mRNA amount of Fas in high-stage or high-grade tumors was significantly higher than that in low-stage or low-grade tumors (Fig. 3). In terms of cell type, the mRNA amount of bcl-2 was significantly lower in the clear cell type in comparison with the chromophobe cell types (Fig. 4).

image

Figure 3. Comparison of Fas mRNA amounts between low-stage (pT1 and 2) or low-grade (G1 and 2) tumors and high-stage (pT3a, 3b, 4) or high-grade (G3 and 4) tumors. The data of all cases are plotted. The horizontal bars indicate the mean ratios of Fas/β-actin mRNA of the groups.

Download figure to PowerPoint

image

Figure 4. Comparison of bcl-2 mRNA amounts between the tumors of clear and chromophobe cell types. The data of all cases are plotted. The horizontal bars indicate the mean ratios of bcl-2/β-actin mRNA of the groups.

Download figure to PowerPoint

The characteristics and evaluation of immunohistochemistry

Immunohistochemically, the characteristics of Fas and bcl-2 in normal tissues were similar. Namely, relatively strong immunoreactivity was observed on the cell membrane and within the cytoplasm of proximal and distal tubules and collecting ducts; however, no or weak immunoreactivity was observed in the glomeruli cells. In the tumor tissues, immunoreactivity was observed on the cell membrane and within the cytoplasm of cancer cells. All cases showed immunoreactivity both in normal and tumor tissues. Almost all normal tubule cells showed positive staining; however, the percentile of positive staining cancer cells varied among cases. Thus, the relative immunoreactivity was defined as the staining intensity of normal tubular cells, and was graded as follows: +, weak positive; ++, moderately positive; and +++, strong positive. In tumor tissues, immunoreactivity was defined as the percentile of positive staining cells, and was graded as follows: +, <10% positive cells; ++, 10–50% positive cells; and +++, >50% positive cells by observation of at least 1000 cells in each section. All slides were examined by two physicians who were familiar with urological pathology of oncology. The typical IHC microphotographs are shown in Figure 5.

image

Figure 5. Typical microphotographs of immunohistochemistry (IHC) for Fas and bcl-2. (a) Fas expression in normal tissue (parenchyma; ×200). (b) bcl-2 expression in normal tissue (×100). (c) Fas expression in tumor tissue (×400). (d) bcl-2 expression in tumor tissue (×400).

Download figure to PowerPoint

The protein expression in tumor and normal tissues by immunohistochemistry analysis

Because the staining characteristics were clearly different between the tubule cells and glomeruli cells in normal tissues, we considered a simple comparison of immunoreactivity between normal and cancer cells to be problematic. We analysed whether or not IHC analysis could support the results obtained by RT–PCR analysis. Thus, the differences of case distribution with regard to immunoreactivity between the group of positive mRNA expression and the group of negative expression in tumor and normal tissues were analysed (Fig. 6). In the tumor tissues, there were significant differences of case distribution with regard to immunoreactivity between the mRNA positive and negative cases in the Fas and bcl-2 analysis. In normal tissues, same statistical difference of case distribution was recognized in the Fas analysis. In other words, stronger immunoreactivity was recognized in the Fas or bcl-2 mRNA positive tumor tissues than in the mRNA negative tumor tissues, and the same result was recognized in normal tissues in the Fas analysis. Although the comparison of case distribution in normal tissues of bcl-2 analysis did not reach statistical differences, the tendency for stronger immunoreactivity to be recognized in the mRNA positive tissues than in the negative tissues was obvious. In general, IHC analysis demonstrated that higher immunoreactivity was recognized in the mRNA positive cases than in the negative cases.

image

Figure 6. The scatter graph demonstrates the case distribution with regard to immunoreactivity between the group of positive mRNA expression and the group of negative expression in Fas and bcl-2 analysis.

Download figure to PowerPoint

Comparison between pathological parameters of tumors and protein expression by immunohistochemistry analysis

In Fas analysis (Fig. 7), there was a significant difference of case distribution with regard to immunoreactivity between G1, 2 and G3, 4 tumors. Thus, the immunoreactivity in G3, 4 tumors was stronger than in G1, 2 tumors. The same tendency was recognized in the comparison between pT1, 2 and pT3a, 3b, 4 tumors; however, it did not reach statistical significance. In bcl-2 analysis (Fig. 8), there was a significant difference of case distribution between clear and chromophobe cell tumors. Thus, the immunoreactivity in chromophobe cell tumors was stronger than in clear cell tumors. The same tendency was recognized in the comparison between clear cell tumors and tumors of the other cell types; however, it did not reach the statistical difference.

image

Figure 7. The case distribution with regard to immunoreactivity of Fas between high and low-stage, and between high and low-grade tumors is shown.

Download figure to PowerPoint

image

Figure 8. The case distribution with regard to immunoreactivity of bcl-2 between the tumors of clear and chromophobe cell types is shown.

Download figure to PowerPoint

Relationships of the clinical courses of patients with mRNA and protein expression

The median follow-up duration was 22.5 months (range 4–50). If patients were divided into high amount of Fas mRNA expression (above 0.11 of Fas/β-actin ratio) and low amount of expression (below 0.11) groups, the former showed a significantly poor cause-specific survival rate compared with that of the latter (Fig. 9). No statistical difference in cause-specific survival was obtained from the data of Fas protein, and of bcl-2 mRNA and protein analysis. We analysed each patient's survival according to pathological parameters by univariate analysis. There were statistical differences between following parameters: pT1, 2 versus pT3, 4; G1, 2 versus G3, 4. Multivariate analysis showed that tumor grade G3, 4 was the independent poor prognostic factor (hazard ratio, 3.783; 95% confidence interval [CI], 1.082–13.862; P = 0.0333).

image

Figure 9. Cause-specific survival according to quantitation of Fas mRNA. There is a significant difference in survival rate between the case with a Fas/β-actin ratio of >0.11 and <0.11. P = 0.034.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

We compared the mRNA and protein expression of Fas and bcl-2 in RCC tissues with those in normal tissues. Needless to say, p53 is also one of the important apoptotic regulatory molecules in regard to carcinogenesis and tumor progression. However, we have determined less importance about p53 expression in RCC because of its rare positivity by IHC methods.8 Thus, we focused our study on the expression of Fas and bcl-2 in RCC. Our results showing a significantly reduced Fas mRNA and protein expression in tumors as compared to those in normal tissues are similar to the results of Kim et al.,9 who investigated Fas expression in tumor and normal tissues by immunohistochemistry. To our knowledge, our results are the first report about detailed comparison by molecular analysis of Fas mRNA expression between RCC and normal tissue. However, our results showing markedly reduced Fas mRNA expression in RCC versus normal tissues agrees with the results for gynecologic malignancies.10 The resultant loss of Fas may make it easier for tumors to survive and grow in spite of the existence of various antitumor mechanisms in vivo.11 On the other hand, our other results regarding mRNA quantitation for Fas demonstrated the statistical relationships of high amounts of mRNA with high-stage or grade tumors, and poor prognosis. IHC for Fas also demonstrated the similar results. RCC tumors have less positive Fas expression, compared with normal kidney. Nevertheless, higher grade or higher stage RCC showed higher Fas. This does not mean that higher grade or higher stage RCC is close to the normal tissue. We have preliminary determined relatively high Fas protein expression and low incidence of cellular apoptosis in normal kidney. On the other hand, cellular apoptosis was compatible with Fas expression in RCC tissue, indicating different cellular apoptotic environments between normal and tumor tissues.12 It was reported that tumor apoptosis occurs more frequently in metastatic foci than in primary lesions of human colorectal carcinomas.13 These findings indicated that tumor progression is often accompanied by an up-regulated occurrence of both cancer cell apoptosis and proliferation. Another interpretation is that Fas expression might have been modulated by the host's immune mechanism. Weller et al. indicated the possibility of Fas induction in tumors by interferon γ as a part of the host's defense mechanism.3 Horie et al. showed that Fas was expressed predominantly in advanced tumors of RCC, and tumors that had metastasized expressed high levels of Fas.14 Anti-Fas monoclonal antibody (mAb) kills Fas-expressing cells by apoptosis. Some chemotherapeutic agents were reported to be useful to enhance cytotoxicity of anti-Fas antibody against RCC cells.15 However, considering our result that normal kidney has much Fas expression compared with RCC tumors, our results do not necessarily contribute to the possibility of clinical use of anti-Fas antibody against RCC.

The results of our comparison of bcl-2 mRNA and protein expression in tumor and normal tissues were similar to those reported by Olive et al. in which it was demonstrated by a semiquantitative RT–PCR that bcl-2 expression was well represented in normal kidney, but markedly decreased in RCC.16 In regard to the antiapoptotic function of bcl-2, the reduced expression of bcl-2 in RCC would be expected to enhance the susceptibility of RCC to CTL killing and have an unfavorable outcome on tumor cell growth. In the comparison of Fas and bcl-2 expression between tumor and normal tissues, our results demonstrated a favorable gene environment of reduced Fas expression for tumor cell growth, and also demonstrated an unfavorable gene environment of reduced bcl-2 expression. Further study is necessary to determine the predominant relevance that the alteration of these two molecules has for carcinogenesis. The results of our mRNA quantitation and IHC for bcl-2 demonstrated no correlation with pathological or clinical parameters except that the expression in clear cell tumors was significantly reduced than in chromophobe cell tumors. The correlations between bcl-2 expression in RCC and clinicopathological factors were analysed by immunohistochemistry.17–19 Paraf et al. demonstrated that tubulopapillary carcinomas strongly expressed bcl-2 immunostaining whereas only a few bcl-2 positive cells were observed in clear cell carcinomas.17 However, the other two studies demonstrated no significant correlations between them.18,19 To our knowledge, no study of bcl-2 mRNA quantitation in RCC has been performed. Taking into account results of ours and the previous reports, both the mRNA and protein levels of bcl-2 seem to have no relation with RCC progression. mRNA quantitation as well as mRNA positivity analysis of bcl-2 might be less meaningful than the same analyses for Fas in determining the tumor progression of RCC because of the reduced expression of bcl-2 in tumors of the clear cell type, which is the most dominant cell type in RCC.

In conclusion, we have shown the alteration of Fas and bcl-2 expression between tumor and normal tissues, and during tumor progression by evaluating the relationship between this expression and the pathological and clinical characteristics in RCC. It is suggested that the reduced expression of Fas and bcl-2 in RCC compared with the expression in normal kidney is prominent alteration of apoptotic regulatory molecules. The alteration of the up-regulated Fas expression might be characterized during the tumor progression stage. It is also suggested that the effect of alteration of bcl-2 expression might be minimal during the tumor progression stage because of the reduced expression in tumors of the clear cell type, which is the most dominant cell type in RCC.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The authors would like to thank Dr Toshiyuki Yamamoto and Ms Etsuko Ueta, the Gene Research Center, Tottori University, Yonago, Japan, for their laboratory directions for competitive RT–PCR. This paper was presented on 28 April 2003, at the 98th American Urological Association Meeting in Chicago, IL, USA.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References