Interferon-g and its functional receptors overexpression in benign prostatic hyperplasia and prostatic carcinoma: parallelism with c-myc and p53 expression.

ARTICLE

INTRODUCTION

Interferons are a group of proteins that trigger multiple responses including prevention of viral replication, inhibition of cell growth, and modulation of cell differentiation [1]. They also show anti-tumour properties, and are currently used for the treatment of cancers [2]. Type I interferons include interferon-alpha (IFN-alpha, leukocyte-derived) and interferon-beta (IFN-beta, fibroblast-derived), which are produced during viral or bacterial infection. Type II or immune interferon (IFN-gamma) is produced primarily by activated T lymphocytes upon mitogen or antigen stimulation, shows little or no homology with type I interferons [3], and its receptor differs from that of the type I interferons [4].

The functional IFN-gamma receptor consists of two subunits: alpha and beta. Concomitant expression of the two receptors is required for transcriptional activation after IFN-gamma binding [5]. The IFN-gamma-Ralpha subunit is required for ligand binding and signal transduction [6] and the IFN-gamma-Rbeta subunit plays a critical role in signalling and is species-matched to the extracellular domain of the alpha-chain [5]. To our knowledge, there are no immunohistochemical studies of IFN-gamma and its receptors in the prostate, although several authors have reported the therapeutic potential of this factor in PC, since it inhibits the proliferation of PC cell lines both in vitro and in vivo [7-9].

Cancer has been described as a disease of the cell cycle, and interferons have been reported to inhibit cell replication by blocking their progression through the cell cycle [9]. The genes which either increase cell proliferation (oncogenes) or decrease it (tumour suppressor genes) are implicated in the control of cell proliferation [10]. c-myc is considered an oncogene [11] and a perturbation of its normal function, usually due to amplification or a translocation, has been implicated in several human tumour types, including breast, lung, colon and prostate carcinomas [12, 13]. p53 is a transcription factor capable of DNA sequence-specific binding, and is also a regulator of the G1/S checkpoint of the cell cycle, a regulator of programmed cell death, and an inhibitor of SV40 DNA synthesis [14]. This factor is considered an anti-oncogene or tumour suppressor gene, and mutations in the p53 gene are among the most common genetic alterations in human cancers [15]. A co-expression of c-myc and the mutant form of p53 has been reported in PC [16]. In addition, in the human prostatic cell line JCA-1, the expression of wild form p53 was found to be up-regulated by addition of IFN-gamma to the cell cultures [8].

The aim of the present study was to investigate the expression of IFN-gamma and its receptor components (IFN-gamma-Ralpha and IFN-gamma-Rbeta) in normal prostate, benign prostatic hyperplasia (BPH), and prostatic cancer (PC), as well as the possible relationship between this factor and the products of the p53 gene (the wild and mutant forms) and the oncogene c-myc, by means of immunochemical techniques (Western blot, ELISA, and quantification of immunostaining in histological sections).

MATERIALS AND METHODS

The prostates were obtained from: (a) transurethral resection from 25 men (aged from 53 to 88 years) diagnosed clinically and histopathologically with BPH; (b) radical prostatectomies from 25 men (aged from 54 to 69 years) diagnosed with PC, dominant Gleason grade 3, and without metastasis or lymph node infiltration at the time of surgery; and (c) histologically normal prostates obtained at autopsy (8-10 hours after death) from 15 men (aged from 20 to 38 years) without histories of reproductive, endocrine or related diseases. Each sample was divided intro three portions: one portion was immediately processed for immunohistochemistry, and the other two portions were frozen in liquid nitrogen and maintained at ­ 80° C for enzyme-linked immunoassay (ELISA) and Western blotting analysis.

For immunohistochemistry, tissues were fixed in a 0.1 M phosphate-buffered 10% formaldehyde solution for 24 hours, dehydrated, and embedded in paraffin. Sections (5 mum thick) were processed following the avidin-biotin-peroxidase complex method. Briefly, after deparaffinisation, the sections were hydrated and incubated for 30 min in 0.3% H₂O₂ in methanol to inhibit endogenous peroxidase activity. The sections were incubated overnight at 4° C with the primary antibodies (all from Santa Cruz Biotechnologies, Santa Cruz, CA, USA) diluted in TBS containing 1% BSA. The primary antibody dilutions found to be optimal for this study were: IFN-gamma, 1:30; IFN-gamma-Ralpha, 1:150; IFN-gamma-Rbeta, 1:40; p53 (Pab 1801), 1:20; p53 (Pab 240), 1:20; and c-myc, 1:20. The sections were then washed in TBS, and incubated with either donkey anti-goat (for IFN-gamma), mouse anti-rabbit (for IFN-gamma-Ralpha and IFN-gamma-Rbeta), or rabbit anti-mouse (for c-myc, p53-Pab 1801, and p53-Pab 240) biotinylated immunoglobulins (Vector Laboratories, Burlingame, CA, USA), in all cases for 1 hour at room temperature. Later on, the sections were incubated with avidin-biotin-peroxidase complex (Vector) for 30 min at room temperature and developed with diaminobenzidine (DAB, Sigma, Barcelona, Spain). After this, sections were dehydrated and mounted in DePex (Probus, Badalona, Spain). Care was always taken to develop the sections of the different pathological and non-pathological conditions exactly the same time in each immunohistochemical reaction.

The specificity of the immunohistochemical procedures was checked using negative and positive control sections. For the negative control of the immunoreactions, adjacent sections of each type (normal, BPH, and PC) were incubated with preimmune goat, mouse or rabbit serum according to the first antibody, at the same immunoglobulin concentrations used for each antibody. As positive controls, sections of human thymus (for IFN-gamma, IFN-gamma-Ralpha, and IFN-gamma-Rbeta) and skin (for p53-Pab 1801, and c-myc) were incubated with the same antibodies.

A histological, comparative quantification of immunolabelling density in normal, hyperplastic, and neoplastic prostates was performed for each antibody. Of each normal prostate specimen, 6 histologic sections of each region (central, intermediate and peripheral) were selected at random and the staining intensity (optical density) per unit surface area of the epithelium was measured with an automatic image analyser (MIP4 version 4.4, Consulting Image Digital, Barcelona, Spain) in 5 light microscopic fields using the X40 objective. For each positively immunostained section one negative control section (the following in a series of consecutive sections) was also used, and the optical density of this control section was substracted from that of the stained section. Of the average values obtained for each prostate specimen the means ± SD for the normal prostate group were calculated. The same quantitative study was carried out in the hyperplastic and neoplastic prostates, although the number of sections used was greater (23 in BPH and 29 in PC), and all these sections were taken from the impaired zone. In the study, the number of sections and microscopic fields in each section necessary for calculation were determined by successive approaches to obtain the minimum number required to reach the lowest SD. The statistical significance between the mean values obtained for each prostate group and antibody were assessed by the ANOVA test and the two-sample t test with unequal variances and unequal sample sizes (Fisher and Behrens' test).

For Western blot analysis, tissues were homogenised in the extraction buffer (0.05 M Tris-HCl, pH 8) with the addition of a cocktail of protease inhibitors (10 mM iodoacetamide, 100 mM phenylmethyl sulphonic fluoride, 0.01 mg/ml of soybean trypsin inhibitor and 1 mul/ml of leupeptin) in the presence of 0.5 % Triton X-100. Homogenates were centrifuged for 10 min at 10,000 rpm. Supernatants were mixed with an equivalent volume of SDS loading buffer (10% SDS in Tris/HCl pH 8 containing 50% glycerol, 0.1 mM 2-beta-mercaptoetanol and 0.1% bromophenol blue). Then the mixture was denatured for 5 min at 100° C, and aliquots of 10 microliters of homogenate were separated in SDS-polyacrylamide slab minigels (15% gradient gels). Separated proteins were transferred in the transfer buffer (25 mM Tris-HCl, 192 mM glycine, 0.1% SDS and 20% methanol). Nitrocellulose membranes (0.2 mum) were blocked overnight at 37° C with 3% BSA dissolved in TBST buffer (10 mM Tris-HCl, 150 mM NaCl, 0.05% Tween-20 pH 8), and then incubated for 3 h with the primary antibodies at 1:150 (p53 Pab 1801), 1:200 (IFN-gamma), 1:250 (p53 Pab 240 and c-myc), or 1:500 (IFN-gamma-Ralpha and IFN-gamma-Rbeta) dilution in blocking solution. After extensive washing with PBS/Tween 20, the membranes were incubated with a biotin-conjugated donkey anti-goat (IFN-gamma), anti-rabbit (IFN-gamma-Ralpha and IFN-gamma-Rbeta), or anti-mouse (p53-Pab 1801, p53-Pab 240, and c-myc) (Vector) immunoglobulins for 1 h, and then washed and incubated with the avidin-biotin-peroxidase complex (Vector) at 1:1000 dilution. The filters were developed with an enhanced chemiluminescence (ECL) kit, following the procedure described by the manufacturer (Amersham, Buckinghamshire, UK).

For the ELISA study, the protein concentration of homogenates was calculated by the Bradford method. The different antigens were coated on 96-well multiplates overnight at 4° C. The plates were washed with TBS containing 0.05% Tween 20 and blocked with 1% BSA in TBS for 1 hour at room temperature, and incubated with the different antibodies (anti-IFN-gamma, IFN-gamma-Ralpha, IFN-gamma-Rbeta, Pab 1801 for wild-p53 detection, Pab 240 for mutant-p53 detection, and c-myc) for 3 hours, also at room temperature. After another wash, the biotin-conjugated anti-rabbit or anti-mouse immunoglobulins (Vector) or donkey anti-goat immunoglobulins (Santa Cruz) were added to each well, incubated for 1 hour at room temperature, and then washed and incubated with the avidin-biotin-peroxidase complex (Vector). The interactions were visualised with 0.05% 2,2 azino di-3-ethylbenziatioazoline sulphonic acid (ABTS) (Sigma) in 100 mM citrate buffer, and were measured (optical density at 405 nm) in a spectrophotometer (Multiskan Bichromatic, Labsystems, Finland). From the average values obtained for each prostate, the means for each prostate group and antigen were obtained.

In order to know if the source of material (surgery or autopsy) could be responsible for changes in the immunohistochemical pattern, five prostatic biopsies (which were taken because of the suspicion of prostatic disease and their histological study revealed a normal histological pattern) were processed for immunohistochemistry. The results of the quantitative immunohistochemical study in these biopsies were compared with that performed in autopsy prostates.

RESULTS

The Western blot analysis showed a single band for each antigen, at the corresponding molecular weight, except for the mutant form of p53 (240) and the c-myc gene in the normal prostates which showed no reaction for these two antigens (Figure 1). For wild p53, a weakly stained band was observed in the three groups of prostates. For the remaining antibodies, the bands appeared more intensely stained in BPH and PC specimens than in normal prostates.

The ELISA showed a linear correlation between the optical density and increasing concentrations for each antibody analysed, and corroborates the results of Western blot analysis (Figure 2). Except for the wild form p53 (Pab 1801), which only showed a scanty reaction in the three groups of prostates, the highest optical densities were found in PC samples, and the lowest optical densities corresponded to normal prostates.

The immunohistochemical study showed no immunoreaction in the negative controls incubated with the preimmune sera (Figure 3). Immunostainings of thymus (IFN-gamma, IFN-gamma-Ralpha, IFN-gamma-Rbeta) and skin (c-myc) sections were always positive. Skin immunoreaction to wild p53 was very weak. The results obtained fom the comparison of the optical densities for the three groups of specimens (normal prostates, BPH and PC) for each immunohistochemical stain are shown in Table 1. No significant histological or quantitative immunohistochemical differences between the two subgroups of normal prostates (biopsies and autopsies) were observed.

In the normal prostates, IFN-gamma immunostaining was positive in the basal cells of the epithelium, and weak in some stromal cells (Figure 4a). In the BPH specimens, labelling of this protein increased significantly (P ¾ 0.05) in both basal cells and stromal cells, and, in addition, the columnar cells of the epithelium were also immunostained (Figure 4b). In the cancerous prostates studied, the three cell types were immunolabelled (Figure 4c) and the intensity of epithelial immunostaining was significantly higher (P ¾ 0.05) than in BPH specimens.

Positive immunostaining for IFN-gamma-Ralpha was detected in basal cells and some stromal cells in normal prostates (Figure 5a). In BPH specimens, immunostaining was observed in the same locations (Figure 5b), although the intensity of basal cell labelling was significantly higher (P ¾ 0.05) than in normal prostates. In PC, all epithelial cells and some stromal cells were immunostained (Figure 5c), and the intensities of both epithelial and stromal cell immunostaining were significantly increased (P ¾ 0.05) with regard to those of BPH specimens.

The reaction to anti-IFN-gamma-Rbeta was similar to that found for IFN-gamma-Ralpha. The only qualitative difference between both receptor types was that, in BPH, columnar cells were also immunostained for IFN-gamma-Rbeta (Figure 6). The quantitative study revealed that the intensity of epithelial immunostaining was similar in normal prostates and BPH, and significantly higher (P ¾ 0.05) in PC. The wild-type p53 (Pab 1801) content determined by immunohistochemistry was low in the three groups of prostates studied. Only the nuclei of stromal cells were positively stained.

No immunoreactiveness to mutant p53 (Pab 240) was observed in normal prostates (Figure 7a). In BPH (Figure 7b) and PC specimens (Figure 7c), positive immunoreaction was identified in the nuclei of epithelial cells and some stromal cells, although the intensity of immunostaining was significantly higher (P ¾ 0.05) in PC.

Myc protein could not be detected by means of immunohistochemistry in normal prostates (Figure 8a). In BPH, immunoreaction was identified in the nuclei of all epithelial cells and some stromal cells (Figure 8b). In PC, immunostaining also appeared in epithelial cells and some stromal cells (Figure 8c), although the intensity of immunostaining in both structures was significantly higher (P ¾ 0.05) than in BPH.

DISCUSSION

Interferons have been widely used in cancer therapy. In urological cancer, they have revealed promising results in experimental model systems as well as in phase I and II clinical studies [17]. However, there have been no reports concerning the IFN-gamma content or its receptors in normal prostates, BPH or PC. Present studies provide information about in vivo conditions and cell interactions before starting IFN-gamma therapy in cancer patients.

IFN-gamma is produced by immune cells [3, 4]. IFN-gamma synthesis has only been mentioned in the porcine trophectoderm and rat early spermatids [18], although many cell types seem to have receptors for this factor [6]. Present results reveal the basal cells of normal prostatic epithelium show positive immunostaining for
IFN-gamma and its two receptors. This suggests that these cells produce IFN-gamma, which would act as an autocrine factor, probably to avoid an excessive proliferation of the epithelial cells. Stromal cells also express complete receptors, and because of the scant amount of IFN-gamma in these cells, it could be hypothesized that basal cells also exert a paracrine regulation on stromal cell proliferation.

In BPH, in addition to these cell-to-cell communication patterns, the secretory columnar cells also demonstrate IFN-gamma and IFN-gamma-Rbeta but not IFN-gamma-Ralpha, and thus, they could be producing the IFN-gamma themselves, although they are not receptive to this factor. This suggests that the IFN-gamma produced by the secretory columnar cells might cause a paracrine overstimulation (in this case inhibition of proliferation) of basal cells.

In PC, immunoreactivity to IFN-gamma and its receptors was increased with respect to normal prostate, and all epithelial cells exhibited complete receptors together with IFN-gamma. Therefore, an autocrine regulation of all epithelial cells could also occur in PC. In addition, in these patients, the increased IFN-gamma immunoreactivity of stromal cells might also mean that these cells perform an autocrine regulation which would be added to the paracrine regulation of stromal cells by epithelial cells.

All these observations suggest that, in BPH, and even more so in PC, increased IFN-gamma secretion is a response to the increased epithelial proliferation, as an attempt to inhibit it. These findings agree with the results obtained in colon carcinoma cell lines by Aguet et al. [6] who found that IFN-gamma receptors are expressed to a lesser extent on normal cells than on tumour cells.

The increased expression of IFN-gamma and its receptors in PC is surprising because it has been demonstrated that IFN-gamma inhibits in vitro proliferation of several cancer cell lines including the human prostatic lines
PC-3 [19, 20], JCA-1 [20] and DU145 [9]. Similar results were obtained after administration of IFN-gamma to PC-3 and DU145 xenografts transplanted in nude mice [7]. Nevertheless, this antiproliferative effect seems to occur only for high IFN-gamma levels [7, 8]. In the JCA-1 line (which contains the wild form of p53 but not the mutant form), Oya et al. [8] observed that increased IFN-gamma levels induced increased expression of both wild-p53 and p21. In the DU145 line, which present the mutant form of p53, Hobeika et al. [9] also found that the inhibitory effect of IFN-gamma was associated with an increase in p21 expression. In our study, the wild form of p53 was poorly expressed, even in normal prostate, because this protein has a short half-life (about 20 min) [21] and, consequently, is not usually detected by immunohistochemical staining [22]. In contrast, mutant-p53, which has a longer half-life [23], was detected in the PC specimens studied here. This mutant form has also been reported in many immunohistochemical studies of PC [17, 24], although no statistically significant association between p53 positivity and prostate tumour grade has been demonstrated [25-27].

The present results revealed that mutant-p53 is also present in BPH. This agrees with Meyers et al. [28] who suggested that mutant-p53 overexpression might be associated with the known proliferative capacity of basal cells in BPH glands, and that mutations of p53 might play a role in the pathogenesis of a subset of high-grade prostate adenocarcinomas [29].

Elevated c-myc expression has been observed in both BPH and PC [13, 30], and this expression has been correlated with increasing tumour grade [31]. This is consistent with the present results, where c-myc immunoreaction was found in BPH and PC but not in normal prostates, although a correlation of immunohistochemical changes with the tumour grade could not be made because all PC specimens corresponded to the Gleason score 3. Hermeking and Eick [32] have suggested that p53 is stabilised by c-myc activation, and that these two events cooperate in tumour development.

A relevant finding of the present study is that the immunophenotype of BPH cells is more similar to that observed in PC than to that of normal prostate. This suggests that, at least in relation to some of the factors that regulate the cell cycle (IFN-gamma and its receptors, mutant-p53, and c-myc), the reaction of BPH cells is rather like that of PC cells, although milder. This finding suggests that BPH might be a pre-malignant stage of PC. If the expression of genes that increase cell proliferation, such a mutant-p53 and c-myc increases to values that cannot be controlled by tumour suppressor genes such as IFN-gamma, benign hyperplasia could evolue to malignant neoplasia.

CONCLUSION

REFERENCES

1. Borden E C, Balkwill F R. 1984. Preclinical and clinical studies of interferons and interferon inducers in breast cancer. Cancer 53: 783.

2. Witherspoon L R. 1984. The interferons-clinical applications and new technologies. J. Nucl. Med. 25: 817.

3. Balkwill F R. 1989. Interferons. (Review) Lancet I: 1060.

4. Branca A A, Baglioni C. 1981. Evidence that types I and II interferons have different receptors. Nature 294: 768.

5. Hemmi S, Böhni R, Stark G, Di Marco F, Aguet M. 1994. A novel member of the interferon receptor family complements functionality of the murine interferon gamma receptor in human cells. Cell 76: 803.

6. Aguet M, Dembic Z, Merlin G. 1988 Molecular cloning and expression of the human interferon-gamma receptor. Cell 55: 273.

7. Van Moorselaar R J, van Stratum P, Borm G, Debruyne F M, Schalken J A. 1991. Differential antiproliferative activities of alpha- and gamma-interferon and tumor necrosis factor alone or in combination against two prostate cancer xenografts transplanted in nude mice. Prostate 18: 331.

8. Oya M, Konno S, Chen Y, Tazaki H, Mallouh C, Wu J M. 1997. Control of cell cycle regulatory proteins and modulation of STAT1 proteins by IFN-gamma in human prostatic JCA-1 cells. Int. J. Oncol. 10: 835.

9. Hobeika A C, Subramanian P S, Johnson H M. 1997. IFN-alpha induces the expression of the cyclin-dependent kinase inhibitor p21 in human prostate cancer cells. Oncogene 14: 1165.

10. Bussemakers M J, van de Ven W J, Debruyne F M, Schalken J A. 1991. Identification of high mobility group protein I(Y) as potential progression marker for prostate cancer by differential hybridation analysis. Cancer Res. 15: 606.

11. Cole M D. 1986. The myc oncogene: its role in transformation and differentiation. (Review) Annu. Rev. Genet. 20: 361.

12. El-Deiry W S. 1998. The p53 pathway and cancer therapy. Cancer J. 11: 229.

13. Thompson A M, Steel C M, Chetty U, Hawicins R A, Miller W R, Carter D C, Forrest A P, Evans H J. 1990. p53 gene mRNA expression and chromosome 17p allele loss in breast cancer. Br. J. Cancer 61: 74.

14. Vogelstein B, Kinzler K W. 1992. p53 function and dysfunction. (Review) Cell 70: 523.

15. Levine A J, Momand J, Finlay C A. 1991. The p53 tumor suppressor gene. (Review) Nature 351: 453.

16. Fox S B, Persad R A, Royds J, Kore R N, Silcocks P B, Collins C C. 1993. p53 and c-myc expression in stage A1 prostatic adenocarcinoma: useful prognostic determinants? J. Urol. 150: 490.

17. Horoszewicz J S, Murphy G P. 1989. An assessment of the current use of human interferons in therapy of urological cancers. (Review) J. Urol. 142: 1173.

18. Dejucq N, Dugast I, Ruffault A, van der Meide P H, Jegou B. 1995. Interferon-alpha and -gamma expression in the rat testis. Endocrinology 136: 4925.

19. Liu S, Ewing M W, Anglard P, Trahan E, Larrocca R V, Myers C E, Linehan W M. 1991. The effect of suramin, tumor necrosis factor and interferon-gamma on human prostate carcinoma. J. Urol. 145: 389.

20. Nakajima Y, DelliPizzi A, Mallouh C, Ferreri N R. 1995. Effect of tumor necrosis factor-alpha and interferon-gamma on the growth of human prostate cancer cell lines. Urol. Res. 23: 205.

21. Rogel A, Popliker M, Webb C G, Oren M. 1985. p53 cellular tumor antigen: analysis of mRNA levels in normal adult tissues, embryos and tumors. Mol. Cell Biol. 5: 2851.

22. Iggo R, Gatter K, Bartek J, Lane D, Harris A L. 1990. Increased expression of mutant forms of p53 oncogene in primary lung cancer. Lancet 335: 675.

23. Sturzbecher H W, Chumakov P, Welch W J, Jenkins J R. 1987. Mutant p53 proteins bind hsp 72/73 cellular heat shock-related proteins in SV40-transformed monkey cells. Oncogene 1: 201.

24. Effert P J, Neubauer A, Walther P J, Liu E T. 1992. Alterations of the p53 gene are associated with the progression of a human prostate carcinoma. J. Urol .147: 789.

25. Visakorpi T, Kallioniemi O P, Heikkinen A, Koivula T, Isola J. 1992. Small subgroup of aggressive highly proliferative prostatic carcinomas defined by p53 accumulation. J. Natl. Cancer Inst. 84: 883.

26. Mellon K, Thompson S, Charlton R G, Marsh C, Robinson M, Lane D P, Harris A L, Home C H, Neal D E. 1992. p53, c-erbB2 and the epidermal growth factor receptor in the benign and malignant prostate. J. Urol. 147: 496.

27. Soini Y, Paakko P, Nuorva K, Kamel D, Lane D P, Vahakangas K. 1992. Comparative analysis of p53 protein immunoreactivity in prostatic, lung and breast carcinomas. Virchows Arch. A. Pathol. Anat. Histopathol. 421: 223.

28. Meyers F J, Chi S G, Fishman J R, deVere White R W, Gumerlock P H. 1993. p53 mutations in benign prostatic hyperplasia. J. Natl. Cancer Inst. 85: 1856.

29. Kallakury B V, Figge J, Ross J S, Fisher H A, Figge H L, Jennings T A. 1994. Association of p53 immunoreactivity with high Gleason tumor grade in prostatic adenocarcinoma. Hum. Pathol. 25: 92.

30. Marcu K B, Bossone S A, Patel A J. 1992. myc function and regulation. (Review) Annu. Rev. Biochem. 61: 809.

31. Jenkins R B, Qian J, Lieber M M, Bostwick D G. 1997. Detection of c-myc oncogene amplification and chromosomal anomalies in metastatic prostatic carcinoma by fluorescence in situ hybridization. Cancer Res. 57: 524.

32. Hermeking H, Eick D. 1994. Mediated of c-myc-induced apoptosis by p53. Science 265: 2091.

33. Wagner A J, Kokontis J M, Hay N. 1994. myc-mediated apoptosis requires wild-type p53 in a manner independent of cell cycle arrest and the ability of p53 to induce p21^waf1/cip1. Genes. Dev. 8: 2817.