Bcl-2 proto-oncogene expression in low- and high-grade prostatic intraepithelial neoplasia
Objective To determine the incidence of bcl-2 protein expression in low- and high-grade prostatic intra-epithelial neoplasia (PIN) lesions, and to explore the role of bcl-2 in prostatic tumorigenesis.
Materials and methods Immunoreactivity for bcl-2 was examined in 10 samples of benign prostatic hyperplasia (BPH), 13 of primary prostatic adenocarcinoma, 15 of high-grade PIN and 18 of low-grade PIN. All immunostaining results were scored for the approximate percentage of positive tumour cells and relative immunostaining intensity (score range 0–12).
Results In all BPH samples, bcl-2 staining was detected consistently in the basal cell layer of the ducts and acini, but no staining was ever apparent in luminal cells. The immunoreactivity for bcl-2 was heterogeneous in the prostatic carcinomas and bcl-2 protein expression was present in six samples. In these six bcl-2-positive tumours, the mean (range) staining score was 1.15 (1–6). There was detectable expression of bcl-2 in low- and high-grade PIN in all cell layers; immunoreactivity was present in 10 of 15 high-grade PIN lesions, with a mean (range) score of 1.14 (1–4), and in 12 of 18 samples of low-grade PIN, with a mean (range) score of 1.77 (1–6).
Conclusions The high incidence of bcl-2 protein expression in low- and high-grade PIN lesions suggests that bcl-2 protein expression is associated with early prostate tumorigenesis.
The bcl-2 oncoprotein is known to inhibit programmed cell death or apoptosis . Within benign or normal prostatic epithelium the proto-oncogene bcl-2 is normally expressed in androgen-independent basal cells but not in the differentiated luminal secretory cells that are androgen-dependant [2,3]. In prostate cancer, it has been suggested that bcl-2 plays a role in hormone resistance and several studies have shown enhanced expression of bcl-2 in hormone-refractory prostatic adenocarcinomas [3–5]. Bcl-2 immunoreactivity has been reported in 25–62% of untreated prostate cancers and staining in the cancer areas is heterogeneous [3–6]. However, conflicting results have been reported for the incidence of bcl-2 immunoreactivity in high-grade prostatic intraepithelial neoplasia (PIN), which has been shown by several authors to be a precursor of prostatic adenocarcinoma [3,6–9]. In two studies strong bcl-2 immunoreactivity was observed in all foci of high-grade PIN [3,7] but in three studies, only 17–35% of high-grade PIN lesions were bcl-2 immunopositive [6,8,9]. However, there are no reports of the expression of bcl-2 protein in low-grade PIN. The purpose of the present study was to investigate the incidence of bcl-2 staining in low- and high-grade PIN lesions, as bcl-2 protein expression may be important in prostate tumorigenesis.
Materials and methods
Formalin-fixed, paraffin-embedded tissue samples from 10 cases of BPH, 18 low-grade PIN lesions, 15 high-grade PIN lesions and 13 primary prostatic adenocarcinomas were examined. The BPH samples were obtained from 10 suprapubic or retropubic prostatectomies performed for BPH. The adenocarcinomas were removed by radical prostatectomy from 13 patients who had received no prior therapy. The 18 cases of low-grade and seven of high-grade PIN were identified in needle-biopsy specimens. Because eight of the radical prostatectomy specimens also contained high-grade PIN coexisting with carcinoma, there were 15 high-grade PIN lesions available. Tissue samples were re-examined by one pathologist and the tumours graded according to the Gleason system. PIN was classified according to nomenclature described by an international consensus conference held in 1989 .
Sections of 5 µm were mounted on poly-l-lysine-coated slides and deparaffinized. Antigens were retrieved by subjecting the slides to microwaves (700 W) for 4 × 5 min in 1 L of citrate solution. Sections were then exposed to anti-bcl-2-124 monoclonal primary antibody (Biogenex, Santa Barbara, CA, USA; dilution 1 : 20) for 1 h. After washing with PBS, the sections were incubated with biotinylated secondary antibody (Dako, Carpenteria, USA) for 15 min. After incubation for 15 min with streptavidin-peroxidase complex (Dako), sections were exposed to AEC chromogen (Dako) for 5 min; they were counterstained with Mayer’s haematoxylin. Sections of human tonsil tissue were concurrently immunostained with the anti-bcl-2 antibody to provide a positive control and substitution of the anti-bcl-2 antibody with normal mouse serum was used as a negative control.
The immunostaining results were scored as described by Krajewska et al.; the percentage of positive tumour cells was graded as: 0, none; 1, 25%; 2, 26–50%; 3, 51–75%; and 4, 76–100%. The immunostaining intensity was rated as: 0, none; 1, weak; 2, moderate; and 3, intense. Specimens were considered immunopositive when 1% of the tumour cells had clear evidence of immunostaining. As the tumours were sometimes heterogeneous, for some analyses a score was calculated in which the percentage positive rating was multiplied by the intensity rating (score range 0–12). The Kruskal–Wallis anova was used to compare the groups, with P < 0.05 considered to indicate significant differences in all tests applied.
The present study, like others, showed that bcl-2 is normally expressed in the basal cell layer of hyperplastic glands [3,7]. Considering previous reports that the basal epithelial cells, in contrast to the secretory cells, are resistant to the programmed cell death induced by androgen withdrawal , the present results are consistent with the interpretation that the basal cells represent the stem cell layer in the adult prostate gland.
The present study, using an immunohistochemical approach, is the first to characterize the expression of bcl-2 protein in low-grade PIN. Although five studies have reported the incidence of bcl-2 immunoreactivity in high-grade PIN [3,6–9], there have been no reports to our knowledge of bcl-2 expression in low-grade PIN.
In the present study, bcl-2 immunoreactivity was detected in 12 of 15 high-grade PIN samples; in two other studies strong bcl-2 immunoreactivity in all high-grade PIN samples assessed [3,7] was reported, whereas in another three studies, only 17–35% of high-grade PIN lesions were bcl-2 positive [6,8,9]. This variation might be attributable to differences in the antibodies used and the relative preservation of antigen epitopes. The microwave treatment method for antigen unmasking used in here may increase antigenic preservation, resulting in higher bcl-2 positivity.
Of the 13 untreated primary prostatic adenocarci-nomas, there was bcl-2 immunoreactivity in six; the previously published incidence varies widely, at 25–62%[3–6]. Some of this variation might be attributable to the threshold level for positivity or to differences in the patient populations presenting to different institutions. Alternatively, it might also reflect differences in the antibodies used and the relative preservation of antigen epitopes. However, in all studies, including this, bcl-2 staining was always heterogeneous in the cancer areas.
There are conflicting results about the relationship between bcl-2 immunoreactivity and tumour grade. In two studies the expression of bcl-2 tended to be more frequent in high-grade tumours [6,7], but in two other studies [3,12], there was no correlation between these variables. In the present study it was not possible to compare the Gleason score and bcl-2 immunoreactivity, as the range of Gleason scores was too narrow.
Although interobserver variability with low-grade PIN limits its use clinically utility, and many pathologists do not report this finding , low-grade PIN corresponds to ‘very mild’ to ‘mild’ dysplasia. The present results show that two-thirds of low-grade PIN samples were bcl-2 immunoreactive and like high-grade PIN lesions, bcl-2 was expressed in all cell layers. However, the evidence supporting a relationship between high-grade PIN and invasive carcinoma of the prostate is strong, and is supported by histological, immunohistochemical and genetic studies. Several authors suggested high-grade PIN to be a precursor of prostate cancer [14,15]. Although Myers and Grizzle  reported that strong expression of bcl-2 typically occurs in advanced-stage prostatic adenocarcinomas, and therefore probably represents late events in the development of prostatic adenocarcinoma, we suggest (as did Stattin et al. and Colombel et al.) that bcl-2 expression, probably by prolonging cell survival, renders the PIN cells more vulnerable to other oncogenes, which can induce further steps in tumorigenesis. Thus bcl-2 expression may be involved in early prostate tumorigenesis.
Androgen deprivation has been shown to induce bcl-2 expression in the rat ventral prostate in the short-term  and several studies have shown enhanced expression of bcl-2 protein in hormone-refractory prostatic adenocarcinomas, suggesting a role in the development of androgen insensitivity [3,4]. In contrast, H?ggman et al. and Ferguson et al. reported a reduced frequency and extent of PIN after maximal androgen blockade, which suggests that PIN is androgen-dependent. However, we and others have shown that 17–100% of high-grade PIN lesions express bcl-2 immuno-reactivity [3,6–9]. Additionally, van der Kwast  reported that PIN was still present in a few prostatectomy specimens of patients treated for 3 months with androgen-ablation therapy. Thus, as suggested by Stattin et al., the question of whether bcl-2-expressing prostatic cancer cells are hormonally dependent seems to be complex. These apparently contradictory results suggest that the interaction between bcl-2 and other anti-apoptotic genes, e.g. bcl-x and mcl-1, and the pro-apoptotic gene bax, determines the difference in androgen dependence in PIN and tumour cells .
In the present study, the expression of bcl-2 was not evaluated by in situ hybridization. A recent study described discrepancies between the detection of bcl-2 by in situ hybridization and immunocytochemistry in human prostate cancer tissues , and this point should be considered in future studies.
In conclusion, although the present study included only relatively few specimens of low- and high-grade PIN, the results suggested that bcl-2 is expressed in both grades of PIN. The high incidence of bcl-2 immuno-reactivity in PIN lesions suggests a role for bcl-2 in early prostate tumorigenesis. Further studies with this proto-oncogene and other bcl-2 family genes are needed to understand the role of these genes in prostate tumorigenesis. Such studies with a large series of patients may also address the value of bcl-2 as a prognostic marker.
S. Baltaci, MD, Associate Professor.
D. Orhan, MD, Assistant Professor.
G. Özer, MD, Urologist.
Ö. Tolunay, MD, Professor.
O. Gög?üs, MD, Professor.
Correspondence. Dr Sümer Baltaci, Fahrettin Altay Cad., Güzin Sok. No:50/6, 06130 Aydinlikevler, Ankara, Turkey. e-mail: email@example.com