| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Departments of Urology [E. O. O-O., D. P. S. S., T. R. T.] and Pathology [E. H. M.], Leicester General Hospital, Leicester LE5 4PW; Department of Epidemiology and Public Health, University of Leicester, Leicester LE1 6TP [N. A. T.]; and University Department of Surgery, Western General Hospital, Edinburgh EH4 2XU [F. K. H.], United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the earliest studies, hAR content of prostatic tissues was assessed using biochemical assays, but these studies were considered nonspecific because the substrates analyzed were a heterogeneous mixture of stromal elements and benign and malignant epithelial cells (2) . The recent development of anti-hAR-specific antibodies has made it possible to determine hAR expression in the individual glands and cellular compartments within prostatic tissue sections. Studies using this newer methodology have concentrated mainly on the tumor cells and have reported conflicting findings. Whereas initial reports suggested that hAR protein expression in epithelial nuclei was inversely proportional to the histological grade of the tissues (3 , 4) , more recent studies have detected high expression and increased heterogeneity of the receptor in advanced cancers (5 , 6) . These later reports are supported by scrutiny of prostate cancer tissues using reverse transcriptase techniques, which has revealed amplification of the hAR gene in hormone-refractory tumors (7 , 8) .
hAR transmission of hormonal control of prostatic growth occurs via the regulation of the expression of the peptide growth factors and receptors involved in the stromal-epithelial interactions that mediate the influence of androgens on the gland (9 , 10) . Proliferation and differentiation of prostatic epithelium is initiated by the binding of epithelial cell membrane receptors by growth factors expressed by stromal cells (11) . This intercellular relationship indicates that prostatic stroma plays an important intermediary role in the transmission of androgen-induced stimuli to the adjacent epithelium. Malignant transformation of prostatic epithelial cells (and the development of hormone resistance) may therefore involve abnormalities of a pathway that leads from the stromal nucleus through receptors on the epithelial cell membrane to the epithelial cell nucleus.
Little is known about hAR expression in the stroma of prostate cancer because, in contrast to other aspects of the androgen-regulatory pathway, it has received little specific attention from researchers. We postulated that transformation of prostatic epithelial cells could be associated with changes in hAR expression in the nuclei of surrounding stroma. To test this hypothesis, we assessed the differential expression of hAR in epithelial and stromal nuclei in archival sections of benign and malignant prostatic tissue using immunohistochemical methods.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
hAR Immunostaining.
Using a commercially available monoclonal anti-hAR antibody (NCL-2F-AR; Novocastra), hAR immunostaining was done as described previously (13)
. Briefly, the archival sections were deparaffinized and rehydrated in xylene and graded alcohols. They were equilibrated in TBS, and then intrinsic peroxidase activity was quenched with hydrogen peroxide. Exposure of hAR epitopes was achieved by microwaving the sections in 10 mM citrate buffer (pH 6.0). All sections were then blocked with normal goat serum and incubated overnight at 4°C with a prediluted solution of mouse anti-hAR. The sections were incubated with a goat secondary antimouse antibody, and then the streptavidin-biotin complex was applied, followed by a freshly prepared chromogen mix of buffered 3',3'-diaminobenzidine tetrahydrochloride (with heavy metal enhancement to give a black-positive reaction). The sections were dehydrated, cleared, and mounted with XAM neutral mounting medium (BDH Poole) without counterstaining. All staining runs included positive controls (CAP sections known to be hAR positive) and negative controls (sections of transitional bladder carcinoma incubated with anti-hAR and prostatic sections incubated with antimouse IgG instead of the mouse anti-hAR). As we described previously (14)
, all sections were stained within 10 days of preparation to exclude the loss of immunoreactivity due to storage-induced antigen degradation.
Evaluation of Immunochemistry.
hAR-positive immunoreaction was defined as black nuclear staining that was present in test sections stained with anti-hAR antibody but absent in negative controls. False positivity due to formalin pigments was excluded by viewing the slides with polarized light. Also, areas of inflammation, infarction, and tissue autolysis were not assessed. The hAR score was independently determined by two separate observers (E. O. O-O. and E. H. M.) by counting 400 luminal epithelial and stromal cells in normal, benign hyperplastic, and malignant glands and their surrounding stroma at x400 magnification from selected areas of the sections (see below). The number of cells with a positive nuclear reaction was then expressed as a percentage (15)
. The final score was the average of the scores awarded by the two observers. In cases where the scores given differed by a margin of >5, the sections were jointly reexamined, and a score was agreed upon. The final scores were classified as follows: 0, negative; 1–33%, weak expression; 34–66%, moderate expression; >66%, high/strong expression.
The level of hAR expression was first determined in areas of histologically homogeneous BPH and CAP. The effect of neoplastic epithelium on the surrounding stroma was also determined by assessing hAR positivity in areas of normal/atrophic glands and nodular (stromal) hyperplasia in BPH sections and in adjacent and distant benign (normal/atrophic and benign hyperplastic) glands and their surrounding stroma in CAP sections. Adjacent glands were defined as those within the same high-powered field, whereas distant glands were defined as being more than two high-powered fields from a malignant gland (16) .
Statistical Methods.
The results were reported as proportions within histological groups. The unpaired t test was used to compare the mean hAR scores of the different histological groups when the distributions were approximately normal, and the Mann-Whitney test or Fisher’s exact test was used otherwise. Comparison of the mean hAR scores in the different histological areas within the same sections was done using the paired t test or McNemar’s test as appropriate. The univariate association between stromal and epithelial hAR expression and hormone resistance was examined using Spearman’s correlation coefficient. All tests were two-sided and performed at the 0.05 level of significance.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
|
.
|
Stroma
Heterogeneous hAR expression was seen in the nuclei of stromal cells in all BPH sections. However, hAR immunostaining was highest in areas of nodular stromal hyperplasia (Fig. 2, A and B)
and decreased progressively in stroma surrounding normal/atrophic and BPH glands (P = 0.02 and 0.0004, respectively).
|
. However, hAR expression was significantly higher in normal/atrophic glands than in BPH glands within the same sections (P = 0.002).
Androgen Receptor Expression in CAP Sections
hAR staining in both epithelial and stromal nuclei in CAP sections was more heterogeneous than that in BPH sections (Fig. 3, A and B)
. Whereas all histologically benign epithelium within our sections expressed the receptor, 5 of 39 CAP glands had foci of malignancy that were hAR negative (Fig. 3C)
. One of these glands was a well-differentiated (G1) cancer, whereas two each were G2 and G3 tumors.
|
. Although the lack of stromal expression of hAR was more apparent in the higher-grade tumors, the difference between the mean scores of adjacent histological grades was not statistically significant (P = 0.28 and 0.69, respectively). Similar to CAP glands, the stroma around HG PIN glands in the majority of sections lacked hAR expression (HG PIN versus CAP, P = 0.61).
Stroma Surrounding Adjacent and Distant Benign Glands in CAP Sections.
Benign glands adjacent to CAP glands were also surrounded by hAR-depleted stroma, with a gradual reappearance of the receptor in the stroma around the more distant benign epithelium (Fig. 3, A and D)
. This gradual change was more easily seen in whole mount sections of radical prostatectomy specimens. The weak hAR expression in the stroma surrounding the adjacent benign glands was slightly higher than but not significantly different from that of stroma within homogeneous areas of malignancy within the same sections (P = 0.30). However, it was significantly less than the expression of the receptor by the stroma around distant benign glands of the same section (P = 0.006). In addition, hAR expression in the stroma surrounding the peripheral benign glands in CAP sections was statistically similar to that of the stroma around benign glands in BPH sections (P = 0.40; Figs. 2B
and 3D
).
Glandular Tissue
CAP Epithelium.
Despite the increased heterogeneity of hAR staining in malignant prostatic epithelium, the mean hAR expression in CAP glands was significantly less than that in normal/atrophic and BPH glands in BPH sections (P < 0.0001). Although the mean hAR expression was higher in the advanced cancers, there was no significant difference between the mean scores of adjacent histological grades (P = 0.39 and 0.62, respectively).
HG PIN Epithelium.
Mean hAR expression in HG PIN glands was intermediate between that of benign and malignant glands. However, whereas the HG PIN hAR mean score was significantly less than that of the BPH glands (P = 0.005), it was statistically similar to that of CAP glands (P = 0.5).
Adjacent and Distant Benign Epithelium in CAP Sections.
hAR expression was lower in adjacent histologically benign glands than it was in distant histologically benign glands in our malignant sections, but the difference was not statistically significant (P = 0.10; Fig. 3, A and D
). However, hAR immunopositivity was significantly less in these benign glands than in CAP or HG PIN glands within the same sections (P = 0.012 and 0.007, respectively; Fig. 3A
). Furthermore, the mean hAR score of benign glands in CAP sections was markedly lower than that of benign glands in BPH sections (P = 0.0001).
Stromal and Epithelial hAR versus Hormone Resistance in CAP
Comparison of hAR positivity in untreated and hormone-refractory CAP glands showed similar expression of the receptor in both the epithelium and the stroma of both types of glands (Table 2)
. Subsequent examination of the Spearman correlation coefficients of hormone insensitivity with hAR expression in the epithelial and stromal compartments by univariate analysis also revealed no significant association between the indices (P = 0.59 and 0.15, respectively).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In general, nuclear hAR expression in the epithelium of our specimens decreased as the tissues became increasingly neoplastic, whereas the heterogeneity of the staining increased. Interestingly, hAR positivity in benign glands in our CAP sections was significantly lower than that seen in the epithelium in BPH sections, but the significance of this previously unreported finding is presently unknown. Within the cancers however, the mean expression of the receptor was increased in the poorly differentiated (G3) tumors. The lower mean hAR score recorded in our moderately well-differentiated G2 tumors as compared to the well-differentiated cancers is probably due in part to the disproportionately higher number of androgen receptor-negative tumors within the group [20% (G2) versus 10% (G1) and 11% (G3), respectively]. However, we found no significant correlation between the expression of the receptor by epithelial nuclei and hormone resistance in our specimens. In addition, hAR expression was significantly higher in the epithelium than in the adjacent stroma, irrespective of glandular histotype. Because these findings are in keeping with those of previous studies (5 , 6 , 15) , the rest of this discussion will focus on the varied expression of the receptor in the stroma of our specimens.
We believe that this is the first definitive report of a more pronounced decrease in expression of hAR in stromal cells (as compared to the epithelium) as prostatic tissues dedifferentiate. We found only one report in the literature that noted the absence of the receptor in the stroma of a minority of sections from untreated CAP glands, and although the proportion increased after hormonal ablation, the difference was not significant (17) . This is in contrast to comments from other studies which, although directed at testing other hypotheses, reported some heterogeneous staining of stromal cells irrespective of the histology of the adjacent epithelium (3 , 15 , 18) . The validity of our results is supported by the demonstrable specificity of the antibody used and the similarity of the pattern of hAR expression in the epithelial compartment of our specimens with those of previous studies. In addition, technical artifacts were excluded by the presence of heterogeneous staining in the adjacent malignant epithelium.
Several reasons may be responsible for the novelty of our findings, and these would include: (a) the specific aim, ab initio, to determine the expression of hAR in the stromal and epithelial cells in our sections; (b) the use of archival sections of BPH and CAP in which the preservation of tissue architecture made it possible to evaluate staining in the two cellular compartments; (c) the use of a monoclonal anti-hAR antibody that recognizes antigens on archival material; and (d) the evaluation of the effect of proximity to a neoplastic epithelium on hAR expression in the stroma in both BPH and CAP sections.
We are presently unable to explain the absence of hAR positivity in the stroma around neoplastic epithelium in our sections, but one possible cause is the presence of mutant receptors that are not recognized by the antibody used. However, although there have been several reports of aberrant receptors in cancerous prostate tissues (19, 20, 21) , in general, they are considered to be uncommon in primary prostatic tumors. The finding may also be due to abnormal interactions between the stromal and epithelial elements of the malignant prostate gland. Previous experiments have shown that mitotic human epithelium is capable of inducing neoplastic changes in the adjacent stroma (22 , 23) , although the responsible mechanisms are unknown. The graded reduction of hAR expression in the stroma in the vicinity of altered prostatic glands, culminating in the total absence of the receptor in the stroma surrounding fully malignant cells, may therefore be due to the effect of as yet undetermined tumorigenic signals from transformed epithelial cells. The similarity of the depletion of hAR in the stroma surrounding both cancerous and adjacent benign glands in our CAP sections, whereas the stroma surrounding distant benign epithelium in the same sections was hAR positive may also be important, because it suggests that the expression of the receptor in the stromal nuclei may be determined primarily by the proximity of a malignant focus, and not the histology of the adjacent gland. This in vivo evidence is in support of previous suggestions from human and animal studies that local extension of primary and metastatic cancers may rely, at least in part, on the mutagenic field effect of the altered surrounding stroma (24, 25, 26) . When considered with the detection of a decreased hAR expression in adjacent benign glands in CAP sections as compared to the glands in BPH sections, it seems possible that there is a similar oncogenic field effect in the epithelial compartment as well. However, further research is required to confirm these findings.
There was no significant association between the loss of stromal hAR expression and hormone resistance in our specimens, and this is probably due to the absence of expression of the receptor in the stroma of untreated and hormone-refractory prostatic glands alike. However, the loss or mutation of hormone receptors is thought to be one of the pathways by which hormone-sensitive tumors acquire hormone resistance. The possibility that changes in stromal hAR expression (such as those demonstrated in this study) in prostatic neoplasms could be significant in this regard is strengthened by the importance of the intermediary role of this cell type in the mediation of androgenic control of the gland (24) . Furthermore, because hAR is required for the expression of peptide growth factors, it is possible that with the loss of this receptor, stromal cells cease to express these proteins early in the malignant transformation of the prostate. The survival of malignant cells would then rely on their ability to switch to hormone-independent cellular pathways (including those responsible for autonomous epithelial proliferation such as we have recently described4 ) sooner rather than later following transformation. In contrast, hAR gene amplification in epithelial cells has been detected only in tumors that have gradually become hormone resistant, i.e., later in the natural history of the disease (6 , 7) . Thus, it seems possible that those alterations in the molecular biology of the stromal compartment of the prostate involved in the progression to hormone-refractory disease occur during or soon after malignant transformation of the epithelium, perhaps preceding the genetic changes in the epithelial cells that are thought to contribute to the same phenotype.
We have found that hAR expression decreased in human prostatic epithelial and stromal cells in proportion to the degree of dedifferentiation of the tissues, with the decline being more pronounced in the stroma. In view of the hypothesis that mutagenic epithelia induce changes in the surrounding stroma, this report of a loss of expression of the receptor in the stroma surrounding CAP glands may be the first in vivo evidence of the possible mechanisms by which tumor stroma is formed. This abnormal stroma may then play an important role in both the progression of malignant transformation within the gland and the transmission of the potential for hormone-independent growth. We believe that the similarity in lack of hAR expression in the stroma in untreated and hormone-resistant CAP cases suggests that the cellular events responsible for the development of hormone resistance may occur during or shortly after carcinogenesis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work is dedicated to Prof. T. F. Solanke (contemporary). These results were presented in part at the British Association of Urological Surgeons Meeting in Bournemouth, United Kingdom, June 21–24, 1997 and at the Schilling Research Conference, Hormones and Cancer in Santa Cruz, California, September 18–21, 1997. 
2 To whom requests for reprints should be addressed, at Department of Urology, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW, United Kingdom.
3 The abbreviations used are: hAR, human androgen receptor; CAP, carcinoma of the prostate; BPH, benign prostatic hyperplasia; HG, high grade; PIN, prostatic intraepithelial neoplasia.
4 E. O. Olapade-Olaopa, D. K. Moscatello, E. H. MacKay, T. Horsburgh, D. P. S. Sandhu, T. R. Terry, A. J. Wong, and F. K. Habib, submitted for publication.
Received 6/26/98; revised 9/25/98; accepted 8/12/98.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M. Shen and C. Abate-Shen Pten Inactivation and the Emergence of Androgen-Independent Prostate Cancer Cancer Res., July 15, 2007; 67(14): 6535 - 6538. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Cano, A. Godoy, R. Escamilla, R. Dhir, and S. A. Onate Stromal-Epithelial Cell Interactions and Androgen Receptor-Coregulator Recruitment Is Altered in the Tissue Microenvironment of Prostate Cancer Cancer Res., January 15, 2007; 67(2): 511 - 519. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Duff and I. J. McEwan Mutation of Histidine 874 in the Androgen Receptor Ligand-Binding Domain Leads to Promiscuous Ligand Activation and Altered p160 Coactivator Interactions Mol. Endocrinol., December 1, 2005; 19(12): 2943 - 2954. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Y. Liu, M. P. Roudier, and L. D. True Heterogeneity in Primary and Metastatic Prostate Cancer as Defined by Cell Surface CD Profile Am. J. Pathol., November 1, 2004; 165(5): 1543 - 1556. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. K. Habib, M. Ross, C. W. Bayne, P. Bollina, K. Grigor, and K. Chapman The Loss of 5{alpha}-Reductase Type I and Type II mRNA Expression in Metastatic Prostate Cancer to Bone and Lymph Node Metastasis Clin. Cancer Res., May 1, 2003; 9(5): 1815 - 1819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Cinar, K. S. Koeneman, M. Edlund, G. S. Prins, H. E. Zhau, and L. W. K. Chung Androgen Receptor Mediates the Reduced Tumor Growth, Enhanced Androgen Responsiveness, and Selected Target Gene Transactivation in a Human Prostate Cancer Cell Line Cancer Res., October 1, 2001; 61(19): 7310 - 7317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Gao, J. T. Arnold, and J. T. Isaacs Conversion from a Paracrine to an Autocrine Mechanism of Androgen-stimulated Growth during Malignant Transformation of Prostatic Epithelial Cells Cancer Res., July 1, 2001; 61(13): 5038 - 5044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Henshall, D. I. Quinn, C. S. Lee, D. R. Head, D. Golovsky, P. C. Brenner, W. Delprado, P. D. Stricker, J. J. Grygiel, and R. L. Sutherland Altered Expression of Androgen Receptor in the Malignant Epithelium and Adjacent Stroma Is Associated with Early Relapse in Prostate Cancer Cancer Res., January 1, 2001; 61(2): 423 - 427. [Abstract] [Full Text]
|
|
|
|
|
|
A. Krishan, A. Oppenheimer, W. You, R. Dubbin, D. Sharma, and B. L. Lokeshwar Flow Cytometric Analysis of Androgen Receptor Expression in Human Prostate Tumors and Benign Tissues Clin. Cancer Res., May 1, 2000; 6(5): 1922 - 1930. [Abstract] [Full Text]
|
|
|
|
 |
|
 C. E. Petre, Y. B. Wetherill, M. Danielsen, and K. E. Knudsen Cyclin D1: Mechanism and Consequence of Androgen Receptor Co-repressor Activity J. Biol. Chem., January 11, 2002; 277(3): 2207 - 2215. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
