Cytometry (Clinical Cytometry) 50:25–30 (2002) DOI 10.1002/cyto.10050
Androgen Receptor Expression and DNA Content of Paraffin-Embedded Archival Human Prostate Tumors Satish Kumar Adiga,1 Ilia Andritsch,2 Ravikala Vittal Rao,3 and Awtar Krishan2* 1
Department of Obstetrics and Gynecology, Kasturba Medical College, Manipal, India Division of Experimental Therapeutics, Department of Radiation Oncology, Miami, Florida 3 Department of Pathology, Kasturba Medical College, Manipal, India
2
Background: Androgen receptors (AR) are expressed in human prostate cells and immunohistochemistry has been used for qualitative analysis of AR expression in prostate tumor cells. Quantitative and multiparametric analysis of receptor expression could be of diagnostic and prognostic value in the management of patients on antiandrogen therapy. Multiparametric flow cytometric methods have been developed for analysis of hormone receptor expression and DNA content in nuclei isolated from formalin-fixed/paraffinembedded human solid tumors. The present study was undertaken for analysis of AR expression and DNA content in archival human prostate tumors. Methods: AR expression and DNA content were measured in nuclei isolated by enzyme digestion from thick sections cut from 51 paraffin-embedded human prostate tumors. AR expression in different subpopulations was studied by gated analysis. The relationship among AR activity, DNA content, and histopathological grade was analyzed. Results: Distinct aneuploid populations were observed in 23% of tumors examined. AR activity was observed in all the specimens and the percentage of AR- positive nuclei in the 48 samples analyzed was 51% (n ⴝ 5). Tumor subpopulations with aneuploid DNA content had higher AR expression (percent AR-positive cells and mean log fluorescence) than the diploid subpopulations. No strong correlation was seen between AR expression and histopathological grade of the tumors. Conclusions: Flow cytometric analysis of archival prostate tumor can be used for rapid determination of aneuploid DNA content and AR expression in subpopulations of nuclei isolated from formalin-fixed/paraffin-embedded prostate tumor blocks. Cytometry (Clin. Cytometry) 50:25–30, 2002. © 2002 Wiley-Liss, Inc. Key terms: androgen receptors; prostate cancer; flow cytometry; DNA ploidy
Intracellular action of the male sex hormone (androgen) is mediated by the androgen receptor (AR), which is a key element of the nuclear hormonal signal transduction cascade and a target of endocrine therapy. In view of its importance as a cellular marker of clinical importance, precise and accurate quantitation of AR expression in prostate tumor cells may be important. AR expression in human prostate tumor cells can be determined by radioligand binding assays performed on tissue homogenates or by immunohistochemical staining of sections or smears. The biochemical methods for quantitation of AR expression in cytosolic fractions of tissue homogenates cannot discriminate between AR expression of the tumor cells and that of the nonmalignant epithelial and stromal cells. With the availability of AR antibodies, immunohistochemistry (IHC) has become the primary tool for detection of AR expression in tumor cells (1,2). However, IHC methods in general are not quantitative and cannot determine simultaneously the expression of AR and other cellular markers (3,4).
© 2002 Wiley-Liss, Inc.
Recently, multiparametric flow cytometric methods have been developed successfully for quantitative determination of estrogen, progesterone, and AR expression in nuclei isolated from paraffin-embedded human breast and prostate tumors (5,6). The advantage of flow cytometry over IHC methods is that it can provide quantitative results on antigen expression and it can determine the percentage of nuclei with positive receptor expression. Furthermore, multiparametric flow cytometric analysis allows for simultaneous monitoring and correlation of several different cellular markers of diagnostic and prognostic significance (e.g., DNA content, AR, estrogen receptor, progesterone receptor, and vitamin D receptor expression). However, the major disadvantage of flow cytometry
*Correspondence to: Awtar Krishan, Ph.D., Division of Experimental Therapeutics (R-71), Department of Radiation Oncology, University of Miami Medical School, P.O. Box 016960, Miami, FL 33101. E-mail:
[email protected] Received 4 October 2001; Accepted 29 November 2001
26
ADIGA ET AL.
is that it does not allow morphological examination of the cells in relation to their receptor expression under a microscope. Recently, we described a flow cytometric method for the analysis of AR expression in formalin-fixed/paraffinembedded archival prostate tumors (6). The present study was undertaken to study AR expression and the DNA content of nuclei isolated from 51 archival formalin-fixed/ paraffin-embedded human prostate tumors and to determine if any correlation exists among AR expression, DNA content, and the histopathology of the tumors. MATERIALS AND METHODS Human Prostate Tissues Formalin-fixed/paraffin-embedded tissue blocks obtained from the Department of Pathology, Kasturba Medical College (Manipal, India) were used for the present study. From each selected block, 50-m sections (for enzyme digestion) and 10-m sections (for histological examination) were obtained. Gleason histological grading system was used to rate the tumors on a scale of 1–5, from simple round closely packed glands to anaplastic adenocarcinomas (7). A hematoxylin/eosin-stained slide was used to confirm the presence of tumor cells in sections used for enzyme digestion and flow cytometric analysis. Antigen Retrieval and Enzyme Digestion Thick sections were deparaffinized in xylene, rehydrated in a descending ethanol series, and resuspended in 0.01 M citrate buffer (pH 6.0). Antigen retrieval was performed by heating the sections in a water bath at 80°C for 45 min. After 30 min of cooling at room temperature, the sections were washed in phosphate-buffered saline (PBS) and collected by centrifugation. Tissue digestion was done in 0.05% pepsin (Sigma, St. Louis, MO, catalog no. 7012, pH 1.65) in a water bath at 37°C. After 30 min, proteolytic action was terminated by the addition of 3% fetal bovine serum (FBS) in PBS. The resulting digest was filtered through a 40-m nylon mesh, washed in PBS, and stored at 4°C. Staining for AR Expression and DNA Analysis The isolated nuclei were incubated overnight with 150 l of anti-AR antibody (MU256-UC, clone F39.4.1, Biogenex, San Ramon, CA) at a 1:45 dilution in PBS at 37°C in the dark. The negative isotype control used for the anti-AR antibody was normal mouse IgG1 (MOPC 21, Sigma catalog no. M5284) at a 1:180 dilution in PBS. Samples were washed with 0.05% Triton-X in PBS and stained with 150 l of secondary antibody (fluorescein isothiocyanate [FITC]-conjugated anti-mouse IgG antibody, catalog no. 4143, Sigma) at a 1:80 dilution in PBS for 45 min at room temperature in the dark. After centrifugation, nuclei were washed with 0.05% Triton-X in PBS and the resulting pellet was resuspended in PBS for flow cytometric analysis. For simultaneous monitoring of AR expression and nuclear DNA content, AR-FITC-stained samples were incubated with propidium iodide (PI; 25 g/ml ⫹ 0.5 mg/ml RNAse) for 30 min at 37°C.
FIG. 1. Plot of DI of the isolated nuclei versus Gleason grade of the tumor biopsy. Correlation coefficient value of 0.16 indicates that aneuploidy was more common in tumors with a higher score.
Flow Cytometry Samples were analyzed on an XL-MCL flow cytometer (Beckman/Coulter, Miami, FL) with the standard argon ion laser excitation and filter selection for the FITC/PI dye combination. The relative fluorescence intensity was determined by dividing the mean log fluorescence (MCF) value of the antibody-reacted samples by that of the isotype controls. To determine the percent of AR-positive nuclei in a sample, the subtract function (by Overton’s method) in EXPO-32 software (Beckman/Coulter) was used for comparing the AR-mAB-stained and the isotype samples. ModFit.Lt program from Verity House (Topsham, ME) was used for cell cycle analysis. SigmaPlot 4.01 software (SPSS, Chicago, IL) was used for graphics and determination of correlation coefficients. RESULTS Of the 51 paraffin-embedded prostate tumors analyzed in the present study, dual parametric data on both AR expression and DNA content were obtained from 48 samples. In the three remaining samples, repeated attempts at generation of analyzable DNA histograms did not succeed even though positive staining for AR expression was excellent. Of the 48 samples analyzed, 38 had typical DNA distribution histograms of diploid tumors with distinct G0/G1 and G2/M peaks. In 10 of 48 samples, a distinct subpopulation of cells had aneuploid DNA content. Four of these had a DNA index (DI) of 1.5–1.67 whereas 6 of 10 had a DI of 1.72–2.0. Of the 10 tumors with aneuploid DNA content, 6 were Gleason histopathological grade 1, 3 were grade 5, and 1 was grade 3. Figure 1 shows that the DI increased with increasing Gleason grade and that aneuploidy was common in the more advanced grade tumors. AR activity was observed in all the specimens examined for dual parametric expression of DNA content and AR expression. Figure 2 shows three representative overlay histograms with low, medium, and high AR expression.
FLOW CYTOMETRIC ANALYSIS OF AR
27
FIG. 3. Plot of DI versus the percent of AR-positive nuclei. Lack of a strong correlation was indicated by the correlation coefficient of 5.99⫺4.
grade of the tumors examined. Figure 3 shows that the correlation between the DI and the percent AR-positive nuclei in the samples examined was not very good (r ⫽ 5.99-4). Similarly, there was not a strong correlation between the percent of AR-positive nuclei and the Gleason grade of the tumor (Fig. 4) or between the MCF value and the Gleason grade (Fig. 5). AR Expression in Diploid Versus Aneuploid Nuclei We used gated analysis by WinMidi software (version 2.8, Joseph Trotter,
[email protected]) to compare AR expression of the diploid and the aneuploid nuclei in a heterogeneous tumor population. The DNA histograms in Figure 6 are of representative tumor samples with diploid and aneuploid subpopulations. Overlay histograms (E-H) are of the isotype and the mAB-treated total samples (diploid and aneuploid nuclei). Figure 6I-L and M-P shows the AR reactivity of the gated diploid and the aneuploid populations, respectively. A comparison of the subtraction values obtained (MAB minus isotype percent of AR-posi-
FIG. 2. Overlay histograms of AR expression in the isotype (light line) and in the AR antibody-treated samples (dark line). The percentage of AR-positive cells varied from 10.5% to 68% with the ratio of the MCF ranging from 1.71 to 7.5 in three representative samples.
The number of positive nuclei in these three specimens varied from a low of 10.5% to 68% with MCF ratios (isotype versus mAB stained) of 1.71–7.5. The percentage of AR-positive nuclei in the 48 samples analyzed was ⬍10% (n ⫽ 4), 11–50% (n ⫽ 39), and ⬎51% (n ⫽ 5). Based on AR expression data showing the percent of AR-positive nuclei and the ratio of the MCF peaks, we sought to correlate these values with the DI and Gleason
FIG. 4. Plot of the Gleason grade of the prostate tumor versus the percent of AR-positive nuclei seen by flow analysis.
28
ADIGA ET AL.
FIG. 5. Plot of the Gleason grade of the tumor biopsy versus the MCF ratio of AR expression.
tive nuclei) shows that, in general, the aneuploid populations had more AR-positive nuclei than the diploid populations (Table 1). DISCUSSION Initially, prostate cancer is an androgen-dependent disease and hormonal ablation therapy is used to control its growth. However, tumor progression is often associated with hormone refractory disease (7). In view of the diagnostic and prognostic value of estrogen receptor expression in breast cancer, it was expected that AR status might be a useful clinical indicator of the disease course and of similar value for the management of prostate cancer patients (4). However, studies on the evaluation of AR expression in human prostate tumors have been controversial. Although several authors have reported heterogeneous or decreased AR expression in a subset of hormone refractory tumors (8,9), others have reported high AR expression (10,11). Recent studies have also indicated that stromal and epithelial cells in prostate tumors may have altered AR expression compared with that of cells in the normal prostate. Olapade-Olaopa et al. (12) reported that AR expression in the normal tissues surrounding the tumor cells is muted. In a recent study, very high AR expression was observed in androgen-independent prostate cancer, suggesting that the AR signaling pathway is important in the progression of prostate cancer during endocrine treatment (13). Similarly, Henshall et al. (14) evaluated the pattern of AR expression in the stroma and epithelium. They found concurrent overexpression of AR (⬎70% positive nuclei) in the malignant epithelium and loss of AR immunoreactivity in the adjacent periepithelial stroma (⬍30%) associated with poor clinical outcome in prostate cancer. IHC is a standard and universal method for monitoring AR expression in prostate cells. However, in most laboratories, these data are semiquantitative, which makes comparison of the data from the various studies difficult. In contrast, laser flow cytometric evaluation of hormone
receptor expression in tumors offers the advantage that quantitative data on percent of AR-positive cells and antigen density of a population (MCF) can be collected for correlative and comparative studies. Flow cytometry also offers the advantage that multiparametric determination of receptor or other cellular marker expression can be performed in gated subsets of a heterogeneous tumor population. As shown in the present study and our earlier report (6), hormone receptor expression of diploid versus aneuploid cells in a multiploid tumor can be determined and compared. Data from the present study show that archival formalin-fixed/paraffin-embedded prostate tumor blocks can be analyzed readily for flow cytometric determination of AR expression and its correlation with the DNA content of the individual isolated nuclei. AR expression was observed in most of the archival prostate tumors analyzed and the AR activity was significantly higher in the aneuploid nuclei compared with that of the nuclei with the diploid DNA content. This might be due to the underexpression of AR activity in diploid cells rather than to the overexpression seen in aneuploid cells in advanced tumors as normal prostatic cells and cells from benign prostatic hyperplasia (BPH) have very high AR expression (6). In formalin-fixed tissues, antigen retrieval is often necessary to improve reactivity with the anti-AR antibodies. However, use of antigen retrieval methods for processing formalin-fixed archival tissues can result in loss of DNA and generation of poor DNA histograms. A similar problem was observed in our previous study on the detection of estrogen and progesterone receptor expression in breast cancer cells from archival specimens (5). The loss of DNA reactivity may be caused by heating the cells in the citrate buffer at an acidic pH. However, as noted in the present study, in some of the archival samples, we did not obtain good DNA histograms even though we did get good AR expression profiles. We believe these samples may not have been fixed properly in formalin. We had hoped that data generated on AR expression in the present study could be used for positive correlation with pathological grading of the tumors. However, we did not find any significant correlation between the percent of AR-positive cells or the MCF of the tumors analyzed with their Gleason grade. These observations are in agreement with previous studies by Brolin et al. (1) and Ruizeveld de Winter et al. (15) who also did not find a positive correlation between percent of AR-positive tumor cells determined by IHC and tumor grade or disease stage. The overall incidence of aneuploid tumor in the present study was 23% and AR expression among the aneuploid subpopulation was significantly higher than that of the diploid cells. A review of the published literature shows that aneuploidy is often seen in advanced prostate tumors and does seem to correlate with disease progression and tumor grade (16). Hussain et al. (17) reported that 22 of 97 (23%) radical prostatectomy patients had aneuploid tumors. Aneuploidy was present in 15 of 97 (18.9%) tumors with a Gleason score 5–7 versus 71.4% of those with a Gleason score of 8 –10. Similarly, Shankey et al. (18) re-
FLOW CYTOMETRIC ANALYSIS OF AR
29
FIG. 6. DNA distribution histograms of the representative tumors analyzed (A–D). E–H: Overlay histograms of the AR expression of the isotype and the AR antibody-treated total population. I–L, M–P: Overlay histograms of AR expression in the diploid and the aneuploid components of the total population, respectively. The percent of AR-positive cells was generally higher in the aneuploid populations than in the diploid or the total population.
ported that early prostate tumors have diploid DNA content and a low S-phase and that these tumors likely evolve into DNA tetraploid tumors with a similar low S-phase fraction. Warzynski et al. (19) observed that 30 of 75 (40%) patients showed DNA heterogeneity in multiple samples of the same prostate. Overexpression of the AR receptor in aneuploid and metastatic tumors probably indicates the hormone refrac-
tory nature of these advanced tumors. The general consensus is that mutation frequency increases in most advanced stages of the disease. Recent studies also suggest that estrogen and its receptor and other circulating hormones may be involved in the interaction with the mutated and overexpressed AR receptor (20,21). In conclusion, flow cytometric analysis of archival prostate tumors may allow the identification of aneuploid
30
ADIGA ET AL.
Table 1 AR Distribution in Relation to DNA Ploidy Sp. no. 1 2 3 4 5 6 a
DNA contenta Diploid Aneuploid 95 64 56 87 75 65
5 36 44 13 25 35
DI of aneuploid cells
Total
1.95 1.72 1.5 1.73 1.67 1.5
2.66 25.69 10.51 60.96 17.32 36.84
AR-positive cells Diploid Aneuploid 5.32 25.98 24.23 67.01 18.67 27.99
8.16 44.94 29.92 85.92 31.12 50.94
Diploid 1.32 1.74 1.95 8.9 1.33 3.1
MCF ratio Aneuploid 1.66 1.93 2.2 8.54 1.38 3.0
Percent of population with diploid/aneuploid DNA content.
tumors as well as the quantitative determination of their AR expression. Thus, it may be possible to carry out retrospective studies correlating hormone receptor expression status with mutations and the clinical response to hormonal and other therapeutic modalities. ACKNOWLEDGMENTS Dr. Adiga was supported as a visiting research fellow by a grant from the American Cancer Society (Florida Division). LITERATURE CITED 1. Brolin J, Skoog L, Ekman P. Immunohistochemistry and biochemistry in detection of androgen, progesterone, and estrogen receptors in benign and malignant human prostatic tissue. Prostate 1992;20:281– 295. 2. Sweat SD, Pacelli A, Bergstralh EJ, Slezak JM, Bostwick DG.Androgen receptor expression in prostatic intraepithelial neoplasia and cancer. J Urol 1999;161:1229 –1232. 3. Tilley WD, Lim-Tio SS, Horsfall DJ, Aspinall JO, Marshall VR, Skinner JM. Detection of discrete androgen receptor epitopes in prostate cancer by immunostaining: measurement by color video image analysis. Cancer Res 1994;54:4096 – 4102. 4. Prins GS, Sklarew RJ, Pertschuk LP. Image analysis of androgen receptor immunostaining in prostate cancer accurately predicts response to hormonal therapy. J Urol 1998;159:641– 649. 5. Redkar AA, Krishan A. Flow cytometric analysis of estrogen, progesterone receptor expression and DNA content in formalin-fixed, paraffin-embedded human breast tumors. Cytometry 1999;38:61– 69. 6. Krishan A, Oppenheimer A, You W, Dubbin R, Sharma D, Lokeshwar BL. Flow cytometric analysis of androgen receptor expression in human prostate tumors and benign tissues. Clin Cancer Res 2000;6: 1922–1930. 7. Gleason DF. Histologic grading of prostate cancer: a perspective. Hum Pathol 1992:23:273–279. 8. Hobisch A, Culig Z, Radmayr C, Bartsch G, Klocker H, Hittmair A. Distant metastases from prostatic carcinoma express androgen receptor protein. Cancer Res 1995;55:3068 –3072. 9. Kinoshita H, Shi Y, Sandefur C, Meisner LF, Chang C, Choon A, Reznikoff CR, Bova GS, Friedl A, Jarrard DF. Methylation of the androgen receptor minimal promoter silences transcription in human prostate cancer. Cancer Res 2000;60:3623–3630. 10. Visakorpi T, Hyytinen E, Koivisto P, Tanner M, Keinanen R, Palmberg C, Palotie A, Tammela T, Isola J, Kallioniemi OP. In vivo amplification
11.
12.
13.
14.
15.
16. 17.
18. 19.
20. 21.
of the androgen receptor gene and progression of human prostate cancer. Nat Genet 1995;9:401– 406. Koivisto P, Kononen J, Palmberg C, Tammela T, Hyytinen E, Isola J, Trapman J, Cleutjens K, Noordzij A, Visakorpi T, Kallioniemi OP. Androgen receptor gene amplification: a possible molecular mechanism for androgen deprivation therapy failure in prostate cancer. Cancer Res 1997;57:314 –319. Olapade-Olaopa EO, MacKay EH, Taub NA, Sandhu DP, Terry TR, Habib FK.Malignant transformation of human prostatic epithelium is associated with the loss of androgen receptor immunoreactivity in the surrounding stroma. Clin Cancer Res 1999;5:569 –576. Linja MJ, Savinainen KJ, Saramaki OR, Tammela TL, Vessella RL, Visakorpi T. Amplification and overexpression of androgen receptor gene in hormone-refractory prostate cancer. Cancer Res 2001;61: 3550 –3555. Henshall SM, Quinn DI, Lee CS, Head DR, Golovsky D, Brenner PC, Delprado W, Stricker PD, Grygiel JJ, Sutherland RL. Altered expression of androgen receptor in the malignant epithelium and adjacent stroma is associated with early relapse in prostate cancer. Cancer Res 2001;61:423– 427. Ruizeveld de Winter JA, Janssen PJ, Sleddens HM, Verleun-Mooijman MC, Trapman J, Brinkmann AO, Santerse AB, Schroder FH, van der Kwast TH. Androgen receptor status in localized and locally progressive hormone refractory human prostate cancer. Am J Pathol 1994; 144:735–746. Zattoni F, Prayer Galetti T, Vianello F, Bertoldin R, Panozzo M, Blandamura S, Costantin G. Comparison between nuclear DNA and histologic grade of prostatic carcinoma. J Urol 1992;98:192–195. Hussain MH, Powell I, Zaki N, Maciorowski Z, Sakr W, KuKuruga M, Visscher D, Haas GP, Pontes JE, Ensley JF. Flow cytometric DNA analysis of fresh prostatic resections. Correlation with conventional prognostic parameters in patients with prostate cancer. Cancer 1993; 72:3012–3019. Shankey TV, Jin JK, Dougherty S, Flanigan RC, Graham S, Pyle JM. DNA ploidy and proliferation heterogeneity in human prostate cancers. Cytometry 1995;21:30 –39. Warzynski MJ, Soechtig CE, Maatman TJ, Goldsmith LC, Grobbel MA, Carothers GG, Shockley KF. DNA heterogeneity determined by flow cytometry in prostatic adenocarcinoma—necessitating multiple site analysis. Prostate 1995;27:329 –335. Li LC, Chui R, Nakajima K, Oh BR, Au HC, Dahiya R. Frequent methylation of estrogen receptor in prostate cancer: correlation with tumor progression. Cancer Res 2000;60:702–706. Zhao XY, Malloy PJ, Krishnan AV, Swami S, Navone NM, Peehl DM, Feldman D. Glucocorticoids can promote androgen-independent growth of prostate cancer cells through a mutated androgen receptor. Nat Med 2000;6:703–706.
