Correlation of Biochemical (Receptors, Endogenous Tissue Hormones) and Quantitative Morphologic (Stereologic) Findings in Normal and hyperplastic Human Prostates

0022-5347 /87 ('._3'?3-D558$02,0/0 -;loL 13'!, 1\d:a:rch

TEE JOURNAL OF UROL0G~i

Copyxight © 1987 by The iNJliams & \ffilkins Co.

Printed in U.S.A.

CORRELATION BIOCHElVHCAL (RECEPTORS, ENDOGENOUS TISSUE HORMONES) QUANTITATIVE MORPHOLOGIC (STEREOLOGIC) FINDINGS IN NORMAL AND HYPERPLASTIC HUMAN PROSTATES G. BARTSCH,* F. KEEN, G. DAXENBICHLER, CH. MARTH, R MARGREITER, A. BRUNGGER, T. SUTTER, AND H. P. ROHR From the Department of Uro/,ogy, Gyneco/,ogy and Surgery, University of Innsbruck, Av..stria and the Department of Pathology, University of Bas/,e, Switzerland

ABSTRACT

In previous light and electron-microscopic analyses human benign prostatic hyperplasia was shown to be predominantly a stromal disease; the aim of the present study was to correlate the stereological data with the levels of the endogenous tissue hormones (androgens, estrogens, progesterone) in normal (N) and hype:rplastic human prostatic tissues (BPH). BPH tissue specimens were obtained by open prostatectomy (n = 25); normal prostatic tissue was obtained from kidney donors (n=

No statistically significant difference was found between normal and hyperplastic tissue. Testosterone BPH 0.69 ± 0.44, N 0.25 ± 0.12; 5a-dihydrotestosterone BPH 7,0 ± 2.9, N 4.2 ± 0.7; progesterone BPH 0.059 ± 0.022, N 0.058 ± 0.005; estrone BPH 0.10 ± 0.03, N 0.14 ± 0.03; estradiol BPH 0.07 ± 0.02, N 0.05 ± 0.02; estriol BPH 0.02 ± 0.01, N 0.04 ± 0.02. Using a Spearman rank correlation coefficient a statistical analysis was performed for age, weight of the prostate, absolute stereological data and the endogenous ~".~~<-n<-,,~ hormones. As can be seen from the statistical analysis there is a poor correlation for 5a-dihydrotestosterone and the amount of the glandular epithelium; otherwise no correlation of the endogenous tissue hormones with the stereological data investigated was found. These data show that the stromal overgrowth of benign hyperplasia is not reflected in the tissue hormone levels. In previous light- and electron-microscopic analyses human benign prostatic hyperplasia was shown to be primarily a stromal disease with actions on the glandular part of the hyperplastic tissue. 1 • 2 Sixty per cent of the hyperplastic tissue in benign prostatic hyperplasia consists of stromal tissue including collagen fibers, fibroblasts and smooth muscle cells, whereas the glandular cells account for 12 per cent of the hyperplastic tissue. 1 • 2 In human benign an activation of the smooth muscle cells of the stroma was shown. 2 Neubauer and Mawhinney showed that the stroma of the male accessory sex organs has a high concentration of estrogen molecules; Twcmo""rl that estradiol contribute physiologically to the growth and function of male accessory smooth muscle tissue. 3 In castrated dogs the stroma can be activated by high dosages of 17/3-estradiol. The initial studies of Siiteri and WilsoP. in 1979 showed that 5a-dihydrotestosterone levels are elevated in cases of spontaneous development of human benign prostatic hyperplasia. 5 However, in 1984 Walsh et al. reported for the first time that the 5a-dihydrotestosterone levels in normal and hyperplastic prostatic tissue obtained open surgical procedures were similar. 6 At present there are strong indications that it is primarily the androgens which control the function of the stromal part of the human prostate. 7- 10 On the other hand, it is proposed that estrogens may contribute to the growth and function of the male accessory gland smooth muscle cells. The Accepted for publication September 30, 1986. * Requests for reprints: Universitiitsklinik fur Urologie, A-6020 Innsbruck, Anichstrasse 35, Austria. Supported by Swiss National Foundation grant no. 3.190.-0.82 and Fonds zur Fiirderung der wissenschaftlichen Forschung Nr. 4020, Austria. 559

aim of the present study was to correlate the quantitative morphologic (stereologic) and the biochemical (receptors, endogenous tissue hormones) values of normal and hyperplastic tissues to determine if the altered steroid metabolism in BPH is reflected in the morphologic architecture. MATERIALS AND METHODS

Benign prostatic hyperplasia specimens were obtained open suprapubic prostatectomy; normal prostatic tissue was obtained from kidney donors. Fixation. The specimens were fixed in buffered formalin (pH Specimens used for paraffin blocks were taken as cross sections at four different levels of the prostate gland in a way that on one side there was epithelium of the urethra and on the other side the capsule of the outer part of the prostate could be seen. Each section from these blocks is a representative transverse crnss section of the inner and outer parts of the normal prostate. The tissue specimens from human benign prostatic hyperplasia were taken at three different levels of the enucleated material of the hyperplastic gland and fixed in buffered formalin (pH 7.4). In that way three paraffin blocks were prepared for each patient. Quantitative morphologic analysis. In order to evaluate the prostatic gland and its components in stereologic terms, a morphometric model of the human prostate was developed. 1 • 2 • 11 Fig. 1 shows how the human prostate was divided into morphologically defined compartments. Essentially the model has two major divisions-the stromal part (ST) including connective tissue, blood vessels, nerves and smooth muscle fibers, and the glandular part (EP) including the lumina of the acini and the glandular epithelial cells. Stereological evaluations were done on a magnification level of 80X.

560

BARTSCH AND ASSOCIATES

FIG. 1. Sterological model of prostatic tissue.

The following absolute stereological parameters were determined: Surface density of the glandular epithelium (S) Volume density of the stromal part (VsT) Volume density of the epithelial part (VEP) Volume density of the glandular lumen (Vw) Length density of the glandular tubules (L) Height of the glandular epithelium (H) Diameter of the glandular lumen (Dw) Diameter of the glandular tubules (DAP) Mean free distance between the glandular tubules (LAP) Mean distance between the centers of the glandular tubules (ScA).

"The volume densities were determined according to Rohr et al." P,

Vv,

= Pr

where i is the tissue component under consideration, Pi is the number of test points in the test system associated with i and PT is the number of points of the test system. To relate the volume densities of the acinar parenchyma, glandular cells, acinar lumina and interstitial (stromal) tissue the primary volume densities determined by point counting are converted into the volume densities of the reference system prostatic gland. Sampling procedures. There were four paraffin blocks used for stereological measurements of the normal human prostate and three used for measurements of benign prostatic hyperplasia. The paraffin blocks were taken at different levels of the prostate. For determination of the volume densities of the acinar parenchyma, glandular cells, acinar lumina and interstitial fibromuscular tissue of the normal human prostate, at least four, and of prostatic hyperplasia six paraffin sections stained with hematoxylin and eosin were selected. For each section stained with hematoxylin and eosin in benign prostatic hyperplasia at least 30 randomized test areas were analyzed with the sampling microscope of Wild by superimposing test lattice systems (121 and 1,089 test points). For sampling normal human prostate 20 randomized test areas of the inner part and 20 of the outer part were analyzed for each section. Therefore, from one patient 160 test areas were evaluated for the normal human prostate (inner and outer part) and 90 test areas were evaluated for benign prostatic hyperplasia. Data processing. Calculations were done with a HewlettPackard 9815A desk computer. Receptor and endogenous hormone analyses. Chemicals. (6, 73H) methyltrienolone (R1881), 55.5 Ci/mmol. and (6, 7- 3H) promogestone (R5020), 51.4 Ci/mmol., were purchased from New England Nuclear, Boston, MA; (2, 4, 5, 7-3H) estradiol,

(E2) 100 Ci/mmol., was purchased from the Radiochemical Centre, Amersham, England. Non-radioactive R1881 and R5020 were purchased from New England Nuclear; unlabelled diethylstilbestrol (DES) and triamcinolone acetonide (TA) were purchased from Serva, Heidelberg, W. Germany. Tissue. Benign prostatic hyperplastic tissue specimens were obtained by open suprapubic prostatectomy. The average age of the patients was 70 ± 9 years. The specimens were transported from the operating room to the laboratory within 15 minutes after surgical removal, washed in physiologic saline to remove any blood, weighed, dissected into pieces of appropriate size and then frozen in liquid nitrogen. The frozen tissue was placed in plastic vials and stored at -70C until assay. The time interval between tissue removal and tissue processing never exceeded three weeks. Normal prostatic tissue was obtained from kidney donors who had suffered brain death. The average weight of the normal prostates was 21.4 ± 3.6 gm; the average age of the donors was 24 ± 2 years. Extraction of steroids from tissue. One gm. of tissue was homogenized in two ml. of ethanol with an Ultra Turrax. The homogenizer was washed with two ml. of cold ether, which was then added to the homogenate. The extraction was performed overnight by vertical rotation of the tubes containing the homogenate. After centrifugation the supernatant was transferred to a second tube and the pellet was washed with four ml. of ether. Then the supernatant was again centrifuged and added to the first one. The combined extracts were dried with a stream of nitrogen. The residue was dissolved in two ml. of methanol and water (volume 90/10) and extracted with five ml. of petrol ether. This procedure serves to remove most of the lipid. Two ml. of water was added to the methanol-ether layer and the mixture was extracted three times with two ml of ether. Before being combined each ether extract was washed with the same two ml. of water. Finally, the combined ether extracts were dried with nitrogen. The recovery for testosterone was 70%, for 5a-dihydrotestosterone 69%, for estrone 85%, for estradiol 89% and for estriol 93%. Radioimmunoassay. Testosterone and 5a-dihydrotestosterone were separated and purified using celite microcolumns and then measured by radioimmunoassay. Estradiol, estriol and estrone were separated on sephadex LH 20 columns. Estradiol and estriol were analyzed by radioimmunoassay with an antiserum reacting with estrone and estradiol to the same extent; the estriol radioimmunoassay was performed with a specific estriol antiserum. Preparation of cytosol and crude nuclear extract. The frozen tissue was pulverized under liquid nitrogen and the resulting powder was weighed and transferred to homogenizing flask. TEMG buffer (20 mM of TRIS, 1.5 mM of EDTA, 5.0 mM of monothioglycerol, 10% v/v glycerol, pH 7.4, 20C) containing 20 mM of sodium molybdate was added to the tissue powder in a ratio of 5:1 (vol/wt.) and homogenized with two to three UltraTurrax bursts, each of which lasted for 10 seconds (Janke and Kunkel, KF, Stanten in BN, W. Germany). The homogenizing flask was cooled for two min. between each burst. The crude homogenate was centrifuged at 800 g for 15 min. (4C) in an IEC Centra-7R Refrigerated Centrifuge (International Equipment Company, USA). The resulting supernatant was centrifuged at 150,000 g for 30 min. at 4C in a Beckman Model L8-55 Ultracentrifuge using a Type-65 fixed angle rotor. This supernatant was termed the cytosol fraction. The "crude nuclear pellet" resulting from the 800 g centrifugation was washed by resuspension in six volumes of TEMG buffer and subsequent centrifugation at 800 g for 10 min. This procedure was repeated twice and the remaining pellet was extracted with six volumes of TKMG buffer (20 mM of TRIS, 0.6 M ofKCl, 5 mM ofmonothioglycerol, 10% glycerol vol/vol, pH 8.5, 20C) for 30 min. in an ice-bath and subjected to frequent vortexing. The nuclear pellet-KC! mixture was centrifuged for 30 min. at 140,000 g and the supernatant

a

561

BIOCHEMICAL AND MORPHOLOGICAL FINDINGS IN NORMAL AND HYPERPLASTIC ff!JMAN PROSTATES

represented the nuclear extract. Binding studies. All receptors were measured means of titration analyses. Radioactive and non-radioactive steroids were prepared in ethanol solutions. The appropriate volume was pipetted into glass tubes and evaporated under vacuum. Two hundred µl. of cytosol or nuclear extract were incubated under the conditions described below. The androgen receptors were determined with the help of (3H) Rl881 in final concentrations from 0.5 to 8.0 nM either alone (total binding) or with radioactive steroid in the presence of 0.8 µM of nonradioactive hormone (non-specific binding). In all incubations there was a 1000-fold excess of TA, which inhibits the binding of Rl881 to the prostatic progesterone receptors. The cytosol and nuclear extract mixtures were incubated at 4C for 16 to 20 hours. The progesterone receptors were measured by means of (3H) R5020 alone (total binding) or (3 H) R5020 in the presence of 0.8 µM of un-labeled R5020. The concentrations used for saturation analyses ranged from 0.5 to 8.0 nM (final concentration). Both cytosol and nuclear extract were incubated at 4C for 16 to 20 hours. The estrogen receptors were measured by means of 3Hestradiol in the presence of 0.8 µM of diethylstilbestrol (nonspecific binding). 5a-dihydrotestosterone in a 100-fold molar excess was added to all tubes in order to prevent any binding of 3H-estradiol to testosterone-estradiol-binding globulin (TEBG}. The solutions were incubated at 30C for four hours. After incubation the unbound and loosely bound steroids were separated from the receptor-bound steroid by means of the dextran-coated-charcoal (DCC) method when determining the cytoplasmic receptors and by means of hydroxylapatite (HAP) adsorption when measuring the receptors in the nuclear extract. For the DCC assay 500 µl. of a 0.7% dextran-coated-charcoal solution (0.5% activated charcoal and 0.05% dextran T-70 in TEMG buffer, pH 7.6, 20C) were added to each assay tube; the suspensions were then mixed by means of a vortex mixer and allowed to incubate for 15 minutes at 4C. Subsequently they were centrifuged at 800 g for 10 min, Five hundred µl. aliquots of the supernatant were transferred to scintillation vials and radioactivity was counted. HAP adsorption was utilized for the measurement of receptors in the nuclear extract. Hydroxylapatite (DNA grade, Bio Rad, Richmond, CA, USA) was suspended in TEMG buffer (pH 7.6, 20C), allowed to settle, and then the fine particles were removed. This procedure was repeated three times and subsequently the washed HAP was suspended in two volumes of TEMG buffer. For the assay 500 ml. of HAP suspension were added to each incubation tube and incubated at 4C for 30 min. with gentle mixing at 10 minute-intervals. The suspensions were centrifuged at 800 g for 10 min. and the supernatant was discarded. The HAP pellets were washed twice with TEMG buffer containing 0.1 % Tween 80 (Serva, HeidelTABLE

Case

1 2 3 4 5

Age

8

80 72 63 52 73 76 72 79

9 10

62

6

7

71

m±SD 70 ± 9

1 2 3

24 26 21

Weight

s

(gm.)

Xl05 mm. 2

87 42 36 79 62 204 42 56 50 84 74 ± 49

19 24 22

m±SD 24 ± 2 21.7 ± 2.5

7.63 2.58 2.98 5.96 3.32 17,90 34.60 3.92 3.49 1.94

mm. 3

43.2 8.7 8.9 14.5 7.0 78,6 13.5 16.7 13.9 10.9

8.4 ± 10.3 21.6 ± 22.5

2.30 3.17 2.49 2.6 ± 0.4

5.32 6.78 5.57 5.9 ± 0.8

VEP Xl03 mm. 3

12.9 4.9 4.9 11.7 7.9 29.1 6.4 6.4 7.5 3.4 9.5 ± 7.5

4.23 7.38 4.55 5.3± 1.7

were berg, W. Germany). After the final wash the extracted with the help of 2.0 mL of pure ethanol one hour at room temperature and mixed frequently with a vortex mixer. These mixtures were centrifuged at 800 g for 10 min. and the radioactivity in the ethanol extract supernatant was measured. The binding site concentration and the apparent dissociation constant (Kd) were determined by means of the Scatchard plot analysis for which a program developed for a Hewlett Packard Model 9815 A programable desk top calculator was used. The receptor values are expressed in femtomol bound per gram of prostatic tissue (fm./gm.). RESULTS

Stereologic results. The absolute data show that in benign prostatic hyperplasia there is a statistically significant increase in the volumetric amounts of strnmal tissue and glandular lumina as compared to normal prostatic tissue; on the other hand no difference between normal and hyperplastic tissue can be found regarding the volumetric amounts of glandular epithelium; on account of the differing weights of the hyperplastic prostates great variation was observed. There is furthermore no difference regarding the surface density (S) and the height of epithelial cells (H). As a consequence of the highly significant increase in stromal tissue a significant increase in the mean free distance between the glandular tubules (LAP) and the centers of the glandular tubules (ScA) was found in hyperplastic tissue (table 1). Tissue steroid levels. Androgens. The endogenous concentrations of 5a-dihydrotestosterone and testosterone were measured in 10 patients with BPH and five normal prostates. A stereologic analysis was also performed on these specimens. Furthermore, the endogenous testosterone and 5a-dihydrotestosterone concentrations were determined in another 15 samples of BPH. No statistically significant difference was found between normal and hyperplastic tissues (normal tissue: 4.3 ± 0.8 ng./ml., BPH: 10 specimens: 7.0 ± 2.9 ng./ml., BPH: 15 specimens: 4.3 ± 0.6 ng./ml.) (table 2, 3). Progesterone and estrogens. Regarding the endogenous tissue levels, no significant differences were found between the estrone, estradiol, estriol and progesterone levels of normal and hyperplastic tissues (table 2). Receptor determination. High-affinity cytoplasmic R1881 binding was found in all BPH and normal speciments (fig. 2, table 4). The mean receptor concentration was 2710 ± 1455 fmol./gm. in hyperplastic tissue and 2192 ± 300 fmol./gm. in normal prostatic tissue. Concerning the receptor content there was no sig-.11ificant difference between these two groups. The mean equilibrium dissociation constant (KD) was 2.4 ± 1.6 nM

1. 8tereologic analysis

VsT Xl0 3 mm. 3

L

H

Dw

DAP

LAP

ScA

Xl05 mm.

I'

I'

I'

I'

I'

17 19 16 20

226 134 120 97

24 16 19 16 22 18

176 156 170 159 225

260 173 153 137 132 208 193 203 202 261

199 479 332 392 613 249 298 378 371 368

HYPERPLASTIC 31.4 18.3 28.4 8.1 22.3 7.4 52.6 15.0 13.9 46.9 97.0 50.0 22.6 11.4 33.8 13.2 12.l 28.6 16.1 5.6

84

459

652 485 529 745 457 492 581 574 629

38 ± 23.5 15.5 ± 12.7 18.7 ± 2.7 154.7 ± 47.9 192.2 ± 45.1 367.9 ± 116.4 560.3 ± 94.3

6.7 10.2 9.6

NORMAL 7.9 5.1 5.5

8.8 ± 1.8

6.2 ± 1.5

18 23 18 19.7 ± 2.7

93 85 89 89.0 ± 4.0

129 132 126 129.0 ± 3.0

152 176 190

382 308 316

172.6 ± 19.2

335.3 ± 40.6

562

BARTSCH AND ASSOCIATES TABLE 2. Endogenous steroid levels in normal and hyperp/,astic human prostatic tissue

Age

Case

1 2 3 4 5 6 7 8 9 10

Weight (gm.)

Testosterone (ng./gm.)

87 42 36 79 62 204 42 56 50

80 72 63 52 73 76 72 79

m±SD

62

84

70± 9

74±49

0.69 ± 0.44

24 26 21 24 26

16.2 24.4 19.7 25.0 22.0

0.17 0.14 0.21 0.43 0.33

1 2 3 4 5

24± 2

m±SD

HYPERPLASTIC 8.0 5.2 3.6 6.5 7.5 8.0 8.5 3.9 13.8 5.1

0.21 0.36 0.33 0.29 0.85 0.91 0.94 0.45 0.95 1.60

71

21.5 ± 3.6

Dihydrotestosterone (ng./gm.)

7.0 ± 2.9 NORMAL 3.4 4.7 5.4 4.0 3.9

0.25 ± 0.12

4.3 ± 0.8

Tissue Levels Case

Age

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

72 74 82 68 63 76 78 83 77 75 73 72 82 69 73

m±SD

74 ±5

Weight (gm.)

Testosterone (ng./gm.)

o----o specific binding

o---o non specific binding

4.4 3.5 5.1 4.0 3.8 3.6 5.1 4.7 4.7 3.5 3.6 4.4 4.8 5.4 4.4 4.3 ±0.6

0.75 ± 0.75

• - - • total binding B/F

12

0.12

JO

0.10

8

0.08

== 6 0..

0.06

';"

Dihydrotestosterone (ng.fgm.)

HYPERPLASTIC 67.4 0.21 31.6 0.31 83.8 0.16 70.0 0.20 97.2 0.30 42.1 0.30 85.1 0.60 102.2 0.30 79.4 0.30 73.2 0.70 116.2 2.80 123.4 1.40 92.0 0.90 0.70 25.7 36.1 0.20 75.0 ± 30.1

Nuclear R1881 Binding BPHt 2 Binding sites • 826 rmol / gram Ko • 3.8 x I0 - 9 M

:::: X

u

0.04 2

0.02

• 2

3 4 5 6 3H - R1881 nM

7

8

Estradiol (ng./gm.)

Estriol (ng./ gm.)

Progesterone (ng./gm.)

0.90 0.14 0.12 0.08 0.o7 0.13 0.04 0.12 0.09 0.14

0.05 0.06 0.06 0.03 0.06 0.07 0.04 0.06 0.04 0.11

0.03 0.02 0.02 0.02 0.02 0.03 0.04 0.02 0.02 0.03

0.770 0.074 0.055 0.050 0.083 0.055 0.048 0.048 0.036 O.o40

0.10 ± 0.03

0.06 ± 0.02

0.02 ± 0.01

0.059 ± 0.022

0.19 0.13 0.10 0.12 0.15

0.03 0.03 0.06 0.07 0.05

0.02 0.02 0.o7 0.03 0.04

0.055 0.065 0.063 0.061 0.048

0.14 ± 0.03

0.05 ± 0.02

0.036 ± 0.021

0.058 ± 0.007

± 11 fmol./gm.). This represents a significant difference. The

3. Androgens

TABLE

Estrone (ng./gm.)

0.02

0.08

0.06

0.10

B

Fm. 2. Saturation analysis of 3H R1881 binding to androgen receptor.

in BPH samples and 1.7 ± 0.1 nM in the normal specimens. These values are not significantly different. R1881 binding was detected in the nuclear extracts of eight out of 10 BPH (802 ± 320 fmol./gm.) and of two out of three normal prostates (392

Ko values for both of the aforementioned groups, which were 3.2 ± 0.9 nM (BPH) and 1.8 ± 0.1 nM (normal), did not differ significantly. Cytoplasmic R5020 binding was observed in all hyperplastic (2262 ± 634 fmol./gm.) and normal prostates (921 ± 174 fmol./ gm.), (fig. 3 and tab. 4). As can be seen from table 4 the levels of progesterone receptors detected in BPH samples were significantly higher than in .normal prostates. The average Ko value for the hyperplastic group was 4.8 ± 2.3 x 10-10 nM and 14.0 ± 2 X 10-10 in normal specimens, which represents approximately a three-fold difference. No high-affinity R5020 binding was detected in the high salt nuclear extracts of either group. Under the assay conditions employed no estrogen receptors were detectable in the cytosol or high salt nuclear extracts of BPH specimens. Estrogen binding was m.easured in all three normal prostatic cytosols (433 ± 274 fmol./gm., Ko 2.9 ± 0.8 nM) (fig. 4, tab. 4). No estrogen binding was detectable in the nuclear extracts of normal specimens. Correlation of stereologic and biochemical results. With the help of the Spearman rank correlation coefficient a statistic analysis wasperformed for age, weight of the prostate, absolute stereologic (S, Vw, VEP, VsT, L, H, Dw, DAP, LAP, ScA) and the following biochemical . data: endogenous __1>rostatic horinones (5a-cfihydrotestosterone, testosterone, progesterone, estradiol, estrorie and estriol): five samples; and receptors (androgens, estrogens and progesterone): three samples. As can be seen from the statistical analysis the endogenous 5a-dihydrotestosterone concentration and the age of the patient as well as the weight of the prostate do not show a statistically significant correlation coefficient; a statistically significant correlation could only be found for the glandular tissue and the surface density of glandular epithelium. Estriol, estradiol, estrone an.d endogenous progesterone steroid levels did not correlate significantly with the morphometric data investigated; regarding the receptor levels a statistically significant correlation was found between the nuclear androgen receptor and the volumetric amount of stromal tissue and as a consequence of stromal overgrowth to the mean free distance between the glandular tubules (LAP); regarding the cytoplasmic progesterone receptors, the diameters of the glandular lumina (Dw) and the diameters of the glandular tubules (DAP) were found to correlate significantly (table 5). DISCUSSION

In recent years stereology has proven to be an important tool for correlating morphologic with biochemical data obtained on

TABLE ,t Receptor analyses in normal and hyperplastic human prostatic tissue Androgens

Estrogens

Nuclear Case Age Cytoplasmic K 0 X 10-9 Receptor Ko Receptor (fmol./gm.) (fmol./gm.)

X

10-9

Cytoplasmic Receptor (fmol./gm.)

Nuclear Receptor Ko (fmol./gm.)

Ko+ 10-s

Prostereone X

10-9

Cytoplasmic Ko Receptor (fmol./gm.)

X

NucleaT Receptor K 0 (fmol./gm.)

10- 10

X

10- 10

HYPERPLASTIC 1 2

80 72 63 52 73 76 72 79

3 4

5 6 7 8

m±SD n.d.

3.2 ± 0.9

n.d.

n.d.

2662 ± 634

1.8 1.9

n.d. 385 400

1.9 1.7

122 642 535

NORMAL 3.8 2.3 2.8

n.d. n.d. n.d.

1098 750 915

14 13 17

n.d. 532 n.d.

14

1.7 ± 0.1

261 ± 226

1.8 ± 0.1

433 ± 274

2.9 ± 0.8

n.d.

921 ± 174

14 ± 2

532

14

2710 ± 1455

2.4 ± 1.6

641 ± 440

2489 1889 2200

1.6

24 26 21

1 2 3

2222 2513 2800 2673 1401 3200 2487 2348 3557 2900

795 950 n.d. 409 1471 540

62

m±SD

2.9 4.5 3.7

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

2.0 0.6 2.2 4.6 2.7 5.0 4.1 0.9 1.3 1.1

71

9 10

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

2251 1170 2600 1575 2062 3700 6262 1786 2800 2900

2192 ± 300

3.8 2.3 2.5 2.0 4.5

775

0-------0

4

~

/.

0 ~

~

u

s,F

specific binding

o----o non specific binding

~·

0.5

Cytoplasmic R 5020 BINDING BPH • 8

BINDING SITES • 2348 lmol / gram Ko, 9. 6, 10 -JO M

0.2

D.l

0.2

3H - R 5020 lnMI

0.3

0.4

0.5

B

FIG. 3. Saturation analysis of 3H R5020 binding to progesterone receptor.

j

B1

•--@

total binding

D------D

specific binding

0--------0

non specific binding

O.bB 0.07

14

0.06 ·

20

0.05

16

0.04

'.:::' 11

0.03

';-' a

± 2.3

n.d.

1.7

4.8

Cytoplasmic E2 BINDING Normal Prostate ,1, 2 Binding sites ~ 642 fmol / gram

\''"'" .,

X

~

0.01 0.01

2

4

6

3H - Estradiol I nM I

TABLE 5. Corre/,ation of endogenous tissue and receptor determinations and stereological data in normal and hyperplastic human prostatic tissue

Testosterone 5a-dihydrotestosterone Estradiol Estrone Estriol Progesterone Androgen Receptor (N) Androgen Receptor (C) Estrogen Receptor (N) Estrogen Receptor (C) Progesterone Receptor (C)

0. 1

1ll

3.3 3.2 2.5 5.5 9.6 4.7 4.5

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Age Weight Vw VEP VsT S L H Dw DAP LAP ScA

o., 0. 3

.I

32

6.5 6.9

= not detectable. • - - • total binding

-

n.d. 826 650

0.02

0.04

0.06 B

0.08

0.10

FIG. 4. Saturation analysis of 3H-estradiol binding to estrogen re-

ceptor.

the prostate. 1 ' 2, 4, 11 In our previous light-microscopic stereologic analysis we reported that human benign prostatic hyperplasia is primarily attributable to an increase in stromal tissue. 2 The ultrastructural analysis of the smooth muscle cells of pros tatic

0

0

0

0

0

0

O

0

O

p < 0.01. N: Nucleus. 0 p < 0.05. C: Cytoplasm. All other values do not show a statistically significant correlation. ~

stroma shows an activation of these cells in benign prostatic hyperplasia as compared to normal human prostates. 2 Activated smooth muscle cells were also found in beagle dogs after administration of a high dosage of estrogen; this finding indicates that estrogens play a role in regulating stromal growth. 12 The present extensive stereologic analysis on another five patients again yields a significant absolute increase in stromal tissue in BPH specimens; on the other hand, when comparing hyperplastic and normal tissue no difference in the absolute amounts (=g) of glandular cells can be seen, However, in hyperplastic tissue a significant increase in the glandular lumina can be observed; these results support our previous results showing that human BPH is primarily a stromal disease with secondary interactions on the glandular part, The etiologic factors underlying the development of benign prostatic hyperplasia in human beings are not yet clearly understood; however, human benign prostatic hyperplasia is supposed to be under testicular influence, since it does not develop in individuals who have been castrated at an early age. The prostate does not develop in patients with pseudohermaphroditism due to 5a-reductase deficiency or androgen insensitivity. Since the initial studies of Siiteri and Wilson in 1970, 5 it has become apparent that 5a-dihydrotestosterone levels are elevated in spontaneous development of human BPH. Quantitative differences between normal prostatic tissue and BPH tissue can be demonstrated for androgen-binding capacity, metabolism, and endogenous androgen concentrations. The influence exerted by estrogens on the prostatic metabolism has

564

BARTSCH AND ASSOCIATES

not been clearly defined; however, it has been demonstrated that estradiol administered in combination with 3a-androstanediol can induce hyperplasia in castrated canine prostates.13· 14 Regarding the endogenous concentrations of estrogens, progesterone and androgens, no significant differences between normal and hyperplastic tissues were found in our study. Concerning the 5a-dihydrotestosterone levels, our study confirmed the studies by Walsh and coworkers, who found that the levels of 5a-dihydrotestosterone were similar in normal and benign prostatic hyperplastic tissue obtained by open surgical procedures. 6 When the 5a-dihydrotestosterone contents in normal peripheral and benign hyperplastic prostatic tissues were determined (obtained at open surgical procedures on 20 men aged 36 to 82 years) they were similar in both groups (mean± SE, 5.1 ± 0.4 ng./gm. tissue). 6 When reviewing the literature to find out why we have been unable to confirm the findings of other investigations, it was shown that the elevated levels of DHT in BPH reported in the literature were based on measurements performed in surgically removed specimens, whereas the "low levels" of DHT in normal tissue were based on measurements performed in tissues obtained at autopsy. 6 In the study performed by Walsh normal prostatic tissue was obtained by cystectomy or radical prostatectomy; 6 in our studies normal prostatic tissue was obtained from kidney donors aged between 21 and 25 years. In both studies no difference between normal and hyperplastic prostatic tissues concerning the 5a-dihydrotestosterone concentration was found. Therefore, the finding that the 5a-dihydrotestosterone contents are identical in BPH and normal prostates questions the role of 5a-dihydrotestosterone as the causative factor in the etiology of human benign prostatic hyperplasia. When the synthetic radioligand Rl881 (6, 7- 3H) methyltrienolone and triamcinolone, which inhibits the binding of R1881 to the progesterone-binding component, and proteolytic inhibitors such as phenylmethylsulfonylfluoride were used, cytoplasmic and nuclear receptors were detectable in human BPH as well as in normal prostatic tissue. 15 ·16 Similar results were obtained in this study (BPH: cytosolic 2710 ± 1455 fmol./gm. and nuclear 641 ± 440 fmol./gm.; normal tissue: cytosolic 2182 ± 300 fmol./gm. and nuclear 261 ± 226 fmol./gm.). Concerning the correlation of stereologic and biochemical results this study shows that prostatic overgrowth in the aging male is poorly reflected in the endogenous hormonal levels of the whole prostate; however, it should be stressed, that in the present study receptors and steroid levels were measured in whole homogenates; since the ratio of stromal and glandular tissue varies in normal and hyperplastic prostatic tissue, future studies should focus oh the heterogeneity of the prostatic gland. In order to clarify the etiology of this disease other factors than steroids must be studied, among them factors of the extracellular matrix which possibly controls overgrowth and function of the glandular epithelium; on the other hand, the great amount of secretory products in large BPH samples should be investigated with regard to neurotransmitters; further factors will have to be identified which may lead to an accelerated growth in the human prostate.

REFERENCES

1. Bartsch, G., Frick, J., Ruegg, I., Bucher, M., Bolliger, 0., Oberholzer, M. and Rohr, H. P.: Electron microscopic stereological analysis of the normal human prostate and of benign prostate hyperplasia. J. Urol., 122: 481, 1979. 2. Bartsch, G., Muller, H. R., Oberholzer, M. and Rohr, H. P.: Light microscopic stereological analysis of the normal human prostate and of benign prostate hyperplasia. J. Urol., 122: 487, 1979. 3. Neubauer, B. and Mawhinney, M. G.: Actions of androgen and estrogen on guinea pig seminal vesicle epithelium and muscle. Endocrinology, 108: 680, 1981. 4. Rohr, H. P. and Bartsch, G.: Human benign prostatic hyperplasia: a stromal disease; new perspectives by quantitative morphology. Urology, 16: 625, 1980. 5. Siiteri, P. K. and Wilson, J. D.: Dihydrotestosterone in prostatic hypertrophy. I. The formation and content of dihydrotestosterone in the hypertrophic prostate of man. J. Clin. Invest., 49: 1737, 1970. 6. Walsh, P. C., Hutchins, G. M. and Ewing, L. L.: Tissue content of dihydrotestosterone in human prostatic hyperplasia is not supranormal. J. Clin. Invest., 72: 1772, 1983. 7. Cowan, R. A., Cowan, S. K., Grant, J. K. and Elder, H. Y.: Biochemical investigations of separated epithelium and stroma from benign prostatic hyperplastic tissue. J. Endocrinol., 74: 111, 1977. 8. Krieg, M., Klotzl, G., Kaufmann, J. and Voigt, K. D.: Stroma of human benign prostatic hyperplasia: preferential tissue for androgen metabolism and estrogen binding. Acta Endocrinol. (Copenh.), 96: 422, 1981. 9. Romijn, J. C., Oishi, K., Belt de Vries, J., Schweikert, H. U., Mulder, E. and Schroder, F. H.: Androgen metabolism and androgen receptors in separated epithelium and stroma of the human prostate. In: Steroid Receptors, Metabolism and Prostatic Cancer. Edited by Schroder, F. H. and de Voogt, H. J. Excerpta Medica, Amsterdam, pp. 134-139, 1980. 10. Bruchovsky, N., Rennie, P. S. and Wilkin, R. P.: New aspects of androgen action in prostatic cells: stromal localisation of 5areductase, nuclear abundance of androstanolone and binding of receptor to linker deoxyribonucleic acid. In: Steroid Receptors, Metabolism and Prostatic Cancer. Edited by Schroder, F. H. and de Voogt, H.J. Excerpta Medica, Ansterdam, pp. 57-76, 1980. 11. Rohr, H.P., Oberholzer, M., Bartsch, G. and Keller, M.: Morphometry in experimental pathology: methods, baseline data, and applications. Int. Rev. Exp. Pathol., 15: 233, 1976. 12. Rohr, H. P., Naef, H. F., Bolliger, 0., Oberholzer, M., Ibach, B., Weissbach, L. and Bartsch, G.: The effect of estrogen and stromal growth of the dog prostate. Urol. Res., 9: 201, 1981. 13. Walsh, P. C. and Wilson, J. D.: The induction of prostatic hypertrophy in the dog with androstanediol. J. Clin. Invest., 57: 1093, 1976. 14. DeKlerk, D. P., Coffey, D. S., Ewing, L. L., McDermott, I. R., Reiner, W. G., Rooinson, Ch., Scott, W. W., Standberg, J. D., Talakay, P., Walsh, P. C., Wheaton, L. G. and Zirkin, B. R.: A comparison of spontaneous and experimentally induced canine prostatic hyperplasia. J. Clin. Invest., 64: 842, 1979. 15. Walsh, P. C. and Hicks, L. L.: Characterisation and measurement of androgen receptors in human prostatic tissue. Prog. Clin. Biol. Res., 33: 51, 1979. 16. Trachtenberg, J. and Walsh, P. C.: Androgen receptor content of normal and hyperplastic human prostate. American Urological Association, Program and Abstracts, p. 98, 1981.