Immunohistochemical localization of androgen receptors with mono- and polyclonal antibodies to androgen receptor H.

Takeda, G. Chodak, S. Mutchnik, T. Nakamoto and C. Chang

Department of Surgery/Urology and Ben May Institute, University of Chicago, Box 403, Chicago, Illinois 60637 (Requests for offprints should be addressed to C. Chang) received

5841 S.

Maryland Avenue,

4 October 1989

ABSTRACT

Rat, human, and mouse tissues were stained immuno-

and polyclonal androgen histochemically using antibodies. Monoclonal antibodies were receptor raised in rats and used to stain human and mouse tissues; polyclonal antibodies were raised in rabbits and used to stain rat tissues. Frozen tissue sections were incubated with the appropriate androgen receptor antibody and staining was completed by the indirect avidin-biotin peroxidase method. A comprehensive survey of rat and mouse tissues was performed. Antibody staining was found exclusively in the nucleus of certain specific cell types, suggesting that the androgen receptor is a nuclear protein. All male sexual organs in the rat showed strong positive nuclear staining for androgen receptor. mono-

INTRODUCTION

have a great variety of effects on many target tissues. They induce development in malespecific organs, such as the prostate and seminal

Androgens

vesicle (Jost, 1965), and later in life they control the functional activity of target tissues (Mooradian, Morley & Korenman, 1987). In fact, specific receptors for androgens have been demonstrated in repro¬ ductive organs (Liao, 1975), central nervous tissue (McEwen, 1980), skeletal muscle (Michel & Baulieu, 1980) and other organs (Gustafsson & Pousette, 1975). So far the detection of androgen receptors has been carried out by biochemical binding assay (Chang

& Liao, 1987) and autoradiographic techniques (Stumpf & Sar, 1975). Although the biochemical bind¬ ing assay can detect minute amounts of the androgen receptor protein in tissue homogenate, it lacks the fine resolution necessary for showing the cellular or subcellular distribution of receptors in the tissue. This defect may be serious in clinical use because the tissue

Weaker positive reactions were seen in kidney, liver, adrenal cortex and pituitary gland. Furthermore, positive staining for androgen receptor was exhibited in skeletal, cardiac and smooth muscle cells, and central nervous tissue. Female reproductive organs also contained androgen receptor-positive cells. The spleen was found to be the only organ examined which did not stain for androgen receptor. The monoclonal antibody could also demonstrate androgen receptor-positive cells in a human prostatic cancer and in a prostate with benign hyperplasia. These data demonstrate the use of antibodies in revealing cellular/subcellular distribution of androgen receptor in target tissues. Journal of Endocrinology (1990) 126, 17\p=n-\25

specimens

obtained from

patients

sometimes show

great heterogeneity (Gorelic, Lamm, Ramzy et al. 1987). On the other hand, autoradiographic analysis can provide the precise cellular distribution of recep¬

tors, but at the expense of very long exposure times 1-12 months). Moreover, the use of radiolabelled hormones limits greatly the clinical use of this method. Immunocytochemical analysis has a greater advantage in that it allows visualization of androgen receptor at a cellular level after relatively easy and quick treatment of the tissue sections. Until recently, however, specific monoclonal antibodies against androgen receptors were not available due to difficulties in the purification of androgen receptors. Since cDNAs have been constructed which encode full-length rat and human androgen receptors (Chang, Kokontis & Liao, 1988a and b), it has been possible to make monospecific antibodies against androgen receptors by molecular fusion techniques. These techniques were employed to obtain large quan¬ tities of specific androgen receptor peptides (Chang,

(usually

Whelan, Popovich

et al. 1989 ). Here we report the of these antibodies to demonstrate the nuclear localization of androgen receptor in rat, mouse and human tissues and compare our results to those obtained by biochemical binding assay and use

stored at 70 °C until use. The normal tissues taken from three to five animals were examined in the —

present study.

autoradiographic analysis.

Two human prostatic tissues were also examined, from a patient with benign prostatic hyperplasia and the other from a patient with well-differentiated prostatic carcinoma.

MATERIALS AND METHODS

Immunohistochemical staining of androgen receptor

Antigens and antibodies The preparation of mono-

Polyclonal staining

and

polyclonal androgen

receptor antibodies used in this study has been described previously (Chang et al. 1989 ). Briefly, rats (for the monoclonal antibody) and rabbits (for the polyclonal antibody) were immunized with a fusion protein expressed by bacteria which contain a part of the human androgen receptor (N-terminal domain). Androgen receptor fusion proteins were purified as follows. Escherichia coli RR1 with TrpE-androgen receptor fusion pATH vector were grown in M9 mini¬ mal medium and induced by ß-indoleacrylic acid.

Cells were harvested, washed, lysed with lysozyme and sonicated at 370 W for 4-10 s. The protein pellet was isolated, resuspended in a small volume of Tris-Cl (10 mmol/1) and separated by SDS-PAGE. The 62 kDa fragment was cut from the gel and underwent electroelution for 24 h at 50 V in a dialysis bag con¬ taining SDS-PAGE chamber buffer (25 mmol Tris/1, 250 mmol glycine/1 and 1 % (w/v) SDS). The electroeluted protein was precipitated with 6 volumes of acetone, and the pellet was air-dried and resuspended in a small volume of TE buffer (10 mmol Tris-HCl/1, pH 80, 1 mmol EDTA/1). Mono- and polyclonal androgen receptor antibodies were purified through an anti-rat or anti-rabbit IgG agarose column and their receptor specificity was confirmed by gradient

centrifugation and immuno-precipitation assays (Chang et al. 1989 ). It was also confirmed by Western blot

analysis

one

that these antibodies reacted with the

androgen receptor fusion protein (62 kDa) which was used for immunization (data not shown). Animals and tissues

Cri: CD-I (ICR) mice were purchased from Charles River Inc. (Wilmington, MA, U.S.A.) and SpragueDawley rats were from Taconic (Germantown, NY, U.S.A.). The following tissues were examined: cerebellum, pituitary gland, heart, adrenal and kidney (each from male and female animals), ventral prostate, seminal vesicle, coagulating gland, testis, epididymis, ovary, uterus and vagina. The tissues taken from the animals were placed in O.C.T. compound (Miles Laboratories Inc., Elkhart, IN, U.S.A.) and immedi¬ ately frozen on solid C02. The frozen tissues were

The frozen tissues embedded in O.C.T. compound cut at 5 µ thickness in a cryostat and mounted onto poly-L-lysine-coated slides. Polyclonal sections were fixed for 1 min with fixative (2% (w/v) paraformaldehyde and 15% (v/v) saturated picric acid in phosphate buffer) at room temperature and rinsed with 50 mmol phosphate buffer/1 for 30 min. The sections were then stained following the peroxidasestreptavidin-biotin method (Hsu, Raine & Fanger, 1981). All procedures were carried out at room tem¬ perature except for the incubation with first antibody. To prevent the non-specific binding of avidin or biotinylated second antibody, the polyclonal sections were treated with an avidin D solution containing 3% (v/v) goat blocking serum for 15 min, followed by treatment for 15 min with a biotin solution (avidin/ biotin blocking kit; VECTOR Laboratories Inc., Burlingame, CA., U.S.A.). After a brief wash with the buffer, the sections were incubated with first anti¬ body, at a concentration of 30µg/ml overnight at 4 °C. The sections were then washed with the buffer and incubated with biotinylated second antibody (1:100 dilution, goat anti-rabbit IgG(H + L),F(ab') ; ZYMED Laboratories Inc., San Francisco, CA, U.S.A.) for 30 min, followed by a wash with the buffer and incubated with peroxidase-streptavidin for 30 min (1:100 dilution; ZYMED). Finally, the sections were incubated for 5 min with the substrate solution (001% H202, 005% 3,3-diaminobenzidine tetrahydrochloride; Sigma). The sites with enzyme activity (indicating antibody binding) became brown in the sections following this treatment. were

Monoclonal staining Frozen sections for monoclonal staining were fixed in a freshly made 4% paraformaldehyde solution for 1 min. After a 30-min wash in the phosphate buffer, the sections were incubated in 1% (w/v) bovine serum albumin solution in order to block the non-specific binding of the biotinylated second antibody. The first monoclonal antibody was then applied to the tissue sections at a concentration of 15 µg/ml for overnight incubation at 4 °C. After washing, the sections were incubated with the biotinylated second antibody (1:500 dilution, rabbit anti-rat IgG, F(abc); VECTOR Laboratories Inc.) for 45 min. This was followed by

incubation with

peroxidase-streptavidin

for 45 min

(1:200 dilution, ZYMED). For monoclonal sections, incubation was for 15 min with the substrate solution.

Both normal preimmune serum and absorbed immune serum were used as a negative control. To make the absorbed serum, the purified androgen receptor fusion protein was added to immune serum at a final ratio of antigen : antibody 1:1, and the supernatant of the mixture was used after centrifuga tion. For evaluation, the intensity of the positive nuclear staining was expressed in terms of three relative intensities: high (+ + +), moderate (+·+) and low ( + ) (see Table 1). The relative intensity of each tissue was determined by comparison with the following representative tissues. The strongest reaction of prostatic epithelium, adrenal cortex (zona fasciculata) and ovary (luteal cells) represent, respectively, high, moderate and low intensities. =

Since our monoclonal antibodies were raised in rat, the rat tissue sections stained with the monoclonal antibody showed relatively high background staining. The sections from rat tissues were, therefore, stained with the polyclonal antibody and those from mouse and human tissues were stained with both mono- and polyclonal antibodies. The mono- and polyclonal antibodies used produced essentially the same pattern of staining in both mouse and human tissues. The androgen receptor localization in the rat tissue with polyclonal antibody also showed the same pattern as that in the mouse tissue stained with the monoclonal

antibody (PI. 1, figs 1-3). Positive staining for androgen receptor appeared

in nuclei but not in cytoplasm from all the tissues examined except for the spleen which showed no staining of androgen receptor (Table 1). No immunostaining was observed in the nuclei when the sections were incubated with preimmune serum or absorbed serum (PI. 1, fig. 2). Pretreatment of the sections with H202, normally used to reduce endogenous

peroxidase activity, was

completely Male

while cuboidal cells showed less. A few stromal cells also showed positive staining. Seminal vesicle The pseudostratified secretory epithelium, which forms a highly convoluted gland, showed nuclear staining; a few stromal cells were also positive. The seminal vesicle is surrounded by a prominent smooth muscle layer which showed a strong positive reaction in its nuclei (PI. 1, fig. 4).

Coagulating gland

Glandular epithelial cells and stromal cells exhibited weak positive staining. The intensity was weaker than that seen in the prostate and seminal vesicle. Outer smooth muscle also showed nuclear staining.

Epididymis In epididymis,

RESULTS

since it

although the intensity of the staining varied depend¬ ing on the region (PI. 1, figs 1 and 3). Tall columnar cells tended to show strong positive nuclear staining

was

omitted from the procedure

found to abolish in our sections.

positive nuclear staining

reproductive organs Prostate Prostatic alveoli consist of columnar or cuboidal secretory epithelium lining a large lumen. The stroma shows a fibroblastic connective tissue which sur¬ rounds the epithelium. Androgen receptor is localized predominantly in the nuclei of epithelial cells,

epithelium shows a gradual tall, pseudostratified columnar form proximally to a shorter pseudostratified form distally leading to the ductus deferens. The intensity of epithelial staining was highest in the proximal part (PI. 1, fig. 5), decreased gradually in the middle part, the

transition from

a

and increased

again

tissue, positive cells

epididymis.

in the distal part. In stromal observed in all parts of the

were

Testis The testis consists of spermatogenic cells, Sertoli cells, peritubular cells and Leydig cells. Except for the spermatogenic cells, all other cells showed positive nuclear staining (PI. 1, fig. 6). However, the intensity was weaker than that of other male reproductive organs.

Female

reproductive organs

Vagina

The androgen receptor-positive staining was observed in the nuclei of stromal cells and stratified squamous cells in the basal and parabasal epithelial cell layers of the vagina. However, intermediate and keratinized epithelial cell layers were negative. The smooth muscle layers also showed positive staining (PI. 1, figs 7 and 8). The intensity was weaker than that in male

reproductive organs. Uterus

Myometrium and stromal cells of the uterus showed weak positive staining, while staining was negative or very weak in the endometrium and uterine glands.

1. Summary of the localization and relative intensity of positive staining of androgen receptor in the nuclei of rat tissues stained with the polyclonal androgen receptor antibody. The designations appear as high (+ + +), moderate ( + + ), low ( + ), or no ( ) relative positive staining. The use of two symbols ( / + ) symbolizes hetero¬ geneity of positive staining in that tissue or cell type. For example, nuclei of prostatic stromal cells show positive staining but the intensity varies among the cells from negative ( ) to moderately positive ( + + ) table







Tissue

Relative staining

Prostate

Epithelium

Stroma Smooth muscle

+/+ + + -/+ + -/ +

+

Tissue

Relative staining

Ovary

Germinal cell Granulosa cell Luteal cell Thecal cell

Seminal vesicle

Epithelium

++

Stroma Smooth muscle

+++

Coagulating gland Epithelium Stroma Smooth muscle

Epididymis Epithelium

Stroma Smooth muscle

Vagina Epithelium

Stroma Smooth muscle

-/+ +

-/+ + + /+ +

Stroma Smooth muscle

+ /+ + + /+ + ++

Muscle Skeletal muscle Cardiac muscle

-/ +

++ + /+ +

+ /+ + ++

-/ + + + +

Purkinje cell

+

Granule cell

Pituitary gland

Endocrine cells

Stroma Smooth muscle

+ /+ + + +

Testis Germinal cell Leydig cell Peritubularcell Sertoli cell

+ + /+ + + /+ +

Epithelium

luteum, major components of

parenchyma (follicles, stroma, interstitial negative for androgen receptor staining. However, granulosa luteal cells in corpus luteum are stained weakly (PI. 2, fig. 9). tissue)

Liver Male hepatocyte Female hepatocyte Adrenal gland Zona glomerulosa Zona fasciculate Zona reticulata Zona medula

Cervix

the ovarian

+ + /+ + + /+ +

Brain

Epithelium

corpus

Glomerulus Proximal tubule Distal tubule

+ /+ + + -/+ + -/ +

Uterus

Ovary Except for the

Kidney

-/+ + -/ +

Pituicyte

cell types, such

+/+ + +

as

sinusoidal cells and Kupffer cells, of female tissues were

negative. Hepatocytes stained less intensely.

were

were

Other organs Liver

Although liver tissue showed relative high back¬ ground staining, weak positive staining was detected in the hepatic parenchymal cells (hepatocytes). Other

Adrenal The adrenal medulla was totally negative for andro¬ gen receptor staining, while the adrenal cortex did contain positive cells. The adrenal cortex is divided into three histological regions: zona glomerulosa, zona fasciculata, and zona reticularis. The cells in both the zona fasciculata and reticularis showed posi¬ tive staining, whereas those in the zona glomerulosa

were

stained less

staining intensity ally weaker from

or

(PI. 2, fig. 10). The positive cells became gradu¬

not at all

of the

the zona fasciculata towards the reticularis. The male and female adrenal showed the same pattern of staining.

zona

is in agreement with results of previous work on the prostate (Tan, Joseph, Quarmby et al. 1988; Chang, Chodak, Sarac et al. 19896), and demonstrates that the androgen receptor, like the oestrogen receptor and the progesterone receptor (Press & Greene, 1988), is a nuclear receptor protein

Kidney

Positive staining was observed in the epithelial cells of convoluted tubules and in some of the parietal cells of Bowman's capsule (PI. 2, fig. 11). No difference was evident between male and female tissues.

Skeletal and cardiac muscle A positive reaction was seen in nearly every nucleus of skeletal muscle. In cardiac muscle, most nuclei of the ventricular and atrial myocardial cells were positive (PI. 2, fig. 12). Nuclear staining was slightly weaker in female than in male tissues.

Spleen

No positive reaction could be detected in any cell type in the spleen. Anterior pituitary gland The anterior pituitary

consists of epithelial endocrine cells. These cells showed positive staining although the intensity varied among the cells (PI. 2, fig. 13). The pars intermedia also contained weakly positive cells. The posterior pituitary gland contains axons from neurosecretory cells and pituicytes, the function and structure of which are similar to the neuroglial cells of the central nervous system. Some of the pituicytes showed positive staining. No difference was observed between the male and female glands.

gland

Brain tissues

Rat cerebellum

was

stained with the antibodies.

Androgen receptor-positive cells were observed in the Purkinje cell and granule cell layers (PI. 2, fig. 14). Human prostate Human prostates

stained with monoclonal androgen receptor antibody. The epithelial cells in benign prostatic hyperplasia and in adenocarcinoma showed a positive reaction for androgen receptor (PI. 2, figs 15 and 16). Most positive staining was observed in epithelial cells, with little or no staining in stromal cells. were

DISCUSSION

In the present study, the positive reaction appeared in the nuclei of cells from most of the tissues examined.

This indicates that the androgen receptor is localized in nuclei, but not in cytoplasm. This

mainly finding

reproductive organs The cellular distribution of androgen receptors has been demonstrated by earlier autoradiographic studies in prostate, epididymis and seminal vesicle (Schleicher, Stumpf, Thiedemann & Drews, 1988). The androgen dependency of the accessory sex organs, such as the prostate, epididymis and seminal vesicle, is well established; castration results in regression of the secretory epithelium. In the present study, the nuclei of secretory epithelium of the reproductive tissues all showed positive staining for Male

androgen receptor. Most epithelial cells of the prostate contained androgen receptor. The nuclei of the tall columnar cells tended to show much stronger positive staining than those of the cuboidal cells. It is well known that the functional activity of prostatic epithelial cells correlates with cell height (Brandes, 1966); hence it can be said that the prostatic cells showing higher

secretory activity possess

more

androgen receptors

than inactive cells. The prostatic stroma also con¬ tained a few androgen receptor-positive cells. How¬ ever, the positive reactions were detected mainly in the epithelial compartment in the prostate gland. This is in contrast to the fetal prostate, in which autoradio¬ graphic staining is mainly in the stroma (Takeda, Mizuno & Lasnitzki, 1985). The intense staining of androgen receptor in the epithelium is also in contrast with the abundance of 5a-reductase in the stroma, which converts testosterone into dihydrotestosterone (DHT) in the prostate (Wilson, Bruchovsky, Shnitka et al. 1980). It is possible that the epithelium utilizes DHT supplied by the stroma. In the seminal vesicle and the coagulating gland, the strongest reactions were observed in the smooth muscle layer which surrounds the whole gland. This suggests that not only the secretory activity of the epithelium, but also the activities of the muscle layer, such as contraction during ejaculation, may be under the control of androgens. In the epithelium of the epididymis, the staining intensity was higher in the proximal segment, lower in the middle and again higher in the distal part. This observation is consistent with the results of biochemi¬ cal (Pujol & Bayard, 1979) and autoradiographic (Schleicher, Drews, Stumpf & Sar, 1984) analyses. The differential distribution of androgen receptors in the epididymal epithelium may explain the earlier

findings that the middle segment is the least sensitive to the maintenance of epithelial function and to the induction of RNA/protein synthesis by androgens (Kanka & Kopency, 1977).

Germinal cells of the testis were negative for andro¬ cells, peritubular cells and Leydig cells were positive. It has been reported that in mature animals, androgens alone, without folliclestimulating hormone, can maintain spermatogenesis (Steinberger, 1971). It has been postulated that andro¬ gen actions on spermatogenesis are mediated through the Sertoli cells. Furthermore, it has been shown that peritubular cells produce a factor under androgen control which modulates Sertoli cell function (Skinner & Fritz, 1985). In fact, the androgen receptor has been demonstrated biochemically in the enriched fractions from Sertoli, peritubular and Leydig cells of the testis (Burek & Sanborn, 1988). Androgens may, there¬ fore, control Sertoli cell function both directly and indirectly through surrounding peritubular cells. Also, the presence of androgen receptors in Leydig cells suggests that androgens regulate the function of Leydig cells such as the secretion and synthesis of the hormones. In general, our data show that the intensity of androgen receptor staining in male reproductive tissue was higher in prostate, seminal vesicle and epididymis, which is in agreement with receptor binding assays in these tissues. gen receptor, while Sertoli

Female

reproductive organs

Although

the staining intensity was weaker in female than in male reproductive organs, we found androgen receptor-positive cells in both the stroma and the epithelium of the vagina, the stroma of the uterus and the corpus lutea of the ovary. Giannopoulos (1973) observed that immature rat uterus concentrated androgens following the administration of [3H]androgens. Takeda et al. (1985) have demonstrated by autoradiography that androgen receptor-positive cells are localized in the stroma but not in the epithelium of the vagina of fetal and newborn female rats. Our present data show the presence of androgen receptors in the adult female reproductive tract. These results suggest that the female reproductive tract maintains sensitivity to androgens during adulthood. At present, the role of androgen receptor in the ovary is poorly understood.

castration results in a reduction of enzyme activity, and androgen administration restores activity to a normal level. However, our immunostaining data and previous Northern analysis (Chang et al. 19896) shows that the expression of androgen receptor in liver is quite low. Previous autoradiographic analysis also failed to show nuclear concentrations of andro¬ gens in hepatocytes (Winborn, Sheridan & McGill,

1987).

Adrenal

The zona fasciculata and zona reticularis of the adrenal cortex contained a large amount of androgen receptor-positive cells. It is believed that adrenal androgens are secreted by zona reticularis, mineralocorticoids by zona glomerulosa and glucocorticoids by zona fasciculata. The androgen receptor in androgen-producing cells may be involved in an autocrine function.

Kidney The epithelial cells of convoluted tubules and parietal cells of Bowman's capsule showed weak or moderate positive staining. The effects of androgens on the kidney have been examined extensively (Catterall, Kontula, Watson et al. 1986). Castration results in both atrophy of the kidney and a decrease in the activity of some enzymes (such as ß-glucuronidase); administration of androgens causes hypertrophy and an increase in renal enzyme activity (Kochakian,

1977).

Muscle

Androgen receptors in androgen-sensitive muscle, such as levator ani, have been studied extensively (Krieg, Szalay & Voigt, 1974). Other skeletal muscle also has androgen receptor with similar binding characteristics to other target organs (Michel & Baulieu, 1980). Our data confirmed these results and we were able to demonstrate nuclear staining in the thigh skeletal muscle which does not show clear sexual dimorphism. In fact, various effects of androgens on skeletal muscle have been reported, including increases in protein synthesis (Rogoskin, 1979) and in glycogen synthesis (Bergamini, 1975). Some of these effects may involve the interaction with androgen receptors in the nuclei of muscle cells.

McGill, Anselmo, Buchanan & Sheridan (1980)

Liver

have demonstrated,

Steroid hormone receptors in liver have been well studied (Kyakumoto, Sato, Nemoto et al. 1984) and the sexual dimorphism of their functions has been established (Kochakian, 1959). For certain hepatic enzymes, such as fumarase in mice (Adams, 1952),

with our immunohistochemical data; most nuclei of cardiac muscle showed positive staining, even stronger than that in skeletal muscle. It is well known

by both biochemical and auto¬ radiographic methods, that atrial and ventricular myocardial cells contain androgen receptors in the female rhesus monkey. Their findings are consistent

induce morphological (Behrendt, and biochemical 1977) changes (Chainy & Knungo, 1978) in cardiac muscle. It will be interesting to study whether the presence of androgen receptors may explain the differences in the prevalence of heart disease between men and women. Androgen receptors in smooth muscle have not been studied extensively. We showed here that most of the smooth muscle layers of reproductive organs of both sexes contain androgen receptor-positive cells. Furthermore, Winborn et al. (1987) have revealed by autoradiography that the smooth muscle of the diges¬ tive tract is positive for androgen receptor. Therefore, it is conceivable that in general, muscle cells, including skeletal, cardiac and smooth muscle cells, possess androgen receptors in their nuclei. that

androgens can

Pituitary gland We found androgen receptor-positive cells in the pituitary gland of both sexes. Sar & Stumpf (1977 ) have demonstrated by autoradiography and immunocytochemical techniques that androgen receptors are present in gonadotrophs. This has been confirmed biochemically by Thieulant & Duval (1985) using gonadotroph-enriched cell populations. Therefore, most of the positive cells in our experiments may be gonadotrophs. Also, Wehrenberg, Baird, Ying & Ling (1985) found that androgens increase the pituitary growth hormone (GH) response to GH-releasing fac¬

tor, and suggested the presence of androgen receptor in somatotrophs. Brain tissue can also be applied to tissue. In the cerebellum, Purkinje with some granule cells around them,

Immunostaining techniques

central

nervous

cells, together

showed strong

positive

staining. This finding

autoradiographic

is consistent with

Stumpf (19776). Human

nuclear

data of Sar &

prostate

staining benign prostatic Nuclear

observed in the epithelium of hyperplasia and prostatic adeno-

was

carcinoma. Several

previous experiments measuring

prostate have been carried out to predict the response of such tissue to endocrine therapy (Benson, Gorman, O'Brien et al. 1987; Peters & Barrack, 1987). However, thus far no clear result has been obtained, probably due to the heterogeneity of tissue samples. Immunohistochemical analysis, therefore, can be a very useful method for prediction of the effects of endocrine therapy. An extensive immunohistochemical survey of human

androgen receptor

diseased prostate

content in diseased

using

our

monoclonal

androgen

receptor antibody is

now

in progress in

our

laboratory.

The data presented in this paper demonstrate the of antibodies to reveal the cellular distribution of androgen receptor in target tissues. It is expected that, together with biochemical assay, immunohistochemical techniques can provide very reliable information on the cellular/subcellular distribution of androgen receptor in a wide range of tissues. use

ACKNOWLEDGEMENTS

We thank L. Acakpo-Satchivi for his expert technical assistance and Dr A. Saltzman for his valuable discus¬ sion. This work was supported by Grant 89-28 from the American Cancer Society, the Chicago Cancer Research Foundation, the Whirlpool Foundation, the Milhein Foundation and the Cancer and Urology Research Endowment at the University of Chicago. REFERENCES

Adams, D. H. (1952). Hormonal factors influencing liver catalase

activity in mice: testicular and adrenal factors. Biochemical Journal 50,486-493. Behrendt, H. (1977). Effect of anabolic steroids on rat heart muscle cells. I. Intermediate filaments. Cell and Tissue Research 180, 303-315. Benson, R. C, Jr, Gorman, P. ., O'Brien, P. C, Holicky, E. L. & Veneziale, C. M. (1987). Relationship between androgen receptor binding activity in human prostate cancer and clinical response to endocrine therapy. Cancer 59, 1599-1606. Bergamini, E. (1975). Different mechanisms in testosterone action on glycogen metabolism in rat perineal and skeletal muscles. Endocrinology 96, 77-84. Brandes, D. (1966). The fine structure and histochemistry of prostatic glands in relation to sex hormones. International Review ofCytology 20,207-276.

Burek, S. W. & Sanborn, . M. (1988). Increase in testicular androgen receptor during sexual maturation in the rat. Biology of Reproduction 39, 39^9. Catterall, J. F„ Kontula, K. K., Watson, C. S., Sepannen, P. J„ Funkenstein, ., Melanitou, E., Hickkok, . J., Bardin, C. W. & Janne, O. A. (1986). Regulation of gene expression by androgens in murine kidney. Recent Progress in Hormone Research 42, 71-109. Chainy, G. B. N. & Knungo, M. S. (1978). Effects of estradiol and testosterone on the activity of pyruvate kinase of the cardiac and skeletal muscles of rats as a function of age and sex. Biochimica et Biophysica Acta 540,65-72. Chang, G, Chodak, G, Sarac, E., Takeda, H. & Liao, S. (19896). Prostate androgen receptor: immunohistological localization and mRNA characterization. Journal of Steroid Biochemistry 34,

391-395.

Chang, G, Kokontis, J. & Liao, S. (1988a). Molecular cloning of human and rat complementary DNA encoding androgen receptors. Science 240,324-326.

Chang, C, Kokontis, J. & Liao, S. (19886). Structural analysis of complementary DNA and amino acid sequences of human and rat androgen receptors. Proceedings of the National Academy of Sciences of the U.S.A. 85,7211-7215.

Chang, C. & Liao, S. (1987). Topographic recognition of cyclic hydrocarbons and related compounds by receptors for andro¬ gens, estrogens and glucocorticoids. Journal of Steroid Biochemistry 27', 123-131. Chang, C, Whelan, C. T., Popovich, T. C, Kokontis, J. & Liao, S. ( 1989a). Fusion proteins containing androgen receptor sequences and their use in the production of poly- and monoclonal antiandrogen receptor antibodies. Endocrinology 125, 1097-1099. Giannopoulos, G. (1973). Binding of testosterone to uterine components of the immature rat. Journal of Biological Chemistry 248,1004-1010. Gorelic, L. S., Lamm, D. L., Ramzy, I., Radwin, H. M. &Shain, S. A. (1987). Androgen receptors in biopsy specimens of prostate adenocarcinoma.

Heterogeneity of distribution and relation to prognostic significance of receptor measurements for survival of advanced cancer patients. Cancer 60,211-219.

Gustafsson, J.-A. & Pousette, A. (1975). Demonstration and partial characterization of cytosol receptor for testosterone. Biochemistry 14, 309+3101. Hsu, S., Raine, L. & Fänger, H. (1981). Useofavidin-biotin-

peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. Journal of Histochemistrv and Cytochemistry 29, 577-580.

Jost, A. ( 1965). Gonadal hormones in the sex differentiation of the mammalian fetus. In Organogénesis, pp. 611-628. Eds R. L. Haan & H. Ursprung. New York: Holt, Rinehart & Wilson. Kanka, J. & Kopency, V. (1977). An autoradiographic study of macromolecular synthesis in the epithelium of ductus epididymis in the mouse. I. DNA, RNA and protein. Biology ofReproduction

16,421-427.

Kochakian, C. D. (1959). Mechanisms of androgen actions. Laboratory Investigation 8, 538-556. Kochakian, C. D. (1977). Regulation of kidney growth by andro¬ gens. Advances in Steroid and Biochemistry and Pharmacology 6, 1-34.

Krieg, M., Szalay, R. & Voigt, K. D. (1974). Binding and metabolism of testosterone and 5ot-dihydrotestosterone in bulbocavernosus/levator ani (BCLA) of male rats: in vivo and in vitro studies. Journal of Steroid Biochemistry 5,453—459. Kyakumoto, S., Sato, N., Nemoto, T., Ohara-Nemoto, Y. & Ota, M. (1984). Binding of [3H]methyltrienolone to androgen receptor in rat liver. Biochimica et Biophysica Ada 800,214-219. Liao, S. (1975). Cellular receptors and mechanism of action of steroid hormone. International Review ofCytology 41,87-172. McEwen, B. S. (1980). Binding and metabolism of sex steroids by the hypothalamic-pituitary unit; physiological implications. Annual Review ofPhysiology 42,97-110. McGill, H. C, Anselmo, V. C, Buchanan, J. M. & Sheridan, P. J. (1980). The heart isa target organ for androgen. Science 207, 775-777.

Michel, G & Baulieu, E.-E. (1980). Androgen receptor in rat

skeletal muscle: characterization and physiological variation.

Endocrinology 107,2088-2098. Mooradian, A. D., Morley, J. E. & Korenman, S. G. (1987). Biological actions of androgens. Endocrine Reviews 8,1-28.

Peters, C. . & Barrack, . R. (1987). Androgen receptor localiza¬ tion in the human prostate: demonstration of heterogeneity using a new method of steroid receptor autoradiography. Journal of Steroid Biochemistry 27, 533-541. Press, M. F. &Greene, G. L. (1988). Immunohistochemical localization of estrogen and progesterone receptors. In Advances

in Immunohistochemistry, pp. 341-361. Ed. R. A. DeLellis. New York: Raven Press. Pujol, A. & Bayard, F. (1979). Androgen receptors in the rat epididymis and their hormonal control. Journal ofReproduction and Fertility 56,217-222. Rogoskin, V. (1979). Metabolic effects of anabolic steroids on skeletal muscle. Medicine and Science in Sports and Exercise 11, 160-163. Sar, M. & Stumpf, W. . (1977a). Distribution of androgen target cells in rat forebrain and pituitary after [3H]-dihydrotestosterone administration. Journal of Steroid Chemistry 8,1131-1135. Sar, M. & Stumpf, W. E. (19776). Androgen concentration in motor neurons of cranial nerves and spinal cord. Science 197,77-79. Schleicher, G, Drews, U„ Stumpf, W. E. & Sar, M. (1984). Differ¬ ential distribution of dihydrotestosterone and estradiol binding sites in the epididymis of the mouse. Histochemistry 81, 139-147. Schleicher, G., Stumpf, W. E., Thiedemann, K. U. & Drews, U. ( 1988). Intersex mice composed of androgen insensitive Tfm and wild-type cells analysed by 3H-dihydrotestosterone autoradiography. Anatomy and Embryology 178, 521-528. Skinner, M. K. & Fritz, I. B. (1985). Testicular peritubular cells secrete a protein under androgen control that modulates Sertoli cell function. Proceedings of the National Academy of Sciences of the U.S.A. 82,114-118. Steinberger, E. (1971). Hormonal control of mammalian spermatogenesis. Physiological Reviews 51, 1-22. Stumpf, W. E. & Sar, M. (1975). Autoradiographic techniques for localizing steroid hormones. Methods in Enzymology 36, 135-156. Takeda, H., Mizuno, T. & Lasnitzki, I. (1985). Autoradiographic studies of androgen-binding sites in the rat urogenital sinus and postnatal prostate. Journal ofEndocrinology 104, 87-92. Tan, J., Joseph, D. R., Quarmby, V. E„ Jubahn, D. B., Sar, M., French, F. S. & Wilson, E. M. (1988). The rat androgen receptor: primary structure, autoregulation of its messenger ribonucleic acid and immunocytochemical localization of the receptor

protein. Molecular Endocrinology!, 1276-1285. Thieulant, M. L. & Duval, J. (1985). Differential distribution of androgen and estrogen receptors in rat pituitary cell populations separated by centrifugal elutriation. Endocrinology 116, 1299-1303.

Wehrenberg, W. B., Baird, ., Ying, S.-Y. & Ling, N. (1985). The effects of testosterone and estrogen on the pituitary growth hormone response to growth hormone-releasing factor. Endocrinology 32, 369—375. Wilson, R. P., Bruchovsky, N., Shnitka, T. K., Rennie, P. S. & Comeau, T. L. (1980). Stromal 5a-reductase activity is elevated in benign prostatic hyperplasia. Acta Endocrinologica 94,284-288. Winborn, W. B., Sheridan, P. J. & McGill, H. C. (1987). Sex steroid receptors in the stomach, liver, pancreas, and gastrointestinal tract of the baboon. Gastroenterology 92,23-32.

DESCRIPTION OF PLATES Plate 1 figure

Plate 2 1. Immunohistochemical localization of androgen

receptor in rat prostate stained with polyclonal antibody against androgen receptor. Androgen receptor is present

in most of the epithelial cells as indicated by the diaminobenzidine reaction product (dark areas) within cell nuclei. A few stromal cells surrounding the epithelium also show

positive staining. L, Lumen; E, epithelium; S, stroma. Bar=10C^m; 180.

figure

the corpus luteum (granulosa luteal cells) show weak posi¬ tive staining, whereas the cells surrounding the corpus luteum (theca luteal cells) are negative. G, Granulosa luteal cells; T, theca luteal cells. Bar= 100 µ ; 180. figure

2. Rat prostate stained with control serum. The polyclonal antibody absorbed with the antigen (androgen receptor fusion protein) was used as the control serum. No nuclear staining is present. Bar 100 µ ; 180. figure

=

figure

3. Immunohistochemical localization of androgen

receptor in mouse prostate stained with the monoclonal antibody against androgen receptor. There is no significant difference between monoclonal staining and polyclonal staining (compare with Fig. 1). Bar= 100 µ ; 180. figure

4. Immunohistochemical localization of androgen

receptor in rat seminal vesicle stained with the polyclonal antibody against androgen receptor. The nuclei of the

epithelium show positive staining. Smooth muscle layers which surround the gland also show nuclear staining. SM, Smooth muscle. Bar= 100 µ ;

220.

=

contains androgen receptor-positive cells while the medulla (MD) is negative. 3 =100µ ; 180. figure

cells (arrow) and a few cells of glomeruli are positive for androgen receptor staining. CT, convoluted tubules; GM, glomerulus. Bar= 100 µ ; 220. figure

muscle cells (SM) and stromal cells are also positive. 3 =100µ ; 180.

8. Rat vagina stained with control serum. The vagina shows relatively high background staining. No nuclear reaction is observed. Bar= 100 µ ; 180. figure

12. Immunohistochemical localization of androgen

receptor in rat cardiac muscle stained with the polyclonal antibody against androgen receptor. The nuclei of myocar-

dial cells show positive staining. Bar= 100 µ ;

370.

13. Immunohistochemical localization of androgen

receptor in rat anterior pituitary gland stained with the polyclonal antibody against androgen receptor. Many cells

in the pituitary gland show positive staining. Bar= 100 µ ; x220. 14. Immunohistochemical localization of androgen

receptor in rat cerebellum stained with the polyclonal antibody against androgen receptor. Positive cells are

observed in the Purkinje cell and the granule cell layers. ML, Molecular layer; GL, granule cell layer. Bar 100 µ ; xl80. =

figure

7. Immunohistochemical localization of androgen receptor in rat vagina stained with the polyclonal antibody against androgen receptor. Although the epithelium of vagina shows relatively high background staining, nuclear staining is present in the basal and parabasal layers (com¬ pare with the control section in Fig. 8). Nuclei of smooth figure

11. Immunohistochemical localization of androgen

receptor in rat kidney stained with the polyclonal antibody against androgen receptor. The nuclei of convoluted tubule

figure

6. Immunohistochemical localization of androgen receptor in rat testis stained with the polyclonal antibody against androgen receptor. Germ cells are negative for androgen receptor staining. On the other hand, Sertoli cells (St), peritubular cells (P) and Leydig cells (Lg) show positive staining. Bar= 100 µ ; 220. figure

10. Immunohistochemical localization of androgen

receptor in mouse adrenal stained with the monoclonal anti¬ body against androgen receptor. The adrenal cortex (CX)

figure

5. Immunohistochemical localization of androgen receptor in mouse epididymis (proximal segment) stained with monoclonal antibody against androgen receptor. Most of the nuclei in the epithelium are positive. Bar 100 µ ; xl80. figure

9. Immunohistochemical localization of androgen

receptor in corpus luteum of rat ovary stained with the polyclonal antibody against androgen receptor. The cells in

15. Immunohistochemical localization of androgen

receptor in human prostate stained with the monoclonal antibody against androgen receptor. The tissue was

obtained from a patient with benign prostatic hyperplasia. The nuclei of most epithelial cells show positive staining; some stromal cells are also positive. Bar 100 µ ; 180. =

figure

16. Immunohistochemical localization of androgen

receptor in human prostate stained with the monoclonal antibody against androgen receptor. The tissue was

obtained from a patient with prostatic adenocarcinoma. The cells show positive nuclear staining for androgen receptor. Bar= 100 µ ; 180.

cancer

(Facing p. 26)

Immunohistochemical localization of androgen receptors with mono- and polyclonal antibodies to androgen receptor.

Rat, human, and mouse tissues were stained immunohistochemically using mono- and polyclonal androgen receptor antibodies. Monoclonal antibodies were r...
11MB Sizes 0 Downloads 0 Views