0022-15541911$3.30 Thejournal of Histochernistry and Cytochemistry Copyright © 1991 by The Histochemical Society,

Vol. 39, No. 7, pp. 927-936, 1991 Printedin USA.

Inc.

Original

Artide

Androgen Receptor Expression in Human An Immunohistochemical Study JACOBUS A. RUIZEVELD EPPO MULDER, NEUY Departments

ofPathology

Immunology

(NDZ),

Received

WINTER,’ JAN TRAPMAN, D. ZEGERS, and THEODORUS

Medical

for publication

Biological

November

MARCEL

DE

(JARdWJ7MV,TbHvdK)

and Biochemistry

Laboratory-TNO,

12,

Rijswijk,

and in revised

1990

form

The cellular localization of the human androgen receptor was visualized immunohistochemically using a mouse monodonal antibody (MAb) F39.4, directed against a fragment of the N-terminal domain of the androgen receptor. The nudearimmunoreactivity ofvaxious human tissues with F39.4 was generally consistent with earlier biochemical and autoradiographic data. However, previously suggested androgen receptor expression in thyroid, pancreatic, gastrointestinal, and bladder tissues was not confirmed immunohistochemically. Stratified squamous epitheia of vagina and cervix showed selective immunostaining of the basal cell layer, whereas in the preputial epitheium the intensity of immunoreactivity decreased gradually with maturation. In contrast, glandular epithelia of the sweat glands, male acces-

H (EM),

Tissues:

H.

University,

February

4, 1991;

accepted

but did not provide

The

physiological

arid

sues. Androgen be mediated

via the

androgen

responsive

estrogen

receptor

transcription

receptor

belongs

accessory

sex

play an and tis-

regulators,

which

receptors,

are expressed

at variable,

(25).

to the

receptors, the thyroid hormone mone receptors (7,12). Androgen which

ofmale

Similarly, androgens of several other organs

action in these organs and tissues is believed to either directly by the androgen receptor or, after

aromatization, The

function

on androgens. functioning

superfamily includes

and

receptors

the

retinoic

acid

several

steroid

hor-

are

relatively

of ligand-

nuclear

low levels

in a number

Studies on tissue formed thus far made receptors, either of autoradiography. titative

1

data

distribution use ofthe

in tissue homogenates or in tissue sections, Biochemical exchange assays generated

on the androgen

Correspondence

of the androgen receptor perligand binding properties of these

receptor

to: Dr. ).A.

ogy, Ee 1032, Erasmus University terdam, The Netherlands.

content

Ruizeveld Rotterdam,

of tissue

de Winter, P0

Box

homogenates

Dept. 1738,

by use quan-

3000

of PatholDR Rot-

of

February

18,

1991

(0A2132).

information

of this receptor. time-consuming

about

[3H]-steroid procedure

cellular

and

dis-

subcellular

autoradiography and does not

is a laborialways lead to

detailed cellular localization of steroid receptors in tissues to background labeling. Moreover, fluorescein-conjugated gens are Therefore,

of no use in the it was anticipated

histochemical cellular

being

could

contribute

and

subcellular

The isolation

more

sensitive

belonging

distribution

of the

and characterization

to the

tor superfamily

thyroid

exhibit

ofthe

androgen

(5,12,13).

After

receptor.

androgen

recep-

acid/steroid

functional

identification

than detailed

amino acid sequence Although all recep-

hormone/retinoic

a similar

specific ofthe

human

domain

N-terminal transactivation domain of the plays a low degree of homology with the receptors

and

to understanding

tor cDNA permitted the establishment ofthe ofthe androgen receptor protein (9,14,22,37). tors

owing andro-

detection of androgen receptors (4). that an easily performed immuno-

technique,

autoradiography,

proteins

of tissues.

and Department

sory sex organs, and female breast showed nearly exdusive F39.4 staining ofthe inner cylindric layer. In the testis, Sertoli cells, peritubular myoid cells, and interstitial cells were immunoreactive with MA], F39.4. Expression ofthe androgen receptor by smooth muscle tissue was largely confined to the male reproductive organs. The specificity and sensitivity of this simple and rapidly performed immunohistochemical technique in the detection of the human androgen receptor at the cellular and subcellular level makes it worthwhile to study tissue androgen receptor expression by immunohistochemistry in physiological and pathological states. (J Hiswcbcm Cytocbem 39:927-936, 1991) KEY WORDS: Human androgen receptor; Immunohistochemistry; Human tissues; Male accessory sex glands.

tribution ous and

development

KWAST

Rotterdam,

The Netherlands.

Introduction organs are dependent important role in the

DER

VAN

Erasmus

VERMEY,

recep-

structure,

of unique

immunogenic

sequences prepared

in the N-terminal domain ofthe androgen receptor, androgen receptor-specific polyclonal and monoclonal

tibodies

(39,46).

Monoclonal

antibody

(MAb)

F39.4

biochemical

assays

with

high

specificity

and

we an-

proved

to be

both and

in imin 5ev-

very effective in demonstrating the androgen receptor, munohistochemistry on formalin-fixed frozen sections eral

the

androgen receptor disother steroid hormone

affinity

(38,46). 927

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

928

DE WINTER,

RUIZEVELD

Since

we were

interested

in the immunohistochemical

charac-

teristics of MAb F39.4 with regard to detection of the androgen receptor in non-reproductive and reproductive tissues at the cellular and subcellular level, we undertook a body survey with this MAb.

TRAPMAN,

Antibodies. MAb F39.4 was prepared and characterized as described previously (38,46). In brief, Balb/c mice were immunized with a synthetic peptide conjugate corresponding to amino acids 301-320 ofthe N-terminal domain of the human androgen receptor conjugated to keyhole limpet hemocyanin. Murine spleen cells were fused with Sp2/0 myeloma cells. Hybridomas were screened for specific antibody production in an EUSA. Supematants selected by ELISA were subsequently screened in the rabbit antimouse immunoglobulin (RaM) agarose screening assay for their ability to adsorb [3H]-R1881-androgen receptor complexes from nuclear extracts of the androgen receptor-positive human prostate carcinoma cell line LNCaP (16). Hybridoma F39.4 was selected on the basis ofits strong reactivity with radioactively labeled ligand-bound nuclear extract from the LNCaP cells in a sucrose gradient density centrifugation assay and its good performance in imrnunohistochemistry is reactive

with

(46).

a protein

with

Western

blot

an apparent

analysis

size

showed

of 1 10

KD

that

F39.4

in a lysate

from

the LNCaP cell line (38). Specificity ofF39.4 for the human androgen receptor was established both in immunoprecipitation assays (46) and in a goat anti-mouse immunoglobulin (GriM) agarose test using nuclear extracts from LNCaP (androgen receptor), NHIK (glucocorticoid receptor), MCF7 (estrogen receptor), and T47D (progesterone receptor), and radiolabeled R1881, dexarnethasone, estradiol, R5020, respectively, as ligand (Figure 1). Slight crossreactivity

body

with

estrogen

receptors

was

seen

only

at a very

high

anti-

concentration.

Tissue Sources. Fresh human dergoing surgery. One cryptorchid

tissues were collected testis was included

from patients unin the study. Non-

ZEGERS,

MULDER,

DER

VAN

KWAST

Immunohistochemical Techniques. Frozen tissues were mounted on cryostat chucks in Tissue-TekllOClcompound obtained from Miles Laboratories (Naperville, IL). Air-dried 6-jim thick cryostat sections of fresh frozen tissues were fixed in 4% formalin (10 mm, 4C) and dehydrated in chilled methanol (4 mm, - 20C) and acetone (2 mm, - 20C). Preliminary

and Methods

Materials

VERMEY,

experiments

indicated

that

the

epitope

of androgen

receptor

defined

by F39.4 is sensitive to formalin fixation, which prohibits its use on routimely fixed paraffin-embedded material. Formalin fixation lasting for more than 10 mm or omission of methanol in the fixation procedure led to deterioration ofstaining. To minimize nonspecific binding ofreagents in subsequent steps, sections were first treated with normal rabbit serum diluted 1:10 in PBS (pH

7.4)for

15 mm

at room

temperature.

Incubation

with

MAb

F39.4 (hybridoma culture supernatant), diluted 1:100 in PBS containing 0.5% BSA and 0.1% sodium azide (pH 7.8), was performed overnight at 4’C. After washing in PBS for 10 mm, sections were incubated with horseradish peroxidase-conjugated RaM IgG (Dakopatts; Glostrup, Denmark) diluted 1:100 in PBS, for 30 mm at 37C. Alternatively, the indirect unconjugated peroxidase-antiperoxidase complex method (PAP) was used employing unconjugated RaM (Dakopatts) as linking reagent and PAP complexes (Sigma; Munchen, FRG) as tertiary reagent (20). Both reagents were diluted 1:100 in PBS. Incubations with RaM and PAP complexes were performed at 37C for 30 mm. Because the PAP method led to significantly better staining results, the indirect conjugated method was employed dnly for staining of male accessory sex glands, as they appeared to have a high androgen receptor content. After rinsing of the sections in PBS, they were incubated with 3,3’diaminobenzidine (DAB; Sigma) containing 0.03% hydrogen peroxide for 7 mm at room temperature. Nuclear staining intensity ofthe prostate secretory

epithelial

cells

in glandular

used for comparison sues.

Sections

used

as negative

hyperplasia

with staining

of lymph

nodes

control

for

intensity

incubated

was considered

high

found

organs

with

the

in other primary

and

antibody

was

or tiswere

F39.4.

autopsies were performed within 6 hr after decease of the patient. The tissues investigated are listed in Lble 1. Tissue specimens were transported

To confirm androgen receptor expression by smooth muscle cells in male reproductive organs, a double immunoenzymatic staining on androgen receptor and desmin was performed sequentially using horseradish peroxidase and alkaline phosphatase as enzymatic labels (27). After overnight incuba-

to the laboratory within 15 mm nitrogen until further use.

daze method,

diseased

testis

and

epididymis

were

obtained

after

from

surgery,

and

autopsy

were

cases.

stored

These

in liquid

tion

with

F39.4

desmin

Figure 1: Cross reactivity of F39.4 in GaM agarose assay.

and

antibody

1 . Human

Table

immunostaining

with

slides were subsequently (Sanbio;

tissues

Uden,

The

the

indirect

incubated Netherlands)

examineda,

conjugated

peroxi-

with monoclonal diluted

1:20

antiin PBS

b

Tissue

M

F

Tissue

Myocardium

2

2

Skeletal

Lung

2

0

Prostate

Esophagus

1

1

Seminal

Stomach

1

1

Vas

Colon Salivary Liver Pancreas

1

2

1

M

F

50

muscle

3

2

2

-

2

-

2

-

Epididymis

3

-

3

Testis

3

-

2 0

2

Prepuce

2

-

2

Skin

2

3

39 40

30

20

10

0,3

AR/R1881

ER/Eslradol (3H1s1eod

PR/R5020 ecep1or

0

GR/Deamelhason

compe*es

gland

1. Specificity

of F39.4forthe androgen receptor tested in a GeM agarose assay using nuclear extracts from LNCaP, MCF7, T47D, and NHIK as a source for androgen receptor, estrogen receptor, progesterone receptor, and glucocor-

ticoid receptor, respectively.

deferens

Thyroid

gland

2

0

Vagina

-

1

Adrenal Kidney Ureter Bladder

gland

2

0

Cervix

-

4

Thymus Figure

vesicle

Lymph a b

node

1

2

Uterus

0 2

2 0

Mammary

0

3

Sebaceous

1

1

M. male; F. female. Figures represent the number

Sweat

gland

gland

of cases investigated.

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

gland

-

4

0

3

1

3

2

2

IMMUNOHISTOCHEMISThY

for

30 mm

direct

at 37’C.

conjugated

strate

OF

Visualization alkaline

yielding

THE

of desmin

phosphatase

an alcohol-resistant

RECEFIDR

ANDROGEN

was performed

method

with

(Dakopatts),

red precipitate

(Vector;

929

the

using

in-

Digestive,

Burlingame,

CA).

No

staining

ofthe chi,

Reproductive

Tissues

F39.4

yielded

tive organs, nuclear

staining

the basal androgen

epithelial receptor

basal

Stromal

tive

reproduc-

kidney,

in the columnar

vesicle,

intensity

epithelial

and the epididymis,

cells

peritubular

are

intense

nuclear

seminal

vesicle,

staining

and

reaction.

epididymis

drogen

receptor

cells

male

accessory

2A). cryptorchid

cells and

expression

form in cryptorchid matozoa did not

sex organs

testis,

(34).

F39.4

F39.4

testicular

were

gave

also

reac-

a moderate

stained

fibroblasts.

was variable

nuclei

Leydig

in normal

testis. Spermatogonia, stain with MAb F39.4

In the stratified squamous highest reactivity with F39.4

cell an-

testis

but

spermatids, (Figure 2D).

and

epithelium of the penile was in the basal cell layers,

of preputium, cervix, and moderate reactivity (Figures

vagina 3C and

unisper-

prepuce, gradually

also displayed van3D). No immuno-

staining was observed in the reserve cells and columnar the cervical glands. No significant staining with F39.4 uteri

liver

to the nuclei

the

constituents

pancreas,

bron-

comparatively

of the hepatocytes

weak

(Table

2).

biopsy

specimens

androgen

from

receptor

two

male

expression

patients,

low-

was found,

which

receptor was not detectable in skeletal muscle cell nuclei SB). Smooth muscle of bronchi, intestines, and bladder detectable

immunoreactivity.

Other

Tissues

Lymph nodes, thymus, thyroid gland, nal gland revealed no immunoreactivity larly, cells,

no staining was found

Table

2.

of peripheral in the tissues

Androgen

parathyroid gland, and adrewith F39.4 (Table 2). Simi-

neural tissue, examined.

receptor

including

ganglion

immunoreactivitya

of

diminishing in the more superficial layers (Figure 3A). In the vagina and cervix, F39.4 yielded detectable reactivity only in the basal cell layer ofthe stratified squamous epithelium (Table 2). Stromal fibroblasts able and

In the

tissue

colon,

dis-

Immunoenzymatic double stainproportion of (desmin-positive)

of these

myoid

Tracts of the

stomach,

or bladder.

nuclear

lacked

an

prostate,

Sertoli

in any

contrasts with the negative staining results obtained on myocardial biopsy specimens from the two female patients (Figure 5A). The

whereas

staining reaction with the nuclei of a circle of cells located close to the spermatogonia. This localization ofthe immunoreactive nuclei within the transversely cut seminiferous tubules indicates that the stained

Urinary

found

esophagus,

was confined

In myocardial

the exclusive

showed

cells

was

glands,

alveoli,

androgen (Figure

with F39.4 (Figure In both normal and

in the

salivary

cells of these organs did not show detectable expression (Figure 2). In the vas deferens, occa-

of the

muscle

in all male

2). In general,

was observed

played variable immunoreactivity. ing showed that a substantial smooth

reaction

(ikble

the seminal

cells

cells

staining

the testis

reaction

prostate,

and

F39.4

Muscle

an intense

excluding

cells ofthe

sional

with

reactivity

Results

MAb

Respiratory,

a sub-

cells lining was observed

examined.

Testis Sertoli cells Germ cells Myoid

cells

Leydig

cells

Thymus Lymph

+

+

nodes

Skeletal

muscle

Cardiac

muscle

+

Fibroblasts Skin Prostate,

seminal

vesicle,

Keratinocytes

vas deferens,’’ epididyrnis Secretory cells Basal cells

Fibroblasts

Fibroblasts Smooth

Sebaceous glands Sweat glands

++

++ +

+1-

+

muscle

cells

+

+1-

Respiratory

tract

Foreskin Keratinocytes’

+

+

/

Urinary

+

tract

Fibroblasts

Skin,

Its Appendages,

Although

in non-genital

did not reveal

detectable

and Mammary skin

the epidermis

nuclear

staining

Gastrointestinal

Gland and

with

the hair

MAb

F39.4

follicles (Figure

3B), sebaceous glands, sweat glands, and ducts were immunopositive (Figures 4A and 4B; Table 2). In sebaceous glands most nuclei were positive, whereas in sweat glands and ducts only a proportion ofcells

were

detectable mal

stained. F39.4

Notably, reactivity.

or in subcutaneous

the myoepithelial Expression

mesenchymal

was not

cells did not show detectable

in den-

cells.

Immunostaining ofthe mammary skin and associated glands did not differ widely from skin staining patterns

sebaceous previously

mentioned

specimens

studied,

(Table the nuclei

Some ofthe a moderate (Figure

4C).

nuclei staining

2). In one ofmammary

of three acinar

mammary

gland

cells were labeled

with

F39.4.

ofthe inner ductal epithelial lining cells showed reaction. Myoepithelial cells were not stained

Ectocervix, vagina Squamous cells”

tract

Salivary

Fibroblasts

glands

Hepatocytes

Endocervix

+

Bile ducts Exocnine pancreas

-

Uterus

Mammary

gland

Ducts

+

+

/

-

Acini

+

+

/

-

Myoepitheliurn a

able ( b

-

+ I

,

A significant

staining. Androgen cell layers. d Only the

Pancreatic islets Adrenal gland

-

Signs designate + + I

Thyroid

high ( -

+ +

),

moderate

) immunoreactivity.

proportion

of basal

(

+

),

no (

cells in the

-

)

immunoreactivity

vas deferens

showed

or vanintense

nuclear

C

receptor basal

expression

cell layer

both

is reactive

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

in the with

basal and in the more F39.4.

superficial

RUIZEVELD

930

DE

WINTER,

TRAPMAN,

VERMEY,

MULDER,

ZEGERS,

VAN

DER

KWAST

#{149}1

.

‘

S

4. -.

,

c

0

.,

,

,,, .

:,

. *

....

:

.f fr,.

.

0

,

,i

1%

*.

.,

:

a’.**

I, 4

.

.#{149}u;4

.

.

.

#{149},

.

.

4_

I

‘51

‘

:

,

, :#{149} #{149} .

#{149}‘4’1’

3

.

::

1

,

#{149}

,,

.‘,

:

,

I

p

.

:

..4

f ,.

I

%

I

I..

‘p

“7 I

1

I,,

...‘

I

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

.

.

.,&

*

*‘

4

MUNOHIS1CHEMISThY

OF ThE

ANDROGEN

931

RECEP’KR

_

a ‘I a’,.

9.

0

A

B

-

4

.,

C

1 *,b,.

1..

S 4.

p

a’

0’

4

*#{248}#{149}..:

\

I

0

1

V. .

#{149}

a’

.

p 4:.

‘P -,

‘5

a

‘4 ‘;

I

a

ID Figure a lmmunostaining ofstratified squamous ficial epithelial cells of the foreskin and variable ofthe basal cell layer and nonspecific membrane fibroblasts. No nuclear counterstaining. Original

epithelia with F39.4 visualized

withthe

indirect PAPtechnique.

(A) Immunoreactivity

ofthe basal and more super-

staining of preputial fibroblasts. (B) Epidermis without detectable reactivity. (C) Ectocervix with faint reactivity staining of superficial cells. (D) Vagina with faint reactivity of the basal cells and rather intense labeling of stromal magnification x 310. Bar = 20 jim.

. Figure 2. Immunostaining of male sex organs for androgen receptor expression with F39.4 visualized with (A) the indirect PAP technique or with (B-D) the indirect conjugated peroxidase technique. (A) Prostatic hyperplastic tissue stained with immunoperoxidase for F39.4 reactivity and with immunoalkaline phosphatase for expression ofdesmin. Both the inner layer ofsecretory epithelial cells and desmin-positive cells are reactive with F39.4. (B) Vesicula seminalis with nuclear staining of the inner layer. (C) Epididymis with selective nuclear staining of the cylindric cells. (D) Testis in which Sertoli cells and peritubular myoid cells are stained. Nuclear counterstaining with Mayer’s hematoxylin. Original magnification x 310. Bar 20 jim.

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

932

DE WINTER,

RUIZEVELD

r

P;

,

#{149}

I

.

-ft

TRAPMAN,

.

0

‘.

I

,

S

.

$

#{149}4\

..:.

-

.a’

A” i*I

‘

x:

-

..

4

VERMEY,

#{149}

.

.1

t

aw

4

t

. I

e

.

,

.

‘

t

‘

y

S

MULDER,

.1,t4A4

#{149} 4,

‘

ti,--

h

. 1b

j

“-

.‘;

I

-

.

:

,

-

.

‘,

‘

,

/:41

:4

,

S

.

A r-

4#{149}

.

,r

I

4

#{149}:.-

.

,

.

‘

r’#{149}’5:i

‘a

,

,

.

...

,.....

‘

,

.

a’

10

.

:

V.

y._

.

5

a

r

‘

-4-

..

&L:.::

F.’

4 ‘

-

.

.

is. a

,jir5

.

.

..

I

.

r’1’

J#{149}l

,.-#{149},,‘a.

d’

I

I

‘?‘

-

\:L

1f/

.a’

-

4IpIa

..

::19_ -.

a.

4

0

‘

lv#

.,

?4

#{149}

#{149}11

‘:.

:,I!

;

-

elf.

t

‘-.

1

a

:

Iv.

. .-.

;,:k; ,w,

#{149}F,

4#{149}

‘

,

#{149}

-.

1

‘4..

ao’;.a

‘)

#{149}‘!“‘

4

4a

4

B

.i:’:&/i”

i:

p.

*

.-

#{246}a

..

-a

t;

a-’ I’*

.

I

$‘

4,,1a

‘

a

V

I

‘p

4

#{149},‘

w.

‘

t

.

“

%

1.

#{149}

. .

.

.,w

-

‘i

ti’#{149}

. ,

4I

_.J

4

a’

-p

,.

4.

I

.

*,p

a

-I

-j

l4it4:I/

.

VAN DER

KWAST

Figure 4. Immunostaining of skin appendages and mammarygland with F39.4visuaiized with the indirect PAP technique. (A) Intense staining of sebaceousgiandand no reactivitywith hair fe. (B) Variable but intensestainingof sweat glands. (C) Mammary acini with staining of the innerc Nuclearcounterstainlngwlth Mayet’s hematoxylin. Original magnification x 310. Bar -20jirn

I.

ZEGERS,

.,1

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

IMMUNOHISTOCHEMISThY

OF

THE

ANDROGEN

RECEPTOR

933

Figure 5. lmmunostaining of (A) cardiac muscle and (B) skeletal muscle with F39.4, visualized with the indirect PAP technique. Reactivity is found only in cardiac muscle. No nuclear counterstaining. Original magnification x 310. Bar - 20 jim.

4

A

a

S -

Discussion The

immunohistochemical

ton localization previously with

data

in human biochemical

ography

(1-3,17,29,31,33,45).

on tissue

homogenates

They in that

several tissues and organs In all tissues examined clear localization ofthe The nuclear localization

here

reported

on androgen

tissues largely confirm those ligand binding assays and extend

the observations

the exact cellular

localization

recepobtained autoradimade within

has now been visualized (Table 2). immunohistochemically the selective nu-

androgen receptor has now been established. of androgen receptor was previously de-

scribed labeled

in tissues androgens

polyclonal receptor,

ofvanious species by autoradiography (3,17,24,26,29,35,36,40-44). Using

antibodies other

authors

and

a monoclonal

similarly

antibody

found

ligand The

to the androgen

an exclusively

calization in the human prostate secretory epithelial This subcellular localization ofthe androgen receptor the reported roid receptors,

with radiowell-defined nuclear

predominantly nuclear localization of other in both the presence and the absence ofthe

sex stespecific

(30). absence

ofimmunostaining

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

ofthe

thyroid

lo-

cells (8,18,23). is in line with

gland,

pancreas,

934

RUIZEVELD

salivary

gland,

urinary

tract

and

with

interstitial

tissues

our MAb

in these

organs

or with

autoradiography

differences to detect

may in part explain the androgen receptor

pancreatic

with

F39.4

and

drogen

gastrointestinal

to positive

binding

assays

on

tissue

(10,11,26,36,40,43).

renal

receptor

radioactively

ligand

of the

is in contrast

tissue

content labeled

be

of these

ligands

attributed

tissues.

the

species inability in human

very

low

binding may

to proteins

(36).

Similarly,

labeling precluded

unrelated

of hepatic tissue the unequivocal

in this organ

Since

of

have

led

only

on

receptor

presence

receptor

may

nonspecific

interfere

receptor

with expression

of androgen

and

F39.4

made

in adrenal receptor

dig cells is consistent cells and interstitial

with Leydig

with

occur

it possible

gland

in Sertoli data cells

these castrated

and

studies

were Im-

to study

cells

and

interstitial

from cultures of purified (1). The immunohistochemically

cate

that

synthesis. cells and

androgens

their

effect

phism

we have not found conclusive in the expression of the androgen

sues. Although in the four specimens it was noted that only the two specimens strated

detectable

difference hypertrophy

androgen

on spermatogenesis

could also be attributed in these two specimens.

biochemical key cardiac

ligand muscle

binding data demonstrated

both male and female labeling experiments smooth muscle gans. Expression and smooth dependence The layer

in-

this

expression

to the presence of myocardial Earlier autoradiographic and on both baboon and Rhesus monandrogen receptor expression in

of androgen

squamous

staining

receptor

epithelia

intensity

for androgen

superficial epidermis of the penile drogen receptor in the differentiation

foreskin ofthe

in the

of cervix suggest epithelial

organs. As no immunostaining was observed in the ofthe glandular cervical epithelium, whereas the latter ment

to differentiate

of androgen

sia of the

transition

receptor zone

into

squamous

can be postulated of the

cervix.

epithelium, in squamous

the

of the androgen

androgen

insensitivity

squamous

ofpreputial

receptor

owing

of cultured receptor

syndrome.

However,

epithelial

cells

fibroblasts

with

to in vitro

foreskin properties

the high

condi-

fibroblasts in cases of

androgen

recep-

ton levels in keratinocytes suggests the use of these cells rather than fibroblasts in such studies. The demonstration of androgen receptor in sebaceous glands of male

and

female

binding

skin

studies

confirms

(33).

previous

Studies

data

dermal

papilla

studies

are required

of the

pilosebaceous

follicle

striking

was the

proportions

(15).

thermore, nign (19)

beard

F39.4

It is obvious

that

expression

of breast

Similarly,

carcinomas

further

in cells of the

in androgen-dependent

of androgen

In conclusion, antibody directed able

lining

receptor

epithelial

do indeed

a role of androgen

the genesis of hepatocellular tionally inflicted by this

re-

in the

acne.

detection inner

on andro-

of scalp, with

androgen receptor may be involved and malignant neoplasms of the

proportion (21).

of the

role

its

biochemical

recently unit

receptor

unit to evaluate as alopecia and

such

processes

hair

on androgen

from

performed

cells

in van-

androgen

has been

carcinomas, latter malignancy

and sugFur-

in the genesis of bebreast. A substantial

express

receptor

of mammary

receptor

suggested

as males (28).

for

are dispropor-

the availability of a highly specific monoclonal against the human androgen receptor will en-

the study of androgen physiological and of a simple

receptor pathological and

expression at the cellular level processes of human organs

rapid

immunohistochemical

tech-

nique.

Acknowledgments iVe wish to thank Ms PC. Delfos

Mr C.CJ. van Vroonhoven for photography.

for technical

assistance

and

double of

basal

and vagina

receptor

content

by the

Selection

tions may explain the suitability in the in vitro study of androgen

by application

sex

muscle cells is consistent with the known androgen of fibromuscular hyperplasia of the prostate.

of the stratified

the potential

seeming

(24,35). Immunoenzymatic that high-level immunostaining

expression (6,31).

in both

was largely confined to the male reproductive orof the androgen receptor in prostate fibroblasts

selective

the declining

animals revealed

expression,

of

ducts and acini of females and in sweat glands of both males females. The presence of androgen receptor in these structures gests a functional role of androgens that is hitherto unknown.

of myocardial tissue tested from male patients demon-

receptor

expression

LeySertoli

evidence for sexual dimorreceptor in various tis-

far,

receptor foreskin

penile

Most

directly. Thus

receptor

KWAST

able

The observed androgen receptor exlack of its expression in germ cells mdi-

mediate

VAN DER

The

detectable expression of 5a-reductase in adult rat germ cells has led some authors (32) to suggest a direct influence of testosterone in promoting and controlling the activity of initiation factors for germ cell protein pression in Sertoli

androgen

ofthe

andro-

as well.

variable

ZEGERS,

gen receptor content of the pilosebaceous gion, and pubic skin revealed immunostaining

animals.

testis

MULDER,

of androgen

ligand

background

autoradiographic

expression,

adrenalectomized

munohistochemistry gen

high

by radiolabeled dihydrotestosterone identification ofthe androgen receptor

androgens

of androgen

performed

that

(43).

endogenous

examination

to the androgen

it was reported

highly

a high

an-

to false-positive labeling of tissue sections. Labeling of both thyroid follicle lining epithelial cells and colloid in baboon thyroid glands further suggests that nonspecific binding of tnitiated androgens

The

VERMEY,

subepithelial foreskin fibroblasts contrasts not only with the reported androgen receptor content of these cells but also with the high level

the

Alternatively, proteins

TRAPMAN,

and

homogenates

to

to non-receptor

WINTER,

findings

Although

these discrepancies, immunohistochemically

may

DE

Literature

Cited

1. Anthony

cell and

in the more a role of ancells of these reserve cells cells do have an involvemetapla-

CT, Kovacs WJ, Skinner MK: Analysis ofthe ton in isolated testicular cell types with a microassay finity ligand. Endocrinology 125:2628, 1989

2. Bannister tion ofthe 1985

P, Sheridan human

P, Losowsky

hepatic

androgen

MS: Identification receptor.

andnogen

recep-

that

an af-

uses

and characteniza-

Clin

Endocninol

23:495,

WC, Fried FA, Stafford SJ, Mickey DD: Autoradiographic localization oftnitium labeled dihydrotestosterone in human vas deferens. 3 Androl 7:197, 1986

3. Beckman

4. Berns EMJJ, Mulder E, Rommerts FFG, van den Molen stein RA, Bolt-dc Vnies ), de Goeij TFPM: Fluorescent

nivatives do not discriminate between androgen -negative human tumor cell lines. Prostate 5:425,

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

H),

Blanken-

androgen receptor-positive 1984

deand

IMMUNOHISTOCHEMISTRY

OF THE

ANDROGEN

RECEPTOR

935

5. Bninkmann AO, Kuiper GGJM, van den KorputJAGM, Bolt), de Boer W, Smit A, Faber PW, van Rooij HCJ, Geurts van Kessel A, Voorhorst MM, Mulden E, Trapman J: Structure and function of the andnogen receptor. Urol Res 17:87, 1989 6. Brown TR, Migeon C): Androgen binding in nuclear genital skin fibroblasts from patients with androgen drome. J Clin Endocninol Metab 62:542, 1986 7. Carson-Junica ily:

MA, Schrader

structure

8. Chang

and

functions.

C, Chodak

G,

WT, O’Malley Endocr

Sarac

receptor: immunohistochemical tion. J Steroid Biochem 34:311,

of human insensitivity syn-

Lao

localization

25.

and mRNA

characteniza-

Liao 5: Structural analysis ofcomplementary acid sequences of human and rat androgen receptors.

Nail Acad

Sci USA

85:7211,

Corbishley

TP, Iqbal

M), Wilkinson

DNA Proc

13.

ML, Williams

R: Andnogen

pancreatic

tissue

and

thyroid

hormone

receptor

Life sciences.

Vol 4. Basel,

Faber PW, Kuiper GGJM, Rooij HCJ, van den KorputJAGM, mann AO, Trapman J: The N-terminal domain of the human gen receptor is encoded by one large exon. Mol Cell Endocninol

Brinkandro61:257,

Hodgins

MB,

Choudry

R, Parker

G, Oliver

RF, )ahoda

CAB,

papilla cells of scalp Acad Sci, in press

hair

follicles

in male

16. Horoszewicz J, Leong 5, Chu T, Wasjman L, Kim U, Chiu L, Kakati S, Arya 5, Sandberg new model Res 37:11, 17.

18.

20.

on human

prostatic

baldness.

Z, Friedman

Ann

H, Papsidero

A: The LNCaP cell line-a carcinoma. Prog Clin Biol

WC, Checkoway H, DiFernando G, Hammond DD, StumpfWE, Clark TD: Androgen receptors detected by autoradiography in prostatic carcinoma and benign prostatic hyperplastic tissue. Prostate 10:223, 1987

Hussmann

DA, Wilson

Lai LC, Ghilchik

MW,

between

epidermal

sulphate

in breast

Lansdorp

MW, McPhaul

MJ, Tilley WD, WilsonJD:

An-

antibodies to two distinct regions of the androgen receptor the receptor protein to the nuclei of target cells in the rat and prostate. Endocrinology 126:2359, 1990 Shaikh

NA,

Reed

MJ,James

VHT:

factor and dehydroepiandrostenone cyst fluid. Br J Cancer 60:320, 1989

PM, Van den Kwast ThH,

wise amplified

immunoperoxidase

ogy in relation

to membrane

Dc Boer M, Zeijlemaker (PAP)

J

markers.

staining.

I. Cellular

Histochem

Lea OA, Kvinnsland S, Thorsen gen receptors in human breast

22.

Lubahn DB, )oseph FS, Wilson EM: The

McGill HC, Sheridan P): Nuclear the cardiovascular system of the Mooradian

AD,

Endocr

MorrellJI,

),

Higgs receptor:

Morley)E,

Rev 8:1, Gresik

Korenman

of sex steroid hormones Circ Res 48:238, 1981

SG: Biological

actions

in

of andro-

1987

EW,

Barka

binding organs

uptake baboon.

T: Autoradiographic localization of diglands and other androgenCytochem 35:1053, 1987

in the major salivary ofthe mouse. ) Histochem

ER: Androgen

receptor

localization

in the human

using a new method of steBiochem 27:533, 1987

Press MF, Xu 5, Wang), Greene GL: gen receptor and progesterone receptor Am) Pathol 135:857, 1989

31.

Roehrborn CG, Lange amount and intracellular foreskin as a function

32.

Sanfilippo S, Imbesi RM: Is the spermatogonium an androgen target cell? An histochemical, immunocytochemical and ultrastructural study in the rat. In Motta PM, ed. Progress in clinical and biological research. Vol 296. New York, AR Liss, 1989, 177

33.

Sawaya

ME,

in sebaceous

Honig glands

Subcellular with and

distribution of estrowithout specific ligand.

)L,

George FW, Wilson )D: Changes in the distribution of androgen receptor in human of age. ) Clin Invest 79:44, 1987

IS, Hsia SL: Increased androgen binding ofmale-pattern baldness.J Invest Dermatol

capacity 92:91,

1988

34. Schulze

W, Rehder

seminiferous 35.

36.

U: Organization

epithelium.

Sheridan chemical Cytochem

Cell

Tissue

and morphogenesis Res

237:395,

ofthe

P), Buchanan )M, Anselmo V: Autoradiographic studies of hormone receptor localization. 29:195, 1981

Sheridan P), McGill roid gland contains 1984

human

1984

HC, Lissitzky)C, Martin PM: The receptors for androgens. Endocrinology

and

bio-

) Histochem

primate thy115:2154,

37. Trapman),

Klaassen P, Kuipen GGJM, van der Korput)AGM, Faber PW, van Rooij HC), Geurts van Kessel A, Voorhonst MM, Mulder E, Brinkmann AO: Cloning, structure and expression of a cDNA encoding the human androgen receptor. Biochem Biophys Res Commun 153:241, 1988

and its 38.

WP: Step-

cells.

morphol-

Cytochem

32:172,

HN, Larson RE, complementary

French deoxy-

Van Laar)H,

Bolt-dc Vnies), Zegers ND, Trapman), Bninkmann receptor heterogeneity and phosphorylation in human Biochem Biophys Res Commun 166:193, 1990

Androgen

AO:

LNCaP

39.

Van Laar)H, Voorhorst-Ogink MM, Zegers ND, Boersma W)A, Claassen E, van den KorputJAGM, Ruizeveld de WinterJA, van der Kwast ThH, Mulder E, Trapman), Bninkmann AO: Characterization of polyclonal antibodies against the N-terminal domain ofthe human androgen receptor. Mol Cell Endocninol 67:29 1989

40.

Weaker

T Improved measurement of androcancer. Cancer Res 49:7162, 1989

DR, Sar M, Tan human andnogen

Ann NY Acad Sci

30.

1984 21.

tract tissues.

FS, Wilreceptors

1989

prostate: demonstration of heterogeneity roid receptor autoradiography. ) Steroid

Relationship

growth

DR, French rat androgen

Peters CA, Barrack

NY

Hulka BS, Beckman JE, Fried FA, Mickey

tipeptide localize human 19.

for studies 1980

pattern

reproductive

and

29.

Withers

AP, Brinkmann AO, Boersma W)A, Van der Kwast ThH, Lammers KM, Wong TK, Wawrzyniak C), Warren R: Androgen receptors in dermal

human

Nagasue N, Kohno H, Chang Y, Hayashi T, Utsumi Y, Nakamura T, Yukaya H: Androgen and estrogen receptors in hepatocellular carcinoma and the surrounding liver in women. Cancer 63:112, 1989

1989 15.

VE, Sar M, Joseph of the

in pros-

28.

superfamily.

1989,

gene expression

Mullink H, Henzen-Logmans SC, Alons-van Kordelaar))M, Ldema TM, Meijer CJLM: Simultaneous immunoenzyme staining of vimentin and cytokeratins with monoclonal antibodies as an aid in the differential diagnosis of malignant mesothelioma from pulmonary carcinoma. Virchows Arch [BJ 52:55, 1986

recep-

Birkh#{228}user Verlag,

in male

and

27.

1988

Reports.

analysis

responsive

and cell lines.

Faber PW, Kuiper GGJM, van Rooij HCJ, van der KorputJAGM, Brinkmann AO, TrapmanJ: Characterization ofthe human androgen recepton. In Carlstedt-Duke J, Eniksson H, Gustafsson jA, eds. Birkh#{228}user Congress 169

14.

steroid

26.

1988

1 1. Corrales JJ, Pastor I, Garcia LC, Gonzalez )M, Montero J, Dc Castro 5, Minalles)M: Androgen receptor content in cytosol from non-tumoral human kidney and its relation to steroid hormone environment. Urol Int 40:307, 1985

12. Evans RM: The Science 240:889,

Quarmby

Structural

hydrotesterone

C, KokontisJ,

ton in human normal and malignant Cancer 57:1992, 1986

DB, Tan),

son EM:

gens.

androgen

1989

Lubahn

564:48,

fam-

1990 5: Prostate

acid cloning, sequence analysis Endocninol 2:1265, 1988

and expression 24.

receptor

and amino

9. Chang

10.

11:201,

H,

23.

matrix

BW: Steroid

Rev

E, Takeda

nibonucleic tate. Mol

F), Herbert

DC,

Sheridan

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015

PJ: Autoradiographic

demonstra-

936

RUIZEVELD

tion of binding sites for oestnadiol and dihydrotestosterone nary tract of male and female baboons. Unol Res 11:127, 41.

Weaken

IJ, Sheridan

testosterone

P): Autonadiognaphic

in the reproductive

localization

organs ofbaboons.

DE WINTER,

in the un-

44.

1983

115:244,

Weaker

hormones

Sheridan

P): Autonadiognaphic

in the lymphatic

organs

ofbaboons.

P), McGill

HC:

localization of sex steroid Cell Tissue Res 231:593,

1983 43.

Winbonn

WB,

Sheridan

stomach, liven, pancreas and gastrointestinal tnoenterology 92:23, 1987

Sex steroid

receptors

tract of the baboon.

in the

Gas-

WB,

ogy 84:1563,

nifenous

kJ

46.

Zegers

VERME

PJ, McGill

neceptons

ZEGERS,

HC:

VAN DER

Autoradiographic

in the esophagus

ofbaboons.

KWAST

demonstraGastroenterol-

1983

SJ, Troen tubules ND,

MUIDER,

Sheridan

ofestrogen

45. Winters

1983 42.

Winborn tion

of 3H-dihydro-

Acta Anat

TRAPMAN,

P: Evidence

for an androgen

receptor

of the human

testis. J Steroid

Biochem

Claassen

E, Neelen

MM, Benrevoets CA, Bninkmann de WintenjA, TrapmanJ, Boersma

C, Mulden

E, van

in the semi-

21:315, 1984 LaanJH, Voorhorst

AO, van den Kwast ThH, W)A: Epitope prediction

Ruizeveld and con-

firmation for the human androgen receptor: generation of monoclonal antibodies for multi assay performance following the synthetic peptide strategy. Biochem Biophys Acta 1973:23, 1991

Downloaded from jhc.sagepub.com at Scott Memorial Library @ Thomas Jefferson University on March 11, 2015