Molecular Human Reproduction vol.3 no.11 pp. 933–940, 1997
Region-specific expression of the androgen receptor in the human
epididymis
Hendrik Ungefroren1,3, Richard Ivell2 and Süleyman Ergün1,4
1Institute
of Anatomy, University of Hamburg, Martinistrasse 52, D-22046 Hamburg and 2Institute for Hormone and
Fertility Research, Grandweg 64, D-22529 Hamburg, Germany
3Present address: Research Unit Molecular Oncology, Clinic for General Surgery, Christian-Albrechts-University,
Arnold Heller Strasse 7, D-24105 Kiel, Germany
4To
whom correspondence should be addressed
The expression of the androgen receptor in the human epididymis was analysed by ribonuclease protection,
in-situ hybridization and immunohistochemistry. Androgen receptor mRNA and protein could be detected
throughout the entire organ, albeit in different quantities, in the caput, corpus and cauda regions, respectively.
Also positive, though only weakly, was the ductus deferens, while the efferent ducts were devoid of specific
signals. In-situ transcript hybridization and immunocytochemistry localized androgen receptor mRNA and
protein primarily to the epithelium of the epididymal duct. In the ductal epithelial cells androgen receptor
immunoreactivity showed a distinct nuclear distribution. While peritubular cells occasionally displayed weak
signals, interstitial cells as well as blood vessels were consistently negative throughout the entire organ. The
observed pattern of androgen receptor expression in the human epididymis supports the notion that the
structure and function of the epididymis is differentially controlled by androgens in a region-specific manner,
whereas it would not seem compatible with a direct role for androgens in the regulation of epididymal
blood flow.
Key words: androgen receptor/blood flow/epididymis/human/principal cells
Introduction
The mammalian epididymis is a highly differentiated organ,
whose various functional activities, e.g. absorption of testicular
fluid or secretion of proteins, serve the maturation and storage
of spermatozoa produced in the testis. Histologically, the
epididymis can be divided into three major parts: the caput,
the corpus and the cauda. In contrast to the epididymis of
other mammalian species the caput of the human organ is
composed predominantly of efferent ducts whereas corpus and
cauda are formed by the epididymal duct (Holstein, 1969;
Yeung et al., 1991). The efferent ducts contain non-ciliated
cells and various types of ciliated cells, while the epididymal
duct is lined by principal, light, basal and clear cells. All these
cell types have been attributed different functional activities
(Robaire and Hermo, 1988). The regional differentiation of
the human epididymis is also reflected by the pattern of specific
gene expression, which also varies along the length of the
epididymal duct (Krull et al., 1993).
Earlier studies have established that the epididymis strictly
requires testicular androgens for its development and maintenance of tissue homeostasis (Brooks, 1979; Orgebin-Christ,
1996). Androgens have also been implicated in the regulation
of epididymal blood flow (Setchell et al., 1964; Brown and
Waites, 1972). Androgen effects are principally mediated by
the androgen receptor (AR), a member of the steroid hormone
receptor superfamily (Carson-Jurica et al., 1990). Upon ligand
binding the hormone–receptor complex is translocated into the
© European Society for Human Reproduction and Embryology
nucleus, where it binds to DNA response elements in the
promoter region of target genes, thereby modulating their
transcriptional activity (Carson-Jurica et al., 1990).
The epididymal localization of the AR has been reported
for various species using biochemical (Tindall et al., 1975;
Carreau et al., 1984; Tekpetey et al., 1989; Paris et al., 1994)
as well as molecular genetic approaches (Cooke et al., 1991;
Blok et al., 1992; Viger and Robaire, 1995). However, although
the presence of AR in human epididymis is well documented
(Vazquez et al., 1986; Dankbar et al., 1995), its precise cellular
distribution in the adult organ has not yet been analysed. Since
this is a prerequisite for discerning direct androgen effects
from indirect ones and for understanding the morphological
and functional alterations eventually arising in this organ as a
result of a disturbed androgen supply and/or AR malfunction,
we have characterized AR expression in the human epididymis,
both at the RNA and protein levels.
Material and methods
Tissue samples
Human epididymides were obtained from 15 patients ranging in age
from 40 to 84 years. All were undergoing orchidectomy because of
prostatic carcinoma, but had not received any anti-androgen treatment
prior to removal of the testes and epididymides. None of the patients
were known to be infertile. A histological inspection of the tissues
revealed no pathological alterations in the testes and epididymides.
All epididymides were snap-frozen in liquid nitrogen ~1 h after
933
H.Ungefroren, R.Ivell and S.Ergün
surgery and stored at –80°C. Three of these that had previously been
dissected into caput, corpus and cauda regions were homogenized for
RNA extraction. The remaining 12 were cryosectioned in a longitudinal direction, such as to contain cross-sections through all major
segments within the same tissue section. Sections from 10 out of
these 12 epididymides were used for the immunohistochemical
and two for the in-situ hybridization experiments. In all cases
the Helsinki declaration regarding the use of human tissues was
strictly obeyed.
RNA isolation and ribonuclease protection assays (RPA)
Total RNA was extracted from homogenized epididymal tissues
with RNA Clean (Angewandte Gentechnologie Systeme, Heidelberg,
Germany) according to the manufacturer’s instructions.
The application of ribonuclease protection assays, a technique that
has previously been used for the detection and (semi)quantification
of AR mRNA transcripts in tissue samples (Dankbar et al., 1995),
served a dual purpose: to semiquantify AR mRNA in the head and
the body of the human epididymis, and to establish a RNA probe
suitable for subsequent use in the in-situ hybridization experiments.
RPA were performed with the RPA II kit (Ambion, Austin, TX) and
essentially carried out as described previously (Ungefroren et al.,
Figure 1. Semiquantitative assessment by ribonuclease protection
assay of relative androgen receptor mRNA levels present in the
caput, corpus and cauda regions of human epididymides. The
epididymal RNAs were from two different patients (patient 1: lanes
1, 3, and 7; patient 2: lanes 2 and 4). Lanes 1 and 2, caput
epididymis; lanes 3–4, corpus epididymis; lane 5, whole testis; lane
6, pancreatic carcinoma cell line BxPC3; lane 7, cauda epididymis;
lane 8, yeast RNA (negative control); lane 9, molecular weight
marker (PhiX174/HinfI); lane 10, undigested probe. The large
arrowhead points to the position of the protected fragments (415 b
in length), the small one to that of the undigested probe (463 b in
length).
1994). The probe used was a 463 bp EcoRI-PstI cDNA fragment
containing 415 bp (nucleotides 2482–2896 according to Chang et al.,
1988 and nucleotides 2796–3210 according to Lubahn et al., 1988)
from the ligand binding domain of the human AR. This fragment
was excised from the entire 3.6 kb cDNA inserted in the pGEM-3Z
plasmid vector (generously provided by Dr E.M.Wilson, North
Carolina, USA) and subcloned into the Bluescribe plasmid (Stratagene,
Heidelberg, Germany). After linearizing the plasmid with EcoRI,
antisense transcripts were produced using T3 RNA polymerase in
the presence of [α-32P]CTP (800 Ci/mmol, Amersham Buchler,
Braunschweig, Germany) to a specific activity of 53108 c.p.m./µg.
The corresponding sense transcripts (used as a negative control) were
synthesized with T7 polymerase after linearization of the plasmid
with Hind III. 1–23105 c.p.m. of the gel-purified probe were incubated
overnight at 42°C with the indicated amounts of total RNA. The
protected fragments were fractionated on a 5% polyacrylamide
sequencing gel and visualized by autoradiography on X-ray film
(Kodak X-Omat AR, Kodak, Rochester, NY).
In-situ hybridization
In-situ transcript hybridization followed a protocol that had earlier
been optimized for human testicular tissue (Ungefroren et al., 1995).
Briefly, 10 µm serial cryosections from whole epididymides were
fixed in 4% paraformaldehyde, acetylated and prehybridized in 20 mM
Tris–Cl, pH 7.5; 0.3 M NaCl; 1 mM EDTA, pH 8.0; 0.1 M
dithiothreitol; 13 Denhardt’s; 50% deionized formamide; 500 µg/ml
yeast tRNA; 500 µg/ml denatured salmon sperm DNA. Hybridization
was carried out overnight at 52°C in 40 µl prehybridization buffer
1 10% dextran sulphate containing ~13106 cpm antisense or sense
cRNA per slide (the same cRNA used in the ribonuclease protection
assay but labelled with 35S). The washing procedure and RNase A
digestion were as described previously (Ungefroren et al., 1995). The
optimal stringency of washing was determined empirically and found
to be 70°C. Slides were dried and exposed overnight to Kodak XAR
5 X-ray film to check for the optimal signal-to-noise ratio and to
obtain an estimate for subsequent long-term exposure under NTB 2
nuclear emulsion (Kodak). Slides coated with this emulsion were
then exposed in the dark for 73 days, developed, counterstained or
not in Mayer’s haemalum and photographed in darkfield reflectance
and brightfield illumination using a Nikon Epiphot microscope.
Immunohistochemistry
Cryostat sections (8 µm) through the entire epididymis were mounted
on chrome-gelatine-precoated slides, air-dried, and fixed in 4%
paraformaldehyde for 15 min at room temperature. Following a
pretreatment with 1.2% H2O2 in absolute methanol for 30 min (to
inhibit the endogenous peroxidase activity) sections were incubated
with either 2% normal swine serum (when a polyclonal anti-AR
antibody was to be used subsequently) or 2% normal rabbit serum
(when a monoclonal anti-AR-antibody was to be used subsequently)
to block non-specific binding sites. Incubation with the primary
Figure 2. In-situ hybridization of androgen receptor mRNA in the human epididymis. All photographs (a–e, g and h in darkfield reflectance,
f in brightfield) were taken from the same tissue section. (a) Proximal segments of the efferent ducts showing no specific label above the
background level (original magnification 3100). (b) Distal segments of the efferent ducts with some being negative and others weakly
positive (arrowhead, original magnification 3100). (c) Initial segment of the epididymal duct (d.e.); only cells of the epithelial layer are
positive (original magnification 3100). (d) Proximal/central part of the epididymal corpus, the epithelium which has increased in thickness
is strongly labelled (original magnification 350). (e, f) Enlargement (original magnification 3200) of (d); the heavily labelled epithelial cell
layer contrasts on the one hand with the peritubular cell layer (p) that appears only weakly positive and on the other hand with the blood
vessels (b) that are devoid of signal. (g) Negative control; corresponding area in an adjacent section treated with the sense cRNA (original
magnification 3200). (h) Caudal segment of the epididymal duct; the flat epithelium shows only few silver grains (arrowhead, original
magnification 3100); All sections were counterstained with haemalum and eosin.
934
Androgen receptor in human epididymis
Figure 2.
935
H.Ungefroren, R.Ivell and S.Ergün
antibody, either a rabbit polyclonal antiserum (AP-52) against AR
peptide (both generously provided by Dr E.M.Wilson), or a mouse
monoclonal antibody (MA-110) against human synthetic AR peptide
(both from Affinity Bio Reagents, Neshanic Station, USA) were
carried out for 48 h at 4°C in a humid chamber. Subsequently,
sections were processed for visualization of the antigen by use of the
peroxidase–antiperoxidase (PAP) method in combination with avidin–
biotin–peroxidase complex (ABC) procedure (Davidoff and Schulze,
1990). A 1 h incubation with a second (bridge) antibody, a biotinylated
anti-rabbit IgG or a biotinylated anti-mouse IgG was followed by an
incubation with the rabbit or mouse PAP complex and the ABC, each
for 30 min at room temperature. The peroxidase reaction was
developed with 3,39-diaminobenzidine and H2O2 in 50 mM Tris–HCl
(pH 7.6) for 10 min and the sections counterstained with Mayer’s
haemalum. The following controls were performed: (i) the primary
antiserum was replaced by antiserum that had been preadsorbed with
a synthetic peptide at 4°C for 24 h; (ii) the primary antiserum was
replaced by normal rabbit or mouse serum, respectively.
Results
To assess the relative amount of AR mRNA in each epididymal
region, we performed ribonuclease protection assays on RNA
extracted separately from the head, body and tail and compared
the AR signal intensities relative to those for the housekeeping
enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
(Figure 1). Upon hybridization with the antisense cRNA,
protected fragments of AR mRNA were present in the epididymal caput (lanes 1 and 2), corpus (lanes 3 and 4), and cauda/
vas deferens regions (lane 7), with the largest amount in the
corpus region (lanes 3 and 4). The signals obtained were
considered specific, because they are (i) of the predicted size
(415 nucleotides), (ii) present in human testis (positive control,
lane 5) but absent from yeast (lane 8) and pancreatic carcinoma
cell line BxPC3 RNA (lane 6) used as negative controls, and
(iii) absent from epididymal RNA when using the corresponding sense cRNA as a probe (data not shown).
Next, we attempted to obtain an assessment of the relative
steady-state mRNA levels for the AR in the various epididymal
segments and their flanking tissue, the efferent ducts and the
ductus deferens. However, application of the RPA was hampered by the close proximity of efferent ducts and part of the
proximal epididymal duct in the caput, which did not allow a
good separation of these two tissue components of the epididy-
mal head. As a result, cross-contamination with RNA from
the neighbouring region(s) would be expected to obscure/
dilute putative differences in AR mRNA content. Therefore
we used in-situ transcript hybridization, a technique that
has been employed successfully for the characterization of
epididymal gene expression (Krull et al., 1993; Pera et al.,
1994) as well as for semiquantification of AR mRNA levels
in the prostate (Takeda et al., 1991). This allowed a more
detailed comparison of AR mRNA levels among the various
epididymal segments, besides locating the AR mRNA on a
cellular level.
Microscopical inspection under darkfield and brightfield
illumination showed clear differences in signal intensities
among the various epididymal segments, obviating the need
for a rigorous quantification of autoradiographic signals by
image analysis. There were numerous silver grains present
over the epithelial cells of the epididymal duct in the proximal
corpus (Figure 2d); these were slightly less abundant over the
proximal epididymal duct in the distal caput (Figure 2c). Weak
signals were seen in the transition zones between the proximal
part of the epididymal duct and the efferent ductules on the
one hand (Figure 2b) and the epididymal tail/ductus deferens
on the other hand (Figure 2h). When compared to Figure 2d
and e, it is apparent that in Figure 2h, although AR mRNA is
still present, it is much less abundant as judged by the density
of silver grains per square unit. The efferent ducts, connecting
the epididymis with the rete testis, displayed no signal above
the background level (Figure 2a). Also negative were connective tissue cells and blood vessels (Figure 2e and f), while
peritubular cells appeared occasionally positive, particularly
those of the inner layers (Figure 2e). The specific localization
of the transcripts within the ductular epithelium (Figure 2e
and f) was confirmed by the absence of silver grains (beyond
a background level of signal) when adjacent tissue sections
were hybridized with [35S]CTP-labelled sense cRNAs as negative controls (Figure 2g).
Consistent with the in-situ hybridization results AR immunoreactivity was found along the entire epididymal duct, primarily
in the principal epithelial cells but also in the basal cells of
the epididymal duct (Figure 3b–h). The nuclei were labelled
much more strongly than the cytoplasm [please note in the
counterstained sections of Figure 3 (b, d, f and h) the difference
Figure 3. Distribution of androgen receptor (AR) immunostaining in various parts of the adult human epididymis. In contrast to the sections
in a, c, e and g, those in b, d, f, and h were counterstained with haemalum and eosin. Please note that nuclei of AR-positive cells appear
brown, whereas those of AR-negative cells have a pink colour. (a) Proximal and middle parts of efferent ducts lacking cytosolic and nuclear
AR staining (original magnification 3200). (b) Transitional zone of efferent ducts faint cytosolic and strong nuclear immunostaining is
visible, regions closer to the epididymal duct (right half) display a higher percentage of positive cells. Some peritubular cells (arrowhead)
and interstitial cells (arrow) show a positive immunoreaction in the cytosol (original magnification 3200). (c) Nuclear labelling in the basal
and principal cells of the proximal part of the epididymal duct. Blood vessels are negative (star) (original magnification 3200). (d) In this
counterstained section of the epididymal duct strong nuclear AR immunostaining is present in the basal and principal cells. Cytosolic
immunostaining is weak in peritubular cells and absent from blood vessels (star) (original magnification 3200). (e) Epididymal duct from
the corpus region. The nuclear immune reaction for AR in basal and principal cells is not as prominent as in the proximal region (original
magnification 3200). (f) Higher magnification of a counterstained section from the corpus region show considerable immunoreactivity in
basal and principal cells, while that over peritubular cells is faint and absent from blood vessels (star, original magnification 3200). (g and
h) Caudal region of the epididymal duct. Compared with the proximal part and the corpus region the cytosolic and nuclear AR
immunoreactivity is much weaker (original magnification 3200). All photographs were taken from the same longitudinal section of the
epididymis analysed in Figure 2.
936
Androgen receptor in human epididymis
Figure 3.
937
H.Ungefroren, R.Ivell and S.Ergün
in colour between AR-positive cells (nuclei appear brown)
and AR-negative cells (nuclei appear pink)]. In agreement
with the AR mRNA data, immunoreactivity was generally not
seen over the interstitial tissue (apart from transitional zone
of the caput), whereas the peritubular cells occasionally displayed a diffuse immunostaining (Figure 3b and f). In all
specimens examined AR immunostaining of epithelial cells
was more prominent in the proximal regions of the epididymal
duct when compared with the distal ones. No immunoreactivity
could be detected in the proximal region of the efferent ducts,
either in the epithelial or in the stromal cells (Figure 3a), but
in the distal part a considerable proportion of epithelial cells
displayed already a nuclear and cytoplasmic immunostaining
(Figure 3b). A comparison of AR staining between the initial
segment (Figure 3c and d) and the caudal part (Figure 3g and
h) of the epididymis revealed firstly that, in contrast to the
initial segment and proximal corpus region, a fraction of
epithelial cells in the caudal segment was not stained and,
secondly, that the nuclear staining in the proximal corpus
region was stronger. In all segments of the caput and corpus,
blood vessels (including endothelial and vascular smooth
muscle cells of arteries, arterioles and veins) were devoid of
specific label (Figure 3d and f).
The pattern of AR expression in the various parts of the
human epididymis has been summarized schematically in
Figure 4.
Discussion
The results from in-situ hybridization and immunocytochemistry presented in this study identified the epithelium of
the epididymal duct as the primary site of AR expression,
while the connective tissue stroma and the blood vessels
lacked specific signals throughout the organ. Furthermore, by
employing RPA and in-situ hybridization this study also
showed a region-specific quantitative variation of AR mRNA
and protein, being high in the distal caput and proximal corpus
but comparatively low in the proximal caput and the caudal
region including the vas deferens. In contrast, the efferent
ducts were found to be negative in both the in-situ hybridization
and immunohistochemical experiments except for some
epithelial cells in the transitional zone.
When compared with equivalent data from other species
some common features as well as some discrepancies were
evident. As in these results from the human the epithelial cells
were consistently identified as the major site of expression
(by immunocytochemistry and protein binding studies) and
exhibited a nuclear antigen distribution in all species examined
so far [adult rhesus macaque: Roselli et al. (1991); rat: Sar
et al. (1990), Paris et al. (1994); ram: Tekpetey et al. (1989)].
However, unlike results from the present study, stromal cells
were reported to be positive in the rhesus macaque (Roselli
et al., 1991), the mouse (Cooke et al., 1991) and the rat
(Sar et al., 1990). Also, in contrast to the human and ram
epididymides (Tekpetey et al., 1989), the efferent ducts showed
an intense staining for the antigen in the adult macaque (Roselli
et al., 1991), and the mouse (Cooke et al., 1991) and the rat
(Bentvelsen et al., 1995) embryo, although in the latter two
938
Figure 4. Diagrammatic representation of androgen receptor
expression in the human epididymis. The structure of the human
epididymis as it is given here is based on the histological studies
by Holstein (1969) and Yeung et al. (1991).
(developmental) studies it was not investigated whether the
efferent ducts retain AR expression during adulthood.
Furthermore, the region-specific variation of AR expression,
demonstrated here for the first time in the human epididymis,
has been described in the adult rhesus macaque (Roselli et al.,
1991), the ram (Tekpetey and Amann, 1988; Tekpetey et al.,
1989), and the rabbit (Toney and Danzo, 1988), but seems to
be absent from the rat (Sar et al., 1990) and the fetal mouse
epididymis (Cooke et al., 1991). The differences observed
may be due to technical heterogeneities, e.g. the use of different
antibodies (Tekpetey et al., 1989; Roselli et al., 1991) and/or
may reflect true species-specific variations.
The epididymal distribution of AR contrasts sharply with
that of the oestrogen receptor. Recent immunohistochemical
studies from our laboratory revealed strong expression in the
efferent ducts and the rete testis but only a marginal expression
in the epididymal duct (Ergün et al., 1997). This probably
reflects the different developmental history/embryonic origin
of the efferent ducts (derived from mesonephric tissue) and
the epididymal duct (derived from the Wolffian duct).
The heterogeneous signal distribution for AR expression
along the human epididymis corresponds well with a regionally
different dependence of this organ on androgens. In order to
Androgen receptor in human epididymis
study which regions are androgen-dependent, several experimental and physiological situations of androgen depletion have
been studied; these include orchidectomy (Goyal et al., 1994;
Fawcett and Hoffer, 1979), ligation of extratesticular rete testis
or efferent ducts (Abe and Takano, 1989; Fawcett and Hoffer,
1979; Goyal et al., 1994), chemical destruction of Leydig cells
(Blok et al., 1992), antiandrogen treatment (Paris et al., 1994),
each with or without testosterone replacement, or an agerelated (Viger and Robaire, 1995) or season-related
(Schindelmeiser et al., 1988) drop in androgen concentrations.
Structural and biochemical/functional alterations occurred most
dramatically in the initial segment of the epididymal duct
located predominantly in the distal caput and proximal corpus
region of the human epididymis, with little or no effect in the
distal corpus and cauda. The structural changes have been
reported in detail in the goat epididymis (Fawcett and Hoffer,
1979; Abe and Takano, 1989; Goyal et al., 1994) and included
involution/degeneration of epithelial cells, loss of stereocilia
and, as a result, disintegration of the epithelial cell layer
(changes that possibly indicate cell death via apoptosis).
Functional consequences of androgen withdrawal were
apparent, e.g. as a reduction in the protein-synthesizing ability
of the initial segment (Jones et al., 1981), the requirement of
mitochondrial enzymes of the rat cauda epithelium for a lower
amount of androgen than those in the caput to express maximal
activity (Brooks, 1979), or a differential reduction in nitric
oxide synthase (NOS) activity after castration (Chamness et
al., 1995).
Studies in the goat epididymis have shown that the androgen
requirement does not only differ depending on the region of
the epididymis but also on the source of androgen. Androgen
delivery to the epididymis is principally via two main routes,
the general circulation (CA), and, as part of the luminal fluid,
through the efferent ducts (LA). A well-conducted study in
the goat underscored the relative importance for rete fluid and,
additionally, for LA (Fawcett and Hoffer, 1979) in maintaining
the epithelial structure of the epididymal head compared with
the more distal regions (Goyal et al., 1994). This is in
concordance with experiments concerning ligation of the
extratesticular rete testis (specific block of LA) (Brooks, 1981)
although the relative contribution/importance of these routes
for androgen supply of the human epididymis and thus maintenance of its function still requires further elucidation.
Earlier work suggested that androgens themselves are
involved in blood flow regulation based on the observations
that the higher blood flow observed in areas with strong
absorptive and secretory activity disappears after orchidectomy
or ligation of extratesticular efferent ducts (Setchell et al.,
1964; Brown and Waites, 1972). We have shown here for the
first time that blood vessels (including arteries, arterioles and
capillaries as well as epididymal veins) lacked positive signals
for AR gene products, both in the in-situ hybridization and
immunohistochemical experiments. Thus it is likely that CA
or LA, rather than acting directly on blood vessels, affect the
expression or the release of vasoactive agents such as NO
(Chamness et al., 1995; Burnett et al., 1995) and probably
endothelin, the latter of which has been co-localized immunohistochemically with AR protein in the epididymal duct
(S.Ergün, unpublished). This assumption is supported by the
recently discovered positive effects of testosterone on NOsynthetase production (Chamness et al., 1995).
Acknowledgements
We thank S.Schwarz for excellent technical assistence and Dr
A.F.Holstein for his continuous interest and support. This work
was supported by the Bundesminister für Bildung, Wissenschaft,
Forschung und Technologie (BMBF), Bonn, Germany, as a part of a
larger concerted project ‘Fertilitätsstörungen’ (01 KY 9103).
References
Abe, K. and Takano, H. (1989) Early degeneration of the epithelial cells in
the initial segment of the epididymal duct in mice after efferent duct
cutting. Arch. Histol. Cytol., 52, 299–310.
Bentvelsen, F.M., Brinkmann, A.O., van der Schoot, P. et al. (1995)
Developmental pattern and regulation by androgens of androgen receptor
expression in the urogenital tract of the rat. Mol. Cell. Endocrinol., 113,
245–253.
Blok, L.J., Bartlett, J.M., Bolt De Vries, J. et al. (1992) Effect of testosterone
on expression of the androgen receptor in rat prostate, epididymis and
testis. Int. J. Androl., 15, 182–198.
Brooks, D.E. (1979) Influence of androgens on the weights of the male
accessory reproductive organs and on the activities of mitochondrial
enzymes in the epididymis of the rat. J. Endocr., 82, 293–303.
Brooks, D.E. (1981) Metabolic activity in the epididymis and its regulation
by androgens. Physiol. Rev., 61, 515–555.
Brown, P.D.C. and Waites, G.M.H. (1972) Regional blood flow in the
epididymis of the rat and rabbit: effect of efferent duct ligation and
orchidectomy. J. Reprod. Fertil., 28, 221–233.
Burnett, A.L., Ricker, D.D., Chamness, S.L. et al. (1995) Localization of
nitric oxide synthase in the reproductive organs of the male rat. Biol.
Reprod., 52, 1–7.
Carreau, S., Drosdowsky, M.A. and Courot, M. (1984) Androgen-binding
protein in sheep epididymis: characterization of a cytoplasmic receptor in
the ram epididymis. J. Endocrinol., 103, 273–279.
Carson-Jurica, M.A., Schrader, W.T. and O’Malley, B.W. (1990) Steroid
receptor family: Structure and functions. Endocrine Rev., 11, 201–220.
Chamness, S.L., Ricker, D.D., Crone, J.K. et al. (1995) The effect of androgen
on nitric oxide synthase in the male reproductive tract of the rat. Fertil.
Steril., 63, 1101–1107.
Chang, C., Kokontis, J. and Liao, S. (1988) Molecular cloning of human and
rat complementary DNA encoding androgen receptors. Science, 240,
324–326.
Cooke, P.S., Young, P. and Cunha, G.R. (1991) Androgen receptor expression
in developing male reproductive organs. Endocrinology, 128, 2867–2873.
Dankbar, B., Sohn, M., Nieschlag, E. et al. (1995) Quantification of androgen
receptor and follicle-stimulating hormone receptor mRNA levels in human
and monkey testes by a ribonuclease-protection assay. Int. J. Androl., 18,
88–96.
Davidoff, M.S. and Schulze, W. (1990) Combination of the peroxidase antiperoxidase (PAP)- and avidin–biotin–peroxidase complex (ABC)techniques: an amplification alternative in immunocytochemical staining.
Histochemistry, 93, 531–536.
Ergün, S., Ungefroren, H., Holstein, A.F. and Davidoff, M.S. (1997) Estrogen
and progesterone receptors and estrogen receptor related antigen (ER-D5)
in human epididymis. Mol. Reprod. Dev., 47, 448–455.
Fawcett, D.W. and Hoffer, A.P. (1979) Failure of exogenous androgen to
prevent regression of the initial segment of the rat epididymis after efferent
duct ligation or orchidectomy. Biol. Reprod., 20, 162–181.
Goyal, H.O., Hutto, V. and Maloney, M.A. (1994) Effects of androgen
deprivation in the goat epididymis. Acta Anat., 150, 127–135.
Holstein, A.F. (1969) Morphologische Studien am Nebenhoden des Menschen.
In Bargmann, W. and Doerr, W. (eds), Zwanglose Abhandlungen aus dem
Gebiet der normalen und pathologischen Anatomie. Georg Thieme Verlag,
Stuttgart, pp. 1–91.
Krull, N.B., Ivell, R., Osterhoff, C. and Kirchhoff, C. (1993) Region-specific
variation of gene expression in the human epididymis as revealed by in
situ hybridization with tissue-specific cDNAs. Mol. Reprod. Dev., 34, 16–24.
Lubahn, D.B., Joseph, D.R., Sar, M. et al. (1988) The human androgen
939
H.Ungefroren, R.Ivell and S.Ergün
receptor: complementary deoxyribonucleic acid cloning, sequence analysis
and gene expression in prostate. Mol. Endocrinol., 2, 1265–1275.
Orgebin-Christ, M.C. (1996) Androgens and epididymal function. In Shalender,
X., Basin, S. et al. (eds), Pharmacology, Biology, and Clinical Applications
of Androgens. Wiley, New York.
Paris, F., Weinbauer, G.F., Blum, V. and Nieschlag, E. (1994) The effect of
androgens and antiandrogens on the immunohistochemical localization of
the androgen receptor in accessory reproductive organs of the male rat.
J. Steroid Biochem. Mol. Biol., 48, 129–137.
Pera, I., Ivell, R. and Kirchhoff, C. (1994) Regional variation of specific
gene expression in the dog epididymis as revealed by in-situ transcript
hybridization. Int. J. Androl., 17, 324–330.
Robaire, B. and Hermo, L. (1988) Efferent ducts, epididymis, and vas deferens:
structure, functions, and their regulation. In Knobil, E., Neill, J. et al. (eds),
The Physiology of Reproduction. Raven Press, New York, pp. 999–1080.
Roselli, C.E., West, N.B. and Brenner, R.M. (1991) Androgen receptor and 5
alpha-reductase activity in the ductuli efferentes and epididymis of adult
rhesus macaques. Biol. Reprod., 44, 739–745.
Sar, M., Lubahn, D.B., French, F.S. and Wilson, E.M. (1990)
Immunohistochemical localization of the androgen receptor in rat and
human tissues. Endocrinology, 127, 3180–3186.
Schindelmeiser, J., Kutzner, M., Bergmann, M. et al. (1988) Influence of long
and short photoperiods on the morphology and androgen receptor levels of
the epididymis and the ductus deferens of Phodopus sungorus. Andrologia,
20, 507–515.
Setchell, B.P., Waites, G.M.H. and Till, A.R. (1964) Variations in blood flow
within the epididymis and testis of the sheep and rat. Nature, 203, 317–318.
Takeda, H., Nakamoto, T., Kokontis, J. et al. (1991) Autoregulation of
androgen receptor expression in rodent prostate: immunohistochemical and
in situ hybridization analysis. Biochem. Biophys. Res. Commun., 177,
488–496.
Tekpetey, F.R. and Amann, R.P. (1988) Regional and seasonal differences in
concentrations of androgen and estrogen receptors in ram epididymis. Biol.
Reprod., 38, 1051–1060.
Tekpetey, F.R., Veeramachaneni, D.N. and Amann, R.P. (1989) Localization
of androgen receptors in ram epididymal principal cells. J. Reprod. Fertil.,
87, 311–319.
Tindall, D.J., Hansson, V., McLean, W.S. et al. (1975) Androgen-binding
proteins in rat epididymis: properties of a cytoplasmic receptor for androgen
similar to the androgen receptor in ventral prostate and different from
androgen-binding protein (ABP). Mol. Cell. Endocrinol., 3, 83–101.
Toney, T.W. and Danzo, B.J. (1988) Developmental changes in and hormonal
regulation of estrogen and androgen receptors present in he rabbit
epididymis. Biol. Reprod., 39, 818–828.
Ungefroren, H., Wathes, D.C., Walther, N. and Ivell, R. (1994) Structure of
the alpha-inhibin gene and its regulation in the ruminant gonad: inverse
relationship to oxytocin gene expression. Biol. Reprod., 50, 401–412.
Ungefroren, H., Ergün, S., Krull, N.B. and Holstein, A.F. (1995) Expression
of the small proteoglycans biglycan and decorin in the adult human testis.
Biol. Reprod., 52, 1095–1105.
Vazquez, M.H., de Larminat, M.A. and Blaquier, J.A. (1986) Effect of androgen
on androgen receptors in cultured human epididymis. J. Endocrinol., 111,
343–348.
Viger, R.S. and Robaire, B. (1995) Gene expression in the aging brown
Norway rat epididymis. J. Androl., 16, 108–117.
Yeung, C.H., Cooper, T.G., Bergmann, M. and Schulze, H. (1991) Organization
of tubules in the human caput epididymis and the ultrastructure of their
epithelia. Am. J. Anat., 191, 261–279.
Received on March 3, 1997; accepted on August 14, 1997
940