| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
, ß, and 
, and Retinoid X Receptors , ß, and in the Mouse Testis1
Department of Cell Biology, Utrecht University Medical School (I.C.G., A.M.M.v.P., D.G.d.R.), and Hubrecht Laboratory, Netherlands Institute for Developmental Biology (P.T.v.d.S.), Utrecht; and the Department of Endocrinology and Reproduction, Erasmus University Rotterdam (J.W.H., A.P.N.T.), Rotterdam, The Netherlands
Address all correspondence and requests for reprints to: Dr. D. G. de Rooij, Department of Cell Biology, Medical School, Utrecht University, Postbus 80.157, 3508 TD Utrecht, The Netherlands. E-mail: d.g.derooij{at}med.ruu.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,
-ß, and -
, was studied in normal mice and in vitamin A-deficient
mice after the administration of all-trans-retinoic acid
(ATRA). All three types of RARs were expressed in normal and/or vitamin
A-deficient testes. Only the expression of RARß messenger RNA was
transiently induced within 24 h after ATRA injection. ATRA-induced
RARß expression was also found in purified Sertoli cells, suggesting
that these cells mediate at least part of the effect of retinoids on
germ cells. When an equimolar amount of retinol was administered
instead of ATRA, no induction of RARß was seen at the point of
maximal induction by ATRA, suggesting that the effect of retinol was
delayed and probably less.
The related nuclear receptors, RXR, -ß, and, for the first time,
, were also shown to be present in the mouse testis. Upon
administration of ATRA, messenger RNA expression of RXR and -ß did
not change significantly. The expression of RXR was too low to allow
quantification.
Finally, the effect of the retinoid metabolism inhibitor liarozole on ATRA-induced proliferation of A spermatogonia was examined. The labeling index of A spermatogonia, 24 h after the administration of 0.25 mg ATRA, was significantly lowered by liarozole due to a shift of the maximal 5-bromo-deoxyuridine incorporation to an earlier point (20 h). This indicates that liarozole delays retinoid metabolism, thereby increasing the actual ATRA concentration, and more importantly, that ATRA by itself is an active retinoid in spermatogenesis. Apparently, ATRA does not need to be metabolized to 4-oxo-RA, which was previously shown to be a more potent inducer of spermatogonial proliferation than ATRA, to be effective.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In both rats and mice, spermatogenesis can be reinitiated by the administration of retinol (1, 2, 3, 4, 5, 6) or ATRA (7, 8), resulting in a synchronized formation of A1 spermatogonia that start to proliferate. Recently, we found that 4-oxo-RA, a metabolite of ATRA, is able to reinitiate spermatogenesis in an even more efficient way than ATRA (8). At present, it is not clear which retinoid is the most important in supporting spermatogenesis in vivo. It also has not been established whether the effects of retinoids on spermatogonial proliferation and differentiation are direct or indirect, for example via Sertoli cells.
Binding proteins and nuclear receptors may play an important role in
mediating the effects of retinoids. In the rat testis, both cellular
retinol-binding proteins and cellular retinoic acid-binding proteins
are present (9, 10, 11, 12, 13) as are the nuclear retinoic acid receptors, RAR,
-ß, and - and retinoid X receptor (RXR) (12, 14, 15, 16, 17, 18, 19). Retinoid
status and administration of retinoids or testosterone were shown to
affect the expression of some of these genes (15, 16, 17, 18, 19). The involvement
of RARs in spermatogenesis was also shown by the generation of RAR null
mutant mice. Due to differential promoter usage and alternative
splicing, several isoforms of each nuclear retinoid receptor exist
(reviewed in Ref.20). Although RAR1-,
RARß2-, and RAR2 null mutants appeared
phenotypically normal, indicating a functional redundancy among the
RARs, complete RAR or - null mutants displayed male sterility
among other defects (21, 22, 23). However, mice lacking all RARß isoforms
develop apparently normally and are fertile (24). This does not
necessarily mean that RARß is not involved in spermatogenesis,
because the response of a substituting receptor may be adequate.
RARs were shown to bind to responsive elements in the promoters of some genes, not as homodimers but as heterodimers with RXRs (25, 26, 27, 28, 29, 30). Both ATRA and 9-cis-RA are ligands for the RARs, but 9-cis-RA is also the ligand for the RXRs (31). A rapid isomer exchange between ATRA and 9-cis-RA, however, is possible (32).
In the present study, the testicular messenger RNA (mRNA) expression of RARs after ATRA administration to VAD mice was studied in an attempt to elucidate mechanisms by which retinoids affect spermatogenesis. It was shown for the first time that all three RXRs are present in the adult VAD mouse testis. By blocking the metabolism of ATRA to 4-oxo-RA with the inhibitor liarozole, it was determined whether ATRA itself is able to reinitiate spermatogenesis or whether it has to be metabolized to 4-oxo-RA first.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
VAD animals in Exp 1 each received an ip injection of 0.5 mg ATRA [in 16% dimethylsulfoxide (DMSO)-H2O]. Control VAD animals received an equal amount of 16% DMSO-H2O. Control normal mice were killed without any treatment. At several time intervals after ATRA administration, animals were killed by CO2. VAD animals in Exp 2 each received an ip injection of 0.5 mg ATRA (in 16% DMSO-H2O) or 0.5 mg retinol acetate (in 16% DMSO-H2O). Animals were killed by CO2, 6 h after retinoid administration. From all animals in Exp 1 and 2, both testes were removed, which, after isolation of a small part for histological examination, were frozen in liquid nitrogen and stored at -80 C.
In Exp 3, VAD mice each received 0 mg (control) or 0.5 mg ATRA (in 16% DMSO/H2O), ip, and were killed 6 h later by CO2. From each animal both testes were removed for Sertoli cell isolation (three or four mice per group).
In Exp 4A, VAD animals were injected ip with liarozole (40 mg/kg BW; in 20% ethanol-sesame oil) (33) 1 h before ip injection of 0, 0.25, or 0.5 mg ATRA (in 16% DMSO-H2O). Twenty-four hours after ATRA administration, animals were killed by CO2 (at least four animals per group). In Exp 4B, four VAD animals were injected ip with liarozole (40 mg/kg BW) 1 h before ip injection of 0.25 mg ATRA (in 16% DMSO-H2O). Animals were killed 20 h after ATRA administration by CO2. All animals in Exp 4A and 4B received an ip injection of 5-bromo-deoxyuridine (BrdU; 100 mg/kg BW; Sigma Chemical Co., St. Louis, MO) 2 h before they were killed. During all experiments animals remained on the VAD diet.
Retinoids and chemicals
Retinol acetate and all-trans-RA were purchased from
Sigma. All retinoids were dissolved at 10-1 M
in DMSO and stored in the dark at -20 C. Liarozole (R 75251) was a
gift from Dr. W. Wouters (Janssen Research Foundation, Beerse,
Belgium). It was dissolved at 10-1 M in
ethanol and stored at -20 C.
Histology
Small parts of testes from animals in Exp 1 and 2 were fixed in
Bouin’s fluid and embedded in glycol methacrylate (Technovit 7100,
Kulzer, Wehrheim, Germany). Three-micron sections were made and stained
with periodic acid-Schiff and Gill’s hematoxylin 3. All sections were
examined using light microscopy to check the VAD status of the animals
used.
Sertoli cell isolation
Cell preparations were made of testes of VAD mice and of mice
killed 6 h after ATRA administration. Two or three animals were
used for each cell preparation. The method used was, in principle, a
combination of the methods of Themmen et al. (initial steps)
(34) and Van Pelt et al. (final steps) (35). Testes were
decapsulated and incubated in a shaking water bath (45 min, 33 C, 90
rpm) in 10 ml PBS containing 1.1 mM CaCl2, 0.5
mM MgCl2, 1 mg/ml trypsin (Worthington
Biochemcial Corp., Freehold, NJ; 192 U/mg), 1 mg/ml collagenase (type
IV, Worthington; 213 U/mg), and 2 mg/ml hyaluronidase (Sigma; 450
U/mg). After sedimentation, the tubules were washed twice in PBS
without Ca/Mg. In this way a large amount of the interstitial cells was
already removed, and single tubule fragments were obtained. After
repeated pipetting, a second incubation in 10 ml PBS (without Ca/Mg and
enzymes) was carried out for 10 min in a shaking water bath at 33 C and
120 rpm. In this way, most of the tight junctions between the Sertoli
cells were disconnected, and a cell suspension was obtained. The
remaining large cell clusters (interstitial cells and/or Sertoli cells)
were removed by swirling them around a Pasteur pipette. After this, the
volume of the cell suspension was brought to 50 ml with PBS and Ca/Mg
and centrifuged for 3 min at 175 x g. The cell pellet
was resuspended in 3 ml MEM and filtered over a 200-µm filter to
remove remaining cell clusters. Cells were pelleted by centrifugation
at 110 x g for 5 min and resuspended in 5 ml MEM. This
was repeated three times. Cells were then plated on a peanut
agglutinin-coated petri dish for 1 h at 34 C to remove any
(degenerating) spermatids that were occasionally present in the VAD
mouse testes. The free cells were loaded on a discontinuous Percoll
gradient (65%, 50%, 40%, 36%, 34%, 32%, 30%, 28%, and 20%
Percoll) to separate the remaining cell types (35). After
centrifugation, fractions were collected and examined using Nomarski
microscopy. In addition, a small part of the cells in each fraction was
used to make cytospin preparations on poly-L-lysine coated
slides, which were fixed in Bouin’s fluid, stained with Mayer’s
hematoxylin, and examined by light microscopy. Two fractions containing
purified Sertoli cells were pooled. The integrity of the cells appeared
normal, and the viability of the cells was not checked, as the cells
were intended for (immediate) RNA isolation.
RNA isolation
Total RNA was isolated using the LiCl-urea method (36). Briefly,
frozen testes or isolated cells were lysed and homogenized in 3
M LiCl-6 M urea with a Polytron (Brinkmann
Instruments, Westbury, NY) and left on ice for at least 3 h. The
RNA was pelleted through ultracentrifugation (30,000 rpm, 25 min;
Beckman TLA 100.3 rotor, Palo Alto, CA). The pellet was dissolved in
0.1% SDS, and protein was removed by repeated phenol extractions.
After ethanol precipitation, the amount of RNA was determined
spectrophotometrically. After a second ethanol precipitation, the RNA
was dissolved at 5 µg/µl.
Ribonuclease (RNase) protection assay probes
Complementary RNA was obtained by in vitro
transcription of pBluescript II SK- plasmid (Stratagene,
La Jolla, CA) containing one of the following murine complementary DNA
fragments [numbering according to Zelent et al. (14) (RARs)
and Leid et al. (29) (RXRs)]: RAR, 129 bp (nucleotides
433–562); RARß, 116 bp (nucleotides 1226–1342); RAR, 202 bp
(nucleotides 1170–1372); RXR, 275 bp (nucleotides 1500–1775);
RXRß, 176 bp (nucleotides 2007–2183); RXR, 222 bp (nucleotides
1488–1710); glyceraldehyde-3-phosphate dehydrogenase (GAPD), 58 bp,
equivalent to nucleotides 346–404 of rat GAPD ]these clones were
described by Jonk (36a)]. To exclude cross-detection, the RAR and RXR
probes were derived from regions of their complementary DNA that
exhibited high sequence divergence. The probe detecting transcripts
from the uniformly expressed GAPD gene was included as an internal
control to which all quantification data were normalized.
Transcriptions were performed in the presence of [32P]UTP
(800 Ci/mmol; Amersham Corp., Arlington Heights, IL). After synthesis,
the DNA templates were degraded with 10 U deoxyribonuclease I
(Boehringer Mannheim, Indianapolis, IN) for 20 min, and the synthesized
RNAs were further purified by electrophoresis on a 6% polyacrylamide-7
M urea-Tris-borate-EDTA gel.
RNase protection assay
Equal amounts of total RNA each were hybridized overnight at 47
C with 5 x 104 cpm RAR or RXR probe and 1 x
104 cpm GAPD probe (labeled to a 3-fold lower specific
activity) in 30 µl 80% formamide, 400 mM NaCl, 40
mM PIPES (pH 6.4), and 1 mM EDTA. After
hybridization, nonhybridized RNA was digested by adding 350 µl RNA
digestion buffer [10 mM Tris-HCl (pH 7.5), 300
mM NaCl, and 5 mM EDTA]. Per ml, this buffer
contained 1300 U RNase T1 (Life Technologies, Gaithersburg, MD) for
RARß and RXRß; 1950 U for RAR, RAR, and RXR; and 2600 U
for RXR, respectively. Samples were incubated for 60 min at 37 C.
The remaining (hybridized) RNA was purified by proteinase K digestion,
phenol-chloroform extraction, ethanol precipitation, and
electrophoresis on 6% polyacrylamide-7 M
urea-Tris-borate-EDTA gels. Gels were fixed, dried, and exposed to a
phosphor screen (Molecular Dynamics, Sunnyvale, CA), followed by
exposure to Fuji film at -80 C. Quantification of the protected bands
was performed using a Molecular Dynamics PhosphorImager with ImageQuant
software. The transfer RNA (tRNA) signal was used as a background
value. All data were normalized against GAPD values.
Determination of labeling index
To determine the BrdU labeling index, testes of each animal in
Exp 4A and 4B were fixed in Carnoy’s fixative and embedded in
Technovit 7100. Three-micron sections were made and processed
immunohistochemically using the BrdU-Immunogold-silver staining
(BrdU-IGSS) as described by Van de Kant and De Rooij (37). Using an
epipolarization microscope, the percentage of BrdU-labeled A
spermatogonia was determined. Per testis, at least 200 A spermatogonia
were counted.
Statistics
After testing for homoscedasticity, using the Bartlett’s test
for homogeneity of variances, variances proved to be homogeneous, and
the Tukey-Kramer method was used. Data are expressed as the mean
± SD.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, but not RAR, mRNA could be detected in normal mouse
testes by RNase protection assays (results not shown). In VAD mouse
testes, relatively more RAR mRNA was found, and in these testes,
RAR mRNA was detected also. No significant change in the expression
of RAR or - was observed within 24 h after retinoid
administration to VAD mice. A representative autoradiograph is shown in
Fig. 1
.
|
). In contrast, very weak expression of
RARß was found in VAD mouse testes (Fig. 2, a and b). However, RARß
mRNA was rapidly induced by ATRA administration, with a maximum
expression 6–8 h after ATRA injection. At these time points, the
expression of RARß was induced by a factor of approximately 4.5
compared to the expression in VAD testes. In multiple experiments, this
induction was always found, although the extent of the induction
varied, probably due to the problem of quantifying the weak signal in
VAD mice. Twenty-four hours after ATRA administration, the expression
was almost comparable to that in the VAD situation.
|
>> RAR >>
RARß. In contrast, in normal mouse testes, RARß mRNA expression was
the highest.
RXR mRNA expression in VAD mouse testis after ATRA
administration
Using RNase protection assays, expression of RXR, -ß, and
- in VAD mouse testes before and after ATRA administration was
studied. RXR and -ß were expressed at a relatively high level in
almost equal quantities. A representative autoradiograph is shown in
Fig. 3. Within 24 h after ATRA injection, no change
in their mRNA expression was observed. In normal mouse testis,
expression of RXR and -ß mRNA was also detected, at a level
similar to that found in VAD mouse testes.
|
was detected in all RXR RNase protection experiments performed.
However, it was expressed at an extremely low level in both VAD and
normal mouse testes (results not shown), and quantification of the
signal proved difficult. Because of this, by RNase protection assays we
cannot assess whether RXR mRNA expression is regulated by ATRA
administration.
Localization of RARß mRNA expression
Sertoli cells were isolated from adult VAD mice that had received
no or 0.5 mg ATRA 6 h before they were killed. Five Sertoli cell
isolations were performed. Although the yields were low using the
described cell isolation method, the purity of the isolated Sertoli
cells was high. A normal cell isolation contained more than 95%
Sertoli cells. Only twice were A spermatogonia (1–2%), characterized
by a large nucleus containing multiple nucleoli, encountered as
contaminating cells. In the other three isolations, only small round
cells (<5%) were found as contaminating cells in the Sertoli cell
fraction. Based on their morphology, these cells were presumably
peritubular cells or interstitial cells (but not Leydig cells). The
general problem of germ cell contamination in Sertoli cell isolations
did not occur when VAD mice testes were used, because the only germ
cells present in these testes were A spermatogonia. These cells were
separated from the Sertoli cells by the Percoll gradient. Using this
method, 10–15 µg RNA/six testes could be obtained from the purified
Sertoli cells, which was enough to examine the mRNA expression of
RARß by the sensitive RNase protection assay. A clear signal was
found with RNA from Sertoli cells of ATRA-treated animals. A much
weaker or no signal was found with RNA from Sertoli cells of untreated
VAD animals (Fig. 4).
|
).
|
), but administration of liarozole before
injection of 0.5 mg ATRA resulted in a labeling index significantly
lower than that when only ATRA was given. Twenty-four hours after
retinoid administration, these indexes were, respectively, 33 ±
3% and 49 ± 5% (mean ± SD; Fig. 6). At lower
ATRA concentrations, however, the injection of liarozole had no
significant effect on the labeling index of the A spermatogonia at
24 h (Fig. 6). As we previously had shown that at high retinoid
concentrations the maximum BrdU incorporation occurs sooner (8), these
results suggested that liarozole not only influenced the metabolism of
ATRA to 4-oxo-RA, but inhibited ATRA metabolism in general. To test
this hypothesis, VAD mice that had received liarozole and 0.25 mg ATRA
were killed 20 h after retinoid administration. In accordance with
our hypothesis, the labeling index in the presence of liarozole 20
h after retinoid administration was significantly (P <
0.01) higher than that 24 h after ATRA injection (49 ± 3%
vs. 31 ± 3%).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and RARß mRNA,
but not RAR mRNA, were found. In comparison, in the VAD mouse
testis, the amount of RAR mRNA was higher, and in these testes,
RAR mRNA was also detected. These data suggest that the expression
of RAR and - correlates with the presence of somatic cells and/or
A spermatogonia, as these cells are selected for in the VAD testis.
RARß mRNA expression was very low in the VAD testis. When ATRA was
administered to VAD mice there was no change in RAR and - mRNA,
but RARß mRNA expression was transiently induced by ATRA, suggesting
that this receptor is involved in mediating the early effects of ATRA
on the reinitiation of spermatogenesis in VAD mice. In addition, in
purified Sertoli cells of VAD mice, a similar induction of RARß mRNA
expression by ATRA was found. This suggests that Sertoli cells are
functional in the reinitiation of spermatogenesis in VAD mice.
Previously, the regulation of RAR expression by ATRA in
vitro has been studied in cell lines of murine origin (38, 39).
The results of these studies resemble ours in that the expression level
of RARß, compared to those of RAR and -, is influenced the most
by ATRA. All three RARs have also been detected in the adult mouse
testis in vivo (Refs. 14 and 40, and the present results),
but to our knowledge there is only one study of the regulation of these
genes in adult mice in vivo, in mouse cervical epithelia
(41). This study, like ours, showed a lower gene expression of RARß
in VAD mice compared to that in normal animals and no influence of
retinoid status on the expression of RAR and -. The direct
responsiveness of the RARß gene to ATRA may well be explained by the
retinoic acid-responsive element in the RARß2 promoter
(42, 43, 44).
In the normal mouse testis to date only RXR and -ß have been
detected, using Northern blot, in situ hybridization, or
immunohistochemical analyses (45, 46). RXRß probably plays a role in
spermatogenesis, because RXRß null mutant mice have an abnormal
spermiogenesis (46). We were also able to show a relatively strong
expression of both RXR and -ß in the normal mouse testis using
RNase protection assays. In addition, comparable expression was found
in VAD mouse testes, which did not change shortly after the
administration of ATRA. These results suggest that these RXRs are
expressed in both somatic and germ cells or that their expression in
somatic cells is up-regulated with the appearance of germ cells. This
is noteworthy, as Kastner et al. (46), in an in
situ hybridization study, found RXR transcripts to be
restricted to round spermatids in a subset of tubules. Nevertheless, we
found a strong mRNA expression of RXR in the VAD testis despite the
fact that the VAD testis does not contain spermatids. Apparently, cells
other than spermatids can express RXR too.
In the present study for the first time a low, but apparent, mRNA
expression of RXR was found in both the normal and the VAD mouse
testis. That we were able to detect RXR transcripts in the testis,
whereas other investigators (46) could not, may be due to the greater
sensitivity of RNase protection assays compared to the in
situ hybridization technique. Due to the low signal strength,
quantification of RXR was not possible. Therefore, we cannot say
whether testicular RXR expression is influenced by ATRA, although we
can exclude any substantial increase in the expression of this gene.
Regulation of RXR gene expression by retinoids has been studied
in vitro in F9 and P19 cell lines (47, 48) and in mouse
cervical epithelia by Darwiche et al. (41). In accordance
with our data, the latter showed a strong expression of RXR and -ß
in both the normal and VAD cervix, but RXR could not be detected in
cervical epithelia. Even though the amount of RXR mRNA does not seem to
change, it is still possible that RXRs influence retinoid signaling.
RAR/RXR heterodimers are considered to be the active components in RAR
signaling (25, 26, 27, 28, 29, 30), and the formation and/or activation of these dimers
may be influenced by binding of ATRA or 9-cis-RA (reviewed
in Ref.49).
Although ATRA is capable of restoring spermatogenesis in VAD mice (8), it is still not clear which retinoid is the most important in supporting spermatogenesis. The ATRA precursor, retinol, and 4-oxo-RA, a metabolite of ATRA, are both capable of reinitiating spermatogenesis, but less and more effectively than ATRA, respectively (8). To further study this issue, we examined the influence of retinol on RARß gene expression and the effect of liarozole, an inhibitor of retinoid metabolism, on the ATRA-induced proliferative activity of A spermatogonia in VAD mouse testes. Compared to ATRA, an equimolar amount of retinol was not able to induce RARß gene expression 6 h after administration to VAD mice. This is in accordance with the lower labeling index of A spermatogonia that we have found in these mice after the administration of retinol (8). Another precursor of ATRA, ß-carotene, was previously shown to induce proliferation of A spermatogonia, but with a delay of several days and at a lower level than that produced by ATRA (8). Probably, induction of the RARß gene by retinol does take place, but is delayed and less strong. Comparable results were reported by Kato et al. (18), who found that retinol administration to VAD rats caused an induction of RARß mRNA expression in several tissues, but it was delayed and less than that caused by ATRA. Van Pelt et al. (16) found induction by retinol of all three RARs in the VAD rat testis, but, again, less strongly than with ATRA.
In vivo, ATRA is quickly converted to a number of polar metabolites (among them, 4-oxo-RA) by cytochrome P450-dependent enzymes (50, 51, 52). Liarozole can quantitatively inhibit this metabolism in a dose-dependent way, thereby enhancing endogenous ATRA levels (33, 51, 52). If oxidation of ATRA to 4-oxo-RA is necessary to induce the differentiation and subsequent proliferation of A spermatogonia in the VAD mouse testis, then liarozole should cause a reduction in the labeling index of these spermatogonia at all ATRA doses used. We found that coadministration of liarozole and ATRA did lower the labeling index of A spermatogonia 24 h after injection of the highest retinoid dose used, but no significant effect of liarozole could be observed 24 h after administration of a lower ATRA dose. In contrast, liarozole did increase the percentage of BrdU-labeled A spermatogonia 20 h after the administration of this lower ATRA dose. This indicated that the maximum of BrdU labeling had shifted to an earlier time and had become higher, an effect we previously observed after the administration of increasing retinoid doses (8). Apparently, liarozole had indeed slowed down the metabolism of ATRA, resulting in an increase in the general ATRA concentration. The present results indicate that although 4-oxo-RA may be more potent in inducing the proliferation of A spermatogonia in the VAD mouse testis (8), ATRA in itself is an active retinoid as well, even when it cannot be metabolized to 4-oxo-RA. As the administration of liarozole alone had no effect on the labeling index of the A spermatogonia in VAD mice, it can be concluded that the effect of liarozole was related to the presence of ATRA. There have been reports of the inhibition of in vitro testicular androgen synthesis by liarozole (53), but such an effect has not been found in vivo (in rats) (54). Furthermore, it is difficult to explain the effects of liarozole from possible changes in testosterone levels, as these are already decreased in the VAD testis (55), and administration of testosterone to VAD rats does not restore spermatogenesis in these animals (56).
The mRNA expression of RARs and RXR has been studied in VAD rat
testes by several investigators, but their results are not consistent.
Induction of all three RARs and RXR in the VAD rat testis after
retinol or ATRA injection was found by Van Pelt et al. (16).
Kim and Griswold (15) found that upon retinol injection of VAD rats,
RAR in the testis was induced, whereas RARß expression did not
change. Similar to our results in mice, Kato et al. (18) did
not find an effect of retinol administration to VAD rats on testicular
RAR and - mRNA expression, but these researchers could not detect
RARß in the testis. The reason for the variability of the results is
not clear. Possibly, variations in the severity of the vitamin A
deficiency and/or strain differences play a role. Besides the fact that
these variable results make it difficult to compare rats with mice,
this would be hazardous because the effect of the vitamin A deficiency
on the seminiferous epithelium differs between these species. In
contrast to rats, VAD mouse testes do not contain spermatocytes, and
the numbers of A spermatogonia that remain are higher in mice than in
rats.
In conclusion, all three types of RARs and RXRs were expressed in normal and/or VAD mouse testes. However, only RARß mRNA expression was enhanced upon ATRA administration to VAD mice both in the testis as a whole and in purified Sertoli cells. The latter suggests that the activating effect of ATRA on the spermatogonia in the VAD testis is mediated by Sertoli cells. Nevertheless, a role for other RARs and RXRs cannot be excluded, as the receptors are present in the VAD testis and may be activated by the appearance of their ligand. Finally, it was found that ATRA does not need to be metabolized to 4-oxo-RA before reinitiating spermatogenesis, because blocking of the metabolic pathway to 4-oxo-RA does not abolish spermatogonial activation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 10, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, cellular retinol-binding protein, and cellular retinoic
acid-binding protein in rat testis–evidence for germ cell specific
mRNAs. Biol Reprod 44:53–61[Abstract]
and ß retinoic acid receptors and a
novel receptor predominantly expressed in skin. Nature 339:714–717[CrossRef][Medline]
, ß and
retinoic acid receptor mRNA levels in various rat tissues. Biochem J 286:755–760
in the mouse. Cell 73:643–658[CrossRef][Medline]
mutant mice. Proc Natl Acad
Sci USA 90:7225–7229, a promiscious partner of retinoic acid and thyroid hormone
receptors. EMBO J 11:1409–1418[Medline]
, ß and
mRNAs in F9 teratocarcinoma cells. Mol Cell Biol 10:391–396 gene: genomic organization and evidence for functional
isoforms. Mol Endocrinol 7:651–658This article has been cited by other articles:
 |
|
 |
|
  M. A. Glozak, Y. Li, R. Reuille, K. H. Kim, M.-N. Vo, and M. B. Rogers Trapping and Characterization of Novel Retinoid Response Elements Mol. Endocrinol., January 1, 2003; 17(1): 27 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Guo, P. L. Morris, and L. J. Gudas Follicle-Stimulating Hormone and Leukemia Inhibitory Factor Regulate Sertoli Cell Retinol Metabolism Endocrinology, March 1, 2001; 142(3): 1024 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Fu, H. Kato, K. Sugahara, and T. Kubo Retinoic Acid Accelerates the Development of Reproductive Organs and Egg Production in Japanese Quail (Coturnix coturnix japonica) Biol Reprod, December 1, 2000; 63(6): 1795 - 1800. [Abstract] [Full Text]
|
|
|
|
|
|
B. Boulogne, C. Levacher, P. Durand, and R. Habert Retinoic Acid Receptors and Retinoid X Receptors in the Rat Testis During Fetal and Postnatal Development: Immunolocalization and Implication in the Control of the Number of Gonocytes Biol Reprod, December 1, 1999; 61(6): 1548 - 1557. [Abstract] [Full Text]
|
|
|
|
|
|
I. C. Gaemers, E. Sonneveld, A. M. M. van Pelt, B. H. G. J. Schrans, A. P. N. Themmen, P. T. van der Saag, and D. G. de Rooij The Effect of 9-cis-Retinoic Acid on Proliferation and Differentiation of A Spermatogonia and Retinoid Receptor Gene Expression in the Vitamin A-Deficient Mouse Testis Endocrinology, October 1, 1998; 139(10): 4269 - 4276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Chinpaisal, C.-H. Lee, and L.-N. Wei Mechanisms of the Mouse Orphan Nuclear Receptor TR2-11-mediated Gene Suppression J. Biol. Chem., July 17, 1998; 273(29): 18077 - 18085. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
