| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
UMR Centre National de la Recherche Scientifique 6547, Reproduction et Développement, Université Blaise Pascal Clermont II, Complexe Universitaire des Cézeaux, and Unité de Transfert de gènes dans les cellules Animals et Humaines, Complexe Universitaire des Cézeaux (S.G.), 63177 Aubiere Cedex; and INSERM U-129, Physiologie et Pathologie Génétiques et Moléculaires, Institut Cochin de Génétique Moléculaire, Université René Descartes (A.K.), 75014 Paris, France
Address all correspondence and requests for reprints to: Dr. Antoine Martinez, Reproduction et Développement, Université Blaise Pascal Clermont II, Complexe Universitaire des Cézeaux, 24 avenue des Landais, 63177 Aubiere Cedex, France. E-mail: martinez{at}cicsun.univ-bpclermont.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We show that at least two regions ensure proper gene regulation in vivo. The first, located within the 1.8-kb promoter fragment, directs tissue specificity; positive elements necessary for vas deferens and adrenal expression lay within positions -1804 to -510 and -510 to +41, respectively. The second, located within the 3.5-kb intragenic fragment spanning intron 1 to intron 2, increases percentage of expressing lines and behaves as a vas deferens-specific enhancer. Hormonal and developmental control of transgenes closely parallel endogenous gene regulation. Androgen and ACTH responsiveness in adrenals is conferred by 0.5-kb promoter, whereas in vas deferens, full androgenic response of the 1.8-kb promoter required the 3.5-kb intragenic fragment. Thus, vas deferens and adrenals use distinct cis-acting elements to direct and regulate the expression of the MVDP gene.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies from our laboratory focused on the cis- and trans-acting phenomena controlling the androgenic response of the MVDP gene in vitro. Transient transfections demonstrated that a 1.8-kb promoter region (nucleotides -1804 to +41) encompassing two putative androgen-responsive elements (AREs) at positions -1171 and -97, respectively, was capable of mediating androgen responsiveness in T47D mammary tumor cells (8). In these experiments, a minimal -121/+41 region was identified as being sufficient to ensure androgen regulation via interaction of the hormonal receptor to the proximal ARE (-97). The full androgen responsiveness occurred in vitro through a composite responsive unit requiring the interplay of the proximal ARE and adjacent binding sites for ubiquitous NF1 and Sp1 factors (9). Because vas deferens epithelial cells in subculture were resistant to transfection, these previous experiments had been carried out with androgen-responsive cell lines that did not express endogenous MVDP. Thus, transfection analyses cannot take into consideration either the tissue-specific aspects or the developmental regulation of the MVDP gene androgen responsiveness occurring under physiological conditions. Analysis of introduced genes in transgenic mice provides a means by which the total aspects of gene regulation may be analyzed.
Functional analysis of transcriptional mechanisms leading to the androgen responsiveness of genes expressed in the genital tract has not been investigated extensively in vivo. Indeed, previous transgenic studies had focused on the characterization of regulatory sequences ensuring hormonal and tissue-specific regulation of androgen-responsive genes expressed in the kidney, i.e. the Adh-1 gene (10), the KAP gene (11), and the Gus-s gene (12). Paradoxically, although the male genital tract is a major target of androgen action, transgenic mouse studies concern only prostate-specific genes, i.e. the rat C3 (1) gene (13, 14), the rat probasin gene (15, 16), and the human prostate-specific antigen (17). Finally, understanding of the mechanisms controlling adrenal cortex-specific expression has been hampered due to a lack of in vivo efficient adrenal-specific transgenes (18).
To gain an understanding of the cis-regulatory regions responsible for hormonal control of the MVDP gene in the vas deferens and the adrenals in vivo, we have generated transgenic mice harboring different MVDP-chloramphenicol acetyltransferase (MVDP-CAT) fusion genes. We show herein that 1) distinct 5'-flanking sequences of the MVDP gene are necessary to target vas deferens and adrenal expression of the CAT reporter; 2) intragenic regions cooperate with these upstream regions to enhance the probability of transgene expression; and 3) hormonal and developmental control of the transgenes reveals that different cis-acting regions ensure proper hormonal regulation of the endogenous gene in the two major MVDP-expressing tissues.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The 1.8M-CAT and
0.5M-CAT transgenes were previously described (8) and contained,
respectively, the -1804/+41 and -510/+41 upstream sequences of the
MVDP gene fused to a reporter construct composed of the CAT-coding
sequence and the simian virus 40 (SV40) small t intron and early
polyadenylation signal sequences. A 3.5-kb PvuII fragment
covering most of the first intron, all of the second exon, and most of
the second intron of the MVDP gene was excised from the 
17a phage
clone (5). The resulting genomic fragment was cloned, in an inverted
position, into the SmaI polylinker site located downstream
of the SV40 polyadenylation signal of the MVDP-CAT constructs described
above to create the 1.8M-CAT-Int and 0.5M-CAT-Int transgenes. Note that
PvuII digestion removes the splicing donor site and the
acceptor site of introns 1 and 2, respectively (5).
|
)
containing the chimeric genes were purified from preparative agarose
gel electrophoresis as previously described (20) and microinjected into
fertilized B6D2 mouse eggs according to the method of Hogan et
al. (21). Transgenic founders and offsprings were identified by
Southern blot analysis after BamHI digestion of tails DNA
from 10-day-old mice. Southern blots were hybridized with a CAT/SV40
probe (a 948-bp NcoI/HpaI fragment covering half
of the CAT-coding sequence and most of the SV40 sequences). Lines were
established by outbreeding founder F0 mice to generate
heterozygous mice. All subsequent studies were performed on
F1 or F2 mice, except for developmental studies
in which F3 mice were used. These F3 mice have
received the transgene by outbreeding F2 homozygous mice
with nontransgenic B6D2 mice.
Analysis of transgenes and endogenous gene expression
Frozen individual tissue samples were homogenized in 100–300
µl 250 mM Tris-HCl (pH 7.5) and 0.4 mM
phenylmethylsulfonylfluoride using a glass-glass tissue homogenizer and
centrifuged in a microfuge at 13,000 rpm and 4 C for 15 min. For CAT
assays, the supernatants were incubated at 65 C for 10 min and then
centrifuged at 13,000 rpm and 4 C for 15 min. For Western blot analysis
of MVDP contents, a 25-µl aliquot was collected before heat
treatment. The concentration of soluble proteins in heat-treated and
untreated supernatants was determined by the Bradford method
(Bio-Rad Laboratories, Inc., Richmond, CA). Proteins from
heat treated samples (5 or 20 µg) were incubated for 3 h at 37 C
in 200 µl 250 mM Tris-HCl (pH 7.5), 0.4 mM
acetyl coenzyme A, 0.12 µCi [14C]chloramphenicol
(2–2.4 nmol). Five micrograms (line 70) and 20 µg (other lines)
corresponded to the protein quantity necessary to obtain a fraction of
acetylated chloramphenicol that was linear with respect to enzyme
activity. The products were separated by TLC. Acetylated forms of
[14C]chloramphenicol were excised individually, and
radioactivity was counted in a scintillation counter. Specific
activities were determined as disintegrations per min/µmol
substrate/min reaction time and mg/protein.
Western blot analyses of MVDP tissue contents were performed with 10 or 20 µg proteins separated by SDS-10% (wt/vol) PAGE and transferred to nitrocellulose membranes in Bio-Rad Laboratories apparatus according to the manufacturer’s instructions. After blocking of nonspecific protein-binding sites for 1 h at room temperature in TBST [50 mM Tris-HCl (pH 8), 150 mM NaCl, and 0.1% (vol/vol) Tween-20] containing 5% (wt/vol) nonfat dry milk, the blots were incubated with primary monoclonal antibody (ascite B263) and anti-MVDP (22) at a 1:700 dilution for 1 h at room temperature. After three washes of 15 min each in TBST, secondary antibody (peroxidase-conjugated antimouse) was added at a 1:15,000 dilution and left for 1 h at room temperature. The blots were washed again as described above, and peroxidase activity was detected by autoradiography with the enhanced chemiluminescence (ECL) system (Amersham, Arlington Heights, IL).
Hormonal regulation of transgene expression
F1 or F2 transgenic mice
(8–12-week-old) were castrated and killed 20 days later. Intact males
of the same age served as controls. Where appropriate, intact males
were injected with corticosterone (75 µg twice daily; Sigma Chemical Co.), 20-day castrated males were supplemented with
heptylate testosterone or progesterone, and intact females
were injected with heptylate testosterone (75 µg twice daily;
Theramex Laboratories, Monaco). All steroids were diluted in sesame oil
as vehicle and administered for 7 days (9 days for testosterone-treated
females), and animals were killed 6 h after the last injection.
Control mice received injection of vehicle only. Tissues were removed,
frozen in liquid nitrogen, and tested for CAT activity and endogenous
gene expression by Western blot as described above.
Immunocytochemistry
Transgenic (8- to 12-week-old) mice were killed and tissues
immediately removed and fixed. Vasa deferentia and adrenals were fixed
in 75% ethanol and 25% acetic acid for 30 min, dehydrated in absolute
ethanol, embedded in paraffin, and sectioned at 6 µm. After dewaxing,
tissue sections were exposed to primary antibodies diluted in PBS
containing 5% (wt/vol) nonfat dry milk [1:50 dilution for anti-CAT
antibody purchased from 5 Prime, 3 Prime, Inc. (Paoli, PA) or 1:80
dilution of anti-MVDP polyclonal antibody (23)] and incubated
overnight at 4 C. Sections were then rinsed three times in PBS and
treated for 2 h at room temperature with antirabbit-peroxidase
antibody (1:400 dilution; Boehringer Mannheim, Meylan, France).
Peroxidase activity was revealed using standard immunoperoxidase
procedures, as previously described (24). Control slides were obtained
by omission of primary antibody and using the complete technique on
wild-type tissues.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). These constructs belong to two
groups depending on whether they contain intronic sequences of the MVDP
gene. The first group corresponds to constructs containing -1804/+41
or -510/+41 MVDP 5'-flanking sequences linked to the CAT/SV40 reporter
gene (1.8M-CAT and 0.5M-CAT). In the second group, in addition to these
promoter sequences, an intragenic 3.5-kb fragment bearing the MVDP
first and second introns and the second exon was linked immediately
downstream the reporter as described in Materials and
Methods (1.8M-CAT-Int and 0.5M-CAT-Int). The constructs of the first group were chosen for our initial investigations because they both have been shown to be responsive to androgens via ARE in previous transfection studies (8). In addition to information concerning the influence of upstream sequences, the constructs of the second group allowed us to test a possible enhancer activity harbored by MVDP intragenic sequences.
The 4 constructs were microinjected in fertilized eggs to produce
founders of independent transgenic lines (Table 1). Thirty-one founder transgenic mice
were obtained. The number of transgenes integrated into the DNA of each
founder varied from 2–50 copies/cell. All founders except 2 (1 for
0.5M-CAT-Int and 1 for 1.8M-CAT-Int) transmitted the transgene to their
offsprings in a Mendelian fashion. CAT activity was determined in
tissue extracts from all of the male founder mice or male offsprings of
the female founders. No expression of the transgene was detected in all
5 lines bearing the 0.5M-CAT construct. The 1.8M-CAT transgene was
expressed in only 2 of 10 lines. However, when adding MVDP intronic
sequences, transgene expression was detected in 3 of the 4 lines
bearing the 0.5M-CAT-Int construct and in 9 of the 12 lines bearing the
1.8M-CAT-Int construct.
|
). As shown by the typical Western blot
experiment presented in Fig. 2A, MVDP was abundantly detected in the
vas deferens and in lower amounts in adrenal glands from both sexes.
Note that male and female adrenals expressed MVDP in similar
amounts.
|
. It showed
that, at least in male mice, this construct reproduced almost perfectly
the pattern of expression of the endogenous MVDP gene. All of the nine
lines positive for the transgene activity expressed CAT according to
this pattern (data not shown). The CAT transgene was actively expressed
in the vas deferens and to a lesser extent in the adrenals, but no
expression was detected in other parts of the genital tract or in the
liver. However, significant ectopic expression of CAT was detected in
kidney and lung from both sexes (Table 2). Several tissues that normally do not
express MVDP (heart, muscle, eye, brain, spleen, and intestine) were
also assayed for CAT expression. No expression of CAT was detected in
these organs (data not shown).
|
, the
1.8M-CAT transgene lacking MVDP intronic sequences exhibited a similar
expression pattern as 1.8M-CAT-Int (the same pattern was found in a
second line at a markedly reduced level; Table 2). Targeted expression
of the reporter to the vas deferens and adrenals was maintained, and
except for ectopic expression in kidney and lung, no significant CAT
activity was detected in all other tissues assayed. However, two major
differences can be emphasized: 1) the CAT activity detected in the vas
deferens was 6- to 60-fold lower than that in 1.8M-CAT-Int transgenic
lines (Table 2); and 2) the penetrance of the phenotype was severely
reduced (Table 1; 2 of 10 compared with 9 of 12).
Tissue-specific expression was similar in all of the expressing lines
harboring either the 1.8M-CAT-Int or 1.8M-CAT construct, indicating
that the sequence -1804 to +41 contains the necessary information to
direct vas deferens and adrenal expression of MVDP gene in male mice.
However, for the two constructs, the adrenal expression of the
transgene was always barely detected in females (Table 2). This result
contrasts with the similar expression of endogenous MVDP in the
adrenals from both sexes, suggesting that important
cis-acting elements for female expression were absent from
the transgenes or were not active.
0.5M-CAT-Int transgene activity. The 0.5M-CAT-Int transgene,
lacking sequence -1804 to -510, was barely expressed in the vas
deferens, but showed an enhanced activity in adrenals (Fig. 2C and
Table 2). These results indicate that the deleted fragment is essential
for vas deferens specificity of the MVDP gene. As for the other
MVDP-CAT constructs, ectopic CAT expression was also observed in kidney
and lung from both sexes. Again, no CAT activity was detected in all
other tissues assayed. A similar pattern of expression was observed for
all 0.5M-CAT-Int-expressing lines, indicating that the sequence -510
to +41 contains the necessary information to direct adrenal expression
of MVDP gene in male mice. As for longer constructs, the adrenal
expression of the 0.5M-CAT-Int transgene was always poorly detected in
females, suggesting that important cis-acting elements for
female expression were also absent from this construct or were not
active.
The penetrance of the CAT-expressing phenotypes was greatly enhanced by
the addition of the 3.5-kb intronic segment of MVDP gene (reaching
about 75% penetrance; Table 1), suggesting that elements present in
the first intron, second exon, or second intron were able to cooperate
with either -510/+41 or -1804/+41 upstream sequences to enhance the
probability of transgene expression. However, the 3.5-kb fragment is
not sufficient to confer an expression independent of the integration
site, as 100% penetrance was not observed. In addition, CAT activity
was not related to the number of integrated copies (Table 2). Together,
our results indicate that the 1.8M-CAT-Int transgene contains necessary
information to direct vas deferens and adrenal expression of the MVDP
gene: the -1804 to -510 region ensures vas deferens specificity, the
-510 to +41 region is necessary to adrenal expression, and the
intragenic fragment enhances the probability of transgene expression.
However, additional cis-acting elements are required for
full activation of the transgene in a copy number-dependent/integration
site-independent manner.
Cell-specific expression of MVDP-CAT transgenes
Immunocytochemistry was used to address the cellular localization
of CAT within the vas deferens and adrenals of 8- to 12-week-old
transgenic mice (Fig. 3). As shown in
Fig. 3A, section of the vas deferens from the 1.8M-CAT-Int transgenic
line 70 exhibited positive staining strictly limited to the cytoplasm
of all epithelial cells and was absent from the muscular wall. No
positive staining was detected in either the epithelium or the muscular
wall in the control section obtained from the 0.5M-CAT-Int transgenic
line 62, which did not efficiently express CAT in the vas deferens
(Fig. 3C). This pattern of epithelial cell expression coincided with
that of the endogenous MVDP and was consistent with previous data (22)
(Fig. 3A).
|
), and no differences were detected between males and females (data
not shown). Note that no staining was present in the zona glomerulosa
or the medulla. No positive CAT staining was detected in a section of
male adrenal gland from the 0.5M-CAT-Int transgenic line 62, probably
due to an insufficient level of CAT expression in that line (Fig. 3B)
and to the low sensitivity of the CAT antigen immunodetection technique
(20). Indeed, CAT staining was also negative in vas deferens from
1.8M-CAT-Int line 28, which exhibited a CAT-specific activity in this
organ similar to that observed in adrenal extracts from line 62 (data
not shown and Table 2). We conclude from these observations that the
1.8M-CAT-Int transgene displays the cis-acting elements
needed to direct expression of MVDP gene in the epithelial cells of vas
deferens.
Hormonal regulation of the MVDP-CAT transgenes
To determine whether the 1.8M-CAT-Int constructs are regulated by
androgens in vivo, sexually mature transgenic male mice were
castrated, and CAT activity was determined in the MVDP-expressing
organs (vas deferens and adrenals) and a negative organ (kidney) after
castration alone or androgen replacement. As shown in Fig. 4A (left side), transgene
activity decreased dramatically in the vas deferens and adrenals and
returned partially to precastration levels after testosterone
administration. As exemplified by Western blot analysis, endogenous
MVDP and the transgene followed the same androgenic control in vas
deferens and adrenal glands. In addition, this androgenic control was
tissue specific, because ectopic expression of CAT in kidney was
unaffected by androgenic status. To determine whether the low level of
CAT activity in adrenal glands from females may be due to the lack of
androgens, adult female transgenic mice were treated for 9 days with
testosterone, and CAT activity was assayed in either the adrenals or
kidneys. As shown in Fig. 4A (right side), neither transgene
expression nor endogenous MVDP accumulation could be induced in the
female adrenals by an extensive androgenic treatment (9 days).
|
, transgene regulation behaved
differently in the two MVDP-expressing tissues. Indeed, as observed
above for longer constructs, the 1.8M-CAT transgene was responsive to
castration and androgen replacement in the adrenals. However, in the
vas deferens, transgene activity was not affected by castration and
exhibited scarcely a 3-fold stimulation after testosterone treatment.
Again, ectopic expression of CAT in kidney was unaffected by androgenic
status, and the regulation of endogenous MVDP, assayed by Western blot
in the same samples, displayed an androgen-dependent pattern in the vas
deferens and adrenals. Taken together, these observations allow us to conclude that the sequences involved in the androgenic control of the MVDP gene in the vas deferens and adrenals are present in the 1.8M-CAT-Int construct. However, the cis-acting elements required could be different in part in the two tissues. Indeed, although the presence of the MVDP intronic segment was clearly dispensable to impart androgenic control of the transgene in the adrenals, the effects of castration were abolished and the response to androgen replacement was impaired in the ductus, suggesting the involvement of intronic cis-acting elements for regulated androgenic response in the vas deferens.
To understand the hormonal requirements of the
adrenal-specific 0.5M-CAT-Int transgene, sexually mature males were
castrated, and CAT activity was measured in MVDP-expressing tissues
(vas deferens and adrenals) and negative tissue (lung) after castration
alone and after androgen replacement. In addition, some castrated mice
were treated with progesterone, a known intermediate
product of adrenal steroidogenesis, to test the hormonal specificity of
the gene and transgene responses. As ACTH is an essential regulator of
gene expression in the adrenal cortex (25), male mice were treated with
corticosterone to ensure the negative feedback of ACTH production. As
shown in Fig. 4C, neither CAT ectopic expression in lung nor basal CAT
activity in ductus was affected by any treatment. After castration, the
transgene activity as the MVDP accumulation decreased dramatically in
the adrenals and returned partially to precastration levels after
testosterone administration. Note that progesterone had
very little effect on either transgene or endogenous gene induction in
these castrated mice. The chronic administration of corticosterone gave
rise to a 3-fold inhibition of transgene activity in the adrenals and
nearly completely abolished immunodetected endogenous MVDP. This
response was tissue specific, because endogenous MVDP levels in the vas
deferens were unaffected by glucocorticoid treatment.
We conclude that the sequences involved in the androgenic and ACTH-dependent control of the MVDP gene in the adrenals are present in the 0.5M-CAT-Int construct. Because the distal ARE (nucleotides -1186 to -1171) is absent from the 0.5M-CAT-Int construct, we conclude that this element was not required for androgenic regulation of the MVDP gene in the adrenals.
Developmental regulation of 1.8M-CAT-Int transgene
The expression of the MVDP gene has been shown to be a marker for
vas deferens differentiation in the developing mouse (2) and for
epithelial cell differentiation in subcultures (3, 26). However, until
now, no information concerning the developmental pattern of MVDP in
adrenals has been assessed. To determine the pattern of 1.8M-CAT-Int
transgene expression during postnatal development, extracts were
prepared from the vasa deferentia and adrenals (from both sexes) of
mice between 6–90 days of age. Developmental patterns of CAT activity
were compared with those of the endogenous MVDP gene (estimated by
Western blot) in the two tissues. As shown in Fig. 5A, in the vas deferens, the transgene
and MVDP showed similar developmental patterns of expression; both were
barely expressed at 6 days, easily detected yet at low levels between
10–20 days, and increased sharply between 20–60 days, a time
corresponding to the progressive establishment of sexual maturity.
|
, adrenal expression of the MVDP
gene was high at 6 days of age, was abruptly reduced between 6–10
days, remained low until 15 days, and increased thereafter until
adulthood (60–90 days). It is important to note that postnatal
expression of the MVDP gene in adrenals followed the same pattern in
males and females. CAT activity followed a very similar pattern, but
only in males. Indeed, male and female CAT expression patterns were
identical until 20–30 days, after which transgene expression in the
female remained low. We conclude from these findings that the 1.8M-CAT-Int transgene contains the sequences necessary for postnatal regulation of the MVDP gene in the vas deferens and male adrenals. However, these data also suggest that in females, the transgene lacks information necessary for the induction of the MVDP gene occurring in the adrenals after 15 days of age.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our reports demonstrate that the 1.8-kb MVDP promoter (-1804/+41 bp)
contains enough information to direct vas deferens- and
adrenal-specific expression of the transgene. However, additional
cis-acting elements, present within the 3.5-kb intragenic
fragment spanning intron 1 to intron 2, are required together with the
1.8- or 0.5-kb MVDP promoters for increasing the frequency and/or the
level of transgene activity in the vas deferens and adrenals,
respectively. Because the 3.5-kb intragenic fragment lacks splicing
sites and is linked downstream of the polyadenylation signal of the
reporter gene, it can only act as a cis-acting enhancer that
contains control elements capable of partially overcoming some
integration site-effects and also to stimulate promoter activity in the
vas deferens. The low frequency of transgene activity in the mice
transgenic for 1.8M-CAT or 0.5M-CAT constructs suggests that these
promoter sequences are unable to protect transgene from the influence
of neighbor regulatory elements when the 3.5-kb intronic region is
absent. However, in some rare cases (2 of 10 for 1.8M-CAT mice), the
transgene was inserted into a permissive chromatin region where the
tissue-specific cis-acting elements laying within the
promoter can be recruited. However, even in the presence of the 3.5-kb
intronic element, expression of the transgene is not observed in 100%
of the lines and is not copy number dependent, indicating that it
contains neither the locus control region nor the so-called insulators
(27, 28). Although further analyses are required, it is tempting to
propose that the large first intron (2.8 kb) would have a higher
probability of carrying the enhancer activity. Such a cooperation
between proximal promoters and first intron enhancers to ensure a high
level of transgene expression and/or fewer position effects has been
described for other genes, e.g. the 
2(XI) collagen gene
(29), the aldolase B gene (20), and the RIß subunit gene of protein
kinase of cAMP-dependent protein kinase (30).
Our data evidence that some of the cis-acting elements involved in vas deferens and adrenal expression are distinct. Deletion up to -510 nearly completely abolishes expression in the vas deferens, indicating that putative vas deferens-specific cis-acting elements of the MVDP gene are located between -1804 and -510 bp. However, in that case, a strong expression is observed in the adrenals, suggesting that the sequence downstream from position -510 is sufficient for specific expression in the adrenals. Interestingly, the -1804/-510 region contains five potential recognition sites for the PEA3/Ets1 factor (positions -691 to -598) known to be enriched in some parts of the male genital tract (31). In addition, the -510/+41 sequence encloses a putative binding site (position -453) for steroidogenic factor 1, which is a key regulator of steroid hydroxylases genes expressed in the adrenal cortex (32). As the endogenous gene, the 1.8M-CAT-Int transgene expression is exclusively targeted to the epithelial cells of the vas deferens. To our knowledge, the 1.8-kb MVDP promoter (together with the 3.5-kb intronic region) is the first promoter construct able to direct vas deferens epithelium-specific expression within the genital tract of transgenic mice. We show that MVDP expression in the adrenals is restricted to the zona fasciculata. Previously, regulatory regions of three steroid hydroxylase genes specifically expressed in the adrenal cortex, i.e. 21-hydroxylase, 11ß-hydroxylase, and aldosterone synthase, have been studied with respect to their potential application in the development of adrenocortical-specific transgenes (18). Only the 6.4-kb 21-hydroxylase promoter linked to the ß-galactosidase reporter was able to direct transgene activity in the adrenals. In this regard, the 0.5-kb MVDP promoter (together with the 3.5-kb intronic region) that has demonstrated ability to target expression of the heterologous gene to the adrenals provides a new adrenal-specific promoter model suitable for transgenic approaches.
Except for the abnormal targeting in kidney and lung, the 1.8M-CAT-Int and 0.5M-CAT-Int transgenes were only intensively expressed in the vas deferens and adrenals, respectively, mimicking the endogenous MVDP expression. The ectopic expression of the transgenes is not due to position artifacts, because this same pattern was found in MVDP-CAT transgenic lines that differ in both copy number and integration site. This could, instead, reflect a common lineage of the origin of these tissues. The vas deferens and kidney differentiate from the intermediate mesoderm, and the adrenals develop from the splanchnic mesoderm. Therefore, the transgene ectopic expression could reflect the common mesodermal origin of these organs. However, the transgene expression in lung raises the question of whether this tissue of endodermal origin also shares a common route in development.
In male mice, the developmental and hormonal control of the
1.8M-CAT-Int transgene closely follows that of the endogenous gene,
indicating that the transgene contains the sequences needed for both
vas deferens and adrenal postnatal regulation of the MVDP gene.
However, our data suggest that the hormonal requirements may be partly
different in these two organs. Indeed, in the vas deferens, the
developmental expression of both transgene and endogenous gene follows
a biphasic pattern, consisting of a low expression in immature animals
followed by a rise that parallels sexual maturation. This observation
is consistent with previous data demonstrating that although androgens
are the essential activator of the gene at adulthood, the first
expression of MVDP (6–10 days) is independent of androgens and that
the induction taking place from 20 days of age parallels the increasing
serum androgen levels seen during sexual maturation (2). In agreement
with the known transcriptional regulation of the MVDP gene by androgens
(7), hormonal manipulations consisting of castration and testosterone
replacement of mature males clearly demonstrate that the 1.8M-CAT-Int
transgene is also responsive to androgens in the vas deferens. In
adrenals, the developmental expression of both MVDP gene and the
1.8M-CAT-Int transgene follows a triphasic pattern: first, both are
highly expressed in 6-day-old mice, then decrease sharply until day 15,
and 6-day-old like levels are progressively restored in maturing
animals. These changes in MVDP expression correlate well with the
changes in the steroidogenic activity of the adrenal cortex during
development in the rat (reviewed in Ref. 33). Indeed, corticosterone
biosynthesis and plasma ACTH concentrations are high during the first
days after birth, then decrease during the first 15 days of life and
increase thereafter to reach adult values. A comparable pattern is
observed for known ACTH-regulated enzymes involved in the biosynthesis
of corticosterone, the P450 side-chain cholesterol cleavage enzyme, and
the 3ß-hydroxysteroid
dehydrogenase-
5-4-isomerase (34, 35). The
decrease in both MVDP gene and transgene expression after negative
feedback of the hypothalamic-pituitary-adrenal axis, is consistent with
the observed triphasic developmental pattern and suggests strongly that
ACTH exerts a positive control on MVDP gene transcription in the
adrenals. In agreement with these in vivo data, we have
recently demonstrated that cAMP, a known cellular mediator of ACTH
action, was able to stimulate MVDP mRNA levels in adrenocortical cells
cultures (36). Furthermore, our experiments provide the first evidence
that androgens are also able to control MVDP gene transcription in the
adrenals. Indeed, both transgenes and endogenous gene expression are
strongly reduced after castration, and precastration levels are
restored by testosterone replacement. The potential role of sexual
steroids on adrenal cortex function has been previously suggested by
the presence of high levels of androgen (37, 38) and estrogen receptors
(39, 40, 41) in rodents as well as in primates (42). More recently,
evidence from in vivo studies showed that both testosterone
and estradiol increase the responsiveness of the adrenal cortex to ACTH
stimulation in humans and rats, respectively (43, 44). However, until
now the characterization of target genes for sexual steroid action in
the adrenal glands remains poorly documented. Interestingly,
diazepam-binding inhibitor/acyl-coenzyme A-binding protein
a protein involved in mitochondrial steroidogenesis by stimulating
cholesterol delivery to the inner membrane (45), was also shown to be
under androgenic control in the rat adrenals (46).
The presence of similar amounts of MVDP in adrenals from both male and female adult mice together with the absence of sexual dimorphism in the developmental expression patterns raise the question of the role of sexual steroids in the control of the MVDP gene in females. Androgens of adrenal origin are not involved, because rodent adrenals lack P450C17 activity and so do not synthesize androgens (reviewed in Ref. 47). By contrast with the endogenous gene, all of the transgenic constructs active in male adrenals are found to be inactive in the adult female. This behavior originates in the postnatal evolution of transgene activity, which differs in male and female mice; both patterns overlap until 20–30 days, but the rise in transgene activity occurring in maturing males does not take place in maturing females. Because the transgene activity cannot be stimulated by androgenic treatment of the transgenic females, the low expression of the constructs cannot be ascribed to the lack of androgens in the females. Together, these results suggest that 1) male and female share common cis-acting elements to ensure expression of the gene in young animals and that 2) positive response of the gene to maturing signals required distinct, sex-specific, cis-acting elements (1.8M-CAT-Int transgene bearing only male-specific elements). We propose that in the case of male mice, this maturing signal might be the increasing plasma androgens levels seen during puberty. In the case of female mice, it is tempting to hypothesize that ovarian hormones and particularly estrogens, which act through responsive elements distinct from those of androgens, could constitute this maturing signal.
The transgenic approach gave us the opportunity to consider together tissue-specific and hormonal aspects of MVDP gene transcription. In vivo, androgen-regulated transcription of the MVDP gene in the vas deferens requires, on the one hand, cis-acting elements located within the 1.8-kb promoter and, on the other hand, elements laying within the 3.5-kb intronic fragment (1.8M-CAT-Int), whereas in the adrenals the upstream elements are clearly sufficient to ensure androgen responsiveness (1.8M-CAT). Previously, the MVDP 1.8-kb promoter was shown to contain two potential AREs, a distal ARE (nucleotides -1186 to -1171) and a proximal ARE (nucleotides -111 to -96). Transfection studies established that a minimal -121/+41 promoter region was sufficient to confer androgen-regulated expression via the proximal ARE (8). Although we cannot infer from our transgenic experiments whether the distal ARE is required in vivo, the presence of the two AREs within the 1.8M-CAT construct is not sufficient to confer hormonal responsiveness in the vas deferens, whereas the distal ARE is clearly dispensable for androgenic regulation of the MVDP gene in the adrenals (0.5M-CAT-Int). Interestingly, kidney expression and androgen response of the ß-glucuronidase structural gene (Gus-s) were shown to occur in transgenic mice through the involvement of two regions; the first, located within the 3.8-kb promoter sequence, supported kidney expression but ensured a low frequency of androgen responsiveness, and the second, located within a 6.4-kb intragenic fragment, exerted a permissive action by protecting the androgen responsiveness of the transgene from repressive influences of the insertion site (12). Similarly, the observation that androgen responsiveness occurs in the vas deferens only when the 3.5-kb intronic fragment is included in the transgene (1.8M-CAT-Int) suggests that these additional sequences are capable of partially overcoming some integration site-effects as discussed previously, and also behave as a vas deferens-specific androgen-dependent enhancer. Although previous transfection studies indicated that the androgen receptor can mediate induction through the direct binding to responsive elements, the possibility should be considered that some of the effects of androgens in vivo might be indirectly mediated by androgen-induced changes in other factors. Thus, direct and indirect mechanisms might coexist. In vas deferens cell subcultures, androgen stimulation of the MVDP gene was abolished by cycloheximide, suggesting that androgen action depends on continuous protein synthesis (26). Intragenic regulatory elements now have to be characterized with respect to their ability to induce, in cooperation with upstream AREs, a specific androgen response in the vas deferens.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 27, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2(XI) collagen gene expression in the
embryonic mouse cartilage. J Cell Biol 134:1573–15825–3ß-hydroxysteroid
deshydrogenase in rat adrenals in the early postnatal period. Arch Int
Physiol Biochem 76:533–537[Medline]
-hydroxylase and 3ß-hydroxysteroid dehydrogenase in the
integration of gonadal and adrenal steroidogenesis via the
5 and 4 pathways of steroidogenesis in
mammals. Biol Reprod 56:789–799[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  B. Ragazzon, A.-M. Lefrancois-Martinez, P. Val, I. Sahut-Barnola, C. Tournaire, C. Chambon, J.-L. Gachancard-Bouya, R.-J. Begue, G. Veyssiere, and A. Martinez Adrenocorticotropin-Dependent Changes in SF-1/DAX-1 Ratio Influence Steroidogenic Genes Expression in a Novel Model of Glucocorticoid-Producing Adrenocortical Cell Lines Derived from Targeted Tumorigenesis Endocrinology, April 1, 2006; 147(4): 1805 - 1818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-M. Lefrancois-Martinez, J. Bertherat, P. Val, C. Tournaire, N. Gallo-Payet, D. Hyndman, G. Veyssiere, X. Bertagna, C. Jean, and A. Martinez Decreased Expression of Cyclic Adenosine Monophosphate-Regulated Aldose Reductase (AKR1B1) Is Associated with Malignancy in Human Sporadic Adrenocortical Tumors J. Clin. Endocrinol. Metab., June 1, 2004; 89(6): 3010 - 3019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Val, C. Aigueperse, B. Ragazzon, G. Veyssiere, A.-M. Lefrancois-Martinez, and A. Martinez Adrenocorticotropin/3',5'-Cyclic AMP-Mediated Transcription of the Scavenger akr1-b7 Gene in Adrenocortical Cells Is Dependent on Three Functionally Distinct Steroidogenic Factor-1-Responsive Elements Endocrinology, February 1, 2004; 145(2): 508 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Martinez, P. Val, I. Sahut-Barnola, C. Aigueperse, G. Veyssiere, and A.-M. Lefrancois-Martinez Steroidogenic Factor-1 Controls the Aldose Reductase akr1b7 Gene Promoter in Transgenic Mice through an Atypical Binding Site Endocrinology, May 1, 2003; 144(5): 2111 - 2120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Lane Diabetic kidney disease: impact of puberty Am J Physiol Renal Physiol, October 1, 2002; 283(4): F589 - F600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Val, A. Martinez, I. Sahut-Barnola, C. Jean, G. Veyssiere, and A.-M. Lefrancois-Martinez A 77-Base Pair LINE-Like Sequence Elicits Androgen-Dependent mvdp/akr1-b7 Expression in Mouse Vas Deferens, But Is Dispensable for Adrenal Expression in Rats Endocrinology, September 1, 2002; 143(9): 3435 - 3448. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
