| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
U-368, INSERM (P.T., S.S.-M., G.L., P.C.) Biologie
Moléculaire du Développement Ecole Normale
Supérieure 75230 Paris, France
Laboratory of
Molecular Biology (G.L.) Advanced Biotechnology Center-IST
Genova, Italy
URA CNRS 1414 (A.T.) Ecole Normale
Supérieure 75230 Paris, France
U-159, INSERM
(D.G.) Centre Paul Broca 75014 Paris, France
Institut
National de la Recherche Agronomique Physiologie de la Reproduction
des Mammifères Domestiques (M.-A.D.) 37380 Nouzilly,
France
Laboratory of Molecular Reproductive Biology and
Medicine Department of Obstetrics and Gynecology (C.V.R.)
University of Louisville Health Science Center Louisville, Kentucky
40292
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-glycoprotein hormone subunit (GSU) in association with
specific ß-subunits. Proliferation of the different cell types and
their secretory activities are subject to multiple controls, including
neurotransmitters and neuropeptides from the hypothalamus and feedback
regulations from target organs. Major advances have recently been made
in elucidating the cascades of genes involved in the regulation of
pituitary ontogeny (1, 2). In particular, the transcription factor gene
Pit-1, which is disrupted in the mouse Snell and Jackson
dwarf mutations, has been shown to be required for the specification of
three lineages: somatotropes, lactotropes, and thyrotropes (3, 4).
Interestingly, this transcription factor is necessary both for the
initial activation of the expression of the corresponding hormones and
for the proliferation and survival of the three cell types (5, 6, 7). More
recently, analysis of another murine pituitary-dependent dwarfism (Ames
dwarf) has led to the identification of a novel transcription factor
gene, Prophet of Pit-1. Its mutation prevents
Pit-1 activation and the initial determination of Pit-1 cell
lineage and, therefore, also results in failure to produce GH, PRL, and
TSH (8). The Lim homeobox gene Lhx3, which is expressed in
the pituitary throughout development, is likely to act upstream of this
transcription factor cascade, since its mutation affects the
determination of all pituitary cell lineages except the corticotropes
(9). Finally, the nuclear receptor steroidogenic factor-1 has been
shown to play an essential role in the gonadotrope cell lineage, as the
disruption of its gene prevents the expression of gonadotrope-specific
markers in the pituitary (10). Although these data reflect important
progress, our understanding of the pituitary regulatory cascade is
still very fragmentary, and several important regulators remain to be
discovered. Krox-24, also known as Egr-1, NGFI-A, and Zif268, is a zinc finger phosphoprotein that was originally identified as an immediate-early serum response or nerve growth factor response gene product (11, 12, 13, 14). This factor was subsequently shown to be rapidly induced by various stimuli (reviewed in Ref.15). Krox-24 belongs to a multigene family encoding closely related transcription factors (16, 17, 18) that bind to very similar or identical cognate GC-rich sequences and can activate transcription of a nearby gene (19, 20, 21, 22, 23). The expression pattern of Krox-24 during development is rather widespread; the sites of expression include the endothelial system, thymus, muscle, cartilage, bone, and part of the central and peripheral nervous system (24, 25). On the basis of antisense RNA experiments, the gene has been implicated in differentiation of the macrophage lineage and in control of proliferation of T lymphocytes (26, 27). Genomic footprinting experiments have also suggested its involvement in control of transcription of the platelet-derived growth factor-B gene in endothelial cells (28). However, these observations have not yet been corroborated by the analysis of mice carrying a targeted mutation of Krox-24 (29, 30, 31). Furthermore, these latter studies did not confirm the expected involvement of Krox-24 in the control of cell proliferation.
To further investigate the function of Krox-24, we have created a novel targeted allele in which the lacZ gene was introduced into the Krox-24 locus and expected to recapitulate the normal expression of this gene. Observation of these Krox-24-/- mice revealed major differences compared with the previously described mutant animals. Both males and females are sterile, whereas only females were affected in the previous study. Furthermore, in contrast to the other allele, the fertility of the mutant females cannot be restore by injection of LH. Finally, the present mutant allele leads to a reduction in body size and weight, whereas no such defect was reported in the previous study. We show that these defects result from abnormal development of both the anterior pituitary and the ovary, leading to the absence of LH and a significant reduction in GH secretion as well as the absence of LH receptor expression in the ovary. In the pituitary, our data are consistent with a direct involvement of Krox-24 in the control of LHß gene transcription and of the proliferation and/or survival of the somatotrope lineage. Therefore, Krox-24 appears to constitute a novel key regulator of anterior pituitary physiology and to play important, but different, roles in at least two distinct cell lineages.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). This modification was expected to
prevent efficient transcription and translation of the
Krox-24 gene and to place the lacZ gene under the
control of the Krox-24 cis-acting elements, making it a
useful reporter. In addition, to exclude any low level synthesis of a
functional Krox-24 protein from a putative bicistronic or spliced
messenger RNA (mRNA), a frameshift mutation was introduced upstream of
the DNA-binding domain. The mutant allele was, therefore, expected to
be null.
|
, A and B). Among
350 G418-resistant colonies analyzed, 24 (7%) showed a pattern
suggesting the recombination of one Krox-24 allele. These
clones were further studied by Southern blot analysis using internal
and 5'-external probes and were analyzed for the presence of the
frameshift mutation (data not shown). Eleven clones were found to
correspond to the expected replacement of the wild-type allele by a
single copy of the disrupted allele with conservation of the frameshift
mutation. Their karyotypes were established and were found to be
normal. Four of these clones were used for injection into C57Bl6
blastocysts, and all of them led to germ-line transmission. Two
chimeras, deriving from different clones, were selected to establish
families of heterozygous carriers by crossing with C57Bl6 females. The
presence of wild-type or Krox-24- alleles was
determined by PCR amplification using three primers located in the
Krox-24 and lacZ sequences (Fig. 1C). The
heterozygous embryos or animals did not show any overt abnormality,
except that 16% of the males (18 of 115 animals) were sterile,
although they were able to produce vaginal plugs. The absence of
Krox-24 mRNA and protein in homozygous mutant mice was verified by
in situ hybridization and immunocytochemical analyses
performed in several organs (Fig. 2, E
and F, and data not shown). All of the phenotypes associated with the
homozygous mutation that are described below were observed with mice
derived from either clone. Therefore, they are likely to result as a
direct consequence of the inactivation of Krox-24.
|
). At around 2 months the difference was
on the order of 20% and was maintained in older animals. This
differential growth was observed in both males and females (Fig. 3).
|
Krox-24 Is Involved in the Control of LH and
GH Production in the Anterior Pituitary
We tested whether a slower growth rate and sterility in mutant
mice might result from the loss of pituitary functions. Morphological
and histological analyses of the pituitary in adult
Krox-24-/- animals revealed that the gland was
grossly affected by the mutation. Its size was significantly
diminished, the reduction affected specifically the anterior lobe,
while the posterior or neural lobe appeared normal (Fig. 2A). Hormone
expression in the anterior pituitary was subsequently studied by
immunocytochemistry. Pituitary sections from 12-week-old wild-type and
Krox-24-/- mice were analyzed with antibodies
directed against each of the six anterior pituitary hormones. Whereas
the proportions of ACTH-, TSHß-, FSHß-, and PRL-positive cells were
not significantly different in wild-type and homozygous mutant animals,
the mutation led to a clear reduction in the number of GH-positive
cells and to the complete inhibition of LHß expression (Fig. 4). These data were confirmed by
analyzing the hormone expression pattern of pituitary cells that has
been dispersed and cultured for 24 h. In these experiments, the
proportions of cells immunoreactive for ACTH or PRL were similar in
wild-type and homozygous Krox-24 mutants, whereas in the
mutants, the proportion of GH-positive cells was reduced 2-fold
compared with that in wild-type animals, and no cells expressing LHß
were detected (data not shown).
|
and data not shown). Densitometric
analysis of the autoradiograms indicated that the levels of PRL and
GSU in the mutant were not affected. In contrast, the level of GH
was decreased approximately 4-fold in males and 3-fold in females
compared with that in the wild-type animals. LHß could not be
detected in males or females. Hormone immunoassays were performed, in
addition, on older animals. In 2- to 4-month-old males, a 2.5-fold
reduction of the GH content of the pituitary was observed in
Krox-24-/- animals compared with that in
controls [wild-type, 14.8 ± 0.8 µg (n = 4); homozygous
mutant, 5.9 ± 0.8 µg (n = 4); P < 0.0003,
by Student’s t test]. In females of the same age, the
reduction observed in the homozygous mutants was less dramatic
[wild-type, 7.7 ± 1 µg (n = 4); homozygous mutant,
5.2 ± 0.4 µg (n = 4); P < 0.06]. Taken
together, these data indicate a clear decrease in GH content in
homozygous pituitaries and suggest that this phenotype becomes less
severe as the animals age. LH levels were below the detection limit of
our RIA in the serum of both male and female 1- to 4-month-old
Krox-24-/- mice, indicating an at least 6-fold
reduction compared with levels in wild-type littermates (data not
shown).
|
). Surprisingly, the majority of the protein
appeared to be localized in the cytoplasm (Fig. 2E). As expected, no
expression was observed in homozygous mutant animals (Fig. 2F). In
accordance with the immunocytochemical data, analysis of
ß-galactosidase activity in heterozygous animals indicated a very
high level of expression in the anterior lobe (Fig. 2D), confirming
that the expression of this reporter faithfully reflects normal
Krox-24 expression. ß-Galactosidase activity was also
detected in the intermediate lobe, although at a lower level, whereas
the neural lobe was completely negative except in the vascular system.
To precisely correlate hormone and Krox-24 expression in the
anterior lobe, double labeling experiments were performed on
heterozygous mutant pituitaries. Specific hormone synthesis was
revealed by immunocytochemistry, and Krox-24 expression was
detected by X-gal staining, and the sections were examined at high
magnification. X-gal staining was highly localized within the cytoplasm
(Fig. 4), as previously observed in the case of another knock-in
experiment (32). The observations suggested that most, and possibly
all, GH-, LHß-, FSHß-, or TSHß-positive cells expressed
lacZ (Fig. 4, N–Q). In contrast, ACTH-positive cells were
not labeled by X-gal (Fig. 4M). Similarly, most PRL-positive cells did
not express lacZ (Fig. 4R). Nonetheless, a few cells
expressing PRL at a low level were also found to be X-gal positive. It
is possible that these latter cells produce both PRL and GH (33, 34). In conclusion, our data suggest that Krox-24 expression in the anterior pituitary is restricted to specific cell types, including those expressing GH and LH. This supports the idea that the defect in GH and LH production in the homozygous mutant animal is a consequence of the absence of the Krox-24 protein in the corresponding cells.
Krox-24 Is Required for Normal Development of the Anterior
Pituitary
If Krox-24 is continuously required for efficient expression of LH
and GH in adult mice, its transcriptional activation should precede
that of these hormones during pituitary development. Transcription of
GH and LHß has been shown to be initiated around 15.5 and 16.5 days
postcoitum (dpc), respectively (35). We have determined the pattern of
expression of Krox-24 in the pituitary during development by
X-gal staining in heterozygous mutant mice. We find that the gene is
activated around 14.5 dpc in the anterior pituitary, before
GH and LHß activation as anticipated (Fig. 2B).
Interestingly, at this stage Krox-24 is expressed at a much
higher level in the neural lobe. At 18.5 dpc, expression has increased
dramatically in the anterior lobe, is maintained in the neural lobe,
and begins to appear in the intermediate lobe (Fig. 2C). Later,
expression disappears in the neural lobe, while it continues to
increase in the anterior and intermediate lobes (Fig. 2D). Therefore,
the pattern of expression of Krox-24, as deduced by X-gal
staining, is consistent with its involvement in the control of GH and
LH expression during development.
Morphological analysis of the Krox-24-/-
pituitary indicated a dramatic and specific reduction of the anterior
lobe (Fig. 2A). To determine the precise cytological consequences of
the mutation, an ultrastructural analysis was performed. In 4- to
6-week-old mutant mice, no obvious alteration was
observed for the small secretory granule-bearing cells likely to
correspond to corticotropes, gonadotropes, and thyrotropes. In
contrast, an abnormal number of large granule-bearing cells displayed
dilated endoplasmic reticulum (ER) saculae (Fig. 6, A and B). These cells were identified
by immunogold labeling as somatotropes (Fig. 6B, inset). The
ER alteration in these cells varied from slight dilatations of
individual ER cisternae to extremely large vacuole-like dilatations
occupying most of the cytoplasm of the cell; the few remaining
secretory granules were restricted to the cell periphery. Such cells
were found to coexist with somatotropes displaying a normal appearance.
The distribution of the cells presenting a dilated ER complex varied
within the anterior lobe of Krox-24-/- mice;
some areas contained many, and others fewer. We also noted that among
the pituitaries originating from homozygous
Krox-24-/- mice, smaller pituitaries contained
more of those abnormal cells compared with larger glands. Such a
variability prevented a precise determination of the ratio of
somatotropes presenting the abnormal phenotype. Nevertheless, although
some somatotropes displaying a slightly dilated ER were also observed
in wild-type pituitaries, it was unambiguous that the number of these
cells was greatly enhanced in Krox-24-/-
pituitaries. Moreover, cells displaying the most pronounced dilatations
were only observed in homozygous mutant animals. In conclusion, these
data are consistent with a lesion specifically affecting a subset of
the somatotropes and possibly explaining the reduction of the number of
immunoreactive GH-producing cells in mutant animals.
|
, A and D–G). At 5
months, the testes weights of the homozygous mutants were reduced by
approximately 38% compared with those of the wild-type animals
[wild-type, 0.125 ± 0.005 g (n = 12); homozygous mutant,
0.077 ± 0.007 (n = 12); P < 0.0001],
whereas general body weight was only reduced by 20%. In the testes,
the sizes of the seminiferous tubules and of the clusters of Leydig
cells as well as the number of spermatozoa were reduced (Fig. 7, D–G).
The amount of seminal fluid in the seminal vesicles was also
dramatically diminished (Fig. 7, H and I). The sites of
Krox-24 expression in the male reproductive apparatus
revealed by X-gal staining of Krox-24+/-
animals (epithelium of the epididymis and some Sertoli cells in the
seminiferous tubules; Fig. 7, B and C) are not likely to provide a
basis for all of these different phenotypes. In contrast, these
phenotypes may all derive from the deficit in LH that is involved in
controlling the differentiation of Leydig cells, in particular the
synthesis of the androgens that, in turn, are required for normal
development of the male reproductive system (Ref. 36 and references
therein). Indeed, RIA analysis of the level of testosterone in the
serum of 4- to 6-week-old males revealed a significant reduction in
Krox-24-/- animals compared with the control
value [wild-type, 0.72 ± 0.12 ng/ml (n = 5); homozygous
mutant, 0.30 ± 0.03 ng/ml (n = 5); P <
0.04]. Furthermore, a single injection of LH in
Krox-24-/- males restored fertility in about
half of the animals (data not shown), establishing that LH deficiency
is the main cause of male sterility.
|
, A, C, and D). In the uterus, the
reduction affected both the endometrium and the longitudinal and
circular layers of smooth muscle (Fig. 8, C and D). Histological
analysis of the ovary indicated that the antral follicles were
consistently smaller in the mutant, never exceeding 330 µm in
diameter, with no evidence of corpora lutea, whereas the largest
follicles in the wild-type littermates had diameters between 400–470
µm (Fig. 8, E and F, and data not shown). This suggests that the
follicles remain immature in Krox-24-/-
females and that ovulation does not occur, explaining the infertility.
Although such a block is consistent with the absence of LH, the
possibility of an additional direct role of Krox-24 in the gonad was
not excluded, as analysis of lacZ and Krox-24
expression in Krox-24+/- and wild-type ovaries,
respectively, revealed expression in the oocytes, in the thecal and
granulosa cells of mature antral follicles, and in the corpora lutea
(Fig. 8B).
|
, G and H, and data not shown). These enlarged
follicles, however, lacked signs of luteinization (i.e.
enlarged cells in the thecal layer and dissociation of the cumulus
cells surrounding the oocyte). In addition, no corpora lutea were
observed; instead, the largest follicles appeared highly hemorrhagic
(Fig. 8H). This could be due to a defect in follicular steroidogenesis
and/or in the receptivity of follicular cells to LH. To investigate
these possibilities, an immunohistochemical analysis was performed on
ovarian sections from homozygous mutant and wild-type litter mates
using specific antibodies directed against 17-hydroxylase, a
LH-induced steroidogenic enzyme, and LH receptor. Both mutant and
wild-type follicles presented an obvious labeling in the theca of the
largest follicles after staining with the 17-hydroxylase antibody
(data not shown). In contrast, the amount of LH receptor on somatic
cells of the ovarian follicles was severely reduced in
Krox-24-/- mice (Fig. 8, I and J). In
conclusion, the incapacity to rescue the sterility phenotype of
Krox-24-/- females by injection of LH is due
to an additional defect in the ovary, involving a reduction in the
level of LH receptor. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
As thyrotropes, in contrast to somatotropes and gonadotropes, are apparently unaffected by the mutation, Krox-24 may have no role in this cell type. Alternatively, its absence could be compensated by another factor(s), possibly a member of the Krox-24 multigene family. The presence of apparently normal GH-positive cells in Krox-24-/- mice may be related to a heterogeneity of the somatotrope population and also reflect such a compensation. In line with this hypothesis, we observed that Krox-20 is expressed in a subpopulation of GH-positive cells (P. Topilko, S. Garel, and P. Charnay, unpublished observation), raising the possibility that this closely related gene may compensate for Krox-24 inactivation in this cell subpopulation.
Krox-24 Plays Essential, but Different, Roles in the Gonadotrope
and Somatotrope Lineages
The data presented in this report indicate that Krox-24 is
required for synthesis of LHß by gonadotropes in the anterior
pituitary. This is consistent with the analysis of another mutation of
the gene (30), although in the latter case, a low level of LHß was
maintained in males, whereas with the present Krox-24
allele, LHß expression was completely eliminated in both males and
females. In contrast to LHß, GSU, FSHß, and TSHß are normally
synthesized in Krox-24 mutants. As LHß and FSHß are
synthesized in the same cells (38, 39), these data indicate that the
survival and proliferation of the gonadotrope cells are not affected by
the Krox-24 mutation, and that LHß synthesis is
specifically prevented. In addition, these data reveal that the
phenotype does not originate in the hypothalamus, which produces GnRH,
as such a defect would affect both LH and FSH synthesis. Indeed, Lee
and collaborators (30) have also shown that the LHß
promoter contains a Krox-24 binding site and that cotransfection of a
reporter construct driven by the LHß promoter with a
NGFI-A (Krox-24) expression vector leads to an
elevation of LHß promoter activity and that this elevation
is dependent on the integrity of the Krox-24 binding site. Together
with the gene inactivation studies, these data suggest that the
LHß gene constitutes a direct transcriptional target of
Krox-24.
We have observed that Krox-24 inactivation leads to a severe
reduction in the level of GH in the pituitary of young animals, which
may account for the smaller size of the homozygous mutant animals. This
reduction in GH level is accompanied by a significant decrease
(
2-fold) in the number of GH-positive cells, whereas the numbers of
PRL-, FSHß-, TSHß-, and ACTH-positive cells are not affected.
Furthermore, the size of the anterior pituitary is reduced in
Krox-24-/- mutant mice, and ultrastructural
analysis has revealed dramatic cytological abnormalities in many
somatotropes. Such abnormalities were not observed in the little, Ames,
and Snell dwarf mice (40). They may be a consequence of cell
death and/or impairment of normal cell proliferation. In such a case,
the Krox-24 mutation would lead to a partial, but selective,
elimination of the somatotrope compartment in the anterior pituitary.
The absence of effect on the lactotropes suggests that the mutation
might not impair the common precursors of the somatotrope and
lactotrope lineages.
Although our data do not exclude a direct involvement of Krox-24 in the transcriptional activation of the GH gene, they suggest that a large part of the decrease in GH is due to a reduction of the number of cells responsible for its expression. An argument suggesting that Krox-24 is not required for the direct transcriptional activation of the GH gene is provided by the analysis of the mutation introduced by Lee et al. (30). This mutation is likely to eliminate Krox-24 transcription factor activity (see below); nevertheless, it does not affect GH expression.
In conclusion, the Krox-24 mutation results in a deficit in somatotropes and, in turn, a low level of GH secretion. This situation is clearly different from the gonadotrope lineage, where the Krox-24 mutation does not significantly affect the number of cells, as revealed by analysis of FSH-positive cells. Therefore, Krox-24 performs different roles in gonadotrope and somatotrope lineages. In the former, it appears to act as a transcription factor required for the synthesis of one of the hormones produced by these cells, whereas in the latter, it is required for cell survival and/or proliferation, and this function is not likely to be related to its transcription factor activity.
The Two Krox-24 Mutations Are Associated with Different
Phenotypes
In several respects the phenotypes associated with the
Krox-24 mutation analyzed in this study contrast those
reported by Lee and collaborators (30), who created a distinct mutant
allele. These differences include the following. 1) There was a
decrease in the level of GH in the homozygous mutants, which probably
accounts for the significant reduction in the size of these animals. No
such phenotypes were described in the report by Lee and collaborators.
2) Although both mutations affect LHß synthesis, the present mutation
appears more stringent, because LH is eliminated in both females and
males whereas in the case of Lee and collaborators a low level of
LH is maintained in males. This residual level of LH in males might
explain why the male reproductive apparatus is only marginally affected
in their study, whereas the consequences are much more dramatic with
our allele, resulting in sterility. Consistent with this
interpretation, we were able to restore the fertility of
Krox-24-/- males by injection of LH. 3) In our
study, the female sterility phenotype was not rescued by injection of
LH and FSH. We have shown that this is due at least in part to an
additional defect affecting the ovaries: a severe reduction in the
level of LH receptor expression in somatic cells of the ovarian
follicles. This is probably a direct consequence of Krox-24
inactivation in the ovaries because the gene is expressed in these
cells. Therefore, Krox-24 appears to control the development and
physiology of the female reproductive system at two levels along the
pituitary-gonadal axis: the synthesis of LH in the anterior pituitary
and its receptor in the ovarian follicles. This observation is
intriguing, and it will be interesting to assess its significance in
terms of physiology and evolution. The fact that the male sterility
phenotype can be rescued by LH injection suggests that LH receptor gene
expression may not be affected by the Krox-24 mutation in
male gonads; therefore, its regulation may be different in males and
females.
The Dual Roles of Krox-24 in the Pituitary May Correspond to
Different Molecular Functions of the Protein
It is unlikely that the differences in phenotypes associated with
the two Krox-24 mutant alleles could be explained by
differences in the genetic background, as the two mutations have been
maintained in the same hybrid background (129/C57Bl6) (31). Although
this is not a pure background, all of the phenotypes observed in
homozygous mutant animals presented a full penetrance, except male
sterility, with one case of fertility of 40 homozygous males. The
differences in phenotypes are therefore likely to originate from the
mutations themselves. Their comparison may provide us with some clues
on the dual function of the protein. The present mutation involved the
insertion of a lacZ-neo cassette containing a
polyadenylation site in between the Krox-24-coding sequence
and its promoter. This was expected to prevent transcription of the
gene. In addition, a frameshift mutation was introduced into the coding
sequence at the level of a NdeI restriction site
corresponding to the beginning of the DNA-binding domain (Fig. 1A).
Therefore, the resulting allele was expected to be null, and indeed, we
were unable to detect either Krox-24 protein (Fig. 2F and data not
shown) or its mRNA (data not shown) in pituitaries and other organs
from Krox-24-/- mice. In contrast, the
targeting construction of Lee and collaborators involved the insertion
of a neo cassette at the same NdeI site. Although
this is likely to eliminate the DNA-binding activity of the protein,
the presence of a truncated protein corresponding to the N-terminal
half of Krox-24 cannot be excluded. Therefore, we propose the following
interpretation of the phenotypic differences associated with the two
mutations in somatotropes. The Krox-24 protein carries at least two
molecular activities. The first one, requiring the DNA-binding domain,
results in transcriptional activation of target genes, for example
LHß in gonadotropes. The second activity requires
exclusively the N-terminal half of the protein and controls cell
proliferation or survival in somatotropes by unknown mechanisms. We
propose that the latter activity is not completely eliminated in the
case of the mutation generated by Lee and collaborators, and that this
is sufficient to allow normal survival and function of somatotropes.
This model, however, does not account for the other phenotypic
differences between the two alleles.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
ES Cell Culture, Transfection, and Screening
ES cells (CK35 line derived from the 129/SV Pasteur strain) (41)
were maintained on a feeder layer of mouse embryonic fibroblasts that
were mitotically inactivated by mitomycin treatment in DMEM containing
0.05 mM ß-mercaptoethanol, 1 mM pyruvate, 2
mM glutamine, 100 U/ml penicillin/streptomycin, and 1000
U/ml leukemia inhibitory factor (all products from Life Technologies,
Gaithersburg, MD) and supplemented with 15% FCS (Techgene, Les
Ulis, France). Electroporation and G418 selection were performed as
described previously (42). From a total of 6 x 107
transfected ES cells, 350 G418-resistant colonies were picked and
transferred into separate 1.5-cm wells. These clones were subsequently
expanded into two 35-mm diameter petri dishes for freezing and DNA
analysis. Genomic DNA extraction and Southern blot analysis were
performed as described previously (42).
Generation and Screening of Krox-24 Mutant Mice
Krox-24+/- ES cells were injected into
C57Bl/6J blastocysts to obtain chimeras (42). Male chimeras were bred
with C57Bl/6J females to test for germ-line transmission of the agouti
coat color marker. The presence of the Krox-24-disrupted
allele was detected by PCR analysis of tail genomic DNA. A mixture of
three oligonucleotides was used for PCR amplification: a common
5'-primer mapping in the Krox-24 coding sequence
(5'-GAGTGTGCCCTCAGTAGCTT-3') and two different 3'-primers, mapping in
the Krox-24-coding sequence (5'-GGTGCTCATAGGGTTGTTCGCT-3')
and in the lacZ-coding sequence
(5'-AACGACTGTCCTGGCCGTAACC-3'), respectively. PCR amplification was
performed with Taq polymerase (Eurobio) under the conditions
recommended by the supplier and involved 30 cycles consisting of 1 min
at 92 C, 2 min at 60 C, and 2 min at 72 C. Animal studies were
conducted in accord with The Endocrine Society Guidelines for the Care
and Use of Experimental Animals.
ß-Galactosidase Staining and Histological and
Immunohistochemical Analyses
For analysis of the lacZ expression pattern in whole
embryos (14.5–18.5 dpc), the animals were fixed for 30 min in Kryofix
(Merck, Rahway, NJ), cryoprotected in 30% sucrose, embedded in
Tissue-Tek, (Sakuraus Finetek, Torrance, CA), and frozen at -60 C.
Ten-micron cryostat sections were prepared and stained with X-gal as
described previously (22). Testes, ovaries, and pituitaries from
postnatal animals were dissected; fixed overnight in 35% methanol,
35% acetone, and 5% acetic acid; embedded in paraffin; and sectioned
at 6 µm. Some sections were stained with hematoxylin and eosin.
Immunohistochemistry was performed at the indicated dilutions with
polyclonal antisera against mouse PRL (AFP 131078; 1:300), rat GH (AFP
411S; 1:500), rat LHß (AFP 22238790; 1:300), rat FSHß
(85GP9691bFSHb; 1:300), rat TSHß (AFP 1274789; 1:1000), rat ACTH (AFP
39013082; 1:1000; all antibodies from the National Hormone and
Pituitary Program, NIDDK, Bethesda, MD), the 19 amino acids mapping at
the C-terminal terminus of mouse Krox-24 (1:50; Santa Cruz
Biotechnology, Santa Cruz, CA), and rat LH receptor (amino acids
15–38; 1:500). Alkaline phosphatase-conjugated secondary antibodies
(guinea pig, rabbit, and human kits, Vector Laboratories, Burlingame,
CA) were used with nitro blue tetrazolium and
5-bromo-4-chloro-3-indolyl-phosphate as chromogen. Normal serum was
substituted for the primary antibody in negative controls. In the case
of the LH receptor, immunostaining was performed with an
avidin-biotin-immunoperoxidase procedure.
Western Blotting
Single pituitary glands from wild-type and
Krox-24-/- mice were homogenized directly in
50 µl Laemmli sample buffer and boiled for 3 min. Aliquots of the
cleared samples were analyzed by SDS-PAGE (43), and the total protein
concentration was estimated by Coomassie blue staining of the gels and
densitometric analysis. The protein concentration was then normalized
by addition of Laemmli sample buffer. The samples were boiled again,
and the proteins were resolved by SDS-PAGE on 15% polyacrylamide gel
and then electrophoretically transferred to a polyvinylidene difluoride
membrane (Amersham, Arlington Heights, IL). The presence of the
hormones was revealed by incubating the membrane overnight with
antihormone antibodies at the appropriate dilution followed by 1-h
treatment with a peroxidase-conjugated secondary antibody and the ECL
(Amersham) detection system. Densitometric analysis was performed with
an Eagle Eye image analysis system (Stratagene, La Jolla, CA).
Hormone Immunoassays
For the different hormones, samples from the same litter were
analyzed within the same assay. GH enzyme immunoassay was performed on
pituitary extracts. The pituitaries were snap-frozen and stored at -80
C until analysis. They were homogenized in 10 mM phosphate
buffer (pH 7.4), 0.9% NaCl supplemented with sodium deoxycholate
(0.5%), Nonidet P-40 (0.5%), 200 µM
phenylmethylsulfonylfluoride, and 1 µM pepstatin. The
lysates were centrifuged (15,000 x g, 10 min, 4 C)
after a brief ultrasonic disruption. Aliquots of the supernatants were
analyzed using a rat GH double antibody enzyme immunoassay for rat GH,
validated for mouse GH (Ezan E. et al., submitted for
publication). Results were expressed as micrograms of rGH equivalent of
the NIDDK standard rat GH RP-2 per anterior pituitary. Free LHß in
the serum was quantitated using a Reprokit (Sanofi, Milan, Italy).
Testosterone was quantified in the blood obtained after decapitation
using the RIA described and validated by Hochereau de Reviers et
al. (44). The intraassay coefficient of variation was 8.5%, and
the sensitivity of the assay was 0.1 ng/ml.
Electron Microscopy
Pituitaries were dissected from 4- to 6-week-old mice, cut in
small pieces, and immersed for 3 h at 4 C in the fixative solution
[2% paraformaldehyde and 1% glutaraldehyde diluted in 0.1
M phosphate buffer, pH 7.5 (PB)]. In some experiments,
mice were fixed by intraaortic perfusion with the same fixative before
being postfixed by immersion as described above. After fixation, the
tissues were rinsed in PBS, postfixed in 1% osmium tetroxide in PB for
1 h, rinsed in 80 mM maleate buffer (pH 4.45), and
treated en bloc with uranyl acetate (2% in 85
mM maleate buffer, pH 4.75) before being dehydrated and
embedded in Araldite. For immunogold detection of GH, ultrathin
sections were obtained and pretreated with 5% hydrogen peroxide
diluted in PBS (5 min) and with 50 mM glycine in PBS (10
min), then blocked in PBS containing 5% BSA and 10% FCS (30 min).
Sections were incubated with an anti-GH rabbit antiserum (Chemical
Credential (Aurora, OH) 1/100 in PBS containing 1% BSA, 2% FCS, and
0.1% Tween-20) for 3 h at room temperature, followed by a 15-nm
colloidal gold-conjugated goat antirabbit antiserum (Biocell, Cardiff,
U.K.; 1:100 dilution in PBS containing 1% BSA, 2% FCS, and 0.01%
gelatin, 1 h at room temperature). Each incubation step was
followed by three rinses in PBS. Finally, the sections were postfixed
with 1% glutaraldehyde in PBS, rinsed in water, and contrasted using
uranyl acetate and lead citrate.
|
ACKNOWLEDGMENTS |
|---|
-hydroxylase, Dr. R. Counis for the gift of the antibodies against
LHß and GSU, Dr. P. Roche for providing the LH receptor
antibodies, Drs. S. Bao and Z. M. Lei for LH receptor
immunohistochemical analysis, Dr. A. Tixier-Vidal for useful
discussions, and Drs. C. Kordon, J. Ghislain and M. Hausser for
critical reading of the manuscript. We acknowledge the National Hormone
and Pituitary Program, the NIDDK, the NICHHD, and the USDA for the
gifts of antibodies directed against pituitary hormones. We acknowledge
the expert technical help of the staff of the animal house. |
FOOTNOTES |
|---|
This work was supported by grants from Institut National de la Santé et de la Recherche (INSERM), Ministère de l’Enseignement Superieure et de la Recherche (MESR), European Economic Community (EEC), Associaton de Recherche sur le Cancer (ARC), Ligue Nationale Française Contre le Cancer (LNFCC), Associazione Italiana per la Ricerca sul Cancer (AIRC), Associazione Italiana Sclerosi Multiple (AISM), and Telethon (Italy; project D33).
Revision received October 10, 1997. Accepted for publication October 16, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Shima, M. Zubair, T. Komatsu, S. Oka, C. Yokoyama, T. Tachibana, T. A. Hjalt, J. Drouin, and K.-i. Morohashi Pituitary Homeobox 2 Regulates Adrenal4 Binding Protein/Steroidogenic Factor-1 Gene Transcription in the Pituitary Gonadotrope through Interaction with the Intronic Enhancer Mol. Endocrinol., July 1, 2008; 22(7): 1633 - 1646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Salisbury, A. K. Binder, and J. H. Nilson Welcoming {beta}-Catenin to the Gonadotropin-Releasing Hormone Transcriptional Network in Gonadotropes Mol. Endocrinol., June 1, 2008; 22(6): 1295 - 1303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Reynolds, S. D. Kasteler, M. G. Cosio, A. Sturrock, T. Huecksteadt, and J. R. Hoidal RAGE: developmental expression and positive feedback regulation by Egr-1 during cigarette smoke exposure in pulmonary epithelial cells Am J Physiol Lung Cell Mol Physiol, June 1, 2008; 294(6): L1094 - L1101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. L Henderson, J. Townsend, and D. J Tortonese Direct effects of prolactin and dopamine on the gonadotroph response to GnRH J. Endocrinol., May 1, 2008; 197(2): 343 - 350. [Abstract] [Full Text] [PDF]
|
|
