| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Endocrinology (J.A.M., V.L., J.N., M.N., W.E., J.L.W.Y.,
J.R.S., K.E.C.) University of Edinburgh Molecular Medicine
Centre Western General Hospital Edinburgh, EH4 2XU, United
Kingdom
Millennium Pharmaceuticals, Inc. (M.D.J.)
Cambridge, Massachusetts 02139
Developmental
Neuroendocrinology Laboratory (S.W., M.J.M.) Douglas Research
Center Departments of Psychiatry, Neurology, and Neurosurgery
McGill University Montreal, H4H IR3, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Much evidence suggests that GR gene transcription is, in part, permanently determined or programmed by perinatal events, again in a cell-specific manner. Thus, animals exposed to short periods of infantile stimulation (handling) have, as adults, permanently elevated GR expression selectively in hippocampal neurons (12). The hippocampus is a site of glucocorticoid feedback inhibition upon the hypothalamic-pituitary-adrenal (HPA) axis, and adult rats handled as neonates are therefore more sensitive to glucocorticoid-negative feedback with decreased HPA responsivity to stress throughout life (12, 13). In contrast, prenatal treatment of rats with the synthetic glucocorticoid dexamethasone permanently reduces GR mRNA in the hippocampus (14), but increases GR mRNA in the liver (15). These animals have permanently elevated levels of plasma corticosterone, fasting hyperglycemia (attributable to elevated levels in liver of the glucocorticoid-inducible enzyme, phosphoenolpyruvate carboxykinase, the rate limiting step in gluconeogenesis), hyperinsulinemia (15), and hypertension (14). A key question therefore, is how can GR mRNA levels be regulated in a cell-specific and even opposite manner during adult life and particularly by prenatal manipulations?
Surprisingly little is known of the mechanisms that control GR gene transcription. The GR gene spans more than 80 kb and contains 8 coding exons (exons 2 to 9) (16, 17). The human (16, 18), mouse (17, 19), and rat (20) (M. D. Jacobson and K. R. Yamamoto, unpublished data) GR gene promoter regions have been cloned and partially characterized. A single promoter has been described for the human GR gene (16, 18). In mouse, expression of the GR gene is controlled by at least 3 promoters, resulting in GR transcripts with different 5'-untranslated exons designated exons 1A (restricted to T cell lines), 1B, and 1C (the latter is homologous to the exon 1 present in the human GR cDNA) (17), and very recent evidence suggests the existence of 2 more (21). It has been suggested that rat GR mRNA might also exhibit 5'-heterogeneity (20). Little is known about GR promoter usage in tissues in vivo. Here we demonstrate tissue-specific 5'-heterogeneity of rat GR mRNA and present compelling evidence for early-life environmental programming of specific GR gene promoters in the hippocampus.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
).
The majority of this region corresponds to a CpG island (68% CG, with
a CG/GC ratio of 
0.8 between -1620 and -4520 relative to the
translation start at +1, within exon 2) and contains the exon 1
sequence present in the published rat GR cDNA sequence (22) (-3269 to
-3322; here designated exon 16). The sequence is
highly conserved when compared with the mouse GR gene (19) (91%
identity throughout the whole region), including the mouse exons 1B and
1C (17). The rat sequence shows moderate conservation with the
corresponding human GR gene sequence (18, 23) (
70% identity over
the CpG island; nucleotides -1600 to -4220, but only 40% identity
over the region between -50 and -1600), including exon 1 present in
the human GR cDNA (24).
|
).
The majority of clones (31) from rat hippocampal GR mRNA contained exon
110 (corresponding to exon 1C of the mouse GR
gene and the exon 1 sequence present in the human GR cDNA sequence). In
addition, we found 7 novel exon 1 sequences present in rat hippocampal
GR mRNA, exons 13, 14,
15, 17,
18, 19, and
111 (Table 1); exons 15 and
111 are likely to correspond to the recently
described mouse exons 1D and 1E (21), respectively. Of the minor exon 1
species, exons 17 and 111
represented the major variants, present in 8 (exon
17) and 5 (exon 111) of the
independent clones. Exons 14,
15, and 19 were present in,
respectively, 3, 2, and 3 of the independent 5'-RACE clones. Two of the
classes (13 and 18) were
represented each by a single independent clone. No clones were found
corresponding to the exon 1 sequence present in the published rat GR
cDNA sequence (22) (but see below). Of the 4 independent clones
produced from 5'-RACE PCR of thymus RNA, 2 contained exon
110. In addition, a further 2 novel variants of
rat GR mRNA were identified in thymus RNA, containing exons
11 and 12 (Table 1). Exon
11 is likely to correspond to mouse exon 1A (76%
identity) (25), reportedly specific to T lymphocytes (17).
|
; summarized in
Fig. 2). Each of the exons 1 that mapped
to this region was flanked, at the 3' end, by the conserved GT
dinucleotide splice donor site (Table 1). Sequences corresponding to
exons 11, 12, and
13 were not found within the sequenced region,
and Southern blot hybridization showed they were not located within
208 (see Materials and Methods for details of 208),
indicating that they lie at least 15 kb 5' of exon 2 (data not
shown).
|
, Table 2, and data not shown). Interestingly,
although exons 15 and 17
were both present in hippocampal GR mRNA (together accounting for
17% of GR mRNA in hippocampus), in liver and thymus they were below
the limit of detection of the RNase protection assay
(i.e. <1% of the total) (Table 2). Furthermore, exon
111-containing transcripts were relatively more
abundant in hippocampus than in liver RNA but were below the limit of
detection of the assay in thymus (Table 2). Exon
11, identified in a 5'-RACE clone from thymus,
was detectable only in thymus RNA by RNase protection analysis (Fig. 3B
and Table 2). Exon 16 was not detected among the
products of 5'-RACE PCR on hippocampal RNA, possibly as a consequence
of the hybridization of the 5'-RACE UAP-anchor primer [which contains
a G(GGIIG)3 sequence] to a
C8 sequence close to the 3'-end of exon
16; 5'-RACE products likely to contain only very
short amounts of exon 1 sequence were not analyzed. However, exon
16 mRNA transcripts were expressed in all three
tissues, representing around 10–20% of total GR mRNA (Table 2).
Levels of exon 14-- and exon
12-containing GR mRNA were below the limit of
detection of the RNase protection assay (Table 2).
|
|
). Using cRNA probes
specific to the major alternate first exons employed in the
hippocampus, we found that the predominant exon
110-containing transcript was distributed very
similarly to total GR mRNA, with high expression in the dentate gyrus
and CA1 and lower expression in CA3 and CA4 (Fig. 4E). In contrast, GR
mRNA transcripts containing exons 15,
17, or 111 showed a more
homogeneous distribution, although in each case expression was highest
in the dentate gyrus and CA1 region of hippocampus (Fig. 4F, C, and G,
respectively).
|
). Luciferase activity therefore arises
from chimeric RNA transcripts encoding part of an alternate exon 1 of
the GR gene at the 5'-end and represents the activity of the promoter
that directs transcription through that individual exon 1. Although
transcription may additionally originate from alternate
promoters present on the same genomic DNA fragment, these transcripts
will not be transcriptional fusions to luciferase and will not,
therefore, result in luciferase activity (there is no splice acceptor
site upstream of the luciferase gene in these constructs). To measure
promoter activity of the whole CpG island, including exons
14-111, plasmid P2 was
constructed in which the GR gene is joined to luciferase within exon 2
(just before the translation start) (Fig. 5A). In P2, RNA initiating at
any of the transcription start sites will be spliced, from the donor
site at the 3'-end of the respective exon 1 onto the acceptor site at
the 5'-end of the common exon 2; luciferase reporter activity therefore
reflects the total promoter activity of the DNA fragment inserted into
P2.
|
). Activity of all constructs was
similar in all three cell lines with the exception of
P17. P17 was, apart from
P2, the most active construct in B103 and C6 cells but was relatively
less active in HepG2 cells (Fig. 5B). Promoter activity of
P110 was high and P111
activity low in all three cell lines (Fig. 5B).
Differential Regulation of Variant GR mRNA Transcripts by
Early-Life Events
Neonatal handling causes marked and permanent increases in GR mRNA
expression in hippocampus (27). Strikingly, neonatal handling induced
expression of GR mRNA containing the hippocampus-specific exon
17 by 2.5- to 3-fold selectively across all
hippocampal subfields (Fig. 6A), whereas
expression of exon 17-containing GR mRNA in
cortex, where neonatal handling has no effect on expression of GR, was
unchanged by the manipulation (Fig. 6A). In contrast, the level of
expression and distribution of the major exon
110, and the other hippocampus-specific exons
15- and 111-containing GR
mRNAs were unaffected by neonatal handling (Fig. 6B and data not
shown). To start to examine whether this effect was confined to the
hippocampus, we examined rats exposed prenatally to dexamethasone
(dexamethasone administered during week 3 of gestation). This
manipulation selectively and permanently increases hepatic GR mRNA
levels by 25% (15). In these animals RNase protection assays show a
significant decrease in the relative amount of exon
110-containing GR mRNA in the liver in the
dexamethasone-treated group (73 ± 3%; n = 10) compared with
controls (82 ± 2%; n = 9, P < 0.05) (an
increase in one of the minor GR mRNA species would reduce the level of
exon 110-containing RNA as a percentage of the
total). These data suggest that prenatal dexamethasone treatment
induces one of the minor mRNA variants. However, RNase protection
assays demonstrated that the level of exon
16-containing GR mRNA was unchanged in livers
from adult rats treated with dexamethasone prenatally (9 ± 1%;
n = 4), compared with controls (9 ± 1%; n = 5). Exon
11, 15,
17, and 111-containing GR
mRNAs remained very low or were undetectable, suggesting that an as yet
unidentified exon 1-containing GR mRNA is induced in these animals.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
208 containing 15 kb of DNA 5' of exon 2) and
is probably in a similar position to the corresponding mouse exon. All
the other alternate exons 1 identified here are novel. Sequence
analysis of DNA flanking the 5'-end of exon 2 revealed that most of the
novel exons 1 lie within a CpG island highly conserved between rat and
mouse. The 3-kb CG-rich region, therefore, contains at least 8
alternate exons 1 (including exon 16, present in
the published rat GR cDNA sequence), at least 5 of which are conserved
in the mouse. The CpG island is also conserved in the human GR gene,
indicating that the use of alternate exons 1 in GR gene expression may
also occur in humans. At least 6 of the alternate exons 1 are present in vivo in rat GR mRNA. In all adult rat tissues examined, GR mRNA containing exon 110 predominated, accounting for at least half of total GR transcripts. Exon 16 was also present ubiquitously, in a substantial minority of total GR mRNA. All other alternate exons 1 were, to varying extents, expressed in a tissue-specific manner. Exon 11 was well represented in GR mRNA in thymus, but was absent from hippocampus and liver. Preliminary data suggest that exon 11 is not restricted to a specific subset of cells in thymus, but is expressed similarly in thymocytes and thymic epithelium (A. Dammermann, C. Blackburn, and K.E. Chapman, unpublished observations). Hippocampal RNA contained significant levels of the minor exon 15-, 17-, and 111-containing GR mRNA variants that were expressed at either low or undetectable levels in liver and thymus. Five other exon 1 variants (12, 13, 14, 18, and 19) are unlikely to be of significance as they were poorly represented in the 5'-RACE PCR or were undetectable by RNase protection assays. It is unlikely that any further GR mRNA variants are present at significant levels in hippocampus as the sum of the exon 1 variants examined was close to 100% of total mRNA. Interestingly, exon 110-, 16-, and 111-containing transcripts accounted for only 90% of the GR mRNA in liver, suggesting that additional novel exon 1 sequences may be present.
In transient transfection experiments, a construct encoding the whole CpG island of the GR gene, including 8 of the alternate exons 1 and the splice acceptor site within the intron 5' of exon 2, fused to a luciferase reporter gene within exon 2 (P2), exhibited substantial promoter activity in all cell lines tested. This activity results from transcripts originating at any point within the CpG island that are spliced from an appropriate donor site onto the splice acceptor site 5' to exon 2, and represents the sum of the activity of individual promoters on the genomic DNA fragment. Promoter activity was also associated with particular regions of the CpG island, where the fusion to luciferase was made within specific exon 1 sequences. In these cases, no splice acceptor site is available within the luciferase gene, and a transcriptional fusion is generated between the specific exon 1 and the luciferase reporter; luciferase activity therefore reflects transcription through the specific exon 1. Relative activity of these constructs in different cell types was similar with one notable exception, P17 (see below). The low activity of P111, compared with the shorter constructs or to P2, probably results from promoter competition by the stronger promoters directing transcription of exon 16 and exon 110, neither of which will generate productive RNA transcripts encoding luciferase, due to the lack of a splice acceptor site. Interestingly, P17, fused to luciferase within exon 17, had the highest activity of any single promoter construct (P2 activity reflecting activity of the whole region) in B103 and C6 cells, both CNS derived. The activity of this construct was low in hepatic cells, in which P16 and P110 had the highest activity. In vivo, GR mRNA transcripts containing exon 17 were present at significant levels in hippocampus, but absent from liver, suggesting that factors present in cells of CNS origin are responsible for transcription initiation at the promoter upstream of exon 17 in rat hippocampus.
Neonatal handling induces an increase of approximately 50% in total GR mRNA levels in all subfields of the hippocampus, but not in cortex (27). Only the 17 variant GR mRNA was induced in the hippocampus by handling, with a 2- to 3-fold increase, also across all fields of the hippocampus. RNase protection assays, carried out on RNA extracted from the whole hippocampus (which will include glia and interneurons, as well as pyramidal cells and the granule cells of the dentate gyrus) showed that exon 17-containing GR mRNA is normally present in approximately 10% of total GR mRNA in hippocampus. The observed induction of 17 may not appear sufficient to account for the overall increase of approximately 50% in steady state GR mRNA levels after handling (27). However, the heterogeneity of the hippocampus as a whole may have resulted in a dilution of exon 17-containing GR mRNA if it is expressed predominantly in the pyramidal cell layers of the hippocampus and granule cells of the dentate gyrus, thereby lowering the estimate of the amount of exon 17-containing GR mRNA present in these hippocampal neuronal layers obtained by RNase protection assays of whole hippocampus. Indeed, we have previously noted a similar discrepancy between the magnitude of change in mRNA encoding the type I inositol 1,4,5-triphosphate receptor during human pregnancy measured by RNase protection assays and in situ mRNA hybridization (28). Although we cannot exclude the possibility that an additional minor variant of GR mRNA is induced by neonatal handling, none of the other main variant GR transcripts were altered by handling. These data suggest that neonatal handling programs increased hippocampal GR via increased transcription from a novel promoter, 17, active predominantly in CNS-derived cells. A similar permanent induction of a minor promoter of the GR gene appears likely in the liver after prenatal dexamethasone exposure. Within the overall increase in GR mRNA in liver of prenatally treated rats, the proportion containing the predominant exon 110 fell, although we were unable to identify a specific transcript induced. Nevertheless, the clear implication is that early life programming events selectively alter otherwise minor tissue-specific GR gene transcripts, whereas the major and ubiquitous promoters are unaffected, thus programming GR levels for the lifetime of the animal in a tissue-specific manner. Conversely, it is possible that manipulations that decrease GR levels may decrease the levels of the minor GR mRNA variants.
5HT appears crucial in mediating the effects of neonatal handling upon GR expression in hippocampus (29, 30), with subsequent effects upon HPA axis activity (31, 32). The transcription factors NGFI-A and AP2 have been implicated in the induction of GR in the hippocampus after handling or with 5HT (33). Intriguingly, a sequence in the human GR gene that binds AP2 in vitro (34) is completely conserved in the rat GR gene (at -2718). Additionally, within the CpG island, the GR gene contains 16 GC boxes (GGGCGG), which form the core consensus Sp1 site (35) and which may also bind NGFI-A; indeed, there is a sequence exactly matching the consensus binding site for the family of zinc finger proteins that includes NGFI-A (36) immediately upstream of exon 17. We speculate that the increases in AP2 and NGFI-A induced by neonatal handling cause increased transcription from a promoter adjacent to exon 17, leading to increased total GR mRNA.
It remains possible that transcription may originate at a common
promoter further upstream, resulting in a common exon 0, which is then
spliced upstream of the alternate exons 1. We consider this to be
extremely unlikely for the following reasons. First, sequence analysis
of 58 independent 5'-RACE clones neither provided evidence for a common
5'-leader sequence nor revealed any lack of colinearity with the
genomic sequence. Second, the predominant rat exon
110 is homologous to exon 1 of the human GR gene
(16, 18) for which a transcription start site has been mapped. A number
of transcription start sites exist (typically for a TATA-less CG-rich
promoter), but all are located within the region corresponding to rat
exon 110, and all appear to extend to the same
3'-splice site (16, 18). Similarly, sequencing of our 5'-RACE clones
and RNase protection analysis (to map transcription start sites)
suggests that a number of transcription starts exist at least for exon
110 and probably for other exons 1 also (J.
A. McCormick, V. Lyons, and K. E. Chapman, unpublished
observations). Indeed, the 5'-end of the longest of our 5'-RACE clones
containing exon 110 corresponds exactly to one of
the transcription starts mapped for human GR mRNA (18) (Fig. 1).
Finally, our transfection data suggest that promoter activity is
associated with the 5'-flanking regions of specific exons 1. Thus, it
is most probable that alternate exon 1 usage results from transcription
initiation at a number of predominant transcription start sites within
the CpG island, associated with promoter activity. CpG islands are
frequently associated with multiple transcription initiation sites,
often spread over a distance of up to 1 kb, resulting in transcripts
with differing exons 1, all of which, however, are spliced at the same
3'-splice donor site onto exon 2 (e.g. Refs. 37, 38).
Multiple 5'-ends are not usually associated with alternate splice donor
sites, giving rise to discrete alternate exons 1, as we have observed
for the rat GR gene. It is possible that, as the CpG island in the GR
gene is very large, transcription initiates at a large number of
initiation sites spread over the entire 3-kb region. The probability of
splice donor sites occurring within such a large region is high, and
the splicing machinery associated with the RNA polymerase complex may
simply splice from the first appropriate splice donor site that occurs
to the splice acceptor site before exon 2. This hypothesis is supported
by the sequence of mouse exon 1E (corresponding to exon
111) which, at the 5'-end, includes a portion of
exon 1C (corresponding to exon 110) as well as
the intervening genomic DNA (21). Possibly, transcription originated
too far 3' within exon 110 to utilize the exon
110 splice donor site; splicing then occurred at
the next available splice donor site, 3' of exon
111. Certain sites within the CpG island will be
favored for transcription initiation, and this will probably vary in a
tissue-specific manner. Indeed, we see the highest number of variant
exons 1 in hippocampus, a tissue exhibiting a high degree of
complexity. This hypothesis predicts that more alternative exons 1 may
exist in the CpG island if more splice donor sites are predicted, and
we have preliminary evidence that this is the case (V. Lyons and
K. E. Chapman, unpublished observations).
The use of multiple and tissue-specific promoters provides a flexible mechanism for distinct tissue-specific regulation of individual promoters by hormonal signals and has been described for other members of the steroid receptor family (39, 40, 41). GRs are widely expressed in virtually all cell types, although expression levels and functions vary considerably. The complex organization of the 5'-end of the GR gene may reflect this need for diverse tissue-specific regulation. We speculate that exon 110 is constitutively expressed in all tissues, providing a basal or minimal constitutive level of GR gene transcription (e.g. Ref. 8). The existence of tissue-specific promoters (e.g. 11 in thymus and 17 in hippocampus) permits differential regulation of GR in specific cell types and may explain the opposite regulatory effects of glucocorticoid hormones on the levels of GR in T lymphocytes and hippocampus (10, 42). In addition, the presence of several minor promoters clustered together may permit regulation of one or more by signal transduction pathways, resulting in moderate, but biologically significant, changes in total GR mRNA in a specific cell type and thus, ultimately, the glucocorticoid signal on the target genes. Our data illustrate the complexity of transcriptional regulation of GR and provide a basis to understand tissue-specific effects of early-life programming.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
All studies involving animals described herein were approved of by the UK Home Office and were performed in strict accordance with the UK Home Office Animals (Scientific Procedures) Act, 1986.
Isolation of RNA
Total RNA was isolated using the guanidinium isothiocyanate
method (43). Integrity was verified by electrophoresis on
formaldehyde-agarose gels.
5'-RACE PCR
5'-RACE PCR was performed using a commercial kit (Life Technologies , Gaithersburg, MD) according to the
manufacturer’s instructions. First-strand synthesis of GR cDNA was
carried out at 42 C for 30 min on 5 µg total RNA using 8 U/µl
SuperScript II RT and 100 nM primer GSP1
(5'-AAGGGATGCTGTATTCA-3') in a 25 µl reaction containing 20
mM Tris HCl (pH 8.4), 50 mM KCl, 3
mM MgCl2, 10 mM
dithiothreitol, 400 µM deoxynucleoside
triphosphates (dNTPs). After RNase H treatment, cDNA was added to a 24
µl reaction containing 20 mM Tris HCl (pH 8.4), 50
mM KCl, 1.5 mM MgCl2, 200
µM dCTP, and 0.4 U/µl terminal deoxynucleotidyl
transferase. dC-tailed cDNA (5 µl) was used in a PCR reaction with
400 nM anchor primer
(5'-CUACUACUACUAGGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3'), 440
nM primer GSP2 (5'-ACTCCAAATCCTTCAAGAGGTCA-3'), 20
mM Tris HCl (pH 8.4), 50 mM KCl, 1.5
mM MgCl2, 200 µM dNTPs,
and 2.5 U Taq DNA polymerase (Promega Corp.,
Madison, WI), with 35 cycles of PCR amplification (96 C, 45 sec; 45 C,
40 sec; 72 C, 1.5 min), followed by 10 min, 72 C. A nested PCR was
carried out on the products of the first PCR reaction, under the same
conditions with the following primers: UAP
(5'-CUACUACUACUAGGCCACGCGTCGACTAGTAC-3') and GSP3
(5'-TTGGAATCT-GCCTGAGAAGC-3'). PCR products were cloned into pGEM-T or
pGEM-T-easy (Promega Corp.) and sequenced using GSP3.
Subcloning and Sequence Analysis of the Rat GR Promoter
208 contains exon 2 and approximately 15 kb of the rat GR
gene flanking the 5'-end of exon 2 (M. D. Jacobson, unpublished
data). The sequence between -4600 and +500 (the translation start
close to the 5'-end of exon 2 is designated +1) was determined from
restriction fragments subcloned from 208 on both strands using the
Sequenase II system (Amersham International,
Buckinghamshire, UK) or the Thermo Sequenase
33P-radiolabeled terminator cycle sequencing
system (Amersham International). Sequence analysis,
including identification of putative transcription factor-binding sites
was carried out using computer software available at the UK MRC Human
Genome Mapping Project Resource Centre.
Accession Number
The nucleotide sequence data reported in this paper will appear
in the EMBL, GenBank, and DDBJ Nucleotide Sequence Databases under
the accession number AJ271870.
Transfection Analysis Of GR Promoter Activity
Plasmids that fused the rat GR gene to a luciferase reporter
gene were constructed from appropriate restriction fragments ligated
into pGL3-Basic (Promega Corp.) containing a modified
polylinker as follows: P2, a HindIII/SspI
fragment encoding rat GR from -4572 to -9 (the ATG translation start
is designated +1); P2(rev), the same fragment in the reverse
orientation with respect to luciferase; P16, a
HindIII/PstI fragment encoding -4572 to -3336;
P17, a HindIII/BglI
fragment encoding -4572 to -2931; P110, a
HindIII/HincII fragment encoding -4572 to
-2318, and P111, a
HindIII/PstI fragment encoding -4572 to -1767.
Plasmid DNAs used for transfections were purified by CsCl density
gradient centrifugation.
HepG2 (human hepatoma), C6 (rat glioma), and B103 (rat neuroblastoma) cells were maintained in DMEM supplemented with 10% (vol/vol) FBS, 100 IU/ml penicillin, and 100 µg/ml streptomycin. Twenty four hours prior to transfection, HepG2 and C6 cells were seeded at 5–7 x 105 cells per 60-mm dish and B103 cells at 2 x 105 cells per 60-mm dish. Cells were transfected using the calcium phosphate procedure (44) with 1 µg modified pGL3-Basic or an equimolar amount of GR promoter-luciferase plasmid (plasmids varied markedly in size), 1 µg of the ß-galactosidase expression plasmid pCH110 (Pharmacia Biotech, Piscataway, NJ), and carrier pGEM-3 (Promega Corp.) to a total of 10 µg. Forty eight hours after transfection, cells were lysed and luciferase activity determined as previously described (44). ß-Galactosidase activity was determined using the Tropix Galacto-Light kit (Cambridge Bioscience, Cambridge, UK), and luciferase activity/ß-galactosidase activity was calculated. Transfections were carried out in triplicate; each experiment was repeated at least twice and two independently prepared plasmid DNAs were used for each promoter construct.
RNase Protection Assays
With the exception of exon 16 (see below),
exon 1-specific cRNA probes were synthesized from corresponding 5'-RACE
subclones, linearized, and transcribed with either T7 or SP6 phage
polymerase, as appropriate, in the presence of either
[
-32P]-UTP or
[-32P]-GTP (3000 Ci/mmol; Amersham International). The template used to synthesize an exon
16-specific cRNA probe was made by subcloning
into pGEM-T-easy an RT-PCR product generated from total rat liver RNA
using GSP3 (complementary to exon 2) and 5'-primer (5'-ACC-
TGGCGGCACGCGA-3').
RNase protection assays were carried out using a HybSpeed RPA kit (Ambion, Inc., Austin, TX). Hybridization conditions were optimized in preliminary experiments using synthetic RNA templates. Total RNA (50 µg) was coprecipitated with 5–10 x 105 cpm cRNA probe, resuspended in 20 µl hybridization buffer (supplied with the kit) at 95 C, and incubated at 68 C for 1 h. Reactions were incubated with RNase A/T1 (1:25 dilution) for 30 min, 37 C, and RNA products were separated on a 4% polyacrylamide gel containing 7 M urea and visualized using autoradiography or a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA). Data were analyzed using Student’s t test. Significance was set at P < 0.05.
In Situ mRNA Hybridizations
[35S]UTP-labeled RNA probes were
synthesized as previously described (26). After DNase I treatment,
unincorporated nucleotides were removed by passage over a Sephadex G-50
Nick column (Pharmacia Biotech, St Albans, UK). Exon
15-, 17-,
111-, and exon 2-specific templates were
generated by PCR carried out on subclones of 208 using the following
oligonucleotides tagged with sequences encoding either a T3 promoter
(to make sense RNA) or a T7 promoter (to make cRNA or antisense RNA):
15, 5'-primer
(5'-TATTAACCCTCACTAAAGGGTAAGAGGAGGGCGGACT-3'), 3'-primer
(5'-TTAAT-ACGACTCACTATAGGGCCAGCGCGCTCACACT-3');
17, 5'-primer
(5'-CATTAACC-CTCACTAAAGGGC-ACCGTTTCCGTGCAT-3'),
3'-primer (5'-TTAATACGACT-CAC-TATAGGGCAGCGTGTGCCGACCT-3');
111, 5'-primer
(5'-TATTAACCCTCACTAAA-GGGAGCGGCGTCTGGACC-3'), 3'-primer
(5'-TTAATACGACTCACTATAGGGCTA-GCGCTCAAGTTGTC-3') and exon 2, 5'-primer
(5'-ATTAACCCTCACTAAAGGGCC-AATGGACTCCAAAGAA-3') and 3'-primer
(5'-ATAATACGACTCACTATAGGGAA-TCTGCCTGAGAAGC-3'). The template used to
synthesize exon 110-specific cRNA was generated
by PCR from an exon 110 5'-RACE clone using UAP
and 3'-primer (5'-ATAATACGACTCACTATAGGGCTTTGGAGTCCA-
TTGGCA-3').
In situ-hybridization histochemistry was carried out as previously described (26, 45). Silver grains were counted under bright-field illumination using an image analysis system (MCID, Research Imaging, St. Catherine’s, Ontario, Canada). Results were analyzed blind and background, counted over adjacent areas of neuropil, was subtracted. Data were analyzed using Student’s t test. Significance was set at P < 0.05.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported by a Wellcome Trust programme grant and Senior Clinical Research Fellowship (J.R.S). J.A.M. is supported by a studentship from the Medical Research Council.
Received for publication September 14, 1999. Revision received November 29, 1999. Accepted for publication January 5, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
gene are generated by alternative splicing
and promoter usage. Mol Endocrinol 12:1939–1954This article has been cited by other articles:
 |
|
 |
|
  F. D. Provenza What does it mean to be locally adapted and who cares anyway? J Anim Sci, April 1, 2008; 86(14_suppl): E271 - E284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. X. L. Zhang, T. R. Searcy, Y. Wu, D. Gozal, and Y. Wang Alternative promoter usage and alternative splicing contribute to mRNA heterogeneity of mouse monocarboxylate transporter 2 Physiol Genomics, December 19, 2007; 32(1): 95 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Szyf The Dynamic Epigenome and its Implications in Toxicology Toxicol. Sci., November 1, 2007; 100(1): 7 - 23. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Yehuda, M. H. Teicher, J. R. Seckl, R. A. Grossman, A. Morris, and L. M. Bierer Parental Posttraumatic Stress Disorder as a Vulnerability Factor for Low Cortisol Trait in Offspring of Holocaust Survivors Arch Gen Psychiatry, September 1, 2007; 64(9): 1040 - 1048. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Wyrwoll, P. J. Mark, and B. J. Waddell Developmental Programming of Renal Glucocorticoid Sensitivity and the Renin-Angiotensin System Hypertension, September 1, 2007; 50(3): 579 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. C. G. Weaver, A. C. D'Alessio, S. E. Brown, I. C. Hellstrom, S. Dymov, S. Sharma, M. Szyf, and M. J. Meaney The Transcription Factor Nerve Growth Factor-Inducible Protein A Mediates Epigenetic Programming: Altering Epigenetic Marks by Immediate-Early Genes J. Neurosci., February 14, 2007; 27(7): 1756 - 1768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Presul, S. Schmidt, R. Kofler, and A. Helmberg Identification, tissue expression, and glucocorticoid responsiveness of alternative first exons of the human glucocorticoid receptor J. Mol. Endocrinol., January 1, 2007; 38(1): 79 - 90. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Russcher, V. A S H Dalm, F. H de Jong, A. O Brinkmann, L. J Hofland, S. W J Lamberts, and J. W Koper Associations between promoter usage and alternative splicing of the glucocorticoid receptor gene J. Mol. Endocrinol., January 1, 2007; 38(1): 91 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sola, J. D. Amaral, P. M. Borralho, R. M. Ramalho, R. E. Castro, M. M. Aranha, C. J. Steer, and C. M. P. Rodrigues Functional Modulation of Nuclear Steroid Receptors by Tauroursodeoxycholic Acid Reduces Amyloid {beta}-Peptide-Induced Apoptosis Mol. Endocrinol., October 1, 2006; 20(10): 2292 - 2303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. YEHUDA Advances in Understanding Neuroendocrine Alterations in PTSD and Their Therapeutic Implications Ann. N.Y. Acad. Sci., July 1, 2006; 1071(1): 137 - 166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. SECKL and M. J. MEANEY Glucocorticoid "Programming" and PTSD Risk Ann. N.Y. Acad. Sci., July 1, 2006; 1071(1) |
indicates the splice acceptor site in the intron 5' of exon 2. B,
Promoter activity of regions of the GR gene in three cell lines, HepG2
hepatoma cells, C6 glioma cells, and B103 neuroblastoma cells. Activity
of P2 (spanning the whole CpG island, fused to luciferase within exon
2) was nominally set to 100% for each cell line, and activity of the
other constructs was expressed relative to this value. Values represent
means ± SEM.