| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Molecular Biology and Research Center for Cell Differentiation Seoul National University Seoul 151–742, Korea
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
4,300 bases), resulting in a mature mRNA of about 560
bases. However, despite a low level expression, the mature GnRH mRNA
has been detected in numerous extrahypothalamic tissues such as the
pituitary, ovary, lymphocytes, and certain brain regions (3, 4, 5, 6). GnRH
RNA species retaining the intron A are expressed in human and
monkey reproductive tissues (7, 8), and the GnRH primary transcript
appears to be more prevalent than the mature mRNA in the rat ovary (9).
Recently, Zhen et al. (10) demonstrated that there are
alternative GnRH splicing variants in the mouse olfactory,
hypothalamus, and immortalized GnRH cell lines (Gn11 and NLT). Roberts and colleagues (11, 12, 13) found a relatively high prevalence of the GnRH primary transcript and its splicing intermediates in the rat and mouse POA (10–20% of the total gene transcript). In addition, the rat basal olfactory area, which contains a small amount of GnRH neurons, showed about 40% of molar ratio of the primary transcript vs. cytoplasmic mRNA and had a similar amount of the primary transcript compared with that of the POA (11). In contrast to the tissues, in immortalized GnRH-producing cells (GT1), the primary transcript and its splicing intermediates make up less than 2% of the total GnRH gene transcripts (12). The discrepancy of why the tissues contain unusually high portions of splicing intermediates is not yet clearly understood. In the present study, we found that although a variety of neural and peripheral tissues expressed much smaller amounts of the mature mRNA than in the POA, they expressed similar amounts of the GnRH primary transcript and its splicing intermediates. Together with the previous results, this result suggests the possibility that splicing intermediates may be arrested in the splicing process in non-GnRH-producing tissues and that some unspliced GnRH transcripts in the POA may originate from non-GnRH-producing cells. Since accurate and efficient splicing is obviously critical for maintaining the normal function of GnRH-producing cells, in the present study, we examined the splicing arrest of the GnRH primary transcript in non-GnRH-producing tissues and whether this arrest could be relieved by the GT1 nuclear extract (NE).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Long-term exposure (2 weeks)
revealed additional signals of the mature GnRH mRNA in the olfactory
bulb (OB) and piriform cortex (PC) (Fig. 1B). Interestingly, several
putative GnRH splicing intermediates longer than mature mRNA were seen
in most tissues examined. To further define the putative GnRH splicing
intermediates in these tissues, RT-PCR analyses with several sets of
primers were performed. To exclude the possible DNA contamination, we
used deoxyribonuclease (DNase)-treated RNA samples, and PCR reaction
without RT showed no DNA or other contamination in all tissues examined
(data not shown). When the RT-PCR was performed with e1-up primer
located at the 5'-end of the first exon and e3-down primer near the
3'-end of the third exon, the mature GnRH mRNA signals were strongly
observed in the POA, OB, and PC, which is consistent with the Northern
data. In other tissues, the faint mature GnRH mRNA signals were also
detected. In addition, a longer putative GnRH splicing intermediate was
detected in several tissues. To identify this putative GnRH splicing
intermediate, PCR products were applied to Southern hybridization with
the GnRH cDNA, intron A, or intron B probes, respectively. The GnRH
cDNA probe hybridized with both the mature mRNA and the longer PCR
product while intron A probe hybridized only with the longer one. No
hybridization signals were observed in any tissues when intron B probe
was used (Fig. 2A). These results,
confirmed by DNA sequencing analysis, indicate that the splicing
intermediate contains intron A but not intron B, hence the 1A23
fragment. When the RT-PCR was carried out with the iB-up primer located
in intron B and e4-down primer located in exon 4, two types of splicing
intermediates were observed (Fig. 2B). These PCR products were
subjected to Southern blot hybridization with GnRH cDNA, intron B, or
intron C probes, showing that the long intermediate contains introns B
and C, and the shorter one contains intron B but not intron C.
|
|
. Low level expression of this splicing intermediate in the POA may
be due to PCR competition. Because in the POA the mature mRNA is much
more abundant than other splicing intermediates, only the mature mRNA
may be strongly amplified. This is supported by data in Fig. 2C showing
similar expression levels of the splicing intermediate (A23 fragment)
among the tissues when iA-up and e3-down primers were used.
Furthermore, from the difference in the ratio of the mature mRNA to the
splicing intermediates among the POA and other tissues, it can be
speculated that the splicing process of premature GnRH splicing
intermediates may be arrested or minimally active in non-GnRH-producing
tissues.
Intrinsic Splicing Weakness of GnRH Intron A
To examine the splicing activity of each GnRH intron, splicing
substrates containing each intron and its neighboring exons were
constructed. Unnecessary inner regions of the introns were partially
truncated (these were denoted 1
A2, 2B3, and 3C4,
respectively), due to the sizes of native introns being too long to be
easily transcribed in vitro. Since the HeLa nuclear extract
(NE) contains the general splicing machinery, the HeLa NE has been used
to investigate the splicing mechanism of RNAs derived even from
different species such as fly, chicken, and mouse (15, 16, 17). HeLa NE can
also serve as a control NE compared with other NE containing specific
splicing factors (17). Application of 2B3 and 3C4 RNA transcripts
produced spliced exons (Fig. 3) and their
splicing lariats, indicating that introns B and C could be efficiently
spliced by the HeLa NE. However, 1A2 could not be or was marginally
spliced in this system (Fig. 3). This result suggests the weakness of
the GnRH intron A splice sites. Note that spliced RNAs and their
splicing lariats were identified by RNA size markers and running on
higher percentage gels. Spliced RNAs were also confirmed by RT-PCR.
There were several unexpected bands that appear to be cryptic splicing
intermediates (see Fig. 3 and following figures).
|
). These results imply that the low
splicing activity of intron A may be due to the intrinsic weakness of
the 3'-splice site of intron A, resulting in an arrest of splicing
intermediates such as 1A234 seen in Fig. 2A. However, it cannot be
concluded that splicing of intron A is nonexistent in
non-GnRH-producing tissues, since Fig. 2A showed that low-level
expression of the mature GnRH mRNA was observed even in
non-GnRH-producing tissues. Therefore, PCR data shown in Fig. 2A are in
part inconsistent with the in vitro splicing result in which
no splicing of intron A was seen. A recent report demonstrated that
suboptimal 5'- and 3'-splice sites cause retention of an intron in
bovine GH pre-mRNA and that the efficient splicing of this intron is
ensured by the presence of a purine-rich exonic splicing enhancer (22, 23). This postulates that enhancer-like activity of other exons (exons
3 and 4) may enhance splicing of intron A when these exons are
connected to exon 2.
|
A234 was digested by several
restriction enzymes (HincII, SfuI,
SacI, or BamHI) (Fig. 5), and their RNA
transcripts were applied to the splicing reaction. Although little of
1AH was spliced, others (1ASf, 1ASc, and 1ABam) that
contain exon 3 and/or exon 4 showed a slightly increased splicing (Fig. 6). To further
delineate the possible role of ESEs in exon 3 (ESE3) and exon 4 (ESE4)
in the splicing activity, we connected each ESE to the 3'-splice site
of intron A. Placing the ESE3 close to the 3'-splice site (1A(CC)Sf)
showed a very slight splicing effect. The weaker splicing of
1A(CC)Sf compared with 1ASf or 1ASc was unexpected. However,
it is possible that when exon 3 joins to exon 2, the ESE3 directly
links to the five purine bases (AAGAG) at the 3'-end of exon 2 and
strengthens the ESE3. 1A(CSc)Bam containing only the ESE4 showed
increased activities (Fig. 7A). In
1A(CSc)Bam, although the distance between ESE4 and intron A is
longer than that in 1A(CC)Bam, similar splicing activity was
detected, which may be due to the strong activity of the ESE4 and/or an
additive effect of the ESE3 and ESE4. To examine the role of ESE4 in
the absence of ESE3, p1A234 was digested with HincII in
exon 2 and SfuI or SacI site in exon 3. These
gene constructs thus lack ESE3, and the distance between ESE4 and
intron A 3'-splice site of 1ABam (251 bases) was shortened to 112
bases or 58 bases in 1A(HSf)Bam and 1A(HSc)Bam RNA, respectively.
Application of these RNA transcripts showed that shortening the
distance between the 3'-splice site of intron A and the ESE4 led to an
increase in splicing activity and that ESE4 alone was strong enough for
enhancing the intron A splicing (Fig. 7B). These results indicate that
the ESEs located in exon 3 and exon 4 may participate in the splicing
of intron A, and such a partial splicing of intron A from the 1A234 may
account for the extremely low-level expression of the mature GnRH mRNA
in non-GnRH-producing cells (Fig. 2). However, this result could not
explain the mechanism by which GnRH neurons efficiently splice intron A
from splicing intermediates.
|
|
|
ABam
RNA that contains two ESEs was examined in the absence or presence of
GT1 NE. Because of the difficulty in obtaining a high concentration of
GT1 NE, we were unable to totally replace HeLa NE with GT1 NE at the
same concentration. Rather, we retained the total volume (10 µl) but
replaced HeLa NE (50 µg) with 10 or 20 µg of GT1 NE as shown in
Fig. 8. Although we did not use a
constant concentration of NE, we found that this replacement
drastically increased splicing activity of 1ABam in a dose-dependent
manner, but not in 3C4Bam construct (please compare A and B in Fig. 8). This result shows that GT1 NE specifically recognizes the 1ABam
structure rather than 3C4Bam. When 3C4Bam RNA in which the ESE4
is close to the 3'-splice site was subjected, GT1 NE did not show an
additive effect on splicing activity of 3C4Bam RNA, but rather
slightly decreased the splicing activity (Fig. 8B). Although the reason
for the lack of increase in splicing of 3C4Bam by GT1 NE is not
clear, GT1 NE would not be required for the splicing process of highly
conserved 3'-splice site such as intron C.
|
AH, 1ASf,
1ASc, and 1ABam RNA were employed to the splicing reaction in the
presence of GT1 NE. Adding the GT1 NE increased the splicing activity
of only 1ABam RNA, but not in other RNAs (Fig. 9). This result indicates that putative
regulatory splicing factor(s) from the GT1 NE specifically interacts
with the ESE4 even though the ESE4 is located far from the
3'-splice site of intron A. We also examined whether enhanced splicing
activity of 1ABam RNA is due to ESE4 alone or a combinational effect
of ESE4 and neighboring ESE3. Splicing activities of pre-mRNAs
containing ESE3 and ESE4 (1ABam, 1A(CC)Bam) and containing ESE4
alone (1A(CSc)Bam, 1A(HSf) Bam, and 1A(HSc)Bam) in the
presence or absence of GT1 NE were examined. Complementation of GT1 NE
increased the splicing activity of all constructs (Fig. 10), indicating that ESE4 alone is
enough for interaction with GT1 NE. To further elucidate the role of
ESE3 and ESE4 in intron A splicing, we mutated some purines to
pyrimidines as shown in the diagram in Fig. 11. Mutation in ESE3 reduced the intron
A splicing, suggesting a possible additive effect of ESE3 with the
combination of ESE4 on intron A splicing. Mutations in ESE4 were much
more effective than ESE3 mutation, and there was no difference in
mutation effect between ESE4 mutations, indicating that ESE4 sequence
is very specific, and thereby changes in some purine bases critically
disrupted ESE4 activity (Fig. 11).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The low level transcription of GnRH mRNA as well as its unspliced
transcripts has been demonstrated in several neural and peripheral
tissues by the sensitive RT-PCR method (3, 4, 5, 6, 7, 8, 9). Figures 1 and 2 show
that among splicing intermediates a very small portion appears to
undergo further processing to mature mRNAs in non-POA tissues, which
are detectable by the RT-PCR method. Although the biological role of
such a low-level expression of mature mRNA in non-POA tissues remains
to be investigated, several reports have suggested that they contribute
to produce local GnRH peptides that are involved in autocrine and/or
paracrine regulation in several tissues, such as pituitary, gonads,
spleen, and the olfactory system (3, 4, 5, 6, 9). We cannot rule out the
possibility that some part of intermediates would be released into the
cytoplasm without further processing to mature GnRH mRNA (7, 8).
Although a number of studies on GnRH transcript retaining intron A or
lacking exon 2 have been published, their biological consequence
has been poorly understood (2, 7, 8, 10).
One interesting finding of the present study is that the GnRH primary
transcript and splicing intermediates are more prevalent than the
mature GnRH mRNA in non-GnRH-producing tissues. This result may be
closely related to the unusually high portion of the GnRH primary
transcript and its splicing intermediates in the animal POA when
compared with GT1 cells showing less than 2% of the premature splicing
intermediates (11, 12, 13). One possible explanation for the discrepancy
between GT1 cells and animal tissues is that more than 10% of the
premature splicing intermediates in the POA may originate from
non-GnRH-producing cells in the POA because other tissues also express
premature splicing intermediates. According to the in vitro
splicing finding that intron A, but not intron B and C, is
poorly spliced, a predominant accumulation of 1A234 RNA in
vivo would be predicted. Surprisingly, as revealed by Northern
blot analysis, it is not the case; rather, there are many splicing
intermediates (Fig. 1). Currently we do not have a clear explanation
for this result. It is difficult to reconcile in vitro data
with an in vivo situation. It is, however, possible that the
arrest of intron A splicing may cause a rapid degradation of this
splicing intermediate. Indeed, several investigators have described
that nuclear posttranscriptional regulation involves changes in the
stability and accumulation of primary or unspliced transcripts
(26, 27, 28). One recent report showed that a large portion of the
unspliced RNA is degraded before processing and transport to the
cytoplasm without changes in transcriptional rate in certain conditions
(29). It appears that the lack of predominant accumulation of 1A234 RNA
is likely due to degradation of these intermediates. However, this
possibility remains to be examined.
Recently, GnRH splicing variants containing intron A have been found in the human placenta and monkey reproductive tissues (7, 8), along with alternative GnRH splicing variants with deletion of the exon 2 in the OB and caudal hypothalamus (10). The present study showed a lesser splicing activity of intron A than those of introns B and C. Based on the consensus sequence of splice sites (15), the 5'-sequence of intron A shows two base mismatches, and the 3'-splice site has many purines within the polypyrimidine tract. More importantly, a putative BPS exists within the polypyrimidine tract, although usually the BPS locates the upstream region of the polypyrimidine tract. Recognition of the polypyrimidine tract by U2 auxiliary factor (U2AF) is essential for the binding of U2 small nuclear ribonucleoprotein (snRNP) with the BPS (30). One of the splicing factors, BBP (branchpoint bridging protein), can also interact specifically with the pre-mRNA BPS (21). The increased splicing activity of intron A, when BPS was moved to the upstream region of the pyrimidine tract, may raise the possibility of steric hindrance among U2 snRNP, U2AF, BBP, and other splicing factors in the splicing of intron A. However, this possibility should be further examined. Together with this BPS switch experiment, increased splicing activity of intron A when intron A was point-mutated to strengthen the 5'- and 3'-splice site apparently indicates the intrinsic weakness of the intron A splice sites.
Sequence analysis of GnRH exons 3 and 4 revealed that there are two
purine-rich sequences, one located in the 5'-region of exon 3 and the
other located in the 5'-region of exon 4. The purine-rich sequences are
well known splicing enhancers that exist within exon sequences located
downstream from introns containing a weak 3'-splice site in a variety
of eukaryotic RNAs (15, 31, 32). The enhancers can increase the
splicing activity of a weak intron without the aid of specific splicing
factor(s), depending on its distance from the 3'-splice site: they must
be located within 100 nucleotides of the regulated intron (15). The
ESE3 appears to be very weak because shortening the distance between
the ESE3 and the 3'-splice site of intron A had little effect on
splicing activity. However, the ESE4 is much stronger than the ESE3.
The activity of the ESE4 is highly dependent on the distance: the
closer the ESE to the 3'-splice site, the stronger the splicing
activity became. Usually, enhancer sequences are recognized by general
splicing factors, such as SR proteins and U1 snRNP. The interaction of
general splicing factors with the enhancer is enough to strengthen the
binding of U2AF to the polypyrimidine tract resulting in an efficient
splicing of the weak intron when the enhancer is near the 3'-splice
site (18, 32). More importantly, efficient splicing activity was
observed despite the long distance between ESE4 and the 3'-splice site
of intron A (250 bases) in the presence of GT1 NE. This finding
suggests that some specific splicing factor(s) in GT1 NE are required
to maintain the ESE activity because of the long distance from the ESE
to intron A. The regulatory protein-dependent splicing activity of the
enhancer sequence was shown in Drosophila doublesex (dsx)
gene whose alternative splicing relies on the specific splicing
factors, Tra and Tra2 (15). Recently, a neuron-specific splicing
enhancer-binding protein has been identified (17) along with
neuron-specific RNA-binding proteins, Hu proteins (33). Interestingly,
Hu proteins showed strong homology with the Drosophila
splicing factor sxl, and expression of Hu gene products revealed
developmental stage specificity and different spatial distribution in
the central nervous system. It was also suggested that different
combinations of Hu proteins in individual neurons determine different
neuron-specific aspects of posttranscriptional RNA regulation (33).
These findings raise the possibility that there is a GnRH
neuron-specific splicing system and associated specific splicing
factors.
It is also notable that the GnRH gene uses the ESE4 to
facilitate the splicing of intron A, since other ESEs are usually
located downstream of the neighboring intron (24, 32). Thus, the ESE4
may not be active in the primary transcript status, but following the
splicing of introns B and C from the primary transcript, ESE4 may act
on the intron A splicing. Therefore, it can be assumed that there is a
specified order in the GnRH transcript splicing system. Enhanced
splicing activity by the interaction of GnRH neuron-specific splicing
factor(s) to the ESE seems to take place in a certain condition
especially when the 3'-splice site is weak. Failure of enhanced
splicing of 3C4Bam pre-mRNA with GT1 NE indicates that such an
interaction between GnRH neuron-specific splicing and ESE4 appears not
to be required for the splicing of introns that show highly conserved
3'-splice site like intron C. It is postulated that the proper and
temporal association of the specific splicing factor(s) with the ESE4
may be another regulation mechanism to control the efficient splicing
of the primary transcript. It is also possible that GT1-specific
activity may require the presence of both ESE4 and ESE3. In this case,
ESE3 may serve as a bridge for the interaction of ESE4 and a weak
3'-splice site. It remains to be resolved whether GT1-specific splicing
factors directly bind to ESE3 and/or ESE4 or to other proteins that are
already bound to ESEs.
Mass production of the mature GnRH mRNA by an efficient and accurate splicing process is required for the production of intact GnRH peptide, which is obviously critical for the normal function of GnRH neurons. Therefore, it can be assumed that the presence of the GnRH neuron-specific splicing factor(s) is one of the selection systems to maintain the specificity of GnRH neurons. Issues such as identification of the splicing factors involved in the facilitated splicing of intron A in GnRH-producing cells, and the interaction of splicing factors with GnRH exonic enhancers and their protein-protein-snRNP interaction during the splicing of intron A, should be addressed for the elucidation of splicing mechanisms in the mammalian brain.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Northern Blot Hybridization Assay
Total RNA was extracted with the acid guanidinium
thiocyanate-phenol-chloroform method (34). RNA (30 µg) was dissolved
in distilled water and denatured in 50% formamide, 6.2% formaldehyde,
20 mM MOPS (3-[N-morpholino]propanesulfonic
acid), 5 mM sodium acetate, and 1 mM EDTA at 60
C for 5 min. Electrophoresis was performed at 100 V for 1.5 h in a
submarine 1.2% formaldehyde agarose gel. RNA was transferred to Nytran
filter (pore size: 0.45 µm, Schleicher & Schuell, Inc.,
Dachen, Germany) for 18 h by capillary transfer. The GnRH RNA
probe labeled with 32P-UTP (Amersham Pharmacia Biotech, Little Chalfort, UK) was prepared by in
vitro transcription of the rat GnRH cDNA clone inserted into
plasmid pGEM4, which was generously provided by Dr. Kelly Mayo
(Northwestern University, Evanston, IL). Hybridization procedures and
rehybridization with 18S cDNA were performed as previously described
(35).
Oligonucleotides
e1-up 5'-CACTATGGTCACCAGCGGGG-3' (20 mer) e3-down
5'-AGAGCTCCTCGCAGATCCCTAAGC-3' (24 mer) e4-down
5'-GCTGCTGGGTATAGAAATGCG-3' (21 mer) iA-up
5'-CCCTCTGTGTCTTGATGTCCC-3' (21 mer) iB-up
5'-CATCACTTCTCCACCCCTTG-3' (20 mer)
RT-PCR Analyses
RT-PCR was carried out as previously described (36, 37). RNA
samples (1 µg) were applied to the RT reaction mixture containing 200
U of RNaseH- moloney murine leukemia virus (MMLV) reverse
transcriptase (Life Technologies, Inc., Gaithersburg, MD),
50 pmol of random hexamer, 20 U of RNase inhibitor (Promega Corp., Madison, WI), 10 mM Tris-HCl, pH 8.3, 50
mM KCl, 5 mM MgCl2, and 1
mM deoxynucleoside triphosphate. The RT mixture was
overlaid by 50 µl of light mineral oil. The RT reaction was carried
out at 37 C for 30 min followed by a 5-min denaturation period at 99 C.
Subsequently, 80 µl of the PCR reaction mixture containing 10
mM Tris-HCl, pH 8.3, 50 mM KCl, 2
mM MgCl2, 50 pmol of upstream and downstream
primers, and 2.5 U of Taq DNA polymerase (Perkin Elmer Corp., Norwalk, CT) were added. Forty cycles of PCR
amplification were carried out with the condition of denaturation at 94
C for 1 min, primer annealing at 60 C for 1 min, and primer extension
at 72 C for 2 min. Ten-microliter aliquots of PCR products were
electrophoresed on an 1.5% agarose gel in Tris-acetate-EDTA
buffer, stained with ethidium bromide, and photographed under UV
illumination with Polaroid 667 film (Polaroid, Cambridge, MA).
Southern Blot Hybridization
After rinsing with bidistilled water, the gel was soaked in 0.5
M NaOH, 1 M NaCl solution for denaturation of
double-strand DNA followed by resoaking in 0.5 M Tris-HCl
(pH 7.4), 1.5 M NaCl solution for neutralization. The DNA
was then blotted onto Nytran membrane by capillary transfer method. The
Nytran membrane was prehybridized in 50% formamide, 10 x
Denhardt solution, 6 x SSPE, 1% SDS, 50 µg/ml salmon sperm
DNA. Hybridization was carried out overnight at 42 C in the same buffer
including 32P-labeled probes. The membrane was washed twice
in 6x SSPE, 0.5% SDS at room temperature for 15 min and twice in 1x
SSPE, 1% SDS at 42 C for 15 min. The washed filter was exposed to
x-ray film at -70 C for 1 day (36). The probes used in this study are:
the intron A probe made by cutting HincII and
PstI sites in the intron A; the intron B probe containing
the DNA fragment from iB-up primer to SacI site in intron B;
the intron C probe consisting of the SacI site in the 3'-end
of exon 3 to the NcoI site in intron C; and the GnRH cDNA
probe containing all exons.
DNA Constructions and In Vitro Splicing
Substrates
The GnRH DNA fragment containing exon 1, intron A, exons 2 and 3
(1A23) was obtained by RT-PCR with e1-up and e3-down primers and
subsequently cloned into the pGEM-T vector (Promega Corp.)
(This gene construct was denoted as p1A23.) DNA fragment 1A23Sf
obtained by digestion with SphI and SfuI from
p1A23 was cloned into the p1234 digested with SphI and
SfuI, generating p1A234. Subsequently, StyI and
PstI sites within intron A were digested and ligated,
resulting in p1A234. This plasmid was linearized by
HincII, SfuI, SacI, or
BamHI and used as a template for in vitro
transcription. In vitro transcribed RNAs from these
templates were denoted as 1AH, 1ASf, 1ASc, and 1ABam,
respectively (Fig. 5). To shorten the distance of the ESEs from the
3'-splice site of intron A, we artificially made ClaI sites
near (3 bases from 5'-end) the 5'-end of exon 2 and the 5'-end of exon
3. This was cut by ClaI and ligated, which resulted in a
truncation of most of exon 2 (denoted as p1A(CC)34). The
p1A(CC)34 was further digested by ClaI and
SacI and ligated to make p1A(CSc)4. The p1A(CC)34 was
linearized by SfuI or BamHI, whose in
vitro transcript was 1A(CC)Sf or 1A(CC)Bam. The p1A(CSc)4
was linearized by BamHI, resulting in the 1A(CSc)Bam RNA
transcript (Fig. 7A). The p1A234 was digested with HincII
and SacI, and ligated, generating the p1A2(HSc)4. The
p1A234 was digested with HincII and Sfu1,
which produced the p1A2(HSf)34. These plasmids were linearized by
BamHI, producing 1A(HSc)Bam and 1A(HSf)Bam RNA
transcripts, respectively (Fig. 7B). The A2B3 DNA was obtained by PCR
from rat brain genomic DNA with iA-up and e3-down primers and cloned
into the pGEM-T vector (denoted as pA2B3). EcoRI site in the
middle of intron B was cut, and this linearized DNA was treated with
the ExoIII nuclease and S1 nuclease followed by ligation, generating
the pA2B3. The HincII site in exon 2 and SacI
site in exon 3 were cut, and this DNA fragment was subcloned into the
pGEM4Z vector, resulting in the p2B3. Using iB-up and e4-down
primers, the pB3C4 was cloned. Intron B and a part of exon 3 were
removed by digestion with ApaI and SfuI (p3C4).
This clone was further digested by PstI and NcoI
to remove the inner part of intron C, generating the p3C4 (Fig. 3).
Site-directed mutation of the 5'- and 3'-splice site of intron A was
performed by PCR with mutated primers. The sequence of the 5'-splice
site of intron A (GUAAAA) was substituted to
GUAAGU (1A(5'm)H) and the sequence of the 3'-splice site
(UGUGUCUUGAUGUCCCUUAG)
was replaced by
UCUCUCUUGAUCUCCCUCAG
(1A(3'm)H). Mutation of both sites was denoted as 1A(5'3'm)H. The
1A(3'mBPS)H RNA has a sequence of
UACUGAUCCCUCUCUCUCUUUUUCUCCCUCAG.
Underlines show the changed bases from the 1A(3'm)H, and
bold characters indicate the BPS (Fig. 4). Site-directed
mutation of the ESE3 and ESE4 was performed by PCR with mutated primers
as shown in diagram of Fig. 11. All the final products were
sequenced by Sanger dideoxy method (Sequenase 2.0, Amersham Pharmacia Biotech).
In Vitro Splicing Reactions
Pre-mRNA substrates were synthesized in 25 µl in
vitro transcription reactions containing 200 ng of template DNA,
10 U of T7 RNA polymerase (Roche Molecular Biochemicals,
Nutley, NJ), 0.5 mM diguanosinetriphosphate (Roche Molecular Biochemicals), 0.5 mM ATP and CTP, 0.05
mM GTP, 15 µM UTP, and 2.5 µl of
[32P]UTP (Amersham Pharmacia Biotech). The
RNAs were purified by electrophoresis on a 6% polyacrylamide gel
containing 8 M urea. The purified RNA substrates (usually
about 20,000 cpm) were applied to the splicing reaction in a total
volume of 25 µl containing 5 mM HEPES, pH 7.9, 20
mM creatine phosphate, 0.4 mM ATP, 0.6%
polyvinyl alcohol, 3 mM MgCl2, 100 µg HeLa
NE, and 20 U RNasin according to the manufacturer’s protocols
(Promega Corp.). In the case of replacement experiments in
Figs. 8–11, the half-volume (50 µg) of HeLa NE was replaced by NE
storage buffer, 10 µg or 20 µg of GT1 NE. The splicing reaction was
performed at 30 C for 3 h and stopped by the addition of the 2x
stop mix containing 0.5% SDS, 2 mM EDTA, 3 µg/ml tRNA,
0.03 M sodium acetate, and 0.03 M Tris-HCl, pH
7.4. After the phenol/chloroform extraction and subsequent ethanol
precipitation, the RNAs were analyzed on 6% polyacrylamide gels
containing 8 M urea (15).
GT1 NE Preparation
GT1–1 cells (kindly provided by R.I. Weiner, University of
California at San Francisco, San Francisco, CA) were maintained in DMEM
with 10% FCS under a humidifying atmosphere containing 5%
CO2 at 37 C. About 3 x 108 GT1–1 cells
were harvested from the cell culture media. The NE was prepared as
described by Dignam et al. (38) with a slight modification.
Several protease inhibitors (1 µg/ml aprotinin, 0.5 µg/ml
leupeptin, 0.7 µg/ml pepstatin A) were included in the normal buffer
containing 0.3 M HEPES (pH 7.9), 25% (vol/vol) glycerol,
0.42 M NaCl, 1.5 mM MgCl2, 0.2
mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride
(PMSF), and 0.5 mM dithiothreitol (DTT). Instead of buffer
D, a nuclear storage buffer containing 40 mM Tris-HCl (pH
7.4), 100 mM KCl, 0.2 mM EDTA, 0.5
mM DTT, 0.5 mM PMSF, 25% (vol/vol) glycerol
was used. The protein concentration was usually 4–5 mg per
ml.
|
FOOTNOTES |
|---|
The present study was supported in part by the Korea Science and Engineering Foundation (KOSEF) through the Research Center for Cell Differentiation and by Ministry of Science and Technology through Korea Brain Science Program.
Received for publication November 24, 1998. Revision received April 6, 1999. Accepted for publication August 4, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
messenger
ribonucleic acid in the ovary. Endocrinology 127:2350–2356
-crystallin pre-mRNA in a
mammalian nuclear extract. J Biochem 102:1289–1301 expression
in mitogen-activated primary T lymphocytes. EMBO J 9:3831–3837[Medline]
-amidating monooxygenase messenger ribonucleic
acid levels by a nuclear posttranscriptional event. Endocrinology 138:5256–5265
-immunoglobulin
precursor mRNA. Mol Cell Biol 8:2610–2619This article has been cited by other articles:
 |
|
 |
|
  E. Park, M. S. Lee, S. M. Baik, E. B. Cho, G. H. Son, J. Y. Seong, K. H. Lee, and K. Kim Nova-1 Mediates Glucocorticoid-induced Inhibition of Pre-mRNA Splicing of Gonadotropin-releasing Hormone Transcripts J. Biol. Chem., May 8, 2009; 284(19): 12792 - 12800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Yue, T. A. Ponzio, R. L. Fields, and H. Gainer Oxytocin and vasopressin gene expression and RNA splicing patterns in the rat supraoptic nucleus Physiol Genomics, November 12, 2008; 35(3): 231 - 242. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Xu, D. Walker, A. Bernardo, J. Brodbeck, M. E. Balestra, and Y. Huang Intron-3 Retention/Splicing Controls Neuronal Expression of Apolipoprotein E in the CNS J. Neurosci., February 6, 2008; 28(6): 1452 - 1459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Park, J. Han, G. H. Son, M. S. Lee, S. Chung, S. H. Park, K. Park, K. H. Lee, S. Choi, J. Y. Seong, et al. Cooperative Actions of Tra2{alpha} with 9G8 and SRp30c in the RNA Splicing of the Gonadotropin-releasing Hormone Gene Transcript J. Biol. Chem., January 6, 2006; 281(1): 401 - 409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. H. Son, H. Jung, J. Y. Seong, Y. Choe, D. Geum, and K. Kim Excision of the First Intron from the Gonadotropin-releasing Hormone (GnRH) Transcript Serves as a Key Regulatory Step for GnRH Biosynthesis J. Biol. Chem., May 9, 2003; 278(20): 18037 - 18044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Seong, J. Han, S. Park, W. Wuttke, H. Jarry, and K. Kim Exonic Splicing Enhancer-Dependent Splicing of the Gonadotropin-Releasing Hormone Premessenger Ribonucleic Acid Is Mediated by Tra2{alpha}, a 40-Kilodalton Serine/Arginine-Rich Protein Mol. Endocrinol., November 1, 2002; 16(11): 2426 - 2438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Seong, B. W. Kim, S. Park, G. H. Son, and K. Kim First Intron Excision of GnRH Pre-mRNA During Postnatal Development of Normal Mice and Adult Hypogonadal Mice Endocrinology, October 1, 2001; 142(10): 4454 - 4461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Romano, R. Marcucci, and F. E. Baralle Splicing of constitutive upstream introns is essential for the recognition of intra-exonic suboptimal splice sites in the thrombopoietin gene Nucleic Acids Res., February 15, 2001; 29(4): 886 - 894. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
