(Received for publication, October 7, 1996, and in revised form, January 13, 1997)
From the Department of Medicine, Gonadotropin-releasing hormone (GnRH) is encoded
by the proGnRH gene which contains four exons and three introns. In
this study, two immortalized GnRH-expressing cell lines (Gn11 and NLT) were characterized. The NLT and Gn11 cells, derived from a same brain
tumor in a transgenic mouse, display neuronal morphology and
neuron-specific markers. However, NLT cells secrete much higher levels
of GnRH than Gn11 cells. To delineate the mechanism underlying this
difference, reverse transcriptase-polymerase chain reaction and RNase
protection assays were performed to examine proGnRH gene expression.
While the mature proGnRH mRNA was predominately expressed in NLT
cells, Gn11 cells express an abundant short transcript. Sequence
analysis revealed that this short transcript contains exons 1, 3, and
4, but not exon 2, which encodes the GnRH decapeptide. RNase protection
assays demonstrated that NLT cells express much higher levels of mature
proGnRH mRNA than Gn11 cells. The lower level of GnRH secreting
capacity in Gn11 cells is due, in part, to decreased expression of
mature proGnRH mRNA. When proGnRH gene expression in the mouse
brain was examined, the same short splicing variant was observed in the
olfactory area and preoptic area-anterior hypothalamus. But the
prevalent transcript in these regions was the mature proGnRH mRNA.
In contrast, only the mature proGnRH mRNA was found in the caudal
hypothalamus. These results suggest that alternative splicing may be
one of the mechanisms regulating proGnRH gene expression in the animal
brain.
GnRH1 neurons in the hypothalamus play
an essential role in the regulation of mammalian reproduction. These
neurons originate from precursor cells in the olfactory placode and
migrate to their target sites during embryonic development (1, 2). The
most prominent axonal projection of GnRH neurons is to the median
eminence, where GnRH is released and transported via the hypothalamic
hypophyseal portal vessels to the anterior pituitary to stimulate the
synthesis and release of luteinizing hormone and follicle-stimulating
hormone (3). The regulation of GnRH neuronal activities has been
difficult to study, however, due to their scattered distribution and
paucity in cell number (3-5). Recently, this laboratory and Mellon
et al. (6, 7) have generated immortalized GnRH-expressing
neuronal cell lines by targeted tumorigenesis. In our laboratory,
targeting the expression of the simian virus 40 T antigen (SV40-Tag) to the GnRH neurons with the human GnRH gene 5 Although the NLT and Gn11 cells were derived from the same tumor, RIA
measurement of GnRH concentrations demonstrated that the NLT cells
secrete about 10 times higher levels of GnRH than the Gn11 cells. In an
attempt to understand the molecular basis responsible for generating
this difference, we characterized the expression of the proGnRH gene in
these cell lines by RT-PCR and solution hybridization-RNase protection
assays. In addition, the expression of proGnRH gene expression in the
mouse olfactory and preoptic area-hypothalamus was also examined. Our
results demonstrate that a splicing variant lacking exon 2 of the
proGnRH gene was present in Gn11 cells but not in NLT cells. This same
splicing variant was also found in the olfactory and preoptic
area-anterior hypothalamus, but not in the caudal hypothalamus of the
mouse brain. This is the first demonstration that alternative splicing of the primary proGnRH gene transcript occurs in the immortalized GnRH-expressing neuronal cell lines and in the animal forebrain.
Gn11 and NLT cells were maintained in
Dulbecco's modified essential medium supplemented with 10% fetal
bovine serum. For GnRH measurement, culture medium was replaced 4 h before sample collection from dishes containing 90% confluent cells,
and a 0.5-ml medium sample was collected from each dish and kept at
The levels of GnRH in the samples were determined
by RIA as described (10). Cells in the culture dish were lysed for
determination of protein concentrations as described (Sigma) (11). GnRH
values were expressed as picograms/ml of medium/µg of protein. The
sensitivity of the GnRH RIA was 0.2 pg/tube, and the interassay
coefficient of variation at 20 pg/tube was 14%.
Adult CD1 mice maintained on a 12/12 h
light/dark schedule (lights on at 07:00 h) were sacrificed by
CO2 inhalation and cervical dislocation. The olfactory
cortex was isolated to include both olfactory bulbs and the tissue
rostral to the hypothalamic section, with a dorsal incision 3 mm deep.
Preoptic area-anterior hypothalamus was dissected from the anterior
edge of the mammillary bodies to the anterior of the optic chiasm,
laterally 1 mm beyond the lateral aspect of the median eminence and 3 mm dorsally. The caudal hypothalamus included the rostral mammillary
bodies to a point 1 mm caudal to the mammillary bodies, with the same
dorsal, and lateral parameters as those taken for the rostral
hypothalamus. The brain tissues were kept at Cells growing on glass coverslips were
fixed, permeabilized, and stained as described by the manufacturer
(Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD).
Primary antibodies were used at the following dilutions: anti-MAP-2
antibody 1:500, anti-Tau antibody 1:500 (Sigma). Labeled cells were
examined on a microscope using a × 100 lens.
RNA was extracted from
cells and the brain tissues according to the manufacturer's manual
(Molecular Research Center, Inc., Cincinnati, OH). The resultant RNA
was suspended in 20 µl of RNase-free diethyl pyrocarbonate water for
RNase protection assay and RT-PCR analysis.
First-strand cDNA synthesis was performed by
using oligo(dT)16 as the primer and Moloney murine leukemia
virus reverse transcriptase (Boehringer Mannheim). PCR was performed by
using primers complementary to sequences in exon 1 and exon 4 of the
mouse proGnRH gene. The 5 PCR reaction mixtures were
separated by electrophoresis in a 1.5% agarose gel, and the two bands
corresponding to 352 and 210 bp were cloned into pGEM-T plasmids
(Promega, Madison, WI), and sequenced by Sanger's method using T7/SP6
primers and Sequenase® according to the manual (U. S. Biochemical
Corp.).
The PCR products were
electrophoresed in a 1.5% agarose gel and transferred to a
nitrocellulose membrane. Probes for the Southern blot were obtained
from the above pGEM-T plasmids through restriction enzyme digestion.
The cDNA fragments were [ Fifteen µg of total RNA from NLT,
Gn11, or CV-1 cells were electrophoresed on a 1% agarose gel and
transferred to a nitrocellulose membrane. After hybridizing to a
32P-labeled probe specific for Tau, the membrane was washed
three times with 0.1 × SSC, 0.1% SDS at 45 °C. For
autoradiography, the membrane was exposed to x-ray film with an
intensifying screen at The assay was
performed with the RPA II ribonuclease protection assay kit (Ambion,
Austin, TX). Template cDNA inserted in the pGEM-T plasmids was the
same as that used for making Southern blot probes. The template plasmid
was linearized with the restriction enzyme NcoI, and
in vitro transcription was performed by using [ Immortalized NLT and Gn11 GnRH-expressing neuronal cells
were generated by targeting the expression of the SV40-Tag to GnRH neurons with the human GnRH gene promoter (7). The hybrid gene containing the human GnRH gene promoter from The NLT and Gn11 cells have a doubling time about 24-48 h. Both cell
types displayed neuron-like morphology with fusiform or multipolar
processes and had extensive contacts as shown in Fig.
1a. Some thin processes extended far away
from cell body and resembled neurosecretory axons by exhibiting
varicosities, while other processes exhibited dendrite-like appearance.
Immunocytochemical staining with polyclonal antibodies against either
MAP-2 or Tau proteins, two neuron-specific markers (8, 9), demonstrated the expression of both antigens in NLT and Gn11 cells (Fig.
1a). Shown in Fig. 1b is a Northern blot
analysis, indicating the expression of Tau mRNA in both NLT and
Gn11 but not in CV1 cells.
RIA measurement of GnRH concentrations in the media collected from the
culture dishes indicated that both NLT and Gn11 cells secrete GnRH.
However, NLT cells secrete about 10 times higher levels of GnRH than
Gn11 cells (Fig. 1c). These results suggest that although
both the NLT and Gn11 cells are GnRH-expressing neuronal cells derived
from the same tumor, a significant difference exists in their capacity
to secrete GnRH.
In an attempt to delineate the molecular basis responsible
for generating this difference in GnRH secreting capacity between these
two cell lines, we utilized RT-PCR to characterize proGnRH gene
expression in these two cell lines. Interestingly, in Gn11 cells, two
different species of PCR products were consistently observed, one at
352 base pairs, corresponding to the complete mouse hypothalamic GnRH
cDNA, and a second more prominent species which was 142 base pairs
shorter, at 210 base pairs long (Fig. 2). A Southern
blot of the same agarose gel was performed and hybridized to the
radiolabeled mouse hypothalamic GnRH cDNA probe. Both species of
transcripts hybridized to the radiolabeled GnRH probe. In contrast,
only the 352-bp transcript was present in the NLT cells (Fig. 2). This
result suggests that in addition to the expression of the mature
proGnRH mRNA which encodes GnRH and GAP, a short transcript is
abundantly expressed in Gn11 but not in NLT cells.
Alternative splicing of the primary proGnRH gene
transcript has been demonstrated in various peripheral tissues
(12-15). To test whether alternative splicing of the primary proGnRH
transcript may account for the occurrence of the two different species
of transcripts found in the Gn11 cells, these two species of
transcripts were isolated and cloned into the pGEM-T vector (Promega).
The cloned plasmids were then subjected to NcoI and
PstI restriction enzyme digestion to release the inserted
fragments for agarose gel electrophoresis. Southern blot analysis was
performed with a probe that contained the mouse hypothalamic proGnRH
cDNA sequence (exons 1-4). As expected, this probe hybridized to
both the larger and smaller inserts (Fig. 3, left
panel). However, when a second probe which contained only the exon
2 sequence of the mouse proGnRH gene was hybridized to the same
Southern blot, only the larger cDNA species hybridized to this
probe (Fig. 3, right panel). The lack of hybridization of
the smaller insert with the exon 2 probe indicates the absence of exon
2 sequence in this cDNA obtained from Gn11 cells.
To confirm the above findings, these two clones were sequenced and the
encoding nucleic acids deduced. As expected, sequencing revealed that
the larger transcript expressed in NLT and Gn11 cells contains exons 1, 2, 3, and 4 of the proGnRH gene and that introns A, B, and C are
spliced out from the primary transcript (Fig.
4a). The ATG (designated
To investigate whether alternative splicing of the primary
GnRH gene transcript in Gn11 cells may result in a decrease in the
level of mature proGnRH mRNA, RNase protection assays were performed to quantify proGnRH mRNA levels in both NLT and Gn11 cells. 32P-labeled antisense mouse proGnRH mRNA probe
was obtained by in vitro transcription of a cDNA
template containing exons 1, 2, 3, and 4 of the mouse proGnRH gene. As
a control, the actin probe was also prepared at the same time. As shown
in Fig. 5, the NLT cells express high levels of proGnRH
mRNA in the RNase protection assay. With the assay sensitivity at
50 fg/well level, proGnRH mRNA in Gn11 cells was undetectable,
although RT-PCR revealed its expression in this cell line. In contrast,
the actin mRNA levels were comparable between the two cell lines.
This finding suggests that the low level of GnRH secreting capacity in
Gn11 cells is due, in part, to decreased expression of the proGnRH mRNA.
Although alternative splicing of the primary GnRH gene
transcript has been demonstrated in various peripheral tissues, little is known about whether this phenomenon also occurs in the animal brain.
To investigate this possibility, brain tissues from the mouse
olfactory, preoptic area-anterior hypothalamus or caudal hypothalamus
were isolated for RNA extraction, and RT-PCR reactions of the RNA were
performed. The PCR products were separated by agarose gel
electrophoresis and blotted to a nylon membrane. Similar to the Gn11
cells, hybridization with a probe that contains the entire proGnRH
cDNA revealed the presence of both species of transcripts in the
olfactory and preoptic area-anterior hypothalamus (Fig. 6, A and B, left), respectively,
with the larger transcript corresponding to the mature proGnRH
mRNA. However, the caudal hypothalamus expressed only the mature
proGnRH mRNA (Fig. 6C, left). The relative amount of the
proGnRH mRNA versus the short splicing variant was
estimated at approximately 20:1 by quantification with a
PhosphorImager. When an exon 2 probe was used, only the proGnRH
mRNA was detected in the olfactory, preoptic area-anterior
hypothalamus and the caudal hypothalamus (Fig. 6, A-C,
right). The absence of hybridization of the smaller transcript in
Fig. 6, A and B, to the exon 2 probe indicates
that exon 2 was spliced out in this species of the transcript. This
result has been confirmed in two independent experiments. Therefore,
these data indicate that alternative splicing of the proGnRH primary
transcript also occurs in the mouse brain. In the caudal hypothalamus,
processing of the primary proGnRH transcript gives rise to the mature
proGnRH mRNA, which is then translated into proGnRH for subsequent
processing to GnRH and GAP. In the olfactory and preoptic area-anterior
hypothalamus, the primary transcript is alternatively processed to
generate either proGnRH mRNA or the shorter splicing variant.
Our data indicate that the NLT and Gn11 cells are neuronal in
phenotype, express proGnRH mRNA, and secrete GnRH into the medium. Since one major difficulty in the study of GnRH neuronal activities is
their low abundance and scattered distribution, these cell lines
provide a convenient in vitro model for the study of GnRH secretion and regulation of gene expression. Using the Gn11 and NLT
cell lines as a model, we found that although both cell lines were
derived from a same tumor in a transgenic mice, they display heterogeneity in the secretion of GnRH and expression of proGnRH gene.
Our data provide evidence that the lower GnRH secreting capacity found
in the Gn11 cells is due, in part, to the prevalence of a splicing
variant lacking exon 2 which encodes the GnRH decapeptide. Moreover, we
extended this finding to the animal forebrain by demonstrating that
this splicing variant is also expressed in the olfactory and preoptic
area/anterior hypothalamus, although the prevalent form of transcripts
in these areas is the mature proGnRH mRNA. Therefore, these results
suggest that alternative splicing of the primary transcript may
contribute to the regulation of proGnRH gene expression in the animal
forebrain.
Various lines of evidence indicate that proGnRH gene expression is
regulated in a tissue-specific pattern. In various peripheral tissues,
including the placenta, mammary gland, testes, and ovary (12-14, 17),
the expression and processing of the primary transcript is
significantly different from that in the hypothalamus in that a greater
proportion of the transcripts in these tissues contains intron A. Moreover, a major transcriptional start site upstream of that used in
the hypothalamus is utilized for transcription of the proGnRH gene in
these tissues (14, 17). In the immature T lymphocyte cell line Nb2, in
addition to the mature GnRH mRNA, the alternatively spliced form of
transcript is also found, which lacks exon 2 that encodes the GnRH
decapeptide (15). Because the splicing process affects the coding
capacity of transcripts directly, its efficiency and accuracy are
obviously critical for normal functions of GnRH neurons. Our present
data demonstrate that even among GnRH neurons located in different
regions of the central nervous system, the primary transcript is
differentially processed. For GnRH neurons located in the olfactory
area where Gn11 and NLT cells were derived, and those in the preoptic
area-anterior hypothalamus, in addition to the prevalent mature
mRNA, a short splicing variant lacking exon 2 of proGnRH gene is
also produced, whereas only the mature proGnRH mRNA is generated by
GnRH neurons located in the caudal hypothalamus. This finding is in
agreement with the emerging evidence that the expression of proGnRH
gene is differentially regulated in different regions of the central nervous system. For example, in the human brain, in situ
hybridization studies suggest the presence of three distinct subtypes
of GnRH neurons with pronounced differences in morphology, labeling
density, and location (18). The number of GnRH neurons detected by
in situ hybridization and immunocytochemistry has also been
shown to vary in response to ovariectomy (19, 20), steroid treatment (21), and during the estrous cycle (22, 23). Recent studies of the rat
preovulatory luteinizing hormone surge demonstrate that the number of
GnRH-expressing cells fluctuates during the periovulatory period, and
peak numbers of GnRH-expressing cells are attained at different time
points in the preoptic area versus GnRH neurons in the more
rostral regions (24). Taken together, these studies support the
presence of heterogeneous populations of GnRH neurons in the mammalian
central nervous system. Our data suggest that alternative splicing may
be one of the mechanisms by which heterogeneous populations of GnRH
neurons can be generated in the animal brain.
Studies of RNA splicing mechanisms suggest that differences in the
activities/amounts of general splicing factors or the presence of
specialized proteins may participate in the regulation of alternative splicing (25). In addition, a number of cis-acting sequences that
influence splice site recognition have been identified, which include
intron size, exon sequence, alternative branch points, pyrimidine
content of 3 Although the shorter transcript found in the Gn11 cells and in GnRH
neurons in the forebrain lacks the normal ATG translation initiation
codon located in exon 2, examination of exon 3 sequence indicates that
another ATG translation initiation codon is present in exon 3. The
methionine at position +25 of the proGnRH peptide may act as a
translation initiation signal for the remaining 45 amino acids of the
GAP. However, the Kozak sequence (16), which is preserved as a purine
(G) located at 3 nucleotides upstream of the ATG in the larger
transcript, is not present in the smaller transcript. Whether this
splicing variant is translated into a protein in Gn11 cells and the
translational efficiency has yet to be determined.
Since the NLT cells express higher levels of mature proGnRH mRNA
and secrete larger quantities of GnRH, these cells have the advantage
over the Gn11 cells for studying the regulation of proGnRH gene
expression and GnRH secretion by various factors. Recent studies from
this laboratory using the NLT cell line have demonstrated that these
cells express type-I receptor for insulin-like growth factor (27) and
the receptor for epidermal growth factor (28). These findings are
interesting since both insulin-like growth factor-1 and epidermal
growth factor have been shown to regulate GnRH neuronal activities
in vivo and have been suggested to play an important role in
the control of pubertal development (29, 30). However, to date, the
molecular mechanisms by which these growth factors regulate GnRH
neuronal activities remain largely unknown, due to the fact that less
than 1500 GnRH neurons are present in the animal brain (4, 5).
Therefore, the availability of NLT cells should greatly enhance our
ability to explore the molecular mechanisms that mediate the regulation
of proGnRH gene expression and GnRH secretion.
We thank Drs. Andrew Wolfe, Laurie
Cohen, and Marjorie Zakaria for comments on the manuscript, Eric Su for
technical assistance, and Drs. William Chin and Rupert Yip in the
Division of Genetics/Brigham & Women's Hospital for support of the
immunocytochemistry.
Department
of Neurobiology and Physiology, Northwestern University, Evanston,
Illinois 60208, and the
Division of Development and
Reproduction, Roslin Institute (Edinburgh), Roslin,
Midlothian, Scotland EH25 9PS
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-upstream regulatory sequence resulted in the development of an olfactory tumor in one of
the transgenic mice (7). Two GnRH immunoreactive cell lines (Gn11 and
NLT) were subsequently derived from this tumor. Characterization of the
NLT and Gn11 cells demonstrated that these cells display neuronal
morphology and neuron-specific markers, such as microtubule-associated
peptide 2 (MAP-2) and Tau protein (8, 9). Solution hybridization-RNase
protection assays and RT-PCR analysis indicated that these cells
express proGnRH mRNA and are able to synthesize and secrete GnRH as
demonstrated by RIA. Therefore, these cell lines provide a suitable
in vitro model for the study of GnRH neuronal activity and
its regulation.
Cell Culture
80 °C for the GnRH RIA.
80 °C before RNA
extraction.
primer sequence was:
5
-GAAGTACTCAACCTACCAA-3
, and the 3
primer was 5
-
GCCATAACAGGTCACAAGCCTC-3
.
-32P]dCTP labeled by
using the Klenow enzyme and random hexanucleotide primer (Boehringer
Mannheim).
80 °C for 1 day.
-32P]UTP and SP6 polymerase to generate the antisense
mouse GnRH probe. As a control, mouse actin antisense probe was also
synthesized. Solution hybridization was performed by mixing sample RNA
(30-40 µg) with 4-6 × 104 cpm of
[
-32P]UTP-labeled riboprobe in a final volume of 30 µl of hybridization solution. Hybridization was carried out at
45 °C overnight, and the mixture was digested with RNase at 37 °C
for 1 h. The hybridized RNA fragments were precipitated,
resuspended in 8 µl of gel loading buffer, and heated at 95 °C for
5 min before electrophoresis in a 6% denaturing polyacrylamide gel.
The amount of radioactivity in the samples was counted using a
PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
The NLT Cells Secrete More GnRH Peptide than the Gn11
Cells
1131 to +5 bp fused to
the SV40-Tag coding region was injected into fertilized one-cell embryos to generate transgenic mice. An olfactory tumor containing Tag-
and GnRH-immunopositive neurons developed in an F1 male. The tumor was
dispersed with collagenase and cultured. Through serial dilution and
cloning, a number of GnRH-immunopositive cells were obtained from this
tumor. Immunocytochemical double labeling indicted the expression of
both Tag and GnRH within these cells (7). In the present study, further
characterization of two of these cell lines, referred to as NLT and
Gn11, was performed.
Fig. 1.
Gn11 and NLT cells express neuron-specific
markers and secrete GnRH. a, immunocytochemical staining of
MAP-2 and Tau proteins in Gn11 and NLT cells. The primary antibody
mixture contained either the polyclonal anti-MAP-2 or anti-Tau antibody
(both at 1:500 dilution). The secondary antibody was a goat anti-rabbit IgG-biotinylated and streptavidin-covalently-coupled to the horseradish peroxidase at dilution of 1:250. A and B, MAP-2
and Tau immunoreactivities, respectively (stained dark pink)
in Gn11 cells; C and D, MAP-2 and Tau
immunoreactivities, respectively, in NLT cells. Magnification: × 100. b, Northern blot analysis of Tau mRNA expression in Gn11 and NLT cells. Ten micrograms of RNA from Gn11, NLT, and CV-1 cells
were electrophoresed, and the positions of 28 and 18 S ribosomal RNA
are shown on the left. After transferring to a
nitrocellulose membrane, the RNAs were hybridized with a
32P-labeled Tau cDNA probe. The position of Tau
mRNA is shown on the right. c,. RIA
measurement of GnRH concentrations in culture media collected from
dishes containing either NLT or Gn11 cells. GnRH values were expressed
as picograms/ml/µg of protein. *, p < 0.05 as
compared with Gn11, Student's unpaired t test.
[View Larger Version of this Image (40K GIF file)]
Fig. 2.
A splicing variant of the proGnRH gene
transcript is present in Gn11 cells. RT-PCR was performed to study
GnRH gene expression in the Gn11 and NLT cells. Two different
transcripts were consistently seen in the Gn11 cells, with one at 352 bp corresponding to the mature proGnRH mRNA and a second species at
210 bp long. In contrast, only the 352-bp transcript was observed in
NLT cells. A Southern blot of this agarose gel was performed and
hybridized to the radiolabeled mouse hypothalamic proGnRH cDNA
probe. Both species of transcripts in Gn11 cells hybridized to the
radiolabeled probe. In contrast, only the 352-bp transcript was seen in
the NLT cells.
[View Larger Version of this Image (24K GIF file)]
Fig. 3.
The splicing variant in Gn11 cells lacks exon
2 of the proGnRH gene. The two different transcripts shown in Fig.
2 were cloned into the pGEM-T vector. These plasmids were digested with
NcoI and PstI and separated by agarose gel
electrophoresis. Southern blotting was performed, and the membrane was
subjected to hybridization with a probe containing the mouse proGnRH
cDNA sequence (exons 1-4). This probe hybridized to
both the larger and smaller inserts. However, when a second probe
containing the exon 2 sequence was used, only the larger insert
hybridized to this probe. The smaller insert showed no hybridization
signal, indicating a lack of exon 2 sequence in the smaller
cDNA.
[View Larger Version of this Image (18K GIF file)]
21 methionine in
the proGnRH peptide) in exon 2 of the proGnRH gene is used for
translation initiation (Fig. 4b). However, in the smaller
transcript present only in the Gn11 cell, exon 2 together with introns
A and B are spliced out. Therefore, the shorter splicing variant in
Gn11 cells contains exons 1, 3, and 4, but does not have exon 2 which
encodes the signal sequence of GnRH, the GnRH decapeptide, and the
first 11 amino acids of GAP. The ATG (designated +25 Met) in exon 3 may be used for translation initiation. However, the Kozak sequence (16),
which is a purine (G) at 3 nucleotides upstream of the ATG in the
proGnRH mRNA (AGTGG/ACATG), is not present in the short splicing variant (CAAACAGATG).
Fig. 4.
Processing of the proGnRH gene primary
transcript in Gn11 and NLT cells. a, a schematic diagram of
the amino acid sequence encoded by the mature proGnRH mRNA. The
relative positions of the signal peptide (21 to
1), GnRH
decapeptide (+1 to +10, underlined), and GAP (+14 to +69)
are shown. b, the larger transcript present in both NLT and
Gn11 cells, and the short splicing variant present only in Gn11 cell,
were cloned into the pGEM-T plasmid. The two cDNA clones were
sequenced by Sanger's method using T7 and SP6 primers. In Gn11 and NLT
cells, processing of the proGnRH gene primary transcript gives rise to
the mature proGnRH mRNA containing exons 1, 2, 3, and 4. The ATG in
exon 2 at
21 is used for translation initiation. In Gn11 cells, in
addition to the mature proGnRH mRNA, processing the primary
transcript yields a short splicing variant in which intron A, exon 2, and intron B are spliced out.
[View Larger Version of this Image (21K GIF file)]
Fig. 5.
Quantification of proGnRH mRNA levels in
NLT and Gn11 cells by RNase protection assay. Template cDNA
containing exons 1, 2, 3, and 4 of the mouse proGnRH gene was inserted
in the pGEM-T plasmid between the NcoI and PstI
sites. The template plasmid was linearized with restriction enzyme
NcoI, and in vitro transcription was performed by
using [-32P]UTP and SP6 polymerase to generate
antisense mouse GnRH probes. As a control, mouse actin antisense probe
was also synthesized. The riboprobes were hybridized with 30-40
µg/tube of total cellular RNA obtained from either NLT or Gn11 cells.
The hybridized RNA fragments were precipitated, electrophoresed in a
6% denaturing polyacrylamide gel, and quantified with a
PhosphorImager. The positions of the proGnRH and actin mRNA signals
are indicated.
[View Larger Version of this Image (63K GIF file)]
Fig. 6.
Expression of the short splicing variant of
the proGnRH gene transcript in the mouse forebrain. Total RNA was
extracted from the mouse olfactory, preoptic area-anterior
hypothalamus, and caudal hypothalamus, respectively. After RT-PCR and
agarose gel electrophoresis, Southern blot hybridization was performed with a probe that contains the proGnRH cDNA. Both the larger and shorter transcripts were present in the olfactory (A) and
preoptic area-anterior hypothalamus (B). However, the caudal
hypothalamic area (C) expressed only the larger transcript.
In the right panel, an exon 2 probe detected only the
full-length proGnRH mRNA in the olfactory and preoptic
area-anterior hypothalamus.
[View Larger Version of this Image (18K GIF file)]
acceptor sites, and secondary structure of the
pre-mRNA (26). The difference in processing the primary proGnRH
gene transcript between the Gn11 versus NLT cells and between the GnRH neurons in the caudal hypothalamus versus
those in the olfactory and preoptic areas may reflect the differential capacity of different GnRH neurons to recognize exon 2 in the proGnRH
pre-mRNA. In the Gn11 cells and GnRH neurons located in the
olfactory area and preoptic area-anterior hypothalamus, the splicing
machinery either recognizes exon 2 in the primary transcript to
generate the mature mRNA or ignores it, resulting in the formation of a shorter species of transcript lacking exon 2. Whether the difference in splicing of the primary transcript is due to the presence
or absence of specific splicing factors in the Gn11 cells versus NLT cells or due to the difference in the activities
in general splicing factors assembled in the spliceosomes inside the
nucleus remains to be determined.
*
This work was supported by National Institutes of Health
Grant HD30040 and by American Cancer Society Grant DB-73786 (to S. R.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
**
To whom correspondence should be addressed: Children's Hospital,
Division of Endocrinology, Enders Bldg., Rm 407, Boston, MA 02115. Tel.: 617-355-6136; Fax: 617-730-0741.
1
The abbreviations used are: GnRH,
gonadotropin-releasing hormone; GAP, GnRH-associated peptide; RT-PCR,
reverse transcriptase-polymerase chain reaction; SV40-Tag, simian virus
40 T antigen; MAP-2, microtubule-associated protein-2; RIA,
radioimmunoassay; bp, base pair(s).
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.