(Received for publication, March 11, 1997)
From the Wellcome Laboratory for Molecular Pharmacology, Department of Pharmacology, University College London, London WC1E 6BT, United Kingdom
The m1 receptor is one of five muscarinic
receptors that mediate the metabotropic actions of acetylcholine in the
nervous system where it is expressed predominantly in the telencephalon and autonomic ganglia. RNase protection, primer extension, and 5-rapid
amplification of cDNA ends analysis of a rat cosmid clone containing the entire m1 gene demonstrated that the rat m1 gene consists of a single 657-base pairs (bp) non-coding exon separated by a
13.5-kilobase (kb) intron from a 2.54-kb coding exon that contains the
entire open reading frame. The splice acceptor for the coding exon
starting at
71 bp relative to the adenine of the initiating
methionine. This genomic structure is similar to that of the m4 gene
(Wood, I. C., Roopra, A., Harrington, C. A., and Buckley, N. J. (1995)
J. Biol. Chem. 270, 30933-30940 and Wood, I. C.,
Roopra, A., and Buckley, N. J. (1996) J. Biol. Chem. 271, 14221-14225). Like the m4 gene, the m1 promoter lacks TATA and
CAAT consensus motifs, and the first exon and 5
-flanking region are
not gc-rich. The 5
-flanking region also contains the consensus
regulatory elements Sp-1, NZF-1, AP-1, AP-2, E-box, NF
B, and Oct-1.
Unike the m4 promoter, there is no evidence of a RE1/NRSE silencer
element in the m1 promoter. Deletional analysis and transient
transfection assays demonstrates that reporter constructs containing
0.9 kb of 5
-flanking sequence and the first exon are sufficient to
drive cell-specific expression of reporter gene in IMR32 neuroblastoma
cells while remaining silent in 3T3 fibrobasts.
G-protein coupled receptors are responsible for mediating a vast
amount of intercellular communication throughout the body, especially
in the nervous system. Current estimates put the size of this gene
superfamily well in excess of 1000, making it one of the largest in the
mammalian genome. In situ hybridization studies indicate
that each individual member probably has a unique expression profile
within the nervous system, yet the factors that determine and direct
the expression patterns of the members of this gene family are largely
unknown. Since the response of any neuron to a neurotransmitter is
determined by the repertoire of receptors expressed at the cell
surface, then it is essential to understand the mechanisms that
determine the types of receptor gene expressed by individual neurons.
The five muscarinic receptors are encoded by a subfamily of this gene
superfamily (1, 2), and their gene products are reponsible for
mediating the metabotropic actions of acetycholine in the nervous
system and its effector tissues (3). Each of the five muscarinic
receptor genes is differentially expressed throughout the central and
autonomic nervous systems both in adulthood (4, 5) and during embryonic development,1 and each of the receptors
exhibits a unique pharmaclogical profile (6, 7). The m1 receptor is the
most abundant subtype in both the central and autonomic nervous systems
and is found predominantly in the telencephalon (4), autonomic ganglion
cells (4, 8, 9), and exocrine tissue (10). Activation of the m1
receptor leads to numerous responses, including stimulation of
phospholipases C (7) and A2 (11), inhibition of cAMP
production (5), activation of K+ and Cl
channels (12), and inhibition of opening of K+M
channels in neural cell lines, and sympathetic and hippcampal neurones
(13, 14). We have previously described the structure of the rat m4 gene
and its promoter (5, 15-17) and have shown that expression is silenced
in non-neuronal cells (16) via a RE1/NRSE (18, 19) type silencing
element. There are many locations, such as the cerebral cortex,
hippocampus, striatum, and autonomic ganglia, that co-express m1 and m4
receptor genes, and equally important in view of the silencing of the
m4 promoter, there are many locations that express neither the m1 nor
m4 genes, such as most regions of the mesencephalon and rhombencephalon
and most non-neuronal tissue. Yet even within areas that co-express m1 and m4, not all individual cells express both m1 and m4 genes. There
is, therefore, an intimate matrix of overlapping and nonoverlapping expression profiles between these homologous family members. We were
thus interested to identify the control regions of the m1 promoter to
ascertain if the m1 and m4 genes shared a common gene structure and
regulatory elements. As a first step to addressing this issue, in the
present study, we present the first description and analysis of the
structure of the m1 muscarinic receptor gene and its promoter.
A rat cosmid library in pWE15
(Stratagene) was screened using a mixture of three
[-32P]dATP tailed coding region oligodeoxynucleotides
(Rm1A, GGCCACCGTCCAGGGACCCTTTCCTGGTGCCAAGACAGTGATGTTGGG; Rm1B,
ACTCAGGGTCCGAGCTGCCTTCTTCTCCTTGACAGTGAGAAGGTCTT; and Rm1C, GGGCGCTTGGGGATCTTGCGCCAGCGCCTCTTGTCCCAGCGGCAGAGC).
Filters were hybridized overnight at 37 °C in buffer
containing 4 × saline/sodium/phosphate/EDTA (150 mM
NaCl, 10 mM NaH2PO4, 1 mM EDTA), 5 × Denhardt's reagent (0.1% Ficoll,
0.1% bovine serum albumin, 0.1% polyvinylpyrrolidone), 250 mg/ml
yeast tRNA, 500 mg/ml denatured salmon sperm DNA, and 0.1% SDS (1 × 106 dpm/ml) before washing at 55 °C in 1 × saline/sodium/phosphate/EDTA, 0.1% SDS, four times for 15 min each.
Filters were then exposed to x-ray film at
70 °C for a week.
Positive colonies were mapped using probes based upon
RACE2 sequence of the upstream
exon.
Reverse transcription-PCR
analyses were carried out as described previously (16). RNA was treated
with RQ DNase I (Promega) prior to reverse transcription to ensure that
subsequent amplification products were derived from RNA and not
contaminating genomic DNA. PCR cycling conditions were (95 °C/30 s;
63 °C/30 s; 72 °C/1 min). Reactions were run for 30 cycles, and
amplified products were separated by electrophoresis through 2%
agarose gels. The primers used for PCR analysis were: species
degenerate primers that amplified rat, human, and mice m1, Rm1 1069s
(5-CTGGTCAAGGAGAAGAAGGCAGCT-3
) and Rm1 1518a
(5
-GTCTCTCTGGGCTGTCCAGGAAGG-3
). To specifically amplify rat m1
sequences, a set of rat specific antisense primers, Rm1 1256a
(5
-TCCCGGAAGGCTTTGTTGCACAGT-3
) and Rm1 1069s, were used. Primers used
to amplify hypoxanthine-guanine phosphoribosyl transferase, (hprt),
were; hprt 231s (5
-CCTGCTGGATTACATTAAAGCACTG-3
) and hprt 567a
(5
-CCTGAAGTACTCATTATAGTCAAGG-3
). PCR analysis of m1 genomic DNA and
rat brain cortex cDNA was performed using the primers 696s, 539s,
461a, 235a, and 132a, all of which are shown in Fig. 1. Cycling
conditions were: 95 °C/5 min followed by 30 cycles of 95 °C/30 s,
55 °C/30 s, 72 °C/60 s, and then 72 °C/10 min. Resulting
amplified products were analyzed on a 2% metaphor gel.
Organization of the rat m1 muscarinic
receptor gene. Panel A shows the relative sizes and
positions of the coding (Ex2) and non-coding
(Ex1) exons on a fragment of the S1-239 cosmid clone while
panel B shows an enlargement of the 5-flanking sequence. B, BamHI; E, EcoRI;
H, HindIII; K, KpnI;
N, NcoI; P, PstI;
S, SalI; Sa, SacI.
Panel C shows the sequence of 1.6 kb of 5
-flanking sequence
and the 657-bp noncoding exon. Positions of HindIII, KpnI, EcoRI, PstI, and SacI
sites are underlined. The position of the
H/P1.8 deletion
is also indicated. Positions of consensus regulatory elements Sp-1,
NZF-1, AP-1, AP-2, E-box, NF
B, and Oct-1 are boxed.
Dashed underlines indicate primers used in PCR analysis.
Transcription start sites identified by RNase protection are indicated
by asterisk; the two guanine residues representing the dominant sites (*) are double underlined,
and the first is assigned position +1. The extension product derived
from primer extension studies is indicated by (+). 5
-flanking sequence
defined from assigning the guanine at +1 is in uppercase
while the non-coding exon is shown in lowercase.
Underlined sequence represents the extent of the longest
5
-RACE cDNA clone.
5
Total RNA was extracted from rat cerebral cortex
according to Chomczynski and Saachi (20) and reverse transcribed using either AMV-RT (Seikakagu) or Tth polymerase (Promega) and a mixture of
random hexamers and oligo(dT). Some reverse transcriptions were carried
out in the presence of methylmercuric hydroxide to denature the RNA.
Reverse transcriptions with Tth polymerase were carried out by
incubating 2 µg of rat cerebral cortex RNA, 200 ng of random
hexamers, and reaction buffer (Promega) for 5 min. 1.8 mM
dNTPs, 1 mM MnCl2, and 6 units of Tth
polymerase (Promega) were then added, and the reaction was allowed to
proceed at 42 °C for 5 min followed by 75 °C for 10 min. 5-RACE
was then carried out essentially according to Frohman (21) using
Taq polymerase and two nested gene-specific primers
(Rm1-7a, ACTGACAGCAGGGGGCACTGAGGT, and Rm1-62a,
ACAGGTTTTTCCTCAGAGAAA) to improve specificity in subsequent rounds of
PCR, and the final PCR products were cloned directly into pCRII
(Invitrogen). DNA from individual colonies was sized and sequenced.
All sequencing was performed using Taq polymerase, flouresceinated dye terminators, and an Applied Biosystems automated DNA sequencer.
RNase Protection AnalysisDNA fragments from the m1 gene
spanning the regions +136 to +527, 805 to +189, and
805 to +418
were generated by PCR and cloned into pCR3 (Invitrogen). Plasmids
containing the appropriate orientation of insert were digested with
HindIII and gel purified. Antisense RNA probes were
synthesized using Sp6 RNA polymerase (Promega) and labeled with
[
-32P]UTP (3000 Ci/mmol, Amersham Life Sciences). The
reaction was run on a 4% polyacrylamide gel and exposed to x-ray film,
full-length probe was excised from the gel, and the gel slice was
incubated overnight in 10 mM Tris, 1 mM EDTA
(pH 7.4), 0.1% SDS to elute the probe. Approximately 1 × 105 dpm of probe was precipitated with RNA and hybridized
overnight in 30 µl of 80% formamide, 400 mM NaCl, 40 mM PIPES (pH 6.8), 1 mM EDTA. Nonhybridized RNA
was digested with RNase A/RNase T1, purified, and analyzed on a 6%
denaturing polyacrylamide gel. A DNA sequencing reaction performed
using Sequenase 2 (U. S. Biochemical Corp.) was run as a size
marker.
Reporter plasmids were
constructed using the luciferase reporter vector, pGL3 basic (Promega).
A 9-kb BamHI fragment including the whole of the non-coding
exon and flanking regions (Fig. 1) was cloned into pGEM3Zf+
(Promega), cut with PstI, and religated to generate
B/P4.0pGEM3Zf+. This in turn was used to generate a 2-kb
HindIII fragment that was cloned into the HindIII
site of pGL3 (Promega) to yield H/P-1.4/+0.6pGL3. A 0.75-kb
Acc65I/EcoRI fragment was excised from
B/P4.0pGEM3Zf+ and cloned into
Acc65I/EcoRI cut H/P-1.4/+0.6pGL3 to yield
K/P-1.6/+0.6pGL3. Exonuclease digestion (Erase-A-Base, Promega) of
H/P-1.4/+0.6pGL3 was used to generate H/P-1.3/+0.6.
S/P-0.88/+0.6pGL3 was generated by excising a 0.5-kb SacI
fragment from H/P-1.4/+0.6pGL3 followed by religation. Qiagen
column-purified DNA (routinely, 250 ng of DNA/10-mm well) was
transfected into cells using lipofectamine (Life Technologies, Inc.).
Transfections were carried out in 10-mm wells using 0.25 µg of DNA
and 1.25 µl of LipofectAMINE applied for 4-16 h. These conditions
gave transfection efficiencies of 5-50% as judged by
-galactosidase histochemistry. One day after transfection, cells
were harvested into Reporter Passive Lysis Buffer (Promega). Luciferase
measurements were carried out using the Promega dual luciferase assay
system according to manufacturer instructions in a Turner TD-20e
luminometer. Cells were cotransfected with pRL-CMV (0.3 ng of DNA/10-mm
well, representing a 1:250 ratio of pRL:test construct) that contains
the Renilla cDNA driven by the CMV promoter and
luminescence measured after quenching luciferase luninescence and
adding Renilla substrate with Stop and Glow (Promega) according to manufacturer instructions. Luciferase luminescence was
then normalized to Renilla luminescence, and results
expressed relative to normalized luminescence driven from the
promoterless pGL3 basic.
IMR32 cells, NBOK1 neuroblastoma, and 3T3 fibroblasts were cultured in 5% CO2 at 37 °C in minimal essential medium (IMR32), in RPMI 1640 (NBOKl), or Dulbecco's modified Eagle's medium (3T3) containing 6 g/l penicillin, 10 g/l streptomycin, and 2 mM L-glutamine, supplemented with 10% fetal calf serum.
Screening of 106
recombinants in a rat cosmid library with coding region primers yielded
a single colony, S1-139. The same colony hybridized to
oligonucleotides derived from 5-RACE cDNA sequence.
Comparison of the genomic sequence upstream of the open reading frame with cDNA sequence demonstrated sequence divergence 70 bp upstream of the adenine of the initiation codon. Inspection of the genomic sequence identified a consensus splice acceptor site at this position, CCTCTCTTTTCA(G/G).
Since repeated screening of cDNA libraries failed to generate any
significant upstream cDNA sequence information, 5-RACE was used to
generate clones corresponding to the 5
-untranslated region of the rat
m1 cDNA. Use of AMV-RT generated clones that routinely terminated
less than 60 bp upstream of the splice acceptor site. Supplementation
of the PCR with dimethyl sulfoxide, formamide, or glycerol failed to
yield any longer clones, and neither did the use of MeHgOH to denature
the RNA prior to reverse transcription. However, use of the
thermostable polymerase Tth (Promega) for the reverse transcription
yielded several longer clones, which were used to generate the
underlined sequence seen in Fig. 1. A long
polypyrimidine tract stretches for about 88 bp between positions +415
and +503 (see Fig. 1), and this may serve as a premature
transcriptional stop during reverse transcription. Interestingly, around this point, the sequence of the rat 5
-untranslated region diverges sharply from its porcine homologue (22). Homology downstream from this point is very high (88% over 294 bases to the initiation codon although in the porcine sequence, the consensus splice site lies
further upstream at
361 with respect to the initiation codon (22).
Both the position of the splice site and homology of the coding exon
are highly conserved between rat and mouse with the exception of a
single base change - CCTTTCTTTTCA(G/G) (23).
Before proceeding to a finer analysis of the promoter of the m1 gene, we first wished to establish whether the S1-239 cosmid contained sufficient information to direct expression of the m1 gene by transfecting the entire cosmid into the human m1 expressing neuroblastoma, NBOK1. Reverse transcription-PCR, using primers derived from noncoding and coding exons specific for the rat m1 sequence, revealed the presence of rat m1 transcripts in the transfectants (data not shown), thus verifying that the cosmid contained all information necessary to drive expression, at least in transient transfection assays. No rat m1 transcripts were detected in untransfected NBOK1 cells.
Identification of the Transcriptional Start SiteTwo
independent strategies were used to identify the position of the
transcription start site of the m1 gene. Initially, primer extensions
were performed on rat brain cortex poly(A)+ RNA using the
two primers, 132a and 235a (see Fig. 1). Despite using several
protocols and different reverse transcriptases, no product was obtained
using the 132a primer. We did, however, obtain a 171-nucleteotide
product using the 235a primer in conjunction with AMV reverse
transcriptase, the length of this extended product corresponding to
position +287 (Fig. 1, indicated by +). To complement the primer
extension, we performed RNase protection experiments using antisense
RNA probes shown in Fig. 2. Using probe A,
the 392-nucleotide protected fragment represents protection of the entire m1 sequence within this probe (each of the probes contain plasmid vector sequence in addition to the m1 sequence shown in Fig. 2,
probe A), indicating that transcription begins upstream of
the 5 end of this probe. Using probe B, two protected bands of 184 and 186 nucleotides were observed, correponding to transcription initiation sites of +1 and +3 (Fig. 1, shown as asterisk and
double underline). Probe C generates a protected
band of 418 nucleotides, which corresponds to the same transcription
initiation site as revealed using Probe B. The remaining two
major protected fragments from probes A and C are
thought not to represent transcription start points but to be
probe-specific artifacts as their 5
ends (Fig. 1, shown as
asterisk) do not correlate with each other. No protected
bands were seen when using yeast tRNA as template. All of these sites
were upstream of the site identified by the primer extension
experiments, leading us to conclude that the latter was probably an
artifactual transcriptional stop and that the upstream sites identified
by RNase protection were more likely to represent true transcriptional
start sites.
To further verify that transcription of the m1 gene initiates at +1 and
the 1st exon does not contain any introns, a comparative PCR between
cDNA from brain cortex and genomic DNA was performed. cDNA was
generated from rat brain cortex RNA that previously had been treated
with DNase to remove genomic DNA. Two different sense primers were
used, 696s, which contains sequence upstream of the proposed
transcription start site, and 539s, which contains sequence downstream
of the proposed transcription start site. Each of the sense primers was
used in PCR in conjunction with the three antisense primers 461a, 235a,
and 132a (the positions of each of the primers is shown in Fig. 1). The
PCR products obtained from this analysis are shown in Fig.
3. Primer pairs 539s/46a, 539s/235a, and 539s/132a generated 71-, 303-, and 412-bp amplified products, respectively, from
both H/P-1.4/+0.6pGL3 and cDNA, indicating that the intervening sequence is exonic and contains no introns. However, primer pairs 696s/46a, 696s/235a, and 696s/132a generated 231-, 461-, and 568-bp amplified products only when H/P-1.4/+0.6pGL3 was used as template. No
signal was seen using cDNA as template, thereby indicating primer
696s must lie upstream of the transcription initiation site. These data
are consistent with the guanine residues at positions +1 and +3 being
the dominant transcription initiation sites. Furthermore, the sizes of
the PCR products obtained using 539s with each of the antisense primers
is the same in cDNA and genomic DNA, indicating that there are no
introns between 539s and 132a.
Sequence Analysis of the 5
Inspection of the sequence of the upstream exon and
5-flanking sequence reveals consensus binding elements for AP-1, NZF-1 (24), AP-2, Oct-1, and NF
B. No TATA or CAAT consensus elements are
present. The sequence flanking the transcription start site show no
homology with any known initiator sequence. No significant homology
with the 5
-flanking region or the promoter of the m4 gene is found,
nor is it found with any other sequence in the data base.
Reverse transcription-PCR analysis demonstrated the
presence of m1 transcripts in IMR32 cells and their absence in 3T3
fibroblasts (Fig. 4). The four reporter constructs
(K/P-1.6/+0.6pGL3, H/P-1.4/+0.6pGL3, H/P-1.3/+0.6pGL3, and
S/P-0.88/+0.6pGL3) used to assay the promoter activity in these cells
all terminated at the PstI site 65 bp upstream of the splice
site and started 1.60, 1.39, 1.26, and 0.88 kb upstream of the
transcriptional start site, respectively. All four constructs expressed
4-5-fold above background in m1-expressing IMR32 cells, and only the
larger K/P-1.6/+0.6pGL3 construct drove expression significantly above
background in the nonexpressing 3T3 fibroblasts.
G-protein coupled receptors are a diverse and widely expressed family of receptors that mediate signaling throughout the body both in development and adulthood. Within the nervous system, they represent one of the most significant sources of phenotypic diversity, yet little is known of the factors and mechanisms and factors that regulate this cell-specific expression. As such, understanding the mechanisms governing the transcriptional regulation of members of this gene family can offer insight into the establishment and maintainance of specific patterns of gene expression within the nervous system. In the present study, we have shown that, in common with several other members of the G-protein-coupled receptor gene family, including the m4 (15, 16, 17, 24), V1a vasopressin (25), D1a dopamine (26, 27), and C5a (28) receptor genes, the m1 muscarinic receptor gene consists of a single coding exon and a single noncoding exon. Another feature shared with most, but by no means all, other G-protein-coupled receptor genes whose promoters have been examined for genes is the absence of TATA, CAAT, or initiator consensus elements. Examples include the promoters of the 5HT1a (29), 5HT2a (30), 5HT1c (31), serotonin receptors, V1a vasopressin receptor (25), D2 and D1a receptors (26, 27, 32), SSTR1 somatostatin receptor (33), and NPY-1 receptor (34).
Inspection of 1.6 kb of 5-flanking sequence revealed several consensus
regulatory elements including one AP-1 site, two AP-2 sites, two NF
B
sites, an E-box and an NZF-1 element. The latter is an element
recognized by a zinc finger protein that is expressed in the developing
nervous system (24). Since it has been shown that two neuronal
proteins, one NF
B-like and one distinct from NF
B (BETA) (35) can
interact with the NF
B recognition sequence and activate
transcription from the proenkephalin and HIV promoters (36), it will be
interesting to ablate these sites and monitor the effect on expression
of the m1 gene. As with the m4 promoter, no CRE elements are found in
the proximal promoter. The existence of a 88-bp
polypyrimidine/polypurine tract in the noncoding exon between +415 and
+503 is intriguing in light of studies on other promoters such as the
malic enzyme (37), EGF receptor (38), and the mouse c-Ki-ras
(39), which have shown deletion of such tracts to decrease promoter
activity. However the role of the polypyrimidine/polypurine tract
in transcription of the m1 muscarinic receptor gene remains to be
examined.
In our earlier studies, we have shown that the core promoter of the m4 muscarinic receptor gene is constitutively active and cell-specific expression is achieved by silencing expression in non-neuronal (15, 16, 40) via a RE1/NRSE-type silencing element (18, 19). Interestingly, inspection of 2.5 kb of flanking sequence of the m1 promoter reveals no RE1/NRSE element. This observation is corroborated by the failure of a radiolabeled single-stranded RE1/NRSE oligodeoxynucleotide to hybridize to a digest of the m1 cosmid under conditions that generate a strong hybridization signal to digests of the R3-6 m4 cosmid (data not shown). Hence, unlike its m4 couterpart, it is unlikey that the m1 gene is under the control of the zinc finger silencer REST/NRSF (41, 42). A recent report describes the promoter of the chick m2 receptor gene (43), and although there is evidence of a silencer region, there is no evidence of a RE1/NRSE element so it may be that quite different factors are involved in driving transcription of each of the muscarinic receptor genes. Nevertheless, a finer analysis of the proximal m1 promoter will be necessary to examine the activity of the core promoter and the role of silencing and activating elements in directing cell specific expression of the m1 gene.
Deletional analysis revealed that constructs containing between 0.6 and
1.4 kb of 5-flanking sequence and the entire noncoding exon were
sufficient to drive reporter gene expression in IMR32 cells, a
neuroblastoma that expresses an endogenous m1 gene. This cell line
expresses more m1 mRNA than any other cell line that we have thus
far screened, but even so, its level of expression compared with rat
cerebral cortex is low (see Fig. 4), at least as judged by reverse
transcription-PCR. All reporter constructs drove 4-5-fold luciferase
expression relative to the promoterless vector, pGL3 Basic (Fig.
5). This modest stimulation of reporter gene activity is
presumably a refelection of the relatively low levels of endogenous m1
expression in IMR32 cells. Only the K/P-1.6/+0.6pGL3 reporter construct
drove reporter gene expression in 3T3 fibroblasts, which express no
endogenous m1 receptor, showing that constructs containing as little as
0.88 kb of 5
-flank and the non-coding exon are sufficient to drive
cell-specific expression, at least in transient transfection assays.
The low level of expression driven by the K/P-1.6/+0.6pGL3 construct
may indicate the presence of a weak non-neuronal activator between the
KpnI site (
1.6 kb) and the HindIII site (
1.4
kb). Since these deletions ablate the E-box, AP-1, and the distal NZF-1
and NF
B sites, then it is clear that they are not necessary for
cell-specific expression. The role, if any, of the NZF-1, NF
B, and
AP-2 sites between the SacI site and the transcriptional
initiation site await determination by a finer analysis of the proximal
promoter. However, interpretation of all transient transfection assays
is limited by the fact that reporter gene expression is driven by
multiple episomal copies of the reporter vector. Consequently, there
are numerous examples of reporter constructs that are capable of
driving apparent cell-specific expression in transient transfections
that nevertheless fail to recapitulate appropriate cell and/or
stage-specific expression in transgenic mice, as in the case of the
dopamine
-hydroxylase gene where reporter constructs containing 0.6 kb of 5
-flanking sequence can drive cell-specific expression in
transient transfection assays (44, 45) but give no expression in
transgenic mice (46). Reporter gene expression of other
G-protein-coupled receptor promoters has revealed that less than 1 kb
of 5
-flanking sequence is sufficient to drive cell-specific expression
of the V1a vasopressin (25), D1 dopamine (27), type-1 angiotensin II
(47), and m4 muscarinic (15, 16, 17) receptor promoter constructs in
transient transfection assays. However, whether such discrete constructs are capable of driving tissue- and stage-specific expression in transgenic mice has not been reported for any members of the G-protein-coupled receptor gene family.
The characterization of two muscarinic receptor promoters now enables us to examine the differential transcriptional regulation of these members of the G-protein-coupled receptor gene family. Future studies are aimed at dissecting the m1 proximal promoter to determine whether the core promoter is constitutively active, as in the case of the m4 promoter, or whether progressive deletions lead to ablation of expression in m1-expressing cells.
We thank Dr. Magali Waelbroeck (Universite' Libre de Bruxelles) for the NBOK1 cells.