Department of Medicine and Center for Molecular Genetics, University of California, and Veterans Affairs San Diego Healthcare System, San Diego, California 92161
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
hypertension
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In humans, two subtypes of receptors serve as binding sites for angiotensin II: type 1 (AGTR1) and type 2 (AGTR2). AGTR1 receptors are widely distributed in the body and mediate almost all of the known systemic effects of angiotensin II (32), whereas AGTR2 receptors are expressed mostly during fetal development, being much less abundant in adult tissues.
Rodent models serve as useful tools in elucidating the genetic mechanisms determining blood pressure (9, 13). It has been shown that the amino acid sequence of human AGTR1 receptor has a high degree of identity to that of the rat Agtr1 receptor (4). In both rats and mice, two isoforms of Agtr1 receptor are expressed: Agtr1a and Agtr1b. These "a" and "b" isoforms share 95% amino acid sequence homology (12, 25) but are encoded by different genetic loci.
The structure of rat Agtr1a has been investigated previously (22). Rat Agtr1a consists of at least three exons, with exon 3 containing the entire coding region along with 5' and 3'-untranslated sequences. This structure was deduced from several isolated cDNA clones from a rat vascular smooth muscle cell cDNA library which shared identical open reading frames (ORF) but differing 5'- and 3'-untranslated sequences (22). In addition, a region of 3'-untranslated sequence further downstream may represent an additional one or more exons, as identified in one of the cDNA clones (31).
However, the molecular and exon/intron structure of the mouse Agtr1a gene has not been described in the literature. Rat and mouse Agtr1a genes are likely to share the same respective gene origins and to have co-evolved (36). Therefore, it is likely that the structure of mouse Agtr1a would be highly analogous to that of rat Agtr1a.
A genome-wide linkage scan for blood pressure loci in mice has been reported by us (35). In that study, we performed intraspecific intercrosses of an inbred hypertensive mouse strain, BPH/2 [Mus musculus genetically high blood pressure strain (BPH/2@J)], with a hypotensive strain, BPL/1 [Mus musculus genetically low blood pressure strain (BPL/1@J)], as well as two interspecific backcrosses involving a wild strain of relatively hypertensive mice, Mus spretus (SPRET). An interval linkage map for systolic blood pressure in a genome scan was created, and significant linkage was observed on chromosome 13 in a SPRET backcross to BPL/1 [(SPRETxBPL/1)xBPL/1; abbreviated "BCBPL," backcross to BPL/1], with a LOD score of 3.3 (Fig. 1). Of note, the support interval for this linkage corresponded to the Agtr1a locus at 16.0 cM on mouse chromosome 13. Unexpectedly, the phenotypic effect of the BPL/1 allele at this locus in the backcross was an increase in systolic blood pressure (35).
|
In this report, we describe the structure of the mouse Agtr1a locus based on the known rat Agtr1a structure, newly obtained mouse genomic sequence, and RT-PCR of the adrenal mRNA from the three mouse strains. We report the sequences of exons 13 and 1 kbp of the upstream promoter region in SPRET, BPL/1, and BPH/2, and we identify polymorphic differences. We performed studies on endogenous Agtr1a gene expression via real-time RT-PCR quantification of mRNA in adrenal glands of all three strains. We also studied gene expression using Agtr1a promoter/reporter constructs transfected into chromaffin cells. Our results suggest substantial structural and functional differences between the Agtr1a promoters of BPL/1 and SPRET, which may account for quantitative differences in gene expression in vivo and hence the blood pressure linkage (35).
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Informatics.
The putative gene structure of the Agtr1a gene in the mouse was determined by analyzing known sequences of rat Agtr1a (rat Agtr1a cDNA, GenBank accession no. X62295; rat Agtr1a genomic DNA, GenBank accession no. S66402) deposited in the National Center for Biotechnology Information database (http://www.ncbi.nlm.nih.gov) and aligning rat exons to mouse genome chromosome 13 contigs surrounding mouse genomic Agtr1a (mCG48688) from the Celera mouse genome database (http://www.celera.com) to detect conserved regions. Multiple sequence alignments were conducted with ClustalW (MacVector; Accelrys, San Diego, CA).
Interspecies (mouse/human) sequence homology plots were generated on the VISTA ("visualization tools for alignment") (20) server at http://www-gsd.lbl.gov/vista, using 100-bp sliding windows. The genomic sequences aligned were mouse Agtr1a (Celera mCG48688, and surrounding genomic DNA contig on chromosome 13; 60- to 500-kbp segments used for alignment) and human AGTR1 (GenBank NT_005616; 883-kbp contig on chromosome 3).
Sequencing PCR primer design and optimization.
Primer pairs were designed for mouse Agtr1a exons and the promoter region using mouse genome nucleotide information from the Celera mouse genome database. Targets for PCR amplification were identified and converted into FASTA format. MacVector (Accelrys) was used to design primer pairs, typically 18- to 24-mers spanning 200- to 1,000-bp target regions (amplicons), and avoiding primer dimer formation. Primer pairs with the lowest practical percent G/C content (typically <60% G/C) were chosen for annealing temperatures between 4565°C.
Optimization of primer pairs was performed on plates containing duplicate wells of template genomic DNA (5 ng in 10 µl). The DNA templates for optimization were genomic DNA from SPRET, BPL/1, BPH/2, and a negative control (no DNA).
Optimization and amplification of exon 3 was performed using AmpliTaq Gold thermostable DNA polymerase (Applied Biosystems, Foster City, CA). The mixture for an individual optimization PCR reaction (10 µl) consisted of the following final concentrations: 400 nM forward/reverse primers (Genset Oligos, La Jolla, CA), 0.2 mM each dNTP (Invitrogen, Carlsbad, CA), 0.5 U AmpliTaq Gold DNA polymerase, 1 mM MgCl2, TM buffer [1 x concentration = 34 mM Tris, pH 8, 25 mM KCl, 2.5 mM MgCl2, 8.3 mM (NH4)2SO4, 0.85 mg/ml bovine serum albumin], supplemented with 18 M deionized/distilled water (Sigma, St. Louis, MO), to achieve the final desired total volume (10 µl/optimization, 50 µl/amplification). PCR was performed by PTC-200 DNA Engine thermal cyclers (MJ Research, Watertown, MA). The Touchdown Profile program (MJ Research) was used, which began at annealing temperatures of 66°C and ran down to 50°C at 1°C/cycle for the first 16 PCR cycles, followed by a uniform three-step amplification profile (94°C denaturing step for 30 s, 50°C annealing step for 30 s, 72°C extension step for 30 s) for another 24 cycles, then finally holding at 10°C.
Optimization and amplification of exon 1, exon 2, and the promoter region were performed with the HotStar Taq Master Mix kit (Qiagen, Valencia, CA). The final mixture for amplification contained the following final concentrations: 1.5 mM MgCl2, 0.2 mM each dNTP, and 100 nM forward/reverse primer (Genset Oligos). The PCR protocol began with a 95°C enzyme-activating step, followed by a Touchdown program (94°C denaturing step for 30 s followed by initial annealing temperature of 70°C, subsequently run down to 55°C at 1°C/cycle, 72°C extension step for 1 min 30 s), followed by a uniform three-step amplification profile (94°C denaturing step for 30 s, 54°C annealing step for 30 s, 72°C extension step for 1 min 30 s) for another 24 cycles, then 72°C for 10 min, and finally held at 4°C.
After PCR, the products were visualized using agarose gels and ethidium bromide stain. PCR products were then purified with the QIAquick PCR purification kit (Qiagen).
Resequencing Agtr1a genomic DNA for polymorphism discovery.
Genomic DNAs were prepared from mouse tails as previously described (35). Purified PCR products served as templates for resequencing, with the exception that purified PCR products inserted into the Promega (Madison, WI) pGL3-Basic Vector (GenBank/EMBL accession no. U47295) were used for resequencing the promoters (including the promoter microsatellites). Resequencing was performed by the UCSD Cancer Center DNA Sequencing Service with the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit version 3 (Applied Biosystems). Areas resequenced included each exon (both coding exon 3 and noncoding exons 1 and 2), exon/intron borders, and the proximal promoter (1,1001,200 bp upstream of the cap site).
Gene expression: RT-PCR for endogenous Agtr1a mRNA abundance.
Total RNA was prepared from freshly frozen adrenal glands of the three strains (typical yield: 67 µg/pair of adrenal glands). RNA was extracted by the RNAzol (guanidinium thiocyanate) method (Tel-Test, Friendswood, TX), followed by RNase-free DNase I (Qiagen) treatment (to eliminate residual genomic DNA). Integrity of the RNA was confirmed by the appearance of 28S and 18S rRNA bands on ethidium bromide-stained gels. First-strand cDNA was prepared from 1 µg of total RNA template by reverse transcription with the SuperScript first-strand cDNA synthesis system for RT-PCR, using SuperScript II reverse transcriptase (Invitrogen), and random hexamer primers. mRNA/cDNA for Agtr1a was quantified by real-time RT-PCR on a ABI-7700 (Applied Biosystems) thermal cycler and fluorescent plate reader, using the Amplifluor universal detection system (Serologicals, Norcross, GA). The mouse Agtr1a PCR primer pairs were selected from coding exon 3 of mouse Agtr1a: upstream (10 nM; beginning at the 347 bp in the ORF) 5'-TGTTCCTGCTCACGTGTCTCA-3', and downstream (100 nM; beginning at 417 bp in the ORF) 5'-GCGGAGGCGAGACTTCATC-3', designed to amplify a 70-bp region. The 18-bp "Z" sequence 5'-ACTGAACCTGACCGTACA-3' was incorporated into the upstream primer (just upstream of the mRNA recognition sequence), to allow incorporation of the fluorophore (5'-FAM-labeled) and quencher (3'-DABSYL-tagged) within a hairpin UniPrimer (100 nM), which only primes after the first PCR cycle. The threshold cycle was defined as that PCR cycle at which fluorescence was 10x the standard deviation of the baseline (initial) fluorescence signal.
Normalization was performed by quantitating the endogenous 18S rRNA (ng equivalents), using the following primer pairs: upstream 5'-CGCCGCTAGAGGTGAAATTC-3' (10 nM; also containing the "Z" sequence, as above), and downstream 5'-TTGGCAAATGCTTTCGCTC-3' (100 nM), and the UniPrimer (100 nM; as above).
As a negative control, when RNA was pretreated with RNase A (Qiagen), the Agtr1a signal in the RT-PCR assay was undetectable. As a second negative control, no PCR product was obtained when cDNA was not first prepared from the mRNA by reverse transcriptase treatment.
To detect expression of endogenous Agtr1a mRNA in rat PC12 cells, RT-PCR was performed with the mouse exon 3 primers (see above), as well as the following rat primers: upstream 5'-CCTTGGCTGACTTATGCTTTTTGC-3' (bp 471494 in rat Agtr1a cDNA clone M74054.1), and downstream 5'-ACTTCATTGGGTGGACGATGG-3' (bp 665645 in the same clone), to amplify a 195-bp region.
Resequencing mouse Agtr1a cDNAs to establish gene structure by RT-PCR.
To establish the exon/intron borders, as well as the number of exons in the mouse Agtr1a gene, we employed RT-PCR on adrenal gland mRNA, with PCR primer pairs chosen to span exon/intron borders (exon 1, intron A, exon 2, intron B, exon 3), as well as putative far-downstream (3') exons 4.1 or 4.2 (22, 31).
Total adrenal gland RNA was isolated from the three strains, and cDNAs were obtained using SuperScript reverse transcriptase, as detailed above. The PCR primer pairs for RT-PCR and sequencing are given in Fig. 2A (see also Table 1).
|
|
The amplified Agtr1a promoter fragments were digested on the 5' end with EcoRI, and the 5'-overhang was made blunt with mung bean nuclease, followed by restriction digestion on the 3' end with XhoI. The promoter fragments were then inserted 5'3' between the MluI (made blunt with mung bean nuclease) and XhoI sites in the polylinker of the firefly luciferase reporter vector pGL3-Basic (Promega), which lacks eukaryotic promoter and enhancer sequences and contains the cDNA for firefly luciferase (luc+). The correct insertion/orientation of the DNA fragments was verified by DNA sequencing.
The inserted promoter regions of the Agtr1a gene were, therefore, SPRET (-1184/+39 bp), BPL/1 (-1112/+39 bp), and BPH/2 (-1107/+39 bp).
The plasmids were purified on columns using a plasmid purification kit (Qiagen), prior to transfection.
Mouse Agtr1a promoter/reporter transfection and luciferase reporter activity assay.
PC12 pheochromocytoma cells [grown in DME high glucose (Invitrogen) with 5% heat-inactivated fetal bovine serum (Gemini Bioproducts, Woodland, CA), 10% heat-inactivated horse serum (Gemini), penicillin G (100 U/ml), and streptomycin (100 µg/ml)] were transfected (at 5060% confluence, 1 day after 1:4 splitting) with the Agtr1a promoter-firefly luciferase reporter plasmids from the three strains, by the liposome method (Superfect; Qiagen), with 1 µg of supercoiled DNA per well (12-well polystyrene plates; 2.1-cm diameter wells, Falcon Multiwell; Becton-Dickinson Labware, Franklin Lakes, NJ). As an internal control in each well for transfection efficiency, the same cells were cotransfected with the Renilla luciferase expression plasmid pRL-CMV, encoding Renilla luciferase, under the control of the strong cytomegalovirus (CMV) promoter. (Promega). The firefly and Renilla luciferase activities in the cell lysates were measured 1621 h after transfection, and the results were expressed as the ratio of firefly/Renilla luciferase activity (Stop & Glo, Promega).
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Exon/intron borders were further established by RT-PCR of adrenal gland mRNA, with PCR primer pairs spanning each exon/intron border (Fig. 2A).
The deduced mouse Agtr1a gene structure includes noncoding exon 1 (135 bp), followed by intron A (8.4 kbp), noncoding exon 2 (84 bp), intron B (
25.9 kbp), and coding exon 3 (1,960 bp). Exon 3 contains the entire 1,077 bp/359 amino acid ORF and possesses a consensus polyadenylation signal [AAUAAA, located 819 bp downstream of the ORFs translational stop codon (TGA)]. The exon/intron borders and splice donor/acceptor sites did not differ among the three mouse strains (BPH/2, BPL/1, and SPRET), as judged by cDNA resequencing.
Our assignment of exon/intron borders was consistent with the sequence of mouse expressed sequence tag (EST) sequences deposited in public databases (e.g., BG969022). Our cDNA sequence information for mouse Agtr1a extends 416 bp beyond (3' of) the most 3' exon 3 sequence information previously available in the EST databases (e.g., BF383260).
One clonal version of the rat Agtr1a mRNA/cDNA also suggests a putative "fourth exon" at the 3'-untranslated portion of the mRNA (22, 31); however, the putative "exon 4" region of the rat Agtr1a cDNA is not consistent from clone to clone, and it may represent repetitive B2 elements (31). When we aligned the putative downstream "exon 4" region of the rat to the 500-kbp mouse genomic chromosome 13 contig spanning Agtr1a, we found only two relatively poor sequence identity matches: 39% identity over
1,057 bp (rat Agtr1a cDNA bp 19212937; Ref. 22) on the mouse genomic contig at 384537385592 bp (
1.3 kbp downstream of exon 3; putative "exon 4.1"); and
48% identity over
129 bp (rat Agtr1a cDNA bp 27602888; Ref. 22) on the mouse genomic contig at 392948393084 bp (
9.7 kbp downstream of exon 3; putative "exon 4.2"). We investigated whether putative exons 4.1 or 4.2 existed, by RT-PCR with primers designed to amplify cDNAs spanning exons 34.1 or 34.2, or 4.1 alone; no evidence of such downstream exons were detected. Neither did any of the approximately eight Agtr1a ESTs in public databases possess "exon 4" sequences. These results, coupled with the consensus polyadenylation signal in exon 3 (Fig. 2A) argue strongly against the existence of a putative "exon 4" in the mouse Agtr1a gene.
Interspecies sequence conservation.
Comparison of mouse Agtr1a with human AGTR1 revealed a region of high (>80%) sequence homology in exon 3, corresponding to the ORF, whereas neither exons 1 nor 2 displayed substantial identity (<75% in each case). The proximal promoter (upstream of exon 1) also lacked substantial interspecies homology.
The genomic organization of human AGTR1 includes five exons spanning at least 37 kbp, with the entire ORF encoded by exon 5 (3, 5). Thus the major ORF alignment (Fig. 2B) is between mouse Agtr1a exon 3 and human AGTR1 exon 5.
Resequencing Agtr1a for polymorphism discovery.
Noncoding exon 1 (135 bp; panel B of the Supplemental Figure) contained three single-nucleotide polymorphisms (SNPs) in SPRET compared with BPL/1 and BPH/2, although BPH/2 and BPL/1 were identical. Two SNPs in part of 3'-flanking intron A also distinguished SPRET from either BPH or BPL.
Noncoding exon 2 (84 bp; Supplemental Figure, panel C) contained three SNPs in SPRET compared with BPL/1 and BPH/2. The flanking 189 bp in 5'-flanking intron A had four SNPs plus a 10-bp insertion/deletion differentiating the species, while the 3'-flanking 60 bp in intron B contained one interspecies SNP.
Coding exon 3 (Supplemental Figure, panel D; 1,960 bp in BPL/1 and BPH/2; 1,966 bp in SPRET) contained 26 interspecific SNPs: 5/26 in the 1,077 bp ORF (all synonymous cSNPs), and 21/26 in the 883-bp flanking regions of exon 3. In addition, there was a 10-bp insertion and 4-bp deletion in SPRET exon 3, downstream of the ORF. In 246 bp of 5'-flanking intron B, interspecific changes were one SNP and two 1-bp insertion/deletions. In 133 bp of 3'-flanking intron C, interspecific changes were three SNPs and one 1-bp insertion/deletion.
Translation of the ORF sequences showed the five SNPs located in the ORF to be synonymous (no changes in the 359 amino acids). See SwissProt/NCBI protein entry P29754.
Resequencing the promoter region (Fig. 3 and Supplemental Figure, panel A) revealed a TATA box (TATAAATAT) at -36/-44 bp in all three strains. There was a long simple sequence repeat region, composed of di- (CT), tri- (CTT), tetra- (CTTT or CTTC), and penta-nucleotide (CCTTT or CCTCT) repeats, from -575/-912 bp upstream of the cap site (SPRET promoter numbering). The length of this simple sequence repeat region varied from strain to strain: the microsatellite was 338 bp long in SPRET, 264 bp in BPL/1, and 260 bp in BPH/2.
|
|
|
|
|
Expression of the Agtr1a promoter in PC12 chromaffin cells.
After subcloning of corresponding Agtr1a promoter fragments into the promoterless luciferase reporter plasmid pGL3-Basic, the promoter/reporter plasmids were transfected into PC12 pheochromocytoma cells; rodent chromaffin cells were chosen because the angiotensin II (type 1) receptor is highly expressed in such cells (8, 17, 21).
Basal (unstimulated) expression (Fig. 5) of the transfected Agtr1a promoter differed significantly (ANOVA F = 58.9, P = 0.0001) among the three strains [SPRET, 6.73 ± 0.26 RLU; BPL/1, 13.16 ± 0.46 RLU; BHP/2, 10.5 ± 0.59 RLU; "RLU" is relative light units (luciferase reporter activity, firefly/Renilla)]. Individual strain comparisons (with Bonferroni post hoc corrections) revealed that the Agtr1a promoter was 48% less efficient in SPRET than BPL/1 (P < 0.001; Fig. 5) and also less efficient in SPRET than BPH/2 (P < 0.01).
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A genetic cross of a low blood pressure strain of Mus musculus (BPL/1) and the relatively high blood pressure (35) Mediterranean mouse species M. spretus (or SPRET) revealed a significant linkage (LOD = 3.3) of systolic blood pressure to a region on mouse chromosome 13, at marker D13Mit198, 16.0 cM from the p-terminus of the chromosome (Fig. 1). Since a mouse locus encoding the angiotensin II (type 1) receptor (Agtr1a) lies directly beneath the LOD peak at 16.0 cM, we decided to investigate the Agtr1a locus as a "positional candidate" for this blood pressure linkage.
Unexpectedly, the BPL/1 allele at 16.0 cM on chromosome 13 was associated with higher blood pressure in the N2 of backcross (BCBPL) offspring than the corresponding allele in SPRET (35). Thus we sought compatible changes in Agtr1a differentiating SPRET and BPL/1: that is, an alteration at Agtr1a causing greater expression in BPL/1 than in SPRET.
Structure of the mouse Agtr1a locus.
A gene structure of Agtr1a has been tentatively described in the rat (31) composed of at least three exons, although putative rat Agtr1a exon 4 is problematic in that the exon 4 sequences may represent repetitive B2 elements (31). Thus only three exons are clear-cut in rat Agtr1a (31).
The exon/intron structure of mouse Agtr1a has not previously been clearly described. Using conserved sequence homologies between rat vascular Agtr1a cDNA clones (GenBank X62295; Ref. 22) and the emerging mouse genome contig assembly, as well as extensive resequencing of RT-PCR products from adrenal gland mRNA, we deduced the exon/intron structure for mouse Agtr1a, similar to that of the rat (Fig. 2A). The structure includes noncoding (5' untranslated) exons 1 and 2, and then exon 3, which encodes the entire ORF, as well as a consensus polyadenylation signal (AAUAAA).
In neither our own cDNA reseqeuncing nor in the available mouse Agtr1a ESTs was there any evidence for a fourth (noncoding, 3' untranslated) exon in the mouse Agtr1a gene.
The human AGTR1 locus also displayed sequence homology with mouse Agtr1a, but substantial homology was limited to the ORF in exon 3 (Fig. 2B).
Mouse Agtr1a polymorphism discovery by resequencing in BPL/1, BPH/2, and SPRET.
Resequencing of the exonic and promoter regions in Agtr1a identified multiple single nucleotide variations among the strains (Supplemental Figure). Several of these polymorphisms were located within the coding region of exon 3 (Supplemental Figure, panel D); however, each of these proved to be synonymous, indicating that qualitative interstrain differences in the Agtr1a protein cannot account for the systolic blood pressure linkage on mouse chromosome 13 (Fig. 1) in the genetic cross of BPL and SPRET (35).
Variations in angiotensin II (type 1) receptor exon/intron splicing occur in both the rat (22, 31) and the human (3). In the human, alternative splicing can give rise to alterations in translational efficiency and thus overall gene expression (3). However, we did not find interstrain variations in consensus (18) splice donor (GU) or acceptor (AG) dinucleotides in intron A (Supplemental Figure, panels B and C) or intron B (panels C and D). In intron A just upstream (5') of exon 2 (panel C), SPRET is deleted for a 10-bp A/T-rich region; although such pyrimidine-rich regions upstream of splice acceptor sites are common in higher organisms (18), the functional significance of this deletion in SPRET intron A is not clear.
Quantitative differences in Agtr1a expression might therefore be responsible for the systolic blood pressure linkage on chromosome 13 between BPL/1 and SPRET (Fig. 1; see also, Ref. 35). We therefore examined the promoter sequence from the cap (transcription initiation) site out to approximately -1100 bp (Supplemental Figure, panel A) in each strain, in which we discovered a complex microsatellite (consisting of variable numbers of di-, tri-, tetra-, and penta-nucleotide repeats), whose length varied among the three strains and was longest in SPRET (Fig. 3; 338 bp long in SPRET, 264 bp in BPL/1, and 260 bp in BPH/2).
The promoter also contained 25 single nucleotide variants and 3 single nucleotide deletions differentiating the 3 strains, 23 of which were unique to SPRET (Tables 24). Of note, some of the single nucleotide variants unique to SPRET altered domains that contained putative transcriptional regulatory sites (see Tables 24): T-55C (creating a longer G/C-rich stretch), A-102G (lengthening a G/C-rich stretch), G-128A (shortening a G/C-rich region), deletion of A at -347 in SPRET (creating a GHF.1 site), G-932A (creating a IgH-1 site), C-1005T (creating a Stat semi-palindrome), and C-1100G (creating a C/EBP site). Murasawa et al. (21) identified an A/T-rich region of the rat Agtr1a promoter (-456/-442 bp) that functions as a negative regulatory element in chromaffin cell expression of Agtr1a.
Agtr1a endogenous gene expression studies: transcript abundance.
Since the angiotensin II (type 1) receptor is expressed in the rodent adrenal gland (8), including chromaffin cells (8), where it may regulate catecholamine release (19), we studied the abundance of Agtr1a transcripts in the adrenal glands of the three strains, using RT-PCR (Fig. 4). The three strains differed significantly in Agtr1a mRNA abundance (ANOVA F = 5.77, P = 0.014). BPL/1 (at 175 ± 2.44 ng equivalents) expressed 67% more (Bonferroni P < 0.05) Agtr1a mRNA than SPRET (at 105 ± 16 ng equivalents). Relative overexpression of Agtr1a in BPL/1 (compared to SPRET) is consistent with the action of the BPL/1 Agtr1a allele to increase systolic blood pressure in the N2 generation of the linkage study of BCBPL [(SPRETxBPL/1)xBPL/1; Ref. 35].
Agtr1a is expressed in not only the adrenal medulla, but also the zona glomerulosa cells of the adrenal cortex, where it regulates mineralocorticoid biosynthesis and release (9, 13). Our RT-PCR results on total adrenal gland mRNA cannot distinguish whether differences in expression between the three strains arose from the medulla vs. the cortex.
Agtr1a promoter variants: promoter/reporter transfections.
Consistent with the mRNA abundance studies (Fig. 4), the three strains transfected promoters differed significantly in activity (ANOVA F = 58.9, P = 0.0001). The transfected Agtr1a promoter from BPL/1 was 95% more active (Bonferroni P < 0.001) than the orthologous region from SPRET (Fig. 5).
In conclusion, Agtr1a endogenous gene expression in vivo was diminished in SPRET compared with BPL/1 (Fig. 4) and the in vitro Agtr1a promoter/reporter experiment showed corresponding changes in SPRET vs. BPL/1 promoter strength (Fig. 5). Such consistent results are compatible with the viewpoint that alterations in the SPRET proximal promoter (Figs. 3 and 5 and Supplemental Figure, panel A) are sufficient to account for the variation in Agtr1a expression in vivo (Fig. 4) between SPRET and BPL/1. Thus Agtr1a proximal promoter variants may underlie the predicted effect of the BPL/1 allele upon blood pressure (a decrease) and hence may explain the role of this region of mouse chromosome 13 (LOD = 3.3) in the genome-wide scan for loci affecting blood pressure (35). The confidence interval for linkage in this cross (Fig. 1) is substantial, spanning perhaps 10 cM and thus many genes in addition to Agtr1a. Even though Agtr1a is directly beneath the LOD peak at 16.0 cM, other genes in this region might have contributed to the linkage result. However, quantitative variation in Agtr1a expression in the mouse is known to result in dose-dependent changes in blood pressure, as evidenced by results of gene targeting studies inactivating one or both copies of Agtr1a (11)
Although we have not yet precisely determined which of the promoter variants (Tables 24, Fig. 3; and Supplemental Figure, panel A) determine the differences in basal Agtr1a expression between BPL/1 and SPRET (Figs. 4 and 5), the variable repeat region in the promoter is particularly striking (Fig. 3). There is emerging evidence that simple sequence repeat polymorphisms in gene promoters may be biologically meaningful and can account for inter-individual trait variability in humans (7). However, whether the promoter strength variation (Fig. 5) results from the variable microsatellite region (Supplemental Figure, panel A; Fig. 3) or the variable SNPs (Tables 24 and Supplemental Figure, panel A) in SPRET is not yet clear from these experiments.
In humans, there is only one AGTR1 locus (on chromosome 3q24), as opposed to two in rodents (Agtr1a and Agtr1b). Human AGTR1 does display polymorphism (2, 24), although its coding region seems to be relatively invariant, without common inter-individual changes in amino acids (2). An SNP in the 3'-untranslated portion of the mRNA (A1166C) has been associated with hypertension, arterial vasoconstriction, cardiac hypertrophy, and myocardial infarction (24, 33, 34), while certain promoter SNPs (2) may be associated with human hypertension, such as -535T (30). Mouse Agtr1a and human AGTR1 display substantial (>80%) homology only in coding exon 3 (Fig. 2B), while the promoter regions are divergent. The human AGTR1 promoter does contain regions of low complexity (Repeat Masker analysis of GenBank clone NM_000685 at http://www.genome.ucsc.edu), while the nearest microsatellite, a [TTTTG]n penta-nucleotide repeat, is 2 kbp upstream from the cap site.
Quantitative variation in other genes encoding components of the renin-angiotensin-aldosterone cascade is well-documented at the human loci. For example, SNP variants in the proximal human AGT promoter are associated with changes in promoter activity in vitro, as well as the circulating concentration of angiotensinogen in vivo (10). At the human ACE locus, an insertion/deletion polymorphism (as well as nearby SNPs) is associated with variations in plasma ACE activity in vivo (37). At the human AGTR1 locus, SNPs in both the promoter region (30) and the 3'-untranslated region of the mRNA (24, 33, 34) associate with clinical phenotypes, despite an invariant amino acid sequence (2). At the aldosterone synthase locus, a promoter SNP also associates with clinical phenotypes (23).
Thus our report on mouse Agtr1a functional promoter polymorphisms adds to the growing body of evidence that quantitative (rather than qualitative, amino acid changing) variations in gene expression at loci encoding components of the renin-angiotensin-aldosterone system may be genetically linked to or associated with physiological alterations in blood pressure in vivo.
![]() |
ACKNOWLEDGMENTS |
---|
The mouse genomic DNA sequence for Agtr1a was established by Celera Genomics and was deposited in the Celera Discovery System at http://www.celera.com. Mouse strains used for this sequencing included 129X1/SvJ, DBA/2J, A/J, C57BL/6J, and 129S/SvlmJ.
![]() |
FOOTNOTES |
---|
Address for reprint requests and other correspondence: D. T. OConnor, Dept. of Medicine and Center for Molecular Genetics, Univ. of California at San Diego (9111H), 3350 La Jolla Village Drive, San Diego, CA 92161 (E-mail: doconnor{at}ucsd.edu; URL: http://medicine.ucsd.edu/hypertension; http://elcapitan.ucsd.edu/hyper).
10.1152/physiolgenomics.00162.2002.
* C. Wong and N. R. Mahapatra contributed equally to this work.
1 The Supplementary Material for this article (panels AD for the Supplemental Figure) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00162.2002/DC1.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|