(Hypertension. 1999;34:625-630.)
© 1999 American Heart Association, Inc.
Scientific Contributions |
From the Division of Human Cancer Genetics and the Comprehensive Cancer Center (F.A.W.), Ohio State University, Columbus, Ohio; the Department of Medicine and Center for Molecular Genetics (D.T.O., L.U.Y.), University of California and VA Healthcare System, San Diego; the Howard Hughes Medical Institute (E.R.), Division of Cellular and Molecular Medicine, University of California, San Diego; the Genetics Program (G.K., G.S.), University of Kansas, Lawrence; and the Department of Family and Preventive Medicine (C.C.B., D.T.), University of California, San Diego.
Correspondence to Fred A. Wright, PhD, Assistant Professor, Division of Human Cancer Genetics, The Ohio State University, 420 W 12th Ave, Room 464A, Columbus, OH 43210. E-mail wright-4{at}medctr.osu.edu
| Abstract |
|---|
|
|
|---|
Key Words: linkage hypertension, genetic mice genes
| Introduction |
|---|
|
|
|---|
Rodent models provide another approach, in which genetic synteny or homology15 may be used to elucidate genetic mechanisms in human hypertension.16 Crosses of the spontaneously hypertensive rat (SHR)17 with normotensive strains and recombinant inbred analyses have provided evidence for blood pressure involvement of several chromosomal regions.6 18 19 Congenic strains have recently provided more definitive evidence in several chromosomal regions by introgressing segments from hypertensive rat strains into a normotensive background.20 21 22 23 Unfortunately, positional cloning strategies in animal models of human hypertension have thus far yielded definitive characterization of few if any loci.24 25
In contrast, few linkage analyses26 have been reported for blood pressure in the mouse, perhaps because of its small size and the difficulty in assessing the phenotype.27 However, when properly performed, noninvasive tail-cuff systolic blood pressure measurements are closely correlated with intra-arterial measurements.28 29 The genetic analysis of newly identified candidate genes in the mouse may provide independent and corroborative evidence of the genetic components of rodent hypertension. Mice of the inbred hypertensive strain BPH/2 and hypotensive strain BPL/1 have been described,27 with comparison of blood pressures and related phenotypes.30 These strains were developed by 2-way selection31 (high blood pressure versus low blood pressure) from an initial cross of 8 inbred strains,27 and inbreeding of the unselected control population resulted in an additional, normotensive strain BPN/3. By use of these strains and a standard biometric method for estimating the minimum number of involved quantitative trait loci (QTLs),32 3 to 5 major genes were estimated to contribute to the 55 mm Hg strain difference in blood pressure, with heritability of 30% to 60%.33 Interestingly, the BPH/2 strain also exhibits elevated pulse rate, left ventricular mass, and hematocrit (percent) compared with BPL/1,27 30 phenotypes that have also been associated with human hypertension and its cardiovascular sequelae.1 34 35 A previous report36 examined cosegregation of blood pressure with 2 candidate genes on chromosomes 8 and 12 in the BPH/2xBPH/1 intercross.
Although the BPH/2 and BPL/1 strains are well described, their genetic similarity and the possibility of random common gene fixation may limit the number of blood pressure genes that vary across the strains. Inbred Mus spretus have relatively high blood pressure, and interspecific crosses of this species with BPL/1 may elucidate additional genes that contribute to variation in blood pressure.37
We undertook a genome-wide QTL linkage scan38 for
systolic blood pressure in intraspecific F2 crosses of 247
BPH/2xBPL/1 mice, 92 interspecific backcross (BC) mice designated
BCBPL (M spretusxBPL/1)xBPL/1, and 84 interspecific
backcross mice designated BCMS (M spretusxBPL/1)xM
spretus. Blood pressure was determined from tail-cuff
measurements, with multiple observations averaged for each mouse. A
microsatellite map39 was used to cover the genome
with 91 markers in the F2s and 95 markers in the backcrosses, at an
average spacing of
16 cM. Additional phenotypes associated
with mouse and human hypertension were also examined for linkage as
available.
| Methods |
|---|
|
|
|---|
Blood Pressure Measurement
Systolic blood pressures of the mice in set A were
measured during the summer of 1994 beginning when the mice reached 100
days old. These determinations were tail-cuff measurements (8-mm ID,
15 mm long) taken on unanesthetized restrained mice with
the Narco-Biosystems Physiograph. Five determinations were made on each
of 3 days at approximately weekly intervals and then measured once or
twice again when the mice were 7 to 8 months old. The blood pressures
of the mice in set B were determined in fall and winter of 1992 on the
same schedule, when the mice reached 100 days old, at 3 weekly
intervals for a total of 15 measurements. All the mice were first
trained to tolerate restraint and mild warming (37.5°C) by being
placed into the restrainer for 1, 2, and 3 minutes on consecutive days.
Most mice had blood pressure measurements on day 4. Some mice required
additional training before blood pressures could be determined. The
multiple observations for each mouse were averaged to produce a single
phenotype value. Because blood pressure in mice varies
seasonally and the F2 crosses were measured over different seasons, all
blood pressures were converted to deviations from the mean (in
mm Hg) within each cross for males and females separately.
Interspecific Backcrosses
Inbred male M spretus were purchased from the Jackson
Laboratory, Bar Harbor, Me, and crossed with BPL/1 females to
produce an F1. The F1 females were in turn crossed to the BPL/1 males
and M spretus males to produce 2 backcross generations:
BCBPL and BCMS, respectively. In total, 51 male and 41 female BCBPLs
and 38 male and 46 female BCMS mice were produced, for which blood
pressures were determined and DNA extracted.
Blood Pressure Measurement
The mice backcrossed to the BPL/1 could have their blood
pressures taken in the usual manner as described above, restrained and
unanesthetized, after considerable training. Five blood
pressure measurements were taken on each mouse on each of 5 or 6 weekly
intervals during the fall and winter of 1992/1993 when the mice were
120 to 150 days old. These values were averaged to form a single
phenotype value for each mouse.
The mice backcrossed to M spretus were relatively "wild" and could not be trained to tolerate restraint. The blood pressure of these mice was measured under metofane anesthetic when the mice were 5 to 8 months old during a 2-year period, 1993 to 1994. The mice were anesthetized for 2 minutes in a glass jar containing cotton soaked with metofane. The mice were then placed on a warming plate (37.5°C) with a nose cone containing cotton and metofane. Within a few minutes, pulse was discernible in the tail, at which time the nose cone was removed. The measurement of blood pressure commenced and was continued at 1-minute intervals until the mouse regained semiconsciousness. As many as 20 blood pressure measurements were taken on some mice. The last few measurements before the mice regained consciousness were likely to be the most comparable across mice, and these were averaged for each mouse to produce a single phenotype for linkage analysis. All procedures used in the crosses were in accordance with guidelines of the University of Kansas.
Additional Phenotypes
The procedures for determining pulse rate, left
ventricular mass, and hematocrit percent are described in
more detail elsewhere.30 Pulse rate was measured each time
blood pressure was determined. The F2 mice in set A were
anesthetized and killed at 150 days old by cervical
dislocation. The heart was removed and cleaned and the left ventricle
separated and weighed on a Mettler balance.
DNA Preparation and Genotyping
A simple 3-step DNA extraction was performed from mouse tail
tissue by use of a Qiagen kit and column (proteinase K step, RNAse A
step, and QIAamp spun column; Qiagen Inc). By this procedure, we
routinely obtained 160 µg of DNA/tail in 400 µL of Tris-EDTA
buffer. Samples were sent to the J.L. Weber laboratory for
genotyping.
Markers were scored by at least 2 independent reviewers. In addition, genotypes were reexamined in regions of significant linkage. Markers were spaced as evenly as possible and extended to within 15 cM of both chromosome ends for all but the X chromosome. The mean±SD for marker interval widths was 16.1±7.1 cM for the F2 intercrosses and 16.2±5.1 cM for the backcrosses. Attempts were made to close any gaps >30 cM, but 3 such chromosomal regions for the F2s and 1 region for the interspecific backcrosses remained uninformative at all microsatellites examined.
Linkage and Statistical Analyses
Linkage analysis was performed for each of the 3 crosses
with the Mapmaker/QTL program 1.1 (Whitehead Institute)40
on a UNIX workstation. The precise phenotype for each cross was
chosen before analysis. Statistical power calculations for a
genome-wide scan were performed by use of both simulation and analytic
approximations,41 which were in close agreement. The F2
intercross data were shown to have >80% power to detect QTLs
contributing 10% of the variation in blood pressure, whereas the
backcrosses BCBPL and BCMS had power of 73% and 70%, respectively, to
detect loci contributing 20% of the phenotype variation. The
crosses were expected to provide little power to detect epistatic
interactions among the loci. Mapmaker/QTL allows the inclusion of
covariates in the linkage analysis,40 and sex was
introduced as an additive effect to reduce error
variation.40 42 The covariate also appears in the
denominator of the likelihood ratio and thus does not inflate the type
I error. Statistical power for the F2s was enhanced by genotyping of
the 98 phenotypically extreme (for blood pressure) mice of set
B,38 with the phenotypes of the remaining 123
nongenotyped mice also entered to provide valid likelihood
estimates and combined with the mice of set A. Pulse rate was
analyzed for the F2s in the same manner. The mice in the
remaining crosses were not chosen on the basis of extreme
phenotype, and all available individuals were used in the
analysis.
Linkage analyses used Haldane's map function and marker map locations reported by Research Genetics, which agrees well with other published maps.39 Additional marker ordering and mapping was performed on the data by use of Mapmaker/EXE.43 Maps of mammalian homology reported by the Jackson Laboratory (http://www.informatics.jax.org) were used to compare syntenic regions. Linkage results are reported according to published conservative standards44 for the maximum LOD in a dense genome scan. LODs of 4.3 and 3.3 are considered significant at the 0.05 level for F2 and backcross populations, respectively. Thresholds of 2.8 and 2.3 are considered "suggestive," in that they are expected to occur approximately once in a full genome scan. To facilitate communication,44 all regions with LODs >2.0 are reported here.
Additional Statistical Analyses
Additional statistical analyses were performed with the
S-Plus 3.4 statistics package (MathSoft, Inc). A
2 contingency table test of Mendelian ratios
revealed a statistically significant imbalance in parental strain
origin of allele marker D13 Mit78 in the F2 cross, in a region not
showing evidence of blood pressure linkage. Linkage results were also
confirmed by interval mapping regression45 and
permutation-based tests of linkage significance.46 ANOVA
tests of 2-way genetic interactions of all pairs of markers were also
performed, with permutation tests used to construct a genome-wide
P value.
| Results |
|---|
|
|
|---|
Interval mapping was performed with Mapmaker/QTL40 for each of the 3 crosses and revealed 10 regions with LOD scores >2.0 (Table 1). Two regions provided significant linkage evidence according to conservative standards44 : the interval D10 Mit123D10 Mit117 (LOD=4.9) in the F2 intercross and D13 Mit198 (LOD=3.3) in the BCBPL backcross. Additional suggestive loci were identified on chromosomes 2, 6, 8, and 18, including a region, D2 Mit192/274, that approached significant linkage in the BCBPL backcross with LOD=3.2. A moderate amount of phenotype variation (10% to 25%) may be attributed to these loci individually, and the summed contribution of the suggestive or significant loci (between 30% and 50% in each cross) is consistent with previous heritability estimates.33 The Figure plots the LOD scores for blood pressure on the 4 chromosomes that achieved suggestive or significant linkage.
|
|
BPL/1 alleles are present in all the crosses, and QTL alleles from this strain would be expected to reduce blood pressure relative to any other parent strain. Although this is true for most of the regions identified, the BPL/1 allele near D13 Mit198 appears to increase blood pressure relative to M spretus. A similar phenomenon has been reported in the rat,47 in which alleles at some loci from the SHR strain appeared to decrease blood pressure. There was no evidence of chromosome 13 involvement in the BCMS cross, suggesting that the BPL/1 allele may be dominant or due to differing genetic background between the strains (ie, other interacting loci).
The LOD score analysis for the significant loci was supported by additional statistical procedures.45 46 However, the LOD score of 4.9 was achieved only in a relatively large interval between markers (although an LOD >3.0 was observed at flanking marker D10 Mit123). Interestingly, the locus on chromosome 10 does not show evidence in either of the interspecific backcrosses, suggesting that M spretus may not have an allele conferring high blood pressure in this region.
Multiple Gene Models
Additive models38 with multiple genes were explored
with Mapmaker/QTL, but no new loci were identified beyond those listed
in Table 1. Using ANOVA with interaction terms,48
we examined all pairs of markers for evidence of epistatic effects on
blood pressure. None of these interaction effects were statistically
significant when permutation testing was used to apply a
multiple-comparison adjustment.
Comparison of Blood Pressures Across Strains
Measurements performed under inhaled metofane anesthetic provide a
uniform condition under which blood pressures can be compared across
strains and crosses. All of the BCMS mice in this report were measured
under metofane anesthetic (see Methods). A few mice from the other
strains were chosen to be measured under metofane anesthetic (in
addition to their unanesthetized measurements) and are
presented for comparison in Table 2. This table exhibits the anticipated
differences across strains, with the exception of the BCMS cross. The
average blood pressure for BCMS females was greater than that of both
of the parent strains (not significant). Metofane phenotypes
for F1 females were unavailable.
|
Additional Phenotypes
In addition to blood pressure, the phenotypes pulse rate,
relative left ventricular mass (left
ventricular mass/body mass), and hematocrit (percent) were
examined. Pulse rate was measured concurrently with blood pressure.
Left ventricular mass was measured in
50% of the F2
mice (n=126) when they were killed at 180 to 250 days old, and
hematocrit (percent) was measured only in males of this group (n=71).
The procedures for obtaining these phenotypes are discussed in
greater detail elsewhere.30 The linkage results are given
in Table 1. None of the regions showed significant linkage,
although a LOD of 4.2 for pulse rate was highly suggestive for a locus
in the interval D3 Mit60/22 among the F2s.
| Discussion |
|---|
|
|
|---|
Other suggestive44 blood pressure linkages were found on chromosomes 2, 6, 8, and 18. Interestingly, only chromosome 2 shows evidence of involvement in multiple crosses (Table 1 and Figure), and 1-LOD-unit support intervals in the 2 backcrosses exhibit some overlap. The overall results are consistent with the complex, polygenic nature of blood pressure.51
Loci encoding components of the renin-angiotensin system are leading candidate genes for hypertension, and affected sib-pair methods have found evidence for linkage of the angiotensinogen (AGT) locus to human hypertension.52 Some alleles at the angiotensin II receptor type I locus (AGTR1) are associated with human hypertension,53 and alleles at the angiotensin-converting enzyme (ACE) locus are associated with adverse myocardial events in hypertension.54
In the spontaneously hypertensive rat, linkage established the role of a major locus (BPH1) near ACE on rat chromosome 1055 56 ; the mouse homologue is probably on chromosome 11. In the mouse, renin-angiotensin components include Agt at 68 cM (angiotensinogen, renin substrate) on mouse chromosome 8, Ren (renin) on mouse chromosome 1 at 70 cM, angiotensin-converting enzyme mouse homologue Ace not assigned but probably on chromosome 11 by synteny, and Agtr2 (angiotensin II receptor type II) on the mouse X chromosome. None of these regions displayed significant linkage to blood pressure in our data. In the mouse genome, there are 2 angiotensin II receptor type I loci57 : Agtr1a, at 16 cM on mouse chromosome 13, and Agtr1b, at 2 cM on mouse chromosome 3. Mouse chromosome 13 showed significant linkage to blood pressure in the BCBPL (backcross to BPL/1; Table 1 and Figure), with the support interval containing Agtr1a.
These syntenic comparisons do not provide support for mouse homologues to the candidate regions in humans and rat models. However, to aid future comparisons, we list in Table 1 the presumed homologous regions in the human and rat genomes to the mouse loci identified in this study. The resolution of syntenic maps varies according to region, and in many cases, the mouse blood pressure locus is not well localized, so the syntenic region may include multiple chromosomal regions in the comparison organism.
In conclusion, this study is the first reported genome-wide linkage scan for blood pressure genes in the mouse. Linked regions of mouse-human chromosomal synteny may suggest novel loci that influence blood pressure in humans. As the genetic maps of mammals improve, investigators can increasingly take a positional candidate approach, testing logical candidate genes in linked regions.
| Acknowledgments |
|---|
Received March 2, 1999; first decision March 29, 1999; accepted June 3, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Y. Deng Positional Cloning of Quantitative Trait Loci for Blood Pressure: How Close Are We?: A Critical Perspective Hypertension, April 1, 2007; 49(4): 740 - 747. [Full Text] [PDF] |
||||
![]() |
C. B.M. Oudejans, J. Mulders, A. M.A. Lachmeijer, M. van Dijk, A. A.M. Konst, B. A. Westerman, I. J. van Wijk, P. A.J. Leegwater, H. D. Kato, T. Matsuda, et al. The parent-of-origin effect of 10q22 in pre-eclamptic females coincides with two regions clustered for genes with down-regulated expression in androgenetic placentas Mol. Hum. Reprod., August 1, 2004; 10(8): 589 - 598. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. S. Fries, P. Mahboubi, N. R. Mahapatra, S. K. Mahata, N. J. Schork, G. W. Schmid-Schoenbein, and D. T. O'Connor Neuroendocrine Transcriptome in Genetic Hypertension: Multiple Changes in Diverse Adrenal Physiological Systems Hypertension, June 1, 2004; 43(6): 1301 - 1311. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. DiPetrillo, S.-W. Tsaih, S. Sheehan, C. Johns, P. Kelmenson, H. Gavras, G. A. Churchill, and B. Paigen Genetic analysis of blood pressure in C3H/HeJ and SWR/J mice Physiol Genomics, April 13, 2004; 17(2): 215 - 220. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Welch, S. Bretschger, P.-Z. Wen, M. Mehrabian, N. Latib, J. Fruchart-Najib, J. C. Fruchart, C. Myrick, and A. J. Lusis Novel QTLs for HDL levels identified in mice by controlling for Apoa2 allelic effects: confirmation of a chromosome 6 locus in a congenic strain Physiol Genomics, March 12, 2004; 17(1): 48 - 59. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. D. L. Woo and I. Kurtz Mapping blood pressure loci in (A/J x B6)F2 mice Physiol Genomics, November 11, 2003; 15(3): 236 - 242. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Hartner, N. Cordasic, B. Klanke, R. Veelken, and K. F. Hilgers Strain differences in the development of hypertension and glomerular lesions induced by deoxycorticosterone acetate salt in mice Nephrol. Dial. Transplant., October 1, 2003; 18(10): 1999 - 2004. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Wong, N. R. Mahapatra, S. Chitbangonsyn, P. Mahboubi, M. Mahata, S. K. Mahata, and D. T. O'Connor The angiotensin II receptor (Agtr1a): functional regulatory polymorphisms in a locus genetically linked to blood pressure variation in the mouse Physiol Genomics, June 24, 2003; 14(1): 83 - 93. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. W. Cowley Jr. Genomics and homeostasis Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2003; 284(3): R611 - R627. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Sugiyama, G. A. Churchill, R. Li, L. J. M. Libby, T. Carver, K.-I. Yagami, S. W. M. John, and B. Paigen QTL associated with blood pressure, heart rate, and heart weight in CBA/CaJ and BALB/cJ mice Physiol Genomics, July 12, 2002; 10(1): 5 - 12. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Ryan, S. P. Didion, D. R. Davis, F. M. Faraci, and C. D. Sigmund Endothelial Dysfunction and Blood Pressure Variability in Selected Inbred Mouse Strains Arterioscler. Thromb. Vasc. Biol., January 1, 2002; 22(1): 42 - 48. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. SAAD, M. R. GARRETT, and J. P. RAPP Multiple blood pressure QTL on rat chromosome 1 defined by Dahl rat congenic strains Physiol Genomics, January 19, 2001; 4(3): 201 - 214. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Stoll, A. E. Kwitek-Black, A. W. Cowley Jr., E. L. Harris, S. B. Harrap, J. E. Krieger, M. P. Printz, A. P. Provoost, J. Sassard, and H. J. Jacob New Target Regions for Human Hypertension via Comparative Genomics Genome Res., April 1, 2000; 10(4): 473 - 482. [Abstract] [Full Text] |
||||
![]() |
J. P. Rapp Genetic Analysis of Inherited Hypertension in the Rat Physiol Rev, January 1, 2000; 80(1): 135 - 172. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. R. Garrett and J. P. Rapp Two closely linked interactive blood pressure QTL on rat chromosome 5 defined using congenic Dahl rats Physiol Genomics, February 28, 2002; 8(2): 81 - 86. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||