Skip Navigation


Human Molecular Genetics Advance Access originally published online on October 3, 2007
Human Molecular Genetics 2008 17(3):331-344; doi:10.1093/hmg/ddm277
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/3/331    most recent
ddm277v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Thifault, S.
Right arrow Articles by Hamet, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thifault, S.
Right arrow Articles by Hamet, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Genetic determinants of emotionality and stress response in AcB/BcA recombinant congenic mice and in silico evidence of convergence with cardiovascular candidate genes

Stéphane Thifault1, Seda Ondrej1, Yulin Sun1, Anny Fortin2, Emil Skamene3, Robert Lalonde1, Johanne Tremblay1 and Pavel Hamet1,*

1 Centre de recherche, Centre hospitalier de l’Université de Montréal (CHUM)–Technopôle Angus, Montréal, Québec, Canada 2 Emerillon Therapeutics Inc., Montréal, Québec, Canada 3 Research Institute, McGill University Health Centre, Montréal, Québec, Canada

* To whom correspondence should be addressed at: CHUM-Technopôle Angus, 4th Floor, Room 401-F, 2901 Rachel Street East, Montréal, Québec, Canada H1W 4A4. Tel: +1 5148908246; Fax: +1 5144127682; Email: pavel.hamet{at}umontreal.ca


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Genomic loci bearing stress-related phenotypes were dissected in recombinant congenic strains (RCS) of mice with C57BL/6J (B6) and A/J progenitors. Adult male mice from 14 A/J and 22 B6 background lines were evaluated for emotional reactivity in open-field (OF) and elevated plus-maze tests. Core temperature was monitored by radio telemetry during immobilization and on standard as well as salt-enriched diets. In addition, urinary electrolytes were measured. Genome-wide linkage analysis of the parameters revealed over 20 significant quantitative trait loci (QTL). The highest logarithm of odds (LOD) scores were within the previously-reported OF emotionality locus on Chr 1 (LOD = 4.6), in the dopa decarboxylase region on Chr 11 for the plus-maze (LOD = 4.7), and within a novel region of calmodulin 1 on Chr 12 for Ca++ excretion after a 24-h salt load (LOD = 4.6). RCS stress QTL overlapped with several candidate loci for cardiovascular (CV) disease. In silico evidence of functional polymorphisms by comparative sequence analysis of progenitor strains assisted to ascertain this convergence. The anxious BcA70 strain showed down regulation of Atp1a2 gene expression in the heart (P < 0.001) and brain (P < 0.05) compared with its parental B6 strain, compatible with the enhanced emotionality described in knock out animals for this gene, also involved in the salt-sensitive component of hypertension. Functional polymorphisms in regulatory elements of candidate genes of the CV/inflammatory/immune systems support the hypothesis of genetically-altered environmental susceptibility in CV disease development.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Animal models provide valuable tools to dissect the ecogenomics of multifactorial traits (1) as they allow significant reduction of the complexity of interacting hereditary and environmental factors. The genomic determinants of stress-related phenotypes (2) and hypertension (3) have begun to be identified in several mouse models, including comprehensive sets of strains derived from A/J and B6. Challenging efforts are being made toward fine-mapping to the quantitative trait nucleotide level and translation into new knowledge for mental (4,5) and cardiovascular (CV) health (6).

The AxB/BxA recombinant inbred (RI) panel was formed by interbreeding B6 and A/J founders (7). When comparing parental strains, the B6 group had higher open-field (OF) activity than the A/J group (8–14), more transitions in the light/dark box (15,16), more hole-poking responses (12), higher open arm duration in the elevated plus-maze (EPM) (14), and a higher number of cued and contextual fear-conditioning responses (17). Since the timid behaviour of the A/J albino strain is apparent in the plus-maze under red light, it is assumed that heightened susceptibility to light is not the main reason for its high anxiety levels relative to other strains (18). Analysis of AxB/BxA RI strains revealed several quantitative trait loci (QTL) for baseline and diazepam-induced activity in OF and light/dark box tests (19) as well as ethanol- (20) and cocaine-induced activation (21). Ambulatory QTL in the OF on Chr 1 (101 cM) and 10 (74 cM) were confirmed in F2 progeny derived from A/J and B6 lines (11).

The AcB/BcA recombinant congenic strains (RCS) used in our study differ from the RI panel in that each line carries a fraction (12.5%) of the genome on the complementary background (87.5%) of the other (7). This model system was designed for genetic dissection of complex diseases characterized by multiple susceptibility factors such as those involved in colon cancer (22). With A/J and B6 as progenitors, QTL were identified for host-immune responses (23,24), the acoustic startle response (25), alcohol intake (26), amphetamine-induced (27), cocaine-induced (28), nicotine-induced (29) and stress-induced activation (2).

The present study complements our previous QTL analyses of the stress response, metabolism and gene expression in mice and RI rat strains (30–34), indicating the pleiotropism of heightened stress susceptibility in genetic hypertension (35–37), which can be amplified and revealed with a high-salt diet as in some human populations (6). Hypothermia in mice corresponds to the initial stage of an acute stress response, evident after high doses of lipopolysaccharide (LPS) (38), similar to the rat hypothermic response (36,39). In seven inbred mouse strains (NM, C3H, Bl/6, Bl/10, Balb/c, A/J and SHM), body temperature changes (hypothermia followed by a recovery phase of thermogenesis) after injection of a high-LPS dose (500 µg/kg) were highly correlated (r = +0.95, P< 0.01) with 15-min immobilization stress (unpublished data).

Here, the strain distribution and genomic determinants associated with behavioural profile (exploration of the OF and EPM) were assessed along with stress-induced physiological measures of body temperature changes, urinary volume and electrolyte excretion on a normal diet and after salt loading. Given the availability of complete genome sequences of the B6 and A/J parental lines, we performed detailed analyses of the identified QTL at the single nucleotide polymorphism (SNP) level, utilizing both private and public databases. Multiple bioinformatic tools were deployed to determine the putative functional implications of individual SNPs. This screening covered interference with functional motifs for both coding and regulatory regions, and changes in mRNA secondary structure or in microRNA target sequences in the 3' untranslated region (3'-UTR).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Exploration of environmental stimuli
For emotionality (Emo) in the OF, broad sense heritability (H2) was 60.9% on the B6 and 50.1% on the A/J background. Polarity between parental strains was highly significant (P< 0.0001), with the B6 lines showing less Emo. The introduction of A/J alleles into the genomic background of the informative strains was associated with increased Emo in the BcA70 and BcA72 lines relative to B6 (P< 0.002) as presented in Figure 1A; conversely, line AcB54 was less emotional than its A/J parental strain (P< 0.004). A similar strain distribution pattern (SDP) was obtained for fecal boli (data not presented), with the same informative BcA70 and AcB54 strains being significant in their respective B6 (H2 = 61.0%) and A/J (H2 = 46.0%) backgrounds. BcA70, BcA72 and AcB54 were also close to significance (P< 0.01) for segment-crossed measurement (H2 = 61.0% in the B6 set; 60% in the A/J background). Significant Emo QTL were found for the B6 set on Chr 1, 3, 4 (2 loci), and 7 (distal) regions, with their respective logarithm of odds (LOD) scores of 4.6, 3.9, (3.5, 3.2) and 3.5 (Table 1), and where albino anxiogenic alleles were detected on BcA70 and BcA72 informative strains. Similar but only suggestive QTL on Chr 1, 3 and 4 were obtained for fecal boli in the B6 set, and no significant QTL were observed with crossed segments.


Figure 1
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Strain distribution pattern of progenitors and all 36 recombinant congenic strains (RCS), ordered by their means (with SEM) for emotional reactivity in the OF (A); emergence latencies in the EPM (B); initial hypothermia under immobilization (C); the thermogenic response to immobilization (D).

 


View this table:
[in this window]
[in a new window]

  		Table 1. Significant stress-related QTL with relevant candidate genes

 
In the EPM, emergence latencies from the initial enclosed arm displayed a heritability estimate of 59.8% on the B6 and 57.6% on the A/J background, with huge polarity between parental strains (P< 0.0001). Figure 1B illustrates that the BcA72, BcA85 and BcA87 lines took more time before starting to explore than B6 (P< 0.002); conversely, AcB51 mice emerged faster than A/J (P< 0.004). The SDP followed a similar pattern for enclosed arm visit duration, but interstrain differences were more prominent in the B6 set (H2 = 53.5 versus 39.5% in the A/J set). Significant linkage signals were detected in the B6 background on Chr 9, 11 and 19 for emergence latencies (LOD = 3.8, 4.7 and 3.9). These QTL corresponded to shorter emergence latencies in lines bearing A/J alleles. However, shorter emergence latencies were associated with the B6 allele in the AcB51 strain contributing to QTL on the upstream region of Chr 1 (LOD = 3.4) found in the A/J set. Finally, for 1 QTL detected on proximal Chr 7 (LOD = 3.8), the introduction of A/J alleles on the BcA69, BcA72 and BcA87 lines was associated with increased enclosed arm duration relative to B6 (P< 0.01).

Body temperature stress response under normal and increased salt diet
In the 30-min immobilization period, two highly inter-correlated phases of response (r = –0.84, P< 0.0001 from the means of all 36 strains) were noted: initial hypothermia, defined as the delta between minimum and baseline temperatures, followed by thermogenesis, measured as the temperature at the last minute of immobilization. Relevantly, each phase of the response was distinct in terms of the genetic determinants. Heritability estimates, which reflect the genetic influence on core temperature interstrain variability, appeared elevated on the A/J background for both hypothermic (H2 = 51.7%) and thermogenesis (H2 = 66.1%) responses compared with the pre-stress period (H2 = 45.8%). The A/J line manifested much less thermogenesis than B6 (P< 0.01), but the parental strains did not differ in terms of hypothermia (P> 0.8). The AcB56 strain differed from A/J for both hypothermic (P< 0.01) and thermogenesis (P< 0.05) responses (Fig. 1C and D). Two significant QTL were found on Chr 2 and 15 in the A/J set (Table 1), where the presence of the B6 differential segment was associated with a reduced hypothermic response. Conversely, in the A/J set, near significant pro-thermogenesis QTL were detected on Chr 3 and 10 (Table 1). In the B6 background, the BcA77, BcA82 and BcA83 lines manifested significantly less thermogenesis than B6 (P< 0.002; Fig. 1D).

Salt loading did not change the polarity of the progenitors, with the A/J line showing less thermogenesis than B6 (P< 0.01); neither was there an ancestor difference on hypothermia (P> 0.7). However, higher intra-strain variability compared to standard diet conditions reflected the environmental component of the hypothermic phase. This was less the case in the B6 background, in which distinct QTL were detected for the hypothermic response, showing a phenotype-specific interaction between salt-loading and genetic background (Table 2). The BcA66 strain was the most sensitive, experiencing even deeper hypothermia than the parental B6 line (P< 0.05, Fig. 2D), and contributing to the QTL revealed on Chr 1, 4 (2 loci), 15 and 18 (LOD = 3.5, 3.8, 3.6, 3.7 and 3.4), where A/J alleles increased the hypothermic response to stress. The BcA66, BcA73 and BcA77 lines displayed less thermogenesis than B6 (P< 0.002, Fig. 2E).


Figure 2
View larger version (49K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. SDP of progenitors and all 36 RCS, ordered by their means (SEM) for urinary volume (A); Ca++ excretion (B); creatinine excretion (C); the initial hypothermic response to immobilization on a high-salt diet (D); and thermogenic response to immobilization on a high-salt diet (E).

 


View this table:
[in this window]
[in a new window]

  		Table 2. Significant diet-induced QTL after 24 h on a high-salt diet, with relevant candidate genes

 
Circadian rhythm of baseline temperature before and during a high-salt diet
Significant main effects of diet and strain factors in 2-way ANOVA were found as a result of lower temperature under the high-salt condition [F (1,15) = 35.35; P< 0.0001] and for the A/J strains [F (1,15) = 79.03; P< 0.0001]. Nevertheless, an interaction between strain and circadian period effects on 24-h baseline temperatures was seen only under the high-salt condition [F (1,15) = 8.69; P< 0.01], in which strain differences were steeper during the dark phase. In concordance with the parental lines, significant interactions between all RCS and circadian period factors were observed for body temperature [F (1,35) = 2.47; P< 0.0001].

Urinary volume and electrolytes after 24 h of a high-salt diet
The A/J strain was characterized by lower urinary volume than B6 (P< 0.01). A QTL for urinary volume (H2 = 59.8% in the B set) was detected at a single site on Chr 11, corresponding to A/J alleles that decreased urinary volume in the BcA80 and BcA85 strains (P< 0.01) relative to B6 (Fig. 2A). Among electrolytes (Na+, K+, Ca++ and creatinine) that were measured, only Ca++ excretion in the A/J set and creatinine excretion in the B6 set (H2 = 51 and 62%, respectively) showed significant QTL, on Chr 1, 2, 4, 8, 11, 12, 13 and 17 for the former, and on Chr 15 for the latter (Table 2). All QTL for Ca++ caused decreased excretion in the already low A/J set, with no concordant parental strain differences (P> 0.8, Fig. 2B). However, caution is warranted concerning Ca++ excretion in the A/J set, because of numerous missing values in some contributing strains as a result of insufficient volumes. The QTL for creatinine excretion corresponded to an increase in the already high-excreting B6 strain (P<0.05, A/J versus B6).

In silico exploration of putative candidate genes
The OF emotionality QTL interval is presented in greater detail because of high statistical significance, spatial resolution and concordance with the literature. Starting from 83 known genes in this region, 38 of these had some SNPs in between parental strains; 5 of these contained major putative changes and could be classified into 2 categories. The CV system-related genes Atp1a2 and Nhlh1 as well as the immune/inflammatory genes Slamf1, Fcgr3 and Crp shared mRNA secondary structure changes with significant impact on binding motifs, and either functional SNP in the promoter or coding regions. A detailed summary of these results is included in Table 3. Figure 3 and Supplementary Material, Figure S1 depict in greater detail the Crp gene to illustrate the nature of the changes that were observed. The same analyses were made for all other QTL, with a detailed summary of these results appearing in Supplementary Material, Table S1, and the relevant genes mentioned in the Discussion. Overall, only a fraction of the genes from Tables 1 and 2 contained putative functional polymorphisms.


Figure 3
View larger version (40K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Integrative scheme of our step-wise approach aimed at narrowing down QTL to QTG. The haplotype box shows the donor alleles inherited from A/J named with an A, and with a B for B6 at the Chr 1 QTL interval for OF Emo. Strains were sorted according to their mean scores. Statistical significance of the QTL was confirmed by t tests on all 36 strains. Crp gene was chosen as a clear example in which each SNP presented has a functional impact either on binding motifs and/or the mRNA secondary structure within a motif (1) or in the coding region (2) (more details are presented in Supplementary Material, Figure S1).

 


View this table:
[in this window]
[in a new window]

  		Table 3. Candidate genes within the 171–174 Mbp region of Chr 1

 
Quantitative real-time (RT)-PCR validation of differential gene expression
Differential expression between progenitors and informative strains was observed only for one gene out of the five candidates. The anxious BcA70 strain showed down regulation of the Atp1a2 gene in the brain (P< 0.05, Fig. 4A), and heart (P < 0.001, Fig. 4B) compared to its parental B6 strain.


Figure 4
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Atp1a2 gene expression profiling in the brain (A), heart (B) and kidney (C) by RT–PCR. Mean mRNA levels of the A/J parental strain (n = 4) were chosen as controls. Expression levels were normalized by reference to 18S rRNA. The level of statistical significance is shown for post hoc comparison versus the B6 progenitor strain as follows: one asterisk for P < 0.05; three asterisks for P < 0.001.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
We have previously reviewed the origin of stress research as a central theme not only for its impact on behaviour but also on CV outcomes. We recalled that ‘Hans Selye came to reflect on human adaptation when he was told that ‘those boys who did not make it in the trenches had unusually large adrenals’ as observed in autopsies. The time was the First Great War and the place was Charles University in Prague where as a young medical student, he overheard this conversation between assistant professors of anatomy and pathology’ (37). Using a model of recombinant congenic mice originating from a cross between two phenotypically distinct strains, we are reporting an extensive set of genomic determinants of stress/emotionality and potentially related immune components that we believe can serve for future exploration of novel pathogenetic, therapeutic and preventive targets in CV diseases with hypertension as its leading risk and global burden (40). We have narrowed down several previously-described larger genomic regions and uncovered novel ones contributing to the mammalian response to environmental disturbances.

Three Chr 1 markers located between 93.3 and 95.8 cM were highly associated with emotional reactivity in the OF for AcB/BcA RCS. Our fine mapping was consistent with a report on intercrosses of chromosome substitution strains derived from A/J and B6 (41). These QTL were near the highest LOD score obtained for the same measure with a F2 intercross from DeFries high- and low-activity strains (42). The same region was associated with haloperidol-induced catalepsy in BxD/DxB RI strains (43). Further analyses of the F2 intercross from DeFries strains revealed the position of the emotional reactivity locus (Emo1) at 95 cM on Chr 1 (44). A similar association on a nearby chromosomal region was obtained with open arm entries and duration as well as enclosed arm entries in the EPM together with latencies before entering the lit side of the light/dark box (45). The same region was suggestively associated with hypoactivity induced by an injection of the anticholinesterase inhibitor paraoxon in BxD/DxB RI strains (46). Likewise, provisional QTL were detected for circadian activity in the home-cage at 92.6 and 94.2 cM with the BxC/CxB RI strain (47).

Many QTL on Chr 1 seem to define a stress-related cluster involved in CV disease development, since QTL for blood pressure (BP) (48,49), atherosclerosis (50) and oxidative stress susceptibility (51) have been described in the vicinity of the Emo1 region. Table 1 illustrates the overlap with BP QTL in A/J and B6 backcrosses known as salt-induced bpq (48), and with BP QTL in A/J X B6 F2, named abbp (3). Because BP measurements were made under immobilization, one could consider these BP QTL as stress-related markers, since under radiotelemetry, mice did not habituate BP and heart rate responses, even after 10 days of exposure to immobilization (52).

Moreover, our most significant OF QTL on mouse Chr 1 was synthenic to a cluster of metabolic phenotypes of hypertension in our studies of French–Canadian families (53) and hypertensive dyslipidemic rats (54–56). Figure 5 presents homology maps related to the Chr 1 QTL. Functional polymorphisms were described in candidate genes of the CV/inflammatory/immune systems, supporting the hypothesis that an altered stress response plays a role in CV disease development (37). Validation by RT-PCR confirmed a down-regulation of the Atp1a2 gene in the informative strain for anxiety (Fig. 4). Brain differential expression of Atp1a2 may thus explain the observed enhanced emotionality (57), since heterogeneous knock out for this gene displays similar behaviour (58,59), while A/J mice show fear memory impairments with deficits in amygdalar long-term potentiation (60). Interestingly, this gene is also involved in the salt-sensitive component of hypertension, exerting a major action on cardiac output and peripheral resistance (61–63). Finally, heterozygous mutation of that gene in humans (Leu764Pro) will cause familial hemiplegic migraine 2 (64). Crp expression was not affected in our adult group. This observation was not unexpected (65), even in the presence of substantial differences between RCS progenitors revealed in our in silico analysis (Fig. 3). Nevertheless, further investigation is warranted, particularly in aged animals in which systolic hypertension can be caused by Crp-related nitric oxide and renin–angiotensin deregulation (66,67) that could possibly be amplified by chronic stress (68,69). Rgs2 fell out of our significant locus boundaries, but nevertheless has already been described as a quantitative trait gene (QTG) for the Emo1 locus on Chr 1 (70,71). Interestingly, this gene was also linked to hypertension development (49,72) and was recently reported to be differentially inherited and expressed in AJ and B6 (73).


Figure 5
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Comparative genomic view of the mouse chromosome 1 (MMU 1) region bearing QTL identified in this study (bars on the right). Selected candidate genes and QTL found in this region and syntenic regions of human Chr 1 (HSA 1) and rat Chr 13 (RNO 13) are shown. BP, blood pressure; TG, triacylglycerol; T2D, type 2 diabetes. Gene and mice QTL names are used according to NCBI’s Entrez Gene Database and Mouse Genome Informatics.

 
In addition to Chr 1 (Table 1), we noted significant linkages for emotional reactivity in the OF on Chr 3 (58.8 cM), 4 (2.5 and 59.1 cM) and 7 (64 cM). A suggestive QTL for paraoxon-induced hypoactivity were detected at the 22.7 cM locus on Chr 3 (46). For Chr 4, a provisional QTL for OF activity was identified with BxD/DxB RI strains at approximately the same site (59 cM) as one of our markers (74). Moreover, QTL on Chr 4 were found in the OF and plus-maze (45) and confirmed in the rat synthenic region on Chr 5 (75–77). From the CV perspective, our QTL on Chr 3 and 7 were near homologous regions of rat Chr 2 for QTL associated with early-onset bradycardia and later-onset tachycardia in response to an airpuff stimulus (78).

In the EPM, QTL were found on Chr 1 (32.8–36.9 cM), 7 (16–18 cM), 9 (17–25 cM), 11 (0–17 cM) and 19 (47 cM). The Chr 9 locus containing Grik4 (glutamate kainate receptor 4) and Chr 19 QTL correspond to those obtained in the OF after saline injection in AxB/BxA RI and AcB/BcA RCS (2). The Chr 9 region is also near a homologous region of rat Chr 8 associated with BP determinants in HXB-BXH RI strains (78,79), the salt-sensitive component of BP in a F2 population (80) and several metabolic syndrome-related features (81). The most significant QTL for the EPM were revealed on Chr 11 bearing Ddc (7.0 cM) that encodes dopa decarboxylase, the monoaminergic-synthesizing enzyme for dopamine, noradrenaline, and 5-hydroxytryptamine synthesis and a relevant candidate gene for affective and attention deficit disorders (82).

With the standard diet, significant QTL for stress-induced hypothermia were identified on Chr 2 (79.7–81.7 cM) and Chr 15 (9.9–24 cM). Two other QTL were associated with thermogenesis on Chr 3 (22 cM) and Chr 10 (44–52 cM). The same Chr 10 region was also detected in the A/J set for emergence latencies in the plus-maze at a lower significance level (LOD = 1.7). The common region might explain the significant correlation between these two variables (r = –0.59, P< 0.0001) and suggest a shared biological pathway. The Chr 10 region contains the Igf1 gene encoding insulin-like growth factor 1 that is likely to be linked to BP regulation (83).

On the salt-diet challenge, QTL for stress-induced hypothermia were elicited on Chr 1 (8.4 cM), 4 (6.5 and 66 cM), 15 (59.2 cM) and 18 (20 cM). The most significant QTL on Chr 4 are a locus for triglyceride level at 6.5 cM (84) near Bpq3 (48) or Abbp2 (3), and within brain cannabinoid receptor 1 gene (Cnr1) involved in both emotional processing (85), metabolic regulation (86) and, more recently, even in BP regulation (87). Both loci on Chr 4 are close to those detected in the OF; the one downstream contains Htr1d receptors that were shown to regulate redox equilibrium and the tumour necrosis factor-alpha cascade in the nervous system (88). Upstream Chr 4 caspase 8-associated protein 2 (Casp8ap2) gives another possible link to the immune response (89). Our Chr 18 QTL (22 cM) is located at the locus of Nr3c1 encoding for the glucocorticoid receptor, providing a possible link between stress and metabolic syndrome (90). Finally, Chr 15 QTL is identical to the one revealed for the acoustic startle response in the same RCS (25). This region is close to the Emo2 QTL on Chr 15 at 43.3 cM (11); Smoller et al. (91) provided evidence of linkage of human homolog Chr 12q13 with panic disorder/agoraphobia. Moreover, this region is synthenic to the rat Hsf1 gene region (http://gapp.gen.gu.se) associated with stress gene expression which we have previously identified through mapping of several Hsps mRNA as the first transcriptomic-based mapping approach reported (33).

The QTL for urinary volume were detected at a single Chr 11 site (44.8 cM). These QTL were concordant with a homologous region of rat Chr 10 associated with stress susceptibility (30,92), salt retention (34) and hypertension (93). The region also encodes Nos2 (45.6 cM), the inducible form of nitric oxide synthase near Abbp4 (58 cM).

The QTL for creatinine excretion on Chr 15 (51.1–54.5 cM) are near the one for stress-induced hypothermia under a high-salt load. The most significant QTL for Ca++ excretion (LOD = 4.6) on Chr 12 (48–50 cM) bear Calm1 encoding calmodulin 1, the overexpression of which induces cardiac hypertrophy (94).

In the context of genetic hypertension, we have also suggested that stress may be considered as a special case of gene/environment interaction in which some gene expression manifests susceptibility to the environment (32,33,39). In this perspective, we report phenotype-specific interactions between salt-loading and genetic background that may recall the marked heterogeneity of neuroendocrine responses to various stressors (95). In contrast, the overlapping linkage of Emo, stress susceptibility and hypertension to homologous genomic regions may be explained by pleiotropy or common pathways related to stress susceptibility; some of them might be grouped in gene clusters (53,96).

In silico sequence analysis (97–99) of selected candidate genes in A/J and B6 strains led us to hypothesis-driven functional genomics which can be a filter to transcriptome-based systematic screening studies (100), and confirms the relevance of considering polymorphisms in regulatory elements. Clearly, the loci determined in the current study represent an important stepping stone for further detailed analyses, as the actual relevance of particular polymorphisms within genes and intergenic regions for the resulting phenotypes is yet to be fully explored. In another report, a multiple regression model was used to trace donor alleles associated with gene expression profiles in RCS (73). SNPs in the 3'-UTR were present in 64% (versus 29%) of genes showing such associations. Some of them are related to transcription factors, such as Usf1, or can be confirmed by an allelic imbalance test, as for Rgs2 gene. Our identification of stress-related QTL and functional genomic variations with relevant informative strains is valuable to the genome-wide cataloguing of regulatory variants in RCS, a powerful tool for the dissection of complex diseases (73). Future perspectives of research include inter-specific F2 crosses or congenic strain production with relevant informative RCS to confirm these provisional QTL in specific regions of interest.

Overall, our studies described here are part of our global pursuit of an integrated approach to the resolution of such complex diseases (1) as hypertension, with its hemodynamic, metabolic, humoral and immune components, forming a link between atherosclerosis, ischemic heart disease, diabetes, dyslipidemias and obesity in a newly proposed concept of a graph-theoretic network of phenotypes and their associated diseases, the human diseasome (101).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Animals
RCS male mice from 14 lines primarily on the A/J background (AcB) and 22 lines primarily on the B6 background (BcA) were randomly provided with their founders by the McGill University Health Centre (MUHC) Research Institute to CHUM—Hôtel-Dieu at 3 months of age (n = 7 per strain). The RCS were originally obtained by backcrossing F1 progeny with either founder line, followed by at least 18 rounds of inbreeding to fixate allelic combinations with more than 99% homozygosity according to genotyping analysis (7). The inbred progeny contained ~12.5% of the genome of one parental line introduced in the other. The mice were group-housed in cages with wood shavings and in a temperature- and humidity-controlled room with a 12 h light–dark cycle (lights on at 0700 h). After at least 1 week of adaptation to their new surroundings and to handling, the mice were evaluated during the diurnal cycle in a separate room. The procedures were in accordance with the guidelines of the Canadian Council on Animal Care and were approved by the Animal Ethics Committees at both universities.

Exploration of environmental stimuli
Motor activity was measured in an OF made of white plastic (30 cm x 25 cm, wall height: 15 cm), with the floor marked by a felt pen into six equally spaced rectangles. The number of segments crossed (4-paw criterion) in a single 4 min session and the number of fecal boli were counted. Each value was standardized to a mean of 0 and a variance of 1. A joint trait of Emo was calculated by subtracting standardized segments crossed from fecal boli and dividing them by 2, as reported previously (44). On the following day, the mice were evaluated in the EPM made of Plexiglas, consisting of 4 arms (length: 24 cm, width: 10 cm, height from floor: 30 cm) and a 10 cm2 central region. Two of the arms were enclosed by opaque walls (height: 10 cm), and the other two were not. The two enclosed arms and the two open arms faced each other on opposite sides. The mice were placed in 1 of the enclosed arms, and their emergence from that arm was measured with a hand-held stopwatch in a single 5 min session, together with the number of entries and the time spent in either enclosed or open arms (4-paw criterion). In addition, open/total arm entries and duration were calculated.

Radiotelemetry measures of stress-induced hypothermia before and after a high-salt diet
The mice were anesthetized by inhalation of Isoflurane at 3% and maintained at 2%. The body of the radiotelemetry probe (1.75 cc) (PhysioTel TA10TA-F20 model, Data Sciences, St. Paul, MN, USA) was implanted in the abdominal cavity. After surgery, the mice received an analgesic (Buprenorphine, 0.05 mg/kg SC, with 0.001 ml/g of 0.9% saline at 37°C) for 3 days and left to recuperate for 18 more days in isolated cages. To stimulate their appetite, they were given chocolate-flavoured Ensure (1.06 mg/ml of sodium, Ross Products Division, Abbott Limited, Montreal, Quebec, Canada) mixed with their regular chow (Purina rat 5012-M, 0.28% sodium), and FruitLoopsTM. After 5 min baseline body temperature recordings taken in their home-cage, the mice were immobilized for 30 min in a restraining device (IITC Life Science Inc., Woodland Hills, CA, USA). Two to 3 days later, they were placed in metabolic cages and fed an 8% salt-rich diet (Purina Modified Lab. Rodent Testdiet 5001C-2). The procedure was then repeated. The number of deaths associated with the surgical and restraint procedures was near zero in this cohort. There was no apparent relationship between survival and genotype after each procedure.

Urinary volume and electrolytes with a salt diet
Twenty-four-hour urine samples were collected from each mouse placed in metabolic cages under the high-salt diet. Urinary volume and electrolyte excretion (Na+, K+, Ca++ and creatinine) were measured by spectrophotometry (102,103), with values corrected for body weight (34).

Genotyping
The strains were genotyped for 625 microsatellite markers with an average distance of 2.6 cM (range 0 –15 cM) throughout the genome, as described previously (7).

Statistical analyses
The SDP of each phenotype was determined by standard descriptive statistics and heritability estimates (interstrain/intrastrain variance ratio), as described by Owen et al. (104). Informative strains, differing from their recurrent parental strain, were identified by an unpaired t-test or with Sattertwaithe-Welsh correction for heterogeneous variances meeting Bonferroni correction criteria for multiple comparisons, i.e. with P< 0.002 for the B6 background (0.05 divided by 22 strains) and P< 0.004 for the A/J background. Only phenotypes with heritability estimates over 50% for either background were retained for further analyses. QTL were first determined by computing single marker regressions with MapManager QTX software, using strain means and variances (105). The map distances were defined for all loci with Mouse Genome Informatics (htpp://www.informatics.jax.org) positions instead of calculated positions, and redundant markers were removed for the best fit to a self-RI model, which implicitly considers each background (albino or B6) as an independent RI set. Permuitation tests (5,000 for single marker additive contributions) allowed us to calculate significant thresholds for each phenotype, adjusted for multiple testing (106). Interval mapping was then performed to obtain a likelihood ratio statistic divided by 4.61 to provide a LOD score. After reinserting the removed markers into the map, fine mapping of QTL intervals was appraised by visual tracking of haplotypes in contributing and non-contributing strains (45,107). In addition, cross confirmation of the ‘direction of effect’ between the two RCS sets was assessed (108). Significant QTL were confirmed by parametric t-tests, with genomic markers serving as the independent variable, and strain means as observations (25,109,110), controlling for equality of variances with a F test ratio (Statview 5.0).

In silico exploration of putative candidate genes
Figure 3 describes our whole approach for narrowing down QTL to QTG. For the Chr 1 QTL interval (171–174 Mb), candidate genes having SNPs polymorphic between AJ and B6 were listed with the help of the Celera Discovery System (http://cds.celera.com). QTL interval boundaries were translated in base pair location according to the National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov/). For all other QTL, candidate genes of interest were selected from evidence-based physiological relevance, starting with the Entrez Gene NCBI database. A/J and B6 progenitors were then screened with the public inbred strain SNP database from Genome Build 36.1 (http://www.ncbi.nlm.nih.gov/SNP/MouseSNP.cgi), to find genes having SNPs within coding and untranslated mRNA sequences, or within 2 kb in the promoter region. The SNPs exact location was validated with the variation map option of NCBI Map Viewer and Blast 2 sequence tools.

Depending on their location in the coding or regulatory regions (promoter or UTRs), the functional significance of the SNPs was respectively assessed by looking for protein domains (PROSITE patterns) or DNA motifs (TRANSFAC vertebrates motif Library) in the relevant database (http://motif.genome.jp/). Moreover, the impact of the SNPs in the 3'-UTR was further estimated with the Vienna RNA-fold program web interface (http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi/). This program allows for the prediction and visualization of occurrence of mRNA secondary structure changes between A/J and B6 haplotypes, covering all SNPs within the gene sequence. In addition, a home-made pattern recognition program was built from 8-mers 3'-UTR target consensus motifs for microRNAs (111), particularly at the location of secondary structure changes. AUUUA (112,113) and AAUAAA (112) motifs were also added to our pattern library. Stem-loop secondary structure, mRNA and regulatory-binding protein are all working as a structural barrier to translation (112,114); we therefore expect these features to potentially affect gene expression.

Preparation of total RNA and RT–PCR
According to haplotype distribution (top right table in Figure 3), BcA76 and AcB58 strains were chosen as negative controls and BcA70 as the informative strain for Emo, in addition to A/J and B6 progenitors. Mouse brain, heart, liver and kidneys were removed and kept at –80°C until RNA extraction. Target organs were selected for a detectable level of gene expression according to NCBI Entrez Gene profiles. Total RNA was extracted with TRIzol reagent (Invitrogen Canada Inc., Burlington, Ontario). Total RNA (1 µg) was reverse-transcribed with random hexamer primer and M-MLV reverse transcriptase (Invitrogen). The Atp1a2 (forward primer 5'-ggcatcaaggtgatcatgg-3', reverse primer 5'-gctgcaatgtcttccacagt-3'), Slamf1 (forward primer 5'-cctcctcagaatcgagtcca-3', reverse primer 5'-ggctggcagtgatttgattt-3'), Crp (forward primer 5'-ggacttttggtcatgaagacat-3', reverse primer 5'-ggtgttcagtggcttctttg-3'), Fcgr3 (forward primer 5'-tgatttctgactggctgctg-3', reverse primer 5'-cattatggaagaacgagatcctg-3') genes and a housekeeping gene (18S) were selected and run in parallel for quantitative RT–PCR analysis. The level of expression of Nhlh1 was insufficient to perform any comparison. Primers were designed with Primer3 (115) and synthesized by Invitrogen. RT–PCRs were performed with Platinum SYBR Green qPCR SuperMix-UDG kits (Invitrogen) in strips of 0.1 ml optical tubes (Corbett Research Pty Ltd., Sydney, Australia) according to the manufacturer’s protocol, using the Rotor-Gene 3000 system (Corbett Research). The results were analysed by the Pfaffl method (116). The expression of a given gene in the A/J ancestor strain was chosen as the control, to which each strain was then compared (n = 4 per strain).


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
FUNDING
 REFERENCES
 
Supplementary Material is available at HMG Online.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
REFERENCES
 
S.T. received a doctoral scholarship from the Canadian Heart and Stroke Foundation. This study was supported by grants from the Canadian Institutes of Health Research to P.H. and to J.T. (CIHR MT-14654, Cardiogene).


   ACKNOWLEDGEMENTS
 
Methodological discussions with A.E. Boyle (from K.J. Gill’s laboratory), K. Morgan and R. Joober (McGill University) were very valuable and helpful. The statistical advice obtained from the SEMQ (Service d’Évaluation en Méthodes Quantitatives), Université de Montréal, is much appreciated, and the editorial assistance of Ovid Da Silva, Research Support Office, Research Centre, CHUM, is acknowledged. RCS were kindly provided by Emerillon Therapeutics Inc.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 

  1. Seda O., Tremblay J., Sedova L., Hamet P. Integrating genomics and transcriptomics with geo-ethnicity and the environment for the resolution of complex cardiovascular diseases. Curr. Opin. Mol. Ther. (2005) 7:583–587.[Web of Science][Medline]

  2. Gill K.J., Boyle A.E. Quantitative trait loci for novelty/stress-induced locomotor activation in recombinant inbred (RI) and recombinant congenic (RC) strains of mice. Behav. Brain Res. (2005) 161:113–124.[CrossRef][Web of Science][Medline]

  3. Woo D.D., Kurtz I. Mapping blood pressure loci in (A/J x B6)F2 mice. Physiol. Genomics (2003) 15:236–242.[Abstract/Free Full Text]

  4. Smoller J.W., Rosenbaum J.F., Biederman J., Susswein L.S., Kennedy J., Kagan J., Snidman N., Laird N., Tsuang M.T., Faraone S.V., et al. Genetic association analysis of behavioral inhibition using candidate loci from mouse models. Am. J. Med. Genet. (2001) 105:226–235.[CrossRef][Web of Science][Medline]

  5. Fullerton J. New approaches to the genetic analysis of neuroticism and anxiety. Behav. Genet. (2006) 36:147–161.[CrossRef][Web of Science][Medline]

  6. Imumorin I.G., Dong Y., Zhu H., Poole J.C., Harshfield G.A., Treiber F.A., Snieder H. A gene-environment interaction model of stress-induced hypertension. Cardiovasc. Toxicol. (2005) 5:109–132.[CrossRef][Web of Science][Medline]

  7. Fortin A., Diez E., Rochefort D., Laroche L., Malo D., Rouleau G.A., Gros P., Skamene E. Recombinant congenic strains derived from A/J and C57BL/6J: a tool for genetic dissection of complex traits. Genomics (2001) 74:21–35.[CrossRef][Web of Science][Medline]

  8. Bolivar V.J., Caldarone B.J., Reilly A.A., Flaherty L. Habituation of activity in an open field: a survey of inbred strains and F1 hybrids. Behav. Genet. (2000) 30:285–293.[CrossRef][Web of Science][Medline]

  9. Garrett K.M., Niekrasz I., Haque D., Parker K.M., Seale T.W. Genotypic differences between C57BL/6 and A inbred mice in anxiolytic and sedative actions of diazepam. Behav. Genet. (1998) 28:125–136.[CrossRef][Web of Science][Medline]

  10. Logue S.F., Owen E.H., Rasmussen D.L., Wehner J.M. Assessment of locomotor activity, acoustic and tactile startle, and prepulse inhibition of startle in inbred mouse strains and F1 hybrids: implications of genetic background for single gene and quantitative trait loci analyses. Neuroscience (1997) 80:1075–1086.[CrossRef][Web of Science][Medline]

  11. Gershenfeld H.K., Neumann P.E., Mathis C., Crawley J.N., Li X., Paul S.M. Mapping quantitative trait loci for open-field behavior in mice. Behav. Genet. (1997) 27:201–210.[CrossRef][Web of Science][Medline]

  12. Thifault S., Lalonde R., Sanon N., Hamet P. Comparisons between C57BL/6J and A/J mice in motor activity and coordination, hole-poking, and spatial learning. Brain Res. Bull. (2002) 58:213–218.[CrossRef][Web of Science][Medline]

  13. Trullas R., Skolnick P. Differences in fear motivated behaviors among inbred mouse strains. Psychopharmacology (Berl.) (1993) 111:323–331.[CrossRef][Medline]

  14. van Gaalen M.M., Steckler T. Behavioural analysis of four mouse strains in an anxiety test battery. Behav. Brain Res. (2000) 115:95–106.[CrossRef][Web of Science][Medline]

  15. Bouwknecht J.A., Paylor R. Behavioral and physiological mouse assays for anxiety: a survey in nine mouse strains. Behav. Brain Res. (2002) 136:489–501.[CrossRef][Web of Science][Medline]

  16. Gershenfeld H.K., Paul S.M. Mapping quantitative trait loci for fear-like behaviors in mice. Genomics (1997) 46:1–8.[CrossRef][Web of Science][Medline]

  17. Bolivar V.J., Pooler O., Flaherty L. Inbred strain variation in contextual and cued fear conditioning behavior. Mamm. Genome (2001) 12:651–656.[CrossRef][Web of Science][Medline]

  18. Cohen R.M., Kang A., Gulick C. Quantitative trait loci affecting the behavior of A/J and CBA/J intercross mice in the elevated plus maze. Mamm. Genome (2001) 12:501–507.[CrossRef][Web of Science][Medline]

  19. Mathis C., Neumann P.E., Gershenfeld H., Paul S.M., Crawley J.N. Genetic analysis of anxiety-related behaviors and responses to benzodiazepine-related drugs in AXB and BXA recombinant inbred mouse strains. Behav. Genet. (1995) 25:557–568.[CrossRef][Web of Science][Medline]

  20. Gill K., Boyle A., Lake K., Desaulniers N. Alcohol-induced locomotor activation in C57BL/6J, A/J, and AXB/BXA recombinant inbred mice: strain distribution patterns and quantitative trait loci analysis. Psychopharmacology (Berl.) (2000) 150:412–421.[CrossRef][Medline]

  21. Boyle A.E., Gill K. Sensitivity of AXB/BXA recombinant inbred lines of mice to the locomotor activating effects of cocaine: a quantitative trait loci analysis. Pharmacogenetics (2001) 11:255–264.[CrossRef][Web of Science][Medline]

  22. Demant P., Hart A.A.M. Recombinant congenic strains—a new tool for analyzing genetic traits determined by more than one gene. Immunogenetics (1986) 24:416–422.[CrossRef][Web of Science][Medline]

  23. Fortin A., Cardon L.R., Tam M., Skamene E., Stevenson M.M., Gros P. Identification of a new malaria susceptibility locus (Char4) in recombinant congenic strains of mice. Proc. Natl Acad. Sci. USA (2001) 98:10793–10798.[Abstract/Free Full Text]

  24. Roy M.F., Riendeau N., Loredo-Osti J.C., Malo D. Complexity in the host response to Salmonella Typhimurium infection in AcB and BcA recombinant congenic strains. Genes Immun. (2006) 7:655–666.[CrossRef][Web of Science][Medline]

  25. Joober R., Zarate J.M., Rouleau G.A., Skamene E., Boksa P. Provisional mapping of quantitative trait loci modulating the acoustic startle response and prepulse inhibition of acoustic startle. Neuropsychopharmacology (2002) 27:765–781.[CrossRef][Web of Science][Medline]

  26. Gill K., Boyle A.E. Genetic analysis of alcohol intake in recombinant inbred and congenic strains derived from A/J and C57BL/6J progenitors. Mamm. Genome (2005) 16:319–331.[CrossRef][Web of Science][Medline]

  27. Torkamanzehi A., Boksa P., Ayoubi M., Fortier M.E., Ng Ying Kin N.M., Skamene E., Rouleau G., Joober R. Identification of informative strains and provisional QTL mapping of amphetamine (AMPH)-induced locomotion in recombinant congenic strains (RCS) of mice. Behav. Genet. (2006) 36:903–913.[CrossRef][Web of Science][Medline]

  28. Gill K.J., Boyle A.E. Confirmation of quantitative trait loci for cocaine-induced activation in the AcB/BcA series of recombinant congenic strains. Pharmacogenetics (2003) 13:329–338.[CrossRef][Web of Science][Medline]

  29. Gill K.J., Boyle A.E. Genetic basis for the psychostimulant effects of nicotine: a quantitative trait locus analysis in AcB/BcA recombinant congenic mice. Genes Brain Behav. (2005) 4:401–411.[CrossRef][Web of Science][Medline]

  30. Dumas P., Pausova Z., Kren V., Krenova D., Pravenec M., Dumont M., Ely D., Turner M., Sun Y., Tremblay J., Hamet P. Contribution of autosomal loci and the Y chromosome to the stress response in rat. Hypertension (2000) 35:568–573.[Abstract/Free Full Text]

  31. Pausova Z., Sedova L., Berube J., Hamet P., Tremblay J., Dumont M., Gaudet D., Pravenec M., Kren V., Kunes J. Segment of rat chromosome 20 regulates diet-induced augmentations in adiposity, glucose intolerance, and blood pressure. Hypertension (2003) 41:1047–1055.[Abstract/Free Full Text]

  32. Malo D., Schlager G., Tremblay J., Hamet P. Thermosensitivity, a possible new locus involved in genetic hypertension. Hypertension (1989) 14:121–128.[Abstract/Free Full Text]

  33. Dumas P., Sun Y., Corbeil G., Tremblay S., Pausova Z., Kren V., Krenova D., Pravenec M., Hamet P., Tremblay J. Mapping of quantitative trait loci (QTL) of differential stress gene expression in rat recombinant inbred strains. J. Hypertens. (2000) 18:545–551.[Web of Science][Medline]

  34. Dumas P., Kren V., Krenova D., Pravenec M., Hamet P., Tremblay J. Identification and chromosomal localization of ecogenetic components of electrolyte excretion. J. Hypertens. (2002) 20:209–217.[CrossRef][Web of Science][Medline]

  35. Hamet P., Kong D., Pravenec M., Kunes J., Kren V., Klir P., Sun Y.L., Tremblay J. Restriction fragment length polymorphism of hsp70 gene, localized in the RT1 complex, is associated with hypertension in spontaneously hypertensive rats. Hypertension (1992) 19:611–614.[Free Full Text]

  36. Pausova Z., Kunes J., Kren V., Krenova D., Pravenec M., Tremblay J., Hamet P. Contribution of the TNF alpha gene region of rat chromosome 20 to the body temperature response to endotoxin. Transplant. Proc. (1999) 31:1622–1623.[CrossRef][Web of Science][Medline]

  37. Hamet P., Tremblay J. Genetic determinants of the stress response in cardiovascular disease. Metabolism (2002) 51:15–24.[CrossRef][Web of Science][Medline]

  38. Leon L.R., Kozak W., Rudolph K., Kluger M.J. An antipyretic role for interleukin-10 in LPS fever in mice. Am. J. Physiol. (1999) 276:R81–R89.[Web of Science][Medline]

  39. Pravenec M., Sun Y.L., Kunes J., Kong D., Kren V., Klir P., Tremblay J., Hamet P. Environmental susceptibility in hypertension: potential role of HSP70 and TNF{alpha} genes. J. Vasc. Med. Biol. (1991) 3:297–302.

  40. Ezzati M., Lopez A.D., Rodgers A., Vander H.S., Murray C.J. Selected major risk factors and global and regional burden of disease. Lancet (2002) 360:1347–1360.[CrossRef][Web of Science][Medline]

  41. Singer J.B., Hill A.E., Nadeau J.H., Lander E.S. Mapping quantitative trait loci for anxiety in chromosome substitution strains of mice. Genetics (2005) 169:855–862.[Abstract/Free Full Text]

  42. Flint J., Corley R., DeFries J.C., Fulker D.W., Gray J.A., Miller S., Collins A.C. A simple genetic basis for a complex psychological trait in laboratory mice. Science (1995) 269:1432–1435.[Abstract/Free Full Text]

  43. Patel N.V., Hitzemann R.J. Detection and mapping of quantitative trait loci for haloperidol-induced catalepsy in a C57BL/6J x DBA/2J F2 intercross. Behav. Genet. (1999) 29:303–310.[CrossRef][Web of Science][Medline]

  44. Turri M.G., Talbot C.J., Radcliffe R.A., Wehner J.M., Flint J. High-resolution mapping of quantitative trait loci for emotionality in selected strains of mice. Mamm. Genome (1999) 10:1098–1101.[CrossRef][Web of Science][Medline]

  45. Turri M.G., Henderson N.D., DeFries J.C., Flint J. Quantitative trait locus mapping in laboratory mice derived from a replicated selection experiment for open-field activity. Genetics (2001) 158:1217–1226.[Abstract/Free Full Text]

  46. Risinger F.O., Quick E., Belknap J.K. Quantitative trait loci for acute behavioral sensitivity to paraoxon. Neurotoxicol. Teratol. (2000) 22:667–674.[CrossRef][Web of Science][Medline]

  47. Mayeda A.R., Hofstetter J.R., Belknap J.K., Nurnberger J.I. Jr. Hypothetical quantitative trait loci (QTL) for circadian period of locomotor activity in CXB recombinant inbred strains of mice. Behav. Genet. (1996) 26:505–511.[CrossRef][Web of Science][Medline]

  48. Sugiyama F., Churchill G.A., Higgins D.C., Johns C., Makaritsis K.P., Gavras H., Paigen B. Concordance of murine quantitative trait loci for salt-induced hypertension with rat and human loci. Genomics (2001) 71:70–77.[CrossRef][Web of Science][Medline]

  49. DiPetrillo K., Tsaih S.W., Sheehan S., Johns C., Kelmenson P., Gavras H., Churchill G.A., Paigen B. Genetic analysis of blood pressure in C3H/HeJ and SWR/J mice. Physiol. Genomics (2004) 17:215–220.[Abstract/Free Full Text]

  50. Phelan S.A., Beier D.R., Higgins D.C., Paigen B. Confirmation and high resolution mapping of an atherosclerosis susceptibility gene in mice on Chromosome 1. Mamm. Genome (2002) 13:548–553.[CrossRef][Web of Science][Medline]

  51. Wang X., Phelan S.A., Forsman-Semb K., Taylor E.F., Petros C., Brown A., Lerner C.P., Paigen B. Mice with targeted mutation of peroxiredoxin 6 develop normally but are susceptible to oxidative stress. J. Biol. Chem. (2003) 278:25179–25190.[Abstract/Free Full Text]

  52. Gross V., Luft F.C. Exercising restraint in measuring blood pressure in conscious mice. Hypertension (2003) 41:879–881.[Free Full Text]

  53. Hamet P., Merlo E., Seda O., Broeckel U., Tremblay J., Kaldunski M., Gaudet D., Bouchard G., Deslauriers B., Gagnon F., et al. Quantitative founder-effect analysis of French Canadian families identifies specific loci contributing to metabolic phenotypes of hypertension. Am. J. Hum. Genet. (2005) 76:815–832.[CrossRef][Web of Science][Medline]

  54. Ueno T., Tremblay J., Kunes J., Zicha J., Dobesova Z., Pausova Z., Deng A.Y., Sun Y.L., Jacob H.J., Hamet P. Rat model of familial combined hyperlipidemia as a result of comparative mapping. Physiol. Genomics (2004) 17:38–47.[Abstract/Free Full Text]

  55. Hubner N., Wallace C.A., Zimdahl H., Petretto E., Schulz H., Maciver F., Mueller M., Hummel O., Monti J., Zidek V., et al. Integrated transcriptional profiling and linkage analysis for identification of genes underlying disease. Nat. Genet. (2005) 37:243–253.[CrossRef][Web of Science][Medline]

  56. Stoll M., Kwitek-Black A.E., Cowley A.W. Jr, Harris E.L., Harrap S.B., Krieger J.E., Printz M.P., Provoost A.P., Sassard J., Jacob H.J. New target regions for human hypertension via comparative genomics. Genome Res. (2000) 10:473–482.[Abstract/Free Full Text]

  57. Matthews D.B., Bhave S.V., Belknap J.K., Brittingham C., Chesler E.J., Hitzemann R.J., Hoffmann P.L., Lu L., McWeeney S., Miles M.F., et al. Complex genetics of interactions of alcohol and CNS function and behavior. Alcohol Clin. Exp. Res. (2005) 29:1706–1719.[CrossRef][Medline]

  58. Ikeda K., Onaka T., Yamakado M., Nakai J., Ishikawa T.O., Taketo M.M., Kawakami K. Degeneration of the amygdala/piriform cortex and enhanced fear/anxiety behaviors in sodium pump alpha2 subunit (Atp1a2)-deficient mice. J. Neurosci. (2003) 23:4667R