Skip Navigation


Human Molecular Genetics Advance Access originally published online on September 30, 2003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
12/23/3097    most recent
ddg333v1
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 arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Le Corvoisier, P.
Right arrow Articles by Rockman, H. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Le Corvoisier, P.
Right arrow Articles by Rockman, H. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 23 3097-3107
DOI: 10.1093/hmg/ddg333
© 2003 Oxford University Press

Multiple quantitative trait loci modify the heart failure phenotype in murine cardiomyopathy

Philippe Le Corvoisier1,2, Hyun-Young Park1,2, Kerri M. Carlson2, Douglas A. Marchuk2 and Howard A. Rockman1,2,*

1Department of Medicine and 2Department of Molecular Genetics and Microbiology, Duke University Medical Center, Durham, NC, 27710, USA

Received July 29, 2003; Revised September 17, 2003; Accepted September 23, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The variability in outcome of heart failure patients depends on a number of factors including differences in their genetic background. To identify novel genes that modify the human heart failure phenotype, we used a strategy of quantitative trait locus (QTL) mapping in an experimental mouse model of dilated cardiomyopathy induced by cardiac-specific overexpression of calsequestrin and characterized by a strong strain-specific variability in the phenotype. We identified two novel QTLs, Hrtfm3 (heart failure modifier 3) on chromosome (Chr) 4 and Hrtfm4 on Chr 18, significantly linked to survival with likelihood ratio statistics (LRS) of 19.9 and 23.6 respectively (corresponding to LOD scores of 4.3 and 5.1). Two other QTLs, Hrtfm5 on Chr 2 and Hrtfm6 on Chr 13, were significantly linked to cardiac function as measured by echocardiographic fractional shortening (LRS 22.1 and 15.2 respectively, LOD score 4.8 and 3.3) and left ventricular end-diastolic diameter (LRS 23.5 and 18.8, LOD score 5.1 and 4.1). Importantly, Hrtfm5 was not significantly linked to survival. A significant interaction was found between Hrtfm4 and two other QTLs (Hrtfm6 and a QTL near to the marker D19Mit88) for fractional shortening with a LRS of 34.6 and 26.5 respectively (LOD score 7.5 and 5.8). These data show that the effect of genetic background on murine heart failure is complex and result from the action of several loci that differentially modify the cardiac phenotype. The identification of these novel modifier genes will serve as strong candidates for the discovery of modifiers in human heart failure.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Heart failure is a major public health problem affecting nearly 5 million patients in the USA (1). Despite the development of novel treatments, the prognosis of these patients remains poor with a one year mortality estimate as high as 42% (2). Heart failure commonly presents with a wide range of symptoms and outcome between patients that cannot be completely explained by environmental factors. Increasing evidence suggests that differences in genetic background can account for an important part of this variability (3–9). The effect of genetic background on outcome is due to the existence of modifier genes whose multiple alleles differentially affect the heart failure phenotype (10).

The identification of modifier genes is a complex task that requires multiple and complementary approaches. The strategy most commonly used is based on association studies between the severity of measured heart failure traits and genetic variation of selected candidate genes (3–8). While some studies have reported a correlation between the survival of heart failure patients and polymorphisms of several genes (3–8), their statistical power is limited since parameters other than genetic background contribute to the phenotypic variability (etiology, treatment and environmental factors) (11). In addition, this strategy cannot identify epistatic interactions between different modifier genes (12). Therefore, it is unlikely that a purely candidate gene approach will be able to identify novel heart failure modifier genes among the ~25 000 cardiovascular genes (13).

By contrast, linkage studies provide an unbiased approach to localize in the genome the genes responsible for a phenotype without prior hypotheses on their function in the pathophysiology of the disease (14). This strategy has allowed the identification of a number of mutations responsible for monogenic diseases such as familial dilated cardiomyopathy (15). However, its application in humans for the identification of modifier genes is more complex because of its low statistical power due to the modest phenotypic effect exerted by modifier genes and to the heterogeneity of the population (11).

To overcome this difficulty, a genetic linkage strategy can be performed using animal models that are sensitized to develop the disease phenotype. The authors and others (9,11,16) have reported that a large number of experimental murine models show strong strain-specific phenotypic variability due to differences in genetic background. A backcross and/or intercross strategy among the different strains using disease-sensitized animals allows mapping of the loci contributing to these phenotypic differences. In addition, this approach allows standardization of environmental variables that might otherwise confound the analysis.

In order to map novel modifier genes affecting the heart failure phenotype, we used an experimental murine model of heart failure induced by cardiac specific overexpression of calsequestrin (CSQ), a sarcoplasmic reticulum calcium binding protein (9,17). We conducted a genome-wide scan in N2 CSQ progeny resulting from a backcross strategy between two genetically distant inbred strains, DBA/2J (DBA) and AKR/J (AKR), and identified a number of quantitative trait loci (QTL) that are linked to the severity of heart failure. Our results should lead to the identification of novel candidates that can be tested as heart failure modifier genes in human studies.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Effect of the genetic background on heart failure phenotype
Cardiac specific overexpression of calsequestrin resulted in dilated cardiomyopathy and premature death in all transgenic mice. However, the severity of the phenotype was strongly dependent on the strain indicating an effect of genetic background (Fig. 1A). While the average survival of DBA CSQ mice was limited to 156.9±49.2 days, the survival of the hybrid F1 CSQ progeny showed a significant increase to 230.5±62.9 days (P<0.001). The improvement in survival between homozygous (DBA/DBA) and heterozygous (AKR/DBA) animals indicates that the AKR genetic background has a protective effect on the phenotype. Compared with the F1 CSQ progeny, N2 CSQ showed an additional increase in average survival from 230.5±62.9 to 290.0±82.9 days (P<0.001), however with a wide distribution of the phenotypic values. The effect of genetic background is estimated to be responsible for 53.7% of the total variability of survival in the N2 CSQ progeny. These results suggest a segregation of the phenotype depending on the genotype for putative modifier genes (AKR/DBA or AKR/AKR), allowing us to map the QTLs responsible for this variability. As for many complex phenotypic traits in a genetically heterogeneous population, the survival of N2 CSQ did not segregate into two distinct groups, but rather showed a continuous distribution indicating the presence of multiple QTLs.



View larger version (42K):
[in this window]
[in a new window]
  		Figure 1. Scatter plots showing the relationship between the phenotype of CSQ mice and their genetic background in an AKRxDBA backcross. A strain-dependent effect was observed for survival (A), fractional shortening (B) and left ventricular end-diastolic diameter (C) with a protective effect of the AKR genetic background. Mice from the N2 CSQ progeny showed a wide distribution of their phenotypic trait values suggesting segregation depending on their genotype at the putative modifier loci. Close circles represent the average±SD for each group. **P<0.001; *P<0.05. (D) A typical echocardiogram for each of the three different generations of this backcross. Note the pulsus alternans characteristic of CSQ overexpression (40).

 
Measurement of cardiac function by echocardiography showed the same strong strain-specific variability as survival (Fig. 1B–D). The average left ventricular end diastolic diameter (LVEDD) and fractional shortening (FS) of F1 CSQ progeny were significantly better compared with mice on a pure DBA genetic background (4.6±0.2 versus 5.5±0.4 mm for LVEDD, P<0.001 and 22.1±4.6 versus 15.3±4.6% for FS, P<0.05). Moreover, additional improvement was found between the F1 CSQ and the N2 CSQ progeny (4.6±0.2 versus 4.2±0.4 for LVEDD, P<0.001 and 22.1±4.6 versus 27.1±7.6 for FS, P<0.05). The genetic variation is estimated to be responsible for 49.7 and 63.0% of the LVEDD and FS variability among the N2 CSQ progeny.

The relationship between survival, LVEDD and FS for the N2 CSQ mice is shown in Figure 2. Survival was significantly correlated with both LVEDD (P<0.0001) and FS (P<0.0001) (r-values -0.42 and 0.49, respectively). However, some mice showed discordance between echocardiographic parameters and survival, indicating that in addition to systolic function, the occurrence of sudden cardiac death likely influenced survival.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. Scatter plots showing the relationship between the survival of N2 CSQ mice and their echocardiographic parameters measured at 12 weeks of age. (A) Fractional shortening; (B) left ventricular end-diastolic diameter (n=114).

 
Importantly, we found no differences between the level of overexpression of CSQ in hearts with a different genetic background, indicating that the variation in the phenotype cannot simply be accounted for by variation in the level of protein expression (Fig. 3A). Finally, the echocardiographic parameters of wild-type DBA or AKR mice were not different, indicating no preexisting differences in cardiac function in the absence of the CSQ transgene (Fig. 3B and C).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 3. (A) Western blot showing an equal level of calsequestrin overexpression in mice with different genetic backgrounds and with different ages. (B and C) Comparison of the echocardiographic parameters measured at 12 weeks of age in wild type mice depending on their genetic background (AKR or DBA). (B) Fractional shortening; (C) left ventricular end-diastolic diameter (n=10). No strain-dependent differences were observed in wild-type mice, suggesting that the effect of the genetic background in CSQ mice reflects differences in the progression of heart failure and not pre-existing strain-dependent variations. The values are reported as mean±SD.

 
Identification of QTLs
Low resolution linkage maps were constructed by genotyping N2 CSQ mice with the most extreme phenotype for survival (n=70), FS (n=71) and LVEDD (n=60). Three loci on chromosomes (Chrs) 2, 13 and 18 were significantly linked with the variability of the different phenotypic traits (Fig. 4). In addition, one locus with a likelihood ratio statistic (LRS) value suggestive for a linkage with survival was localized on Chr 4. On the basis of the preliminary genetic linkage maps, we performed high-resolution mapping within each of the four potential QTLs using 35 additional microsatellite markers and the entire N2 CSQ progeny (n=155 for survival and n=130 for echocardiographic parameters). The results were analyzed by composite interval mapping and are represented in Figures 5 and 6.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 4. Genetic linkage maps performed with the N2 CSQ mice with the most extreme phenotype using survival (A) (n=70), fractional shortening (B) (n=71) and left ventricular end-diastolic diameter (C) (n=60) as phenotypic traits. Four loci on Chrs 2, 4, 13 and 18 showing a suggestive or significant linkage with the different phenotypic trait values were identified. Each vertical bar represents the level of significance for one of the 106 microsatellite markers and is expressed as a likelihood ratio statistic. A LRS can be converted into a LOD score by dividing it by 4.6. The horizontal dotted lines represent the significance thresholds and were determined by permutation testing.

 


View larger version (18K):
[in this window]
[in a new window]
  		Figure 5. Composite interval mapping across the entire Chrs 4 and 18. Two QTLs, Hrtfm3 (centered on D4Mit236) (A) and Hrtfm4 (centered on D18Mit28) (B), showed significant linkage with the survival of heart failure mice (maximal LRS of 19.9 and 23.6 respectively; n=155). While Hrtfm3 was linked only to survival, Hrtfm4 showed an additional significant linkage with fractional shortening (LRS: 17.7; n=130; C). The position of the microsattelite markers is indicated using the function of Kosambi. The limits of each QTL support interval are indicated by the solid box.

 


View larger version (47K):
[in this window]
[in a new window]
  		Figure 6. Composite interval mapping across the entire Chrs 2 and 13. Two QTLs on Chrs 2 and 13 (centered on D2Mit138 and D13Mit213 respectively), named Hrtfm5 and Hrtfm6, were significantly linked with left ventricular end-diastolic diameter (LRS 23.5 and 18.8 respectively) (A and B) and fractional shortening (LRS 22.1 and 15.2 respectively; n=130) (C and D).

 
QTLs linked to mice survival.
We identified two QTLs on Chrs 4 and 18 significantly linked to survival and supported by a LRS of 19.9 and 23.6 respectively (Fig. 5). These QTLs are named Hrtfm3 and Hrtfm4 (Hrtfm for heart failure modifier) and are centered on the microsatellite markers D4Mit236 (support interval, SI, 8–18 cM) and D18Mit28 (SI, 17–38 cM). Hrtfm3 and Hrtfm4 account for ~10 and 12% of the total variance in N2 CSQ survival (18.6 and 22.3% of the genetic variability). Hrtfm3 is linked only to survival and has no statistically significant linkage with cardiac function. By contrast, Hrtfm4 shows significant linkage with FS (LRS, 17.7) and a LRS value suggestive for a linkage with LVEDD (LRS, 9.2). We screened the genome for an interaction between each pair of markers using survival as the phenotypic trait, but no significant interaction was found.

The allele effects of Hrtfm3 and Hrtfm4 on the phenotype are shown in Figure 7. For these QTLs, a homozygous genotype (AKR/AKR) was associated with an improvement of the prognosis compared with a heterozygous genotype (DBA/AKR), explaining the important differences in survival between the three different generations of this study.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 7. Allelic effects of the different QTLs on survival (A), left ventricular end-diastolic diameter (B) and fractional shortening (C). The effect of the interaction between Hrtfm 4 and Hrtfm 6 and between Hrtfm 4 and D19Mit88 is reported in D and E. The allelic effects are shown for homozygous AKR/AKR (AA) and for heterogeneous AKR/DBA (AD) CSQ mice. Values are expressed as mean±SD. *P<10-4 and **P<10-5 determined by t-test for the effect of a single locus. #P<0.05 and ##P<0.001 determined by ANOVA followed by Tukey tests for the interaction between two loci.

 
QTLs linked to the echocardiographic parameters.
We identified two other QTLs on Chrs 2 and 13 significantly linked only to cardiac function (LVEDD, LRS 23.5 and 18.8 respectively; and FS, LRS 22.1 and 15.2 respectively; Fig. 6). They are named Hrtfm5 and Hrtfm6 and are centered on the microsatellite markers D2Mit138 (SI, 67–98 cM) and D13Mit213 (SI, 59–75 cM). Hrtfm5 and Hrtfm6 are estimated to be responsible for 14 and 11% of LVEDD variability (28.2 and 22.1% of the genetic variability, respectively) and 12 and 8% respectively of FS variability (19.0 and 12.7% of the genetic variability, respectively). While Hrtfm5 shows no linkage with survival, the LRS value for Hrtfm6 was suggestive for a linkage (LRS 9.5), indicating that the effect of this QTL on survival is probably too weak to reach the significance level.

Interestingly, we found a statistically significant interaction (P<0.006) between two of the identified QTLs (Hrtfm4 and Hrtfm6) for their effect on FS with a LRS value of 34.6 for the total effect of the two QTLs. In addition, we observed another interaction (P<0.002, LRS for interaction 26.5) between Hrtfm4 and a locus localized near to the microsatellite marker D19Mit88 that had no significant effect on echocardiographic parameters by itself.

The cosegregation of the phenotype with the microsatellite marker located at the center of each QTL and the consequences on the phenotype of the interaction between loci are reported in Figure 7. As with Hrtfm3 and Hrtfm4, the AKR alleles of Hrtfm5 and Hrtfm6 showed a favorable effect on the echocardiographic parameters of heart failure mice.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Using a genome mapping strategy, we identified several novel QTLs that modify the cardiac phenotype in a mouse model of heart failure. Although these loci were identified using the extremes for each phenotype, they are robust since genotyping of all backcrossed animals leads to a further increase in the LRS for the various QTLs. Moreover, this study shows that the relationship between genotype and heart failure phenotype is complex. Indeed, our data demonstrates an interaction between several QTLs, and shows that some of the QTLs have a differential effect on the multiple phenotypic traits of heart failure.

Recent progress in the understanding of the molecular basis of heart failure has demonstrated that a large number of transduction pathways are involved in the development and progression of this disease (18). Given the complexity of heart failure pathophysiology, it is not surprising that we identified several QTLs modifying the expression of the heart failure phenotype in the same genetic cross. These QTLs contain only one of the heart failure modifier genes previously reported in the literature, the ß2 adrenergic receptor (AR) (4). This is probably explained by the absence of polymorphisms within these genes between the two strains in our study.

We have previously reported the effect of genetic background on the heart failure phenotype of CSQ mice in a DBAx C57BL/6 reciprocal backcross (9). In that study, we localized two QTLs affecting survival of CSQ mice on Chrs 2 and 3. Importantly, the interval we previously reported on Chr 2 (Hrtfm1, 38.3–43.5 cM) (9) does not overlap with the SI of the QTL identified in the current study on the same chromosome (Hrtfm5). Our current data extends our previous findings by showing how novel QTLs can be identified when sensitized disease models are backcrossed into disparate genetic backgrounds.

Given the complexity of the pathophysiology in multifactorial diseases, it has been argued that epistatic interactions (interaction between genes) are a universal component of their genetic architecture (12). In this study, we identified an interaction between two pairs of loci (Hrtfm4/Hrtfm6 and Hrtfm4/D19Mit88). Interestingly, the genotype at microsatellite marker D19Mit88 has no effect on fractional shortening by itself, but enhances the effect of Hrtfm4. Such modifier loci with a pure epistatic effect have previously been identified in several diseases (19,20) but have not been reported for heart failure. However, it has been shown that polymorphisms of the ß1AR (Arg389Gly) and {alpha}2AR (alpha2CDel322–325) act synergistically to increase the susceptibility to develop heart failure in patients without cardiovascular disease (21). These results are consistent with the concept that some modifier genes may have no direct effect on the phenotype, but can have important consequences by modifying the effects of other polymorphisms.

One of the important findings in our study is the identification of QTLs that are linked to both survival and cardiac function (Hrtfm4 and Hrtfm6) when other QTLs are linked only to survival (Hrtfm3) or cardiac function (Hrtfm5). The immediate cause of death of heart failure patients can be the consequence of two different mechanisms. In some cases, the death results directly from pump failure. However, recent large scale clinical trials have shown that between 40 and 60% of heart failure patients die suddenly due mainly to ventricular arrhythmias (22,23). Comparable results have been reported in the CSQ heart failure model, where half of the CSQ mice died spontaneously without evidence of pump failure on necropsy (24). While several mutations responsible for rare cases of inherited ventricular arrhythmias have been reported (KVLQT1, minK and SCN5A) (25), none of the modifier genes previously identified in heart failure have been correlated to an increased risk of sudden death.

Our results suggest that Hrtfm3, which correlates only to survival, may affect the phenotype of heart failure by modifying the risk of sudden death. In contrast, the effect of Hrtfm4 and Hrtfm6 on outcome is more likely to be linked to a modification in the progression of left ventricular systolic dysfunction. We hypothesize that the function of the modifier gene(s) localized within the QTLs Hrtfm4 and Hrtfm6 may be quite different from that identified in Hrtfm3. For example, an ionic channel could modify the risk of sudden death, whereas a structural or calcium handling protein could affect both survival and cardiac function. However, since the relationship between Hrtfm3 and the risk of sudden death has only been determined indirectly, we cannot definitively exclude a non cardiac effect of Hrtfm3 on survival.

By contrast, another QTL identified in our study (Hrtfm5) is linked only to echocardiographic parameters without a significant effect on survival. Although the correlation between cardiac function and survival in human heart failure is well established (26), some experimental models, such as knock-out mice of the muscle LIM protein, have shown that the relationship between these two parameters is more complex, and that severe systolic dysfunction may not always be associated with reduced survival (27). Our results demonstrate that linkage studies provide a genetic tool that will enable dissection of the molecular basis underlying the different phenotypic traits of complex diseases such as heart failure.

The identification of modifier genes contained within a QTL remains a challenge that requires the use of several complementary approaches (28–31). The different intervals identified in this study contain numerous genes, with some of them considered good candidates. For example, Hrtfm6 includes the gene for MEK kinase 1 (MEKK1) that is essential for JNK (c-Jun N-terminal kinase) activation in vivo (32) and for the development of cardiac hypertrophy in response to Gq overexpression (33). However, the number of genes in our intervals is still quite large and it would be impracticable to evaluate all genes for polymorphic variations. Strategies to further refine each QTL that include in silico sequence comparison and the generation of several lines of congenic mice, should allow us to isolate the phenotypic effects of each QTL and to reduce the number of candidate genes within the interval.

The survival of heart failure patients has become the standard for the evaluation of interventions in clinical trials. The identification of the gene(s) responsible for modifying survival of mice with severe heart failure would therefore be of major interest and of potential clinical application. Nonetheless, survival may not be the ideal phenotypic trait to perform fine mapping because of the possibility that death in some mice could result from non-cardiac causes. This however, should be overcome with the use of multiple phenotypic traits to map modifier gene(s) that are localized within the QTLs. By contrast, the identification of gene(s) linked only to cardiac function (Hrtfm5) will mainly be useful to provide novel mechanistic insights into the physiopathology of heart failure.

One potential limitation of our study is the experimental model of heart failure we used. It could be hypothesized that the modifier loci identified will be specific to the cardiomyopathy induced by CSQ overexpression. Although expression of CSQ has not been shown to be upregulated in heart failure (34), this model reproduces a number of key features of the disease including modifications of the ß-adrenergic receptor signaling pathway, electrophysiological abnormalities, left ventricle hypertrophy, left ventricular systolic dysfunction and sudden death (17,24,35). Furthermore, a mutation in the CSQ gene has been associated with polymorphic ventricular tachycardia in humans (36). Since QTL mapping requires a reproducible phenotype with a low mouse to mouse variability, we believe that it will be difficult to map modifier genes using other classical models of heart failure such as myocardial infarction by coronary artery ligation (37) or pressure overload (38).

In conclusion, we have identified a number of novel QTLs that modify the severity of the heart failure phenotype. These data will allow a better understanding of the complex relationship between the variability of clinical presentation of heart failure patients and their genetic background.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Animals
Animals were handled according to protocols and animal welfare regulations approved by the Institutional Review Board at Duke University Medical Center. The nomenclature of the crosses is in compliance with the recommendations of the International Committee on Standardized Genetic Nomenclature for Mice. Both DBA and AKR strains were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). These strains were chosen because of the relatively large phenotypic and genotypic differences between them that allowed us to perform linkage analysis. A line of transgenic mice overexpressing CSQ and inbred on a DBA genetic background was generated as previously described (9,17). Briefly, full-length canine cardiac CSQ cDNA was fused to the {alpha}-myosin heavy chain promoter to drive cardiac-targeted expression and micro-injected into fertilized mouse eggs. Animals positive for the transgene were identified at each generation by Polymerase chain reaction (PCR).

Genetic crosses
DBA CSQ males (F0) were mated with wild-type AKR females, leading to a F1 progeny of hybrid mice [AKD2F1/J-Tg(CSQ)]. Fifteen F1 males were selected randomly as breeders and chronically treated with beta-blockers (350 mg/kg of body weight per day) administrated in the drinking water (2 mg/ml) in order to improve their survival and the breeding efficiency. F1 breeders were backcrossed to wild-type AKR females to produce 172 N2 mice [AK(AKD2F1)/J-Tg(CSQ)]. According to Lander and Botstein (39), this number of animals is sufficient to detect three QTLs of equal phenotypic effect with 90% of power when an interval mapping strategy is used to analyze the data.

Western blot
The level of CSQ overexpression was compared between mice with different genetic background or age by western blot. Briefly, left ventricles were homogenized and 10 µg of cytosolic extracts were resolved by 10% SDS–PAGE, transferred to nitrocellulose membranes and probed with antibodies specific for CSQ (1 : 2000; gift from Dr L.R. Jones).

Determination of heart failure phenotype
Transthoracic two-dimensional guided M-mode echocardiography was performed at 12 weeks of age in conscious mice using an HDI 5000 echocardiograph with a 15 MHz probe (ATL, Bothell, Washington, DC, USA) (35). Echocardiograms of good technical quality were obtained in 130 mice and were analyzed off-line using the Access Point 2000 software (Freedland Systems, Kilburn Court, IN, USA). Because all mice presented a pulsus alternans characteristic of CSQ overexpression (40), left ventricular end diastolic diameter, left ventricular end systolic diameter, and left ventricular fractional shortening were measured by averaging six consecutive beats.

All animals were maintained in the same environment with a 10 h light/14 h dark cycle and received unlimited access to the same standard diet. The cages were inspected daily to determine survival. Dead mice were autopsied and animals with evidence for a cause of death other than heart failure (i.e. leukemia, tumors or infection) were excluded from all statistical analysis using survival as the phenotypic trait. The observers gathering the phenotypic data were blinded to the genotype data.

Statistical analysis of the phenotypic traits
All quantitative values were expressed as mean±SD. Differences in phenotypic trait values between mice with a different genetic background were analyzed by analysis of variance followed by Tukey tests. A value of P<0.05 was considered significant. The proportion of the total N2 variance due to variations in genetic background was determined according to the following formula: (total N2 variance-environmental variance)x100/total N2 variance, where the environmental phenotypic variance was estimated by averaging the variance of the two non-segregating generations (F0 and F1) (41).

Genotyping
The genomic DNA was isolated from mouse tails using DNeasy Tissue Kit (Qiagen, Chatsworth, CA, USA). One hundred and six microsatellite markers polymorphic in the DBAxAKR cross and regularly spaced (14.6±3.9 cM) throughout the 19 mouse autosomes were selected using the 2003 mouse genome informatics database from Jackson Laboratory (www.jax.org). Fluorescent labeled primers were purchased from Research Genetics (Carlsbad, CA, USA) and Applied Biosystems (Foster City, CA, USA). PCR was performed following the recommendations of the suppliers, and PCR products were separated by capillary electrophoresis in an ABI PRISM® 3100 Genetic Analyzer. Genotyping was performed using only microsatellite markers that amplified reliably and distinctly the two parental alleles. In order to avoid any misgenotyping, we repeated the PCR for all double recombinants within a short genetic distance.

Linkage analysis
The Map Manager QTX program (version b13) was used to localize the QTLs responsible for the variability of mice heart failure phenotypes (survival, FS, LVEDD). The source codes are available on line at http://mapmgr.roswellpark.org/mmQTX.html. The analysis of the linkage between the phenotype of the N2 CSQ mice and their genotype for the different microsatellite markers was performed in several steps as we previously described (9).

As most of the statistical information required for QTL mapping is located in the extremes of a trait distribution (39,42), we constructed the preliminary genetic linkage maps by genotyping only the N2 progeny with the most extreme phenotype (±0.5 SD from the average). The level of significance of each potential association was estimated using the LRS proposed by Haley and Knott (43). This logarithmic index can be converted to a LOD score by dividing it by 2 ln(10). The significance thresholds for each phenotypic trait were empirically determined by permutation tests performed with all the informative markers and 10 000 permutations of data (44). The results obtained by this method were identical to the thresholds determined by the method proposed by Piepho (45).

On the basis of the preliminary genetic linkage maps, high-density mapping was performed using the entire N2 population within the area suggesting linkage with the phenotype. The composite interval mapping function of Map Manager QTX was used to more accurately determine the position of the QTLs and to estimate the statistical level of significance at multiple analysis point across each interval between two markers. When several QTLs where identified for the same phenotypic trait, one of them was used as genetic background in order to evaluate the effect of each QTL independently (46). The support interval of a QTL was defined as the region where the highest LOD score decrease by a value of 1 on either side of the peak (1-LOD).

In the last statistical analysis, we examined the interaction or epistatic effect between all pairs of marker for each phenotypic trait. Two loci show interaction when the genotype at one locus affects the effect of the other locus (47). The significance of the interaction between the loci was assessed by a permutation tests for the total effect (leading to a threshold comparable to the stringent recommended P-value<10-5) and by a P-value under 0.01 for the interaction itself.

The genes localized within the support interval of the different QTLs have been identified using the available databases (database of the University of California, Santa Cruz; www.genome.ucsc.edu and Celera Discovery system; www.celeradiscoverysystem.com/index.cfm).


   ACKNOWLEDGEMENTS
 
We thank Dr Lan Mao for her expert technique of echocardiography, Kristine Porter for her technical assistance and Nancy Golson, Tracey Howell and Christopher Clayton for their dedicated work on animal handling. We also thank Drs Tom Coffman and Thu Le for active discussion and interpretation of data. This work was supported by National Institutes of Health grant HL69230 (H.A.R.) and by a grant from the French Federation of Cardiology (P.L.). H.A.R. is a recipient of a Burroughs Wellcome Fund Clinical Scientist Award in Translational Research.


   FOOTNOTES
 
* To whom correspondence should be addressed at: Department of Medicine, Molecular Genetics and Microbiology, Duke University Medical Center, CARL Building Room 226, DUMC 3104, Durham, NC 27710, USA. Tel: +1 9196682520; Fax: +1 9196682524; Email: h.rockman{at}duke.edu Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Hunt, S.A., Baker, D.W., Chin, M.H., Cinquegrani, M.P., Feldmanmd, A.M., Francis, G.S., Ganiats, T.G., Goldstein, S., Gregoratos, G., Jessup, M.L. et al. (2001) ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (committee to revise the 1995 guidelines for the evaluation and management of heart failure): developed in collaboration with the International Society for Heart and Lung Transplantation; endorsed by the Heart Failure Society of America. Circulation, 104, 2996–3007.[Free Full Text]

  2. MacIntyre, K., Capewell, S., Stewart, S., Chalmers, J.W., Boyd, J., Finlayson, A., Redpath, A., Pell, J.P. and McMurray, J.J. (2000) Evidence of improving prognosis in heart failure: trends in case fatality in 66 547 patients hospitalized between 1986 and 1995. Circulation, 102, 1126–1131.[Abstract/Free Full Text]

  3. Borjesson, M., Magnusson, Y., Hjalmarson, A. and Andersson, B. (2000) A novel polymorphism in the gene coding for the beta(1)-adrenergic receptor associated with survival in patients with heart failure. Eur. Heart J., 21, 1853–1858.[Abstract/Free Full Text]

  4. Liggett, S.B., Wagoner, L.E., Craft, L.L., Hornung, R.W., Hoit, B.D., McIntosh, T.C. and Walsh, R.A. (1998) The Ile164 beta2-adrenergic receptor polymorphism adversely affects the outcome of congestive heart failure. J. Clin. Invest., 102, 1534–1539.[Web of Science][Medline]

  5. Andersson, B. and Sylven, C. (1996) The DD genotype of the angiotensin-converting enzyme gene is associated with increased mortality in idiopathic heart failure. J. Am. Coll. Cardiol., 28, 162–167.[Abstract]

  6. McNamara, D.M., Holubkov, R., Janosko, K., Palmer, A., Wang, J.J., MacGowan, G.A., Murali, S., Rosenblum, W.D., London, B. and Feldman, A.M. (2001) Pharmacogenetic interactions between beta-blocker therapy and the angiotensin-converting enzyme deletion polymorphism in patients with congestive heart failure. Circulation, 103, 1644–1648.[Abstract/Free Full Text]

  7. Herrmann, S., Schmidt-Petersen, K., Pfeifer, J., Perrot, A., Bit-Avragim, N., Eichhorn, C., Dietz, R., Kreutz, R., Paul, M. and Osterziel, K.J. (2001) A polymorphism in the endothelin-A receptor gene predicts survival in patients with idiopathic dilated cardiomyopathy. Eur. Heart J., 22, 1948–1953.[Abstract/Free Full Text]

  8. Loh, E., Rebbeck, T.R., Mahoney, P.D., DeNofrio, D., Swain, J.L. and Holmes, E.W. (1999) Common variant in AMPD1 gene predicts improved clinical outcome in patients with heart failure. Circulation, 99, 1422–1425.[Abstract/Free Full Text]

  9. Suzuki, M., Carlson, K.M., Marchuk, D.A. and Rockman, H.A. (2002) Genetic modifier loci affecting survival and cardiac function in murine dilated cardiomyopathy. Circulation, 105, 1824–1829.[Abstract/Free Full Text]

  10. Nadeau, J.H. (2001) Modifier genes in mice and humans. Nat. Rev. Genet., 2, 165–174.[Web of Science][Medline]

  11. Montagutelli, X. (2000) Effect of the genetic background on the phenotype of mouse mutations. J. Am. Soc. Nephrol., 11 (Suppl. 16), S101–S105.[Abstract/Free Full Text]

  12. Culverhouse, R., Suarez, B.K., Lin, J. and Reich, T. (2002) A perspective on epistasis: limits of models displaying no main effect. Am. J. Hum. Genet., 70, 461–471.[CrossRef][Web of Science][Medline]

  13. Dempsey, A.A., Dzau, V.J. and Liew, C.C. (2001) Cardiovascular genomics: estimating the total number of genes expressed in the human cardiovascular system. J. Mol. Cell. Cardiol., 33, 1879–1886.[CrossRef][Web of Science][Medline]

  14. Darvasi, A. and Pisante-Shalom, A. (2002) Complexities in the genetic dissection of quantitative trait loci. Trends Genet., 18, 489–491.[CrossRef][Web of Science][Medline]

  15. Durand, J.B., Bachinski, L.L., Bieling, L.C., Czernuszewicz, G.Z., Abchee, A.B., Yu, Q.T., Tapscott, T., Hill, R., Ifegwu, J., Marian, A.J. et al. (1995) Localization of a gene responsible for familial dilated cardiomyopathy to chromosome 1q32. Circulation, 92, 3387–3389.[Abstract/Free Full Text]

  16. Kantachuvesiri, S., Haley, C.S., Fleming, S., Kurian, K., Whitworth, C.E., Wenham, P., Kotelevtsev, Y. and Mullins, J.J. (1999) Genetic mapping of modifier loci affecting malignant hypertension in TGRmRen2 rats. Kidney Int., 56, 414–420.[CrossRef][Web of Science][Medline]

  17. Jones, L.R., Suzuki, Y.J., Wang, W., Kobayashi, Y.M., Ramesh, V., Franzini-Armstrong, C., Cleemann, L. and Morad, M. (1998) Regulation of Ca2+ signaling in transgenic mouse cardiac myocytes overexpressing calsequestrin. J. Clin. Invest., 101, 1385–1393.[Web of Science][Medline]

  18. Towbin, J.A. and Bowles, N.E. (2002) The failing heart. Nature, 415, 227–233.[CrossRef][Medline]

  19. Upadhya, P., Churchill, G., Birkenmeier, E.H., Barker, J.E. and Frankel, W.N. (1999) Genetic modifiers of polycystic kidney disease in intersubspecific KAT2J mutants. Genomics, 58, 129–137.[CrossRef][Web of Science][Medline]

  20. Sugiyama, F., Churchill, G.A., Li, R., Libby, L.J., Carver, T., Yagami, K., John, S.W. and Paigen, B. (2002) QTL associated with blood pressure, heart rate, and heart weight in CBA/CaJ and BALB/cJ mice. Physiol. Genomics, 10, 5–12.[Abstract/Free Full Text]

  21. Small, K.M., Wagoner, L.E., Levin, A.M., Kardia, S.L. and Liggett, S.B. (2002) Synergistic polymorphisms of beta1- and alpha2C-adrenergic receptors and the risk of congestive heart failure. New Engl. J. Med., 347, 1135–1142.[Abstract/Free Full Text]

  22. Poole-Wilson, P.A., Uretsky, B.F., Thygesen, K., Cleland, J.G., Massie, B.M. and Ryden, L. (2003) Mode of death in heart failure: findings from the ATLAS trial. Heart, 89, 42–48.[Abstract/Free Full Text]

  23. Cleland, J.G., Chattopadhyay, S., Khand, A., Houghton, T. and Kaye, G.C. (2002) Prevalence and incidence of arrhythmias and sudden death in heart failure. Heart Fail. Rev., 7, 229–242.[CrossRef][Medline]

  24. Knollmann, B.C., Knollmann-Ritschel, B.E., Weissman, N.J., Jones, L.R. and Morad, M. (2000) Remodelling of ionic currents in hypertrophied and failing hearts of transgenic mice overexpressing calsequestrin. J. Physiol., 525(Pt 2), 483–498.[Abstract/Free Full Text]

  25. Napolitano, C. and Priori, S.G. (2002) Genetics of ventricular tachycardia. Curr. Opin. Cardiol., 17, 222–228.[CrossRef][Web of Science][Medline]

  26. Pfeffer, M.A., Braunwald, E., Moye, L.A., Basta, L., Brown, E.J., Jr., Cuddy, T.E., Davis, B.R., Geltman, E.M., Goldman, S., Flaker, G.C. et al. (1992) Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE Investigators. New Engl. J. Med., 327, 669–677.[Abstract]

  27. Rockman, H.A., Chien, K.R., Choi, D.J., Iaccarino, G., Hunter, J.J., Ross, J., Jr., Lefkowitz, R.J. and Koch, W.J. (1998) Expression of a beta-adrenergic receptor kinase 1 inhibitor prevents the development of myocardial failure in gene-targeted mice. Proc. Natl Acad. Sci. USA, 95, 7000–7005.[Abstract/Free Full Text]

  28. Flint, J. and Mott, R. (2001) Finding the molecular basis of quantitative traits: successes and pitfalls. Nat. Rev. Genet., 2, 437–445.[CrossRef][Web of Science][Medline]

  29. Glazier, A.M., Nadeau, J.H. and Aitman, T.J. (2002) Finding genes that underlie complex traits. Science, 298, 2345–2349.[Abstract/Free Full Text]

  30. Nebert, D.W., Dalton, T.P., Stuart, G.W. and Carvan, M.J. III (2000) ‘Gene-swap knock-in’ cassette in mice to study allelic differences in human genes. Ann. NY Acad. Sci., 919, 148–170.[Medline]

  31. Lyons, P.A., Armitage, N., Argentina, F., Denny, P., Hill, N.J., Lord, C.J., Wilusz, M.B., Peterson, L.B., Wicker, L.S. and Todd, J.A. (2000) Congenic mapping of the type 1 diabetes locus, Idd3, to a 780-kb region of mouse chromosome 3: identification of a candidate segment of ancestral DNA by haplotype mapping. Genome Res., 10, 446–453.[Abstract/Free Full Text]

  32. Yujiri, T., Sather, S., Fanger, G.R. and Johnson, G.L. (1998) Role of MEKK1 in cell survival and activation of JNK and ERK pathways defined by targeted gene disruption. Science, 282, 1911–1914.[Abstract/Free Full Text]

  33. Minamino, T., Yujiri, T., Terada, N., Taffet, G.E., Michael, L.H., Johnson, G.L. and Schneider, M.D. (2002) MEKK1 is essential for cardiac hypertrophy and dysfunction induced by Gq. Proc. Natl Acad. Sci. USA, 99, 3866–3871.[Abstract/Free Full Text]

  34. Prestle, J., Dieterich, S., Preuss, M., Bieligk, U. and Hasenfuss, G. (1999) Heterogeneous transmural gene expression of calcium-handling proteins and natriuretic peptides in the failing human heart. Cardiovasc. Res., 43, 323–331.[Abstract/Free Full Text]

  35. Cho, M.C., Rapacciuolo, A., Koch, W.J., Kobayashi, Y., Jones, L.R. and Rockman, H.A. (1999) Defective beta-adrenergic receptor signaling precedes the development of dilated cardiomyopathy in transgenic mice with calsequestrin overexpression. J. Biol. Chem., 274, 22251–22256.[Abstract/Free Full Text]

  36. Lahat, H., Pras, E., Olender, T., Avidan, N., Ben-Asher, E., Man, O., Levy-Nissenbaum, E., Khoury, A., Lorber, A., Goldman, B. et al. (2001) A missense mutation in a highly conserved region of CASQ2 is associated with autosomal recessive catecholamine-induced polymorphic ventricular tachycardia in Bedouin families from Israel. Am. J. Hum. Genet., 69, 1378–1384.[CrossRef][Web of Science][Medline]

  37. Shah, A.S., White, D.C., Emani, S., Kypson, A.P., Lilly, R.E., Wilson, K., Glower, D.D., Lefkowitz, R.J. and Koch, W.J. (2001) In vivo ventricular gene delivery of a beta-adrenergic receptor kinase inhibitor to the failing heart reverses cardiac dysfunction. Circulation, 103, 1311–1316.[Abstract/Free Full Text]

  38. Esposito, G., Rapacciuolo, A., Naga Prasad, S.V., Takaoka, H., Thomas, S.A., Koch, W.J. and Rockman, H.A. (2002) Genetic alterations that inhibit in vivo pressure-overload hypertrophy prevent cardiac dysfunction despite increased wall stress. Circulation, 105, 85–92.[Abstract/Free Full Text]

  39. Lander, E.S. and Botstein, D. (1989) Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics, 121, 185–199.[Abstract/Free Full Text]

  40. Schmidt, A.G., Kadambi, V.J., Ball, N., Sato, Y., Walsh, R.A., Kranias, E.G. and Hoit, B.D. (2000) Cardiac-specific overexpression of calsequestrin results in left ventricular hypertrophy, depressed force-frequency relation and pulsus alternans in vivo. J. Mol. Cell. Cardiol., 32, 1735–1744.[CrossRef][Web of Science][Medline]

  41. Sebkhi, A., Zhao, L., Lu, L., Haley, C.S., Nunez, D.J. and Wilkins, M.R. (1999) Genetic determination of cardiac mass in normotensive rats: results from an F344xWKY cross. Hypertension, 33, 949–953.[Abstract/Free Full Text]

  42. Darvasi, A. (1997) The effect of selective genotyping on QTL mapping accuracy. Mamm. Genome, 8, 67–68.[CrossRef][Web of Science][Medline]

  43. Haley, C.S. and Knott, S.A. (1992) A simple regression method for mapping quantitative trait loci in line crosses using flanking markers. Heredity, 69, 315–324.[Web of Science][Medline]

  44. Manly, K.F. and Olson, J.M. (1999) Overview of QTL mapping software and introduction to map manager QT. Mamm. Genome, 10, 327–334.[CrossRef][Web of Science][Medline]

  45. Piepho, H.P. (2001) A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics, 157, 425–432.[Abstract/Free Full Text]

  46. Tsujita, Y., Iwai, N., Tamaki, S., Nakamura, Y., Nishimura, M. and Kinoshita, M. (2000) Genetic mapping of quantitative trait loci influencing left ventricular mass in rats. Am. J. Physiol. Heart Circul. Physiol., 279, H2062–H2067.[Abstract/Free Full Text]

  47. Cordell, H.J. (2002) Epistasis: what it means, what it doesn't mean, and statistical methods to detect it in humans. Hum. Mol. Genet., 11, 2463–2468.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
C. J. Barrick, M. Rojas, R. Schoonhoven, S. S. Smyth, and D. W. Threadgill
Cardiac response to pressure overload in 129S1/SvImJ and C57BL/6J mice: temporal- and background-dependent development of concentric left ventricular hypertrophy
Am J Physiol Heart Circ Physiol, May 1, 2007; 292(5): H2119 - H2130.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
H. Ashrafian and H. Watkins
Reviews of Translational Medicine and Genomics in Cardiovascular Disease: New Disease Taxonomy and Therapeutic Implications: Cardiomyopathies: Therapeutics Based on Molecular Phenotype
J. Am. Coll. Cardiol., March 27, 2007; 49(12): 1251 - 1264.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
M. P. Donahue, D. A. Marchuk, and H. A. Rockman
Redefining Heart Failure: The Utility of Genomics
J. Am. Coll. Cardiol., October 3, 2006; 48(7): 1289 - 1298.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
T. P. Maddatu, S. M. Garvey, D. G. Schroeder, W. Zhang, S.-Y. Kim, A. I. Nicholson, C. J. Davis, and G. A. Cox
Dilated cardiomyopathy in the nmd mouse: transgenic rescue and QTLs that improve cardiac function and survival
Hum. Mol. Genet., November 1, 2005; 14(21): 3179 - 3189.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
M. Farrall
Quantitative genetic variation: a post-modern view
Hum. Mol. Genet., April 1, 2004; 13(90001): R1 - 7.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Pharmacol.Home page
M. P. Printz
The Confluence of Population Genetics with Molecular Pharmacology at the Angiotensin II Receptor: Dawn of a New Era or Just a New Wrinkle?
Mol. Pharmacol., March 1, 2004; 65(3): 488 - 491.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
12/23/3097    most recent
ddg333v1
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 arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Le Corvoisier, P.
Right arrow Articles by Rockman, H. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Le Corvoisier, P.
Right arrow Articles by Rockman, H. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?