Exercise Capacity and Response to Training Quantitative Trait Loci in a NZW X 129S1 Intercross and Combined Cross Analysis of Inbred Mouse Strains

Abstract

Genetic factors determining exercise capacity and the magnitude of the response to exercise training are poorly understood. The aim of this study was to identify quantitative trait loci (QTL) associated with exercise training in mice. Based on marked differences in training responses in inbred NZW (-0.65 ± 1.73 min) and 129S1 (6.18 ± 3.81 min) mice, a reciprocal intercross breeding scheme was used to generate 285 F2 mice. All F2 mice completed an exercise performance test before and after a 4-week treadmill running program, resulting in an increase in exercise capacity of 1.54 ± 3.69 min (range = -10 to +12 min). Genome-wide linkage scans were performed for pre-training, post-training, and change in run time. For pre-training exercise time, suggestive QTL were identified on Chromosomes 5 (57.4 cM, 2.5 LOD) and 6 (47.8 cM, 2.9 LOD). A significant QTL for post-training exercise capacity was identified on Chromosome 5 (43.4 cM, 4.1 LOD) and a suggestive QTL on Chromosomes 1 (55.7 cM, 2.3 LOD) and 8 (66.1 cM, 2.2 LOD). A suggestive QTL for the change in run time was identified on Chromosome 6 (37.8 cM, 2.7 LOD). To identify shared QTL, this data set was combined with data from a previous F2 cross between B6 and FVB strains. In the combined cross analysis, significant novel QTL for pre-training exercise time and change in exercise time were identified on Chromosome 12 (54.0 cM, 3.6 LOD) and Chromosome 6 (28.0 cM, 3.7 LOD), respectively. Collectively, these data suggest that combined cross analysis can be used to identify novel QTL and narrow the confidence interval of QTL for exercise capacity and responses to training. Furthermore, these data support the use of larger and more diverse mapping populations to identify the genetic basis for exercise capacity and responses to training.

Fig 1. Genome-wide scan for pre-training exercise time in (NZW x 129S1) F₂ mice.

Scans were performed for the entire population with “sex” as an additive covariate (A) or interactive covariate (B), and in males (C) and females (D) separately. Horizontal lines represent significant (P = 0.05) and suggestive (P = 0.63) logarithm of odds (LOD) thresholds, respectively. LOD thresholds were determined by permutation testing using 1000 permutations.

Fig 2. Genome-wide scan for post-training exercise time in (NZW x 129S1) F₂ mice.

Scans were performed for the entire population with “sex” as an additive covariate (A) or interactive covariate (B), and in males (C) and females (D) separately. Horizontal lines represent significant (P = 0.05) and suggestive (P = 0.63) logarithm of odds (LOD) thresholds, respectively. LOD thresholds were determined by permutation testing using 1000 permutations.

Fig 3. Genome-wide scan for the change exercise time in response to training in (NZW x 129S1) F₂ mice.

Scans were performed for the entire population with “sex” as an additive covariate (A) or interactive covariate (B), and in males (C) and females (D) separately. Horizontal lines represent significant (P = 0.05) and suggestive (P = 0.63) logarithm of odds (LOD) thresholds, respectively. LOD thresholds were determined by permutation testing using 1000 permutations.

Fig 4. Novel QTL for pre-training exercise time identified on Chromosomes 12 and 14 using combined cross analysis.

LOD score plots for pre-training exercise time on Chromosome 12 (A) and 14 (C) are shown for FVB x B6 F₂ (dashed line), NZW x 129S1 F₂ (dotted line) and combined cross (sold line). Horizontal dashed line represents significant (P = 0.05) LOD threshold for the combined cross. Allele effect plots for QTL for pre-training exercise time on Chromosomes 12 (B) and 14 (D). The y-axis in each graph is the z-score transformed exercise time and the x-axis indicates cross-specific genotypes. Homozygous S1, NZW, FVB, and B6 are denoted “SS”, “NN”, “FF”, and “BB”, respectively. Heterozygous alleles are denoted “NS” and “BF” in individual crosses. For combined cross analysis, genotypes were re-coded as “HH” for high performing strains and “LL” for low performing strains, and “LH” for heterozygotes. The strain pattern for high and low performing strains is indicated below allele effect plot for each combined cross QTL. Allele effects are shown at the peak location for each QTL. The solid black line for each genotype represents the mean. *, P < 0.05 compared with alternate homozygous genotype (e.g., BB vs. FF); †, P < 0.05 compared with other genotypes.

Fig 5. LOD score plots (A) and allele effect plots (B) for QTL for change in exercise time on Chromosome 6 using combined cross analysis.

LOD score plots are shown for FVB x B6 F₂ (dashed line), NZW x 129S1 F₂ (dotted line) and combined cross (sold line). Horizontal dashed line represents significant (P = 0.05) LOD threshold for the combined cross. For allele effect plots, the y-axis is the z-score transformed change in exercise time and the x-axis indicates cross-specific genotypes. Homozygous S1, NZW, FVB, and B6 are denoted “SS”, “NN”, “FF”, and “BB”, respectively. Heterozygous alleles are denoted “NS” and “BF” in individual crosses. For combined cross analysis, genotypes were re-coded as “HH” for high performing strains (B6, 129S1) and “LL” for low performing strains (FVB, NZW), and “LH” for heterozygotes. Allele effects are shown at the peak location for the QTL. The solid black line for each genotype represents the mean. *, P < 0.05 compared with alternate homozygous “NN” genotype; †, P < 0.05 compared with other genotypes.