Strain-dependent differences in responses to exercise training in ...

Am J Physiol Regul Integr Comp Physiol 288: R1006 –R1013, 2005. First published December 23, 2004; doi:10.1152/ajpregu.00476.2004.

Strain-dependent differences in responses to exercise training in inbred and hybrid mice Michael P. Massett and Bradford C. Berk Center for Cardiovascular Research, Aab Institute of Biomedical Sciences, University of Rochester, Rochester, New York Submitted 15 July 2004; accepted in final form 16 December 2004

Massett, Michael P., and Bradford C. Berk. Strain-dependent differences in responses to exercise training in inbred and hybrid mice. Am J Physiol Regul Integr Comp Physiol 288: R1006 –R1013, 2005. First published December 23, 2004; doi:10.1152/ajpregu. 00476.2004.—The aim of this study was to characterize the response to exercise training in several mouse strains and estimate the genetic contribution to phenotypic variation in the responses to exercise training. Male mice from three inbred strains [C57Bl/6J (BL6), FVB/NJ (FVB), and Balb/cJ (Balb/c)] and three hybrid F1 strains [CB6F1/J (CB6 ⫽ female Balb/c ⫻ male BL6), B6F F1 (female BL6 ⫻ male FVB), and FB6 F1 (female FVB ⫻ male BL6)] completed an exercise performance test before and after a 4-wk treadmill running program. Distance was used as the primary estimate of endurance exercise performance. FVB mice showed the greatest response to training, with five- to sevenfold greater increases in distance run compared with BL6 and Balb/c strains. Specifically, BL6, FVB, and Balb/c strains increased distance by 33, 172, and 23%, respectively. A similar pattern of changes across strains was observed for run time (17, 87, and 11%) and work (99, 287, and 57%). As a group, F1 hybrid mice derived from BL6 and FVB strains showed an intermediate response to training (61%). However, further analysis indicated that training responses in FB6 F1 mice (80%) were ⬃2.5fold greater than responses in B6F F1 mice (33%, P ⫽ 0.08). A similar pattern of changes between FB6 and B6F F1 mice was observed for run time (44.5 and 17%) and work (141 and 59%). These data demonstrate that there are large strain-dependent differences in training responses among inbred mouse strains, suggesting that genetic background contributes significantly to adaptation to exercise. Furthermore, the contrasting responses in B6F and FB6 F1 strains show that a maternal component strongly influences strain-dependent differences in training responses. treadmill running; hybrid mouse strains; broad-sense heritability SEVERAL MURINE MODELS of chronic human diseases have been developed using genetically modified mice. With greater frequency, exercise and exercise training are being incorporated in investigations of these murine models of chronic disease (4, 24, 25). However, the extent to which the background strain contributes to disease and/or exercise-related phenotypic differences in transgenic mice remains a question (2, 11, 16). Relatively little is known about strain-dependent differences in the responses to exercise and exercise training (4). As a result, phenotypic differences between wild-type and mutant mice can be identified, but comparisons of either strain to normative data are relatively difficult (11, 16, 28). Thus it is important to characterize the responses to exercise training in mouse strains commonly used to derive genetically modified mice.

Address for reprint requests and other correspondence: M. P. Massett, Center for Cardiovascular Research, Univ. of Rochester, 601 Elmwood Ave., Box 679, Rochester NY 14642 (E-mail: michael_massett@urmc. rochester.edu). R1006

In humans, both intrinsic exercise capacity and the response to exercise training are highly variable, such that some individuals may not respond to exercise training (5, 27). Although many of the phenotypic traits associated with exercise training are well known (i.e., increased oxidative metabolism, improved endothelial function; see Refs. 23–25), the genetic factors determining the magnitude of the response to exercise are poorly understood (5, 26). Furthermore, on the basis of previous reports, the genetic factors determining intrinsic exercise capacity are likely different from those determining the response to training (5, 6, 29). Consequently, identifying the genetic factors modulating the adaptations to exercise may provide insight into the prevention and treatment of many chronic diseases. Recently, inbred mouse strains and selectively bred rats and mice have been used to identify the genetic basis for variation in intrinsic endurance exercise performance (3, 12, 19, 21, 22, 32, 34). In rats and mice, there was a two- to fourfold difference in intrinsic exercise performance measured by treadmill running between the highest- and lowestperforming strains (3, 19, 21, 22). Lerman et al. (21) also examined voluntary running performance in inbred mice. Interestingly, they reported that BL6 mice had the highest duration, distance, and running speed during voluntary running but the poorest performance in treadmill running (21). To date, however, there are limited data from animal models regarding the genetic variation in the responses to exercise training (35). Therefore, the aim of this study was to characterize the response to exercise training in three inbred mouse strains [C57Bl/6J (BL6), FVB/NJ (FVB), and Balb/cJ (Balb/c)] and three F1 hybrid strains (B6F, FB6, and CB6) derived from these inbred strains and estimate the genetic contribution to phenotypic variation in the responses to exercise training. In the present study, we demonstrate that training responses in FVB mice were significantly greater than all other strains tested, whereas BL6 and Balb/c strains had relatively small responses to training. Intrinsic exercise capacity or physiometric variables such as body mass, heart mass, or muscle mass were not significantly predictive of the responses to training. However, broadsense heritability estimates indicate that genotype significantly influences the responses to training. Because training responses in hybrid strains derived from BL6 and FVB strains tended to resemble responses in the maternal strain, our data imply that the female parent, either through genetic or maternal effects, can strongly influence the response to training. The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

0363-6119/05 $8.00 Copyright © 2005 the American Physiological Society

http://www.ajpregu.org

EXERCISE TRAINING RESPONSES IN INBRED AND HYBRID MICE METHODS

Animals. The University Committee on Animal Resources at the University of Rochester approved all procedures. Eight-week old male mice from the following strains were randomly assigned to the sedentary (SED) or exercise training (EX) group. Standard inbred strains C57Bl/6J (BL6, n ⫽ 11 EX and 12 SED) and FVB/NJ (FVB, n ⫽ 16 EX and 14 SED) were obtained from an in-residence breeding colony derived from breeders from Jackson Laboratories, and Balb/cJ (Balb/c, n ⫽ 12 EX and 11 SED) and the CB6F1/J (CB6, female Balb/c ⫻ male BL6, n ⫽ 12 EX and 12 SED) hybrids were purchased from Jackson Laboratories and acclimatized for 2 wk before inclusion in the study to minimize any effect of shipping (8). Reciprocal hybrids from a BL6 ⫻ FVB outcross were generated in our vivarium. Hybrid mice generated from mating BL6 females with FVB males were designated B6F F1 (n ⫽ 8 EX and 13 SED), and hybrid mice from the reciprocal cross were designated FB6 F1 (n ⫽ 12 EX and 7 SED). The inbred strains were chosen based on their genetic (1) and phenotypic (exercise capacity; see Refs. 21 and 22) diversity and their frequent use for genetic analysis and development of transgenic animals. All mice were allowed food and water ad libitum and were maintained on a 12:12-h light-dark schedule. Exercise performance test. All mice were familiarized to run on a motorized rodent treadmill with an electric grid at the rear of the treadmill (Columbus Instruments, Columbus, OH) 2 days before completing an exercise performance test. Familiarization runs were 10 min in duration with a treadmill incline of 10°. Treadmill speed on the first day was 10 m/min and 12 m/min on the 2nd day. For the performance test, mice were placed on the treadmill and allowed to adapt to the surroundings for 3–5 min before starting. The treadmill was started at a speed of 8.5 m/min with a 0° incline. After 9 min, the speed and incline were raised to 10 m/min and 5°, respectively, for the second stage of the test. Speed was then increased by 2.5 m/min every 3 min to a maximum of 40 m/min, and the incline progressively increased 5° every 9 min to a maximum of 15° (10, 22). Exercise continued until exhaustion, defined as an inability to maintain running speed despite repeated contact with the electric grid (10). Each mouse was immediately removed from the treadmill when exhaustion was determined and returned to its home cage. This protocol was repeated after completion of the training period to assess the efficacy of the training regime. Two BL6 mice refused to run during the initial performance test and were not included in the study. SED mice were reintroduced to running on the treadmill 2 days before the second exercise performance test. Endurance exercise performance was estimated from each performance test using the following three parameters: the duration of the run (in min), the distance run (in meters, calculated from the run time and speed of the treadmill), and vertical work performed (in kg/m). Vertical work was calculated as the product of body weight (kg) and vertical distance (meters), where vertical distance ⫽ (distance run) (sin␪), where ␪ is equal to the angle of the treadmill from 0° to 15° (3). Distance run was used as the primary estimate of endurance exercise performance. However, the pattern of within (SED vs. EX)- and across-strain differences was similar for all estimates of endurance exercise performance. Exercise training. EX mice participated in a 4-wk training program consisting of treadmill running 5 days/wk, 60 min/day at a final intensity equivalent to ⬃60% of the maximal work load (speed and incline) attained during the exercise performance test. A relative (%maximum) workload was chosen to evoke the maximum response to training for each strain. The intensity (60% of the maximal work load) was chosen because this was the highest intensity that could be completed by all strains over the 4-wk period. Furthermore, this exercise intensity should be sufficient to elicit cardiovascular and skeletal muscle adaptations (9, 13, 15). Treadmill speed and exercise duration were progressively increased over the first 2 wk of training until the target workload and duration could be maintained. The final AJP-Regul Integr Comp Physiol • VOL

R1007

exercise intensity (speed and incline) for each strain was as follows: Balb/c, 15 m/min, 5° incline; BL6, 15 m/min, 5° incline; FVB, 19 m/min, 10° incline; CB6, 17.5 m/min, 5° incline; B6F and FB6 F1, 19 m/min, 10° incline. Each exercise bout was preceded by a 12- to 18-min warm-up period of walking on the treadmill. Body and tissue mass. Body mass in grams was measured before, after, and weekly during the training period. After the final exercise performance test (24 –36 h), all mice were anesthetized by intraperitoneal injection of a ketamine (80 mg/kg)-xylazine (5 mg/kg) cocktail. Plantaris and gastrocnemius muscles and hearts were harvested from all mice, washed in ice-cold (4°C) saline, frozen in liquid nitrogen, and stored at ⫺80°C until further analysis. Tissue wet weights (in mg) were obtained before freezing. Western blot analysis. Frozen tissue samples were weighed, pulverized in liquid nitrogen, and homogenized in 5–20 ␮l/mg tissue of buffer containing 50 mM Tris (pH 8.0), 2 mM EGTA, 2 mM sodium orthovanadate, 0.01% Triton X-100, and protease inhibitor cocktail (104 ␮M AEBSF, 0.08 ␮M aprotinin, 2 ␮M leupeptin, 4 ␮M bestatin, 1.5 ␮M pepstatin A, and 1.4 ␮M E-64; Sigma). The suspension was sonicated and then centrifuged (15,300 g) at 4°C for 15 min. The Bradford assay was used to determine protein content of the supernatant. Protein samples (25–50 ␮g total protein) were diluted 1:1 (vol/vol) in Laemmli buffer (120 mM Tris, 4% SDS, 100 mM dithiothreitol, and 20% glycerol, pH 6.8), boiled for 10 min (100°C), separated by SDS-PAGE (7.5–15%), and transferred to a nitrocellulose membrane. Membranes were blocked for 1 h with 5% nonfat dried milk in PBS plus 0.1% Tween 20 at room temperature and then incubated with primary antibody against mouse monoclonal antieNOS (1:500; Transduction Laboratories) or mouse monoclonal anticytochrome c (1:1,000; BD PharMingen). The membranes were washed and incubated with the appropriate secondary antibody (sheep anti-mouse IgG horseradish peroxidase 1:1,000 or donkey anti-rabbit IgG horseradish peroxidase 1:1,000; Amersham, Arlington Heights, IL) for 1 h at room temperature and visualized using enhanced chemiluminescence (ECL; Amersham). ECL band density was quantified using NIH Image software. Broad-sense heritability. Two estimates of heritability in the broad sense were calculated to provide an estimate of the contribution of genotype to the responses to training (14, 21, 22). The intraclass correlation (rI) is an estimate of the proportion of the total variation that is accounted for by differences between strains and the coefficient of genetic determination (g2), which accounts for the doubling of the additive genetic variance that occurs with inbreeding. The following equations were used to calculate rI and g2: rI ⫽ (MSB ⫺ MSW)/ [MSB ⫹ (n ⫺ 1)MSW] and g2 ⫽ (MSB ⫺ MSW)/[MSB ⫹ (2n ⫺ 1)MSW], where MSB and MSW are the between- and within-mean square, respectively, and n is the number of animals per strain. Because the number of animals per strain differed, n was calculated as n ⫽ (1/a ⫺ 1)(n ⫺ ¥n2i ), where a is the number of strains and ni is the number of animals in the ith strain. Data analyses. Values are expressed as means ⫾ SE. Betweengroup (EX vs. SED) comparisons for each variable (i.e., body mass, exercise test duration) at each time point were made using unpaired Student’s t-tests. Pre- vs. posttraining comparisons for each variable (i.e., body mass, exercise test duration) were made within each strain separately using paired Student’s t-tests. ANOVA followed by a Fisher’s post hoc analysis was used to analyze comparisons across strains for each variable at one time point (i.e., pretraining or posttraining; StatView for MacIntosh, version 5.0.1; see Ref. 36). Statistical significance was set at P ⱕ 0.05. Linear regression analysis was used to determine the relationship between measured variables (i.e., body mass, heart mass) and endurance exercise performance. The contribution of intrinsic (baseline) endurance exercise capacity to the response to training was assessed using linear regression. For the relationship between pre- and postmeasures of exercise capacity, the effect of pretraining (intrinsic) exercise capacity was considered significant if the slope of the regression equation was significantly 288 • APRIL 2005 •

www.ajpregu.org

R1008

EXERCISE TRAINING RESPONSES IN INBRED AND HYBRID MICE

different from 1.0 (35). The slope was considered significantly different from 1.0 if the 95% confidence interval (CI) did not include 1.0. A slope and CI ⬎ 1.0 indicate that mice with low intrinsic capacity improved the most with training, whereas a slope and CI ⬍ 1.0 indicate that mice with a high intrinsic capacity had the greatest response to training. A slope equal to 1.0 indicates that intrinsic capacity does not determine the response to training (35). RESULTS

Endurance exercise performance was compared between EX and SED groups within each strain and across strains. For each section, EX vs. SED comparisons will be described first, followed by interstrain comparisons. Endurance exercise performance. There were no significant differences between EX and SED groups for endurance exercise performance before starting the training program (Tables 1 and 2). In BL6 and FVB inbred strains, posttraining endurance exercise performance, measured as distance run, was significantly higher in EX mice compared with SED mice (Fig. 1). For FVB mice, distance was 2.6-fold higher in EX vs. SED after training (Fig. 1). Run time and work were also significantly higher in EX vs. SED FVB mice (Tables 1 and 2). Run time, but not work, was significantly higher in EX vs. SED BL6 mice (Tables 1 and 2). There were no significant differences between SED and EX groups of Balb/c mice for any exercise parameter measured. In hybrid F1 strains, posttraining endurance exercise performance in EX mice from FB6 F1 and CB6 strains was significantly higher than SED mice for distance (Fig. 1), as well as run time and work (Tables 1 and 2). Conversely, there were no differences between EX and SED mice from the B6F F1 hybrid strain (Fig. 1). Significant strain-dependent differences in endurance exercise performance were evident before and after training (Tables 1 and 2). Among inbred strains, endurance exercise performance was highest in FVB mice and comparable between BL6 and Balb/c mice (Table 1). After training, FVB mice had the highest endurance exercise performance. Pre- and posttraining endurance exercise performance was similar in BL6 and Balb/c mice and significantly lower than the other strains. The magnitude of change in endurance exercise performance after 4 wk of training also varied markedly across strains (Fig. 2). FVB mice showed the greatest improvement in endurance exercise performance (172%) as a result of training, which was five- to sevenfold higher than the other inbred strains for distance (P ⬍

0.05, Fig. 2) and two- to threefold higher than the F1 hybrid strains for distance (P ⬍ 0.05, Fig. 2). Conversely, BL6 and Balb/c strains made relatively small improvements in endurance exercise performance after 4 wk of training (⬃23–33% increase for distance; Fig. 2). The hybrid F1 strains were included in this study to provide some insight into the genetic component of the response to training. The CB6 strain showed significantly greater endurance exercise performance than either parental strain (BL6 and Balb/c) before and after training (Table 1). The response to training (%change) in CB6 was approximately two- to threefold greater for distance than Balb/c and BL6 strains (Fig. 2). Because the responses to training in the BL6 and Balb/c strains were comparable and the responses in the CB6 strain were greater than both parental strains, the reciprocal hybrid to the CB6 was not studied. Conversely, BL6 and FVB parental strains varied markedly in their response to training, and their hybrid offspring showed contrasting responses as well. Pretraining endurance exercise performance was similar between B6F F1 and FB6 F1 strains and significantly higher than parental strains (BL6 and FVB; Table 1). Posttraining values for endurance exercise performance for B6F F1 and FB6 F1 mice were intermediate compared with parental strains; values in F1 mice were significantly higher than BL6 but significantly lower than FVB (Table 1). However, compared with each other, FB6 F1 had significantly higher posttraining values than B6F F1. Thus posttraining endurance exercise capacity in the BL6 ⫻ FVB-derived F1 hybrid strains tended to be similar to their respective maternal strain. When training responses in EX groups are expressed as a percentage change, training responses in B6F and FB6 strains were not different from BL6 mice and significantly less than FVB mice. However, the magnitude of the training responses (i.e., %changes in distance) in FB6 F1 were ⬃2.5-fold greater than those in B6F F1 (Fig. 2, P ⬍ 0.05), suggesting a potential role for a maternal effect. Determinants of the training response. Because of the large strain-dependent differences in intrinsic (baseline) endurance exercise performance (Table 1), linear regression analysis was used to determine whether intrinsic endurance exercise performance related to training responses (Fig. 3). There was a small but significant correlation between pretraining distance and posttraining distance (R2 ⫽ 0.17, P ⬍ 0.0001). However, the slope of the regression line was not significantly different from

Table 1. Comparison of pre- and posttraining run time, distance run, and work in EX inbred and F1 hybrid mouse strains Balb/c (12)

Run time, min Pretraining Posttraining Distance run, meters Pretraining Posttraining Work, kg/m Pretraining Posttraining

BL6 (11)

FVB (16)

B6F F1 (8)

FB6 F1 (12)

CB6 (12)

36.9⫾0.7a*,b*,c 69.1⫾2.9a*,b*,c

42.1⫾1.7a*,b*,c* 48.0⫾5.3a*,b*,c*,d*

41.6⫾1.7a*,b*c* 59.6⫾3.7a*,b*,c*,d

38.8⫾1.0a*,b* 50.4⫾2.5a*,b*c*,d* 691⫾33a*,b* 1,062⫾82a*,b*,c*,d*

33.2⫾0.7a 36.6⫾1.7a

29.2⫾0.8b 33.7⫾1.5b

514⫾19a 622⫾55a

411⫾17a*,b 553⫾45b

626⫾23a*,b*,c 1,629⫾99a*,b*,c

794⫾55a*,b*,c* 992⫾175a*,b*,c*,d*

779⫾55a*,b*,c* 1,362⫾117a*,b*,c*,d

2.05⫾0.12a 3.05⫾0.40a

1.28⫾0.10a*,b 2.28⫾0.33b

2.80⫾0.14a*,b*,c 10.57⫾0.56a*,b*,c

4.23⫾0.37a*,b*,c*,d* 6.21⫾1.32a*,b*,c*,d*

4.11⫾0.36a*,b*,c*,d 9.41⫾1.00a*,b*,c*,d

3.45⫾0.25a*,b*,c*,d* 6.60⫾0.68a*,b*,c*,d*

Values are means ⫾ SE; no. in parentheses is no. of animals/strain. B6F F1, female BL6 ⫻ male FVB; FB6 F1, female FVB ⫻ male BL6; CB6, offspring from female Balb/c ⫻ male BL6. Statistically significant differences (P ⬍ 0.05) between strains (across columns) are indicated by letters (e.g., a vs. a*, b vs. b*, etc.). Posttraining values are significantly different from pretraining values (P ⬍ 0.05) for run time, distance, and work for exercise (EX) group for all strains except B6F F1 and Balb/c. AJP-Regul Integr Comp Physiol • VOL

288 • APRIL 2005 •

www.ajpregu.org

R1009


Table 2. Comparison of pre- and posttraining run time, distance run, and work in SED inbred and F1 hybrid mouse strains

Run time, min Pretraining Posttraining Distance run, meters Pretraining Posttraining Work, kg/m Pretraining Posttraining

Balb/c (11)

BL6 (12)

FVB (14)

B6F F1 (13)

FB6 F1 (7)

CB6 (12)

32.0⫾0.9a 34.3⫾1.3a

29.3⫾1.0b 29.7⫾0.8a*,b

35.9⫾0.9a*,b*,c 36.4⫾0.7b*,c

41.2⫾1.3a*,b*,c* 39.2⫾1.6a*,b*

40.3⫾2.1a*,b*,c* 40.8⫾1.0a*,b*,c*

39.0⫾1.2a*,b* 40.7⫾2.1a*,b*,c*

479⫾24a 550⫾39a

409⫾26b 416⫾23a*,b

559⫾21b*,c 645⫾22b*

765⫾43a*,b*,c* 704⫾53a*,b*

737⫾69a*,b*,c* 753⫾34a*,b*

698⫾41a*,b* 749⫾68a*,b*

1.83⫾0.14a 2.51⫾0.27a

1.30⫾0.13b 1.54⫾0.17a*,b

2.64⫾0.19a*,b*,c 3.00⫾0.17b*,c

3.72⫾0.29a*,b*,c* 3.61⫾0.29a*,b*,d

3.90⫾0.57a*,b*,c* 4.35⫾0.21a*,b*,c*

3.57⫾0.31a*,b*,c* 4.51⫾0.52a*,b*,c*,d*

Values are means ⫾ SE; no. in parentheses is no. of animals/strain. Statistically significant differences (P ⬍ 0.05) between strains (across columns) are indicated by letters (e.g., a vs. a*, b vs. b*, etc.). There were no significant differences between pre- and posttraining values for sedentary (SED) mice.

1.0 (95% CI ⫽ 0.57–1.91; Fig. 3), indicating that pretraining distance was not significantly related to the response to training. When the strains were analyzed individually, pretraining distance was not significantly correlated with posttraining distance for any strain [R2 range ⫽ 0.00005– 0.21, not significant (NS)], and the slopes were not significantly different from 1.0 (range, slope ⫽ ⫺1.3 to 1.9, 95% CI ⫽ ⫺4.2 to 4.1). Collectively, these data suggest that intrinsic endurance exercise performance is not predictive of the strain-dependent differences in the response to exercise training observed in the inbred and hybrid mouse strains included in this study. Body mass, heart mass, gastrocnemius and plantaris muscle mass, and tissue-to-body mass ratios were determined as markers for the efficacy of the training program (Table 3). Over the course of the 4-wk exercise program, body mass increased significantly in all mice (Table 3). Only EX mice from CB6 and B6F F1 strains had body mass values that were significantly different from SED mice posttraining. CB6 EX gained significantly less mass than SED controls (2.4 ⫾ 0.2 vs. 4.1 ⫾ 0.4 g, P ⬍ 0.05), whereas in B6F F1 mice the pretraining difference in mass between EX and SED groups was maintained over the 4-wk period. In BL6 mice, no significant differences were observed between SED and EX groups for heart or muscle mass or tissue-to-body mass ratios. In Balb/c, FVB, and B6F F1 mice, heart mass was significantly higher in EX vs. SED mice; however, only EX mice from the FVB strain had a significantly

Fig. 1. Endurance exercise performance expressed as distance in sedentary control (SED) and exercise-trained (EX) mice after 4 wk of treadmill exercise. Values are means ⫾ SE; n ⫽ 11 (EX) and 12 (SED) for BL6, n ⫽ 16 and 14 for FVB, n ⫽ 8 and 13 for B6F F1, n ⫽ 12 and 7 for FB6 F1, n ⫽ 12 and 11 for Balb/c, and n ⫽ 12 and 12 for CB6. *P ⬍ 0.05 compared with SED. Distance was significantly greater in EX vs. SED for BL6, FVB, FB6, and CB6 strains. There were no significant differences between EX and SED groups for Balb/c and B6F F1 mice. AJP-Regul Integr Comp Physiol • VOL

higher heart mass-to-body mass ratio. Thus cardiac hypertrophy was present only in FVB EX mice (Table 3). Gastrocnemius mass and gastrocnemius mass-to-body mass ratio tended to be higher in EX mice, but only gastrocnemius mass in Balb/c and muscle mass-to-body mass ratio in CB6 mice was significantly higher in EX compared with SED. Conversely, plantaris mass and plantaris mass-to-body mass ratios were similar between EX and SED mice. Only plantaris mass from B6F F1 EX mice was significantly higher than SED mice for that strain (Table 3). When comparisons of body mass were made across all strains within a specific group (EX or SED), there were several strain-dependent differences but very few differences in heart mass, muscle mass, and tissue-to-body mass ratios (Table 3). Of note, FVB mice had the smallest values for plantaris mass and plantaris mass-to-body mass ratio across strains within EX and SED groups. In most cases, however, these differences were small and not related to strain-dependent differences in endurance exercise performance. Using linear regression analysis, the relationship between physiometric parameters and measures of endurance exercise performance (i.e., run time, distance, work) or the training response (post ⫺ pre or %change) was determined. The relationship between pretrain-

Fig. 2. Comparison of %changes in distance among inbred and hybrid mouse strains before and after 4 wk of exercise training. Changes in distance were calculated from pre- and posttraining values measured during an exercise performance test. Training intensity is described in the text. The 3 inbred mouse strains are as follows: C57Bl/6J (BL6, n ⫽ 11), FVB/NJ (FVB, n ⫽ 16), and Balb/cJ (Balb, n ⫽ 12). CB6F1/J (CB6, n ⫽ 12) hybrid mice are derived from a cross of Balb/c and BL6 strains and B6F F1 (n ⫽ 8), and FB6 F1 (n ⫽ 12) mice are derived from a reciprocal cross of BL6 and FVB strains. Values are means ⫾ SE. Statistically significant differences (P ⬍ 0.05) between strains are indicated by letters (e.g., a vs. a*, b vs. b*, etc.). 288 • APRIL 2005 •

www.ajpregu.org

R1010


Fig. 3. Relationship between pretraining distance and posttraining distance after 4 wk of exercise training in inbred and hybrid mouse strains. Distance was measured during an exercise performance test before and after training. EX mice from all strains were included in the linear regression analysis. ■, BL6; F, FVB; Œ, Balb/c; 䊐, B6F F1; E, FB6 F1; ‚, CB6. Dashed line equals line of identity. The slope of the regression equation was not significantly different from 1.0; therefore, the effect of pretraining (intrinsic) exercise capacity on the response to training was not significant.

ing body mass and pretraining distance was highly significant (R2 ⫽ 0.22, P ⬍ 0.001), but pretraining body mass was less related to posttraining distance (R2 ⫽ 0.04, P ⫽ 0.008) and not related to the change in distance (R2 ⫽ 0.002, NS). Heart mass was significantly correlated with all three measures of posttraining endurance exercise performance (range R2: 0.14 – 0.20, P ⱕ 0.002), whereas heart mass-to-body mass ratios were

significantly related to changes (post ⫺ pre) in endurance exercise performance (range R2: 0.07– 0.12, P ⱕ 0.03). Gastrocnemius mass was significantly correlated with posttraining work (R2 ⫽ 0.09, P ⱕ 0.006), whereas gastrocnemius massto-body mass ratio was not significantly correlated with posttraining measures or changes in endurance exercise performance. Neither plantaris mass nor plantaris mass-to-body mass ratio was significantly correlated with posttraining measures or changes in endurance exercise performance (range R2: 0.01– 0.04, not significant). For additional markers of the training response, we chose eNOS and cytochrome c protein expression in skeletal muscle on the basis of previous studies. Plantaris muscle eNOS expression was ⬃23% higher in EX vs. SED (P ⫽ 0.03) for FVB (Fig. 4). In all other strains of mice, plantaris eNOS expression was similar between EX and SED groups. eNOS expression was not significantly different between EX and SED groups for any strain in the gastrocnemius muscle (data not shown). Cytochrome c expression in gastrocnemius and plantaris muscles was measured as a general marker of oxidative metabolism/mitochondrial function (Fig. 5). In the plantaris muscle from FVB mice, cytochrome c expression was nearly double in EX (196 ⫾ 22% of SED, P ⬍ 0.001) compared with SED mice (Fig. 5). Cytochrome c expression in plantaris muscle was also significantly higher in muscle from CB6 EX mice (117 ⫾ 3% of SED, P ⬍ 0.05). In BL6 mice, cytochrome c expression was somewhat higher in EX vs. SED (133 ⫾ 21% of SED, P ⫽ 0.09). There were no differences between SED and EX groups in the other strains. Cytochrome c expression was also not significantly different between EX and SED groups for any strain in the gastrocnemius muscle (data not shown).

Table 3. Physical characteristics of SED and EX mice from inbred and hybrid mouse strains Balb/c (7/9)

Body mass, g Pretraining SED EX Posttraining SED EX Heart mass, mg SED EX HM/BM, mg/g SED EX Gastrochemius mass, mg SED EX GM/BM, mg/g SED EX Plantaris mass, mg SED EX PM/BM, mg/g SED EX

BL6 (6/6)

FVB (9/11)

B6F F1 (10/5)

FB6 F1 (5/10)

CB6 (9/9)

23.7⫾0.3a 24.5⫾0.4a

22.5⫾0.3b 22.6⫾0.6b

25.0⫾0.4b*,c 24.9⫾0.5b*,c

25.0⫾0.6a*,d 27.1⫾0.3†a*,b*

26.9⫾0.7a*,b* 27.1⫾0.5a*,b*,c*

27.2⫾0.5a*,b*,c*,d* 26.5⫾0.6b*

26.8⫾0.4a 27.1⫾0.5a

25.7⫾0.4b 24.9⫾0.6b

27.5⫾0.6c* 27.6⫾0.6b*,c

27.6⫾0.7c* 29.8⫾0.4†a*,b*

29.8⫾0.8a*,b* 30.6⫾0.5a*,b*,c*

31.3⫾0.6a*,b*,c 28.7⫾0.6†b*

127⫾2a* 134⫾2†

125⫾5a* 126⫾6

130⫾2a* 139⫾2†

125⫾3a* 141⫾3†

132⫾5 137⫾3

145⫾3a 140⫾5

4.7⫾0.1 4.9⫾0.1

5.0⫾0.1 5.2⫾0.4

4.7⫾0.1 5.1⫾0.1†

4.7⫾0.1 4.7⫾0.1

4.6⫾0.1 4.6⫾0.1

4.6⫾0.1 4.8⫾0.1

136⫾2a* 144⫾3†

145⫾2 144⫾5a

141⫾5a* 147⫾3

141⫾5a* 157⫾6

153⫾6 160⫾3a*

165⫾5a 153⫾5

5.0⫾0.1a* 5.1⫾0.1a*

5.8⫾0.2a 5.9⫾0.3a

5.1⫾0.1a* 5.4⫾0.1

5.3⫾0.1a* 5.3⫾0.1

5.2⫾0.1a* 5.3⫾0.1

5.2⫾0.1a* 5.5⫾0.1†

21.1⫾0.7a* 22.0⫾0.6†a*

22.4⫾1.0a*,b* 22.4⫾1.6a*

17.4⫾1.0a 19.0⫾1.0a

20.5⫾0.9a*,b* 23.2⫾0.6†a*

26.8⫾1.0a*,b 24.3⫾0.9a*

22.7⫾0.6a*,b* 21.5⫾0.8

0.79⫾0.04a*,b* 0.79⫾0.02

0.87⫾0.04a*,c 0.90⫾0.09a*,b*

0.62⫾0.03a 0.68⫾0.03a

0.75⫾0.03a*,b*,c* 0.77⫾0.02

0.90⫾0.03a*,b 0.80⫾0.04a*

0.73⫾0.03a*,b*,c* 0.76⫾0.04b

Values are means ⫾ SE; no. in parentheses is no. of animals group (SED/EX). HM/BM, heart mass-to-body mass ratio; GM/BM, gastrocnemius mass-to-body mass ratio; PM/BM, plantaris mass-to-body mass ratio. †P ⬍ 0.05 compared with SED from same strain. Statistically significant differences (P ⬍ 0.05) between strains (across columns) are indicated by letters (e.g., a vs. a*, b vs. b*, etc.). Posttraining body mass values were significantly different from pretraining body mass values for SED and EX groups of all strains. AJP-Regul Integr Comp Physiol • VOL

288 • APRIL 2005 •

www.ajpregu.org


Fig. 4. eNOS protein expression in mouse plantaris muscle from SED and EX mice after 4 wk of treadmill exercise. Top: representative Western blot of eNOS protein expression in mouse plantaris muscle from SED and EX groups of FVB mice; bottom: quantitative densitometry of eNOS protein expression in plantaris muscle from 3 inbred (BL6, FVB, and Balb/c) and 3 hybrid F1 (B6F, FB6, and CB6) strains. eNOS protein expression was significantly enhanced in plantaris muscle from EX vs. SED FVB mice. Values are means ⫾ SE; n ⫽ 5– 6 animals/group for each strain. HUVEC, cell lysate from human umbilical vein endothelial cells used as positive control. *P ⬍ 0.05 compared with SED.

Broad-sense heritability. Two estimates of heritability in the broad sense were calculated for each measure of endurance exercise performance to estimate the degree to which endurance exercise performance is determined by genotype (14). The rI for posttraining distance was 0.65, and g2 was 0.48, suggesting that genotypic variance contributes significantly to phenotypic variance. The rI and g2 were somewhat lower for the response to training (change in distance): rI ⫽ 0.56 and g2 ⫽ 0.39. Estimates for posttraining and changes in both run time (posttraining: rI ⫽ 0.64, g2 ⫽ 0.47; change in run time: rI ⫽ 0.55 and g2 ⫽ 0.38) and work (posttraining: rI ⫽ 0.65, g2 ⫽ 0.48; change in work: rI ⫽ 0.55 and g2 ⫽ 0.38) were comparable to those for distance.

R1011

from 47 to 73% (21, 22), whereas narrow-sense heritability estimates ranged from 39 to 64% (3, 12, 19). In the current study, rI and g2 for endurance exercise performance in the sedentary state were comparable to published reports, ranging from 38 to 67%. However, less is known about the heritability of training responses in rodents. In humans, responses to ˙ O2 max) vary widely in individuals comtraining (change in V pleting similar training programs (5, 27). The HERITAGE Family Study reported a heritability of 47% for training responses in 98 two-generation families (5). In the current study, training responses varied considerably across strains, with Balb/c and BL6 mice increasing distance by 23 and 33%, respectively, whereas FVB improved distance run by 172%. Although the improvements in endurance exercise performance in Balb/c and BL6 strains were relatively small, the magnitude of changes based on run time are comparable to that obtained in mice (9, 18), rats (24, 25, 35, 37), pigs (23), and humans (5, 29, 31) in response to exercise training. Moreover, the duration of the training protocol in this study is shorter than protocols used for other species (5, 23, 29, 31, 35, 37), suggesting that mice adapt rapidly to exercise training. In FVB mice, the response to training is markedly greater than many previous findings in other animal models of exercise training. The exaggerated responses in this strain may make these mice ideal for future studies of exercise trainingrelated phenotypes. The F1 hybrid strains also showed wide variation in their response to training. Interestingly, the B6F F1 strain had a relatively small increase in endurance exercise performance, similar to the maternal BL6 strain, whereas the FB6 F1 strain had training responses more closely resembling the maternal FVB strain. This suggests that maternal inheritance may contribute significantly to adaptation responses to exercise training. However, other factors, such as intrauterine

DISCUSSION

Exercise training responses in inbred and hybrid mouse strains were characterized in the present study to examine the role of genetic background on the response to exercise training. The primary finding of this study is that there are significant strain-dependent differences in the adaptation response to exercise. FVB mice showed markedly greater improvements in endurance exercise performance after training compared with BL6 and Balb/c mice. F1 hybrids from a cross of FVB and BL6 mice showed an intermediate response to training, indicating that the adaptation responses to exercise are determined by alleles from both parents. However, further analysis of these mice showed differential responses to training based on the maternal strain, implying that the adaptation responses to training may have a significant maternal component. Recently, several studies have addressed variation in intrinsic (baseline) endurance exercise performance among inbred strains of mice or rats selectively bred for high or low levels of exercise capacity (3, 12, 19, 21, 22, 32, 34). These studies indicate that there are large strain-dependent differences in intrinsic exercise capacity. On the basis of these differences, broad-sense heritability estimates for exercise capacity ranged AJP-Regul Integr Comp Physiol • VOL

Fig. 5. Cytochrome c protein expression in mouse plantaris muscles from SED and EX mice after 4 wk of treadmill exercise. Top: representative Western blot of cytochrome c protein expression in mouse plantaris muscle from SED and EX groups of FVB mice. Bottom: quantitative densitometry of cytochrome c protein expression in plantaris muscle from 3 inbred (BL6, FVB, and Balb/c) and 3 hybrid F1 (B6F, FB6, and CB6) strains. Cytochrome c expression was significantly greater in plantaris muscle from EX vs. SED FVB and CB6 mice. Values are means ⫾ SE; n ⫽ 5– 6 animals/group for each strain. *P ⬍ 0.05 compared with SED. 288 • APRIL 2005 •

www.ajpregu.org

R1012


effects, prenatal and postnatal environment, and maternal care, need to be considered as well (7, 30). The significant contribution of genetic transmission from parent to offspring to the training response is supported by narrow-sense heritability estimates reported for rats selectively bred for high intrinsic exercise capacity (43%; see Ref. 35) and for humans (47%; see Ref. 5). In the current study, broad-sense heritability estimates for posttraining values were similar across all three measures (⬃65% for rI and ⬃47% for g2), suggesting that variance in adaptation responses to exercise in mice are significantly influenced by genotypic variance. Collectively, data indicate that the wide range of training responses in rodents and humans is determined by a significant genetic component. Several physiometrical and biochemical factors were measured to provide an indication of the efficacy of the training program and potential insight into one or more determinants of the response to exercise. Heart mass and heart mass-to-body mass ratio were significantly correlated to posttraining distance or work and changes in distance or work, respectively. The heart-to-body mass ratio only accounted for ⬃10% of the variation in changes in endurance exercise performance, whereas heart mass accounted for 15–20% of the variation in posttraining endurance exercise performance. Of the strains analyzed, the only significant difference for both heart mass and the heart-to-body mass ratio between EX and SED groups was observed for FVB mice. Although no cause-and-effect relationship can be attributed to this, these data suggest that cardiac hypertrophy may underlie some of the strain-dependent differences in adaptation responses to exercise. Furthermore, because hypertrophy was only observed in FVB mice and not associated with marked decreases in body mass, it is unlikely that this is a nonspecific stress response associated with treadmill running. There is evidence indicating that left ventricular mass and changes in cardiac dimension are genotype dependent (20, 33). However, few significant strain-dependent differences were observed for heart mass or heart mass-to-body mass ratio (Table 3). The other parameters measured to indicate differences in training status showed the expected differences between EX and SED groups of FVB mice (Figs. 4 and 5), whereas smaller or no differences were observed between groups in other inbred and F1 strains. Surprisingly, there were few differences in the gastrocnemius muscle for any of the variables measured, despite previous reports that muscle mass and markers of muscle metabolism are good indicators of exercise performance in cross-sectional studies (17, 21). Another factor that could influence the adaptation response to training is initial or intrinsic endurance exercise performance (5, 31, 35). Large differences in treadmill running performance have previously been reported for the strains used in this study (21, 22). We hypothesized that animals with low initial endurance exercise performance may show the greatest response to training. However, our data indicate that endurance exercise performance in the sedentary state was not related to posttraining endurance exercise performance (Fig. 3), implying that the genetic factors determining intrinsic endurance exercise performance are different from those underlying the response to training. Our results are comparable to those reported in the HERITAGE Family Study (5, 31) but contradict those from selectively bred rats completing an 8-wk training program (35). One explanation for this may be the difference in training regimen. In both the HERITAGE Family Study and the current AJP-Regul Integr Comp Physiol • VOL

study, training intensity was based on a percentage of maximum, whereas the selectively bred rats were trained using the same absolute workload (5, 31, 35). However, a study employing a large number of strains with similar intrinsic capacity and varying training responses would be needed to more rigorously address this question. In the current study, a relative workload was used as a training stimulus to try to maximize the training response in all strains. Because there is likely an exercise intensity threshold for inducing training responses (13, 15) and intrinsic endurance exercise capacity was significantly different among strains, using the same absolute workload for all strains might not provide an adequate stimulus for some strains. This would be a concern if the absolute workload were chosen based on the ability of the poorest-performing strain (i.e., BL6; see Ref. 35). Furthermore, the difference in absolute workload cannot explain all of the strain-dependent differences observed in the present study. Significant differences in training responses (%change) and posttraining endurance exercise performance were observed among FVB, B6F F1, and FB6 F1 strains (Fig. 1 and Table 1), all of which trained at the same absolute workload (19 m/min, 10° incline). In addition, CB6 mice were trained at a lower absolute workload (17.5 m/min, 10° incline), yet had similar posttraining values for endurance exercise performance as B6F and FB6 F1 mice (Fig. 1 and Table 1). The training response (%change) in CB6 mice was also approximately twofold greater than B6F F1 mice, although this did not reach statistical significance. Thus the magnitude of the training response is not solely dependent on the absolute workload but most likely needs to be above some relative threshold level. In summary, our data demonstrate significant strain-dependent differences in responses to exercise training in inbred and hybrid mouse strains. Improvements in endurance exercise performance in FVB were five- to sevenfold greater than BL6 and Balb/c mice, indicating that genetic background has a significant effect on training responses. Conversely, intrinsic exercise capacity and physiometric variables, such as heart mass, did not contribute significantly to the training response. The importance of genotype in the response to training in mice is supported by broad-sense heritability estimates, which suggest that ⬃40 – 60% of the variation in training responses is related to genotypic variance. In addition, F1 hybrid strains from BL6 and FVB parental strains showed differential responses to training, with only EX mice from the FB6 F1 strain (FVB mother) showing significant responses to training compared with SED mice. These data suggest that a maternal component may contribute to the interstrain differences in the responses to exercise training. ACKNOWLEDGMENTS We thank Sarah McCarty for help with mouse husbandry and treadmill training and testing. GRANTS This work was supported by funds from an American Heart Association Scientist Development Grant to M. P. Massett and National Heart, Lung, and Blood Institute Grant HL-62826 to B. C. Berk. REFERENCES 1. Atchley WR and Fitch WM. Gene trees and the origins of inbred strains of mice. Science 254: 554 –558, 1991. 288 • APRIL 2005 •

www.ajpregu.org

EXERCISE TRAINING RESPONSES IN INBRED AND HYBRID MICE 2. Banbury Conference on Genetic Background in mice. Mutant mice and neuroscience: recommendations concerning genetic background. Neuron 19: 755–759, 1997. 3. Barbato JC, Koch LG, Darvish A, Cicila GT, Metting PJ, and Britton SL. Spectrum of aerobic endurance running performance in eleven inbred strains of rats. J Appl Physiol 85: 530 –536, 1998. 4. Bernstein D. Exercise assessment of transgenic models of human cardiovascular disease. Physiol Genomics 13: 217–226, 2003. 5. Bouchard C, An P, Rice T, Skinner JS, Wilmore JH, Gagnon J, Perusse L, Leon AS, and Rao DC. Familial aggregation of VO(2max) response to exercise training: results from the HERITAGE Family Study. J Appl Physiol 87: 1003–1008, 1999. 6. Bouchard C, Rankinen T, Chagnon YC, Rice T, Perusse L, Gagnon J, Borecki I, An P, Leon AS, Skinner JS, Wilmore JH, Province M, and Rao DC. Genomic scan for maximal oxygen uptake and its response to training in the HERITAGE Family Study. J Appl Physiol 88: 551–559, 2000. 7. Cowley DE, Pomp D, Atchley WR, Eisen EJ, and Hawkins-Brown D. The impact of maternal uterine genotype on postnatal growth and adult body size in mice. Genetics 122: 193–203, 1989. 8. Crabbe JC, Wahlsten D, and Dudek BC. Genetics of mouse behavior: interactions with laboratory environment. Science 284: 1670 –1672, 1999. 9. De Angelis K, Wichi RB, Jesus WR, Moreira ED, Morris M, Krieger EM, and Irigoyen MC. Exercise training changes autonomic cardiovascular balance in mice. J Appl Physiol 96: 2174 –2178, 2004. 10. Desai KH, Sato R, Schauble E, Barsh GS, Kobilka BK, and Bernstein D. Cardiovascular indexes in the mouse at rest and with exercise: new tools to study models of cardiac disease. Am J Physiol Heart Circ Physiol 272: H1053–H1061, 1997. 11. Deschepper CF, Olson JL, Otis M, and Gallo-Payet N. Characterization of blood pressure and morphological traits in cardiovascular-related organs in 13 different inbred mouse strains. J Appl Physiol 97: 369 –376, 2004. 12. Dohm MR, Hayes JP, and Garland T Jr. The quantitative genetics of maximal and basal rates of oxygen consumption in mice. Genetics 159: 267–277, 2001. 13. Dudley GA, Abraham WM, and Terjung RL. Influence of exercise intensity and duration on biochemical adaptations in skeletal muscle. J Appl Physiol 53: 844 – 850, 1982. 14. Falconer DS. Introduction to Quantitative Genetics. New York: Wiley, 1989. 15. Hildebrandt AL, Pilegaard H, and Neufer PD. Differential transcriptional activation of select metabolic genes in response to variations in exercise intensity and duration. Am J Physiol Endocrinol Metab 285: E1021–E1027, 2003. 16. Hintze TH and Shesely EG. Is a mouse like any other mouse? J Mol Cell Cardiol 34: 1283–1286, 2002. 17. Howlett RA, Gonzalez NC, Wagner HE, Fu Z, Britton SL, Koch LG, and Wagner PD. Selected contribution: skeletal muscle capillarity and enzyme activity in rats selectively bred for running endurance. J Appl Physiol 94: 1682–1688, 2003. 18. Kemi OJ, Loennechen JP, Wisloff U, and Ellingsen O. Intensitycontrolled treadmill running in mice: cardiac and skeletal muscle hypertrophy. J Appl Physiol 93: 1301–1309, 2002. 19. Koch LG, Meredith TA, Fraker TD, Metting PJ, and Britton SL. Heritability of treadmill running endurance in rats. Am J Physiol Regul Integr Comp Physiol 275: R1455–R1460, 1998.

AJP-Regul Integr Comp Physiol • VOL

R1013

20. Landry F, Bouchard C, and Dumesnil J. Cardiac dimension changes with endurance training. Indications of a genotype dependency. JAMA 254: 77– 80, 1985. 21. Lerman I, Harrison BC, Freeman K, Hewett TE, Allen DL, Robbins J, and Leinwand LA. Genetic variability in forced and voluntary endurance exercise performance in seven inbred mouse strains. J Appl Physiol 92: 2245–2255, 2002. 22. Lightfoot JT, Turner MJ, Debate KA, and Kleeberger SR. Interstrain variation in murine aerobic capacity. Med Sci Sports Exerc 33: 2053– 2057, 2001. 23. McAllister RM, Reiter BL, Amann JF, and Laughlin MH. Skeletal muscle biochemical adaptations to exercise training in miniature swine. J Appl Physiol 82: 1862–1868, 1997. 24. Niebauer J, Maxwell AJ, Lin PS, Tsao PS, Kosek J, Bernstein D, and Cooke JP. Impaired aerobic capacity in hypercholesterolemic mice: partial reversal by exercise training. Am J Physiol Heart Circ Physiol 276: H1346 –H1354, 1999. 25. Niebauer J, Maxwell AJ, Lin PS, Wang D, Tsao PS, and Cooke JP. NOS inhibition accelerates atherogenesis: reversal by exercise. Am J Physiol Heart Circ Physiol 285: H535–H540, 2003. 26. Perusse L, Rankinen T, Rauramaa R, Rivera MA, Wolfarth B, and Bouchard C. The human gene map for performance and health-related fitness phenotypes: the 2002 update. Med Sci Sports Exerc 35: 1248 – 1264, 2003. 27. Prud’homme D, Bouchard C, Leblanc C, Landry F, and Fontaine E. Sensitivity of maximal aerobic power to training is genotype-dependent. Med Sci Sports Exerc 16: 489 – 493, 1984. 28. Rao S and Verkman AS. Analysis of organ physiology in transgenic mice. Am J Physiol Cell Physiol 279: C1–C18, 2000. 29. Rico-Sanz J, Rankinen T, Rice T, Leon AS, Skinner JS, Wilmore JH, Rao DC, and Bouchard C. Quantitative trait loci for maximal exercise capacity phenotypes and their responses to training in the HERITAGE Family Study. Physiol Genomics 16: 256 –260, 2004. 30. Rutledge JJ, Robison OW, Eisen EJ, and Legates JE. Dynamics of genetic and maternal effects in mice. J Anim Sci 35: 911–918, 1972. 31. Skinner JS, Jaskolski A, Jaskolska A, Krasnoff J, Gagnon J, Leon AS, Rao DC, Wilmore JH, and Bouchard C. Age, sex, race, initial fitness, and response to training: the HERITAGE Family Study. J Appl Physiol 90: 1770 –1776, 2001. 32. Swallow JG, Garland T Jr, Carter PA, Zhan WZ, and Sieck GC. Effects of voluntary activity and genetic selection on aerobic capacity in house mice (Mus domesticus). J Appl Physiol 84: 69 –76, 1998. 33. Swan L, Birnie DH, Padmanabhan S, Inglis G, Connell JM, and Hillis WS. The genetic determination of left ventricular mass in healthy adults. Eur Heart J 24: 577–582, 2003. 34. Totsuka Y, Nagao Y, Horii T, Yonekawa H, Imai H, Hatta H, Izaike Y, Tokunaga T, and Atomi Y. Physical performance and soleus muscle fiber composition in wild-derived and laboratory-inbred mouse strains. J Appl Physiol 95: 720 –727, 2003. 35. Troxell ML, Britton SL, and Koch LG. Selected contribution: Variation and heritability for the adaptational response to exercise in genetically heterogeneous rats. J Appl Physiol 94: 1674 –1681, 2003. 36. Wallenstein S, Zucker CL, and Fleiss JL. Some statistical methods useful in circulation research. Circ Res 47: 1–9, 1980. 37. Wisloff U, Helgerud J, Kemi OJ, and Ellingsen O. Intensity-controlled ˙ O2 max and cardiac hypertrophy. Am J Physiol treadmill running in rats: V Heart Circ Physiol 280: H1301–H1310, 2001.

288 • APRIL 2005 •

www.ajpregu.org