Selection for high aerobic capacity has no protective effect against obesity in laboratory mice

Accepted Manuscript Selection for high aerobic capacity has no protective effect against obesity in laboratory mice

Julita Sadowska, Andrzej K. Gębczyński, Marek Konarzewski PII: DOI: Reference:

S0031-9384(16)31028-9 doi: 10.1016/j.physbeh.2017.03.034 PHB 11746

To appear in:

Physiology & Behavior

Received date: Revised date: Accepted date:

11 November 2016 11 February 2017 23 March 2017

Please cite this article as: Julita Sadowska, Andrzej K. Gębczyński, Marek Konarzewski , Selection for high aerobic capacity has no protective effect against obesity in laboratory mice. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Phb(2017), doi: 10.1016/j.physbeh.2017.03.034

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Selection for high aerobic capacity has no protective effect against obesity in laboratory mice Julita Sadowska, Andrzej K. Gębczyński, Marek Konarzewski

CR

IP

Running title: Selection for Vo2max does not protect against obesity

T

Institute of Biology, University of Białystok, Ciołkowskiego 1J, 15-245 Białystok, Poland

US

Correspondence author:

AN

Julita Sadowska

ED

Ciołkowskiego 1J, Białystok 15-245

M

[email protected]

CE

PT

Poland

AC

Conflict of interest

The authors declare no conflict of interest.

1

ACCEPTED MANUSCRIPT Abstract Aerobic capacity (Vo2max measured during intensive physical exercise) both trained and intrinsic (i.e. genetically determined) has recently been deemed a good predictor of cardiometabolic risks. However, the underlying mechanisms linking Vo2max and health risk factors are not entirely clear,

IP

T

as it seems that not Vo2max per se, but rather some correlated traits, like spontaneous physical activity (SPA) are responsible for sustaining the lean phenotype. Here we investigated the link

CR

between genetically determined aerobic capacity, SPA and resistance to diet-induced health risks

US

using replicated lines of mice selected for high aerobic capacity during swimming in mid-cold water (25°C) and Randomly Bred control mice. After four months of consumption of the western

AN

type HFat and HCarb diets and no forced nor voluntary training, we found no evidence of

M

protective effects of intrinsic high Vo2max . The Selected mice displayed similar levels of blood glucose, cholesterol, triglycerides and body fat as the Random Bred control animals. Most

ED

notably we found no correlation between Vo2max and SPA levels. Our results therefore call into

PT

question the ubiquity of Vo2max as a predictor of metabolic health and leanness, at least in animal

CE

models.

AC

Keywords: Vo2max , SPA, obesity, selection experiments, diet

2

ACCEPTED MANUSCRIPT Introduction In recent years, several studies demonstrated that low aerobic capacity is a strong predictor of metabolic syndrome [1,2,3,4,5]. Specifically, a lot of attention has been paid to the intrinsic aerobic capacity (untrained, genetically determined aerobic capacity) and correlated traits as

IP

T

possible factors affecting individual susceptibility to obesity [1,6,7,8,9,10]. Obviously Vo2max, affects physical performance as it determines the rate with which muscle can utilize energy and

CR

time an individual can perform without experiencing fatigue. What is perhaps more perplexing is

US

that it also seems to have an effect on energy expenditures during low-intensity periods (e.g. lowactivity periods or recovery periods between training episodes) [11,12,13,14].

AN

Fortunately for most of us this capacity seems to be sensitive to improvement through training,

involving

alternating

bouts

of

M

with particularly good results achieved by the so called High-Intensity Interval Training (HIIT) intensive

exercise

with

low-intensity

recovery

periods

ED

[15,16,17,18,19,20,21]. This protocol clearly suggest that short bouts of intensive exercise

PT

induced Vo2max are likely to have long-lasting effect in terms of elevated metabolism during inactivity periods, in some cases resulting even in mass loss [15,16,17,18,19,20,21].

CE

One of possible explanations of this phenomena is that linking high Vo2max with high

AC

levels of spontaneous physical activity (SPA) exercised during non-training periods. SPA has been shown to significantly affect energy balance [6,22], and therefore was proposed to be the one of the potential factors counteracting obesity and lowering obesity susceptibility [8,9]. The putative positive link between Vo2max and SPA levels was corroborated by the results of recent studies demonstrating an increase in SPA as a correlated response to artificial selection for high aerobic capacity in two selection experiments on rodents [23,24,25,26]. Such experiments created animal models characterized by variation of a trait in question (here Vo2max ) and correlated traits 3

ACCEPTED MANUSCRIPT (SPA) significantly higher than that, observed in randomly bred populations. In particular, Vo2max of selected mouse lines is comparable or higher than that found in mouse lines subject to the equivalent of training (e.g., a 0.4% or 28% HIIT- elicited increase in Vo2max reported by Hafstad et al. [18] and Kruger et al. [27] vs 30% increase resulting from artificial selection in this study).

T

This exaggerated variability facilitates studies on physiological correlations between Vo2max and

IP

other traits, such as SPA. Furthermore, unlike purely phenotypic response elicited by training, a

CR

considerable increase of Vo2max resulting from changes in the underlying genetic makeup ensures consistency in the studied trait unmatched by studies on phenotypic level [28,29].

which alone is by far the most significant determinant of

AN

diet composition [30,31],

US

However, apart from high aerobic capacity, SPA is also affected by other factors, chiefly

cardiometabolic risks [25,32]. Here we investigated the link between genetically determined

M

aerobic capacity, SPA and resistance to diet-induced obesity using mice selected for high aerobic

at 25th generation

ED

capacity during swimming in mid-cold water (25°C) and Randomly Bred control mice. Currently, the selected animals are characterized by a 30% difference in Vo2max However, at 10th generation, selection for swim-induced Vo2max

PT

compared to control mice.

CE

already resulted in a correlated response manifested by larger hearts and gastrocnemius muscle in the Selected animals, along with significantly higher Vo2max , longer distance run till exhaustion

AC

and longer run time - all elicited by forced treadmill running, [33]. To test whether the intrinsic high Vo2max has any protective effects against diet induced metabolic disturbances we exposed the Selected and Random Bred animals to western type diets (high fat and high carb) for four months with no access to running wheels or any other forms of forced nor voluntary physical training. We assessed the diet effects by monitoring body mass, food consumption and energy assimilation, SPA, blood glucose and lipid profile.

4

ACCEPTED MANUSCRIPT Materials and methods Selection animals We used male Swiss Webster mice from the 25th generation of a long term selection experiment

T

designed to produce 4 replicate lines of mice with high maximal metabolic rate elicited by

IP

swimming in 25°C water (from here on called Vo2max ; Selected line types), and mice from 4

CR

Randomly Bred lines which serve as a control for the selection. Briefly, in each isolated line we maintain 10-17 families per generation in 4 Selected and 4 Random Bred line types. At the age of

US

12 weeks mice from all lines are subjected to Vo2max measurements during swimming. To

AN

measure Vo2max we used a cylindrical see-through metabolic chamber supplied with a movable platform. The chamber was partly filled with water leaving a dry volume of 560 ml and placed in

–1

flow rate (Sierra Instruments, Monterey, CA, or Beta Erg, Warsaw,

ED

chamber at 700 ml min

M

a water bath to maintain at 25 ± 0.2°C. Atmospheric air was dried (Drierite), pumped through the

Poland), re-dried (Drierite), scrubbed of CO 2 (Carboabsorb, AS, BDH Laboratory Supplies) and

PT

directed to oxygen analyzer (S-3A/I Applied Electrochemistry). Each mouse was individually

CE

placed just above the water level on the platform, and allowed 10 min for acclimation. The platform was then abruptly submerged to force the animal to swim for a 5 minute period. Vo2max

AC

was defined as the highest body-mass-corrected oxygen consumption averaged over 2 min of 5 minutes of swimming. In the selected line types mice with the highest mass corrected Vo2max are chosen as progenitors for the next generation, and mice from the control line types are bred randomly (excluding sibling breeding). Experimental setup

5

ACCEPTED MANUSCRIPT All procedures were approved by the by the Local Ethical Committee on Testing Animals, Medical University of Białystok, Poland (permit no. 42/2011, 11/2013, 21/2013). Mice were randomly divided into three groups (10 animals per each replicate of Selected and Random Bred line type, equaling 80 animals per diet group). Each group was then randomly

IP

T

assigned a diet protocol: high fat (HFat diet; 40% energy from fat; 104 14.7 kJ g-1 metabolizable energy, 18,3 kJ g-1 caloric value, Labofeed, Kcynia, Poland), high carbohydrate (HCarb diet;

CR

70% energy from carbohydrates; 15.3 kJ g-1 metabolizable energy, 17.2 kJ

g-1 caloric value,

US

Labofeed, Kcynia, Poland) or a standard control (C diet; 12.8 kJ g-1 metabolizable energy, 17.0 kJ g-1 caloric value Labofeed, Kcynia, Poland; for details see Supplement Information) diet on

AN

which the animals were then maintained for the following 16 weeks of the experiment. Mice

M

were housed individually in cages with sawdust bedding, unlimited access to food and water and standard housing conditions for this colony of mice (23°C, 12L:12D). Body mass was recorded

PT

Energy assimilation

ED

weekly.

CE

At the end of the first and last month we measured energy assimilation from food for all animals in the three diet groups. For this we measured food consumption by placing animals in cages

AC

equipped with plastic grids. Food remains from the bottom of the cage were collected and separated from feces, then dried in an oven at 65°C, and weighed to the nearest 0.01 g. Average food intake was calculated individually for each mouse during two consecutive 2-day trials as the mass of food disappearing from the food dispenser minus orts. Caloric value of food and feces was estimated by oxygen bomb calorimetry (IKA Werke 7000 calorimeter, Germany). Using

6

ACCEPTED MANUSCRIPT these values daily energy assimilation was calculated for each animal as follows: ((mass of food consumed × caloric value of food) – (mass of feces × caloric value of feces))/2.

T

Activity measurements

IP

Spontaneous activity was measured four times (at the end of each month) throughout the

CR

experiment. Unfortunately, due to a computer malfunction data on voluntary activity from the 4 th month were lost. For the measurements we used passive infrared sensors (TL-xpress, Crow

US

Electronics Engineering, Fort Lee, NJ, USA). Sensors were installed over each cage and were

AN

monitored each 1 s by a computer (PCL-711 analog–digital interface, Advantech, Cincinnati, OH, USA). Activity was measured during two full days, but data for the analysis comprised a

M

24h period. Following Copes et al. [24] and Brzęk et al. [34] we analysed each animal’s activity

ED

in three ways: as (1) total SPA, the daily sum of all active periods (total activity; [34,35]), (2) the duration of SPA calculated as the number of 1-minute intervals at which any SPA was recorded;

PT

(3) as activity intensity, the average amount of activity per minute when any home-cage activity

CE

was occurring, calculated as total SPA divided by the duration of SPA.

AC

Blood and tissue sample analysis In the week preceding the end experiment we measured the fasting blood glucose levels. We used blood glucose test strips (Optium Xido, Abbott, UK). The tail vein was punctured with a sterile needle and a blood droplet was used immediately to perform the test. In the final 16 week of the experiment blood samples from each mouse were collected via orbital sinus puncture, centrifuged immediately. Blood samples were immediately centrifuged, serum

7

ACCEPTED MANUSCRIPT was collected and stored at -80°C. The total cholesterol as well as HDL and LDL/VLDL cholesterol fractions were measured with the BioAssay Systems HDL and LDL/VLDL Assay Kit (E2HL-100). The triglyceride blood concentration was measured with the BioAssay Systems Triglyceride Assay Kit (ETGA-200).

IP

T

Afterwards animals were killed and dissected. Carcasses were frozen for further body fat

CR

analysis.

US

Body fat analysis

Thawed carcasses were dried at 65°C to a constant mass and then homogenized with an electric

AN

mill. Fat was extracted from homogenate weighted samples with petroleum ether in a Soxhlet

M

extractor. The residues remaining after extraction were then re-dried, and the fat content was

ED

calculated as the mass lost during extraction [36].

PT

Statistics

CE

Vo2max and organ masses were analysed with ANCOVA model with line type (Selected vs. Random Bred) and diet protocol as fixed factors, replication nested within the line type and

AC

family affiliation nested within the line type and replication as random factors and body mass as a covariate. Body mass gain rate (expressed as regression coefficients of individual mass gain for weeks 2-16) and body fat were analysed with ANOVA with line type and diet as fixed factors, replication nested within the line type and family affiliation nested within the line type and replication as random factor. Spontaneous activity was analysed with repeated measures ANOVA

8

ACCEPTED MANUSCRIPT with line type, diet and sensor number as fixed factors, family affiliation nested within the line type and replication as a random factors. Food consumption and assimilation for the 4 months was calculated with RM-ANCOVA with line type and diet protocol as fixed factors, family affiliation nested within the line type and

T

replication as a random factors, and body mass as a covariate.

IP

Blood glucose, triglyceride and cholesterol concentrations (total, HDL and LDL/VLDL fractions)

CR

were analysed with ANOVA with line type and diet type as fixed factors,and family affiliation nested within the line type and replication as a random factors.

AN

US

All statistical analyses were carried out with SAS 9.3 software (SAS Institute, Cary NC, USA).

Results

M

Vo2max , organ, body and dry body fat mass

ED

Vo2max differed significantly between the Selected and RB line types (F1,6 = 78.65; P < 0.001; Selected: 323.1 ± 3.3 ml O 2 /h; Random Bred: 247.2 ± 3.2 ml O 2 /h). There were no differences in

PT

initial body mass between the Selected and RB line types (F 1,6 = 0.77; P = 0.463), as well as no

CE

differences between the experimental groups. Body mass gain rate showed no significant differences between the line types (F 1,6 = 0.14; P =

AC

0.796) but was affected by the diet (F2,122 = 12.83; P < 0.001). There was also no line type × diet interaction (F2,122 = 0.29; P = 0.747). Dry body fat mass was not affected by the line type affiliation (F 1,6 = 3.82; P = 0.098; Figure 1), diet protocol (F2,22 = 2.93; P = 0.074) nor body mass (F115,22 = 1.33; P = 0.793). There was no line type × diet interaction (F2,22 = 0.23; P = 0.793).

Energy assimilation 9

ACCEPTED MANUSCRIPT The line type × diet × measurement order interaction for energy assimilation was significant (F7,365 = 2.99; P = 0.004), therefore we analysed the data for the two measurements separately. In both the first and last month we found no between-line type differences, but a significantly higher assimilation of energy from the HCarb diet in mice from both line types (Table 1). There were no

IP

T

line type × diet interactions.

CR

Spontaneous activity

Total SPA analysis revealed a line type × diet × measurement order interaction (F 12,519 = 2.84; P

US

= 0.001) therefore we analysed the data for each measurement separately. In all three

AN

measurements there were no between-line type differences (F1,6 = 0.01; P = 0.989; F1,66 = 0.01; P = 0.927; F1,6 = 0.04; P = 0.853; Figure 2A). The HCarb and HFat diet protocols however seemed

M

to elicit a drop in total SPA in all three months (F 2,110 = 27.95; P < 0.001; F2,83 = 17.76; P <

ED

0.001; F2,105 = 4.24; P = 0.017). The line type × diet interaction was significant only in the first measurement (F2,110 = 6.54; P = 0.002; F2,83 = 1.92; P = 0.154; F2,105 = 0.46; P = 0.634).

PT

SPA duration was not affected by the line type affiliation (F1,6 = 0.25; P = 0.635) nor the

CE

measurement order (F2,519 = 0.38; P = 0.687), but there was a significant effect of the diet type (F2,519 = 119.96; P < 0.001; Figure 2B). There was no line type × diet × measurement order

AC

interaction (F12,519 = 1.72; P = 0.059). For SPA intensity we found (Figure 2C) a line type × diet × measurement order interaction (F 12,519 = 6.06; P < 0.001), therefore we also analysed this parameter separately for each measurement. We found no between-line type differences in all three measurements (F1,6 = 2.38; P = 0.173; F1,6 = 0.13; P = 0.733; F1,6 = 0.12; P = 0.741), but a significant effect of the diet (F 2,110 = 26.15; P < 0.001; F2,83 = 9.54; P = 0.001; F2,105 = 8.50; P = 0.001). The line type × diet interaction was however significant in all three measurements (F 2,110 = 11.37; P < 0.001; F2,83 = 12.08; P < 0.001; F2,105 = 5.05; P = 0.008). 10

ACCEPTED MANUSCRIPT

Lipid profile and blood glucose Cholesterol fractions were not affected by the line type (total cholesterol: F1,6 = 0.1, P = 0.917; HDL: F1,6 = 0.04, P = 0.852; LDL/VLDL: F 1,6 = 0.02, P = 0.901; Figure 3), but significantly

T

affected by diet (total cholesterol: F2,115 = 103.83, P < 0.001; HDL: F2,115 = 103.89, P < 0.001;

IP

LDL/VLDL: F2,120 = 4.68, P = 0.011; Figure 3). We found no line type × diet interactions (total:

CR

F2,115 = 0.75, P = 0.476; HDL: F2,115 = 0.71, P = 0.491; LDL/VLDL: F2,120 = 0.12, P = 0.886). Likewise, triglyceride blood level was not affected by the line type affiliation (F 1,6 = 0.20, P =

US

0.674), but we found a significant effect of the diet protocol (F 2,126 = 6.53, P = 0.002; Figure 3).

AN

The line type × diet interaction was however at the verge of significance (F 2,126 = 3.08, P = 0.050).

M

Fasting blood glucose showed no between-line type differences (F1,6 = 1.8; F = 0.226; Figure 4).

ED

It was however affected by the diet (F2,133 = 12.68; P < 0.001) and body mass (F1,133 = 8.17; F

CE

Discussion

PT

= 0.005). We found no line type × diet interaction (F2,133 = 0.97; P = 0.379).

AC

Aerobic capacity determines one’s ability for physical exercise and endurance but apparently also lowers the risk of developing metabolic syndrome even in non-active or mildly active human and animal subjects [1,2,3,4,5,26,37]. Here, despite a 30% difference in genetically determined aerobic capacity the Selected line types were not any less prone to the effects of HFat and HCarb diets than the Random Bred mice. After four months under both regimens emulating the westerntype diets and a “sedentary lifestyle” with no forced nor voluntary physical training they were displaying higher cholesterol, triglyceride and blood glucose levels compared to the control 11

ACCEPTED MANUSCRIPT feeding regimen (Figure 3, 4). Both line types showed similar body mass gain under the obesogenic diets, but the Selected mice displayed slightly higher dry body fat mass % values. However this result has not attained statistical significance (Figure 1). Altogether these observations contradicted our expectations. Nevertheless, since our study was based on a robust,

T

replicated selection experiment, the lack of expected between line-type differences merits closer

IP

examination.

CR

In contrast to the results presented therein, in an un-replicated selection experiment, in which rats

US

were divergently bred for high and low treadmill-running capacity (and thus high/low Vo2max ) a suite of metabolic and cardiovascular related risk factors (increased body mass, elevated blood

AN

glucose and triglyceride levels, lower lipid oxidation rates) started to emerge in the low runner

M

animals just after 6 generations under normal and obesity-inducing feeding regimen, whereas high runner rats remained resistant to such disturbances [1,7,37,38]. It therefore seems that while

ED

low Vo2max constitutes a risk factor for metabolic syndrome, high intrinsic Vo2max alone has no

PT

beneficial effects, as demonstrated by our study. The lack of such effects seems to contrast with the positive effects of HIIT in rats [12] and mice [18,14], but also humans. In both physically

CE

active and inactive humans a structured HIIT training of even as little as few minutes applied a

AC

couple times per week can induce an increase in spontaneous physical activity (SPA), and thus in daily energy expenditures that has a lasting effect [39,40]. This points to the high intensity, low volume training as a potent tool in amplifying the innate role of Vo2max and maintaining metabolic health through elevated SPA. SPA (also called the non-exercise activity thermogenesis, or occupation-related activity) is the activity associated with everyday tasks (e.g. walking). Although often underestimated, SPA might have a significant impact on the daily energy expenditures, particularly in sedentary 12

ACCEPTED MANUSCRIPT populations [30,41]. Its significance has been recently tested as a correlated response in artificial selection experiments targeting metabolic rates in laboratory rodents. They revealed that SPA is heritable [22,24,34], albeit complex trait [24,30,34,42]. Total SPA is a sum of movement intensity and duration of the active phase - both of these components can be affected by different

T

factors (e.g. diet) and respond independently from one another [24]. Therefore, as demonstrated

IP

by recent literature it is a good practice to analyze both total SPA as well as its components

CR

[24,34].

US

Here, despite a highly significant between-line type genetically determined differences in Vo2max we found no correlated changes in total SPA nor its duration or intensity. The only visible

AN

modification was the diet-induced drop in total SPA, SPA duration and intensity in the HCarb fed

M

group (Figure 2). This is in disagreement with the results reported by Koch and Britton’s team, who found that high intrinsic aerobic capacity is a predictor of high SPA in lines of rats selected

ED

on a high/low treadmill-running capacity [25,26,43]. It is important to note, however, that in the

PT

absence of comparison with the control (Randomly Bred lines) a positive relationship between intrinsic aerobic capacity and SPA in rats selected by Koch and Britton is dubious. Furthermore,

CE

mice selected for high voluntary running (and indirectly higher Vo2max ) exhibited elevated SPA

AC

mainly in the absence of running wheels [8,23]. The same parent strain of mice, when selected for high home cage activity (SPA), showed no correlated response in wheel running [22]. Conversely, in our own selection experiment for divergent BMR levels (high-BMR and lowBMR line types derived from Swiss Webster mice) we observed consistent between-line type differences in SPA in subsequent generations of mice, with high-BMR animals being consistently more active under a series of different conditions [34,35]. Although these line types show a 50% difference in BMR, and a correlated response in SPA, there are no differences in aerobic capacity

13

ACCEPTED MANUSCRIPT between them (measured during forced treadmill running; [35]). Interestingly, selection for Vo2max

in animals used here did not elicit any changes in BMR in comparison to the Random

Bred mice, which suggests the absence of genetic correlation between the two traits [33]. The above cited studies indicate that genetic correlation between aerobic capacity and SPA is not

IP

T

a common occurrence. One may argue that the applied selection protocol may be of ample importance, as it may target different components of aerobic performance, and thus affect

CR

different sets of genes, with varying levels of co-occurrence with SPA. From one hand, Koch and

US

Britton select for high and low running aerobic capacity measured during forced treadmill exercise [1,7], and thus are targeting directly physical performance and endurance. On the other

AN

hand, Garland’s selection experiment targets high voluntary activity and under this protocol

M

animals are not forced to exercise and use the wheels as they wish [44,45]. In this case the main trait being selected is running behavior, not the physiological mechanisms underlying physical

ED

performance itself. Our protocol involves physical exercise – swimming, but in a 25°C water,

PT

which is significantly below thermoneutality. This means that the aerobic capacity measured and selected for has two physiological components: physical performance and thermoregulatory

CE

response. So far this dual selection triggered a significant response in thermogenic capacity

AC

(Gębczyński unpublished; [33]) in the Selected line types, but more importantly resulted not only in higher Vo2max of the Selected mice measured during swimming, but also significantly higher endurance during treadmill running tests (as higher Vo2max , duration of running and 18% difference in distance run till exhaustion measured in generation 10; [33]). It is also important to note that apart from metabolic challenge, forced swimming elicits emotional stress manifested by so called swim- stress induced analgesia (SSIA) [46,47], which may be also associated with an increased propensity to depressive states negatively affecting SPA [48,49]. However, the

14

ACCEPTED MANUSCRIPT propensity to emotional stress is negatively associated with Vo2max, so our mice having high swim-elicited Vo2max are unlikely to manifest correlated behavioural responses, which could reduce their SPA. This additionally validates the relevance of our model of choice for testing hypotheses on any effects of intrinsic Vo2max in retaining metabolic health.

IP

T

In conclusion, despite significant between line type differences in Vo2max , and correlated differences in physical endurance (implicated in dietary responses) we were unable to detect a

CR

protective effect of Vo2max against diet induced health risks. This calls into question the ubiquity

US

of Vo2max per se as a trait positively genetically correlated with metabolic health and leanness, at least in animal models. In contrast, in the companion paper (Sadowska, Gębczyński,

AN

Konarzewski unpublished) we demonstrated a positive link between high BMR and metabolic

M

fitness mediated by a positive correlation between BMR and SPA. Overall, our results point to BMR, rather than Vo2max , as a metabolic trait related to obesity and emphasize the need to study

ED

different models for understanding the complex genetic relationships between SPA and other

CE

PT

traits.

Acknowledgments

AC

This study was funded by the National Science Centre of Poland (DEC-2011/01/N/NZ8/02009). Special thanks to M. Lewoc, S. Płonowski and B. Lewończuk for help with animal maintenance. We are grateful to Max Plank Institute for Ornithology in Radolfzell Germany for the calorimetry equipment. JS is supported by the Foundation for Polish Science (FNP).

15

ACCEPTED MANUSCRIPT References [1] U. Wisloff , S.M. Najjar, O. Elligsen, P.M. Haram, S. Swoap, Q. Al-Share, M. Fernstrom, K. Rezaei, S.J. Lee, L.G. Koch, S.L. Britton, Cardiovascular risk factors emerge after artificial selection for low aerobic capacity, Science 307 (2005) 418 – 420.

IP

T

[2] S. Kodama, K. Saito, S. Tanaka, M. Maki, Y. Yachi, M. Asumi, A. Sugawara, K. Totsuka, H. Shimano, Y. Ohashi, N. Yamada, H. Sone, Cardiorespiratory Fitness as a Quantitative

CR

Predictor of All-Cause Mortality and Cardiovascular Events in Healthy Men and

US

Women. A Meta-analysis, JAMA 301 (2009) 2024 – 2035.

[3] D.A. Rivas, S.J. Lessard, M. Saito, A.M. Friedhuber, L.G. Koch, S.L. Britton, B.B. Yaspelkis,

AN

J.A. Hawley, Low intrinsic running capacity is associated with reduced skeletal muscle

M

substrate oxidation and lower mitochondrial content in white skeletal muscle, Am. J. Physiol. Regul. Integr. Comp. Physiol. 300 (2011) R835–R843.

ED

[4] T. Yokota, S. Kinugawa, K. Okita, K. Hirabayashi, T. Suga, M. Hattori, Y. Nakagawa, N.

PT

Oyama-Manabe, H. Shirato, H. Tsutsui, Lower aerobic capacity was associated with abnormal intramuscular energetics in patients with metabolic syndrome. Hypertens. Res.

CE

34 (2011) 1029–1034.

AC

[5] L. Nyboe, C.H. Vestergaard, M.K. Moeller, H. Lund, P. Videbech, Metabolic syndrome and aerobic fitness in patients with first-episode schizophrenia, including a 1-year follow-up, Schizophr. Res. 168 (2015) 381 – 387. [6] A.W. Thorburn, J. Proietto, Biological determinants of spontaneous physical activity, Obes. Rev. 1 (2000) 87 – 94.

16

ACCEPTED MANUSCRIPT [7] L.G. Koch, S.L.

Britton, Artificial selection for intrinsic aerobic endurance running capacity

in rats, Physiol. Gen. 5 (2001) 45 – 52. [8] L.M. Vaanholt, I. Jonas, M. Doornbos, C. Nyakas, T. Garland, G.H. Visser, G. van Dijk, Metabolic and behavioral responses to high-fat feeding in mice selectively bred for high

IP

T

wheel-running activity. Int. J. Obes. 32 (2008) 1566 – 1575. [9] T.H. Meek, J.C. Eisenmann, B.K. Keeney, R.M. Hannon, E.M. Dlugosz, T. Garland, Effects

CR

of early-life exposure to western diet and wheel access on metabolic syndrome profiles

US

in mice bred for high voluntary exercise. Genes, Brain and Behavior 13 (2014) 322 – 332.

AN

[10] M.E. Smyers, K.Z. Bachir, S.L. Britton, L.G. Koch, C.M. Novak. 2015. Physically active

M

rats lose more weight during calorie restriction. Physiol. Behav.139 (2015) 303–313.

ED

[11] R.A. Jacobs, D. Fluck, T.C. Bonne, S. Burgi, P. Moller Christensen, M. Toigo, C. Lundby, Improvements in exercise performance with high-intensity interval training coincide

PT

with an increase in skeletal muscle mitochondrial content and function. J. Appl. Physiol.

CE

115 (2013) 785 – 793.

[12] M.A. Linden, J.A. Fletcher, E.M. Morris, G.M. Meers, M.H. Laughlin, F.W. Booth, J.R.

AC

Sowers, J.A. Ibdah, J.P. Thyfault, R.S. Rector, Treating NAFLD in OLETF Rats with Vigorous-Intensity Interval Exercise Training. Med. Sci. Sports Exerc. 47 (2015) 556– 567.

[13] M. Pimenta, I. Bringjenti, V. Souza-Mello, I.K. dos Santos Mendes, M.B. Aguila, C. Mandarim-de-Lecerda, High-intensity interval training beneficial effects on body mass, blood pressure, and oxidative stress in diet-induced obesity in ovariectomized mice. Life Sci. 139 (2015) 75 – 82. 17

ACCEPTED MANUSCRIPT [14] V.F. Motta, M.B. Aguila, C.A. Mandarim-de-Lecerda, High-intensity interval training (swimming) significantly improves the adverse metabolism and comorbidities in dietinduced obese mice. J. Sports Med. Phys. Fitness 56 (2016) 655- 663. [15] K.A. Burgomaster, S.C. Hughes, G.J. F. Heigenhauser, S.N. Bradwell, M.J. Gibala, Six

T

sessions of sprint interval training increases muscle oxidative potential and cycle

IP

endurance capacity in humans. J. Appl. Physiol. 98 (2005) 1985–1990.

CR

[16] P.M. Haram, O.J. Kemi, S.J. Lee, M.O. Bandheim, Q. Al-Share, H.L. Waldum, L.J. Giligan, L.G. Koch, S.L. Britton, S.M. Najjar, U. Wisloff,. Aerobic interval training vs.

US

continuous moderate exercise in the metabolic syndrome of rats artificially selected for

AN

low aerobic capacity, Cardiovas. Res. 81 (2009) 723 – 732. [17] D. Stensvold, A.E. Tjønna, E.A. Skaug, S. Aspenes, T. Stølen, U. Wisløff, S.A. Slørdahl,

M

Strength training versus aerobic interval training to modify risk factors of metabolic

ED

syndrome. J. Appl. Physiol. 108 (2010) 804–810.

PT

[18] A.D. Hafstad, J. Lund, E. Hadler-Olsen, A.C. Höper, T.S. Larsen, E. Aasum, High- and Moderate-Intensity Training Normalizes Ventricular Function and Mechanoenergetics in

CE

Mice With Diet-Induced Obesity. Diabetes 62 (2013) 2287 – 2294.

AC

[19] M. Weston, K.L. Taylor, A.M. Batterham, W.G. Hopkins, Effects of Low-Volume HighIntensity Interval Training (HIT) on Fitness in Adults: A Meta-Analysis of Controlled and Non-Controlled Trials. Sports. Med. 44 (2014) 1005–1017. [20] T.A. Astorino, R.M. Edmundus, A. Clark, L. King, R.M. Gallant, S. Namm, A. Fischer, K.A. Wood, High-intensity interval training increases cardiac output and Vo 2max . Madicine & Science in Sports & Exercise (2016) doi.10,1249/MSs/0000000000001099

18

ACCEPTED MANUSCRIPT [21] Z. Kong, S. Sun, M. Liu, Q. Shi, Short-term high-intensity interval training on body composition and blood glucose in overweight and obese young women. Journal of Diabet. Res. (2016) 4073618 [22] J.A. Zombeck, E.K. DeYong, W. Brzezinska, J.S. Rhodes. Selective breeding for increased

T

home cage physical activity in collaborative cross and Hsd:ICR mice. Behav. Genet. 41

IP

(2011) 571-582.

CR

[23] W. Acosta, T.H. Meek, H. Schutz, E.M. Dlugosz, K.T. Vu, T. Garland, Effects of earlyonset voluntary exercise on adult physical activity and associated phenotypes in mice.

US

Physiol. Behav. 149 (2015) 279 – 286.

AN

[24] L.E. Copes, H. Schutz, E.M. Dlugosz, W. Acosta, M.A. Chappell, T. Garland, Effects of voluntary exercise on spontaneous physical activity and food consumption in mice:

M

results from anartificial selection experiment. Physiol. Behav. 149 (2015) 86–94.

ED

[25] C.M. Novak, C. Escande, P.R. Burghardt, M. Zhang, M.T. Barbosa, E.N. Chini, S.L. Britton, L.G. Koch, H. Akil, J.A. Levine, Spontaneous activity, economy of activity, and

PT

resistance to diet-induced obesity in rats bred for high intrinsic aerobic capacity,

CE

Hormones Behav. 58 (2010) 355 -367. [26] S. Karvinen, K. Waller, M. Silvennoinen, L.G. Koch, S.L. Britton, J. Kaprio, H.

AC

Kainulainen, U.M. Kujala, Physical activity in adulthood: genes and mortality. Sci. Rep. 5 (2015) 18258. [27] K. Kruger, D.K. Gessner, M. Seimetz, J. Banisch, R. Ringseis, K. Eder, N. Weissmann, F.C. Mooren, Functional and muscular adaptations in an experimental model for isomeric strength training in mice. Plos One 8 (2013) e79069. [28] T.J. Morgan, T. Garland, P.A. Carter, Ontogenesis in mice selected for high voluntary wheel-running activity. I. Mean ontogenesis, Evolution 57 (2003) 646 – 657. 19

ACCEPTED MANUSCRIPT [29] J. Sadowska, A.K. Gębczyński, M. Konarzewski. Effect of reproduction on the consistency of the between-line type divergence in laboratory mice selected on BMR. Physiol. Biochem. Zool. 88 (2015) DOI: 10.1086/680167 [30] J.A. Teske, C.J. Billington, M.A. Kuskowski, C.M. Kotz, Spontaneous physical activity

IP

T

protects against mass gain. Int. J. Obes. (Lond) 36 (2012) 603–613. [31] S. Guidotti, N. Meyer, E. Przybyt, A.J.W. Scheurink, M.C. Harmsen, T. Garald, G. van Dijk,

CR

Diet-induced obesity resistance of adult female mice selectively bred fir increased

US

wheel-running behavior is reversed by single perinatal exposure to a high-energy diet. Physiol. Behav. 157 (2016) 246 – 257.

AN

[32] S.K. Keadle, K. Lyden, J. Staudenmayer, A. Hickey, R. Viskochil, B. Braun, P.S. Freedson,

M

The independent and combined effects of exercise training and reducing sedentary behavior on cardiometabolic risk factors. Appl. Physiol. Nutr. Metab. 39 (2014) 770-

ED

780.

PT

[33] A.K. Gębczyński, M. Konarzewski, Metabolic correlates of selection on aerobic capacity in laboratory mice: a test of the model for the evolution of endothermy, J Exp Biol 212

CE

(2009) 2872-2878.

AC

[34] P. Brzęk, A.K. Gębczyński, A. Książek, M. Konarzewski, Effect of calorie restriction on spontaneous physical activity and body mass in mice divergently selected for basalmetabolic rate (BMR), Physiol Behav (2016) 10.1016/j.physbeh.2016.04.022. [35] A.K. Gębczyński, M. Konarzewski, Locomotor activity of mice divergently selected for basal metabolic rate: a test of hypotheses on the evolution of endothermy. J Evol Biol 22 (2009) 1212-1220.

20

ACCEPTED MANUSCRIPT [36] K. Sawicka-Kapusta, Fat extraction in Soxhlet apparatus. In Methods for Ecological Energetics (ed. W. Grodzinski, R. Z. Klekowski and A. Duncan), (1975) pp 288- 293. Oxford: Blackwell Scientific Publications. [37] L.G. Koch, O.J. Kemi, N. Qi, S.X. Leng, P. Bijma, L.J. Gilligan, J.E. Wilkinson, H. Wisløff,

T

M.A. Høydal, N. Rolim, P.M. Abadir, I. Van Grevenhof, G.L. Smith, C.F. Burant, Ø.

IP

Ellingsen, S.L. Britton, U. Wisløff, Intrinsic aerobic capacity sets a divide for aging and

CR

longevity: Koch & Kemi-Rat models link exercise capacity with mortality. Circ Res 109 (2011) 1162–1172.

US

[38] E.M. Morris, M.R. Jackman, G.C. Johnson, T-W. Liu, J.L. Lopes, M.L. Kearney, J.A.

AN

Fletcher, G.M.E. Meers, L.G.Koch, S.L. Britton, R.S. Rector, J.A. Ibdah, P.S. MacLean, J.P. Thyfault, Intrinsic aerobic capacity impacts susceptibility to acute high-fat diet-

M

induced hepatic steatosis. Am J Physiol. Endocrinol. Metab. 307 (2014) E355 – E364.

ED

[39] U. Eiholzer, Y. Nordmann, D. L’Allemand, M. Schlumpf, S. Schmid, K. KromeyerHauschild, Improving body composition and physical activity in Prader-Willi Syndrome.

PT

J Pediatr . 142 (2003) 7378.

CE

[40] U. Eiholzer, U. Meinhardt, R. Petro, F. Witassek, F. Gutzwiller, T. Gasser, High-intensity training increases spontaneous physical activity in children: a randomized controlled

AC

study, J. Pediatr. 156 (2010) 242-246. [41] T. Garland, H. Schutz, M.A. Chappell, B.K. Keeney, T.H. Meek, L.E. Copes, W. Acosta, C. Drenowatz, R.C. Maciel, G. van Dijk, C.M. Kotz, J.C. Eisenmann , The biological control of voluntary

exercise,

spontaneous

physical activity

and

daily

energy

expenditure in relation to obesity: human and rodent perspectives. J. Exp. Biol. 214 (2011) 206 – 229.

21

ACCEPTED MANUSCRIPT [42] P. Majdak, P.J. Bucko, A.L. Holloway, T.K. Bhattacharya, E.K. DeYoung, C.N. Kilby, J.A. Zombeck, J.S. Rhodes, Behavioral and pharmacological evaluation of selectively bred mouse model of home cage hyperactivity. Behav. Genet. 44 (2014) 516 – 534. [43] C.K. Gavini, S. Mukherjee, C. Shukla, S.L. Britton, L.G. Koch, H. Shi, C.M. Novak,

T

Leanness and heightened nonrusting energy expenditure: role of skeletal muscle activity

IP

thermogenesis. Am. J. Physiol. Endocrinol. Metab. 306 (2014) E635 – E647.

CR

[44] J.G. Swallow, T. Garland, P.A. Carter, W-Z. Zhan, G.C. Sieck, Effects of voluntary activity and genetic selection on aerobic capacity in house mice (Mus domesticus). J. Appl.

US

Physiol. 84 (1998) 69–76.

AN

[45] V. Careau, M.E. Wolak, P.A. Carter, T. Garland, Limits to behavioral evolution: the

M

quantitative genetics of a complex trait under directional selection. Evolution 67 (2013) 3102 – 3119.

ED

[46] I.B. Łapo, M. Konarzewski, B. Sadowski. Differential metabolic capacity of mice selected

PT

for magnitude of swim stress-induced analgesia. J. Appl. Physiol. 94 (2003) 677-784. [47] I.B. Łapo, M. Konarzewski, B. Sadowski. Analgesia induced by swim stress in the mouse: a

CE

relationship between analgesic and thermoregulatory mechanisms. Pflügers Arch. 446

[48]

AC

(2003) 463-369.

A. Can, D.T. Dao, C.E. Terrillion, T.D. Gould. The mouse forced swim test. J. Vis. Exp. 59 (2012) e3638.

[49]

Y.S. Mineur, C. Belzung, W.E. Crusio. Effects of unpredictable chronic mild stress on anxiety and depression-like behavior in mice. Behav. Brain Res. 175 (2006) 43 – 50.

22

ACCEPTED MANUSCRIPT Table 1. ANOVA results and mean values of food assimilation of mice form the Selected and Random Bred line types fed a HFat, HCarb and Control diet.

Line type × Month

Line type

Diet

I R

Body mass

(kJ day-1 )

diet

F1,141 =

F2,141 =

20.88

6.51

F1,6 = 1.14 1

HCarb P = 0.354

A

P = 0.012

F2,132 =

F2,132 =

50.73

15.25

Control

D E

T P

F1,6 = 0.89 4

U N

F2,141 = 1.05 P < 0.001

C S

HFat

P = 0.327

T P

S

M

HFat

RB

Between-diet

(kJ day-1 )

differences

79.86 ± 2.91

72.93 ± 2.90

a

85.56 ± 2.90

82.96 ± 2.90

b

68.03 ± 2.91

68.13 ± 2.91

c

77.79 ± 2.90

72.27 ± 2.96

a

71.82 ± 2.95

68.07 ± 2.97

b

60.66 ± 2.93

59.45 ± 2.96

c

F2,132 = 1.06

P = 0.381

E C

P < 0.001

HCarb

P = 0.349

P < 0.001

Control

C A

23

ACCEPTED MANUSCRIPT Figure 1. Dry body fat mass % in Selected (black symbols) and Random Bred (open symbols) line types of mice fed a HFat, HCarb and Control diet. Symbols represent adjusted means for

AC

CE

PT

ED

M

AN

US

CR

IP

T

each replicate of the respective line type.

24

ACCEPTED MANUSCRIPT Figure 2 . Total SPA (A), SPA duration (B) and SPA intensity (C) in three consevutive months in Selected (S) and Randomly Bred (RB) line types of mice fed a HCarb, HFat and Cotnrol diet.

AC

CE

PT

ED

M

AN

US

CR

IP

T

Symbols indicate adjusted means for each line type (± s.e.m.).

25

ACCEPTED MANUSCRIPT Figure 3. Total (a), HDL (b), LDL (c) cholesterol and triglyceride (d) levels in Selected (black symbols) and Random Bred (open symbols) mice fed the HFat, HCarb and Control diet. Symbols

AC

CE

PT

ED

M

AN

US

CR

IP

T

represent adjusted means for each replicate of the respective line type.

26

ACCEPTED MANUSCRIPT Figure 4. Fasting blood glucose in the Selected (black symbols) and Random Bred (open symbols) mice fed the HFat, HCarb and Control diet. Symbols represent adjusted means for each

AC

CE

PT

ED

M

AN

US

CR

IP

T

replicate of the respective line type.

27

AC

CE

PT

ED

M

AN

US

CR

IP

T

ACCEPTED MANUSCRIPT

28

AC

CE

PT

ED

M

AN

US

CR

IP

T

ACCEPTED MANUSCRIPT

29

AC

CE

PT

ED

M

AN

US

CR

IP

T

ACCEPTED MANUSCRIPT

30

AC

CE

PT

ED

M

AN

US

CR

IP

T

ACCEPTED MANUSCRIPT

31

ACCEPTED MANUSCRIPT Selection for high aerobic capacity has no protective effect against obesity in laboratory mice Julita Sadowska. Andrzej K. Gębczyński, Marek Konarzewski

T

Highlights The role of intrinsic role of Vo2max in obesity resistance is studied.

-

Genetically determined high Vo2max does not protect against diet-induced obesity.

-

High Vo2max is not correlated with SPA levels.

AC

CE

PT

ED

M

AN

US

CR

IP

-

32