The use of mouse models to unravel genetic architecture of physical activity: A review

Department of Neuroscience and Pharmacology, Brain Center Rudolf Magnus, UMC Utrecht, Utrecht, the Netherlands.
Genes Brain and Behavior (Impact Factor: 3.66). 10/2013; 13(1). DOI: 10.1111/gbb.12091
Source: PubMed
The discovery of genetic variants that underlie a complex phenotype is challenging. One possible approach to facilitate this endeavor is to identify quantitative trait loci (QTL) that contribute to the phenotype and consequently unravel the candidate genes within these loci. Each proposed candidate locus contains multiple genes and, therefore, further analysis is required to choose plausible candidate genes. One of such methods is to use comparative genomics in order to narrow down the QTL to a region containing only few genes. We illustrate this strategy by applying it to genetic findings regarding physical activity (PA) in mice and human. Here, we show that PA is a complex phenotype with a strong biological basis and complex genetic architecture. Furthermore, we provide considerations for the translatability of this phenotype between species. Finally, we review studies which point to candidate genetic regions for PA in humans (genetic association and linkage studies) or use mouse models of PA (QTL studies) and we identify candidate genetic regions that overlap between species. Based on a large variety of studies in mice and human, statistical analysis reveals that the number of overlapping regions is not higher than expected on a chance level. We conclude that the discovery of new candidate genes for complex phenotypes, such as PA levels, is hampered by various factors, including genetic background differences, phenotype definition and a wide variety of methodological differences between studies.


Available from: Martien J Kas
Genes, Brain and Behavior (2013)
doi: 10.1111/gbb.12091
The use of mouse models to unravel genetic
architecture of physical activity: a review
E. Kostrzewa and M. J. Kas
Department of Translational Neuroscience, Brain Center Rudolf
Magnus, University Medical Center Utrecht, Utrecht, the
*Corresponding author: M. J. Kas, PhD, Department of
Translational Neuroscience, Brain Center Rudolf Magnus,
University Medical Center Utrecht, Universiteitsweg 100, 3584
CG Utrecht, the Netherlands. E-mail:
The discovery of genetic variants that underlie a complex
phenotype is challenging. One possible approach to
facilitate this endeavor is to identify quantitative trait loci
(QTL) that contribute to the phenotype and consequently
unravel the candidate genes within these loci. Each
proposed candidate locus contains multiple genes and,
therefore, further analysis is required to choose plausible
candidate genes. One of such methods is to use
comparative genomics in order to narrow down the QTL
to a region containing only a few genes. We illustrate
this strategy by applying it to genetic findings regarding
physical activity (PA) in mice and human. Here, we show
that PA is a complex phenotype with a strong biological
basis and complex genetic architecture. Furthermore,
we provide considerations for the translatability of this
phenotype between species. Finally, we review studies
which point to candidate genetic regions for PA in
humans (genetic association and linkage studies) or
use mouse models of PA (QTL studies) and we identify
candidate genetic regions that overlap between species.
On the basis of a large variety of studies in mice and
human, statistical analysis reveals that the number of
overlapping regions is not higher than expected on a
chance level. We conclude that the discovery of n ew
candidate genes for complex phenotypes, such as PA
levels, is hampered by various factors, including genetic
background differences, phenotype definition and a
wide variety of methodological differences between
Keywords: Complex phenotypes, genetics, home cage
activity, mouse models, open field test, physical activity,
QTL, running wheel activity, spontaneous physical activity,
voluntary exercise
Received 5 July 2013 , revised 15 August 2013 and 16
September 2013, accepted for publication 1 October 2013
The genetic basis of complex phenotypes is difficult to
unravel due to the high number of genetic and environmental
factors that shape these phenotypes across different
developmental stages. Multiple genetic loci influence
complex phenotypes and the influence of each one of these
is further modulated by genetic background, gender, age
and environment. Furthermore, most of these genetic loci
(called quantitative trait locus or QTL) have small (Flint and
Mott 2008) and pleiotropic effects on multiple (complex)
phenotypes (Flint 2003; Garland
et al.
2011b). One may
think of genetic loci contributing to the occurrence of a
complex phenotype more as ‘modifiers’ than causal factors
as each of them has only a small effect size (Botstein and
Risch 2003). Animal models are often used to facilitate the
process of discovering novel QTLs and candidate genes
because they enable the study of complex phenotypes in a
controlled environmental and genetic background conditions
et al.
2012). Multiple QTLs are discovered in
this way, all pointing with some confidence interval to a
possible candidate genetic region which contains one or
more candidate genes for the phenotype. However, choosing
a specific candidate gene remains difficult (DiPetrillo
et al.
2005). To narrow down a candidate gene region to the level
of a few possible candidate genes one may use various
methods, including comparative genomics. This method
makes use of interspecies genetic homology between QTLs
coming from various species (DiPetrillo
et al.
2005). In this
review, we will consider this method of discovering possible
candidate genetic regions by using an example of human and
mouse studies regarding the genetic basis of physical activity
(PA). PA is broadly defined as ‘bodily movement produced
by contraction of skeletal muscles’ (Dishman 2008)
With the increasing concerns over effect of low PA on
health of individuals (Casazza
et al.
2013; Haskell
et al.
et al.
2004; Mokdad
et al.
2004; Myers
et al.
and observations that many people in Western societies
do not engage in sufficient amount of exercise (Kelly and
Pomp 2013; Marcus and Forsyth 1999) numerous studies
addressed the question of the genetic basis of PA. Hopefully,
better prevention programs may be created if we increase the
knowledge of factors contributing to the appropriate amount
of PA. Undoubtedly, an array of psychological, cultural and
other environmental factors has an influence on the levels
of PA. However, there are also premises which suggest that
habitual PA is strongly regulated by evolutionary conserved
genetic factors (Rowland 1998). One of the possible ways
to unravel the biological pathways involved in the regulation
of PA is to investigate the genetic basis of this heritable
2013 John Wiley & Sons Ltd and International Behavioural and Neural Genetics Society
Page 1
Kostrzewa and Kas
phenotype (Bray
et al.
2009; de Vilhena e Santos
et al.
Kelly and Pomp 2013; Stubbe
et al.
As with other complex phenotypes, the expression of PA
levels is under the influence of multiple genetic loci, modified
by genetic background, gender, age and environmental fac-
tors (as will be addressed further in this article). Taking under
consideration this complex network of factors i nfluencing PA,
one may wonder whether it is possible to use animal models
to unravel the genetic basis of this complex trait in humans
et al.
2011b). In the current review, we discuss
which animal models may be applied in an attempt to model
complex genetic architecture of human PA. We consider the
possibility of using already available results of mouse and
human studies to point to plausible candidate genes for this
complex phenotype. In order to address this question, we
review studies which pointed to candidate gene regions for
PA in mice (QTL studies) and humans (genetic association
and linkage studies) and assess the degree of overlap
between candidate genetic regions. Finally, we assess
whether the potential overlap is greater than expected
on the basis of random sampling from a limited pool of
genetic regions. In case of statistically reliable overlap, it
would be possible to narrow down the candidate gene
regions and possibly identify candidate genes influencing
PA levels.
Argument in favor of biological basis of PA
An important argument supporting biological component
in regulation of PA is an essential role of PA in energy
homeostasis (Rowland 1998). According to the energy
balance equation, all the energy taken in by an organism
must be expended or transformed by this organism. Thus,
the total energy intake will be counterbalanced by the
energy expended, needed for reproduction, growth, stored
as fat or glycogen and dispersed due to fecal and urinary
loss. As far as the first component of the equation is
concerned, the energy is expended on basal metabolic
rate, digestion and processing of food, thermoregulation
and PA (Garland
et al.
2011b). It is of note that PA may
constitute a substantial portion of daily energy expenditure
of an organism depending on the species, environment and
age (Rezende
et al.
2009).This function of PA in maintaining
energy homeostasis caused researchers to speculate that
there must be a mechanism controlling this form of energy
expenditure, so called ‘activity-stat’ (Rowland 1998). Various
arguments support this notion, including heritability of
PA, age-dependent decrease of activity (Sallis 2000), sex
differences in activity levels, biorhythmicity (Rowland 1998;
et al.
1973) of PA and finally compensatory increases or
decreases of PA when other elements of the energy balance
equation are changed (Epstein and Wing 1980; Goran and
Poehlman 1992). Some of these arguments are discussed
further on in the current review.
Out of all the components of the energy balance equation,
we concentrate in this article solely on PA, which can
be further subdivided into spontaneous physical activity
(SPA) and voluntary exercise (VE). The VE is defined as
‘locomotor activity that is not directly required for survival
or homeostasis and not directly motivated by any external
factor’ (Garland
et al.
2011b) which could be understood as
sports and fitness-related activities. All remaining PA such as
activities of daily life, non-specific ambulatory behavior and
maintaining posture may be considered SPA (Garland
et al.
2011b; Levine
et al.
1999). It is not always easy to classify
a particular type of PA into one of the two categories;
however, this division enables some conceptualization
and research on effects of PA and VE, in particular
on health.
Measurement of PA in humans
Voluntary exercise
Various methods of assessment may be used to study
PA levels in humans e.g. direct observation, questionnaires
(diaries and retrospective recall), surveys, calorimetry, heart
rate monitors and motion sensors (Garland
et al.
Westerterp 2009). Choice between the methods involves
a trade-off between feasibility and reliability. As a result of
practical difficulties related to the use of objective methods,
many studies involve the self-reported PA assessment
(Dishman 2008) which may have limited reliability and validity
(Shephard 2003). Attempts to validate the self-reported
measurement methods show that the correlations between
objective and subjective measurement methods are rather
low (de Vilhena e Santos
et al.
2012; Dishman 2008; Garland
et al.
2011b). Furthermore, various PA assessment methods
lead to discrepant results regarding genetic associations and
heritability estimates (Butte
et al.
2006; Cai
et al.
2006; Choh
et al.
2009; de Vilhena e Santos
et al.
2012; Seabra
et al.
et al.
2003a). These observations suggest that
various PA assessment methods measure, in fact, partially
different phenotypes or theoretical constructs (de Vilhena e
et al.
2012). A possible solution to this problem, would
be to involve both objective and subjective PA assessment
methods in one study. It would enable comparison of the
methods within one population and give detailed assessment
of different features of PA such as duration, type, context,
motivation and energy expenditure (Dishman 2008).
Spontaneous physical activity
In humans, there is a partial overlap between methods that
can be used to assess VE and SPA. Generally, pedometers,
accelerometer and observation (direct or video recording)
may be used to estimate the levels of SPA (Garland
et al.
Animal models of PA
Animal models are often used to overcome some of the
limitations hampering human studies. They enable standard-
ization of the complex conditions influencing the levels of PA
and methods of PA measurement. Therefore, it is hypothe-
sized that they may be effectively used to assess biological
Genes, Brain and Behavior
Page 2
Unravel genetic architecture of physical activity
factors influencing PA levels. The VE and SPA in humans may
be best modeled by distinct animal models (Eikelboom 1999;
et al.
2011b; Kelly
et al.
2010; Rezende
et al.
2009). In
general, (voluntary) running wheel activity (RWA) is proposed
as a model of VE (Kelly and Pomp 2013; Rezende
et al.
while spontaneous locomotor activity, especially in the home
cage (HC) environment, may be considered a model of SPA
et al.
2011b). The expressions of the two forms of
energy expenditure (RWA and HC activity) are not indepen-
dent. When animals are given access to a running wheel they
reduce their HC activity (Koteja
et al.
1999) while their total
time spent on activity increases (de Visser
et al.
2005; Kas and
Edgar 1998). Furthermore, mice bred for high RWA show also
a reduction in the total HC activity ( Careau
et al.
2012; Malisch
et al.
2009) and the RWA and HC activity weakly correlate
in the outbred strain used in these selection experiments
et al.
It is of note that open field test (OFT) is not an appropriate
model of human PA because it was initially developed to
test anxiety levels in rodents. As a consequence, many
of the QTLs for locomotor activity in OFT are associated
with the motivation to avoid highly anxiogenic novel and
brightly illuminated environments. Indeed, Gershenfeld
et al.
(1997) have shown that QTLs for locomotor activity in
habituated and nonhabituated OFT have only partial overlap.
Furthermore, by using a complex battery of tests, which
included various anxiety tests and a HC, researchers from
the Flint group managed to show that out of multiple QTLs
that influence mouse activity levels in ethological tests, only
QTLs on chromosome 4 and 8 are locomotor activity specific
et al.
2004; Turri
et al.
2004). For these reasons
we do not include OFT data in the current review However,
interested reader may find a list of 15 mouse QTL studies
et al.
2008; Buck
et al.
2000; Crabbe
et al.
et al.
2010; Flint
et al.
1995; Gershenfeld
et al.
1997; Gill and Boyle 2005; Henderson
et al.
2004; Hitze-
et al.
2000, 2003; Kelly
et al.
2003; Koyner
et al.
et al.
1995; Plomin
et al.
1991; Radcliffe
et al.
et al.
1999; Turri
et al.
1999, 2004; Zhang
et al.
of the OFT behavior in Table S1, Supporting Information.
Voluntary exercise
If VE is defined as locomotor activity ‘that is not directly
required for survival or homeostasis and not directly
motivated by any external factor’ (Garland
et al.
then voluntary RWA may be considered the most suitable
rodent behavior to model human VE (Eikelboom 1999;
et al.
2011b; Kelly
et al.
2010; Rezende
et al.
2009). There are several arguments to support this notion.
First, RWA in rodents, similarly to human VE, is performed
without any obvious immediate goals (Eikelboom, 1999).
It may be considered a classic example of self-rewarding
behavior (Brene
et al.
2007; Garland
et al.
2011b; Kagan and
Berkum 1954; Novak
et al.
2012; Sherwin 1998). Indeed,
rodents show conditioned place preference to the location
associated with wheel running (Lett
et al.
2000) and are
willing to perform an instrumental reaction in order to obtain
access to the running wheel (Belke and Garland, 2007; Kagan
and Berkum 1954). Secondly, age-related decline in PA is
observed in both, humans and rodents (Bronikowski
et al.
2006; Ingram 1983; Morgan
et al.
2003; Sallis 2000; Vaanholt
et al.
2010) (see also further in this review). Thirdly, both VE
and RWA may in some situations involve changes in brain
reward systems that are comparable to the ones present in
addictive behavior (Belke and Garland 2007; Garland
et al.
2011b). Finally, voluntary wheel running and VE occur in
environments which per se require low energy expenditure
(laboratory housing in case of mice or modern industrialized
Western culture) (Booth
et al.
2002; Eikelboom 1999;
et al.
It is of note that there are also some arguments against the
validity of RWA as a model of VE. First, gender differences
in PA are observed in both humans and mice, however,
in an opposite direction suggesting important differences
in biological determinants of the behavior (see further in
this review). Second, some researchers believe that wheel
running is merely an artifact resulting from poor laboratory
housing conditions (Sherwin 1998). Third, it was estimated
that laboratory mice went through selection for more than
400 generations which resulted in diminished PA levels.
In contrast, humans from Western cultures are said to go
through similar environmental pressure for no more than 25
generations (Booth
et al.
2002; Garland
et al.
2011b). As a
consequence, it is possible that the natural state for the lab-
oratory mouse is to maintain low PA levels while for human
it is to continue to exercise (Booth
et al.
2002; Garland
et al.
2011b). This raises questions regarding the validity of using
mouse models as a model of normal human behavior and
physiology (Garland
et al.
2011b). Finally, some researchers
point to the ‘behavioral common path’ as an argument
against possibility to use RWA as a model of VE (Sherwin
1998). According to this notion, various internal and external
factors lead to the occurrence of a given animal behavior
because the repertoire of behaviors is limited. Therefore,
it is not possible to deduct underlying mechanisms and
motivations just on the basis of the observed phenotype
(Sherwin 1998).
In summary, substantial arguments support validity of
RWA as a model of VE, however, caution needs to
be taken as far as the translational value of the model
is concerned due to some arguments against its use.
It is possible that although VE and RWA are artifacts
occurring in specific environments, they may result from
evolutionary conserved ability and motivation to stay active
in order to obtain food (despite pain, effort and potential
Spontaneous physical activity
Most of the energy expenditure of a laboratory animal
may be considered an analogue to human SPA e.g.
locomotor activity, rearing and grooming (Garland
et al.
2011b). In an absence of running wheel and mating partners,
any activity not associated with food intake could be
considered SPA. SPA in rodents may be measured using
various methodologies, for example, video analysis, force
plates or infrared beams (Garland
et al.
2011b) which
can be used to detect movement in a novel or familiar
environment. As mentioned above, the behavior of an
Genes, Brain and Behavior
(2013) 3
Page 3
Kostrzewa and Kas
animal in the novel environment (such as in the OFT) is
influenced by novelty-induced anxiety. Furthermore, the
readout often includes only locomotor activity, although
SPA includes also other behaviors such as grooming.
Therefore, the most adequate tool to measure SPA in
mice is HC environment and preferably includes locomotor
activity and total motility of an animal. HC phenotypes can
be assessed under baseline conditions, without human
interference and across the circadian cycle (Kas and
Van Ree 2004).
Heritability of PA in humans and mice
Genetic contribution to physical exercise was partly
supported by human family studies which showed that broad-
sense heritability ranges from 0 to 60% depending on the
population tested, the exact phenotype measured and the
method of measurement. The highest heritability estimates
come from studies using objective measurement of PA such
as accelerometer instead of subjective measurements such
as questionnaires (de Vilhena e Santos
et al.
2012). This may
point to the fact that PA has a strong genetic basis which is
more adequately estimated when more precise phenotyping
methods are used. Further support of heritability of PA comes
from various twin studies which estimated the broad-sense
heritability of PA in adults to fall between 16 and 80%
(Beunen and Thomis 1999; De Moor
et al.
2011; Eriksson
et al.
2006; Kujala
et al.
2002; Lauderdale
et al.
1997; Simonen
et al.
2002; Stubbe
et al.
2006) and in case of children and
adolescent to stay in a range from 0% (Stubbe
et al.
to 92% (Wood
et al.
2008) (for an extensive review see (de
Vilhena e Santos
et al.
Finally, the heritability of PA, measured as RWA and HC
activity, was estimated for various mouse populations.
Heritability estimates for RWA in 13 mouse inbred strains
varied between 12 to 66 percent (Lightfoot
et al.
2008). This was further supported in a study using a panel
of 41 inbred strains which showed similar (but higher) broad-
sense heritability for distance (55%), duration (57%) and
speed (60%) (Lightfoot
et al.
2010). Furthermore, narrow-
sense heritability of mean revolutions run per day on a
RW in an outbred strain (Hsd:ICR) estimated by midparent-
offspring regression was 18% (Swallow
et al.
1998). Finally,
the realized heritability (adjusted for within-family selection)
was estimated to be on average 28% for lines selected for
high RWA (Swallow
et al.
Estimation of heritability for a total PA in HC delivered
divergent results ranging from low value of 14% (Toth and
Williams 1999) to a relatively high value of 62% (Umemori
et al.
2009). Interestingly, it seems that only activity in
the dark, thus the active period of the day, has strong
genetic component (with estimated heritability of 18%) while
heritability of PA during the light phase is neglectable (5%)
(Toth and Williams 1999).
Age effects on heritability and secular trends
for PA
Despite the strong genetic component, PA is not stable
over the life-span. Multiple studies have shown age
effects on PA participation and intensity (Sallis 2000). The
reasons for this phenomenon are not well understood,
however, this finding is very consistent in various human
populations examined (Sallis 2000). In general, childhood
is a life period with highest PA levels, followed by a
strong decline during adolescence and early adulthood,
relative stability of PA in middle aged people and further
decline in elderly (Anderssen
et al.
1996; Bijnen
et al.
et al.
2006; Sallis 2000; Schoenborn 1986; Telama
et al.
As a consequence, the strength of influence of the genes
on levels of PA is modulated by age. Stubbe
et al.
showed, using a sample of more than 1600 twin pairs,
that the heritability of sport participation increased from
0% in adolescents to 85% in young adults. In the group
with low genetic contribution to PA, the most profound
was the effect of shared environmental factors (7884%)
et al.
2005). However, this observation was not
supported by a follow-up study on a Finish cohort in
which the genetic influence on PA seemed to decrease
with age (Aaltonen
et al.
2010). It is possible that some
genetic or environmental factors are responsible for this
Similar age-dependent decline in PA was reported multiple
times for various species (Ingram 2000; Sallis 2000) including
mice for both RWA (Bronikowski
et al.
2006; Ingram 1983;
et al.
2003; Vaanholt
et al.
2010) and HC activity
(Goodrick 1975; Ingram 1983; Ingram and Reynolds 1986). It
is of note that genetic background influences the effect of
age on PA. Mouse inbred strains differ in the temporal pattern
and extent of age-dependent decline for RWA (Lhotellier and
Cohen-Salmon 1989).
Influence of sex on heritability and secular
trends for PA
Age is not the only modulator of genetic influence on PA
levels. Regardless of the age group examined, there is a
strong influence of gender on the genetic architecture of PA
(Beunen and Thomis 1999; Boomsma
et al.
1989; De Moor
et al.
2007a; Maia
et al.
2002). I t was repeatedly shown
that genetic factors are more important in determining the
levels of PA in boys than in girls (Beunen and Thomis 1999;
et al.
1989; De Moor
et al.
2007a; Maia
et al.
2002). For example, in a Portuguese twin population, in case
of boys, genetic factors had a major influence on participation
in sports while for girls genetic, common and unique
environmental factors contributed evenly to the phenotype
Genes, Brain and Behavior
Page 4
Unravel genetic architecture of physical activity
et al.
2002). This points out to the fact that shared
environment (family and peers) and unique environmental
factors may have much stronger influence on exercise
behavior in case of females than in males. The observed
differences in genetic and environmental determinants
are reflected on the level of observed PA phenotype.
There are clear gender differences in the amount of sport
participation with males being more active than girls at all age
groups (Schoenborn 1986). Furthermore, the age-dependent
decrease in PA is more pronounced in women than in men
(Sallis 2000).
In case of mice, the estimations of heritability of PA also differ
between sexes. Analogous to observations from human
studies, male mice are characterized by stronger influence of
genetic factors for the measured aspects of RWA. Studies
on a panel of 13 inbred strains showed that the estimates of
heritability are as follows: for distance moved, male 3148%
and female 12 22%; for duration, male 4461% and female
1221%; for velocity, male 4966% and female 4461%
et al.
2004). Furthermore, in some studies, sex-
specific QTLs for RWA were found (Leamy
et al.
et al.
The sex effects regarding heritability of PA are reflected on
the level of observed phenotype (Garland
et al.
2011a; Leamy
et al.
2010). Contrary to observations in humans, however,
female mice tend to be more active than male mice at least if
the activity is measured in the running wheel (Lightfoot
et al.
2004, 2010). Various sex differences have been observed
in multiple studies, for example, using a panel of 13 inbred
strains Lightfoot
et al.
(2004) showed that females of F
population between C57L/J (high active) and C3H/HeJ (low
active) strains run 49 further, 9% faster and 39% longer
during a day measurement. These results were replicated
in another study using 41 inbred strains (Lightfoot
et al.
2010). When data of all the strains were pooled together,
there was significant effect of strain, sex and the interaction
between strain and sex on the parameters of RWA (distance,
speed and duration); namely, female mice ran 24% further
(24%), 11%faster and 13% longer (13%) than male mice
et al.
2010). However, after correction for multiple
testing, the differences between genders were significant in
case of only 7 strains for distance, 5 strains for speed and 5
strains for duration (Lightfoot
et al.
2010). Finally, Careau
et al.
(2012) could show in an experiment involving lines selectively
bred for high RWA, that female mice from both control and
high running lines run longer and further than male mice.
Interestingly, in these mouse lines selectively bred for high
RWA, both male and female mice increased the RWA levels;
however, the increase in daily energy expenditure was much
higher for male mice (Rezende
et al.
2009) because females
reached higher PA by running faster while males run longer
et al.
Possible mechanisms that may cause the phenotypic and
genetic sex differences in PA in mice may involve the
endocannabinoid system and sex hormones (Keeney
et al.
2008; Lightfoot 2008) as it is known that estrogens strongly
modulate PA (Morgan
et al.
Genetic basics of PA and physical inactivity
Several facts need to be considered before we can proceed
with summing up the findings regarding genetic basis of PA.
First, it is of note that two components influence the levels of
PA, namely the physical ability and motivation to perform PA
(Dishman 2008; Garland
et al.
2011b; Kelly and Pomp 2013).
Furthermore, it is likely that ability depends on physiological
factors such as respiratory endurance (Rankinen
et al.
while motivation is regulated by central nervous system
functioning (Rhodes
et al.
2003). Therefore, these two
components may have specific genetic basis (Garland
et al.
2011b) and work in synergy to produce the final observed
PA level (Kelly and Pomp 2013).
Second, genetic architecture of PA is undoubtedly
complex, as of any other complex phenotype. Numerous
genetic loci with mostly minor effect size contribute to
PA levels (Dishman 2008). The effect sizes of these
loci are most likely influenced by individual- and species-
dependent genetic background. Furthermore, many of the
loci influencing PA are placed in the regulatory elements,
outside of the protein coding genes, which may make it
difficult to translate the findings between species. Finally,
many of the loci may have pleiotropic effects (e.g. (Garland
et al.
2002; Hanson and Hakimi 2008)). Their influence on
a given phenotype may be also mediated by their effect
on other related phenotypes; for example, there is a 40%
overlap between genes which were shown in humans to
influence PA and obesity, two phenotypes that reciprocally
influence their levels (Good
et al.
Third, various authors pointed out that PA and physical
inactivity may be considered distinct phenotypes and not two
poles of one continuum of activity (Bray
et al.
2009; de Vilhena
et al.
2012; Perusse
et al.
1989). This statement may
find some confirmation in a meta-analysis by Viggiano who
has shown that distinct genetic manipulations or brain lesions
lead to either high or low novelty-induced locomotor activity
(Viggiano 2008). Furthermore, it was suggested that the brain
is primarily prone to hyperactivity, and this hyperactive tone
is down-regulated by brain regions and neurotransmitter
systems that decrease the PA (Rowland 1998; Viggiano
2008). It was proposed that the reticular activating system
is responsible for arousal while cerebral cortex is mostly
inhibitory (Rowland 1998). All in all, the interplay of the
activating and inhibiting systems serves the purpose of
maintaining so called ‘sensoristasis’ (term created by Schultz
(1965)) which is an optimal level of sensory stimulation of
the nervous system for each individual (Rowland 1998). An
individual may regulate the level of stimulation by regulating
PA levels but also by engaging in intellectually stimulating
tasks that may require marginal levels of PA. Therefore,
PA and physical inactivity may be two distinct behaviors
which are used to maintain the balance in terms of sensory
stimulation and energy equation. Each of them may have a
distinct biological (including genetic) basis and a complex set
of possible motivations to engage in them.
Finally, the current review includes only the significant
genetic association with duration or intensity of PA. Reviews
on genetic polymorphisms associated with performance
Genes, Brain and Behavior
(2013) 5
Page 5
Kostrzewa and Kas
phenotypes are reviewed elsewhere and are beyond the
scope of the current review (Bray
et al.
2009; Hagberg
et al.
2011; Rankinen
et al.
2010; Roth
et al.
Human studies
de Vilhena e Santos
et al.
(2012), in an excellent review,
gathered studies which assessed the genetic basis of PA
and physical inactivity in humans. The overlap between
genetic regions found in these studies is rather small (de
Vilhena e Santos
et al.
2012). This may be explained by the
following two reasons. First, cited studies used a variety
of measurement methods and were conducted in various
populations and age groups (de Vilhena e Santos
et al.
Second, the number of studies conducted up to date for both
PA and physical inactivity is still very low (especially if one
takes under consideration that PA and physical inactivity are
polygenic traits). A s far as PA is concerned, there are only 16
studies examining its genetic basis: four linkage studies (Cai
et al.
2006; De Moor
et al.
2007a,b; Simonen
et al.
eleven association studies (Berentzen
et al.
2008; Cole
et al.
2010; De Moor
et al.
2009; Fuentes
et al.
2002; Hakanen
et al.
2009; Liu
et al.
2010; Loos
et al.
2005; Lorentzon
et al.
2001; Richert
et al.
2007; Simonen
et al.
2003a; Stefan
et al.
2002; Winnicki
et al.
2004) and one GWAS study (De
et al.
2009). Only a small proportion of the above
mentioned studies assessed the genetic basis of physical
inactivity: two linkage studies (Cai
et al.
2006; Simonen
et al.
2003b) and two association studies (Loos
et al.
et al.
Nevertheless, significant genetic associations were found
for both phenotypes. The significant genetic association for
PA was found in nine of the studies mentioned above, which
identified fifteen candidate genes or genetic regions that may
be involved in the regulation of intensity or duration of PA in
humans (de Vilhena e Santos
et al.
2012). A short summary of
these significant findings is shown in Table 1. Interestingly,
melanocortin 4 receptor (MC4R) gene was repeatedly asso-
ciated with levels of PA, when PA was measured with the
B3DPAR questionnaire (Loos
et al.
2005) or accelerometer
et al.
2010). Furthermore, Cai
et al.
(2006) have shown
a significant association of the genetic region containing
MC4R with the PA measured by accelerometer. As far as
physical inactivity is concerned, all four cited studies showed
a significant association, identifying six candidate genes or
genetic regions involved in regulation of physical inactivity
(Table 2). Interestingly, MC4R gene was associated with
physical inactivity measured with the B3DPAR questionnaire
et al.
2005). Furthermore, the genetic region containing
MC4R gene showed a significant association with sedentary
activity measured by accelerometer (Cai
et al.
2006). The
overlap of the genetic basis of PA and physical inactivity
points to the fact that these two phenotypes partly share
the genetic and physiological basis, and are not completely
unrelated as it might have been understood on the basis
of the previous statements (Bray
et al.
2009; de Vilhena e
et al.
2012; Perusse
et al.
Mouse QTL studies
HC studies
Table 3 summarizes the findings from six different studies
reporting mouse QTLs associated with HC activity. Each
study was conducted using a different mouse strain, different
Table 1: Genetic regions associated in humans with physical activity (PA)
Author Country Assessment method Phenotype Locus Gene/Marker
Linkage studies
et al.
2006) USA Accelerometer TPA, MPA 18q21.3218q21.33 D18S64D18S68
LPA 18q12.218q21.2 D18S1102D18S474
(De Moor
et al.
2007a) The Netherlands Questionnaire Dichotomous: ‘Do you
participate in exercise
19p13.3 D19S247
et al.
2003b) USA Questionnaire TPA 13q22q31 D13S317
et al.
1983) 7p11.2 IGFBP1
9q31.1 D9S938
13q22q31 D13S317
TSPA 11p15.2 C11P15_3
15q13.3 D15S165
Association studies
et al.
2010) USA Accelerometer TPA, MPA, VPA 18q21.32 MC4R
et al.
2005) Canada B3DPAR TPA, MVPA 18q21.32 MC4R
et al.
2001) USA Questionnaire Past year physical activity 3q21.1 CASR
et al.
2002) USA Respiratory chamber 24 h energy expenditure 1p31.3 LEPR
et al.
2004) Italy Questionnaire Type and frequency of
17q23.3 ACE
(De Moor
et al.
2009) The Netherlands Questionnaire Leisure time exercise
2q33.1 DNAPTP6
and USA (type, frequency, duration) 10q23.2 PAPSS2
18p11.32 C18orf2
LPA, low PA; MPA, moderate PA; MVPA, moderate to vigorous PA; TSPA, time spent on PA; TPA, total physical activity; VPA, vigorous PA.
Genes, Brain and Behavior
Page 6
Unravel genetic architecture of physical activity
Table 2: Genetic regions associated in humans with PI
Author Country Assessment method Phenotype Locus Gene/marker
Linkage studies
et al.
2006) USA Accelerometer SA 18q12.2 18q21.2 D18S1102 D18S474
et al.
2003b) USA Questionnaire PI 2p22 p16 D2S2347
B3DPAR 2p22 p16 D2S2347
et al.
1983) 7p11.2 IGFBP1
20q12 PLC1
Association studies
et al.
2005) Canada B3DPAR PI 18q21.32 MC4R
et al.
2004) Italy Questionnaire SA 17q23.3 ACE
PI, physical inactivity; SA, sedentary activity.
breeding methods and different QTL estimation method.
Because of methodological limitations of using recombinant-
inbred strains (Belknap 1992; Flint and Mott 2001), in one of
the studies only provisional QTLs could be detected (Toth and
Williams 1999). There were other substantial methodological
differences in the way the HC experiment was conducted, for
example, the age of mice used differed between the studies
ranging from 50 days in one study (Mayeda and Hofstetter
1999) to 4 months in another study (Kas
et al.
2009). In
most cases, animals were kept in 12 h/12 h light/dark cycle
but in one of the studies mice were housed in constant
darkness (Mayeda and Hofstetter 1999). Furthermore, the
time window for activity measurement ranged from the first
2 h of the dark phase (Henderson
et al.
2004) to 10 days
of continuous data collection (Mayeda and Hofstetter 1999).
In some cases, animals were habituated to the HC (Furuse
et al.
2002; Toth and Williams 1999; Umemori
et al.
while in others the activity in the HC was measured from the
first day (Kas
et al.
2009; Mayeda and Hofstetter 1999).
Finally, various methods of activity measurement were
used between studies: intraperitoneal transmitters (Toth and
Williams 1999), video tracking (Kas
et al.
2009) and infrared
sensors (Furuse
et al.
2002; Henderson
et al.
2004; Mayeda
and Hofstetter 1999; Turri
et al.
2004; Umemori
et al.
Home cage activity is a complex phenotype in which
different components may be separated. With a use of
principle component analysis, de Visser
et al.
(2006) could
differentiate between two factors within total PA in HC:
temporal and velocity components. It is very likely that
different genomic regions regulate temporal pattern and
intensity components of PA in a HC (Umemori
et al.
Furthermore, temporal patterns of activity may be under
control of central nervous system mechanisms regulating
circadian rhythm while the intensity of activity may depend
on neuromuscular properties (Umemori
et al.
2009). This
hypothesis gained support from genetic studies examining
various aspects of locomotor activity in a HC. Umemori
et al.
(2009) showed that the temporal and intensity components
are only weakly correlated. Furthermore, QTLs contribute to
these aspects of HC activity overlap only partially. This partial
overlap of the QTLs suggests that temporal and intensity
components have partly shared and partly divergent genetic
Voluntary RWA studies
Table 4 presents the major findings from eight different
studies reporting mouse QTLs associated with voluntary
RWA. One of the studies included in this summary examined
RWA only in constant darkness conditions (Shimomura
et al.
2001). This difference in light dark cycle assessment
between the studies could be a confounding factor;
however, another study by Suzuki
et al.
(2000) showed a
strong correlation between RWA in the normal lightdark
cycle and in the constant darkness. Furthermore, these
researchers showed complete overlap of QTLs regulating
RWA in these two lighting conditions (Suzuki
et al.
The two studies mentioned here were conducted in
mice with different genetic background, age and gender;
thus one cannot make direct parallels between the two
studies. However, the experiment by Suzuki
et al.
shows that we cannot exclude findings reported by
et al.
(2001) only because of the altered lighting
The influence of strain, gender, age of mice and
exact experimental conditions on the detection of QTLs
cannot be stressed enough. The experimental setup of
the RWA experiments seems much more comparable
between studies than in the case of HC experiments.
However, important consequences may arise from the
present methodological differences such as age and gender
of animals, lighting conditions, size of the running wheel and
the duration of the data collection. In five of the studies,
mice of both genders were used while in three studies only
male mice were examined (Suzuki
et al.
2000; Yang
et al.
2009, 2012). This may have a substantial effect on the QTLs
detected as sex-specific QTLs for RWA were identified in, at
least, one study (Lightfoot
et al.
When given access to running wheels, mice require
few days to adapt before they exhibit stable RWA levels.
Interestingly, this behavioral adaptation is also reflected in a
complex architecture of genetic loci influencing the behavior.
et al.
(2010) showed that some of the QTLs are
associated consistently across days with RWA while multiple
QTLs have an effect on specific days only. Furthermore,
et al.
(2010) found another profound mediating effect
of time on the underlying genetic architecture of RWA. They
conducted a RWA experiment for 21 days and averaged the
behavioral observations into seven intervals lasting 3 days
Genes, Brain and Behavior
(2013) 7
Page 7
Kostrzewa and Kas
Table 3: Mouse QTL studies using a home cage (HC) activity model
Article Phenotype Mouse strain Gender QTL Mouse genomic location
Mouse genomic
location in bp
human region
et al.
(2002) Spontaneous HC activity Backcross between
KJR and BLG2
females Chr3 Between D3Mit86 and
Chr3:147 002 779
150 756 405
Chr1:80 283 988
84 313 096
Backcross between
NJL and BLG2
females no locus
et al.
et al.
(2001) and
et al.
Locomotor activity in
various behavioral tests
of anxiety and HC
intercross between
DeFries lines H1,
H2, L1 and L2
Male and
Chr1 76 cM
Chr4 36 cM
Chr7 32 cM
Chr8 60 cM
Chr15 22 cM
Chr18 28 cM
ChrX 26 cM
Activity in the HC during
first 2 h of the dark
intercross between
DeFries lines H1,
H2, L1 and L2
Male and
Chr4 3141 cM
Chr8 5668 cM
et al.
(2009) HC activity A/JxC57BL/6J CSS
Male and
Chr1 Region containing
Chr1:80 198 699
80 213 944
Chr2:225 243 406
225 266 790
Mayeda and Hofstetter
mean amount of daily
locomotor activity at HC
in continuous darkness
congenic strain
? Chr1 89106 cM (approximate
markers: rs107876071,
chr1:170 940 982
186 261 829
chr1:219 115 791
227 644 727
chr1:240 253 165
247 125 743
chr1:158 516 903
161 661 531
Toth and Williams (1999)
Locomotor activity over
2daysina HC
Recombinant CXB
inbred strains
males Chr3 D3Mit120
Chr12 D12Mit147 (G50Kbp
Chr12:35 723 478
35 723 626
et al.
(2009) Spontaneous HC
activity total activity
cross between
C57BL/6J and
females Chr2
Chr2:84 182 854
149 000 000
Chr11:26 296 397
57 753 949
Chr20:1 746 871
23 642 232
Chr15:34 933 139
51 298 144
Chr2:151 799 021
164 277 226
Chr20:29 889 343
43 767 941
Chr10:111 732 321
118 150 744
Chr12:68 717 060
76 169 323
This congenic strains contains a small DBA/2J genomic insert that covers previously reported region of the provisional QTL on C57BL/6J background.
The results obtained using RI strains enable to point to provisional linkage and not to confirmed QTL.
Genes, Brain and Behavior
Page 8
Unravel genetic architecture of physical activity
Table 4: Mouse QTL studies using running wheel activity model (RWA)
Article Phenotype Mouse strain Gender QTL
Mouse genomic
Mouse genomic
location in bp
human region
et al.
distance and time (day
1 through 6)
Advanced intercross line
between C57BL/6J and
high runner line bred
from outbred Hsd:ICR
Male and
Chr7 99124 Mb Chr7:99 000 000
124 000 000
Chr11:3 587 645
17 316 591
Chr16:18 515 287
25 839 874
Chr11:71 304 680
75 216 964
Chr16:16 764 146
18 232 691
average speed on day 5
and 6
Chr2 81103 Mb Chr2:81 000 000103
000 000
Chr11:34 845 92157
753 949
Chr2:184 648 829
188 410 216
Chr14 69 92 Mb Chr14:69 000 000
92 000 000
Chr13:53 226 033
65 622 467
Chr13:41 471 003
49 799 099
Chr8:20 190 005
24 475 895
Chr12 73 81 Mb Chr12:73 000 000
81 000 000
chr14:61 068 749
70 317 981
maximum speed,
average on day 5 and 6
Chr2 80115 Mb Chr2:80 000 000
115 000 000
Chr11:34 845 921
57 753 949
Chr2:183 423 068
188 410 216
Chr11 714 Mb Chr11:7 000 000
14 000 000
Chr7:45 649 431
52 696 893
Chr5:130 483 100
157 964 527
4561 Mb Chr11:45 000 000
61 000 000
Chr5:177 530 538
180 682 008
et al.
Daily RWA F
intercross between
C57L/J and C3H/HeJ
Male and
Chr13 rs6329684
Chr13:10 065 818
duration Chr13:9 876 613
10 360 803
Chr1:239 792 373-
240 072 720
distance Chr9 rs13480073
Chr13 rs6329684 Chr13:10 065 818 Chr1: 239 792 373-
240 072 720
speed Chr9 rs13480073
Chr13 rs6329684 Chr13: 9 876 613
10 360 803
chr1:239 550 008
240 073 138
et al.
(2010) RWA 41 inbred lines (haplotype
association mapping)
Male and
Chr12 89.35 89.46
Mbp (Nrxn3)
Chr12:89 350 000
89 460 000
Chr14:79 271 856
79 377 115
Chr18 11.57 11.74
Mbp (Rbbp8)
Chr18:11 570 000
11 740 000
Chr18:20 410 696
20 599 048
Chr19 15.77 16.19
Chr19:15 770 000
16 190 000
Chr9:80 574 663
81 084 208
Genes, Brain and Behavior
(2013) 9
Page 9
Kostrzewa and Kas
Table 4:
Article Phenotype Mouse strain Gender QTL
Mouse genomic
Mouse genomic
location in bp
human region
distance Male Chr5 115.03 118.12
Chr5:115 030 000
118 120 000
chr12:117 394 310
121 287 075
Chr6 145.45145.46
Mbp (Ifltd1)
Chr6:145 450 000
145 460 000
Chr12:25 742 744
25 771 753
Chr8 61.3785.40 Mbp Chr8:61 370 000
85 400 000
Chr4:141 255 401
170 039 614
Chr8:17 992 087
20 177 976
Chr22:33 659 180
35 953 121
Chr19:12 757 314
14 683 008
Chr19:17 970 477
19 774 937
Chr13 96.2596.63 Mbp Chr13:96 250 000
96 630 000
Chr5:74 681 475
75 242 696
speed Chr6 119.46119.81
Chr6:119 460 000
119 810 000
Chr12:1 157 035
1 720 652
distance Female Chr8 96.9697.22 Mbp Chr8:96 960 000
97 220 000
Chr8:97 029 176
97 289 176
Chr11 84.2086.67 Mbp Chr11:84 200 000
86 670 000
Chr17:58 204 722
60 306 122
Chr17:34 838 583
35 675 858
Chr17:57 802 952
58 046 693
speed Chr11 83.7186.23 Mbp Chr11:83 710 000
86 230 000
Chr17:34 838 583
36 200 511
Chr17:58 204 722
59 962 753
duration ChrX 106.30108.65
ChrX:106 300 000
108 650 000
ChrX:77 421 042
79 711 308
et al.
Daily RWA Backcross
Male and
Chr7 42.7560.75 cM
average running speed
maximum running speed Chr6 39.7273.00 cM
Chr7 35.7460.74 cM
et al.
RWA in constant
darkness daily activity
level (average running
wheel revolutions over
15 cycles)
cross between
BALB/cJ and C57BL/6J
Female Chr16 and ChrX
1 (Act1) locus)
Chr16:97 830 150
97 830 295
Chr21:43 267 661
43 267 711
Chrx:11 247 634
11 299 757
Chrx:39 164 446
39 221 777
ChrX:39 164 446
39 221 777
Genes, Brain and Behavior
Page 10
Unravel genetic architecture of physical activity
Table 4:
Article Phenotype Mouse strain Gender QTL
Mouse genomic
Mouse genomic
location in bp
human region
et al.
RWA normal light/dark cycle and
constant darkness (average
running wheel revolutions)
23 SMXA recombinant inbred
strains of SM/J and A/J
progenitor strains
Males Chr1 D1Rik136
Chr17 D17Rik98
et al.
(2009) RWA mean activity over 3
weeks o f recording in
habituated mice
population between C57BL/6J
and CSS13 (BxA) strain
Males Chr13 peak: D13Mit254
(narrowed down
in Yang
et al.
Yang et al. (2012) RWA mean activity over 3
weeks o f recording in
habituated mice
ISCS between C57BL/6J and
CSS13 (BxA) strain
males Chr13 38.84 42.60 Mbp Chr13:38 840 000
42 600 000
chr6:8 317 673
12 616 944
ISCS, interval-specific congenic strains.
There are additional QTLs detected for RWA on specific days of testing. For details, please see the original article.
Further refined by subsequent analysis on the same data by Leamy
et al.
(2010, 2011).
Backcross mice selectively bred for high RWA and a strain with a recessive mutation resulting in triceps muscle size reduction.
Suggestive QTLs.
According to authors this QTL overlaps with QTL associated by Gershenfeld
et al.
(1997)with the open field activity.
The region contains only 10 genes. Authors proposed a candidate gene: Tcfap2a.
each. After conducting the QTL analysis for each of these
time windows, they concluded that some of the QTLs affect
the behavior only at the younger or older mice while other
show stable effect over time (Leamy
et al.
2010). These
two studies, taken together, raise questions regarding the
conditions under which RWA behavior can be used to model
VE in humans. One could ask whether there is a threshold for
the duration of the RWA experiment after which the behavior
and underlying QTLs may be considered comparable with the
VE phenotype and its genetic basis. Another valid question
regard whether mouse QTLs with a transient effect on RWA
may be still considered candidate regions for the VE in
humans. Furthermore, it is not clear whether homologous
human regions would also have a transient effect on PA.
In this case one could wonder to what extent the age-
dependent architecture of PA would be comparable between
the species and which human and mouse age groups should
be compared with each other.
Finally, distinct aspects may be used to characterize
RWA, for example speed, duration of activity and distance
moved (counted as a number of revolutions or derived
from it physical distance). Four of the studies assessed
whether these distinct aspects of RWA have a different
genetic basis (Kelly
et al.
2010; Lightfoot
et al.
2008, 2010;
et al.
2010). In an F
cross between C57L/J and
C3H/HeJ, Lightfoot
et al.
(2008) detected one common QTL
on chromosome 13 which contributed to duration, speed
and distance of RWA. They also identified an additional
QTL on chromosome 9 which regulated speed (Lightfoot
et al.
2008) and distance moved on some of the days of
testing (Leamy
et al.
2010). However, the same researchers,
when using a panel of 41 inbred strains, could detect one
QTL regulating the duration of RWA and other three QTLs
regulating the distance in both genders (Lightfoot
et al.
2010). They also detected two sex-specific QTLs for speed
and six sex-specific QTLs for distance (Lightfoot
et al.
Furthermore, Kelly
et al.
(2010) found no overlap between
QTLs regulating speed and distance moved in an advanced
intercross line. Finally, Nehrenberg
et al.
(2010) could localize
only QTLs regulating speed but no QTLs for the duration or
distance moved.
Overlap of genetic findings between species
Two distinct genetic regions were identified independently
by human and mouse studies. Firstly, a QTL on chromosome
7 associated with an average distance run and time spent
on RWA maps to three different syntenic regions in humans:
11p15.4-p15.1, 16p12.3-12.1 and 11q13.4 (Kelly
et al.
The first of these QTL encloses a previously reported locus
on 11p15.2, which in a study by Simonen
et al.
(2003b) was
associated with a total time spent on PA when assessed by
a questionnaire. Secondly, Umemori
et al.
(2009) discovered
a broad QTL (13 Mbp) on mouse chromosome 2 associated
with activity in a HC, which explained 8.9% of the phenotypic
variance. This region encloses a syntenic human region
(20q12) associated with physical inactivity when measured
with a questionnaire (Simonen
et al.
Genes, Brain and Behavior
(2013) 11
Page 11
Kostrzewa and Kas
Evaluating the significance of the interspecies
The number of regions overlapping between species is
low despite the high total number of candidate regions
pointed in both species. This is most probably caused by
the methodological discrepancies and genetic background
influence, as mentioned before in this article. There are
various possible conclusions that may result from this
observation. It may be hypothesized that the regions which
overlap are very significant QTLs with a robust effect on PA
levels (thus occurring multiple times despite the potential
discrepancies). In this case, the overlapping regions could be
used in the comparative genomics approach. It is, however,
also possible that the observed overlap is solely caused
by sampling multiple times from a limited pool of possible
genetic regions. We wanted to test this hypothesis and
assessed the probability of obtaining this QTL overlap by
chance. In order to do so, we conducted a simulation which
aim was to assess the expected overlap levels in case of a
given number of sampling events ( experiments).
In order to proceed with the statistical analysis of the
interspecies overlap, the following choices were made. First,
as most human genetic studies considered VE and not SPE
(only one study), we decided to conduct the simulation only
for the human VE studies in combination with RWA mouse
studies. Second, as PA and physical inactivity in humans
may have different genetic basis, we included only findings
regarding PA in humans. Third, in order to make the analysis
possible, we approximated a finite number of possible
candidate regions (850 chromosomal regions; see
As a result, we chose the following number of studies into
the final analysis. Eight studies on PA in humans pointed
out to 15 candidate regions which mapped to 28 different
chromosomal regions for VE. Furthermore, studies using
RWA pointed to 25 QTLs, out of which all could be translated
into 64 human chromosomal regions. When these results are
taken under consideration then only one region overlapped
between mouse and human studies of VE.
A Monte Carlo simulation was conducted to model the
following situation. A pool of 850 chromosomal regions was
used twice in a sampling with replacement; in the first
experiment 28 regions and in the second experiment 64
regions were picked randomly. This operation was simulated
one million times. We asked the question, how likely would
it be to find the same region in experiments 1 and 2. The
Monte Carlo simulation (Fig. 1a) showed that one could
obtain a double overlap in 29.16% of cases due to a random
sampling. The result suggests that the overlap obtained while
using the real data of mouse and human candidate regions
is not higher than predicted on a chance level.
There are limitations of this simulation that should be
addressed. These limitations result from several assump-
tions and conditions necessary to conduct this thought
experiment. The most important initial condition was the
translation of all the regions into chromosomal regions which
was conducted in order to obtain a finite number of all pos-
sible regions that may be identified as candidate regions.
The simulation would not be possible when using simply
numbers of QTLs discovered because it is not possible to
establish a final number of QTLs that can be obtained. As
a consequence, two properties of this simulation emerged
Figure 1: Results of the Monte Carlo simulation regarding chances of obtaining an X number of overlapping genetic regions
when sampling two times from a limited pool of genetic regions (by conducting human and mouse studies). The
axes depicts
the percentage of times (if the sampling is repeated 1 000 000 times) when one would obtain X times an overlap due to chance. (a) In
this situation the number of chromosomal regions identified in human and mouse studies is taken under consideration. On the basis of
literature review we discovered one region which was identified by both, mouse and human studies. The graph shows that this event
is expected in 29.16% of samplings. Therefore, obtained overlap does not exceed the chance level. (b) In this situation, the number of
QTLs pointed out in human and mouse studies is taken under consideration. On the basis of literature review we discovered overlap
between one QTL from human studies and one QTL from mouse studies. The graph shows that this event is expected in 29.03% of
samplings. Therefore, obtained overlap does not exceed the chance level.
Genes, Brain and Behavior
Page 12
Unravel genetic architecture of physical activity
which may work as a disadvantage for the comparative
genomics approach. First, we neglected the fact that QTLs
have different sizes and therefore harbor different amount
of genes. However, the size of the majority of the QTLs
could not have been established due to a lack of confidence
intervals reported. Therefore, we have actually increased the
chance of a theoretical overlap by translating the QTLs for
which only the peak value was known into chromosomal
regions. Secondly, we neglected the fact that some of the
bands are not independent as they were identified simultane-
ously by one QTL. A way to circumvent this in the simulation
method would be to consider each QTL as identifying only
one candidate region. To test this we conducted additional
simulation in which a pool of 850 chromosomal regions was
used twice in a sampling with replacement; in the first exper-
iment 15 regions, and in the second experiment 25 regions,
were picked randomly. This operation was simulated one
million times and we asked the question: how likely would
it be to find the same region in experiments 1 and 2. The
Monte Carlo simulation showed that one could obtain a dou-
ble overlap in 29.03% of cases due to a random sampling.
Thus, even in this case the obtained overlap could still be
very likely obtained by chance.
It is of note that the inclusion of all 850 chromosomal
regions as the final number of all possible candidate regions
worked in advantage of the comparative genomics approach
in this thought experiment. The number of all possible
candidate regions should be lower than 850, because
otherwise, it would suggest that each of the possible
regions harbors a candidate gene. In this case, the further
studies aiming at finding regions contributing to a complex
phenotype, would be pointless. The use of this high number
of possible candidate regions increased the chance of
considering the existing overlap to be significant. Therefore,
despite the fact that this simulation has limitations, one may
conclude that caution should be taken when interpreting the
meaning of any single overlap of genetic region for VE activity
between human and mouse studies when the comparison is
made on the basis of currently available findings. This is due
to the fact that it is not feasible to distinguish a significant
overlap from a random event when genetic candidate regions
are identified in studies with methodological discrepancies.
Data gathered in the current review point to the difficulties
that are encountered when comparing QTLs for a complex
phenotype, such as PA levels. There are various sources
of these challenges in the field of PA genetics. Firstly,
various definitions and methodologies may be used
within one species to assess the phenotype of interest.
Although these methods are considered to measure one
phenotype, divergent phenotypes may in fact be measured.
Consequently, genetic regions identified by these studies
may regulate various and not one phenotype. Secondly,
accurate translation of complex phenotypes between
species may be difficult. As a consequence, results obtained
using an animal model of a complex phenotype may
identify candidate regions for another phenotype than
initially intended. Thirdly, genetic background has a complex
modulating influence on the contribution of any genetic
region to the occurrence or the level of expression of
a complex phenotype. Therefore, comparison between
various human populations or various mouse strains may
be difficult. As a consequence, only some of the QTLs
identified in different species may be comparable. Finally,
other modifying factors such as sex and age of study subjects
may have a profound influence on the phenotype expressed
and its genetic architecture. Taking under consideration
above mentioned facts, the low level of between species
overlap of genetic regions associated with PA is not
surprising. However, this observation has consequences for
interpretation of the overlaps when they occur. Care needs
to be taken when detecting and overlap between regions as
this overlap may be based on a chance level. Therefore, it
may be difficult to use the comparative genomic strategy to
narrow down candidate genetic regions for PA on the basis
of currently available findings.
It is probable that the increasing number of mouse and
human studies would lead to increased number of observed
overlap; however, increase in the number of conducted
studies results also in an increase in number of regions which
would overlap by chance. Therefore, it may be necessary
to evaluate the exact setup of HC and RWA experiments
more strictly, which would enable their use as models SPA
and VE, respectively. The consideration should regard the
duration of HC or RWA experiment and the time after which
the behavioral outcome is considered stable, representative
and not influenced by novelty. This needs to be corrected
for the possible age influence on QTL architecture in mice
if the experiments w ould take weeks. Finally, it is important
to include age of the subjects (humans or mice) in within
and between species comparisons of detected QTLs. One
could question whether HC and RWA are adequate models
of SPA and VE and whether we should develop more
adequate models of these complex phenotypes. We hoped
that this question could be solved on the basis of genetic
overlap between human and mouse studies. However, due
to the methodological limitations, this kind of comparison
is not informative with the use of data available right now.
Therefore, we propose to consider more focus on phenotype
definitions and on the phenotype assessment methodology
within and between species in order to optimize translational
research for complex traits.
In order to include all mouse QTL studies which identified candidate
genetic regions associated with PA, we conducted a literature search
with a combination of terms: QTL, mouse, mice, locomotor activity,
HC activity, RWA. We also followed references from identified
articles. We included all articles which ever reported a significant
QTL for any of the phenotypes related to PA in mice. In total, for
the review we included six mouse studies identifying 16 QTLs for
the locomotor activity in the HC (Table 3) and seven mouse studies
identifying 31 QLTs for the RWA (Table 4).
As we were interested in the possible refinement of genetic
candidate regions for PA, we aimed at unraveling the amount of
overlap between regions reported up to now from mouse and human
Genes, Brain and Behavior
(2013) 13
Page 13
Kostrzewa and Kas
studies. Therefore, we translated the obtained mouse QTL regions
to syntenic human regions (using UCSC Gene Browser; Human
February 2009 dataset; GRCh37/hg19;
bin/hgGateway), whenever possible (7 QLTs for HC and 24 QTLs for
RWA). The translation was conducted in a following manner:
1 For mouse QTL regions with defined borders we translated
the whole QTL into syntenic region in humans;
2 For QTLs for which only pick marker is given (either Mit
marker, SNP, locus name or JaxLab marker), we used the
MGI database ( or NCBI map
viewer ( to
check the exact location of this marker;
a If the marker was a gene or was placed within a gene,
we translated the location of that particular gene;
b In other cases, we checked whether there was a
gene not further than 50 kbp away from the marker
location. This region between the marker and the gene
(including the gene) was used for the translation.
We expressed the location of obtained hypothetical human
genetic regions as chromosomal regions defined on the basis of
chromosomal bands (e.g. 20q11). This was done in order to define
a finite number of possible genetic regions that could have been
pointed by association or linkage studies. On the basis of the human
chromosome banding this number was set to 850 (possible regions)
et al.
2009). Finally, the significance of the number
of overlapping regions between the mouse QTL and human gene
association studies was estimated by conducting a Monte Carlo
simulation (MATLAB).
Aaltonen, S., Ortega-Alonso, A., Kujala, U.M. & Kaprio, J. (2010)
A longitudinal study on genetic and environmental influences on
leisure time physical activity in the Finnish Twin Cohort.
Twin Res
Hum Genet
13, 475481.
Anderssen, N., Jacobs, D.R. Jr., Sidney, S., Bild, D.E., Sternfeld, B.,
Slattery, M.L. & Hannan, P. (1996) Change and secular trends in
physical activity patterns in young adults: a seven-year longitudinal
follow-up in the Coronary Artery Risk Development in Young Adults
Study (CARDIA).
Am J Epidemiol
143, 351362.
Bailey, J.S., Grabowski-Boase, L., Steffy, B.M., Wiltshire, T.,
Churchill, G.A. & Tarantino, L.M. (2008) Identification of quanti-
tative trait loci for locomotor activation and anxiety using closely
related inbred strains.
Genes Brain Behav
7, 761769.
Belke, T.W. & Garland, T. Jr. (2007) A brief opportunity to run
does not function as a reinforcer for mice selected for high daily
wheel-running rates.
J Exp Anal Behav
88, 199213.
Belknap, J.K. (1992) Empirical estimates of Bonferroni corrections for
use in chromosome mapping studies with the BXD recombinant
inbred strains.
Behav Genet
22, 677684.
Berentzen, T., Kring, S.I., Holst, C., Zimmermann, E., Jess, T.,
Hansen, T., Pedersen, O., Toubro, S., Astrup, A. & Sorensen,
T.I. (2008) Lack of association of fatness-related FTO gene variants
with energy expenditure or physical activity.
J Clin Endocrinol
93, 29042908.
Beunen, G. & Thomis, M. (1999) Genetic determinants of sports
participation and daily physical activity.
Int J Obes Relat Metab
23 (Suppl 3), S55S63.
Bijnen, F.C., Feskens, E.J., Caspersen, C.J., Mosterd, W.L. &
Kromhout, D. (1998) Age, period, and cohort effects on physical
activity among elderly men during 10 years of follow-up: the
Zutphen Elderly Study.
J Gerontol A Biol Sci Med Sci
Boomsma, D.I., van den Bree, M.B., Orlebeke, J.F. & Molenaar, P.C.
(1989) Resemblances of parents and twins in sports participation
and heart rate.
Behav Genet
19, 123141.
Booth, F.W., Chakravarthy, M.V., Gordon, S.E. & Spangenburg, E.E.
(2002) Waging war on physical inactivity: using modern molecular
ammunition against an ancient enemy.
J Appl Physiol
93, 330.
Botstein, D. & Risch, N. (2003) Discovering genotypes underlying
human phenotypes: past successes for Mendelian disease, future
approaches for complex disease.
Nat Genet
33 (Suppl), 228237.
Bouchard, C., Tremblay, A., Leblanc, C., Lortie, G., Savard, R. &
Theriault, G. (1983) A method to assess energy expenditure in
children and adults.
Am J Clin Nutr
37, 461467.
Bray, M.S., Hagberg, J.M., Perusse, L., Rankinen, T., Roth, S.M.,
Wolfarth, B. & Bouchard, C. (2009) The human gene map
for performance and health-related fitness phenotypes: the
20062007 update.
Med Sci Sports Exerc
41, 3573.
Brene, S., Bjornebekk, A., Aberg, E., Mathe, A.A., Olson, L. & Werme,
M. (2007) Running is rewarding and antidepressive.
Physiol Behav
92, 136140.
Bronikowski, A.M., Morgan, T.J., Garland, T. Jr. & Carter, P.A.
(2006) The evolution of aging and age-related physical decline
in mice selectively bred for high voluntary exercise.
Buck, K., Lischka, T., Dorow, J. & Crabbe, J. (2000) Mapping
quantitative trait loci that regulate sensitivity and tolerance to
quinpirole: a dopamine mimetic selective for D(2)/D(3) receptors.
Am J Med Genet
96, 696705.
Butte, N.F., Cai, G., Cole, S.A. & Comuzzie, A.G. (2006) Viva la Familia
Study: genetic and environmental contributions to childhood
obesity and its comorbidities in the Hispanic population.
Am J
Clin Nutr
84, 646654.
Cai, G., Cole, S.A., Butte, N., Bacino, C., Diego, V., Tan, K., Goring,
H.H., O’Rahilly, S., Farooqi, I.S. & Comuzzie, A.G. (2006) A
quantitative trait locus on chromosome 18q for physical activity
and dietary intake in Hispanic children.
Obesity (Silver Spring)
Careau, V., Bininda-Emonds, O.R., Ordonez, G. & Garland, T. Jr.
(2012) Are voluntary wheel running and open-field behavior
correlated in mice? Different answers from comparative and
artificial selection approaches.
Behav Genet
42, 830844.
Casazza, K., Fontaine, K.R., Astrup, A., Birch, L.L., Brown, A.W.,
Bohan Brown, M.M., Durant, N., Dutton, G., Foster, E.M.,
Heymsfield, S.B., McIver, K., Mehta, T., Menachemi, N., Newby,
P.K., Pate, R., Rolls, B.J., Sen, B., Smith, D.L. Jr., Thomas, D.M. &
Allison, D.B. (2013) Myths, presumptions, and facts about obesity.
N Engl J Med
368, 446454.
Choh, A.C., Demerath, E.W., Lee, M., Williams, K.D., Towne, B.,
Siervogel, R.M., Cole, S.A. & Czerwinski, S.A. (2009) Genetic
analysis of self-reported physical activity and adiposity: the
Southwest Ohio Family Study.
Public Health Nutr
12, 10521060.
Cole, S.A., Butte, N.F., Voruganti, V.S., Cai, G., Haack, K., Kent, J.W.
Jr., Blangero, J., Comuzzie, A.G., McPherson, J.D. & Gibbs, R.A.
(2010) Evidence that multiple genetic variants of MC4R play a
functional role in the regulation of energy expenditure and appetite
in Hispanic children.
Am J Clin Nutr
91, 191199.
Crabbe, J.C., Kosobud, A., Young, E.R. & Janowsky, J.S. (1983)
Polygenic and single-gene determination of responses to ethanol
in BXD/Ty recombinant inbred mouse strains.
Neurobehav Toxicol
5, 181187.
De Moor, M.H., Posthuma, D., Hottenga, J.J., Willemsen, G.,
Boomsma, D.I. & de Geus, E.J. (2007a) Genome-wide linkage
scan for exercise p articipation in Dutch sibling pairs.
Eur J Hum
15, 12521259.
De Moor, M.H., Spector, T.D., Cherkas, L.F., Falchi, M., Hottenga,
J.J., Boomsma, D.I. & de Geus, E.J. (2007b) Genome-wide linkage
scan for athlete status in 700 British female DZ twin pairs.
Res Hum Genet
10, 812820.
De Moor, M., Liu, Y.J., Boomsma, D., Li, J., Hamilton, J., Hottenga,
J.J., Levy, S., Liu, X.G., Pei, Y.F., Posthuma, D., Recker, R.,
Sullivan, P., Wang, L., Willemsen, G., Yan, H., De Geus, E. &
Deng, H.W. (2009) Genome-Wide Association Study of exercise
behavior in Dutch and American Adults.
Med Sci Sports Exerc
De Moor, M., Willemsen, G., Rebollo-Mesa, I., Stubbe, J., De Geus,
E. & Boomsma, D. (2011) Exercise participation in adolescents
Genes, Brain and Behavior
Page 14
Unravel genetic architecture of physical activity
and their parents: evidence for genetic and generation specific
environmental effects.
Behav Genet
41, 211222.
DiPetrillo, K., Wang, X., Stylianou, I.M. & Paigen, B. (2005)
Bioinformatics toolbox for narrowing rodent quantitative trait loci.
Trends Genet
21, 683692.
Dishman, R.K. (2008) Gene-physical activity interactions in the
etiology of obesity: behavioral considerations.
Obesity (Silver
16 (Suppl 3), S60S65.
Eikelboom, R. (1999) Human parallel to voluntary wheel running:
Anim Behav
57, F11F12.
Eisener-Dorman, A.F., Grabowski-Boase, L., Steffy, B.M., Wiltshire,
T. & Tarantino, L.M. (2010) Quantitative trait locus and haplotype
mapping in closely related inbred strains identifies a locus for open
field behavior.
Mamm Genome
21, 231246.
Epstein, L.H. & Wing, R.R. (1980) Aerobic exercise and weight.
Addict Behav
5, 371388.
Eriksson, M., Rasmussen, F. & Tynelius, P. (2006) Genetic factors
in physical activity and the equal environment assumption the
Swedish young male twins study.
Behav Genet
36, 238247.
Flint, J. (2003) Analysis of quantitative trait loci that influence animal
J Neurobiol
54, 4677.
Flint, J. & Mott, R. (2001) Finding the molecular basis of quantitative
traits: successes and pitfalls.
Nat Rev Genet
2, 437445.
Flint, J. & Mott, R. (2008) Applying mouse complex-trait resources to
behavioural genetics.
456, 724727.
Flint, J., Corley, R., DeFries, J.C., Fulker, D.W., Gray, J.A., Miller,
S. & Collins, A.C. (1995) A simple genetic basis for a complex
psychological trait in laboratory mice.
269, 14321435.
Fuentes, R.M., Perola, M., Nissinen, A. & Tuomilehto, J. (2002) ACE
gene and physical activity, blood pressure, and hypertension: a
population study in Finland.
J Appl Physiol
92, 25082512.
Furuse, T., Takano-Shimizu, T., Moriwaki, K., Shiroishi, T. & Koide,
T. (2002) QTL analyses of spontaneous activity by using mouse
strains from Mishima battery.
Mamm Genome
13, 411415.
Garland, T. Jr., Morgan, M.T., Swallow, J.G., Rhodes, J.S., Girard,
I., Belter, J.G. & Carter, P.A. (2002) Evolution of a small-muscle
polymorphism in lines of house mice selected for high activity
56, 12671275.
Garland, T. Jr., Kelly, S.A., Malisch, J.L., Kolb, E.M., Hannon, R.M.,
Keeney, B.K., Van Cleave, S.L. & Middleton, K.M. (2011a) How to
run far: multiple solutions and sex-specific responses to selective
breeding for high voluntary activity levels.
Proc Biol Sci
Garland, T. Jr., Schutz, H., Chappell, M.A., Keeney, B.K., Meek, T.H.,
Copes, L.E., Acosta, W., Drenowatz, C., Maciel, R.C., van Dijk,
G., Kotz, C.M. & Eisenmann, J.C. (2011b) The biological control of
voluntary exercise, spontaneous physical activity and daily energy
expenditure in relation to obesity: human and rodent perspectives.
214, 206229.
Gershenfeld, H.K., Neumann, P.E., Mathis, C., Crawley, J.N., Li, X.
& Paul, S.M. (1997) Mapping quantitative trait loci for open-field
behavior in mice.
Behav Genet
27, 201210.
Gill, K.J. & Boyle, A.E. (2005) Quantitative trait loci for novelty/stress-
induced locomotor activation in recombinant inbred (RI) and
recombinant congenic (RC) strains of mice.
Behav Brain Res
Good, D.J., Coyle, C.A. & Fox, D.L. (2008) Nhlh2: a basic helix-loop-
helix transcription factor controlling physical activity.
Exerc Sport
Sci Rev
36, 187192.
Goodrick, C.L. (1975) Behavioral differences in young and aged mice:
strain differences for activity measures, operant learning, sensory
discrimination, and alcohol preference.
Exp Aging Res
1, 191207.
Goran, M.I. & Poehlman, E.T. (1992) Endurance training does not
enhance total energy expenditure in healthy elderly persons.
J Physiol
263, E950 E957.
Hagberg, J.M., Rankinen, T., Loos, R.J., Perusse, L., Roth, S.M.,
Wolfarth, B. & Bouchard, C. (2011) Advances in exercise, fitness,
and performance genomics in 2010.
Med Sci Sports Exerc
Hakanen, M., Raitakari, O.T., Lehtimaki, T., Peltonen, N., Pahkala, K.,
Sillanmaki, L., Lagstrom, H., Viikari, J., Simell, O. & Ronnemaa,
T. (2009) FTO genotype is associated with body mass index after
the age of seven years but not with energy intake or leisure-time
physical activity.
J Clin Endocrinol Metab
94, 12811287.
Hanson, R.W. & Hakimi, P. (2008) Born to run; the story of the
PEPCK-Cmus mouse.
90, 838842.
Haskell, W.L., Lee, I.M., Pate, R.R., Powell, K.E., Blair, S.N., Franklin,
B.A., Macera, C.A., Heath, G.W., Thompson, P.D. & Bauman, A.
(2007) Physical activity and public health: updated recommendation
for adults from the American College of Sports Medicine and the
American Heart Association.
Med Sci Sports Exerc
39, 14231434.
Henderson, N.D., Turri, M.G., DeFries, J.C. & Flint, J. (2004) QTL
analysis of multiple behavioral measures of anxiety in mice.
34, 267293.
Hitzemann, R., Demarest, K., Koyner, J., Cipp, L., Patel, N.,
Rasmussen, E. & McCaughran, J. Jr. (2000) Effect of genetic cross
on the detection of quantitative trait loci and a novel approach to
mapping QTLs.
Pharmacol Biochem Behav
67, 767772.
Hitzemann, R., Malmanger, B., Reed, C., Lawler, M., Hitzemann, B.,
Coulombe, S., Buck, K., Rademacher, B., Walter, N., Polyakov,
Y., Sikela, J., Gensler, B., Burgers, S., Williams, R.W., Manly, K.,
Flint, J. & Talbot, C. (2003) A strategy for the integration of QTL,
gene expression, and sequence analyses.
Mamm Genome
Ingram, D.K. (1983) Toward the behavioral assessment of biological
aging in the laboratory mouse: concepts, terminology, and
Exp Aging Res
9, 225238.
Ingram, D.K. (2000) Age-related decline in physical activity:
generalization to nonhumans.
Med Sci Sports Exerc
Ingram, D.K. & Reynolds, M.A. (1986) Assessing the predictive
validity of psychomotor tests as measures of biological age in
Exp Aging Res
12, 155162.
Kagan, J.K. & Berkum, M. (1954) The reward value of running activity.
J Comp Physiol Psychol
47, 108.
Kas, M.J. & Edgar, D.M. (1998) Crepuscular rhythms of EEG sleep-
wake in a hystricomorph rodent, Octodon degus.
13, 917.
Kas, M.J. & Van Ree, J.M. (2004) Dissecting complex behaviours in
the post-genomic era.
Trends Neurosci
27, 366369.
Kas, M.J., de Mooij-van Malsen, J.G., de, K.M., van Gassen, K.L.,
van Lith, H.A., Olivier, B., Oppelaar, H., Hendriks, J., de, W.M.,
Groot Koerkamp, M.J., Holstege, F.C., van Oost, B.A. & de Graan,
P.N. (2009) High-resolution genetic mapping of mammalian motor
activity levels in mice.
Genes Brain Behav
8, 1322.
Keeney, B.K., Raichlen, D.A., Meek, T.H., Wijeratne, R.S., Middleton,
K.M., Gerdeman, G.L. & Garland, T. Jr. (2008) Differential response
to a selective cannabinoid receptor antagonist (SR141716:
rimonabant) in female mice from lines selectively bred for
high voluntary wheel-running behaviour.
Behav Pharmacol
Kelly, S.A. & Pomp, D. (2013) Genetic determinants of voluntary
Trends Genet
29, 348357.
Kelly, M.A., Low, M.J., Phillips, T.J., Wakeland, E.K. & Yanagisawa,
M. (2003) The mapping of quantitative trait loci underlying strain
differences in locomotor activity between 129S6 and C57BL/6J
Mamm Genome
14, 692702.
Kelly, S.A., Nehrenberg, D.L., Peirce, J.L., Hua, K., Steffy, B.M.,
Wiltshire, T., Pardo-Manuel, D.V., Garland, T. Jr. & Pomp, D.
(2010) Genetic architecture of voluntary exercise in an advanced
intercross line of mice.
Physiol Genomics
42, 190200.
Kosyakova, N., Weise, A., Mrasek, K., Claussen, U., Liehr, T. & Nelle,
H. (2009) The hierarchically organized splitting of chromosomal
bands for all human chromosomes.
Mol Cytogenet
Koteja, P., Garland, T. Jr., Sax, J.K., Swallow, J.G. & Carter, P.A.
(1999) Behaviour of house mice artificially selected for high levels
of voluntary wheel running.
Anim Behav
58, 13071318.
Koyner, J., Demarest, K., McCaughran, J. Jr., Cipp, L. & Hitzemann,
R. (2000) Identification and time dependence of quantitative trait
Genes, Brain and Behavior
(2013) 15
Page 15
Kostrzewa and Kas
loci for basal locomotor activity in the BXD recombinant inbred
series and a B6D2 F2 intercross.
Behav Genet
30, 159170.
Kujala, U.M., Kaprio, J. & Koskenvuo, M. (2002) Modifiable risk
factors as predictors of all-cause mortality: the roles of genetics
and childhood environment.
Am J Epidemiol
156, 985993.
Lauderdale, D.S., Fabsitz, R., Meyer, J.M., Sholinsky, P., Ramakrish-
nan, V. & Goldberg, J. (1997) Familial determinants of moderate
and intense physical activity: a twin study.
Med Sci Sports Exerc
29, 10621068.
Leamy, L.J., Pomp, D. & Lightfoot, J.T. (2010) A search for
quantitative trait loci controlling within-individual variation of
physical activity traits in mice.
BMC Genet
11, 83.
Leamy, L.J., Pomp, D. & Lightfoot, J.T. (2011) Epistatic interactions
of genes influence within-individual variation of physical activity
traits in mice.
139, 813821.
Lett, B.T., Grant, V.L., Byrne, M.J. & Koh, M.T. (2000) Pairings of a
distinctive chamber with the aftereffect of wheel running produce
conditioned place preference.
34, 8794.
Levine, J.A., Eberhardt, N.L. & Jensen, M.D. (1999) Role of
nonexercise activity thermogenesis in resistance to fat gain in
283, 212214.
Lhotellier, L. & Cohen-Salmon, C. (1989) Genetics and senescence.
I. Age-related changes in activity and exploration in three inbred
strains of mice.
Physiol Behav
45, 491493.
Lightfoot, J.T. (2008) Sex hormones’ regulation of rodent physical
activity: a review.
Int J Biol Sci
4, 126132.
Lightfoot, J.T., Turner, M.J., Daves, M., Vordermark, A. & Kleeberger,
S.R. (2004) Genetic influence on daily wheel running activity level.
Physiol Genomics
19, 270276.
Lightfoot, J.T., Turner, M.J., Pomp, D., Kleeberger, S.R. & Leamy,
L.J. (2008) Quantitative trait loci for physical activity traits in mice.
Physiol Genomics
32, 401408.
Lightfoot, J.T., Leamy, L., Pomp, D., Turner, M.J., Fodor, A.A., Knab,
A., Bowen, R.S., Ferguson, D., Moore-Harrison, T. & Hamilton, A.
(2010) Strain screen and haplotype association mapping of wheel
running in inbred mouse strains.
J Appl Physiol
109, 623634.
Liu, G., Zhu, H., Lagou, V., Gutin, B., Stallmann-Jorgensen, I.S.,
Treiber, F.A., Dong, Y. & Snieder, H. (2010) FTO variant rs9939609
is associated with body mass index and waist circumference,
but not with energy intake or physical activity in European- and
African-American youth.
BMC Med Genet
11, 57.
Loos, R.J., Rankinen, T., Tremblay, A., Perusse, L., Chagnon, Y.
& Bouchard, C. (2005) Melanocortin-4 receptor gene and physical
activity in the Quebec Family Study.
Int J Obes (Lond)
29, 420 428.
Lorentzon, M., Lorentzon, R., Lerner, U.H. & Nordstrom, P. (2001)
Calcium sensing receptor gene polymorphism, circulating calcium
concentrations and bone mineral density in healthy adolescent
Eur J Endocrinol
144, 257261.
Maia, J.A., Thomis, M. & Beunen, G. (2002) Genetic factors in
physical activity levels: a twin study.
23, 8791.
Malisch, J.L., Breuner, C.W., Kolb, E.M., Wada, H., Hannon,
R.M., Chappell, M.A., Middleton, K.M. & Garland, T. Jr. (2009)
Behavioral despair and home-cage activity in mice with chronically
elevated baseline corticosterone concentrations.
Behav Genet
Manson, J.E., Skerrett, P.J., Greenland, P. & VanItallie, T.B. (2004)
The escalating pandemics of obesity and sedentary lifestyle. A call
to action for clinicians.
Arch Intern Med
164, 249258.
Marcus, B.H. & Forsyth, L.H. (1999) How are we doing with physical
Am J Health Promot
14, 118124.
Mathis, C., Neumann, P.E., Gershenfeld, H., Paul, S.M. & Crawley,
J.N. (1995) Genetic analysis of anxiety-related behaviors and
responses to benzodiazepine-related drugs in AXB and BXA
recombinant inbred mouse strains.
Behav Genet
25, 557568.
Mayeda, A.R. & Hofstetter, J.R. (1999) A QTL for the genetic variance
in free-running period and level of locomotor activity between
inbred strains of mice.
Behav Genet
29, 171176.
Mokdad, A.H., Marks, J.S., Stroup, D.F. & Gerberding, J.L. (2004)
Actual causes of death in the United States, 2000.
Morgan, T.J., Garland, T. Jr. & Carter, P.A. (2003) Ontogenies in
mice selected for high voluntary wheel-running activity. I. Mean
57, 646657.
Morgan, M.A., Schulkin, J. & Pfaff, D.W. ( 2004) Estrogens and
non-reproductive behaviors related to activity and fear.
Biobehav Rev
28, 5563.
Myers, J., Prakash, M., Froelicher, V., Do, D., Partington, S. &
Atwood, J.E. (2002) Exercise capacity and mortality among men
referred for exercise testing.
N Engl J Med
346, 793801.
Nehrenberg, D.L., Wang, S., Hannon, R.M., Garland, T. Jr. & Pomp,
D. (2010) QTL underlying voluntary exercise in mice: interactions
with the ‘‘mini muscle’’ locus and sex.
J Hered
101, 4253.
Nelson, M.C., Neumark-Stzainer, D., Hannan, P.J., Sirard, J.R. &
Story, M. (2006) Longitudinal and secular trends in physical
activity and sedentary behavior during adolescence.
118, e1627e1634.
Novak, C.M., Burghardt, P.R. & Levine, J.A. (2012) The use of
a running wheel to measure activity in rodents: relationship to
energy balance, general activity, and reward.
Neurosci Biobehav
36, 10011014.
Perusse, L., Tremblay, A., Leblanc, C. & Bouchard, C. (1989) Genetic
and environmental influences on level of habitual physical activity
and exercise participation.
Am J Epidemiol
129, 10121022.
Plomin, R., McClearn, G.E., Gora-Maslak, G. & Neiderhiser, J.M.
(1991) Use of recombinant inbred strains to detect quantitative
trait loci associated with behavior.
Behav Genet
21, 99116.
Radcliffe, R.A., Jones, B.C. & Erwin, V.G. (1998) Mapping of
provisional quantitative trait loci influencing temporal variation in
locomotor activity in the LS x SS recombinant inbred strains.
28, 3947.
Rankinen, T., Roth, S.M., Bray, M.S., Loos, R., Perusse, L., Wolfarth,
B., Hagberg, J.M. & Bouchard, C. (2010) Advances in exercise,
fitness, and performance genomics.
Med Sci Sports Exerc
Rezende, E.L., Gomes, F.R., Chappell, M.A. & Garland, T. Jr. (2009)
Running behavior and its energy cost in mice selectively bred
for high voluntary locomotor activity.
Physiol Biochem Zool
Rhodes, J.S., Garland, T. Jr. & Gammie, S.C. (2003) Patterns of
brain activity associated with variation in voluntary wheel-running
Behav Neurosci
117, 12431256.
Richert, L., Chevalley, T., Manen, D., Bonjour, J.P., Rizzoli, R. &
Ferrari, S. (2007) Bone mass in prepubertal boys is associated with
a Gln223Arg amino acid substitution in the leptin receptor.
J Clin
Endocrinol Metab
92, 43804386.
Roth, S.M., Rankinen, T., Hagberg, J.M., Loos, R.J., Perusse, L.,
Sarzynski, M.A., Wolfarth, B. & Bouchard, C. (2012) Advances in
exercise, fitness, and performance genomics in 2011.
Med Sci
Sports Exerc
44, 809817.
Rowland, T.W. (1998) The biological basis of physical activity.
Sci Sports Exerc
30, 392399.
Sallis, J.F. (2000) Age-related decline in physical activity: a synthesis
of human and animal studies.
Med Sci Sports Exerc
Schoenborn, C.A. (1986) Health habits of U.S. adults, 1985: the
‘‘Alameda 7’’ revisited.
Public Health Rep
101, 571580.
Schughart, K., Libert, C. & Kas, M.J. (2012) Human disease: Strength
to strength for mouse models.
492, 41.
Schultz, D.D. (1965)
Sensory Restriction: Effects on Behavior
Academic Press, New York.
Seabra, A.F., Mendonca, D.M., Goring, H.H., Thomis, M.A. & Maia,
J.A. (2008) Genetic and environmental factors in familial clustering
in physical activity.
Eur J Epidemiol
23, 205211.
Shephard, R.J. (2003) Limits to the measurement of habitual physical
activity by questionnaires.
Br J Sports Med
37, 197206.
Sherwin, C.M. (1998) Voluntary wheel running: a review and novel
Anim Behav
56, 1127.
Shimomura, K., Low-Zeddies, S.S., King, D.P., Steeves, T.D.,
Whiteley, A., Kushla, J., Zemenides, P.D., Lin, A., Vitaterna, M.H.,
Churchill, G.A. & Takahashi, J.S. (2001) Genome-wide epistatic
Genes, Brain and Behavior
Page 16
Unravel genetic architecture of physical activity
interaction analysis reveals complex genetic determinants of
circadian behavior in mice.
Genome Res
11, 959980.
Simonen, R.L., Perusse, L., Rankinen, T., Rice, T., Rao, D.C. &
Bouchard, C. (2002) Familial aggregation of physical activity levels
in the Quebec Family Study.
Med Sci Sports Exerc
34, 1137 1142.
Simonen, R.L., Rankinen, T., Perusse, L., Leon, A.S., Skinner, J.S.,
Wilmore, J.H., Rao, D.C. & Bouchard, C. (2003a) A dopamine D2
receptor gene polymorphism and physical activity in two family
Physiol Behav
78, 751757.
Simonen, R.L., Rankinen, T., Perusse, L., Rice, T., Rao, D.C.,
Chagnon, Y. & Bouchard, C. (2003b) Genome-wide linkage scan
for physical activity levels in the Quebec Family study.
Med Sci
Sports Exerc
35, 13551359.
Stefan, N., Vozarova, B., Del, P.A., Ossowski, V., Thompson, D.B.,
Hanson, R.L., Ravussin, E. & Tataranni, P.A. (2002) The Gln223Arg
polymorphism of the leptin receptor in Pima Indians: influence on
energy expenditure, physical activity and lipid metabolism.
Int J
Obes Relat Metab Disord
26, 16291632.
Stubbe, J.H., Boomsma, D.I. & de Geus, E.J. (2005) Sports
participation during adolescence: a shift from environmental to
genetic factors.
Med Sci Sports Exerc
37, 563570.
Stubbe, J.H., Boomsma, D.I., Vink, J.M., Cornes, B.K., Martin, N.G.,
Skytthe, A., Kyvik, K.O., Rose, R.J., Kujala, U.M., Kaprio, J., Harris,
J.R., Pedersen, N.L., Hunkin, J., Spector, T.D. & de Geus, E.J.
(2006) Genetic influences on exercise participation in 37,051 twin
pairs from seven countries.
PLoS One
1, e22.
Suzuki, T., Ishikawa, A., Nishimura, M., Yoshimura, T., Namikawa,
T. & Ebihara, S. (2000) Mapping quantitative trait loci for circadian
behavioral rhythms in SMXA recombinant inbred strains.
30, 447453.
Swallow, J.G., Carter, P.A. & Garland, T. Jr. (1998) Artificial selection
for increased wheel-running behavior in house mice.
Behav Genet
28, 227237.
Talbot, C.J., Nicod, A., Cherny, S.S., Fulker, D.W., Collins, A.C. &
Flint, J. (1999) High-resolution mapping of quantitative trait loci in
outbred mice.
Nat Genet
21, 305308.
Telama, R., Leskinen, E. & Yang, X. (1996) Stability of habitual physical
activity and sport participation: a longitudinal tracking study.
J Med Sci Sports
6, 371378.
Toth, L.A. & Williams, R.W. (1999) A quantitative genetic analysis of
locomotor activity in CXB recombinant inbred mice.
Behav Genet
29, 319328.
Turri, M.G., Talbot, C.J., Radcliffe, R .A., Wehner, J.M. & Flint,
J. (1999) High-resolution mapping of quantitative trait loci for
emotionality in selected strains of mice.
Mamm Genome
Turri, M.G., Datta, S.R., DeFries, J., Henderson, N.D. & Flint, J.
(2001) QTL analysis identifies multiple behavioral dimensions
in ethological tests of anxiety in laboratory mice.
Curr Biol
Turri, M.G., DeFries, J.C., Henderson, N.D. & Flint, J. (2004)
Multivariate analysis of quantitative trait loci influencing variation
in anxiety-related behavior in laboratory mice.
Mamm Genome
Umemori, J., Nishi, A., Lionikas, A., Sakaguchi, T., Kuriki, S., Blizard,
D.A. & Koide, T. (2009) QTL analyses of temporal and intensity
components of home-cage activity in KJR and C57BL/6J strains.
BMC Genet
10, 40.
Vaanholt, L.M., Daan, S., Garland, T. Jr. & Visser, G.H. (2010)
Exercising for life? Energy metabolism, body composition, and
longevity in mice exercising at different intensities.
Biochem Zool
83, 239251.
Viggiano, D. (2008) The hyperactive syndrome: metanalysis of genetic
alterations, pharmacological treatments and brain lesions which
increase locomotor activity.
Behav Brain Res
194, 114.
de Vilhena e Santos, D.M., Katzmarzyk, P.T., Seabra, A.F. & Maia,
J.A. (2012) Genetics of physical activity and physical inactivity in
Behav Genet
42, 559578.
de Visser, L., van den Bos, R. & Spruijt, B.M. (2005) Automated
home cage observations as a tool to measure the effects of wheel
running on cage floor locomotion.
Behav Brain Res
160, 382 388.
de Visser, L., van den Bos, R., Kuurman, W.W., Kas, M.J. & Spruijt,
B.M. (2006) Novel approach to the behavioural characterization
of inbred mice: automated home cage observations.
Genes Brain
5, 458466.
Wade, M.G., Ellis, M.J. & Bohrer, R.E. (1973) Biorhythms in the
activity of children during free play.
J Exp Anal Behav
20, 155162.
Westerterp, K.R. (2009) Assessment of physical activity: a critical
Eur J Appl Physiol
105, 823828.
Winnicki, M., Accurso, V., Hoffmann, M., Pawlowski, R., Dorigatti,
F., Santonastaso, M., Longo, D., Krupa-Wojciechowska, B.,
Jeunemaitre, X., Pessina, A.C., Somers, V.K. & Palatini, P.
(2004) Physical activity and angiotensin-converting enzyme gene
polymorphism in mild hypertensives.
Am J Med Genet A
Wood, A.C., Rijsdijk, F., Saudino, K.J., Asherson, P. & Kuntsi, J.
(2008) High heritability for a composite index of children’s activity
level measures.
Behav Genet
38, 266276.
Yang, H.S., Vitaterna, M.H., Laposky, A.D., Shimomura, K. & Turek,
F.W. (2009) Genetic analysis of daily physical activity using a mouse
chromosome substitution strain.
Physiol Genomics
39, 4755.
Yang, H.S., Shimomura, K., Vitaterna, M.H. & Turek, F.W. (2012) High-
resolution mapping of a novel genetic locus regulating voluntary
physical activity in mice.
Genes Brain Behav
11, 113124.
Zhang, S., Lou, Y., Amstein, T.M., Anyango, M., Mohibullah, N.,
Osoti, A., Stancliffe, D., King, R., Iraqi, F. & Gershenfeld, H.K.
(2005) Fine mapping of a major locus on chromosome 10 for
exploratory and fear-like behavior in mice.
Mamm Genome
Zombeck, J.A., Deyoung, E.K., Brzezinska, W.J. & Rhodes, J.S. (2011)
Selective breeding for increased home cage physical activity in
collaborative cross and Hsd:ICR mice.
Behav Genet
41, 571582.
These studies were supported by a ZonMW VIDI-grant
(91786327) from The Netherlands Organization for Scientific
Research (NWO; to M.K. The data collection
was supported by Netherlands Organization of Health Research
and Development; contract grant number: ZonMW #94505017.
The funders had no role in study design, data collection and
analysis, decision to publish or preparation of the manuscript. We
would like to thank Wouter Koning for conducting the simulation
Supporting Information
Additional supporting information may be found in the online
version of this article at the publisher’s web-site:
Table S1: Mouse QTL studies using an open field test
(OFT) activity model.
Genes, Brain and Behavior
(2013) 17
Page 17
    • "In the postoperative group, positive weight-related effects that can arise as early as few months after surgery [61][62][63]might have lessened the motivation to behavioral changes in some patients. Furthermore, whether a person is prone to engage in habitual PA may be at least partly determined by genetic [22, 64] or temperamental factors [65]. The present investigation has a number of limitations that have to be considered when interpreting the results. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Physical activity (PA) is considered to have a beneficial influence on executive functioning, including decision-making. Enhanced decision-making after bariatric surgery may strengthen patients’ ability to delay gratification, helping to establish appropriate eating behavior. The objectives of this study were to (1) compare a preoperative group with a postoperative group with regard to daily PA, decision-making, and eating disturbances; and (2) investigate the relationship between these variables. Methods The study included 71 bariatric surgery candidates (55 % women, BMI [kg/m2] M = 46.9, SD = 6.0) and 73 postoperative patients (57 % women, BMI M = 32.0, SD = 4.1; 89 % Roux-en-Y gastric bypass, 11 % sleeve gastrectomy; months postoperative M = 8.2, SD = 3.5; total weight loss [%] M = 33.2, SD = 8.9) who completed SenseWear Pro2 activity monitoring. Decision-making was assessed using a computerized version of the Iowa Gambling Task and eating disorder psychopathology using the Eating Disorder Examination-Questionnaire. Results The number of patients who were classified as physically inactive was similarly high in the pre- and postoperative groups. No group differences emerged with regard to decision-making, but the postoperative group exhibited less eating disturbances than the preoperative group. No significant associations were found between PA, decision-making, and eating behavior. Conclusions Patients after bariatric surgery were not more physically active than bariatric surgery candidates, which should be considered in care programs. Additionally, future research is needed to explore the possible link between PA, patients’ decision-making abilities, and eating disturbances concerning dose-response questions.
    No preview · Article · May 2016 · Obesity Surgery
  • Source
    • "Engaging in exercise itself has been related to changes in dopaminergic transmission [15] and individual differences in the dopaminergic reward system, more specifically in genetic variants that affect the system, have previously been linked to differences in physical activity both in rodents [16] and in humans [19]. Admittedly, some of this previous evidence implicating dopaminergic genes looked at more general forms of physical activity (e.g., parts of [19]) instead of the trait of self-initiated exercise behavior used here [55]. We focused on voluntary exercise behavior for two reasons. "
    [Show abstract] [Hide abstract] ABSTRACT: Purpose: Twin studies provide evidence that genetic influences contribute strongly to individual differences in exercise behavior. We hypothesize that part of this heritability is explained by genetic variation in the dopaminergic reward system. Eight single nucleotide polymorphisms (SNPs in DRD1: rs265981, DRD2: rs6275, rs1800497, DRD3: rs6280, DRD4: rs1800955, DBH: rs1611115, rs2519152, and in COMT: rs4680) and three variable number of tandem repeats (VNTRs in DRD4, upstream of DRD5, and in DAT1) were investigated for an association with regular leisure time exercise behavior. Materials and methods: Data on exercise activities and at least one SNP/VNTR were available for 8,768 individuals aged 7 to 50 years old that were part of the Netherlands Twin Register. Exercise behavior was quantified as weekly metabolic equivalents of task (MET) spent on exercise activities. Mixed models were fitted in SPSS with genetic relatedness as a random effect. Results: None of the genetic variants were associated with exercise behavior (P>.02), despite sufficient power to detect small effects. Discussion and conclusions: We did not confirm that allelic variants involved in dopaminergic function play a role in creating individual differences in exercise behavior. A plea is made for large genome-wide association studies to unravel the genetic pathways that affect this health-enhancing behavior.
    Full-text · Article · Jan 2013 · BioMed Research International
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: Predisposition to engage in exercise is highly variable and simultaneously influenced by the environment, complex genomics, and their interactions. Given the importance of exercise to health, understanding the underlying influences of variability is crucial. Here, we discuss murine systems approaches, focusing on 'omics', relevant to revealing the architecture of voluntary activity.
    Full-text · Article · Nov 2015 · Trends in Endocrinology and Metabolism