Search tips
Search criteria 


Logo of procbThe Royal Society PublishingProceedings BAboutBrowse by SubjectAlertsFree Trial
Proc Biol Sci. 2009 May 7; 276(1662): 1685–1693.
Published online 2009 February 25. doi:  10.1098/rspb.2008.1946
PMCID: PMC2660998

Genetic modulation of energy metabolism in birds through mitochondrial function


Despite their central importance for the evolution of physiological variation, the genetic mechanisms that determine energy expenditure in animals have largely remained unstudied. We used quantitative genetics to confirm that both mass-specific and whole-organism basal metabolic rate (BMR) were heritable in a captive-bred population of stonechats (Saxicola torquata spp.) founded on birds from three wild populations (Europe, Africa and Asia) that differed in BMR. This argues that BMR is at least partially under genetic control by multiple unknown nuclear loci each with a limited effect on the phenotype. We then tested for a genetic effect on BMR based on mitochondrial–nuclear coadaptation using hybrids between ancestral populations with high and low BMR (Europe–Africa and Asia–Europe), with different parental configurations (femalehigh–malelow or femalelow–malehigh) within each combination of populations. Hybrids with different parental configurations have on average identical mixtures of nuclear DNA, but differ in mitochondrial DNA because it is inherited only from the mother. Mass-specific BMR differed between hybrids with different parental configurations, implying that the combination of mitochondrial and nuclear DNA affected metabolic rate. Therefore, our findings implicate mitochondrial function as an important regulator of energy metabolism. In combination with the substantial heritabilities of metabolic rate, and corroborated by genetic differences in the mitochondrial genome, these results set the stage for further investigations of a genetic control mechanism involving both mitochondrial and nuclear genes determining metabolic rate at the whole-organism level.

Keywords: metabolic rate, heritability, selection, mitochondrial DNA, intergenomic coadaptation, mitochondrial respiration

1. Introduction

The genetic mechanisms that determine energy expenditure in animals are largely unknown, despite the central importance of metabolic rate for behaviour, ecology and life history of organisms. Some studies report a heritable component (Lacy & Lynch 1979; Sadowska et al. 2005; Rønning et al. 2007) to basal metabolic rate (BMR), the minimum rate of metabolism measured at rest, in the rest phase and in the thermal neutral zone (King 1974), whereas others do not (Dohm et al. 2001; Nespolo et al. 2003, 2005), and few demonstrate a direct response to selection on BMR (Ksiazek et al. 2004). These results imply that at least some within-species variation in BMR has a genetic basis, although the genetic mechanisms remain unclear. Recent analyses of heritabilities and genetic correlations among body mass, whole-organism BMR and mass-specific BMR in three populations of stonechats (Saxicola torquata spp.) have showed that these traits can evolve independently (B. I. Tieleman, M. A. Versteegh, B. Helm & N. J. Dingemanse 2006, unpublished data). These results raise the possibility that metabolic rate is influenced by multiple genes.

Since BMR, as measured by oxygen consumption, is a product of summed mitochondrial oxygen consumption, mitochondrial function provides an important link between cellular processes and whole-organism metabolic rate, and, in addition, implicates a limited set of candidate genes to impart genetic control. After accounting for body size, variation in BMR has been attributed to differences in size of metabolically highly active organs (Daan et al. 1990; Konarzewski & Diamond 1995), mitochondrial density (Else & Hulbert 1985), inner mitochondrial membrane surface area and/or degree of unsaturation of lipids in membranes affecting proton permeability (Hulbert & Else 2005). However, studies that failed to confirm that organ size accounted for variation in BMR (Song & Wang 2002; Tieleman et al. 2003; Russell & Chappell 2007), or that variation in lipids in membranes was associated with BMR (Brookes et al. 1997; Brzek et al. 2007), leave us with a poor understanding of the connections between cellular processes and whole-organism energy expenditure.

Mitochondria consume approximately 90 per cent of respired oxygen during the production of ATP, the molecule that powers such processes as protein synthesis, ion gradients and actinomyosin ATPase (Rolfe & Brown 1997; Nichols & Ferguson 2002). The myriad of chemical reactions that supply electrons to the electron transport system in mitochondria produce heat as a by-product that contributes to thermoregulation in endotherms (Rolfe & Brown 1997; Nichols & Ferguson 2002). Within mitochondria, the electron transport system consists of large protein complexes (complexes I–V) that sequentially transport electrons and create a proton-motive force that drives ATP synthesis. The electron transport system contains 13 subunits coded by the mitochondrial genome, in addition to at least 70 nuclear-encoded peptides that are imported into the mitochondria and assembled together with the mitochondrially encoded subunits to produce functional complexes (Blier et al. 2001; Ballard & Rand 2005; Das 2006). Proteins containing mitochondrial DNA-encoded polypeptides play crucial roles in the regulation of mitochondrial respiration, in complexes I, III and IV, and in the production of ATP in complex V (Davey & Clark 1996; Bai et al. 2000).

Maternally inherited, mitochondrial DNA has traditionally been used as neutral marker in phylogeographic studies (Zink & Barrowclough 2008), but an increasing number of studies support the idea that climatic conditions can provide a selective ecological gradient that alters mitochondria haplotype, potentially changing mitochondria function and heat production (Matsuura et al. 1993; Mishmar et al. 2003; Ballard & Whitlock 2004; Ruiz-Pesini et al. 2004; Ballard & Rand 2005; Zink 2005; but see Elson et al. 2004). Furthermore, Fontanillas et al. (2005) showed a relationship between mitochondrial haplotype and maximal non-shivering thermogenesis in white-toothed shrews from a lowland and an upland population, suggesting selection on mitochondria in these two populations.

We exploited differences in inheritance of mitochondrial and nuclear genes to obtain insights into the mechanisms of genetic influence on metabolic rate. Using a captive-bred population of stonechats (S. torquata spp.) founded on birds from three distinct wild populations with different BMRs (from, in decreasing BMR order, Asia, Europe and Africa; Klaassen 1995; Wikelski et al. 2003; Tieleman 2007), we first applied quantitative genetics analyses to evaluate the heritable basis of BMR in our dataset of combined inter- and intrapopulation variations. The heritabilities and genetic correlations are compared with previous analyses for each of the three ancestral populations separately (B. I. Tieleman, M. A. Versteegh, B. Helm & N. J. Dingemanse 2006, unpublished data). We then tested three hypotheses about mechanisms of genetic influence on BMR using two types of hybrids, European–African and Asian–European crosses, with reciprocal parental configurations, Europefemale–Africamale versus Africafemale–Europemale and Asiafemale–Europemale versus Europefemale–Asiamale. A priori, we envisioned three possible outcomes: (i) No difference in metabolic rate between hybrids with contrasting parental configurations and hybrid metabolic rates intermediate between ancestral values. This would indicate that metabolic rate results predominantly from nuclear DNA-encoded processes (‘nuclear control’ hypothesis). (ii) Metabolic rate differs between hybrids with contrasting parental configurations and resembles the mother's metabolic rate, pointing to dominance of the mitochondrial DNA-encoded components of the electron transport system in determining metabolism (‘mitochondrial control’ hypothesis). (iii) Metabolic rate differs between hybrids with contrasting parental configurations and does not resemble the mother's metabolic rate, providing evidence for the importance of a match between mitochondrial and nuclear genomes to regulate metabolic rate (‘mitochondrial–nuclear coadaptation’ hypothesis; Rand et al. 2004). Next, we added quantitative genetics analyses to explore maternal and paternal effects on metabolic rate. Finally, we compared sequences of all mitochondrial protein-coding genes between the European and African populations to confirm the presence of amino acid-changing genetic differences in the mitochondrial genome.

2. Material and methods

(a) Birds and housing

Stonechats are small passerines with a wide geographical distribution in Europe, Africa and Asia, ranging from 71° N to 35° S (Urquhart 2002). We studied stonechats from three different populations belonging to three different subspecies, from Europe (Austria, Saxicola torquata rubicola), Africa (Kenya, Saxicola torquata axillaris) and Asia (Kazakhstan, Saxicola torquata maura), in addition to European–African hybrids and European–Asian hybrids. The phylogenetic relationships among these populations and subspecies have been described in detail in Illera et al. (2008). Although our birds came from geographically separate locations, they produce fertile hybrids when interbreeding in captivity. Our study design provided us with two independent tests of the effect of the parental configuration within a lineage on mass-specific BMR of hybrids. For the European–African combination, we measured 80 individuals of four different genotypic pairings, comprising 45 pure Europeans, 14 pure Africans, 11 hybrids with European mothers and 10 hybrids with African mothers. For the study of Europeans and Asians, we measured 29 individuals in addition to the 45 pure Europeans: 15 pure Asians, 5 hybrids with European mothers and 9 hybrids with Asian mothers. Because birds that we measured varied in age from 8 months to 8 years, we incorporated age as a factor in the analyses. Individuals were kept in separate cages, in rooms with controlled temperature (20–22°C) and day length as experienced by European birds in the winter quarters (40° N). We measured each individual one to three times during winter (total 131 measurements), in the periods 9–13 February 2005 or 24 November 2005–13 February 2006. During these periods none of the birds were breeding, moulting or showing migratory restlessness, the latter verified by recordings of night-time activity (Helm et al. 2005).

(b) Metabolic rates

(i) Laboratory set-up and metabolic measurements

One hour before the start of experiments, we removed water and food from cages to ensure birds were post-absorptive (B. I. Tieleman 2005, personal observation). We placed birds in 13.5 l metal metabolic chambers with Plexiglas lids that closed airtight. In the metabolic chambers, birds perched on a wire mesh platform placed over a layer of paraffin oil. We set the metabolic chambers inside a climatic chamber with a constant temperature of 35°C±0.5°C, a temperature within the thermoneutral zone of European stonechats (Tieleman 2007).

We used standard flow through respirometry to measure O2 consumption of stonechats (Gessaman 1987). Compressed air was pumped through columns of silica gel–soda lime–silica gel to remove water and CO2, through calibrated (Levy 1964) mass-flow controllers (model 5850E, Brooks, Hatfield, PA, USA) set at 500 ml min−1, and then into the chambers.

Air that exited the chamber passed through columns of silica gel–ascarite–silica gel, before entering an oxygen analyser (Oxzilla, Sable Systems, Las Vegas, NV, USA) to measure the fractional concentration of oxygen of the air stream. A stream of dry CO2-free air was used as reference air. We recorded atmospheric pressure, air temperature inside each metabolic chamber and delta O2 with a data logger (model CR23X, Campbell Scientific, Logan, UT, USA).

At least 3 hours after birds were placed in the metabolic chambers, and when O2 concentration had been stable for at least 10 min, we calculated the lowest running average over 10 min of data for BMR (Tieleman 2007). To calculate BMR from O2 consumption, we used eqn (2) from Hill (1972) using a conversion factor of 20.1 J ml−1 O2 (Schmidt-Nielsen 1997). Birds were weighed immediately before and after metabolic measurements, and average weights are used in analyses.

(ii) Statistical analysis of metabolic rates

We analysed data for BMR from the European–African hybrids and the European–Asian hybrids separately. For each dataset, we divided birds into four groups—pure Europeans, hybrids with European mothers, hybrids with African (or Asian) mothers and pure Africans (or Asians)—and tested differences in BMR among groups. We used a mixed model with individual entered as random effect to incorporate multiple measurements on one-third of the individuals (using MLwiN v. 2.02). Age, sex and body mass were included as fixed effects, and we used backward elimination of non-significant fixed effects (p>0.05) as our selection procedure (Crawley 1993). We tested all two-way interactions but do not report non-significant results. Age had no significant effect in any of the models (all p>0.29); we therefore omitted age from the models.

(c) Calculation and interpretation of mass-specific BMR

We calculated mass-specific BMR as whole-organism BMR divided by body mass. We interpreted this value as the product of the biochemical processes of an average gram of tissue, i.e. average tissue-specific metabolic rate (Speakman 2005; Tieleman et al. 2006). We assume that body composition and mitochondrial density are the same in the different groups of stonechat hybrids, and interpret differences in mass-specific BMR to result from differences in mitochondrial function. We find it unlikely that the parental configuration of the hybrids would affect body composition or mitochondrial density, but emphasize that we have no data on body composition or mitochondrial density. Hence, these assumptions remain to be tested.

(d) Quantitative genetics analysis with animal models

(i) Pedigrees

Wild-caught stonechats have been added to our captive population over a 15-year period. For our quantitative genetics analyses, we assumed that wild-caught nestlings collected from different nests were not related to each other, and that nestlings collected from the same nest had the same genetic father and mother. In captivity, stonechats bred in aviaries that contained a single pair, thus for all offspring the genetic father and mother were known. After hatching, chicks were fed by the mother for approximately 5 days and thereafter hand-raised. We had a panmictic population of stonechats with known relationships among birds, including crosses among generations and among populations. The resulting pedigree included 50 African, 127 European and 53 Asian individuals, in addition to 21 European–African hybrids and 14 European–Asian hybrids. The birds that we measured belonged to generations 0 (wild-caught) to 4, with the largest numbers from generations 1 and 2. To indicate the genetic basis for variation within the captive populations, we calculated the number of individuals from the wild that produced the first generation of captive individuals: 20 African, 43 European and 18 Asian stonechats (Lynch & Walsh 1998).

(ii) Estimation of genetic parameters and statistical analysis

We estimated variance components, heritabilities and genetic correlations with restricted maximum-likelihood ‘animal models’ (Falconer & Mackay 1996; Kruuk 2004) using the program ASReml v. 2.0 (Gilmour et al. 2006). We used these animal models to partition the total phenotypic variance (VP) of a trait into additive genetic variance (VA) and residual variance (VR). The model included sex as a fixed effect, accounting for differences between sexes in phenotypic means, and animal as random effect, taking into account repeated measures. Heritability (h2) was calculated as VA/VP (Falconer & Mackay 1996). We used bivariate animal models to estimate additive and residual covariances and correlations (r). To test whether VA was significantly larger than zero, we compared the −2×log likelihood of a model that estimated VA with one that did not. To test whether the estimates of rA were significantly different from zero (or from 1 or −1), we compared the −2×log likelihood of models with VA fixed, and rA either fixed at zero (or 1 or −1) or unconstrained. The corresponding significance levels were taken from a Χ2 distribution. To allow comparisons of additive genetic and residual variances across traits, we calculated coefficients of additive genetic variance (CVA) and residual variance (CVR) following Houle (1992).

The maternal inheritance of mitochondria might lead to a maternal genetic effect on offspring metabolism (mitochondrial control hypothesis). To estimate the maternal genetic effects, we constructed an animal model that included mother as additional random effect with the associated variance–covariance matrix determined by the additive genetic relatedness matrix (Kruuk 2004). Likewise, we tested maternal environmental effects (by fitting maternal identity as random effect), paternal genetic and environmental effects, and permanent environment effects (following Kruuk 2004). We calculated the Χ2-distributed difference between the −2×log likelihoods of a model with the effect and one without to establish statistical significance.

(e) Mitochondrial DNA sequences

(i) Generation of mtDNA sequences

DNA was isolated from muscle tissue from two African and two European individuals using a standard phenol–chloroform procedure. Sequences for all coding genes were obtained by amplifying gene regions in roughly 500 bp sections using primers designed by M. Sorenson for sequencing mtDNA in birds (described at, supplemented with additional primers designed by O. Haddrath and A. J. Baker. Both strands of each amplicon were sequenced using a BigDye (v. 3.0) Cycle Sequencing Kit (Applied Biosystems). Cycle-sequenced products were cleaned using Sephadex columns and nucleotide sequences determined using an ABI 3100 genetic analyser.

(ii) Analyses of mitochondrial DNA

Because we were interested in differences among populations in mitochondrial proteins as functional components in the electron transport system, we focused on a combined analysis of genes that contribute proteins to four of the five metabolic complexes in the electron transport system (Davey & Clark 1996; Korzeniewski & Mazat 1996; Bai et al. 2000): complex I (NADH 1–6); complex III (cytochrome b); complex IV (COI–III); and complex V (ATP6 & 8). We explored whether the sequence differences between the European and African populations resulted in (possibly functional) amino acid replacements or in non-functional, synonymous, variation. We present results for individual mitochondrial genes in appendix 1 of the electronic supplementary material.

Finally, we looked for evidence of selection leading to divergence in mitochondrial DNA among stonechat populations, using McDonald–Kreitman (MK) tests (McDonald & Kreitman 1991) to compare the ratio of amino acid replacement variation to synonymous variation within species with the ratio of replacement to synonymous divergence between taxa. Under neutrality these two ratios should be equal, whereas under different types of selection they should differ from one another. The direction and degree of departures from this neutral prediction were quantified with the neutrality index, NI (Rand & Kann 1996). Assuming that synonymous sites are evolving neutrally, NI<1.0 indicates an excess of amino acid divergence, which is inferred to result from positive selection, whereas NI>1.0 indicates an excess of amino acid polymorphism resulting in purifying selection.

3. Results

(a) Heritabilities and genetic correlations of body mass and metabolic rate

Estimates of h2 for mass-specific BMR, whole-organism BMR and body mass varied between 0.36 and 0.70, and all additive genetic variances were significantly larger than zero (table 1). Furthermore, all genetic correlations were significantly different from 1 (or −1 for mass versus mass-specific BMR), indicating that these variables could evolve at least partly independently of each other (table 1).

Table 1
Quantitative genetics parameters for body mass, whole-organism basal metabolic rate (BMR) and mass-specific BMR in stonechats (combined from Europe, Africa and Asia, including hybrids). (a) Estimates of additive genetic variance (VA), residual variance ...

(b) Hybrids with different parental configurations differ in metabolic rate

Mass-specific BMR of hybrid stonechats with different parental configurations differed from each other in both the European–African and the Asian–European pairings (table 2; figure 1). Furthermore, our results confirmed differences in mass-specific BMR, whole-organism BMR and body mass among stonechats of different origin (table 2; figure 1). Analysis of the European–African dataset (figure 1a) showed that mass-specific BMR was significantly affected by group (Χ32=35.88, p<0.0001) and by sex (Χ12=8.76, p=0.003). Post hoc analysis supported the mitochondria–nuclear coadaptation hypothesis and failed to support the mitochondrial control and nuclear control hypotheses: the two hybrid groups differed significantly from each other and from their maternal pure lines (African×European versus African×African, Χ12=19.18, p<0.0001; European×European versus European×African, Χ12=12.27, p=0.0005; European×African versus African×European, Χ12=5.96, p=0.015).

Figure 1
Mass-specific BMR (kJ d−1 g−1) of stonechats with different parental configurations, specified as mother×father. Origin is abbreviated as Eu, Europe; Af, Africa; As, Asia. Bars are s.e.; for sample sizes see table ...
Table 2
Body mass, whole-organism and mass-specific basal metabolic rate (BMR) for four groups of stonechats with different parental configurations, originating from (a) the African and European populations and (b) the European and Asian populations. Values are ...

The analysis of the European and Asian stonechats duplicated the group effects, but not the sex effects, found in the European and African combinations: group significantly affected mass-specific BMR (Χ32=9.56, p=0.023), but sex did not (Χ12=2.80, p=0.09; after deleting the interaction group×sex, Χ32=7.35, p=0.062). Post hoc analysis revealed significant differences between both hybrid types (European×Asian versus Asian×European, Χ12=4.15, p=0.04) and did not support a maternal effect that would indicate mitochondrial control (European×European versus European×Asian, Χ12=4.71, p=0.03; Asian×European versus Asian×Asian, Χ12=3.53, p=0.06).

(c) No maternal and paternal effects on metabolic rate

Maternal (or paternal) effects occur when the phenotype of the mother (or father) affects the offspring phenotype in ways additional to the additive genetic effects, and can be either genetically or environmentally determined, the latter for example through maternal investments in eggs. We tested for a maternal genetic effect in analyses of the mitochondrial control hypothesis. Because visual inspection of results (figure 1) suggested a possible paternal effect, we decided also to test the paternal genetic and environmental effects, maternal environmental effects and permanent environmental effects. None of these effects were significant (table 3).

Table 3
Results of animal model analyses to test for maternal, paternal and permanent environmental effects on mass-specific BMR of a panmictic population of stonechats, with individuals from populations in Europe, Africa and Asia, and from crosses among these ...

(d) Mitochondrial DNA sequences differ between European and African populations

Comparing the mitochondrial DNA of European and African stonechats, we found substantial numbers of amino acid-altering substitutions in genes belonging to each complex, with values ranging from 4 amino acid-altering substitutions for genes in complexes III and IV to 35 for genes in complex I (table 4). Results from the MK tests and values for the NI index indicate that, overall, purifying selection acts on amino acid-altering mutations in stonechat mitochondrial DNA, although our small sample size limits the power of our analysis, particularly for MK tests on individual genes (table 4; appendix 1 in the electronic supplementary material). It is also difficult to determine which mitochondrial gene(s) is/are the direct target(s) of selection based on the gene-by-gene tests (cf. appendix 1 in the electronic supplementary material) owing to the linkage between genes in the non-recombining mitochondrial DNA genome.

Table 4
Numbers of codons, synonymous (S) and non-synonymous (NS) substitutions, neutrality index (NI) values and p values for MK tests for genes, which make up metabolic complexes of mitochondrial subunits. ‘Combined’ refers to a dataset consisting ...

4. Discussion

Hybrids of stonechats from contrasting parental configurations (femalehigh–malelow versus femalelow–malehigh) differed in metabolic rate, verified in two separate datasets with crosses of high and low metabolism populations. These findings provide evidence that the genetic match between mitochondrial DNA and nuclear DNA has consequences for metabolic rate in birds. Hybrids with contrasting parental configurations have on average identical mixtures of nuclear DNA, and differ only in the mitochondrial DNA that they inherited from the mother. Therefore, our findings implicate mitochondrial function as an important regulator of energy metabolism in birds, and are consistent with the mitochondrial–nuclear coadaptation hypothesis. The mitochondrial control hypothesis, with the a priori prediction of a maternal effect, and the nuclear control hypothesis, predicting no differences between hybrids with reciprocal parental configurations, received no support. In combination with the substantial heritabilities of metabolic rate, and corroborated by differences in mitochondrial DNA sequences, these results set the stage for further investigations of the genetic control mechanism determining metabolic rate.

Intergenomic coadaptation between mitochondrial and nuclear DNA, as inferred from our results, has not previously been shown to affect metabolic rate at the whole-organism level in endotherms. Studies on copepods (Tigriopus californicus), however, have shown that interpopulation hybrids, with mismatched mitochondrial and nuclear genomes, have reduced mitochondrial function resulting from the loss of activity of those complexes in the electron transport chain that involve both mitochondrial and nuclear subunits (Ellison & Burton 2006). In these copepods, impaired mitochondrial function negatively affected fitness measured by hatching number, survival and metamorphosis. Studies on fitness consequences for avian hybrids show ambiguous findings, with some indicating effects on migration and breeding, but others not (Avise & Nelson 1989; Helbig et al. 1994; Veen et al. 2007). This ambiguous pattern might be explained if the nature and magnitude of mismatch between mitochondrial and nuclear DNA varies and affects the functional or evolutionary significance in different ways.

Our study was inspired by the possibility that a clear set of candidate genes could affect metabolic rate through mitochondrial function, but our results, especially the resemblance between the hybrids' metabolic rate and that of their father (figure 1), could also be explained by alternative mechanisms. Whereas the animal model analyses do not support paternal and maternal genetic or environmental effects, the sample size is modest and as a result the power of these tests is limited. Alternative explanations fall into three main categories: control by sex chromosomes; genomic imprinting; and non-genetic maternal or paternal effects. Control by sex chromosomes in birds could lead to a sex-specific effect on metabolic rate, with sons resembling fathers, both possessing two Z chromosomes, and daughters resembling mothers, both possessing a set of WZ chromosomes. We find this an unlikely explanation for our results because, although sex has a significant effect in our analyses, in most bird species metabolic rate does not differ among sexes. In our stonechats, the direction of the sex effect differs among populations, with females having a higher mass-specific BMR in the European and African populations, and a lower value in the Asian population, when compared with their male counterparts. Genomic imprinting (Wilkins & Haig 2003; Keverne 2007), the process whereby the gene inherited from one parent dominates over its counterpart from the other parent, can occur on sets of autosomal chromosomes in both female and male offspring or on the set of Z chromosomes in male offspring. Applying these mechanisms to the results in this paper, the paternally inherited genes involved in metabolic rate could dominate over their maternally inherited counterparts, leading to a paternal effect on metabolic rate. Evidence for genomic imprinting in birds is growing, but there are no studies available that point to effects of genomic imprinting on metabolic rate. Finally, non-genetic maternal or paternal effects are possible through a variety of unexplored pathways and cannot be excluded, but no obvious mechanism for such effects in these birds is known at present.

The importance of the genetic match between mitochondrial DNA and nuclear DNA in regulating metabolic rate in hybrid stonechats supplies further evidence that mitochondrial DNA is subjected to selection (e.g. Ballard & Rand 2005; Zink 2005). Our results imply that a coevolutionary response of mitochondrial and nuclear DNA is involved in the selection on mass-specific metabolic rate (see also Blier et al. 2001). This connects with studies that support the ‘thermal adaptation’ hypothesis that climate can be a selective agent that alters mitochondria haplotype, by changing mitochondria function (i.e. ATP production) and heat production (Matsuura et al. 1993; Mishmar et al. 2003; Ruiz-Pesini et al. 2004; Ballard & Rand 2005; Fontanillas et al. 2005; Zink 2005). Under the thermal adaptation hypothesis, organisms living in different thermal environments may be under strong selection through the impact that amino acid-altering substitutions in mitochondrial DNA have on heat production.

Estimates of h2 for mass, whole-organism BMR and mass-specific BMR, and of rA between these traits, are of similar magnitude when based on the current dataset (table 1), which combines intra- and interpopulation variations by the inclusion of hybrids, and on each population separately (B. I. Tieleman, M. A. Versteegh, B. Helm & N. J. Dingemanse 2006, unpublished data). This potentially indicates that within- and among-population variations are largely caused by the same genetic mechanisms. We recognize that our finding of weak genetic correlations between mass, whole-organism BMR and mass-specific BMR is counter-intuitive (table 1), especially in light of the generally significant effects of mass on BMR at the phenotypic level. However counter-intuitive these results might seem, the data indicate that, to a large extent, different suites of genes influence each trait.

In conclusion, we have shown that whole-organism and mass-specific BMRs in stonechats are under genetic control. As implied by the quantitative genetics analyses, metabolism is probably under the influence of multiple genes. The results of the hybrids of reciprocal parental configurations demonstrate that mitochondrial–nuclear coadaptation is at least partly responsible for regulating mass-specific BMR.


We thank W. Jensen, E. Koch, L. Trost, the animal care takers and other staff of the Max Planck Institute for Ornithology for their technical support and animal care. We gratefully acknowledge D. Garant, A. Gilmour, D. Réale and D. Roff for their advice on animal model analyses, J. Diaz and O. Haddrath for their help with laboratory work, and D. Buehler and T. Piersma for their comments on a previous manuscript. This study was financially supported by a grant from The Netherlands Organisation for Scientific Research to B.I.T. (NWO grant no. 863.04.023) and funds from Ohio State University. J.B.W. was supported by NSF grant IBN 0212587. N.J.D. was supported by The Netherlands Organisation for Scientific Research (NWO grant no. 863.05.002).

Supplementary Material

Molecular evolution information per gene:

Numbers of sites analyzed, synonymous (S) and nonsynonymous (NS) substitutions, Neutrality Index (NI) values and p values for MK tests for individual mitochondrial genes. Poly=polymorphic substitutions within populations; Fixed=fixed substitutions between populations.


  • Avise J.C., Nelson W.S. Molecular genetic relationships of the extinct dusky seaside sparrow. Science. 1989;243:646–648. doi:10.1126/science.243.4891.646 [PubMed]
  • Bai Y.D., Shakeley R.M., Attardi G. Tight control of respiration by NADH dehydrogenase ND5 subunit gene expression in mouse mitochondria. Mol. Cell. Biol. 2000;20:805–815. doi:10.1128/MCB.20.3.805-815.2000 [PMC free article] [PubMed]
  • Ballard J.W.O., Rand D.M. The population biology of mitochondrial DNA and its phylogenetic implications. Annu. Rev. Ecol. Evol. Syst. 2005;36:621–642. doi:10.1146/annurev.ecolsys.36.091704.175513
  • Ballard J.W.O., Whitlock M.C. The incomplete natural history of mitochondria. Mol. Ecol. 2004;13:729–744. doi:10.1046/j.1365-294X.2003.02063.x [PubMed]
  • Blier P.U., Dufresne F., Burton R.S. Natural selection and the evolution of mtDNA-encoded peptides: evidence for intergenomic co-adaptation. Trends Genet. 2001;17:400–406. doi:10.1016/S0168-9525(01)02338-1 [PubMed]
  • Brookes P.S., Hulbert A.J., Brand M.D. The proton permeability of liposomes made from mitochondrial inner membrane phospholipids: no effect of fatty acid composition. Biochim. Biophys. Acta Biomembr. 1997;1330:157–164. doi:10.1016/S0005-2736(97)00160-0 [PubMed]
  • Brzek P., Bielawska K., Ksiazek A., Konarzewski M. Anatomic and molecular correlates of divergent selection for basal metabolic rate in laboratory mice. Physiol. Biochem. Zool. 2007;80:491–499. doi:10.1086/520617 [PubMed]
  • Crawley M.J. Blackwell Scientific Publications; Oxford, UK: 1993. GLIM for ecologists.
  • Daan S., Masman D., Groenewold A. Avian basal metabolic rates: their association with body composition and energy expenditure in nature. Am. J. Physiol. 1990;259:R333–R340. [PubMed]
  • Das J. The role of mitochondrial respiration in physiological and evolutionary adaptation. Bioessays. 2006;28:890–901. doi:10.1002/bies.20463 [PubMed]
  • Davey G.P., Clark J.P. Threshold effects and control of oxidative phosphorylation in nonsynaptic rat brain mitochondria. J. Neurochem. 1996;66:1617–1624. [PubMed]
  • Dohm M.R., Hayes J.P., Garland T. The quantitative genetics of maximal and basal rates of oxygen consumption in mice. Genetics. 2001;159:267–277. [PubMed]
  • Ellison C.K., Burton R.S. Disruption of mitochondrial function in interpopulation hybrids of Tigriopus californicus. Evolution. 2006;60:1382–1391. doi:10.1554/06-210.1 [PubMed]
  • Else P.L., Hulbert A.J. An allometric comparison of the mitochondria of mammalian and reptilian tissues—the implications for the evolution of endothermy. J. Comp. Physiol. B Biochem. Syst. Environ. Physiol. 1985;156:3–11. doi:10.1007/BF00692920 [PubMed]
  • Elson J.L., Turnbull D.M., Howell N. Comparative genomics and the evolution of human mitochondrial DNA: assessing the effects of selection. Am. J. Hum. Genet. 2004;74:229–238. doi:10.1086/381505 [PubMed]
  • Falconer D.S., Mackay T.F.C. Longman; New York, NY: 1996. Introduction to quantitative genetics.
  • Fontanillas P., Depraz A., Giorgi M.S., Perrin N. Nonshivering thermogenesis capacity associated to mitochondrial DNA haplotypes and gender in the greater white-toothed shrew, Crocidura russula. Mol. Ecol. 2005;14:661–670. doi:10.1111/j.1365-294X.2004.02414.x [PubMed]
  • Gessaman J.A. Energetics. In: Pendleton B.A., Millsop B.A., Cline K.W., Bird D.M., editors. Raptor management techniques manual. Yale University Press; New Haven, CT: 1987. pp. 289–320.
  • Gilmour A.R., Gogel B.J., Cullis B.R., Welham S.J., Thompson R. VSN International; Hemel Hempstead, UK: 2006. ASreml user guide. Release 1.0.
  • Helbig A.J., Berthold P., MOhr G., Querner U. Inheritance of a novel migratory direction in central European blackcaps. Naturwissenschaften. 1994;81:184–186. doi:10.1007/BF01134540
  • Helm B., Gwinner E., Trost L. Flexible seasonal timing and migratory—behavior results from stonechat breeding programs. Ann. NY Acad. Sci. 2005;1046:216–227. doi:10.1196/annals.1343.019 [PubMed]
  • Hill R.N. Determination of oxygen consumption by use of the paramagnetic oxygen analyzer. J. Appl. Physiol. 1972;33:263. [PubMed]
  • Houle D. Comparing evolvability and variability of quantitative traits. Genetics. 1992;130:195–204. [PubMed]
  • Hulbert A.J., Else P.L. Membranes and the setting of energy demand. J. Exp. Biol. 2005;208:1593–1599. doi:10.1242/jeb.01482 [PubMed]
  • Illera J.C., Richardson D.S., Helm B., Atienza J.C., Emerson B.C. Phylogenetic relationships, biogeography and speciation in the avian genus Saxicola. Mol. Phylogenet. Evol. 2008;48:1145–1154. doi:10.1016/j.ymper.2008.05.016 [PubMed]
  • Keverne E.B. Genomic imprinting and the evolution of sex differences in mammalian reproductive strategies. Adv. Genet. 2007;59:217–243. doi:10.1016/S0065-2660(07)59008-5 [PubMed]
  • King J.R. Seasonal allocation of time and energy resources in birds. In: Paynter R.A. Jr, editor. Avian energetics. Nuttall Ornithological Club; Cambridge, MA: 1974. pp. 4–85.
  • Klaassen M. Moult and basal metabolic costs in males of two subspecies of stonechats: the European Saxicola torquata rubicula and the East African S.t. axillaris. Oecologia. 1995;104:424–432. doi:10.1007/BF00341339
  • Konarzewski M., Diamond J. Evolution of basal metabolic rate and organ masses in laboratory mice. Evolution. 1995;49:1239–1248. doi:10.2307/2410448
  • Korzeniewski B., Mazat J.P. Theoretical studies on the control of oxidative phosphorylation in muscle mitochondria. Application to mitochondrial deficiencies. Biochem. J. 1996;319:143–148. [PubMed]
  • Kruuk L.E.B. Estimating genetic parameters in natural populations using the ‘animal model’ Phil. Trans. R. Soc. Lond. B. 2004;359:873–890. doi:10.1098/rstb.2003.1437 [PMC free article] [PubMed]
  • Ksiazek A., Konarzewski M., Lapo I.B. Anatomic and energetic correlates of divergent selection for basal metabolic rate in laboratory mice. Physiol. Biochem. Zool. 2004;77:890–899. doi:10.1086/425190 [PubMed]
  • Lacy R.C., Lynch C.B. Quantitative genetic analysis of temperature regulation in Mus musculus. I. Partitioning of variance. Genetics. 1979;91:743–753. [PubMed]
  • Levy A. The accuracy of the bubble meter for gas flow measurements. J. Sci. Instrum. 1964;41:449–453. doi:10.1088/0950-7671/41/7/309
  • Lynch M., Walsh J.B. Sinauer Associates, Inc; Sunderland, MA: 1998. Genetics and analysis of quantitative traits.
  • Matsuura E.T., Niki Y., Chigusa S.I. Temperature-dependent selection in the transmission of mitochondrial-DNA in Drosophila. Jpn. J. Genet. 1993;68:127–135. doi:10.1266/jjg.68.127 [PubMed]
  • McDonald J.H., Kreitman M. Adaptive protein evolution at the Adh locus in Drosophila. Nature. 1991;351:652–654. doi:10.1038/351652a0 [PubMed]
  • Mishmar D., et al. Natural selection shaped regional mtDNA variation in humans. Proc. Natl Acad. Sci. USA. 2003;100:171–176. doi:10.1073/pnas.0136972100 [PubMed]
  • Nespolo R.F., Bacigalupe L.D., Bozinovic F. Heritability of energetics in a wild mammal, the leaf-eared mouse (Phyllotis darwini) Evolution. 2003;57:1679–1688. doi:10.1554/02-576 [PubMed]
  • Nespolo R.F., Bustamante D.M., Bacigalupe L.D., Bozinovic F. Quantitative genetics of bioenergetics and growth-related traits in the wild mammal, Phyllotis darwini. Evolution. 2005;59:1829–1837. doi:10.1554/04-408.1 [PubMed]
  • Nichols D.G., Ferguson S.J. Academic Press; London, UK: 2002. Bioenergetics III.
  • Rand D.M., Kann L.M. Excess amino acid polymorphism in mitochondrial DNA: contrasts among genes from Drosophila, mice and humans. Mol. Biol. Evol. 1996;13:735–748. [PubMed]
  • Rand D.M., Haney R.A., Fry A.J. Cytonuclear coevolution: the genomics of cooperation. Trends Ecol. Evol. 2004;19:645–653. doi:10.1016/j.tree.2004.10.003 [PubMed]
  • Rolfe D.F.S., Brown G.C. Cellular energy utilization and molecular origin of standard metabolic rate in mammals. Physiol. Rev. 1997;77:731–758. [PubMed]
  • Rønning B., Jensen H., Moe B., Bech C. Basal metabolic rate: heritability and genetic correlations with morphological traits in the zebra finch. J. Evol. Biol. 2007;20:1815–1822. doi:10.1111/j.1420-9101.2007.01384.x [PubMed]
  • Ruiz-Pesini E., Mishmar D., Brandon M., Procaccio V., Wallace D.C. Effects of purifying and adaptive selection on regional variation in human mtDNA. Science. 2004;303:223–226. doi:10.1126/science.1088434 [PubMed]
  • Russell G.A., Chappell M.A. Is BMR repeatable in deer mice? Organ mass correlates and the effects of cold acclimation and natal altitude. J. Comp. Physiol. B Biochem. Syst. Environ. Physiol. 2007;177:75–87. doi:10.1007/s00360-006-0110-y [PubMed]
  • Sadowska E.T., Labocha M.K., Baliga K., Stanisz A., Wroblewska A.K., Jagusiak W., Koteja P. Genetic correlations between basal and maximum metabolic rates in a wild rodent: consequences for evolution of endothermy. Evolution. 2005;59:672–681. doi:10.1554/04-553 [PubMed]
  • Schmidt-Nielsen K. 5th edn. Cambridge University Press; Cambridge, UK: 1997. Animal physiology: adaptation and environment.
  • Song Z., Wang D. Relationships between metabolic rates and body composition in the Mongolian gerbils (Meriones unguiculatus) Acta Zool. Sin. 2002;48:445–451.
  • Speakman J.R. Body size, energy metabolism and lifespan. J. Exp. Biol. 2005;208:1717–1730. doi:10.1242/jeb.01556 [PubMed]
  • Tieleman B.I. Differences in the physiological responses to temperature among stonechats from three populations reared in a common environment. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 2007;146:194–199. doi:10.1016/j.cbpa.2006.10.011 [PubMed]
  • Tieleman B.I., Williams J.B., Buschur M.E., Brown C.R. Phenotypic variation among and within larks along an aridity gradient: are desert birds more flexible? Ecology. 2003;84:1800–1815. doi:10.1890/0012-9658(2003)084[1800:PVOLAA]2.0.CO;2
  • Tieleman B.I., Dijkstra T.H., Lasky J.R., Mauck R.A., Visser G.H., Williams J.B. Physiological and behavioural correlates of life-history variation: a comparison between tropical and temperate zone house wrens. Funct. Ecol. 2006;20:491–499. doi:10.1111/j.1365-2435.2006.01126.x
  • Urquhart E. Christopher Helm; London, UK: 2002. Stonechats.
  • Veen T., Svedin N., Forsman J.T., Hjernquist M.B., Qvarnström A., Thuman Hjernquist K.A., Träff J., Klaassen M. Does migration of hybrids contribute to post-zygotic isolation in flycatchers? Proc. R. Soc. B. 2007;274:707–712. doi:10.1098/rspb.2006.0058 [PMC free article] [PubMed]
  • Wikelski M., Spinney L., Schelsky W., Scheuerlein A., Gwinner E. Slow pace of life in tropical sedentary birds: a common-garden experiment on four stonechat populations from different latitudes. Proc. R. Soc. Lond. B. 2003;270:2383–2388. doi:10.1098/rspb.2003.2500 [PMC free article] [PubMed]
  • Wilkins J.F., Haig D. What good is genomic imprinting: the function of parent-specific gene expression. Nat. Rev. Genet. 2003;4:359–368. doi:10.1038/nrg1062 [PubMed]
  • Zink R.M. Natural selection on mitochondrial DNA in Parus and its relevance for phylogeographic studies. Proc. R. Soc. B. 2005;272:71–78. doi:10.1098/rspb.2004.2908 [PMC free article] [PubMed]
  • Zink R.M., Barrowclough G.F. Mitochondrial DNA under siege in avian phylogeography. Mol. Ecol. 2008;17:2107–2121. doi:10.1111/j.1365-294X.2008.03737.x [PubMed]

Articles from Proceedings of the Royal Society B: Biological Sciences are provided here courtesy of The Royal Society