Mitochondrial Genotype Affects Fitness in Drosophila simulans

Corresponding author: Avis C. James, 200 Old Biology Bldg., University of Iowa, Iowa City, IA 52242-1324., avis-james{at}uiowa.edu (E-mail)

Communicating editor: S. W. SCHAEFFER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Drosophila simulans is known to harbor three distinct mitochondrial DNA (mtDNA) haplotype groups (siI, -II, and -III) with nearly 3.0% interhaplotypic divergence but <0.06% intrahaplotypic diversity. With the large amount of genetic variation in this system, the potential power to detect intraspecific fitness differences in fly lines that carry distinct haplotypes is great. We test three life-history traits on fly lines with known sequence differences in the mtDNA genome after controlling the nuclear genome by backcrossing. We find that flies with the siI haplotype are fastest developing and have the lowest probability of surviving to three experimental periods (2–6, 12–17, and 34–39 days of age). Wild-type males with siIII mtDNA were more active while disruption of specific coadapted nucleo-mitochondrial complexes caused a significant decrease in activity. These results are discussed in the context of the geographic distribution of each haplotype.

IN humans, more than 50 inherited diseases of metabolism are known to involve mitochondria. Defects in mitochondrial function have also been linked to many of the common diseases of aging, including Alzheimer's dementia, Parkinson's disease, and Huntington's disease (SCHEFFLER 2001 ). Given the demographic realities of the world, where more individuals are living longer and the mean age of the individuals making up the population is steadily increasing, a better understanding of the role mitochondria play in regulating fitness and well-being becomes increasingly critical.

This study tests the fitness of three divergent mitochondrial genotypes in Drosophila, in which evidence has been found for mitochondrial genotype or nucleo-mitochondrial interactions that influence fly fitness traits (HIRAIZUMI 1985 ; CLARK and LYCKEGAARD 1988 ; MACRAE and ANDERSON 1988 ; FOS et al. 1990 ; HUTTER and RAND 1995 ; KILPATRICK and RAND 1995 ; GARCIA-MARTINEZ et al. 1998 ). This report uses Drosophila simulans to investigate mitochondria and fitness. D. simulans is an ideal system to study mitochondrial DNA (mtDNA) fitness effects because it is known to harbor three distinct haplotype groups (siI, -II, and -III) with nearly 3.0% interhaplotypic divergence (SOLIGNAC and MONNEROT 1986 ; SATTA et al. 1987 ; BABA-AISSA et al. 1988 ) but <0.06% intrahaplotypic diversity (BALLARD 2000 ). Flies carrying these three haplotype groups have a distinct geographical subdivision but are not distinct in chosen makers of nuclear DNA [alcohol dehydrogenase related (BALLARD 2000 ), NADH: ubiquinone reductase 75 kD subunit precursor (BALLARD et al. 2002 ), and the period locus (M. D. DEAN, K. J. BALLARD, A. GLASS and J. W. O. BALLARD, unpublished results)]. Furthermore, they cannot be distinguished by genital arch morphology and they mate randomly (BALLARD et al. 2002 ). The geographical subdivision of mitochondrial but not nuclear DNA suggests that cytoplasmic factors influence the distribution of the three mitochondrial haplotypes.

A mitochondrial DNA mutation may be positively selected if it causes a net increase in the fitness of the molecule in a specific local environment. Thus, if there is no recombination, the selective fixation of any mutation in the mtDNA will lead to the fixation of all variants in that genome by genetic hitchhiking (MAYNARD SMITH and HAIGH 1974 ; KAPLAN et al. 1989 ). Patterns of mtDNA variation in D. simulans are not consistent with neutral models of evolution. BALLARD 2000 compared 22 D. simulans and 2 D. melanogaster mitochondrial genomes (minus the A/T-rich region). A maximum-likelihood sliding-window analysis identified regions of the mitochondrial genome with distinct substitution patterns. In addition, Ballard found differences in patterns of polymorphisms (in particular the ratio of nonsynonymous to synonymous substitutions) specific to the distinct lineages, indicating differential selection in the subdivided populations.

Evidence of differential fitness of D. simulans haplotypes comes from micro-injection studies among eggs carrying siI, siII, and siIII mtDNA (DE STORDEUR 1997 ). After micro-injection, the fate of the foreign injected mtDNA was assayed to detect which haplotypes could be accepted by the foreign cells and, if heteroplasmic flies were created, which haplotype went to fixation in the fly line. When the siI haplotype was injected into each of the two other hosts (siII and siIII), heteroplasmic offspring were never detected. When the siII cytoplasm was injected into siI eggs the frequency of siII mtDNA increased in 37 out of 38 lines. When the siII cytoplasm was injected into siIII eggs, the frequency of the siII haplotype tended to increase. When siIII mtDNA was micro-injected into siI flies the frequency of the foreign mtDNA increased in 9 of 10 fly lines and the resident siI mtDNA was lost after three to eight generations. When siIII mtDNA was micro-injected into siII flies the foreign siIII mtDNA was systematically lost in all lines analyzed. Thus the relative fitnesses of the three haplotypes inferred from this assay are siII > siIII > siI.

Additional evidence for fitness differences among D. simulans mtDNA haplotypes comes from a combination of micro-injections and population cage competition studies. NIGRO 1994 included three lines in her study. The first line had the siII mitochondrial DNA, and the second line had siIII mitochondrial DNA. The third line had been created by micro-injection; the siII haplotype had replaced the mitochondria in what was originally a siIII line. In cages where competition was between lines having the original cytonuclear associations, the frequency of the siII mtDNA increased. In cages where the line with siIII mtDNA competed against the transplasmic line, the frequency of the siIII mtDNA increased. These data show that the mtDNA haplotype and the nuclear background in which it resides can influence the frequency of fly haplotypes in competition studies.

While the three haplotypes are not equally fit in microinjection or population cage competition studies, it is quite possible that mitochondrial fitness differences were not causally involved in the divergence of the haplotypes. D. simulans is infected with the -proteobacteria Wolbachia that can induce cytoplasmic incompatibility. In the simplest case incompatibility occurs when an uninfected female mates with an infected male, causing a reduction in the egg hatch rate (see HOFFMANN and TURELLI 1997 for review). D. simulans carries at least four strains of Wolbachia. Sweeps of Wolbachia through populations due to incompatibility will "pull" a mitochondria variant with it, as both the Wolbachia and the mitochondria are maternally inherited. Our laboratory is currently testing the alternate hypotheses that Wolbachia have caused a "pruning" of internal branches in the mtDNA genealogy and that Wolbachia infection caused an increase in the mitochondrial mutation rate.

In this study, we test whether specific known sequence differences in the mtDNA genome of the three distinct D. simulans genotypes confer fitness differences after controlling the nuclear genome by backcrossing. The prediction is that life-history trait differences can be attributed to the mitochondrial genotype only if the effect occurs in flies with both native and introgressed nuclear genomes. Because of the large amount of genetic variation in this system the potential power to distinguish fitness between carriers of these three major haplotypes is considerable. In contrast, the genetic variation in D. melanogaster is nearly an order of magnitude lower than that in D. simulans.

We considered three measures of fitness. First, egg-to-adult development time was measured. Rapid pre-adult development time increases fitness, as these flies develop in ephemeral resources and are vulnerable to deterioration of their food and attack by predators and parasitoids. Second, we measured the activity level of males at three ages. The demonstrated role of mitochondria in the production of 90% of the energy our cells need to function makes activity levels particularly relevant when considering fitness. Third, we measured adult longevity. Specific mitochondrial haplotypes have been found at high frequencies in a population of older humans (DE BENEDICTIS et al. 2000 ), implying that haplotype may influence aging. We find that fly lines that carry the siI haplotype are exceptional in development time and survivorship and that wild-type siIII males are more active.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Construction of introgression lines:
To disrupt any nucleo-mitochondrial gene complexes we introgressed the nuclear background from isofemale fly lines that carry each mitochondrial type into lines that carried each haplotype. The siI (TT01T) line was originally captured in Papeete, Tahiti, in July of 1998. The siII (MD106T) and siIII (RU07T) lines were collected in Antsirabe, Madagascar, and Salazie, Reunion, respectively, in March of 1998. The mitochondrial genomes of these lines (excluding the A + T-rich region) are included in BALLARD 2000 . These lines were chosen because they were recently collected and because they are all close in latitude (18–20° S); thus clinal variation found in a number of species is controlled. All lines were tetracycline treated prior to backcrossing to ensure they were not infected with Wolbachia. Subsequent PCR confirmed all lines were Wolbachia uninfected.

Virgin females from all lines were collected and mated individually to males from each of the other two lines. Virgin F₁ females were collected and mated to males from the same line as their fathers. This was repeated until 12 generations of backcrossing were complete. In theory, 99.975% of the nuclear genome was replaced. However, it is unlikely that this level of replacement was achieved. DERMITZAKIS et al. 2000 crossed D. sechellia males with D. simulans females in 221 independent lines and found that some parts of the D. sechellia genome were never detected in the F₁ hybrids. This could be attributed to incompatibilities of epistatic genes and, in particular, genes that are associated from the D. simulans mitochondrial and the D. sechellia nuclear genomes. Our 12 generations of backcrossing may break coadapted nucleo-mitochondrial gene complexes. Alternatively, it may be that separating such complexes by backcrossing is biologically impossible, making these lines particularly interesting for investigating such genes.

Two independent, replicate introgression lines per nuclear background and mitochondrial haplotype were employed for the experiments described here. The mtDNA of each line was tested and confirmed according to the assay presented at http://www.biology.uiowa.edu/ballard.

Production of experimental animals:
All lines were maintained at 25°. New cultures were started from flies that were <6 days old (HERCUS and HOFFMANN 2000 ). Parents of the experimental animals were placed in "laying pots" and allowed to acclimate for 2–3 days. Egg-laying periods lasted for 4 hr, and 22 hr after the egg lay first instar larvae were collected and placed in fly media vials in groups of 30. Sample sizes and summary statistics of the life-history trait experiments (by mitochondrial haplotype) are listed in Table 1.

Statistical analyses:
In an attempt to show the main effects clearly we analyze these data using a stepwise approach considering a probability level of 0.05 as significant. First we consider only the major difference between these lines, the mitochondrial haplotype (model 1 in Table 2 Table 3 Table 4). This is biologically reasonable because there is no evidence of nuclear subdivision associated with the mitochondrial haplotypes (BALLARD et al. 2002 ). Then we relax this assumption and consider the effects of the mitochondrial haplotype, the nuclear background, and the interaction between the two (model 2 in Table 2 Table 3 Table 4). We consider this important because it is possible that coadapted nucleo-mitochondrial interactions exist.

We assume that the lines we used are representative of the lines within each haplogroup. This is valid, as there is little within-haplogroup divergence [the most divergent haplotype is siI, and BALLARD 2000 found the most dissimilar sequences to differ by 9/14,958 bp] but great between-haplogroup divergence [on average 150/14,958 bp (BALLARD 2000 )]. On the other hand, any significant differences due to nuclear variation could be a general result or a line effect.

Development time:
Times to pupariation and eclosion were recorded every 6 hr on 20 vials per line (but only 9 were collected for one introgression line). Larval development periods were estimated by subtracting the midpoint of the egg-laying period from the average time of pupation for each vial. Egg-to-adult times were determined by subtracting the midpoint of the egg lay from the average eclosion time for each sex. Pupal periods were calculated by subtracting the larval period from the egg-to-adult development time, per vial and sex. Development times were calculated per vial and averaged over each replicate.

Activity:
Virgin males were collected over a 12-hr period. Activity of replicate males was monitored on consecutive days 2–6, 12–17, and 34–39 after virgins were collected.

Activity was monitored with the "Drosophila Activity Monitoring System" (TriKinetics). With this system, flies are put in individual 10-mm tubes with food in one end and cotton in the other. The tubes are placed on a rack that has an infrared beam that bisects the middle of each tube. Computer software counts each time the fly breaks the beam by passing it.

On the day each male was to be tested, they were immobilized with ice and placed in the tubes between 11:30 a.m. and noon. At 2 p.m., the tubes were placed in the rack, and data collection began at 4 p.m. Thus the activity of each male was monitored for 22 hr, from 4 p.m. one day until 2 p.m. the next. On the consecutive days of data collection, tubes with each fly line were rotated (in groups of eight) to offset any bias due to location within the rack.

Data were collected as the total number of movements recorded per male in each monitoring session and analyzed per line over the males/days.

Longevity:
Virgin males that eclosed in a 12-hr period were collected and placed in new vials. Vials were kept at constant temperature (25°). Deaths were recorded between 5 and 7 p.m. every other day, and vials were changed every 4 days.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Development time:
The larval period differed between haplotypes (Table 2, Fig 1A). Larvae harboring the siI haplotype developed faster than those with the siII or siIII haplotype [a posteriori contrasts (t-tests on the least mean squares to detect which pairs are significantly different) of siI comparisons with siII and siIII were t = 4.16 and 5.77, respectively, both P < 0.001, while siII and siIII haplotypes were not significantly different].

View larger version (9K): In this window In a new window Download PPT slide   Figure 1. Development time traits by mitochondrial haplotype (means and 95% confidence limits). (a) Egg to pupariation. (b) Egg to eclosion of males and females. Circles are siI fly lines, triangles are siII lines, and squares are siIII lines. Each haplotype is represented by four introgressed lines and one control line.

In the extended model, the larval period showed significant variation due to the mitochondrial haplotype, the nuclear background, and the interaction between them (Table 2). Flies with the siI haplotype developed faster than those with siII and siIII (a posteriori contrasts, t = 4.11 and 4.18, respectively, both P < 0.001). Flies with siII and siIII did not differ in larval development time (a posteriori contrast, t = 0.49, P = 0.63).

The nuclear effect in the extended model was a result of flies with the RU07T nuclear background developing significantly more quickly than ones with either the MD106T or the TT01T backgrounds (a posteriori contrasts, t = 7.16 and 5.74, respectively, both P < 0.001). Flies with the MD106T and TT01T nuclear background did not differ from each other (a posteriori contrasts, t = 1.46, P = 0.15). This is likely a line-specific result.

Relationships between the siI and siIII haplotypes and their nuclear backgrounds were responsible for the significant interaction term. When the siI haplotype was placed in the nuclear background of a fly line that was originally siIII, the larvae developed significantly faster than the original siI line or lines introgressed with the siII line's nuclear background (a posteriori contrasts: t = 6.11 and 7.44, respectively, P < 0.001 in both cases). In contrast, when the siIII haplotype was placed in the background of a fly line that was originally siI, development was slowed relative to the original siIII line or lines introgressed with the siII line's nuclear background (a posteriori contrasts: t = 5.35 and 4.45, respectively, P < 0.001 in both cases), while there was no difference when the siIII haplotype was placed in the nuclear background of MD106T (a posteriori contrasts: t = 1.71, P = 0.09). There were no differences when the siII haplotype was introgressed.

The pupal period was longer for males than for females (F_{1, 568} = 7990.18, P < 0.001), but there was no difference between haplotypes (F_{2, 568} = 1.39, P = 0.25) or in the interaction between haplotype and sex (F_{2, 568} = 2.24, P = 0.11).

Egg-to-adult development time is simply the larval and pupal periods combined and is not independent of either. Analyses of this trait tell us if the patterns in the larval period are strong enough to be apparent after adding on the pupal times that do not show pattern. Indeed the relationships of the characters show the same trend as the larval period (Fig 1B).

Activity:
Activity measurements of males were not statistically different between haplotypes and age classes or in the interaction between the two (Table 3, model 1). However, a posteriori contrasts of the combined data indicated that siIII flies moved more than siI flies (t = 2.07, P = 0.04); all other contrasts were not significant.

In the extended model, when the nuclear background and associated interaction terms were added, only the nuclear background showed significant differences (Table 3, model 2). The mitochondrial and nuclear effects are almost exclusively the result of low activity of lines with siI mtDNA introgressed into the MD106T nuclear background (Fig 2, a and b). As each of these introgressed lines showed low activity we suggest that these data demonstrate a clear fitness reduction caused by the disruption of coadapted nucleo-mitochondrial complexes rather than by a specific line effect.

View larger version (13K): In this window In a new window Download PPT slide   Figure 2. Activity of males averaged over three time periods (means and 95% confidence limits). (a) By mitochondrial haplotype. Circles are siI fly lines, triangles are siII lines, and squares are siIII lines. Each haplotype is represented by four introgressed lines and one control line. (b) By nuclear background. Circles are lines derived from TT01T, triangles are derived from MD106T, and squares are derived from RU07T. Each nuclear background is represented by four introgressed lines and one control line. (c) Wild-type flies. Circles are three siI fly lines, triangles are three siII lines, and squares are three siIII lines.

To further investigate the activity of males we extended our data by testing two additional recently collected nonintrogressed controls from each haplotype. When data from these lines are included with the data from the three other wild-type lines the activity of males harboring each mtDNA haplotype did not differ significantly (F_{2, 89} = 2.11, P = 0.12). However, males with siIII mtDNA moved more than males with siI mtDNA (a posteriori contrasts: t = 2.04, P = 0.04, Fig 2C); all other contrasts were not significant.

Longevity:
In the longevity study, survivorship curves were significantly different among fly lines with different haplotypes (Fig 3; Wilcoxon test for homogeneity between groups, ² = 15.86, P < 0.001). Comparisons of haplotypes were as follows: siI and siII, ² = 4.65, P = 0.03; siI and siIII, ² = 7.56, P = 0.006; and siII and siIII, ² = 0.67, P = 0.41. Thus, the survivorship curve of siI flies is exceptional.

View larger version (13K): In this window In a new window Download PPT slide   Figure 3. Survivorship curves of fly lines by haplotype. The solid line is siI fly lines, dashed line is siII, and shaded line is siIII. Each haplotype is represented by four introgressed lines and one control line.

We compared the percentage surviving to each of the three age classes in our activity study (2–6, 12–16, and 34–38 days of age). ANOVA on arcsine-transformed data shows that the mitochondrial haplotype and age class influence the percentage surviving (Table 4). The age effect was simply due to the fact that as flies got older, survival decreased. There was no significant haplotype-by-age effect; thus the age classes can be collapsed (Fig 4). Fewer flies with the siI haplotype than fly lines with the siII or siIII haplotypes survived to the time periods (t = 3.83 and 4.12, respectively, P < 0.001 in each case), while fly lines with siII and siIII did not differ from each other (t = 0.26, P = 0.80). The activity data showed that siI flies were least active. One possible explanation for this result is that siI flies were less active because they were physiologically older than siII or siIII flies at each experimental period. This hypothesis is rejected because the activity measurements of males were not statistically different between age classes.

View larger version (10K): In this window In a new window Download PPT slide   Figure 4. Percentage of flies surviving to three age classes by haplotype (means and standard errors). Circles are siI fly lines, triangles are siII lines, and squares are siIII lines. Each haplotype is represented by four introgressed lines and one control line.

Our extended model including the nuclear background shows these haplotypic results were robust (Table 4). There was no effect of the nuclear background, but there was an interesting haplotype-by-nuclear background interaction. Flies from siI lines survive significantly more successfully to our time periods if associated with their own nuclear background than with the nuclear background from the siII or siIII line (a posteriori contrast with siII, t = 3.01, P < 0.01; with siIII, t = 2.22, P < 0.05). Also, the siIII line survives more successfully to our time periods, although not significantly, with its original background than when introgressed with siI (t = 1.45, P = 0.14) or siII (t = 0.70, P = 0.47). Flies with siI and siIII mtDNA have never been collected in sympatry and are the two most genetically divergent haplotypes (BALLARD 2000 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We found significant differences among flies carrying distinct mitochondrial genotypes for both development time and survival. These mitochondrial effects were observed if we considered a simple model testing only the mitochondrial effects and a more complex model that also included the nuclear background, implying nucleo-mitochondrial interactions. Activity of siI males appears to be lower than that of siII or siIII males; however, this result was not robust to a two-way ANOVA including the nuclear genome.

It seems unlikely that Wolbachia infection can fully explain the geographic distribution of the mtDNA haplotypes. In contrast, it seems quite possible that the population subdivision observed in the mitochondrial genome of D. simulans is correlated with the Wolbachia strains that infect this species (SATTA et al. 1988 ; JAMES and BALLARD 2000 ; M. D. DEAN, K. J. BALLARD, A. GLASS and J. W. O. BALLARD, unpublished results). We now consider how the results obtained in this study may reflect the geographic distribution of these flies.

Our results show that flies with the siI haplotype develop faster and live for a shorter period of time than those with the other two haplotypes. Flies with the siI mitochondrial haplotype have been collected on the islands of Hawaii, Tahiti, New Caledonia, and the Seychelles Islands (MERCOT et al. 1995 ; BALLARD 2000 ; JAMES et al. 2002 ). They have not been collected in sympatry with flies harboring siII or siIII mtDNA. It is possible that faster development gives these flies a selective advantage on islands where resources are particularly ephemeral and are vulnerable to deterioration. The breakup of coadapted nuclear-mitochondrial complexes in flies with siI mtDNA may inhibit the invasion of Indian/Pacific Islands by flies harboring siII mtDNA. The activity of males with siI mtDNA introgressed into MD106 flies (formally with siII mtDNA) is significantly reduced when compared to all other lines. Furthermore, flies from siI lines survive more successfully to each of our three experimental periods with their wild-type nuclear background than with the nuclear background of the siII or siIII lines.

The siII haplotype is cosmopolitan, although it has not been collected in Hawaii, Tahiti, or New Caledonia. The siI and siII haplotypes have both been collected in the Seychelles, but not in sympatry. There is nothing exceptional in these data to help explain the wide geographic range of siII. One explanation is that flies with the siII haplotype are tolerant of temperature variations. As noted in MATERIALS AND METHODS, the lines employed here were collected near the equator to control for clinal variation. This was necessary as the siI and siIII lines have not been collected at sites with high latitudes. The transmission rates of Drosophila mtDNA haplotypes in flies made heteroplasmic by micro-injection can be temperature dependent across species (TSUJIMOTO et al. 1991 ; MATSUURA et al. 1993 , MATSUURA et al. 1997 ). Perhaps the relative fitness of flies with different haplotypes is temperature dependent. To address this hypothesis it would be interesting to test a variety of traits at different temperatures. An alternate hypothesis that Wolbachia governs the distribution of the siII haplotype is unlikely. The siII haplotype is associated with two Wolbachia strains, one that consistently induces significant incompatibility (wRi) and one that does not (wAu). If incompatibility were causing the spread of the siII haplotype, it would be expected that all populations would be associated with the virulent strain. In fact, the two Wolbachia strains are equally geographically diverse. Thus, it appears that something about the mitochondrial genotype, and potentially nuclear genes interacting with the haplotype, influences the distribution of siII mtDNA in D. simulans.

The siIII haplotype has been collected from Tanzania, Kenya, Madagascar, and Reunion Island, where it is sympatric with the siII type. At sites where siII and siIII mtDNA have been collected in sympatry the frequency of the siIII type is always 40% (SATTA et al. 1988 ; JAMES and BALLARD 2000 ; M. D. DEAN, K. J. BALLARD, A. GLASS and J. W. O. BALLARD, unpublished results). It is not clear why this is the case. When wild-type fly lines were compared, males with siIII mtDNA were most active. Fly lines with siIII mtDNA were slower developing and had a higher probability of surviving to our time periods than did siI lines, but did not differ in either regard from fly lines with siII mtDNA. Introgression of the siIII haplotype with the siI line produced flies that were slow developing, but no haplotype-by-nuclear DNA interactions were attributed to the siIII haplotype and the nuclear background of the siII line. Competing flies with the siII and siIII haplotype flies in population cages may be an informative method to test the relative fitness of these mitochondrial genomes.

In natural populations that have both the siII and the siIII haplotypes, there is a surprisingly high frequency of heteroplasmic individuals [5.8 and 11.8% reported by MATSUURA et al. 1991 and SATTA et al. 1988 , respectively, on the island of Reunion; and 6.7% in Dar es Salaam, 1.7% in Malindi, Kenya, and 9.6% in Nairobi, Kenya observed by M. D. DEAN, K. J. BALLARD, A. GLASS and J. W. O. BALLARD (unpublished results)]. For heteroplasmy to exist at high frequency, "hybrid" flies should not have reduced fitness. Indeed, we found that the siII/siIII lines with the wild-type nucleo-mitochondrial interactions were no different from each other or lines introgressed between the two. This contrasts with the reduced fitness of males with siI mtDNA in a nuclear background previously harboring siII mtDNA.

In this study, we limited our analyses of activity and longevity to males. As mitochondria are maternally inherited, it is theoretically predicted that negative selection may be greater in males than in females and a stable equilibrium can be maintained. Examples of mitochondria mutations that have a greater effect in males than in females are pointed out by FRANK and HURST 1996 . However, it may be important to look at these components in females, particularly when considering the geographic distribution of mitochondrial haplotypes. Rand et al. 2001 found opposite fitness effects of cytoplasms when in males or females and tested in total fitness competition studies.

D. simulans is a perfect model system for distinguishing differential fitness of mitochondrial variation because it harbors three well-defined haplotypes. The siI haplotype differs from the siII and siIII haplotypes by 39- and 40-amino-acid substitutions, respectively. One or more of these substitutions appears to cause detectable life-history trait differences that influence the geographic subdivision of the species. Further fine tuning of the specific differences that influence fitness will be interesting. Studying specific physiological variations in flies with different haplotypes or nucleo-mitochondrial interactions is of future interest. Products involved in the electron transport chain are potentially significant, as enzymes encoded by both nuclear and mitochondrial genes support this pathway. In particular, oxygen and hydrogen peroxide production (ROSS 2000 ), oxidizing activity (NAGATA and MATSUURA 1991 ), and cytochrome c activity (EDMANDS and BURTON 1998 ) are likely candidates.

	ACKNOWLEDGMENTS

We thank Wayne Johnson (Department of Physiology, University of Iowa) for lending us the "Drosophila Activity Monitoring System" (TriKinetics) and Matt Dean for general assistance. This research was supported by National Science Foundation CAREER grant DEB-9702824 to J. William O. Ballard.

Manuscript received November 5, 2002; Accepted for publication January 16, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BABA-AÏSSA, F., M. SOLIGNAC, N. DENNEBOUY, and J. R. DAVID, 1988  Mitochondrial DNA variability in Drosophila simulans: quasi absence of polymorphism within each of the three cytoplasmic races. Heredity 61:419-426.

BALLARD, J. W. O., 2000  Comparative genomics of mitochondrial DNA in Drosophila simulans.. J. Mol. Evol. 51:64-75.[Medline]

BALLARD, J. W. O., B. CHERNOFF, and A. C. JAMES, 2002  Divergence of mitochondrial DNA is not corroborated by nuclear DNA, morphology, or behavior in Drosophila simulans.. Evolution 56:527-545.[Medline]

CLARK, A. G. and E. M. LYCKEGAARD, 1988  Natural selection with nuclear and cytoplasmic transmission. III. Joint analysis of segregation and mtDNA in Drosophila melanogaster. Genetics 118:471-481.[Abstract/Free Full Text]

DE BENEDICTIS, G., G. CARRIERI, O. VARCASIA, M. BONAFE, and C. FRANCESCHI, 2000  Inherited variability of the mitochondrial genome and successful aging in humans. Ann. NY Acad. Sci. 908:208-218.[Medline]

DERMITZAKIS, E. T., J. P. MASLY, H. M. WALDRIP, and A. G. CLARK, 2000  Non-Mendelian segregation of sex chromosomes in heterospecific Drosophila males. Genetics 154:687-694.[Abstract/Free Full Text]

DE STORDEUR, E., 1997  Nonrandom partition of mitochondria in heteroplasmic Drosophila.. Heredity 79:615-623.

EDMANDS, S. and R. BURTON, 1998  Variation in cytochrome-c oxidase activity is not maternally inherited in the copeod Tigriopus californicus.. Heredity 80:668-674.

FOS, M., M. A. DOMÍNGUEZ, A. LATORRE, and A. MOYA, 1990  Mitochondrial DNA evolution in experimental populations of Drosophila subobscura. Proc. Natl. Acad. Sci. USA 87:4198-4201.[Abstract/Free Full Text]

FRANK, S. A. and L. D. HURST, 1996  Mitochondria and male disease. Nature 383:224.[Medline]

GARCÍA-MARTÍNEZ, J., J. A. CASTRO, M. RAMÓN, A. LATORRE, and A. MOYA, 1998  Mitochondrial DNA haplotype frequencies in natural and experimental populations of Drosophila subobscura.. Genetics 149:1377-1382.[Abstract/Free Full Text]

HERCUS, M. J. and A. A. HOFFMANN, 2000  Maternal and grandmaternal age influence offspring fitness in Drosophila.. Proc. R. Soc. Lond. Ser. B Biol. Sci. 267:2105-2110.[Medline]

HIRAIZUMI, Y., 1985  Genetics of factors affecting the life history of Drosophila melanogaster. I. Female productivity. Genetics 110:452-464.

HOFFMANN, A. A., and M. TURELLI, 1997 Cytoplasmic incompatibility in insects, pp. 42–80 in Influential Passengers: Inherited Microorganisms and Invertebrate Reproduction, edited by S. L. O'NEILL, A. A. HOFFMANN and J. H. WERREN. Oxford University Press, Oxford.

HUTTER, C. M. and D. M. RAND, 1995  Competition between mitochondrial haplotypes in distinct nuclear genetic environments: Drosophila pseudoobscura vs. D. persimilis. Genetics 140:537-548.[Abstract]

JAMES, A. C. and J. W. O. BALLARD, 2000  Expression of cytoplasmic incompatibility in Drosophila simulans and its impact on infection frequencies and distribution of Wolbachia pipientis.. Evolution 54:1661-1672.[Medline]

JAMES, A. C., M. D. DEAN, M. E. MCMAHON, and J. W. O. BALLARD, 2002  Dynamics of double and single Wolbachia infections in Drosophila simulans from New Caledonia. Heredity 88:182-189.[Medline]

KAPLAN, N. L., R. R. HUDSON, and C. H. LANGLEY, 1989  The "hitchhiking effect" revisited. Genetics 123:887-899.[Abstract/Free Full Text]

KILPATRICK, S. T. and D. M. RAND, 1995  Conditional hitchhiking of mitochondrial DNA: frequency shifts of Drosophila melanogaster mtDNA variants depend on nuclear genetic background. Genetics 141:1113-1124.[Abstract]

MACRAE, A. F. and W. W. ANDERSON, 1988  Evidence for non-neutrality of mitochondrial DNA haplotypes in Drosophila pseudoobscura.. Genetics 120:485-494.[Abstract/Free Full Text]

MATSUURA, E. T., H. FUKUDA, and S. I. CHIGUSA, 1991  Mitochondrial DNA heteroplasmy maintained in natural populations of Drosophila simulans in Reunion. Genet. Res. 57:123-126.[Medline]

MATSUURA, E. T., Y. NIKI, and S. I. CHIGUSA, 1993  Temperature-dependent selection in the transmission of mitochondrial DNA in Drosophila.. Jpn. J. Genet. 68:127-135.[Medline]

MATSUURA, E. T., Y. T. TANAKA, and N. YAMAMOTO, 1997  Effects of the nuclear genome on selective transmission of mitochondrial DNA in Drosophila.. Genes Genet. Syst. 72:119-123.[Medline]

MAYNARD SMITH, J. and J. HAIGH, 1974  The hitchhicking effect of a favourable gene. Genet. Res. 23:23-35.[Medline]

MERÇOT, H., B. LLORENTE, M. JACQUES, A. ATLAN, and C. MONTCHAMP-MOREAU, 1995  Variability with the Seychelles cytoplasmic incompatibility system in Drosophila simulans.. Genetics 141:1015-1023.[Abstract]

NAGATA, Y. and E. T. MATSUURA, 1991  Temperature-dependency of electron-transport activity in mitochondria with exogenous mitochondrial DNA in Drosophila.. Jpn. J. Genet. 66:255-261.[Medline]

NIGRO, L., 1994  Nuclear background affects frequency dynamics of mitochondrial DNA variants in Drosophila simulans.. Heredity 72:582-586.

RAND, D. M., A. G. CLARK, and L. M. KANN, 2001  Sexually antagonistic cytonuclear fitness interactions in Drosophila melanogaster.. Genetics 159:173-187.[Abstract/Free Full Text]

ROSS, R. E., 2000  Age-specific decrease in aerobic efficiency associated with increase in oxygen free radical production in Drosophila melanogaster.. J. Insect Physiol. 46:1477-1480.[Medline]

SATTA, Y., H. ISHIWA, and S. I. CHIGUSA, 1987  Analysis of nucleotide substitutions of mitochondrial DNAs in Drosophila melanogaster and its sibling species. Mol. Biol. Evol. 4:638-650.[Abstract]

SATTA, Y., N. TOYOHARA, C. OHTAKA, Y. TATSUNO, and K. WATANABE et al., 1988  Dubious maternal inheritance of mitochondrial DNA in D. simulans and evolution of D. mauritiana.. Mol. Gen. Genet. 52:1-6.

SCHEFFLER, I. E., 2001  A century of mitochondrial research: achievements and perspectives. Mitochondrion 1:3-31.[Medline]

SOLIGNAC, M. and M. MONNEROT, 1986  Race formation and introgression within Drosophila simulans, D. mauritiana and D. sechellia inferred from mitochondrial DNA analysis. Evolution 40:531-539.

TSUJIMOTO, Y., Y. NIKI, and E. T. MATSUURA, 1991  Further study on selective transmission of mitochondrial DNA in heteroplasmic lines of Drosophila melanogaster.. Jpn. J. Genet. 66:609-616.[Medline]

This article has been cited by other articles:

				B. S. Ort and G. H. Pogson Molecular Population Genetics of the Male and Female Mitochondrial DNA Molecules of the California Sea Mussel, Mytilus californianus Genetics, October 1, 2007; 177(2): 1087 - 1099. [Abstract] [Full Text] [PDF]

				D. K. Dowling, U. Friberg, F. Hailer, and G. Arnqvist Intergenomic Epistasis for Fitness: Within-Population Interactions Between Cytoplasmic and Nuclear Genes in Drosophila melanogaster Genetics, January 1, 2007; 175(1): 235 - 244. [Abstract] [Full Text] [PDF]

				C. S. Willett Deleterious Epistatic Interactions Between Electron Transport System Protein-Coding Loci in the Copepod Tigriopus californicus Genetics, July 1, 2006; 173(3): 1465 - 1477. [Abstract] [Full Text] [PDF]

				D. M. Rand, A. Fry, and L. Sheldahl Nuclear-Mitochondrial Epistasis and Drosophila Aging: Introgression of Drosophila simulans mtDNA Modifies Longevity in D. melanogaster Nuclear Backgrounds Genetics, January 1, 2006; 172(1): 329 - 341. [Abstract] [Full Text] [PDF]

				J. W. O. Ballard Sequential Evolution of a Symbiont Inferred From the Host: Wolbachia and Drosophila simulans Mol. Biol. Evol., March 1, 2004; 21(3): 428 - 442. [Abstract] [Full Text] [PDF]