Comparative population genomics of latitudinal variation in Drosophila simulans and Drosophila melanogaster

Examples of clinal variation in phenotypes and genotypes across latitudinal transects have served as important models for understanding how spatially varying selection and demographic forces shape variation within species. Here, we examine the selective and demographic contributions to latitudinal variation through the largest comparative genomic study to date of Drosophila simulans and Drosophila melanogaster, with genomic sequence data from 382 individual fruit flies, collected across a spatial transect of 19 degrees latitude and at multiple time points over 2 years. Consistent with phenotypic studies, we find less clinal variation in D. simulans than D. melanogaster, particularly for the autosomes. Moreover, we find that clinally varying loci in D. simulans are less stable over multiple years than comparable clines in D. melanogaster. D. simulans shows a significantly weaker pattern of isolation by distance than D. melanogaster and we find evidence for a stronger contribution of migration to D. simulans population genetic structure. While population bottlenecks and migration can plausibly explain the differences in stability of clinal variation between the two species, we also observe a significant enrichment of shared clinal genes, suggesting that the selective forces associated with climate are acting on the same genes and phenotypes in D. simulans and D. melanogaster.     

Nucleotide variation at the runt locus in Drosophila melanogaster and Drosophila simulans.

Intra- and interspecific nucleotide variation for the major developmental gene runt in Drosophila was studied in D. melanogaster and D. simulans. The 1.5-kb protein-coding region and the 0.4-kb intron of the runt gene were sequenced for 11 alleles in each species. The D. melanogaster alleles originated from east Africa. Estimated parameters of intraspecific variation in D. melanogaster (exons: theta = 0.018, pi = 0.018; intron: theta = 0.014, pi = 0.014) and D. simulans (exons: theta = 0.007, pi = 0.005; intron: theta = 0.008, pi = 0.005) were below average for other X-linked genes, while divergence between species (exons: D = 0.094; intron: D = 0.069) fell within the normal range for both silent and replacement changes. This estimate for runt, along with published values for three other genes in regions of normal recombination, show east African D. melanogaster to be roughly twice as polymorphic as D. simulans. The majority of nucleotide variation, silent and replacement, in both species was found to be selectively neutral using various statistical tests (HKA, McDonald-Kreitman, Tajima, and Fu and Li tests). Monte Carlo simulations of the coalescent process significantly rejected a Wright-Fisher model with respect to an amino acid polymorphism and the distribution of polymorphic sites among the D. simulans lines. This indicated an old lineage and may reflect ancestral population substructuring in D. simulans.     

Microsatellite variation and differentiation in African and non-African populations of Drosophila simulans

Drosophila simulans originated in sub-Saharan Africa or Madagascar and colonized the rest of the world after the last glaciation about 10 000 years ago. Consistent with this demographic history, sub-Saharan African populations have been shown to harbour higher levels of microsatellite and sequence variation than cosmopolitan populations. Nevertheless, only limited information is available on the population structure of D. simulans. Here, we analysed X-linked and autosomal microsatellite loci in four sub-Saharan African, one North African, one Israeli, and two European D. simulans populations. Bayesian clustering algorithms combined the North African, Israeli, and European populations into a single cosmopolitan group. The four sub-Saharan populations were split into two separate groups. Pairwise F(ST) analysis, however, indicated significant population differentiation between all eight populations surveyed. A significant signal for population reduction in cosmopolitan populations was found only for X-linked loci.     

Contrasting patterns of X-linked and autosomal nucleotide variation in Drosophila melanogaster and Drosophila simulans

Surveys of molecular variation in Drosophila melanogaster and Drosophila simulans have suggested that diversity outside of Africa is a subset of that within Africa. It has been argued that reduced levels of diversity in non-African populations reflect a population bottleneck, adaptation to temperate climates, or both. Here, I summarize the available single-nucleotide polymorphism data for both species. A simple "out of Africa" bottleneck scenario is consistent with geographic patterns for loci on the X chromosome but not with loci on the autosomes. Interestingly, there is a trend toward lower nucleotide diversity on the X chromosome relative to autosomes in non-African populations of D. melanogaster, but the opposite trend is seen in African populations. In African populations, autosomal inversion polymorphisms in D. melanogaster may contribute to reduced autosome diversity relative to the X chromosome. To elucidate the role that selection might play in shaping patterns of variability, I present a summary of within- and between-species patterns of synonymous and replacement variation in both species. Overall, D. melanogaster autosomes harbor an excess of amino acid replacement polymorphisms relative to D. simulans. Interestingly, range expansion from Africa appears to have had little effect on synonymous-to-replacement polymorphism ratios.     

Contrasting patterns of geographic variation in the cosmopolitan sibling species Drosophila melanogaster and Drosophila simulans

An electrophoretic study was carried out to compare the geographic pattern of genetic variation in Drosophila simulans with that of its sibling species, Drosophila melanogaster. An identical set of 32 gene-protein loci was studied in four geographically distant populations of D. simulans and two populations of D. melanogaster, all originating from Europe and Africa. The comparison yielded the following results: (1) tropical populations of D. simulans were, in terms of the number of unique alleles, average heterozygosity per locus, and percentage of loci polymorphic, more variable than conspecific-temperate populations; (2) some loci in both species showed interpopulation differences in allele frequencies that suggest latitudinal clines; and (3) temperate-tropical genetic differentiation between populations was much less in D. simulans than in D. melanogaster. Similar differences between these two species have previously been shown for chromosomal, quantitative, physiological, and middle-repetitive DNA variation. Estimates of N _m (number of migrants per generation) from the spatial distribution of rare alleles suggest that both species have similar levels of interpopulation gene flow. These observations lead us to propose two competing hypotheses: the low level of geographic differentiation in D. simulans is due to its evolutionarily recent worldwide colonization and, alternatively, D. simulans has a narrower niche than D. melanogaster. Geographic variation data on different genetic elements (e.g., mitochondrial DNA, two-dimensional proteins, etc.) are required before these hypotheses can be adequately tested.We thank the Natural Science and Engineering Research Council of Canada for financial support (Grant A0235 to R.S.S.).     

Nucleotide variation in the triosephosphate isomerase (Tpi) locus of Drosophila melanogaster and Drosophila simulans

DNA sequence variation in a 1.1-kb region including the coding portion of the Tpi locus was examined in 25 homozygous third-chromosome lines of Drosophila melanogaster, nine lines of Drosophila simulans, and one line of Drosophila yakuba. Our data show that the widespread allozyme polymorphism observed in cosmopolitan D. melanogaster is due to a glutamic acid substitution occurring in a phylogenetically conserved lysine that has been identified as part of the "hinged-lid" active site of the enzyme. This observation suggests that the replacement polymorphism may have important functional consequences. One replacement polymorphism was also observed in D. simulans, although its functional relevance is more difficult to assess, since it affects a site that is not strongly conserved. This amino acid change in D. simulans is associated with a single lineage possessing seven unique silent substitutions, which may be indicative of balancing selection or population subdivision. The absence of fixed amino acid differences between D. melanogaster and D. simulans and only a single difference with D. yakuba suggests that triose phosphate isomerase is under strong functional constraint. Silent variation is slightly higher for D. melanogaster than for D. simulans. Finally, we outline the general lack of evidence for old balanced polymorphisms at allozyme loci in D. melanogaster.      

Microsatellite variation in North American populations of Drosophila melanogaster.

Computer database searching for microsatellites can be particularly effective for organisms like Drosophila melanogaster for which there are extensive sequence data. Here we demonstrate that 17 out of 18 such microsatellites are also highly polymorphic in natural populations of Drosophila, and that this variation is easily scorable with PCR followed by electrophoresis on high-resolution agarose. This form of variation is likely to be of great value in studies of the genomic distribution of polymorphism, population structure, the relation between intraspecific polymorphism and interspecific divergence and the mutation rate and pattern of mutations of microsatellites. In this preliminary survey of 15 lines, we find that the variance in repeat count is most strongly correlated with the maximum count, that perfect repeats are significantly more variable than imperfect repeats and that repeats which are split by an imperfection have unexpectedly low variance given the size of the perfectly repeated portion.     

Low genic variation in male-reproductive-tract proteins of Drosophila melanogaster and D. simulans

We report results, using two-dimensional gel electrophoresis (2DE), of natural population surveys of allelic variation in approximately 300 male-reproductive-tract polypeptides in both Drosophila melanogaster and its sibling species, D. simulans. Despite our efforts to maximize operational sensitivity of our 2DE gels to polymorphism, variation estimates in both species were low (proportion of polymorphic loci [P] = 9%, and average heterozygosity [H] = 1%-3%), compared with those by one-dimensional gel electrophoresis (1DE) (P = 29%-55%; H = 8%-19%) in the same populations. However, H of polymorphic loci was very similar for 2DE and 1DE proteins; and for 17 of a total of 54 polymorphic proteins, 2DE detected three or four distinct alleles. The results suggest that the differing levels of variability widely seen with 1DE and 2DE are real and reflect differing intensities of functional constraint between different classes of structural loci. However, the alternative possibility remains that 2DE has a greater between-locus unevenness of variant detection sensitivity than does 1DE.     

Recombination and the frequency spectrum in Drosophila melanogaster and Drosophila simulans

Most "tests of neutrality" assess whether particular data sets depart from the predictions of a standard neutral model with no recombination. For Drosophila, where nuclear polymorphism data routinely show evidence of genetic exchange, the assumption of no recombination is often unrealistic. In addition, while conservative, this assumption is made at the cost of a great loss in power. Perhaps as a result, tests of the frequency spectrum based on zero recombination suggest an adequate fit of Drosophila polymorphism data to the predictions of the standard neutral model. Here, we analyze the frequency spectrum of a large number of loci in Drosophila melanogaster and D. simulans using two summary statistics. We use an estimate of the population recombination rate based on a laboratory estimate of the rate of crossing over per physical length and an estimate of the species' effective population size. In contrast to previous studies, we find that roughly half of the loci depart from the predictions of the standard neutral model. The extent of the departure depends on the exact recombination rate, but the global pattern that emerges is robust. Interestingly, these departures from neutral expectations are not unidirectional. The large variance in outcomes may be due to a complex demographic history and inconsistent sampling, or to the pervasive action of natural selection.     

Directional evolution of size coupled with ascertainment bias for variation in Drosophila microsatellites

Species-specific differences in microsatellite locus length and ascertainment bias have both been proposed to explain differences in microsatellite variability and length usually observed when loci isolated in one species are used to survey variation in a related species. Here we provide a simple algebraic approach to independently estimate the contributions of true species-specific length differences and ascertainment bias. We apply this approach to a reciprocal-isolation microsatellite study and show contributions of both ascertainment bias and a true longer average microsatellite length in Drosophila melanogaster compared with D. simulans.     

Multiple Allelism of the net Gene in Drosophila melanogaster and Drosophila simulans

Thenetgene mutations are known to cause abnormal pattern of veining in all wing regions except for the first posterior cells. In natural populations of Drosophila melanogaster, the net alleles were identified, which differ in phenotypic expression from standard mutations. The mutants net-extra-analis from a population Belokurikha-2000 have only a single additional vein in the third posterior cell. A line from Chernobyl-1986 population have another nontypical allele net ^Ch86 and shows a lower degree of abnormalities than that usually observed. About 10% of these flies have an additional vein fragment in the first posterior cell. In both males and females ofD. simulans population Tashkent -2001, which exhibit net ^ST91 mutation, a net of additional veins is formed as a specific additional fragment in the first posterior cell. The pattern of veining conferred by alleles net-extra-analis and net ^Ch86 is altered to a lesser extent; these alleles are dominant with respect to alleles net ^2-45 and net ^ST91, which cause more abnormalities. The heterozygotes for alleles net ^ST9 and net ^Ch86 and for Df(2) net ⁶² deletion have an additional fragment in the first posterior cell and show similarly strong deviations from normal wing vein pattern. The naturalnet alleles correspond, presumably, to different molecular gene defects involved into uncertain local interactions with numerous modifying factors and other genes that specify the wing vein pattern.     

Clonal analysis in hybrids between Drosophila melanogaster and Drosophila simulans

We have analysed the viability of cellular clones induced by mitotic recombination in Drosophila melanogaster/D. simulans hybrid females during larval growth. These clones contain a portion of either melanogaster or simulans genomes in homozygosity. Analysis has been carried out for the X and the second chromosomes, as well as for the 3L chromosome arm. Clones were not found in certain structures, and in others they appeared in a very low frequency. Only in abdominal tergites was a significant number of clones observed, although their frequency was lower than in melanogaster abdomens. The bigger the portion of the genome that is homozygous, the less viable is the recombinant melano-gaster/simulans hybrid clone. The few clones that appeared may represent cases in which mitotic recombination took place in distal chromosome intervals, so that the clones contained a small portion of either melanogaster or simulans chromosomes in homozygosity. Moreover, Lhr, a gene of D. simulans that suppresses the lethality of male and female melanogaster/simulans hybrids, does not suppress the lethality of the recombinant melanogaster/simulans clones. Thus, it appears that there is not just a single gene, but at least one per tested chromosome arm (and maybe more) that cause hybrid lethality. Therefore, the two species, D. melanogaster and D. simulans, have diverged to such a degree that the absence of part of the genome of one species cannot be substituted by the corresponding part of the genome of the other, probably due to the formation of co-adapted gene complexes in both species following their divergent evolution after speciation. The disruption of those coadapted gene complexes would cause the lethality of the recombinant hybrid clones.     

Electrophoretic variability in natural populations of Drosophila melanogaster and Drosophila simulans

Genetic variation at 59 gene loci coding for enzymes (50) and larval proteins (9) has been studied in sympatric populations of Drosophila melanogaster and D. simulans from insular and continental origin. The average number of alleles per locus, the mean proportion of polymorphic loci and the mean heterozygosity are similar both within and between species. There are however some significant differences between D. simulans populations in the genotypic frequencies for four polymorphic loci.     

Multiple allelism of the net gene in Drosophila melanogaster and Drosophila simulans

The net gene mutations are known to cause abnormal pattern of veining in all wing regions except for the first posterior cells. In natural populations of Drosophila melanogaster, the net alleles were identified, which differ in phenotypic expression from standard mutations. The mutants net-extra-analis from a population Belokurikha-2000 have only a single additional vein in the third posterior cell. A line from Chernobyl-1986 population have another nontypical allele netCh86 and shows a lower degree of abnormalities than that usually observed. About 10% of these flies have an additional vein fragment in the first posterior cell. In both males and females of D. simulans population Tashkent-2001, which exhibit netST91 mutation, a net of additional veins is formed as a specific additional fragment in the first posterior cell. The pattern of veining conferred by alleles net-extra-analis and netCh86 is altered to a lesser extent; these alleles are dominant with respect to alleles net2-45 and netST91, which cause more abnormalities. The heterozygotes for alleles netST9 and netCh86 and for Df(2) net62 deletion have an additional fragment in the first posterior cell and show similarly strong deviations from normal wing vein pattern. The natural net alleles correspond, presumably, to different molecular gene defects involved into uncertain local interactions with numerous modifying factors and other genes that specify the wing vein pattern.     

Allozyme variation in Greek wild populations of Drosophila melanogaster and D. simulans along a North-South gradient

Allozyme frequency data from five Greek wild sympatric populations of Drosophila melanogaster and D. simulans along a North-South gradient were analyzed for genotype-environment relationships. The regression coefficient of genetic distance on geographic distance indicates that there is a significant relationship between these parameters for D. melanogaster only. Highly significant differences in specific alleles at certain loci were found between the various local populations studied. The changes in Gpdh ^F of D. melanogaster and Est-6 ^F of D. simulans exhibited clinal patterns in allele frequencies. In addition, analysis of D. melanogaster Gpdh ^Fand Adh ^F allele frequencies shows that the Greek data do not have regression coefficients (regressing allele frequency on degrees North of latitude) of the same sign as East-and West-Coast United-States populations. These contradictory data are discussed in relation to what is known about the maintenance of the Adh and Gpdh polymorphisms.     

Alcohol tolerance: An ecological parameter in the relative success of Drosophila melanogaster and Drosophila simulans

Summary Laboratory experiments have shown D. melanogaster adults to be more tolerant to alcohol in the environment than D. simulans, with the females being more tolerant than the males of their species. Larval development on alcohol supplemented media also demonstrated an increased tolerance by D. melanogaster although the effect was not as clear cut as for the adult survival. Oviposition choice experiments demonstrated a marked rejection of alcohol impregnated laying sites by D. simulans when compared to standard medium sites. D. melanogaster showed a slight preference for alcohol supplemented sites.Collections in the maturation cellar of a vineyard produced, with the exception of a single D. simulans fly, entirely D. melanogaster adults while larvae and pupae from the cellar were also all D. melanogaster. Away from the alcohol resource, outside the cellar, both species were collected with D. simulans being the more common. However, the outside distribution of the two species was affected by alcohol fumes during vintage, as was the distribution of the sexes of D. melanogaster, with the more tolerant species or sex being closer to the source. The field results were thus in agreement with the laboratory predictions that D. melanogaster is better able to utilize an alcohol resource than D. simulans.     

Genetics and the environment in interspecific competition: a study using the sibling species Drosophila melanogaster and Drosophila simulans

The outcome of interspecific competition of two closely related species may depend upon genetic variation in the two species and the environment in which the experiment is carried out. Interspecific competition in the two sibling species, Drosophila melanogaster and D. simulans, is usually investigated using longterm laboratory stocks that often have mutant markers that distinguish them. To examine competition in flies that genetically more closely resemble flies in nature, we utilized freshly caught wildtype isofemale lines of the two species collected at the same site in San Carlos, Mexico. Under ordinary laboratory conditions, D. melanogaster always won in competition. However, in hotter and drier conditions, D. simulans competed much more effectively. In these environmental conditions, there were genetic differences in competitive ability among lines with the outcome of competition primarily dependent upon the line of D. melanogaster used but in some cases also influenced by the line of D. simulans used. Differences in the measures of productivity and developmental time did not explain the differences in competitive ability among lines. This suggests that the outcome of competition was not due to differences in major fitness components among the isofemale lines but to some other attribute(s) that influenced competitive ability. When lines of flies were combined, the outcome of competition was generally consistent with competitive outcomes between pairs of lines. In several cases, the combination of lines performed better than the best of the constituent lines, suggesting that competitive ability was combined heterotically and that the total amount of genetic variation was important in the outcome of interspecific competition.