• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Feb 2007; 175(2): 751–761.
PMCID: PMC1800615

Ecological Zones Rather Than Molecular Forms Predict Genetic Differentiation in the Malaria Vector Anopheles gambiae s.s. in Ghana

Abstract

The malaria mosquito Anopheles gambiae s.s. is rapidly becoming a model for studies on the evolution of reproductive isolation. Debate has centered on the taxonomic status of two forms (denoted M and S) within the nominal taxon identified by point mutations in the X-linked rDNA region. Evidence is accumulating that there are significant barriers to gene flow between these forms, but that the barriers are not complete throughout the entire range of their distribution. We sampled populations from across Ghana and southern Burkina Faso, West Africa, from areas where the molecular forms occurred in both sympatry and allopatry. Neither Bayesian clustering methods nor FST-based analysis of microsatellite data found differentiation between the M and S molecular forms, but revealed strong differentiation among different ecological zones, irrespective of M/S status and with no detectable effect of geographical distance. Although no M/S hybrids were found in the samples, admixture analysis detected evidence of contemporary interform gene flow, arguably most pronounced in southern Ghana where forms occur sympatrically. Thus, in the sampled area of West Africa, lack of differentiation between M and S forms likely reflects substantial introgression, and ecological barriers appear to be of greater importance in restricting gene flow.

WHEN reproductive isolation between species is incomplete (i.e., when F1 hybrids are not completely sterile), genes may pass between species (Machado et al. 2002). Whether such interspecific hybridization is a significant route for the transfer of genetic adaptation remains a major question in evolutionary biology (Barton 2001). The African malaria vector Anopheles gambiae is a unique model for studies of the evolution of reproductive isolation and the importance of hybridization for the transfer of adaptively advantageous genes in incipient species. Within An. gambiae, sequence variation in X-linked rDNA of populations from West Africa led to the description of two molecular forms (termed M and S), which differ in both the transcribed and the nontranscribed spacers in the rDNA repeat unit (della Torre et al. 2001; Gentile et al. 2002). Reproductive isolation between the M and S forms has been supported by a lack of hybrids detected by PCR of the rDNA intergenic and also partially by studies of the distribution of the kdr gene, a mutation involved in resistance to pyrethroid insecticides (Chandre et al. 1999). In general, in West African countries, apart from Benin, the kdr resistance locus is at or near fixation in the S-form populations but occurs in the M form at very low levels, even in sympatry (della Torre et al. 2001; Yawson et al. 2004). Importantly, it has been proposed that the kdr mutation may have introgressed from the S form into the M form (Weill et al. 2000), raising the possibility that there could be numerous adaptively and epidemiologically important genetic exchanges between forms.

Evidence from Drosophila has indicated that isolation does not occur simultaneously for the whole genome (Machado et al. 2002) and that the least likely parts to exchange are associated with hybrid sterility or mate choice (Noor et al. 2001). This mosaic genome structure was first postulated for An. gambiae by della Torre et al. (1997) and more recent evidence was provided by Turner et al. (2005) who used a microarray approach to identify regions of the genome that they term “islands of speciation.” One of the regions that Turner et al. (2005) identified does indeed appear to be intimately involved with the reproductive isolation of these forms (Stump et al. 2005).

Outside the few areas of marked differentiation only slight differences have been observed in allele or haplotype arrays between the M and S forms using a range of genetic markers (Gentile et al. 2002; Lehmann et al. 2003; Donnelly et al. 2004), although microsatellites from regions of the second chromosome associated with putatively selective advantageous chromosomal inversions have been shown to differ between forms (Lanzaro et al. 1998). The lack of marked differentiation throughout most of the genome has been attributed to the recent isolation of the forms whereby insufficient time has elapsed for even noncoding DNA to accrue differences (Gentile et al. 2001). Premating isolation mechanisms apparently result in positive assortative mating when the forms are in sympatry, with 98.8% within form mating (Tripet et al. 2001). Indeed, evidence from a study in Cameroon suggested that reinforcement between forms might occur in sympatry, with FST estimates of 0.035–0.052 between allopatric populations compared to 0.060 between sympatric populations (Wondji et al. 2002).

Given that there is strong positive assortative mating and molecular genetic suggestions of reinforcement in sympatry, the question arises, where does interform hybridization occur? Rates of hybridization are unlikely to be equal across the sympatric distributions of the forms and therefore identifying the geographic regions where genetic exchange does occur is central to an understanding of the evolution of these incipient species. Previous evidence has suggested that the area between Cote d'Ivoire and Benin may be an area of transition between populations that show strong positive assortative mating and populations that are more permissive of interform gene flow (Black and Lanzaro 2001; della Torre et al. 2001). We used microsatellite loci to determine the intra- and interform variation in M and S forms in Ghana and southern Burkina Faso. Our data suggest that interform hybridization occurs at significant levels and that ecological barriers represent far more important barriers to gene flow in this area of West Africa.

MATERIALS AND METHODS

Sample sites and screening:

Sample sites are shown in Figure 1, with a detailed description in Yawson et al. (2004). In brief, M and S forms were sympatric in the mangrove swamps along the coast; in northern Ghana and Burkina Faso, samples were overwhelmingly M form (80–86% in northern Ghana and 80% in Burkina Faso) and the samples from the middle rain forest belt were all S forms (Figure 1). Specimens were identified as to form using a combination of the approaches of Scott et al. (1993) and Fanello et al. (2002). Seven microsatellite loci—two on chromosome X (AgXH7, AgXH99), one on chromosome 2 (Ag2H197), and four on chromosome 3 (Ag3H119, Ag3H812, Ag3H577, 33CI)—were chosen on the basis of physical and linkage maps of An. gambiae (Zheng et al. 1996) and, with the exception of Ag2H197 (inversion 2Ra), were not contained within polymorphic inversions. Previous studies had found these markers to be highly variable and to amplify reliably (Donnelly et al. 2001; Pinto et al. 2002; Wondji et al. 2002). None of these markers are within the “genomic islands of speciation” identified by other authors (Turner et al. 2005) but some of the markers show marked differences in allele frequencies between molecular forms (Pinto et al. 2002; Wondji et al. 2002). Thirty-one individuals were identified per site unless the forms occurred in sympatry, in which case 31 individuals of each form were identified. Microsatellite genotyping methods followed (Donnelly et al. 1999). In total, genotypes were determined for 462 An. gambiae specimens from 11 sites in Ghana and 1 site in Burkina Faso.

Figure 1.
Sampling locations. Ghana and Burkina Faso are marked on the outline map of Africa as shaded and solid, respectively. The more detailed map of Ghana shows the limits of the ecological zones within Ghana. Solid, mangrove strand zone; light shading, coastal ...

Statistical analysis:

Tests of Hardy–Weinberg equilibrium (HWE) and linkage disequilibrium among all pairs of loci within populations were done with GENEPOP 3.4 (Raymond and Rousset 1995). Diversity per locus and per population was assessed using allelic richness (RS), observed (HO) and expected (HE) heterozygosities, and inbreeding coefficients (FIS), and their significance was calculated using FSTAT 2.9.3.2 (Goudet 2001). To explore the impact of putative null alleles on our analyses, we used the program MICRO-CHECKER (Van Oosterhout et al. 2004) to estimate null allele frequencies (using the “Brookfield 2” estimator; Brookfield 1996) for potentially affected loci that showed a consistent significant excess of homozygotes across sample sites. Using these estimated frequencies, an adjusted data set, with nulls recoded as separate alleles, was produced for comparison with the original data set.

Genetic relationships within and between forms were assessed using three complementary approaches: two Bayesian clustering methods and an FST-based summary statistic method. A Bayesian clustering algorithm implemented in the program STRUCTURE (Pritchard et al. 2000) was applied to identify subgroups that have distinctive allele frequencies, without using prior knowledge of sampling sites. We used the admixture model with correlated allele frequencies between populations with a burn-in length of 100,000, and 100,000 Markov chain Monte Carlo replications for each setting of K from 2 to 6 (20 replicate runs of each). To determine the most appropriate number of clusters, we used the approach of Evanno et al. (2005), which is based upon an ad hoc quantity (ΔK) that evaluates the second-order rate of change of the likelihood function with respect to K. For the STRUCTURE analysis, we made three predictions: (i) if there was extensive genetic exchange between the forms or if they had only recently become separated, then we would expect M- and S-form samples to form a single cluster; (ii) if sufficient time had elapsed for the forms to diverge, then we would observe clusters corresponding with each of the forms; and (iii) if there were zones of hybridization between the two forms, then we would observe two clusters, one corresponding with each of the forms and one or more additional clusters containing those samples where introgression was occurring.

A second Bayesian clustering analysis, implemented in the software BAPS 4.13, was also applied (Corander et al. 2003). By contrast to the individual-based algorithm applied in STRUCTURE, we used the group-level option in BAPS 4.13 such that clusters are formed by merging whole samples. Since group-level analysis does not rely on the integrity of individual multilocus genotypes, we were able to use the method to compare solutions from the original and null-allele-corrected data sets. The resulting clusters (from the original data set) were saved, and admixture proportions—estimated for each individual, again using BAPS 4.13 (Corander et al. 2004), with the posterior probability for the null hypothesis of pure ancestry—were computed by permutation. We used clusters, rather than individual sample sites, as the units for the analysis because admixture results can be unreliable when the power to discriminate populations is low (J. Corander, personal communication). Thus, while total contemporary admixture of an individual sample site could be determined, the source of the nonresident admixture proportion(s) could be assigned only to a cluster. Since our aim was to study patterns of admixture, rather than to identify specific potentially migrant individuals, critical α for admixture was set at a posterior probability of 0.05, although we also determined individuals that showed admixture probabilities that remained significant following Bonferroni correction.

FST estimates and pairwise (permutation-based) tests of genic differentiation, which assume unlinked markers but not Hardy–Weinberg equilibrium, were calculated using Genepop 3.4 (Raymond and Rousset 1995). Pairwise FST estimates were linearized following Slatkin (1995) and the matrix was entered into the MEGA 3.1 program for neighbor-joining tree construction (Kumar et al. 2004). We hypothesize that if introgression is ongoing in sympatry, FST values between sympatric populations will be lower than those between interform allopatric comparisons, whereas the converse is expected if reinforcement occurs in sympatry. As an ad hoc test of these hypotheses, groups were resampled using 1000 bootstrap replicates to estimate 95% confidence intervals around the group mean FST using SAS (SAS Institute 1990). Isolation by distance was examined by testing the association of linearized FST with ln-transformed geographical distance using a Mantel test with 5000 permutations implemented by the POPTOOLS add-in for Microsoft EXCEL (written by Greg Hood, CSIRO; available from http://www.cse.csiro.au/poptools), Unless stated otherwise, where multiple tests were used, we adjusted the significance level using a sequential Bonferroni procedure (Holm 1979).

RESULTS

Hardy–Weinberg equilibrium and genotypic disequilibrium:

An excess of homozygotes was very common in the data set (appendix a), with 48 of a total of 105 tests significant at P < 0.05, and 26 remaining so following sequential Bonferroni correction. Hardy–Weinberg deviations were not restricted to particular loci nor geographic samples (appendix a). Of 315 tests for linkage disequilibrium, 16 were significant at P < 0.05, which is exactly the number expected by chance as type I errors at this critical α-level, and following sequential Bonferroni correction, no tests remained significant in any sample or at any locus. Since within-population structure would be expected to cause both excess homozygosity and linkage disequilibrium, the widespread prevalence of the former but the absence of the latter in our data suggested the presence of null alleles. Estimates of null allele frequencies per population and per locus were generated, assuming random mating, using MICRO-CHECKER (Van Oosterhout et al. 2004) wherever a significant excess of homozygotes was detected. The mean estimated null allele frequency of 8.7% was similar to the mean frequency of null alleles (5–8%) estimated for An. gambiae on the basis of X-linked loci in males (Barnes et al. 2005; Stump et al. 2005).

APPENDIX A
Tests for deviation from HWE

Intra- and interform genetic differentiation:

Individual-based Bayesian clustering using STRUCTURE determined the most appropriate number of clusters as two, and successive increases in K did not split the two major groupings into additional clusters (supplemental materials at http://www.genetics.org/supplemental/). The ΔK value provided support for the division into two groups with a marked peak at K = 2 and a rapid decline at higher values. The clusters did not partition the molecular forms, but rather grouped individuals by ecological zone, although not by geographic proximity (see Figure 1). One cluster comprised M- and S-form individuals from the coastal mangrove strand zone, with the other cluster containing all individuals from the other sites, sampled from the northern Sahel savanna, coastal savanna and deciduous forest (Figure 2a).

Figure 2.
(a) Results from an individual-level Bayesian assignment/cluster analysis in which the optimal number of clusters was two. Vertical bars indicate the proportionate assignment of an individual to each cluster. (b) Neighbor-joining tree based on linearized ...

Population differentiation was widespread with 101 of 105 pairwise tests significant following sequential Bonferroni correction (appendix b). The major division in an FST-based neighbor-joining tree was in accord with neither geographic proximity of the sample locations nor with the molecular status of the samples, but represented a primary division between the populations in the mangrove strand zone of southern Ghana (in which M and S are sympatric) and all remaining M and S populations from the north and the south (Figure 2b). Results from group-level Bayesian clustering analysis performed by BAPS also revealed a major split between the mangrove strand zone and other populations, but in addition partitioned the Kumasi (deciduous forest zone) separately (Figure 2b). This optimal clustering solution was well supported since the next most likely solution—of two clusters, in which Kumasi grouped with all other non-mangrove zone samples—could be confidently rejected (P < 0.0001). Nevertheless, the Kumasi site was clearly far more distinct from the mangrove zone samples (mean pairwise FST = 0.10) than from the northern and coastal savanna samples (FST = 0.04).

APPENDIX B
Population differentiation

There was limited evidence to suggest a contribution of molecular form to differentiation. For comparisons made either within or between ecological zones, interform mean FST values were only marginally higher than intraform values and none of these differences were significant (Figure 3). However, since the difference between inter- and intraform comparisons was, if anything, greater in the non-mangrove cluster of allopatric M and S populations (categories B and E in Figure 3) than in the mangrove cluster of sympatric M and S occurrence (categories C and F in Figure 3), our a priori hypothesis of reinforcement in sympatry (see materials and methods) seems unlikely. Evidence for interform hybridization in sympatry was more equivocal since both inter- and intraform differentiation within the mangrove strand cluster were lower than elsewhere. Such generally greater genetic homogeneity in the mangrove zone might be attributable to lower average pairwise separation (mangrove mean = 61 km; non-mangrove mean = 336 km). Yet isolation by distance (analyzed via linearized FST vs. ln geographical distance) was not significant, either over all samples (Mantel's test P = 0.14) or within samples from outside of the mangrove zone (P = 0.092); too few sites were available to permit separate analysis within the mangrove zone. Moreover, one of the sampling sites from the mangrove strand zone was actually closer geographically to those within the coastal savanna than to the other two sites within its own ecological zone (see Figure 1), further suggesting that distance is not an important determinant of differentiation.

Figure 3.
Bootstrapped mean FST values 95% confidence intervals for comparisons made within and among molecular forms (M and S) and ecological zones. Comparison categories are: A, mangrove zone vs. non-mangrove zones between form comparisons; B, non-mangrove zones ...

The concordant sample groupings observed in both the STRUCTURE analysis (Figure 2a) and the FST neighbor-joining tree (Figure 2b) were not unduly influenced by a single locus. The number of loci with FST values significantly different from zero was markedly different in comparisons from within the major clusters (mean = 3.1 loci) compared to comparisons between clusters (mean = 6.2 loci) (appendix b). Similarly, null alleles exerted little influence on our results. A data set in which null alleles were recoded according to their estimated frequencies showed a lower global FST (0.065, compared to 0.076 for the original data set), but the neighbor-joining tree and group-level clustering analysis results were almost indistinguishable between data sets (supplemental materials at http://www.genetics.org/supplemental/).

Genetic diversity and admixture:

The seven microsatellites analyzed were highly polymorphic in all populations (appendix c). Variation in genetic diversity among populations was quite moderate, whether measured as allelic richness (RS, range 7.69–10.01) or expected heterozygosity (HE, range 0.71–0.82). Both RS and HE differed significantly between the two major clusters identified above, with each higher in the mangrove strand zone than in the cluster of other populations (FSTAT 2.9.3.2, 1000 permutations, both P < 0.05).

APPENDIX C
Genetic diversity

Since we have no evidence that the populations in this study are close to migration-drift equilibrium, we applied a nonequilibrium admixture analysis. Admixture analysis allows us to determine whether individuals have recent ancestors originating from more than one sample population. We used this analysis to study patterns of very recent migration, using the three clusters identified by the BAPS group-level method (namely, Kumasi, mangrove strand zone, and savanna clusters) as potential source populations (see materials and methods). Of a total of 411 individuals, 33 were identified as having multilocus genotypes exhibiting significant evidence of population mixing (appendix d), representing a total multilocus admixture level of 4.3%. Individual admixture proportions rarely approached 1.0; that is, no individual could be unequivocally assigned to a population other than the one in which it was sampled (appendix d), suggesting that our samples contained few first-generation (i.e., F0) immigrants. Four of the eight populations composing the savanna cluster contained no individuals showing significant evidence of mixed cluster ancestry, whereas all from the mangrove cluster showed some degree of admixture (Table 2). Indeed, mean admixture proportions were significantly higher in the mangrove than in the savanna cluster (Mann–Whitney U-test, U6,8 = 3.5, P < 0.01). Interestingly, Kumasi, a single-population cluster from deciduous forest, contained no admixed individuals but was identified as the source for more admixture than either of the other clusters (Table 2). Moreover, since the Kumasi sample was composed entirely of the S molecular form, apparent introgression into M-form populations could be identified. Admixture proportions from Kumasi (S) did not differ between M or S recipient populations (U6,8 = 30.0, P = 0.49).

TABLE 2
Results from Bayesian admixture analysis assessing very recent migration among population clusters based on individuals assessed to show significant evidence of admixture
APPENDIX D
Bayesian admixture analysis (using BAPS 4.13) assessing very recent migration among population clusters

DISCUSSION

In this study, we used microsatellites to screen An. gambiae from Ghana and southern Burkina Faso, with samples from two savanna zones and a deciduous forest zone, in which the M and S molecular forms are largely allopatric, and a coastal mangrove strand zone, in which M and S are sympatric. Significant genetic differentiation between sample sites was near ubiquitous, but the major genetic division, supported by both individual-level Bayesian cluster analysis and FST-based analysis, was found between samples from the mangrove strand zone vs. the other ecological zones. A group-level Bayesian analysis supported an additional partition between the single sample from the deciduous forest zone and the savanna zones, although this was far less distinct from the savanna cluster than from the mangrove zone. Differentiation attributable to molecular form was limited and not significant. These patterns were found consistently across loci and were not the result of the major effect of a single locus, nor could they be attributed to null alleles, although these appeared to be common in the data set.

Determinants of genetic structure:

Such high differentiation (FST > 0.1 between the mangrove zone and other areas) is unusual in An. gambiae. To date, estimates of genetic differentiation have been low even between populations separated by many thousands of kilometers (Lehmann et al. 2003; Donnelly et al. 2004), unless there are major hydrographic or geographic barriers to migration (e.g., Pinto et al. 2002; Lehmann et al. 2003). Yet in the south of Ghana (all samples from south of Kumasi) there are no obvious extant or former geographic barriers to gene flow, with all samples from this region taken from a continuous coastal plain that does not rise >100 m above sea level. Moreover, the effect of geographical distance on the patterns of differentiation in our study appears to be negligible.

High levels of differentiation are sometimes detected among the chromosomal forms of An. gambiae. For example, using AFLP markers, Slotman et al. (2006) obtained estimates of ΦST (an FST analog) between ~0.05 and 0.15 among Bamako, Savanna, and Mopti forms. Although estimates of differentiation from microsatellites and AFLPs may not be directly comparable because of the downward bias to the former caused by size homoplasy, the estimates of Slotman et al. (2006) seem compatible with those found among ecological zones in our study. Moreover, different chromosomal forms are often associated with different habitats and are thought to play an important role in environmental adaptation (della Torre et al. 2001). However, studies of the presence and distribution of chromosomal forms in the area that we sampled suggest that it is unlikely that high divergence results from the presence of different chromosomal forms in different ecological zones. Appawu et al. (1994) reported that although the FOREST chromosomal form is dominant in the deciduous forest zone, the SAVANNA and MOPTI chromosomal forms were sympatric in both of the savanna zones. Furthermore, Mbole et al. (2004) found all three of these chromosomal forms within the mangrove strand zone. Since we do not have karyotypic information for the samples screened in this study, and the relationship between molecular and chromosomal forms is often ambiguous (della Torre et al. 2001), we cannot rule out the possibility that the different habitat types are dominated by different and genetically divergent chromosomal forms. Nevertheless, unless there are major chance, spatial, or temporal discontinuities between the chromosomal forms represented in our samples and those obtained by Appawu et al. (1994) and Mbole et al. (2004), atypical chromosomal forms within a cluster should have been readily detectable via our clustering/assignment analyses and would resemble first-generation immigrants. By contrast, very few individuals in sample sites within each of the two clusters (STRUCTURE) or three clusters (BAPS) identified could be interpreted as immigrants from a different cluster (see Figure 2a).

Therefore, the genetic structure that we have identified appears to be unrelated to physical barriers to gene flow, geographic distance, or molecular form and is unlikely to be attributable to habitat specificity of chromosomal forms. A further possibility is that the very high differentiation is temporally unstable and results from a relatively recent colonization (or recolonization) of, and spread within, the mangrove strand zone from a highly divergent source. While considerable restriction of gene flow from the coastal savanna zone would presumably be required for this scenario to persist for more than a few generations, the significantly lower interpopulation differentiation in the mangrove zone is consistent with this colonization–expansion hypothesis. Yet, the significantly higher genetic diversity and admixture within the mangrove zone are the opposite of what would be expected under a colonization–expansion scenario unless multiple differentiated source populations colonized the area, and this is at odds with the relatively low genetic structure therein. Consequently, we think that recent colonization can also be ruled out. Rather, the most likely explanation would seem that habitat-type-specific selection against interzone immigrants represents a barrier to gene flow.

Gene flow between the molecular forms:

Studies of gene flow between the molecular forms of An. gambiae have found differentiation between the two forms at four genomic regions (but see Wondji et al. 2002; Turner et al. 2005). One of these is on chromosome 2L, associated with insecticide resistance (Gentile et al. 2004 and references therein). The other loci (including the rDNA) are at the proximal (centromeric) end of the X chromosome (Wang et al. 2001; Lehmann et al. 2003). However, little differentiation has been found at loci elsewhere in the genome (Gentile et al. 2001; Mukabayire et al. 2001; Wang et al. 2001; Turner et al. 2005). Debate is ongoing as to whether such generally limited differentiation primarily reflects insufficient time for differentiation to have occurred or ongoing introgression among forms. Neither the FST-based analysis nor the Bayesian clustering methods provided conclusive evidence to support or reject our main hypothesis that interform hybridization/introgression would be more common when forms are sympatric than (primarily) allopatric. The alternative hypothesis—that reinforcement might occur between the forms in sympatry—seems very unlikely, given that the difference between the mean level of intra- and interform comparisons was lower for sympatric than for allopatric populations.

Nevertheless, results from our admixture analysis suggest that genetic exchange between forms might be more frequent than previous molecular genetic and assortative mating studies would imply (Tripet et al. 2001; Wondji et al. 2002). The Kumasi population, which is entirely composed of the S form (Yawson et al. 2004), was sufficiently differentiated by the group-level cluster analysis to be identified as a source of admixture. Recipient M and S populations received low but very similar proportionate admixture from Kumasi, suggesting that intra- and interform gene flow is low but equally common from the Kumasi population. In high-differentiation systems, pairwise admixture proportions in the BAPS analysis that we applied appear to be insensitive to unsampled “ghost” populations (see Hanfling and Weetman 2006). By contrast, since the M and S forms are virtually indistinguishable, it is possible that admixture proportions allocated to M recipient populations from our Kumasi S sample might have originated from an unsampled ghost M-form population closely resembling Kumasi. While it is important to recognize this kind of limitation when applying any admixture or assignment analysis, there is no evidence for any M forms occurring within the deciduous forest zone.

A further line of evidence that higher interform exchange occurs in sympatry comes from typing of the pyrethroid insecticide knockdown resistance (kdr) genotype. This segregated according to form: very low frequencies in the M form and near fixation in the S form; but the highest frequencies of the kdr genotype were all found in M forms from the sympatric M/S sample sites (Yawson et al. 2004). The absence of any M/S hybrids in the 935 specimens from which the data presented here are drawn (Yawson et al. 2004) suggests that at least during our collection period direct introgression between M and S forms is rare (Touré et al. 1998; Taylor et al. 2001; Tripet et al. 2001, 2003). Of course, our data represent a short sampling period and there could be a seasonal component to introgression that we have not detected. Collections were made in the rainy season when mosquito numbers were high and it is possible that patterns of hybridization differ during the dry season when mosquito numbers are low and either one of the two forms may predominate.

Conclusion:

Although there is evidence to suggest that the M and S forms of An. gambiae may be separate entities that have been evolving under divergent selection pressures (Stump et al. 2005), data from our study area suggest that interform gene flow is occurring at appreciable levels, perhaps most frequently in the southern mangrove strand zone of M/S sympatry. Our most striking and unexpected finding was of extremely strong genetic structure that correlated with ecological zones, with little, if any, contribution of distance, molecular form, and probably also chromosomal form. Results such as these suggest that climatic modeling and Geographic Information Systems-based approaches will be fruitful in the prediction of gene flow in An. gambiae, the understanding of which is critical for malaria control in Africa.

TABLE 1
Sampling details

Acknowledgments

We thank Bill Black, Harold Townson, Joao Pinto, Charles Wondji, and Kwang Shik Choi for helpful discussions. This study was funded by grants from the Ford Foundation International Fellowships Program and the World Health Organization Special Programme in Research and Training.

References

  • Appawu, M. A., A. Baffoe-Wilmot, E. A. Afari, F. K. Nkrumah and V. Petrarca, 1994. Species composition and inversion polymorphism of the Anopheles gambiae complex in some sites of Ghana, West-Africa. Acta Trop. 56: 15–23. [PubMed]
  • Barnes, M. J., N. F. Lobo, M. B. Coulibaly, N. Sagnon, C. Costantini et al., 2005. SINE insertion polymorphism on the X chromosome differentiates Anopheles gambiae molecular forms. Insect Mol. Biol. 14: 353–363. [PubMed]
  • Barton, N. H., 2001. The role of hybridization in evolution. Mol. Ecol. 10: 551–568. [PubMed]
  • Black, W. C., IV, and G. C. Lanzaro, 2001. Distribution of genetic variation among chromosomal forms of Anopheles gambiae s.s.: Introgressive hybridization, adaptive inversions, or recent reproductive isolation? Insect Mol. Biol. 10: 3–7. [PubMed]
  • Brookfield, J. F. Y., 1996. A simple new method for estimating null allele frequency from heterozygote deficiency. Mol. Ecol. 5: 453–455. [PubMed]
  • Chandre, F., F. Darriet, L. Manga, M. Akogbeto, O. Faye et al., 1999. Status of pyrethroid resistance in Anopheles gambie sensu lato. Bull. World Health Organ. 77: 230–234. [PMC free article] [PubMed]
  • Corander, J., P. Waldmann and M. J. Sillanpää, 2003. Bayesian analysis of genetic differentiation between populations. Genetics 163: 367–374. [PMC free article] [PubMed]
  • Corander, J., P. Waldmann, P. Marttinen and M. J. Sillanpaa, 2004. BAPS 2: enhanced possibilities for the analysis of genetic population structure. Bioinformatics 20: 2363–2369. [PubMed]
  • della Torre, A., L. Merzagora, J. R. Powell and M. Coluzzi, 1997. Selective introgression of paracentric inversions between two sibling species of the Anopheles gambiae complex. Genetics 146: 239–244. [PMC free article] [PubMed]
  • della Torre, A., C. Fanello, M. Akogbeto, J. Dossou-yovo, G. Favia et al., 2001. Molecular evidence of incipient speciation within Anopheles gambiae s.s. in West Africa. Insect Mol. Biol. 10: 9–18. [PubMed]
  • Donnelly, M. J., N. Cuamba, J. D. Charlwood, F. H. Collins and H. Townson, 1999. Population structure in the malaria vector, Anopheles arabiensis Patton, in East Africa. Heredity 83: 408–417. [PubMed]
  • Donnelly, M. J., M. C. Licht and T. Lehmann, 2001. Evidence for recent population expansion in the evolutionary history of the malaria vectors Anopheles arabiensis and Anopheles gambiae. Mol. Biol. Evol. 18: 1353–1364. [PubMed]
  • Donnelly, M. J., J. Pinto, R. Girod, N. J. Besansky and T. Lehmann, 2004. Revisiting the role of introgression vs. shared ancestral polymorphisms as key processes shaping genetic diversity in the recently separated sibling species of the Anopheles gambiae complex. Heredity 92: 61–68. [PubMed]
  • Evanno, G., S. Regnaut and J. Goudet, 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol. Ecol. 14: 2611–2620. [PubMed]
  • Fanello, C., F. Santolamazza and A. della Torre, 2002. Simultaneous identification of species and molecular forms of the Anopheles gambiae complex by PCR-RFLP. Med. Vet. Entomol. 16: 461–464. [PubMed]
  • Gentile, G., M. Slotman, V. Ketmaier, J. R. Powell and A. Caccone, 2001. Attempts to molecularly distinguish cryptic taxa in Anopheles gambiae s.s. Insect Mol. Biol. 10: 25–32. [PubMed]
  • Gentile, G., A. della Torre, B. Maegga, J. R. Powell and A. Caccone, 2002. Genetic differentiation in the African malaria vector, Anopheles gambiae s.s., and the problem of taxonomic status. Genetics 161: 1561–1578. [PMC free article] [PubMed]
  • Gentile, G., F. Santolamazza, C. Fanello, V. Petrarca, A. Caccone et al., 2004. Variation in an intron sequence of the voltage-gated sodium channel gene correlates with genetic differentiation between Anopheles gambiae s.s. molecular forms. Insect Mol. Biol. 13: 371–377. [PubMed]
  • Goudet, J., 2001. FSTAT, a program to estimate and test gene diversities and fixation indices (Version 2.9.3) (http://www2.unil.ch/popgen/softwares/fstat.htm).
  • Hanfling, B., and D. Weetman, 2006. Concordant genetic estimators of migration reveal anthropogenically enhanced source-sink population structure in the River Sculpin, Cottus gobio. Genetics 173: 1487–1501. [PMC free article] [PubMed]
  • Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. Stat. 6: 65–70.
  • Kumar, S., K. Tamura and M. Nei, 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinformatics 5: 150–163. [PubMed]
  • Lanzaro, G. C., Y. T. Toure, J. Carnahan, L. Zheng, G. Dolo et al., 1998. Complexities in the genetic structure of Anopheles gambiae populations in West Africa as revealed by microsatellite DNA analysis. Proc. Natl. Acad. Sci. USA 95: 14260–14265. [PMC free article] [PubMed]
  • Lehmann, T., M. Licht, N. Elissa, B. T. A. Maega, J. M. Chimumbwa et al., 2003. Population structure of Anopheles gambiae in Africa. J. Hered. 94: 133–147. [PubMed]
  • Machado, C. A., R. M. Kliman, J. A. Markert and J. Hey, 2002. Inferring the history of speciation from multilocus DNA sequence data: the case of Drosophila pseudoobscura and close relatives. Mol. Biol. Evol. 19: 472–488. [PubMed]
  • Mbole, N. M., D. A. Boakye, M. A. Appawu and M. D. Wilson, 2004. The distribution of chromosomal forms of Anopheles gambiae s.s. (Diptera:Culicidae) in urban, periurban and rural locations in greater Accra. Am. J. Trop. Med. Hyg. 71: 242.
  • Mukabayire, O., J. Caridi, X. Wang, Y. T. Toure, M. Coluzzi et al., 2001. Patterns of DNA sequence variation in chromosomally recognized taxa of Anopheles gambiae: evidence from rDNA and single-copy loci. Insect Mol. Biol. 10: 33–46. [PubMed]
  • Noor, M. A. F., K. L. Grams, L. A. Bertucci and J. Reiland, 2001. Chromosomal inversions and the reproductive isolation of species. Proc. Natl. Acad. Sci. USA 98: 12084–12088. [PMC free article] [PubMed]
  • Pinto, J., M. J. Donnelly, C. A. Sousa, V. Gil, C. Ferreira et al., 2002. Genetic structure of Anopheles gambiae (Diptera: Culicidae) in Sao Tome and Principe (West Africa): implications for malaria control. Mol. Ecol. 11: 2183–2187. [PubMed]
  • Pritchard, J. K., M. Stephens and P. Donnelly, 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945–959. [PMC free article] [PubMed]
  • Raymond, M., and F. Rousset, 1995. Genepop (version 1.2): population genetics software for exact tests and ecumenicism. J. Hered. 86: 248–249.
  • SAS Institute, 1990. SAS Language: References, Version 6. SAS Institute, Cary, NC.
  • Scott, J. A., W. G. Brogdon and F. H. Collins, 1993. Identification of single specimens of the Anopheles gambiae complex by the polymerase chain reaction. Am. J. Trop. Med. Hyg. 49: 520–529. [PubMed]
  • Slatkin, M., 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139: 457–462. [PMC free article] [PubMed]
  • Slotman, M. A., M. M. Mendez, A. Della Torre, G. Dolo, Y. T. Toure et al., 2006. Genetic differentiation between the Bamako and Savanna chromosomal forms of Anopheles gambiae as indicated by amplified fragment length polymorphism analysis. Am. J. Trop. Med. Hyg. 74: 641–648. [PubMed]
  • Stump, A. D., J. A. Shoener, C. Costantini, N. F. Sagnon and N. J. Besansky, 2005. Sex-linked differentiation between incipient species of Anopheles gambiae. Genetics 169: 1509–1519. [PMC free article] [PubMed]
  • Taylor, C. E., Y. T. Toure, J. Carnahan, D. E. Norris, G. Dolo et al., 2001. Gene flow among populations of the malaria vector, Anopheles gambiae, in Mali, West Africa. Genetics 157: 743–750. [PMC free article] [PubMed]
  • Touré, Y. T., P. V. Traoré, S. F. Coulibaly, A. Maiga, H. M. Sankaré et al., 1998. Distribution of inversion polymorphism of chromosomally recognised taxa of the Anopheles gambiae complex in Mali, West Africa. Parassitologia 40: 477–511. [PubMed]
  • Tripet, F., Y. T. Toure, C. E. Taylor, D. E. Norris, G. Dolo et al., 2001. DNA analysis of transferred sperm reveals significant levels of gene flow between molecular forms of Anopheles gambiae. Mol. Ecol. 10: 1725–1732. [PubMed]
  • Tripet, F., Y. T. Toure, G. Dolo and G. C. Lanzaro, 2003. Frequency of multiple inseminations in field-collected Anopheles gambiae females revealed by DNA analysis of transferred sperm. Am. J. Trop. Med. Hyg. 68: 1–5. [PubMed]
  • Turner, T. L., M. W. Hahn and S. V. Nuzhdin, 2005. Genomic islands of speciation in Anopheles gambiae. PloS Biol. 3: 1572–1578. [PMC free article] [PubMed]
  • Van Oosterhout, C., W. F. Hutchinson, D. P. M. Wills and P. Shipley, 2004. MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4: 535–538.
  • Wang, R., L. Zheng, Y. T. Toure, T. Dandekar and F. C. Kafatos, 2001. When genetic distance matters: measuring genetic differentiation at microsatellite loci in whole-genome scans of recent and incipient mosquito species. Proc. Natl. Acad. Sci. USA 98: 10769–10774. [PMC free article] [PubMed]
  • Weill, M., F. Chandre, C. Brengues, S. Manguin, M. Akogbeto et al., 2000. The kdr mutation occurs in the Mopti form of Anopheles gambiae s.s. through introgression. Insect Mol. Biol. 9: 451–455. [PubMed]
  • Wondji, C., F. Simard and D. Fontenille, 2002. Evidence for genetic differentiation between the molecular forms M and S within the Forest chromosomal form of Anopheles gambiae in an area of sympatry. Insect Mol. Biol. 11: 11–19. [PubMed]
  • Yawson, A. E., P. J. McCall, M. D. Wilson and M. J. Donnelly, 2004. Species abundance and insecticide resistance of Anopheles gambiae in selected areas of Ghana and Burkina Faso. Med. Vet. Entomol. 18: 372–377. [PubMed]
  • Zheng, L., M. Q. Benedict, A. J. Cornel, F. H. Collins and F. C. Kafatos, 1996. An integrated genetic map of the African human malaria vector mosquito, Anopheles gambiae. Genetics 143: 941–952. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...