Drosophila subobscura - Wiley Online Library

Hereditas 145: 154162 (2008)

Molecular variation in the Odh gene in Chilean natural populations of Drosophila subobscura ´ MEZ-BALDO ´ 1, AMPARO LATORRE2, LUI´S SERRA1 and FRANCESC MESTRES1 LAIA GO 1 2

Dept. de Gene`tica, Facultat de Biologia, Universitat de Barcelona, Barcelona, Spain Institut Cavanilles de Biodiversitat i Biologia Evolutiva, Universitat de Vale`ncia, Vale`ncia, Spain

Go´mez-Baldo´, L., Latorre, A. Serra, L. and Mestres, F. 2007. Molecular variation in the Odh gene in Chilean natural populations of Drosophila subobscura. * Hereditas 145: 154162. Lund, Sweden. eISSN 1601-5223. Received October 11, 2007. Accepted December 12, 2007 A 793-nucleotide fragment from the D. subobscura Odh gene was sequenced in 46 lethal chromosomal lines from two South American colonizing populations (18 from Santiago de Chile and 28 from Puerto Montt). Only eight different haplotypes were found and, with just one exception, all had previously been detected in North American samples. The exception probably corresponds to a genetic exchange between two American haplotypes. Our results confirm the loss in genetic variability due to the recent founder event, and the high resemblance between the two colonized hemispheres. One haplotype is entirely associated with the O5 inversion, the same association found in North America. Two different haplotypes were detected in association with the O347 chromosomal arrangement. The chromosomal lines featuring this arrangement carried different lethal genes, although there is no association between the latter genes and the two Odh haplotypes. There may be free exchange in the homokaryotypes (O347 /O347) between the various genetic elements (for e.g. the lethal genes and Odh haplotypes) located inside the O7 inversion. The infrequent O7 inversion was observed in the Puerto Montt population presenting one of the haplotypes found only in the O347 chromosomal arrangement. Thus, we confirmed the hypothesis that the origin of this inversion is based on a recombination event in a heterokaryotype O347/Ost. Although one haplotype has been associated with the O342 arrangement, the latter also presents another haplotype shared with both O3 4 8 and Ost. Finally, similar nucleotide diversity values were observed in both Chilean populations. Francesc Mestres, Dept. Gene`tica. Facultat de Biologia, Universitat de Barcelona, Av. Diagonal, 645, ES-08028 Barcelona, Spain. E-mail: [email protected]

In the genus Drosophila, inversion polymorphism has been used extensively as a model to study the adaptive processes involved in the maintenance of genetic variability (KRIMBAS and POWELL 1992, 2000; POWELL 1997). Classical population genetic models suggest that a possible advantage of inversions lies in their ability to suppress recombination in heterokaryotypes, thereby maintaining favorable epistatic interactions between alleles at linked loci (WASSERMAN 1968; DOBZHANSKY 1970). Despite the evidence for selection on inversions based on both biogeographical patterns (KNIBB 1982; PREVOSTI et al. 1988; MENOZZI and KRIMBAS 1992) and cage experiments (VAN DELDEN and KAMPING 1991), research into molecular variation in inverted and standard chromosomes have identified patterns that are generally consistent with equilibrium neutral and/or historical explanations (reviewed by ANDOLFATTO et al. 1999). However, a departure from neutral predictions for the number of haplotypes can arise as a result of either selection or from demographic shifts (ANDOLFATTO and KREITMAN 2000). We have examined nucleotide variation in a 793nucleotide fragment from the Odh gene in lethal chromosomal lines from two South American populations: Santiago de Chile and Puerto Montt. Based on

in situ hybridization analysis (MESTRES et al. 2004) and according to the KUNZE-MU¨HL and MU¨LLER map (1958), the Odh gene is located within section 84 B of the O chromosome of D. subobscura. Until now seven different O chromosomal arrangements were observed in American populations (Ost, O5, O7, O34, O342, O347 and O348) (PREVOSTI et al. 1985, 1988, 1990; BALANYA` et al. 2003). The O chromosome is usually divided into two regions: segment I (about 1/3 of the total O chromosome length), encompassing the area in which the arrangement O34 is located, and segment II (the remaining 2/3 of the chromosome) (KRIMBAS 1993). The Odh gene is located in segment II, and is included in both the O7 and O5 inversions. Moreover, this gene is closely located (5 chromosomal sections) to the proximal breakpoint of the O2 inversion. In all other chromosomal combinations, the Odh gene could freely recombine. The main aim of our research has been to compare the distribution of Odh haplotypes in the chromosomal arrangements found in these two South American populations with that observed in North America (MESTRES et al. 2004). We wanted to analyze which associations between Odh haplotypes and chromosomal arrangements were still present in both colonized areas. Some of these associations are expected to disappear due to DOI: 10.1111/j.2008.0018-0661.02040.x

Hereditas 145 (2008)

Odh gene haplotypes in Chilean populations of D. subobscura

recombination and other forms of genetic exchange (POWELL 1997). Finally, as the sequenced material constituted a collection of lethal chromosomal lines, we also analyzed the relationships among these three markers; i.e. lethal genes, Odh gene haplotypes and chromosomal inversions. We sought to determine whether lethal genes and Odh haplotypes formed some kind of associations with their chromosomal arrangements. MATERIAL AND METHODS Chromosomal lines Populations of Santiago de Chile and Puerto Montt were collected in October and November 1999, respectively (BALANYA` et al. 2003). Lethal chromosomal lines were obtained with the appropriate pattern of crosses with the chcu homokaryotypic and Va/Ba lethal balanced strains (complete details of this cross pattern can be found in MESTRES et al. 1990). The complete list of lethal chromosomal lines sequenced, including their arrangements and accession numbers are presented in Table 1. Finally, the Odh gene from the balancer chromosome Ba was also sequenced. DNA preparation, PCR amplification, and sequencing Total DNA was isolated from a single individual using the method of PASCUAL et al. (1997). To amplify the Odh gene, we used the primers ODH-F and CD4, which had been designed during previous research work (MESTRES et al. 2004). PCR cycling conditions were as follows: 948C for 5 min; 35 cycles of 948C for 1 min, 558C for 1 min, 728C for 1 min; and a final extension of 4 min at 728C. This PCR product was purified using the QIAquick PCR Purification Kit (QIAGEN) while direct sequenced was carried out using the ODH-F, ODHseq-R, C2 and CD6 primers (MESTRES et al. 2004). Cycling conditions were as follows: 968C for 1 min; 25 cycles of 968C for 10 s, 558C (458C for ODH-F primer) for 5 s, 608C for 4 min; and a final extension of 1 min at 48C. Odh genes were sequenced using an ABI PRISMTM 3700 DNA Analyzer at Unitat de Geno`mica, Serveis Cientificote`cnics of the Universitat de Barcelona. Sequence alignment and phylogenetic analysis Sequence alignments were performed with BioEdit v. 4.8.6 (HALL 1999) and SeqManTMII v. 4.03 (DNA Star Inc. 1999). DNA polymorphism was analyzed using DnaSP v. 3.0 (ROZAS and ROZAS 1999). The Kimura 2-parameter distance model and the neighbor-joining (NJ) tree reconstruction algorithm (SAITOU and NEI 1987) were conducted using MEGA v. 2.0 (KUMAR et al. 2001) and TreeView (PAGE 1996).

155

RESULTS AND DISCUSSION Sequence analysis In all, 47 sequences from a 793-nucleotide fragment of the Odh were obtained. This fragment contains part of exon 2, intron 2, exon 3, intron 3 and part of exon 4. One problem encountered during the sequence analysis stemmed from the fact that the chromosomal lines were heterokaryotypic, carrying one Va balancer chromosome and one wild-type chromosome. Thus, two Odh alleles (one from each chromosome) were amplified and sequenced together. It was essential to distinguish between the Odh allele of the Va chromosome and that of wildtype. Clear polymorphic sites were often detected in the chromatograms, wherein one nucleotide belonged to the Odh allele of the Va chromosome. As the Odh allele of the Va balancer chromosome has been well characterized (MESTRES et al. 2004), we used it as a reference for all of our analyses. No insertion/deletion-type polymorphisms were found, in either the exons or the introns. Altogether, 34 nucleotide polymorphic positions were found: 23 in the exons and 11 in the introns (Fig. 1). The number of substitutions proved significantly higher in the introns than in the exons (x2 7.863, DF1, P B0.01). Out of the 23 substitutions located in coding regions, 19 were situated in third positions (17 producing synonymous substitutions) while only four substitutions where not in third positions and proved to be nonsynonymous and affecting only a restricted number of sequences. Using a test based on the binomial distribution (for a complete description of the test see MESTRES et al. 2001), the number of substitutions in third positions was significantly higher than in the two other positions (k]4, PB0.05). When comparing the number of substitutions between exon 3 and exon 4, we found no significant difference (x2 0.102, DF1, P 0.05). Eleven substitutions were detected in the introns: 4 in intron 2 and 7 in intron 3. There was no significant difference between the number of substitutions in either intron (x2 1.147, DF 1, P 0.05). Nucleotide polymorphism and phylogenetic trees The values of the parameters describing the nucleotide polymorphisms for Santiago de Chile and Puerto Montt are presented in Table 2. The nucleotide diversity recorded here is very similar in both studied populations, albeit lightly higher in Santiago de Chile. These estimates can be also obtained by grouping the lethal chromosomal lines in two sets: those in which the Odh gene is located within an inversion (O5 and O347) and those in which it is not (all other

156

L. Go´mez-Baldo´ et al.


Table 1. Description of lethal chromosomal lines sequenced for the Odh gene. Chromosomal line

Arrangement

Population

Accession number

SC1 SC14 SC21 SC31 SC34 SC35 SC37 SC49 SC52 SC58 SC65 SC87 SC105 SC130 SC153 SC162 SC166 SC174 PM2 PM5 PM8 PM14 PM18 PM20 PM27 PM32 PM35 PM39 PM44 PM46 PM57 PM61 PM63 PM70 PM72 PM77 PM86 PM99 PM110 PM113 PM115 PM119 PM121 PM124 PM140 PM144

O3 4 8 O3 4 2 O3 4 8 O3 4 2 Ost O5 O5 O3 4 8 O3 4 7 O5 O3 4 7 O3 4 2 O3 4 7 Ost O3 4 7 O3 4 8 O3 4 8 O5 O3 4 7 O3 4 2 Ost O5 O5 O5 O3 4 2 O3 4 7 O3 4 7 O3 4 7 O3 4 2 O3 4 7 O7 O5 O5 Ost O3 4 8 O3 4 7 O3 4 8 Ost O3 4 7 O5 O3 4 7 O5 O3 4 7 Ost O5 O5

Santiago de Chile ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ Puerto Montt ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’ ’’

EF463054 AJ496680 AJ496689 AJ496680 AJ496695 AJ496667 AJ496667 AJ496689 AJ496670 AJ496667 AJ496714 AJ496680 AJ496714 AJ496695 AJ496714 AJ496695 AJ496689 AJ496667 AJ496670 AJ496695 AJ496695 AJ496667 AJ496667 AJ496667 AJ496680 AJ496670 AJ496670 AJ496670 AJ496680 AJ496670 AJ496714 AJ496667 AJ496667 AJ496695 AJ496695 AJ496670 AJ496689 AJ496695 AJ496714 AJ496667 AJ496670 AJ496667 AJ496670 AJ496698 AJ496667 AJ496667

Ba

Ost

Va/Ba lab. strain

EF467328

chromosomal arrangements). Using this method, the values of the parameters describing this nucleotide polymorphism are shown in Table 3. In Santiago de Chile, the haplotype diversity (h) was lower in the O5 and O347 group. However, this latter group presented higher values for: nucleotide diversity (p), expected average number of nucleotide differences, both per site and by sequence (u), and average number of nucleotide differences (k). This was not the case in

Puerto Montt, where all such parameters proved slightly lower in the group of O5 and O347 chromosomal arrangements. This low level of nucleotide polymorphism in both analyzed populations can be explained by the strong founder event of the colonization. Other molecular markers also showed a drastic reduction in variability in the American continent (LATORRE et al. 1986; ROZAS et al. 1990; ROZAS and AGUADE´ 1991; PASCUAL et al. 2007). As



157

Fig. 1. Summary of polymorphic variation detected in all chromosomal lines sequenced for the three exons and two introns of the Odh gene, using the Va allele as a reference. Lines are clustered according to their haplotype. Hap. I O3 47: one of the haplotypes present in this arrangement (PM2, PM32, PM35, PM39, PM46, PM77, PM115, PM121 and SC52).¯ Hap. II O3 47: the other haplotype present in this arrangement (PM110, SC65, SC105 and SC153). Hap. O5: haplotype associated ¯ to this inversion (PM14, PM18, PM20, PM61, PM63, PM113, PM119, PM140, PM144, SC35, SC37, SC58 and SC174).

few Odh haplotypes reached the American continent and the colonization event still is recent, no differentiation in the nucleotide polymorphism has been observed between the lethal chromosomal lines in which the Odh gene is located inside an inversion (O5 and O347) and those in which it is not. A neighbor-joining phylogenetic tree was constructed in order to obtain an objective classification of the haplotypes based on their similarities (Fig. 2). The Odh sequences are clustered according to their given association with certain chromosomal arrangements, and independently of their origin (Santiago de Chile or Puerto Montt). A clear group, well supported Table 2. Estimates of nucleotide polymorphism of Odh gene obtained in Santiago de Chile and Puerto Montt. nnumber of sequences; Nhap number of haplotypes; rhaplotype diversity; S number of polymorphic sites; p nucleotide diversity; u expected average number of nucleotide differences; k average number of nucleotide differences. Santiago de Chile n Nhap h9SD S p9SD u9SD (per site) k9SD u9SD (per sequence)

Puerto Montt

18 7 0.88290.037 27 0.0118590.00191 0.0099190.00385

28 7 0.77990.043 28 0.0118390.00074 0.00908 90.00172

9.38691.870 7.85091.510

9.37091.832 7.19591.359

by bootstrap values, was formed by the two haplotypes associated with the O347 and O7 chromosomal arrangement (shown at bottom). We obtained a second cluster for all Odh sequences within the O5 inversion (shown at top). The remaining haplotypes were mixed in the central area of the phylogenetic tree. This topology is similar to that obtained in our analysis of Odh sequences from North American D. subobscura samples (Fig. 3, MESTRES et al. 2004) and most likely represents a reminiscence of the strong and still recent founder event. Odh haplotypes and chromosomal inversions It has been well documented that the distribution areas of D. subobscura, both in North and South America, stem from a recent colonizing event (about 30 years ago) (BRNCIC et al. 1981; BECKENBACH and PREVOSTI 1986; PREVOSTI et al. 1987; AYALA et al. 1989; MESTRES et al. 2005). If we assume that under natural conditions D. subobscura produces an average of 5 generations per year (BEGON 1976a, 1976b), then approximately 150 generations have elapsed since the beginning of the colonization. Furthermore, the genetic variability carried by the original colonizers in both areas has proven dramatically similar (PREVOSTI et al. 1988, 1989; AYALA et al. 1989; MESTRES et al. 1990, 1992, 1995, 2005; BALANYA` et al. 1994). The molecular markers analyzed thus far demonstrate this high similarity (LATORRE et al. 1986; ROZAS et al. 1990; ROZAS and AGUADE´ 1991; PASCUAL et al. 2007). For all of these reasons, a similar pattern of Odh molecular haplotypes should

158



Table 3. Estimates of nucleotide polymorphism of Odh gene within different groups of arrangements, both in Santiago de Chile and Puerto Montt. n number of sequences; Nhap number of haplotypes; hhaplotype diversity; S number of polymorphic sites; pnucleotide diversity; u expected average number of nucleotide differences; k average number of nucleotide differences. Santiago de Chile O5 and O3 47 ¯

n Nhap h9SD S p9SD u9SD (per site) k9SD u9SD (per sequence)

8 3 0.67990.122 16 0.0107390.00165 0.0077990.00195 8.50091.908 6.17192.380

Puerto Montt

Other inversions 10 4 0.80090.076 15 0.0085990.00118 0.0066990.00173 6.80091.6643 5.30291.874

likely be found in both colonized regions. In fact, only eight different haplotypes were found in Chile and only 10 in North America (MESTRES et al. 2004). Out of these sequences, seven were common to both colonized regions, while one was exclusive to Chile and three to North America. These results confirm the strong bottleneck of the colonizing event and the high similarity between North and South American populations. Furthermore, it is well known that if such inversions spread after a colonizing event, they will initially reduce the genetic variability within a population due to the strong disequilibrium between the inversions and the genetic sequences within them (HOFFMAN et al. 2004). This reduction has been observed in the present study; e.g. some inversions that had spread after the colonizing event carried the same haplotype (Fig. 1). Additional information can be gleaned if we examine the association between haplotypes and chromosomal arrangements. These associations could be the product of an historical event (the colonizing bottleneck) or represent an effect of natural selection beginning during the early stages of colonization. In the first instance a progressive degradation of these associations would be expected, while in the latter the recombinants would break the supergene which confers the heterosis and thus would be eliminated. Some associations between Odh haplotype and chromosomal inversions have been detected in Chilean populations (Fig. 1, 2) such as were previously observed in North America (MESTRES et al. 2004). It is worth noting that the samples collected for this study were of more recent origin than those from North America, which explains the increased frequency of genetic exchange we detected. All O5 inversions from Santiago de Chile and Puerto Montt featured the same Odh haplotype, identical to that found in all previously sequenced O5 inversions (MESTRES et al. 2004). The Odh gene is

O5 and O3 47 ¯

18 3 0.58290.061 16 0.0091690.00147 0.0058790.00245 7.25591.644 4.65291.163

Other inversions 10 5 0.75690.130 25 0.0110090.00223 0.0111690.00493 8.71191.884 8.83791.768

located inside the O5 inversion, very close to the proximal breakpoint. Furthermore, in American populations all of these inversions are completely associated with the same lethal gene, as well as to the same Odh haplotype (MESTRES et al. 1990, 1992, 1995, 2004). This is not the case with the O347 arrangement; in fact, two different haplotypes have been detected in Chilean populations. These two haplotypes had been previously found in North American populations (MESTRES et al. 2004), as was a third haplotype, although the latter is present neither in Santiago de Chile nor in Puerto Montt. This particular haplotype was detected only once in the lethal chromosomal line FGF5 (from Gilroy, California). It is very similar to one observed in both colonized areas, albeit with a nucleotide change in position 167 (T in the place of C). This line was sequenced twice in order to confirm this change. If this haplotype was included in the original colonizer sample it has remained in the American populations at a low frequency. Thus its probable fate would be to disappear due to the effects of genetic drift. Other explanations include a mutation in this position or some kind of genetic exchange (either recombination or gene conversion) with certain haplotypes presenting T in this site. SCHAEFFER and ANDERSON (2005) found that gene conversion occurs at a higher rate per base than does the mutation rate. The fragment carrying the T in position 167 could originate from other haplotypes (Fig. 1, MESTRES et al. 2004). This genetic exchange event probably took place only in North America since all Chilean O347 samples analyzed thus far do not present this haplotypes. Our analyses of Chilean lethal chromosomal lines carrying the O347 arrangement revealed another evolutionary event: the lack of any correspondence between lethal allelism and the coincidence of haplotypes in this particular arrangement (Table 4). Whilst some O347 arrangements presented only a partial association with



159

Table 4. Distribution of the two haplotypes found in the O347 lethal chromosomal lines from Santiago de Chile and Puerto Montt. These haplotypes are not associated to a single class of lethal genes. Type of lethal gene Incomplete association (Group I) New allelic group (Group II) Lethal genes not allelic (Group III)

Fig. 2. Neighbor-joining phylogenetic tree obtained using all the Odh nucleotide sequences form Santiago de Chile and Puerto Montt. All O5 lethal chromosomal lines are clustered together (top of the figure), while the two haplotypes for the O3 47 (and O7) chromosomal arrangements are grouped ¯ in the bottom of the figure.

a relevant lethal gene (designated Group I and found both in North and South America see previous studies by MESTRES et al. 1990, 1992, 1995, 2004), others (Group II) contained a new lethal gene found twice in Santiago de Chile. Finally, several lethal genes (Group III) were detected only once in the Chilean samples. However, we found no correspondence between groups of lethal genes and Odh haplotypes, which suggests that a free interchange of genetic information occurs among the O347 arrangements in these populations. The most parsimonious

Haplotype I

Haplotype II

PM32, PM35, PM39, PM46 SC52

PM110, SC105, SC153 SC65

PM2, PM77, PM115, PM121

explanation is that genetic recombination inside the O7 inversion breaks any associations between lethal genes and haplotypes in the O347 chromosomal arrangement. Until now, no recombination with other inversions has been detected; the O347 haplotypes are only detected in this arrangement and the same phenomenon occurs with the lethal genes. Furthermore, when we sequenced a lethal chromosomal line carrying only the O7 inversion (PM57), it proved to contain a haplotype found only in the O347 arrangement. Its lethal gene is also allelic with some lethal genes found only in this arrangement. Thus, this infrequent inversion is most likely generated by a rare recombination event in the heterokaryotypes O347/Ost, as hypothesized by SPERLICH and FEUERBACH-MRAVLAG (1974). Finally, there is another association common to American populations, that between the O342 arrangement and a single haplotype (chromosomal lines PM27, PM44, SC14, SC31, SC87) (Fig. 1), even though this arrangement also presents a different haplotype in one chromosomal line (PM5), which is also found in some Ost and O348 arrangements. We described a similar situation in one of our previous studies (MESTRES et al. 2004). As the Odh gene is not included within the O2 inversion, these results can be explained by recombination events occurring between the O342 and O348 or Ost arrangements. The remaining lethal chromosomal lines were Ost and O348, in which the Odh gene is not encompassed by any inversion. As can be observed in Fig. 1, out of the six Ost lethal chromosomal lines analyzed, five presented the same haplotype (PM8, PM70, PM99, SC34 and SC130). This haplotype was also found in the PM72 line (O348) and PM5 (O342), and was also detected in North American samples (MESTRES et al. 2004). The remaining Ost line (PM124) presented a different haplotype, one also common to North American samples (MESTRES et al. 2004). Finally, with regard to the O348 lethal chromosomal lines, three different haplotypes were observed. One occurs in the

160


chromosomal lines PM86, SC21, SC49 and SC166, while the other is carried in PM72 and SC162, the latter proving identical to that observed in Ost chromosomal lines (Fig. 1). Both haplotypes were previously detected in North American populations (MESTRES et al. 2004). However, the third haplotype presented a new sequence of nucleotides (SC1 chromosomal line) (Fig. 1). There are two alternative explanations of its origin: it could be included in the sample of colonizers, maintaining a low frequency in the populations, or it may have arisen by genetic exchange between two pre-existing haplotypes. This sequence was most likely generated by a genetic exchange between a typical O348 haplotype (e.g. that found in the chromosomal lines PM86, SC49, SC21 and SC166) and a haplotype characteristic of O342 chromosomal lines (e.g. those observed in the chromosomal lines PM27, PM44, SC14, SC31 and SC87). The proximal region of the SC1 haplotype contains the nucleotide sequence of the O348 haplotype, while the distal region is characteristic of the O342 haplotype. The exact origin of the central fragment of this sequence (positions 272 to 455) remains uncertain, as it is common to both of the previously mentioned haplotypes. Thus, micro-evolutionary forces are changing the genetic composition of the Chilean colonizing populations. Most genetic markers are, to some degree, associated with the rich chromosomal polymorphism of this species. Indeed, inversions are expected to have some effects on the levels of nucleotide variability (ANDOLFATTO et al. 2001; HOFFMANN et al. 2004). We know that a limited number of Odh haplotypes were included in the original sample of colonizers, which automatically engendered different degrees of associations. If the inversions persist over time we expect that they will slowly accumulate new nucleotide variations via mutation and, most importantly, by gene conversion and crossing-over (NAVARRO et al. 1997, 2000; SCHAEFFER and ANDERSON 2005). While gene conversion rates are expected to remain constant along the inversion, double crossovers seem to be restricted to its central segments (NAVARRO et al. 1997). Throughout our studies on the Odh gene in American populations of D. subobscura, we have observed two types of association between haplotypes and inversions. One type stems from an historic event (the founder event); in this case the association should decrease over time. This is probably the situation that occurs in the haplotype associated with the O342 arrangement. It appears that recombination has been slowly breaking this association. In the second type, the association is promoted by the heterotic effect of the inversion, as in the case of O5 inversion (MESTRES


et al. 2001). In this inversion, the association with both the Odh haplotype and the lethal gene is complete. The association between certain Odh haplotypes and the O347 arrangement could be considered an intermediate case. At least some O347 arrangements seem to have a heterotic effect (MESTRES et al. 2001). Most likely several chromosomes with this arrangement were included in the initial sample of colonizers, each carrying different lethal genes and Odh haplotypes (MESTRES et al. 1990, 1992, 1995, 2004, 2005). It seems that there is a nonnegligible genetic exchange between the different O7 inversions analyzed thus far (Table 4), but the genetic markers found in these chromosomal lines have not been detected in other arrangements. Thus, this genetic exchange seems to be restricted to O7 inversions. Finally, the haplotype found in the SC1 line was most likely generated by either crossing-over or via gene conversion. In summary, our study has revealed that the Odh gene underwent a strong founder event during the colonizing of D. subobscura populations, in both North (MESTRES et al. 2004) and South America (present work). However, these populations have been evolving since the origin of the colonization. Odh gene sequences confirm this phenomenon, since they have facilitated the detection of different types of genetic exchange, although it should be mentioned that nucleotide variation in a large part of D. subobscura genome is thought to be highly structured (MUNTE´ et al. 2005). The micro-evolutionary changes common to these colonizing populations has also been observed at other levels; e.g. variations in the clinal distribution of chromosomal inversions (BALANYA` et al. 2003) and changes in chromosomal polymorphism over time (BALANYA` et al. 2006). Indeed, new inversions have arisen in American populations (PEGUEROLES et al. 1988; BALANYA` et al. 2003). Clines for quantitative traits have also developed in both colonized areas, and in a period of only two decades (HUEY et al. 2000; GILCHRIST et al. 2004). Finally, certain lethal genes have constantly been generated by mutation and then eliminated by selection (or genetic drift), while others have persisted due to their location in co-adapted systems (MESTRES et al. 2001, 2005). Acknowledgements We thank Profs. M. Pascual, J. Balanya`, R. B. Huey and G. W. Gilchrist for the Santiago de Chile and Puerto Montt collections. We also thank Mr. R. Rycroft (S.A.L. Univ. de Barcelona) for corrections to the English manuscript. This work was supported by grants BOS200305904-C02-02 and CGL2006-13423-C02-02 of M. E. C. (Spain) to LS and SGR2005 00995 from the Generalitat de Catalunya (Spain).



REFERENCES Andolfatto, P. and Kreitman, M. 2000. Molecular variation at the In(2L)t proximal breakpoint site in natural populations of D. melanogaster. Genetics 154: 1681 1691. Andolfatto, P., Wall, J. D. and Kreitman, M. 1999. Unusual haplotype structure at the proximal breakpoint of In(2L)t in a natural population of D. melanogaster. Genetics 153: 129713311. Andolfatto, P., Depaulis, F. and Navarro, A. 2001. Inversion polymorphisms and nucleotide variability in Drosophila. Genet. Res. 77: 18. Ayala, F. J., Serra, L. and Prevosti, A. 1989. A grand experiment in evolution: the D. subobscura colonization of the Americas. Genome 31: 246255. Balanya`, J., Segarra, C., Prevosti, A. et al. 1994. Colonization of America by D. subobscura: the founder event and a rapid expansion. J. Hered. 85: 427432. Balanya`, J., Serra, L., Gilchrist, G. W. et al. 2003. Evolutionary pace of chromosomal polymorphism in colonizing populations of D. subobscura: an evolutionary time series. Evolution 57: 18371845. Balanya`, J., Oller, J. M., Huey, R. B. et al. 2006. Global genetic change tracks global climate warming in D. subobscura. Science 313: 17731775. Beckenbach, A. T. and Prevosti, A. 1986. Colonization of North America by the European species, D. subobscura and D. ambigua. Am. Midl. Nat. 115: 1018. Begon, M. 1976a. Dispersal, density and microdistribution in D. subobscura Collin. J. Anim. Ecol. 45: 441456. Begon, M. 1976b. Temporal variations in the reproductive condition of D. obscura Falleń and D. subobscura Collin. Oecologia 23: 3147. Brncic, D., Prevosti, A., Budnik, M. et al. 1981. Colonization of D. subobscura in Chile. I. First population and cytogenetic studies. Genetica 56: 39. Dobzhansky, Th. 1970. Genetics of the evolutionary process. Columbia Univ. Press. Gilchrist, G. W., Huey, R. B., Balanya`, J. et al. 2004. A time series of evolution in action: A latitudinal cline in wing size in South American D. subobscura. Evolution 58: 768780. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/ 98/NT. Nucleic Acid Series 41: 9598. Hoffmann, A. A., Sgro`, C. M. and Weeks, A. R. 2004. Chromosomal inversion polymorphism and adaptation. Trends Ecol. Evol. 19: 482488. Huey, R. B., Gilchrist, G. W., Carlson, M. L. et al. 2000. Rapid evolution of a geographic cline in size in an introduced fly. Science 287: 308309. Knibb, W. R. 1982. Chromosomal inversion polymorphism in D. melanogaster II. Geographic clines and climatic associations in Australasia, North America and Asia. Genetica 58: 213221. Krimbas, C. B. 1993. D. subobscura: Biology, genetics and inversion polymorphism. Verlag Dr. Kovac, Hamburg. Krimbas, C. B. and Powell, J. R. 1992. Drosophila inversion polymorphism. CRC Press. Krimbas, C. B. and Powell, J. R. 2000. Inversion polymorphisms in Drosophila. In: Singh, R. S. and Krimbas, C. B. (eds), Evolutionary genetics. Cambridge Univ. Press, p. 284299.

161

Kumar, S., Tamura, K., Jakobsen, I. B. et al. 2001. MEGA2: Molecular evolutionary genetics analysis software. Bioinformatics 17: 12441245. Kunze-Mu¨hl, E. and und Mu¨ller, E. 1958. Weitere Untersuchungen u¨ber die chromosomale Struktur und die natu¨rlichen Strukturtypen von D. subobscura. Chromosoma 9: 559570. Latorre, A., Moya, A. and Ayala, F. J. 1986. Evolution of mitochondrial DNA in D. subobscura. Proc. Natl Acad. Sci. USA 83: 86498653. Menozzi, P. and Krimbas, C. B. 1992. The inversion polymorphism of D. subobscura revisited: synthetic maps of gene arrangement frequencies and their interpretation. J. Evol. Biol. 5: 625641. Mestres, F., Pegueroles, G., Prevosti, A. et al. 1990. Colonization of America by D. subobscura: lethal genes and the problem of the O5 inversion. Evolution 44: 18231836. Mestres, F., Balanã`, J., Segarra, C. et al. 1992. Colonization of America by D. suboscura: analysis of the inversions from Europe and America and their implications for the colonizing process. Evolution 46: 15641568. Mestres, F., Serra, L. and Ayala, F. J. 1995. Colonization of the Americas by D. subobscura. Lethal-gene allelism and association with chromosomal arrangements. Genetics 140: 12971305. Mestres, F., Balanya`, J., Arenas, C. et al. 2001. Colonization of America by D. subobscura: heterotic effect of chromosomal arrangements revealed by the persistence of lethal genes. Proc. Natl. Acad. Sci. USA 98: 91679170. Mestres, F., Abad, L., Sabater-Munõz, B. et al. 2004. Colonization of America by D. subobscura: association between Odh gene haplotypes, lethal genes and chromosomal arrangements. Genes Genet. Syst. 79: 233244. Mestres, F., Balanya`, J., Pascual, M. et al. 2005. Lethal genes and the colonization of America by D. subobscura. Curr. Topics Genet. 1: 3157. Munte´, A., Rozas, J., Aguade´, A. et al. 2005. Chromosomal inversion polymorphism leads to extensive genetic structure: a multilocus survey in D. subobscura. Genetics 169: 15731581. Navarro, A., Barbadilla, A. and Ruiz, A. 2000. Effect of inversion polymorphism on the neutral nucleotide variability of linked chromosomal regions in Drosophila. Genetics 155: 685698. Navarro, A., Betrań, E., Barbadilla, A. et al. 1997. Recombination and gene flux caused by gene conversion and crossing-over in inversion heterokaryotypes. Genetics 146: 695709. Page, R. D. M. 1996. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12: 357358. Pascual, M., Balanya`, J., Latorre, A. et al. 1997. Analysis of the variability of D. azteca and D. athabasca populations revealed by randomly amplified polymorphic DNA. J. Zool. Syst. Evol. Res. 35: 159164. Pascual, M., Chapuis, M. P., Mestres, F. et al. 2007. Introduction history of D. subobscura in the New World: a microsatellite-based survey using ABC methods. Mol. Ecol. 16: 30693083. Pegueroles, G., Segarra, C. and Prevosti, A. 1988. A new inversion of the E chromosome in D. subobscura. Dros. Inform. Serv. 67: 6465. Powell, J. R. 1997. Progress and prospects in evolutionary biology: the Drosophila model. Oxford Univ. Press.

162


Prevosti, A., Serra, L., Ribo´, G. et al. 1985. The colonization of D. subobscura in Chile. II. Clines in the chromosomal arrangements. Evolution 39: 838844. Prevosti, A., Serra, L., Monclu´s, M. et al. 1987. Colonizacioń de Ame´rica por D. subobscura. Evol. Biol. Col. 1: 124. Prevosti, A., Ribo´, G., Serra, L. et al. 1988. Colonization of America by D. subobscura: experiment in natural populations that supports the adaptive role of chromosomalinversion polymorphism. Proc. Natl Acad. Sci. USA 85: 55975600. Prevosti, A., Serra, L., Aguade´, M. et al. 1989. Colonization and establishment of the Palearctic species D. subobscura in North and South America. In: Fontdevila, A. (ed.), Evolutionary biology of transient unstable populations. Springer, p. 114129. Prevosti, A., Serra, L., Segarra, C. et al. 1990. Clines of chromosomal arrangements of D. subobscura in South America evolve closer to Old World patterns. Evolution 44: 218221. Rozas, J. and Aguade´, M. 1991. Using restriction-map analysis to characterize the colonizing process of D. subobscura on the American continent. I. rp49 region. Mol. Biol. Evol. 8: 447457.


Rozas, J. and Rozas, R. 1999. DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics 15: 174 175. Rozas, J., Hernańdez, M., Cabrera, V. M. et al. 1990. Colonization of America by D. subobscura: effect of founder event on the mitochondrial DNA polymorphism. Mol. Biol. Evol. 7: 103109. Saitou, N. and Nei, M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406425. Schaeffer, S. W. and Anderson, W. W. 2005. Mechanism of genetic exchange within the chromosomal inversions of D. pseudoobscura. Genetics 171: 17291739. Sperlich, D. and Feuerbach-Mravlag, H. 1974. Epistatic gene interaction, crossing over, and linked and unlinked inversions in D. subobscura. Evolution 28: 67 75. Van Delden, W. and Kamping, A. 1991. Changes in relative fitness with temperature among second chromosome arrangements in D. melanogaster. Genetics 127: 507 514. Wasserman, M. 1968. Recombination-induced chromosomal heterosis. Genetics 58: 125139.