• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jun 6, 2006; 103(23): 8757–8762.
Published online May 24, 2006. doi:  10.1073/pnas.0509590103
PMCID: PMC1482651

Amplification of DNA from preserved specimens shows blowflies were preadapted for the rapid evolution of insecticide resistance


Mutations of esterase 3 confer two forms of organophosphate resistance on contemporary Australasian Lucilia cuprina. One form, called diazinon resistance, is slightly more effective against commonly used insecticides and is now more prevalent than the other form, called malathion resistance. We report here that the single amino acid replacement associated with diazinon resistance and two replacements associated with malathion resistance also occur in esterase 3 in the sibling species Lucilia sericata, suggesting convergent evolution around a finite set of resistance options. We also find parallels between the species in the geographic distributions of the polymorphisms: In both cases, the diazinon-resistance change is absent or rare outside Australasia where insecticide pressure is lower, whereas the changes associated with malathion resistance are widespread. Furthermore, PCR analysis of pinned specimens of Australasian L. cuprina collected before the release of organophosphate insecticides reveals no cases of the diazinon-resistance change but several cases of those associated with malathion resistance. Thus, the early outbreak of resistance in this species can be explained by the preexistence of mutant alleles encoding malathion resistance. The pinned specimen analysis also shows much higher genetic diversity at the locus before organophosphate use, suggesting that the subsequent sweep of diazinon resistance in Australasia has compromised the scope for the locus to respond further to the ongoing challenge of the insecticides.

Keywords: Lucilia cuprina, Lucilia sericata, mutation, convergent evolution, organophosphates

One of the major unresolved issues in our understanding of adaptive evolutionary change is the extent to which the acquisition of newly beneficial traits proceeds by the selection of existing (preadaptive) rare variants, rather than waiting on new mutations to arise once the selection pressure is imposed. These two scenarios could have very different consequences for the nature of the acquired adaptation, its rate of uptake, and the fate of genetic variation in the chromosomal region surrounding the selected locus. Direct evidence discriminating the two possibilities in natural populations of eukaryotes has been lacking because the critical data needed must include samples that have not been exposed to the selection pressure.

The phenomenon of insecticide resistance is proving an informative model for studying microevolutionary processes in eukaryotes because it represents a rapid contemporary acquisition of a major new biochemical phenotype and it generally has a relatively simple genetic basis. The rapidity with which resistance often follows the first use of new pesticides has prompted much speculation about preadaptation at the relevant loci (1, 2). However, many cases have also been recorded where resistant genotypes have fitness costs in the absence of insecticide (3, 4), which would seem at odds with the preadaptation hypothesis. In this paper, we present evidence that tests the hypothesis directly by retrospective sampling of old preserved blowfly specimens. We amplify alleles at the resistance locus in pinned specimens dating back >75 years, to well before the first use of the insecticides.

The sibling species Lucilia cuprina and Lucilia sericata have become primary ectoparasites of sheep in many farming systems around the world (5). Their distributions overlap significantly, although L. cuprina predominates in warm temperate or subtropical climates and L. sericata is more abundant in cool temperate regions. Some of their spread has been relatively recent; L. cuprina is likely to have reached Australia in the early nineteenth century, Hawaii around 1947, and New Zealand in the 1980s. Organophosphate (OP) insecticides were first used to control the two species in Australia in 1955, <5 years after their first use against any pests anywhere in the world (6).

Esterase-based metabolic resistance to OPs was evident in Australian L. cuprina by 1965, with polymorphism for two forms of resistance evident by 1968 (6). The major form, called diazinon resistance, now exceeds frequencies of 90% in Australia and New Zealand, with the minor form, called malathion resistance, around 5–10%, and susceptible variants are quite rare (7, 8). There is much less information available for L. cuprina populations outside Australasia, many of which are much less exposed to OPs, but bioassay data suggest that populations in South Africa also segregate for at least one of these resistances (6, 9). Likewise, there is little information on L. sericata beyond bioassay data for New Zealand populations that suggest that distinct diazinon- and malathion-resistant genotypes also segregate in this species, albeit at lower frequencies than their L. cuprina counterparts (10).

We previously described two single nucleotide changes in the αE7 gene encoding the esterase 3 (E3) that confer the two forms of resistance in Australasian L. cuprina. Diazinon resistance is because of a Gly137Asp mutation in E3, which bestows a new OP hydrolase activity on the enzyme, although it abolishes the enzyme’s native carboxylesterase activity (1113). Malathion resistance is because of a Trp251Leu mutation, which confers a lower level of OP hydrolase activity but has much less effect on carboxylesterase activity (1315). The Asp-137 enzyme is more active on OPs such as diazinon that have diethyl, rather than dimethyl substituents, and this biochemical preference corresponds with higher resistance ratios in flies treated with these compounds. Conversely, the Leu-251 enzyme has greater OP hydrolase activity for compounds, including malathion that have dimethyl substituents, and, again, this biochemical preference is reflected in the corresponding resistance ratios. Malathion also differs from most other dimethyl OPs in that it contains carboxylester bonds, and the ability of Leu-251 E3 to break these bonds as well is associated with particularly high resistance for this compound.

Intriguingly, the same two mutations in an E3 orthologue have been associated with equivalent forms of resistance in the house fly, Musca domestica, although, in this case, a second mutation of Trp-251 to another small amino acid (Ser) is also associated with malathion resistance (16, 17). Consistent with these findings are the results of Heidari et al. (18), who analyzed a comprehensive series of synthetic mutants at the 137 and 251 sites of E3 in vitro. Although no other change at 137 was found to create equivalent levels of OP hydrolase activity, changes at 251 to any of five small residues (Leu, Ser, Gly, Thr, and Ala) all gave significant OP hydrolase and malathion carboxylesterase activities.

Our goal in this study was to elucidate the evolutionary genetic mechanisms by which E3-based resistances had evolved in the two Lucilia species. To this end, we first characterized extant variation in E3 in samples of the two sibling species in populations with widely differing histories of OP exposure collected from around the world. We then deployed PCR protocols that we had developed for the amplification of nuclear genes from preserved specimens to ascertain E3 polymorphisms in pinned L. cuprina collected before the first use of OPs. We found Asp-137 in extant populations of both species, although it was largely confined to Australasia where OP pressure is so high, and we did not find it in any of the pre-OP samples. However, we found that both Leu- and Ser-251 variants were widely distributed in extant populations of both species, and we recovered both Leu- and Thr-251 variants in the pre-OP material. Thus L. cuprina and L. sericata have evolved precisely parallel E3-based resistances, and L. cuprina was preadapted for this development by the prior existence of malathion resistant genotypes.


Post-OP E3 Polymorphism in L. cuprina.

A total of 46 E3 sequences have been obtained from Australasian L. cuprina collected during the last 20 years since resistance rose to high frequencies (22 randomly sampled and also bioassayed for resistance status; 5 randomly sampled but not bioassayed; and 19 preselected for malathion resistance). These sequences (Fig. 1 and Table 1) encompass seven LcαE7 haplotypes comprising three OP-susceptible Gly-137-Trp-251 combinations (LcA, LcB, and LcC), two diazinon-resistant Asp-137-Trp-251 combinations (LcD and LcG), and two malathion-resistant Gly-137-Leu-251 combinations (LcE and LcF). All but one of the haplotypes (LcF) occur in the randomly sampled lines, with the diazinon-resistant LcD haplotype predominating (67%), and the frequencies of the diazinon-(Asp-137) and malathion-(Leu-251) resistance replacements being 78% and 7%, respectively.

Fig. 1.
Haplotypes generated from the near full-length sequence of the αE7 gene from extant global populations of L. cuprina and L. sericata. Full alignments are presented in Fig. 4, which is published as supporting information of the PNAS web site. Relationships ...
Table 1.
Frequencies of amino acid replacements associated with diazinon-(G137D) and malathion-(W251L/S/T) resistance plus nucleotide (pi) and haplotype diversities (Hd) across the sequenced regions

Only two of eight E3 sequences obtained from outside Australasia (Fig. 1 and Table 1; no bioassay data available) have a haplotype seen in the Australasian samples. These sequences are two further occurrences of the diazinon-resistant haplotype LcD from Hawaii, which raises the possibility that resistance in Hawaii might have been imported with immigrant flies of Australasian origin (5, 19). There are no other occurrences of the Asp-137-Trp-251 combination associated with diazinon resistance in this nonAustralasian sample, although there are just two Gly-137-Trp-251 combinations that we would infer to be OP susceptible (LcH and LcI from central Africa). Three sequences covering two haplotypes (LcJ from North America and Africa and LcK from Asia) have the Gly-137-Leu-251 combination associated with malathion resistance. There is also one additional combination, Gly-137-Ser-251 in haplotype LcL from South Africa, previously not seen in L. cuprina but associated with malathion resistance in M. domestica and found to confer OP hydrolase and malathion carboxylesterase activities on the E3 enzyme in vitro (see introduction). Overall, the widespread occurrence and high frequency (≈50%) of Leu- or Ser-251 sequences outside Australasia suggest that malathion resistance is quite common and widespread.

The various resistance-associated haplotypes are well dispersed across the allele phylogeny for E3 in L. cuprina (Fig. 1), but this pattern does not necessarily indicate polyphyletic origins for any of the relevant amino acid replacements. The Hudson and Kaplan (20) test implies a minimum of ten recombination events (Rm = 10) in these data.

Nucleotide and haplotype diversities calculated from the post-OP samples are >50% higher outside than within Australasia (Table 1). One explanation might be a loss of diversity associated with the selective sweep of diazinon-resistant Gly-137-Trp-251 alleles to high frequencies in Australasia. Coalescence simulations concur, showing that haplotype diversity is much lower than expected from the observed level of nucleotide variation in Australasia (Hd = 0.55 vs. 0.95; P < 0.001), but consistent with such expectations in the rest of the world (Hd = 0.93 vs. 0.97; P > 0.05).

Post-OP E3 Polymorphism in L. sericata.

The five L. sericata sequences collected from Australasia during the last 20 years (Fig. 1 and Table 1; no bioassay data available) cover four haplotypes, three of them (LsA, LsB, and LsC) with the Gly-137-Trp-251 combination which the L. cuprina precedent suggests would correspond to OP susceptibility. The other haplotype (LsD), recovered once from New Zealand, has the Asp-137-Trp-251 combination that the precedent predicts would confer diazinon resistance. This finding is consistent with the known occurrence of diazinon resistance in L. sericata in New Zealand (10).

The twelve L. sericata sequences collected from outside Australasia during the last 20 years (Fig. 1 and Table 1; no bioassay data available) represent eight haplotypes, with all but one of them (LsA from North America) not seen in the Australasian sample. Five of these eight (LsA again, plus LsE, LsF, LsG, and LsH) contain the presumptively OP-susceptible Gly-137-Trp-251 combination, and one of these haplotypes (LsE) was found in three of the European samples. The other three have the presumptively malathion-resistant Gly-137-Leu-251 (LsI and LsJ from North America and Europe) and Gly-137-Ser-251 (LsK in two sequences from southern Africa) combinations.

Thus, the L. sericata data show the same polymorphisms, and indeed combinations of polymorphisms, at the key 137 and 251 residues as the L. cuprina data. The combination associated with diazinon resistance is less common and only recovered from Australasia, but combinations associated with malathion resistance are widespread.

There were too few data for L. sericata to compare nucleotide and haplotype diversities geographically, but overall values for this species (Table 1) are similar to those above for L. cuprina outside Australasia, and the overall haplotype diversities for this species are not different from those expected from its nucleotide variation (Hd = 0.93 vs. 0.95; P > 0.05).

Pre-OP Polymorphism in the Codon 137 Region in Australasian L. cuprina.

All 16 of the Australasian L. cuprina specimens collected before the first use of OPs (i.e., before 1950), from which we sequenced the 153-bp region including the 137 codon (Fig. 2 and Table 1), contain a GGT codon specifying the Gly associated with OP susceptibility. Thus, there is no evidence in this admittedly small sample for the preexistence of diazinon resistance before the use of OP insecticides.

Fig. 2.
Haplotypes generated by the ten variable sites in the 153-bp Gly-137 region of LcαE7 in all of the pre- and post-OP samples. Haplotypes only scored in the 153-bp PCR amplicon (i.e., in the pre-OP data) are given a truncated sequence over the diazinon-resistance ...

Elsewhere in this pre-OP amplicon, we found seven single nucleotide polymorphisms, three of them amino acid replacements (Asn122Thr, Leu130Ser, and Gly168Arg), generating seven haplotypes (Fig. 2). None of these three replacements lie near the catalytic machinery of the enzyme, nor would they be expected to impact on its hydrolytic activity (Table 4). Three of the polymorphisms and four of the haplotypes do not occur in the post-OP samples, either in Australasia or elsewhere. Levels of nucleotide and haplotype diversity across the amplicon (Table 1) are approximately twice those seen in this region in Australasia since resistance reached high frequencies, but they are very similar to the levels seen in the post-OP sample from outside Australasia. Thus, there has been substantial loss and restructuring of variation in this 153-bp region in Australasian L. cuprina since resistance swept through the population, and most of the variation missing in the post-OP material is in the form of OP susceptible haplotypes.

The allelic phylogeny of all of the pre- and post-OP sequences across this region (Fig. 2) is consistent with a monophyletic origin of the G137D resistance mutation. Indeed, the two haplotypes in question, LcD and LcG, have identical sequences over this 153-bp region.

Pre-OP Polymorphism in the Codon 251 Region in Australasian L. cuprina.

The 21 pre-OP specimens collected before 1950 that we sequenced across the 251 codon (Fig. 3 and Table 1) segregate for three different versions of this codon, specifying three different amino acids. Seventeen specimens contain the TGG Trp codon associated with OP susceptibility, two specimens contain the TTG Leu codon found to confer malathion resistance, and two specimens contain an ACG codon specifying a Thr residue. No specimens have the TCG Ser codon associated with malathion resistance, although it may have existed in pre-OP populations because it is an intermediate state between the TGG Trp and ACG Thr codons. We expect that Thr-251 sequences would also confer malathion resistance because our earlier experiments with synthetic mutants showed that this residue and four other small residue substitutions at the 251 site confer OP hydrolase and malathion carboxylesterase activities (18). Thus, consistent with the widespread occurrence of Leu- and Ser-251 sequences in post-OP samples not heavily exposed to OPs, it is clear that alleles associated with malathion resistance occurred in Australasian L. cuprina before they were exposed to OP insecticides. Moreover, such alleles were not uncommon (4/21 = 19%).

Fig. 3.
Haplotypes generated by the 14 variable sites in the 130-bp Trp-251 region of LcαE7 in all of the pre- and post-OP samples. Haplotypes only scored in the 130-bp PCR amplicon (i.e., in the pre-OP data) are given a truncated sequence over the malathion-resistance ...

The 130-bp amplicon, including the codon for residue 251, contains a total of 15 single nucleotide polymorphisms in the pre-OP data (Fig. 3). As well as the 251 variation, there is one other amino acid replacement, Leu228Ser, which would not be expected to impact on resistance status (Table 4). These pre-OP polymorphisms are organized in ten haplotypes, of which six are presumptively OP susceptible. Five of the polymorphisms and six of the haplotypes, three of them presumptively OP susceptible, do not occur in any post-OP samples. Conversely, there were two polymorphisms and seven haplotypes (four of them susceptible) across this region that were scored in the post-OP samples but were not seen in the pre-OP material. Unlike the 137 region/diazinon-resistance data, there has thus been no disproportionate loss of malathion-susceptible haplotypes from the 251 region since the use of OP insecticides.

Overall, however, as with the 137 data, the 251 sequences reveal levels of diversity in the pre-OP Australasian material that are approximately twice those found in post-OP Australasian samples and comparable with those found in post-OP samples from outside Australasia (Table 1). This difference is likely to reflect the hitchhiking of haplotypes in this region with those conferring diazinon resistance in the 137 region that now predominate in Australasian L. cuprina. There is certainly sufficient disequilibrium across the two regions to drive such hitchhiking; restriction fragment length polymorphism (RFLP) studies show that such hitchhiking effects could extend for many genes either side of LcαE7 (8, 13).

The allele phylogeny of all of the pre- and post-OP L. cuprina sequences across the 130-bp region, including the 251 codon (Fig. 3), suggests monophyletic origins for Ser 251 (just one distinct haplotype) and Thr 251(two sister group haplotypes) but not for the Leu 251 form of malathion resistance (five haplotypes in at least three lineages). Application of the Hudson and Kaplan (20) test only requires a minimum of one recombination event to explain the data for the 251 region (Rm = 1) but does not preclude the possibility of others, so no firm conclusion about the number of origins for Leu-251 can be made.


Our finding that L. sericata is polymorphic for the same Gly137Asp and Trp251Leu substitutions in E3 that confer diazinon and malathion resistance in L. cuprina parallels the finding of these mutations in the more distantly related M. domestica (16, 17). Although a causal link with resistance has not been established experimentally in L. sericata and M. domestica, the fact that resistance phenotypes with similar bioassay and/or biochemical properties have been reported in these species strongly suggests a causal connection. Assuming causality, the data for the three species constitute evidence for biochemically precise convergent evolution and suggest that the options for evolving esterase-based metabolic resistance to OPs are tightly constrained, at least in these species.

Our findings of Ser-251 and Thr-251 variants of E3 that might also encode malathion resistance in L. cuprina and/or L. sericata also have striking parallels with the biochemical analyses of synthetic E3 variants by Heidari et al. (18), which found that substitution of Trp-251 to any of five small amino acids, including these two residues and Leu, produced an enzyme with some OP hydrolase and malathion carboxylesterase activities. Our data provides further support for their predictions that these changes would encode resistance, and the finding of the Ser substitution across the two Lucilia species and M. domestica (17) adds to the evidence above for convergent resistance evolution across these higher Diptera.

It is also noteworthy that mutations of E3 251 to Leu, Ser, or Thr are quite widespread in post-OP samples of both L. cuprina and L. sericata, and also in the pre-OP samples of L. cuprina. By contrast, the Gly137Asp mutation conferring diazinon resistance is restricted to post-OP samples of the two species that are heavily exposed to the insecticide and is not recovered from the pre-OP L. cuprina material. These results can clearly explain why OP resistance was recorded so early after these flies were first exposed to OP insecticides in Australasia (see introduction). In respect of malathion resistance, they provide unambiguous evidence for the preadaptative occurrence of beneficial genetic variation. For diazinon resistance, however, the data suggests an origin that postdates the introduction of the selection pressure.

So why are the species preadapted for malathion, but not diazinon, resistance, and why has the diazinon-resistance mutation come to predominate, particularly in Australasian populations of L. cuprina? Relevant here is the fact that diazinon has been the most commonly used OP insecticide for Lucilia control in Australasia from the outset, and, although the so-called malathion resistance mutations also give protection against diazinon, that protection is several-fold less effective than afforded by the diazinon-resistance mutation (14, 15). Thus the Lucilia populations have been preadapted with mutations conferring relatively low levels of diazinon resistance, but these alleles are now being overtaken by a more effective form of diazinon resistance.

We suspect that this order of events also reflects the relative effects of the two classes of mutation on the wild-type function of the E3 enzyme. Although that function is not known, it is clear that the malathion-resistance mutations allow the enzyme to retain much of the carboxylesterase activity seen in susceptible forms, whereas the diazinon-resistance mutation abolishes it (12, 18). Further, there is a clear fitness cost for the diazinon-resistance mutation in the absence of the insecticide, which is expressed in the form of developmental instability, but no such cost is documented for the malathion-resistance mutations (3). The diazinon-resistance mutation has only been recovered from populations almost continuously exposed to the insecticide and even then has only risen to its current high frequency after the proliferation of a mutation at a separate Modifier locus, which, by means unknown, redresses its fitness cost in the absence of the insecticide (3). It is not surprising then that the diazinon-resistance mutation was not evident in our pre-OP samples.

However, it remains intriguing as to why some of the malathion-resistance mutations are at appreciable frequencies in our pre-OP material (and in contemporary Lucilia populations less heavily treated with OPs). We cannot resolve this question without knowing the physiological function of the enzyme in the absence of insecticide, but biochemical data suggests that these mutants are not selectively equivalent to the wild-type enzyme under these circumstances. Some of the mutations at site 251 have radical effects on the stereo preferences of the enzymes for various racemic carboxylester substrates in vitro (21). Interestingly, such substrates include various synthetic pyrethroid insecticides. Although pyrethroids are not currently used to control Lucilia, it may be that the flies would also show a level of preadaptation to them on account of these mutations.

Diazinon continues to be the OP of choice for Lucilia control in Australasia, because even the diazinon-resistance mutation affords only poor OP hydrolase activity, which can be overcome by treatment with higher doses of the insecticide; although the mutant enzyme binds with very high affinity, its turnover rate for the OP is slow (Km in the low μM range; kcat ≈3 h−1; see refs. 12 and 18). By implication, there may be ongoing selection for more effective OP detoxification by E3 in the field. Further resistance evolution has indeed occurred, in the form of duplications of the E3 locus to place both malathion- and diazinon-resistance alleles on the one chromosome, but the additional advantage of these genotypes is likely relatively small (8).

One problem for the further evolution of E3-based resistance is the loss of variability at the locus because of the sweep of the diazinon-resistance mutation to high frequency. Compounding this problem is the apparent difficulty of generating mutant E3s with significantly better OP kinetics than the Gly137Asp mutant by single nucleotide changes. This problem is suggested both by the recurrence of the same mutations in natural populations of the different species and by the inability of Heidari et al. (18) to make more effective variants in vitro. Heidari et al. (18) tested many synthetic mutations at the 137 and 251 sites, as well as at other sites around the active site of E3, which structural modeling suggested might affect OP hydrolase activity. None of the mutants, or any intragenic recombinants that combined the resistance-associated states at the 137 and 251 sites, were individually superior to Gly137Asp in their OP hydrolase activities.

There is a stark contrast between the performance of this insect pesticide detoxification system and those of the gene/enzyme systems that have proliferated in the same time frame in soil bacteria to enable them to use OPs as a nutrient source (17, 22). These systems, which are often many substitutions removed from their putative pre-OP ancestors, are many orders of magnitude more efficient kinetically than the E3 mutants we have investigated; indeed, they approach the diffusion-limited theoretical maxima for the relevant kinetic parameters. They have also spread widely both phylogenetically (across Gram-positive and Gram-negative bacteria) and geographically (across several continents). Homologues of some of them exist in insects, but no mutants of these systems with OP hydrolase activity have been recovered from insects (17). The picture thus emerging in the current case of eukaryote biochemical evolution is of opportunistic use of genetic variation already available or accessible by single nucleotide changes that produces an adaptation well short of optimal.

Finally, we note that our success in amplifying nuclear gene sequences from pinned specimens collected up to 75 years ago suggests that invertebrate material preserved in such a way may permit temporal surveys of polymorphisms for a wide variety of traits and species. The sequences we amplified were only ≈150-bp long, and only ≈30% of the older specimens tested produced the desired amplicons. Nevertheless, our success indicates that a range of evolutionary phenomena associated with invertebrate responses to environmental changes during the last century should now be amenable to post hoc genetic analysis.

Materials and Methods

We have previously reported genomic sequence for αE7 from 41 isogenic strains of Australasian L. cuprina collected since resistance reached high frequencies (generally in the early 1990s) and from one L. sericata strain of unknown resistance status collected in Australia in 1993 (11, 13). Additional data reported herein cover a further 13 L. cuprina and 16 L. sericata strains collected in the 1990s. Five of the new L. cuprina strains and four of the new L. sericata strains were from Australasia, the rest from Africa, North America, Europe, Asia, and Hawaii. Details of these strains are in Table 2.

Genomic DNA was extracted from the new strains as described in Stevens et al. (19), and a region of ≈1.2-kb region in the LcαE7 gene (coordinates 334-1080 in LcαE7 cDNA; see ref. 23) was amplified from this DNA as described in Newcomb et al. (13). Resulting PCR products were cloned into pBluescript (Stratagene) and sequenced on both strands by using the Prism Ready Reaction Dye Deoxy Terminator cycle sequencing kit (PerkinElmer) following manufacturer’s instructions. Sequencing reactions were resolved on an Applied Biosystems Model 377 automated DNA sequencer. Multiple clones were sequenced until at least two identical sequences were recovered, resulting in one allele being sampled per fly.

To assess LcαE7 variation before the use of OPs, we sequenced a 153-bp amplicon (coordinates 610–762 in the LcαE7 cDNA) covering the Gly137Asp codon in 16 pinned specimens collected in Australia from 1930–1949 and a nonoverlapping 130-bp amplicon (coordinates 949-1078 in LcαE7 cDNA) covering the Trp251Leu codon in 21 specimens from the same time period. Genomic DNA was extracted from these specimens by using a modification of the methods of Junqueira et al. (24) and the DNEasy or QIAamp DNA tissue kit (both Qiagen). To prevent contamination with DNA from extant material, DNA from the legs of pinned specimens was prepared in a physically separate, E3-naïve laboratory with fresh, sterile glass and plastic ware. PCRs were prepared in a freshly UV-sterilized flowhood by using primers designed to be both species and gene specific. Negative control reactions prepared in this way, but without adding template DNA, were always carried out in parallel with amplifications from pinned material and never yielded E3 amplicons. The genomic regions described above were amplified by using Expand Hi-Fidelity PCR system (Roche) with the primers GGCTCAGAGGATTGTCTATACC and GCTCCCAAACGATATTGTATG for the 153-bp fragment encompassing codon 137 and ACAGTCTTTGGTGAAAGTGCCG and TGGCTAAGGTGAAGGCACGATG for the 130-bp fragment encompassing codon 251. Resulting PCR products were cloned into pTOPO-CR2.1 (Invitrogen) and sequenced.

Phylogenetic trees were constructed by using the Neighbor-Joining method, and Tajima-Nei nucleotide distances implemented in phylip with 1,000 bootstrap replicates (25). Various population genetics statistics were calculated by using dnasp version 3.50 (26). For coalescent simulations of haplotype diversities, 10,000 random simulations were performed under a no recombination model with the number of segregating sites fixed at 100.

Supplementary Material

Supporting Information:


We thank Richard Wall (University of Bristol, Bristol, U.K.), Dianne Gleeson (Landcare, New Zealand), Allan Heath (Ag Research, New Zealand), Rod Mahon (Department of Entomology, Commonwealth Scientific and Industrial Research Organization, Australia), and Theuns van der Linde (University of Free State, South Africa) for samples of L. cuprina and L. sericata from around the world for the first two data sets, and the Australian National Insect Collection for the pinned specimens of historical L. cuprina collections for our third data set. The financial support of Wellcome Trust Biodiversity Fellowship 050808/Z/97 and The Royal Society of New Zealand, Marsden Fund HRT 802 is gratefully acknowledged by R.D.N. and C.G.Y.


esterase 3


Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Data deposition: The sequences reported in this paper are variants of esterase 3, as detailed in Figs. 2–4, and have been deposited in the GenBank database (accession no. U56636).


1. Crow J. F. Annu. Rev. Entomol. 1957;2:227–246.
2. ffrench-Constant R. H., Pittendrigh B., Vaughan A., Anthony N. Philos. Trans. R. Soc. London B. 1998;353:1685–1693. [PMC free article] [PubMed]
3. Batterham P., Davies A. G., Game A. Y., McKenzie J. A. BioEssays. 1996;18:841–845. [PubMed]
4. Coustau C., Chevillon C., ffrench-Constant R. Trends Ecol. Evol. 2000;15:378–383. [PubMed]
5. Gleeson D. M., Sarre S. Bull. Entomol. Res. 1997;87:265–272.
6. Hughes P. B., McKenzie J. A. In: Combating Resistance to Xenobiotics: Biological and Chemical Approaches. Ford M., Holloman D., Khambay B., Sawicki R., editors. Chichester, England: Ellis Horwood; 1987. pp. 162–177.
7. Levot G. W. Int. J. Parasitol. 1995;11:1355–1362. [PubMed]
8. Smyth K. A., Boyce T. M., Russell R. J., Oakeshott J. G. Heredity. 2000;84:63–72. [PubMed]
9. Blackman G. G., Bakker J. A. J. S. Afr. Vet. Assoc. 1975;46:337–379. [PubMed]
10. Wilson J. A. Ph.D. thesis. Wellington, New Zealand: Victoria University of Wellington; 1999.
11. Newcomb R. D., Campbell P. M., Ollis D. L., Cheah E., Russell R. J., Oakeshott J. G. Proc. Natl. Acad. Sci. USA. 1997;94:7464–7468. [PMC free article] [PubMed]
12. Devonshire A. L., Heidari R., Bell K. L., Campbell P. M., Campbell B. E., Odgers W. A., Oakeshott J. G., Russell R. J. Pestic. Biochem. Physiol. 2003;76:1–13.
13. Newcomb R. D., Gleeson D. M., Yong C. G., Russell R. J., Oakeshott J. G. J. Mol. Evol. 2005;60:207–220. [PubMed]
14. Campbell P. M., Yen J. L., Masoumi A., Russell R. J., Batterham P., McKenzie J. A., Oakeshott J. G. J. Econ. Entomol. 1998;91:367–375.
15. Campbell P. M., Newcomb R. D., Russell R. J., Oakeshott J. G. Insect Biochem. Mol. Biol. 1998;28:139–150.
16. Claudianos C., Russell R. J., Oakeshott J. G. Insect Biochem. Mol. Biol. 1999;29:675–686. [PubMed]
17. Claudianos C., Crone E., Coppin C., Russell R., Oakeshott J. In: Agrochemical Resistance: Extent, Mechanism, and Detection. Marshall Clark J., Yamaguchi I., editors. Washington, DC: Am. Chem. Soc.; 2001. pp. 90–101.
18. Heidari R., Devonshire A. L., Campbell B. E., Bell K. L., Dorrian S. J., Oakeshott J. G., Russell R. J. Insect Biochem. Mol. Biol. 2004;34:353–363. [PubMed]
19. Stevens J. R., Wall R., Wells J. D. Insect Mol. Biol. 2002;11:141–148. [PubMed]
20. Hudson R. R., Kaplan N. L. Genetics. 1985;111:147–164. [PMC free article] [PubMed]
21. Heidari R., Devonshire A. L., Campbell B. E., Dorrian S. J., Oakeshott J. G., Russell R. J. Insect Biochem. Mol. Biol. 2005;35:597–609. [PubMed]
22. Horne I., Qiu X. H., Russell R. J., Oakeshott J. G. FEMS Microbiol. Lett. 2003;222:1–8. [PubMed]
23. Newcomb R. D., Campbell P. M., Russell R. J., Oakeshott J. G. Insect Biochem. Mol. Biol. 1997;27:15–25. [PubMed]
24. Junqueira A. C. M., Lessinger A. C., Azeredo-Espin A. M. L. Med. Vet. Entomol. 2002;16:39–45. [PubMed]
25. Felsenstein J. J. Theor. Biol. 1991;152:357–376. [PubMed]
26. Rozas J., Rozas R. Bioinformatics. 1999;15:174–175. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...