Extreme Mitochondrial DNA Variability and Lack of Genetic ... - BioOne

ECOLOGY AND POPULATION BIOLOGY

Extreme Mitochondrial DNA Variability and Lack of Genetic Structure in Populations of Dermatobia hominis (Diptera: Cuterebridae) from Brazil S. R. GEURGAS, M. E. INFANTE-MALACHIAS,

AND

A. M. L. AZEREDO-ESPIN

Departamento de Gene´ tica e Evoluç aõ, Instituto de Biologia e Centro de Biologia Molecular e Engenharia Gene´ tica, Universidade Estadual de Campinas. Box 6109; CEP 13083Ð970, Campinas, Brazil

Ann. Entomol. Soc. Am. 93(5): 1085Ð1094 (2000)

ABSTRACT Restriction fragment-length polymorphism analysis of mitochondrial DNA (mtDNA) was used to characterize the genetic variation and population structure of the human bot ßy, Dermatobia hominis (Diptera: Cuterebridae), in parasite populations from cattle in southeastern, southern and central regions of Brazil. Forty-eight haplotypes with a nucleotide sequence divergence of 2.75% were found among 227 individuals. Haplotypes could be divided into three clades, with interclade variation ranging from 0.94% to 2.00%. The degree of differentiation obtained suggests that the mitochondrial clades may have differentiated in allopatry, and that their contemporary pattern of distribution probably results from secondary contact of isolated populations, reinforced by the introduction and movement of livestock in South America. KEY WORDS Diptera, Dermatobia hominis, restriction fragment-length polymorphism, mitochondrial DNA, genetic structure, variability

THERE ARE SOME 20 species of obligate myiasis ßies worldwide, whose larvae parasitize the living vertebrates. Of these, Dermatobia hominis L. (Guimara˜es et al. 1983), the human bot ßy, is remarkable in its ability to parasite both humans and livestock. Larvae of D. hominis penetrate the hostÕs skin without need for lesions and produce a furuncular myiasis, developing in the subcutaneous tissue of the host during most of their life-time (review in Guimara˜es et al. 1983, Hall and Wall 1995). Although there is usually just one larva per myiasis, infestations can be intense, causing great economic loss and health problems. For this reason, D. hominis is considered the most important agent of primary myiasis in cattle and one of the most important livestock pests in Latin America (Guimara˜es et al. 1983, Sancho 1988, Thomas 1988). This species, an inhabitant of forests, is endemic from northern Mexico to southern Argentina, except Chile (Guimara˜es et al. 1983). Dermatobia hominis has two characteristics that make it distinctive from other myiasis ßies. Although most myiasis ßies are host speciÞc, the human bot ßy may infect a variety of vertebrates. The second and most fascinating characteristic is the use of an intermediate vector for dispersal. Adult females oviposit on the abdomens of carriers, usually zoophilic ßies or mosquitoes, during ßight. When the carrier settles on a warm-blooded animal, the eggs hatch in response to heat, releasing the larvae that rapidly penetrate the skin (Guimara˜es et al. 1983, Hall and Wall 1995). Because adults live only a short time (⬇3 d), dispersal may be dependent on the movement of the carriers and hosts. Although there are no genetic studies of

natural D. hominis populations, the introduction and movement of livestock in South America have probably modiÞed the original D. hominis population structure, as is believed to have happened to other myiasis ßies (Roehrdanz and Johnson 1988, Infante and Azeredo-Espin 1995, Taylor et al. 1996, Stevens and Wall 1996). Because livestock is currently the primary D. hominis host, we assessed the genetic variation and the population structure of this species by analyzing restriction site variation in mitochondrial DNA (mtDNA) using populations derived from cattle. Maternal inheritance, rapid evolution, and high polymorphism among different populations make mtDNA an ideal genetic marker for studying population structure and for estimating variation in species and populations (Avise et al. 1987). The information obtained by analyzing the phylogeny of mtDNA haplotypes and their pattern of distribution within a speciesÕ range provides insights into populational processes across ecological and evolutionary time (Avise et al. 1987, Avise 1994). Materials and Methods Two hundred and twenty-seven larvae of D. hominis were collected from 14 localities in southeastern, southern, and central Brazil (Fig. 1): Aragoiaˆnia (n ⫽ 13) and Piracanjuba (n ⫽ 5) in the State of Goia´s, Alfenas (n ⫽ 37) and Braso´ polis (n ⫽ 22) in the State of Minas Gerais, Ponta Grossa (n ⫽ 20) in the State of Parana´, and Campinas (n ⫽ 8), Garç a (n ⫽ 9), Indaiatuba (n ⫽ 11), Paraibuna (n ⫽ 23), Piracaia (n ⫽ 13), Pirassununga (n ⫽ 21), Presidente Prudente (n ⫽ 15)

0013-8746/00/1085Ð1094$02.00/0 䉷 2000 Entomological Society of America

1086

ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA

Vol. 93, no. 5

based on Nei and Miller (1990). Mitochondrial DNA differentiation between populations was inferred by the Arlequin program (Schneider et al. 1997), using the AMOVA method (analysis of molecular variance; ExcofÞer et al. 1992). F-statistic analogues, which reßect the correlation of haplotypic diversity within and among populations, were estimated. Evolutionary relationships among haplotypes were inferred from restriction site differences. A matrix of genetic distances between pairs of haplotypes was phenetically clustered by the unweighted pair-group method with arithmetic means (UGPMA) (Sneath and Sokal 1973) using the numerical taxonomy and multivariate analysis system program (NTSYS) (Rohlf 1988). No voucher specimens were deposited. Results

Fig. 1. Location of the D. hominis populations used in the restriction fragment analysis. Al, Alfenas; Ar, Aragoiaˆnia; Br, Braso´ polis; Cp, Campinas; GO, Goia´s; Gr, Garç a; Id, Indaiatuba; MG, Minas Gerais; Pb, Paraibuna; Pr, Piracaia; PR, Parana´; Pj, Piracanjuba; Ps, Pirassununga; Pg, Ponta Grossa; Pp, Presidente Prudente; Sc, Saõ Carlos; SP, Saõ Paulo; Ss, Saõ Sebastiaõ.

Saõ Carlos (n ⫽ 13), and Saõ Sebastiaõ (n ⫽ 7) in the State of Saõ Paulo. The larvae were maintained in sawdust at ⬇37⬚C. After pupation, all specimens were stored at ⫺70⬚C. DNA extraction was done as described by Infante and Azeredo-Espin (1995). Total DNA was digested with 13 restriction endonucleases (BamHI, ClaI, EcoRI, EcoRV, HaeIII, HindIII, KpnI, MspI, PstI, PvuII, SstI, XbaI, and XhoI) according to the suppliersÕ recommendations. The resulting fragments were separated by molecular weight on 0.8 Ð1.2% agarose gels. DNA/HindIII from ␭ phage and DNA/HaeIII from ␾X174 were used as size markers. Gels containing the digested total DNA were transferred to nylon membranes (Hybond-N, Amersham) and hybridization was carried out under standard conditions (Sambrook et al. 1989), using a probe of Cochliomyia hominivorax mtDNA labeled with ␣32P. Digestion proÞles were visualized by autoradiography and fragment sizes were determined by comparison with comigrating molecular size standards. The results were transformed into a restriction site matrix using the restriction enzyme analysis package (McElroy et al. 1992). Sequence divergence (d) between haplotypes and haplotypic diversity, h (the probability that randomly drawn pairs of individuals differ in mtDNA haplotype), were calculated using restriction enzyme analysis package following the procedures of Nei and Miller (1990) and Nei (1987), respectively. Nucleotide diversity (mean estimated sequence divergence between individuals) and divergence within and between populations were calculated by the restriction enzyme analysis package,

A total of 51 patterns with 39 polymorphic restriction sites was obtained using the eight informative endonucleases (Table 1). Five endonucleases (KpnI, PstI, PvuII, XbaI, and XhoI) did not cut D. hominis mtDNA and were not included in further analyses. Representative gel proÞles are shown in Fig. 2. Each fragment pattern for a speciÞc endonuclease was assigned a letter in the order they were discovered. Transformation series reßecting probable steps for interconversion of the restriction patterns are shown in Figs. 3 and 4. For most of the endonucleases, differences between the mtDNA digestion proÞles were attributable to single site gains or losses. There were three exceptions, one of them being EcoRI, for which it was necessary to posit three site changes between patterns to account for proÞle interconversions. The second exception was pattern F of MspI, which represented the linearized mtDNA, and could have been derived from any other pattern that had two restriction sites (B, C, D, or E), and the third exception was pattern L of HaeIII, that could have been derived from patterns A, B or I. Three individuals showed patterns F of HaeIII and H of MspI, which differed only by 0.4 kb from pattern A of the respective endonucleases. Partial digestion of pattern A showed that this variation resulted from a difference in molecule length. There was not enough sample from individuals with pattern H of HaeIII for their total haplotype to be analyzed. Forty-eight mtDNA haplotypes were observed among the 227 individuals assayed (Table 2). A transformation series that relates the haplotypes using parsimony criteria is shown in Fig. 5. This cladogram is unrooted and was constructed manually, Þrst clustering those haplotypes that differed by only a single restriction site, and then adding the more peripheral ones. Possible origins of haplotypes 4 and 22, that have pattern F of MspI, are indicated. As shown, there are three groups of haplotypes separated from each other by Þve or eight mutational steps. In some cases, the same site change appeared in haplotypes of different groups, reßecting possible homoplasies. A simpliÞed character matrix containing only the phylogenetically informative restriction sites was used to generate an

September 2000 Table 1.

GEURGAS ET AL.: MTDNA VARIABILITY IN D. hominis

1087

Mitochondrial DNA restriction fragment patterns for Brazilian populations of D. hominis

Enzyme

Restriction fragments, kb

BamHI

A B

10.0, 5.7 16.4

ClaI

A B C D

5.3, 5.3, 4.7, 1.6 5.3, 4.7, 3.5, 1.9, 1.6 5.3, 4.0, 2.6, 2.2, 1.6, 1.4 5.3; 4.7; 4.0; 1.6, 1.4

EcoRI

A B C D

6.7, 3.5, 3.2, 1.9, 1.2, 0.9 6.7, 6.3, 2.0, 1.9 6.3, 4.7, 2.0, 1.6, 1.2, 1.1 8.2, 3.5, 3.2, 1.9, 0.9

EcoRV

A B C D E F

4.6, 4.1, 3.5, 2.6, 2.1 6.6, 4.1, 3.5, 2.6 7.5, 6.6, 2.6 9.2, 4.1, 3.5 8.1, 4.1, 2.6, 2.1 4.6, 3.7, 3.5, 2.6, 2.1, 0.4

HindIII

A B C

5.5, 4.2, 3.4, 3.1 7.3, 5.5, 3.4 8.5, 4.2, 3.4

SstI

A B C D E F

6.6, 4.1, 2.9, 2.4, 0.9, 0.1 11.9, 2.9, 2.4, 0.9, 0.1 6.6, 5.1, 4.1, 0.9, 0.1 6.6, 2.9, 2.4, 2.2, 2.0, 0.9, 0.1 11.9, 5.1, 0.9, 0.1 6.6, 4.1, 2.9, 2.4, 1.0

unweighted pair-group method with arithmetic average phenogram (Fig. 6). Phylogenetic analysis conÞrmed that the haplotypes clustered in three distinct groups, refereed to here as clades I, II, and III. Clade I is consisted of 30 haplotypes (189 individuals) that

Restriction fragments, kb

Enzyme HaeIII

A B C D E F G H I J L M N

8.7, 4.5, 3.8 8.7, 5.3, 3.2 8.7, 3.8, 3.2, 1.7 8.3, 7.9, 0.6 8.2, 4.5, 3.8 8.7, 4.5, 3.4 14.5; 4.5 6.8, 4.5, 3.8, 1.8 8.3, 7.9, 0.9 11.2, 3.8 8.7, 8.3 9.1, 7.9 7.9, 6.5, 1.7; 0.9

MspI

A B C D E F G H I J L M N

6.8, 5.8, 4.1 15.3, 4.1 9.4, 6.8 8.6, 8.1 12.4, 5.8 20.0 7.3, 5.4, 4.1 6.8, 5.8, 3.7 8.6, 5.8, 2.6 15.3, 2.6, 1.7 5.8, 4.1, 4.1, 3.0 8.6, 6.8, 1.3 15.3, 3.6, 0.5

showed pattern A for EcoRI. One haplotype (2) displayed the closely related pattern D. In addition, these haplotypes had pattern A for most of the other endonucleases. Clade II contained 12 haplotypes (19 individuals) characterized by pattern C of EcoRI, in as-

Fig. 2. Representative EcoRI digests of mtDNA from D. hominis. In each gel, the Þrst lane from the left represents the molecular weight standards, some of which are identiÞed (in kbp) in the margins. Indicated patterns are described in Table 1.

1088


Vol. 93, no. 5

Fig. 3. Restriction patterns of D. hominis mtDNA. The most common pattern observed for each endonuclease is shown with its restriction sites. The solid arrows indicate loss of site, relative to the most common pattern, and the dotted arrows indicate gain of site. Arrows do not show the direction of evolution.

sociation with pattern B of BamHI and ClaI. This group showed exclusive patterns for HaeIII. Clade III was represented by six haplotypes (19 individuals), all of them showing pattern B for EcoRI.

When the samples were considered together, the haplotypic diversity (h) was 0.76 and the nucleotide diversity (␲) was 2.75%. For clade I, h ⫽ 0.81 and ␲ ⫽ 0.85%, for clade II, h ⫽ 0.92 and ␲ ⫽ 1.41%, and for

September 2000


1089

Fig. 4. Restriction patterns of D. hominis mtDNA. The most common pattern observed in samples for each endonuclease is shown with its restriction sites. The solid arrows indicate loss of site, relative to the most common pattern, and the dotted arrows indicate gain of site. Arrows do not show the direction of evolution.

clade III, h ⫽ 0.72 and ␲ ⫽ 0.54%. The sequence divergence were 1.37% between clades I and II, 0.94% between clades I and III, and 2.0% between clades II and III, after correction for mean within-clade diversity (Nei 1987).

The distribution of haplotypes among the populations is given in Table 3. Of the 48 mitochondrial haplotypes, 15 were shared by most populations. Five (1, 3, 4, 13, and 18) occurred at a high frequency and 10 were considered rare (⬍5% of the sample). The

1090


Table 2. Descriptions and absolute number (n) of mtDNA haplotypes observed in D. hominis from Brazil Haplotypes Description Clade n Haplotypes Description Clade n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

AAAAAAAA AADAAAAA AAAAAABA AAAAAAFA AAAAAAGA AAAALAEA AABABADB AABALADB BBCBDADA BBCBIAEA AAAAFAHA BBCBDAAA AAAAAABB AAAAEAAA AAAAAADA BBCDMACA AAAAAAAC AAAAAAEA AABALADE AAABAAAA AABALAMB BBCCNAFA AAAEAABA AAAAAACA

I I I I I I III III II II I II I I I II I I III I III II I I

63 6 37 14 1 4 9 5 4 1 3 1 15 1 1 3 1 14 1 2 2 1 2 1

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

AAAAABBA AAAAACAA AAAAGABA BACBDADA BBCAMAEA BBCBIAIA BAAAAAAD AAAFAAAA AAAEAAAA AAAFAABA AABABBDB BBCBIAGA ACAAAAAA AAAACALA ADAAAAAA AAAAAAJA AAAAJAAA BBCBIADA AABALBDB BBCDIAGA BBCBMACA AAAAAABC AAAAAANA AAAAAAEF

I I I II II II I I I I III II I I I I I II III II II I I I

1 2 1 1 1 1 3 1 3 1 1 3 2 1 1 2 1 1 1 1 1 2 2 1

Letters (from left to right) designate multifragment gel proÞles produced by digestions with BamHI, ClaI, EcoRI, EcoRV, HaeIII, HindIII, MspI and SstI.

other 33 haplotypes were restricted to single localities, with 25 of them appearing only once in the sample. Haplotypic diversities within localities ranged from 0 to 0.91, with a mean of 0.81 (Table 4). Nucleotide diversity ranged from 0 to 2.85%, with a mean of 1.67%. The Piracanjuba site was not included in estimates of the mean diversities because the lack of diversity at this location was considered a sampling artifact. The Þve individuals analyzed showed the most common haplotype of the study collected at all locations. In the AMOVA analyses the high level of mtDNA variation in D. hominis was attributed to differences within populations (95.4%); only 4.6% was attributed to differences among populations. The Fst was estimated as 0.05. Discussion The length of D. hominis mtDNA, estimated as 17.0 kb, is close to that of other myiasis ßies (Roehrdanz and Johnson 1988, Goldenthal et al. 1991, Infante and Azeredo-Espin 1995, Azeredo-Espin and Madeira 1996, Valle 1997). The signiÞcant Þnding of this study was the presence of three distinct mtDNA clades in this species in Brazil. The presence of mitochondrial clades was previously described in Cochliomyia hominivorax Coquerel (Taylor et al. 1996). One of the clades was represented in North and Central America, the second one was represented in South America and the third in Jamaica, with a mean divergence within the clades of 0.3%. The clades of D. hominis showed a higher intraclade genetic polymorphism, which increased the variability of the whole species. The nucleotide diversity of 2.75% was greater than in other

Vol. 93, no. 5

myiasis ßies. Indeed, the nucleotide diversities of clades I and II were comparable to those for C. hominivorax in North America (␲ ⫽ 1.1%, Roehrdanz and Johnson 1988) and for Brazilian populations from the State of Saõ Paulo (␲ ⫽ 0.92%, Infante and AzeredoEspin 1995). The nucleotide diversity of clade III was comparable with that for C. macellaria F. in Brazil (␲ ⫽ 0.5%, Valle 1997), and for C. hominivorax in North, Central and South America (␲ ⫽ 0.64%, Taylor et al. 1996). Although most of the restriction patterns were uninformative with respect to clade deÞnition within D. hominis, the patterns of EcoRI shared by subgroups of haplotypes supplied unique genetic markers for the identiÞcation of mitochondrial clades. These patterns differed from each other by at least three mutational steps, which decreases their chances of being homoplasious between individuals within clades. Homoplasies are deÞned as states of resembling characters that originated by independent evolutionary changes, and can introduce ambiguities in tree placements of mtDNA haplotypes. The high mutation rate of mtDNA associated with the independence of mutational events increases the occurrence of homoplasies in distinct mitochondrial lineages proportionally to the time of separation between them (Avise et al. 1987). In this study, there were three patterns that could represent homoplasies between clades. Pattern B of BamHI representing the linearized molecule of mtDNA could have originated from pattern A either by the loss of restriction site B1 or B2. Considering the interconversion of Fig. 5, it is reasonable to suppose that pattern B shown by haplotype 31 of clade I originated from a mutational event different from that of pattern B exhibited by haplotypes of clade II. For the same reason, patterns L of HaeIII and F of MspI were considered homoplasious. To avoid or minimize ambiguities, the patterns considered homoplasious were used to relate haplotypes within clades, but not as characters to establish a relationship between clades in the interconversion of Fig. 5 or to generate the phenogram in Fig. 6. The estimates of divergence between mitochondrial clades of D. hominis, after correction for within clade variability, ranged from 0.94% between clades I and III, to 2.0% between clades II and III. For many animal species, mtDNA clades distinguished by more than a 1% sequence divergence are geographically isolated (phylogeographic category I of Avise et al. 1987). The most likely explanation of their origins involves long-term, extrinsic barriers to gene ßow, or the extinction of intermediate haplotypes in widely distributed species with limited dispersal and gene ßow capacities (Avise et al. 1987, Avise 1994). Although the clades of C. hominivorax did not show complete geographic Þdelity, the overall pattern of distribution Þtted the species in this category (Taylor et al. 1996). In a few species of vertebrates (Avise et al. 1984, 1990, 1992; Wayne et al. 1990), the presence of deep phylogenetic discontinuities in the mtDNA gene tree in the absence of geographic localization of the mtDNA clades (phylogeographic category II) has

September 2000


1091

Fig. 5. Parsimony network describing the interrelationships among the 48 mtDNA haplotypes of D. hominis (speciÞed in Table 2). The tick marks on the connecting lines represent the difference in the number of restriction sites between corresponding haplotypes. Possible origins of haplotypes four and 22 are shown separately.

been observed. Avise et al. (1987) suggested that this situation most likely results from secondary admixture between allopatrically evolved populations from which the deep splits in the gene tree

were derived. Although the mitochondrial clades of D. hominis were sympatric, their frequencies varied among populations, unlike those species belonging to phylogeographic category II. As the boundaries

1092


Vol. 93, no. 5

Fig. 6. Estimated phylogeny for the haplotypes observed in D. hominis, presented as a UPGMA phenogram. Although alternative networks were found during the analyses, all of them involved only minor rearrangements within mitochondrial clades. The consistency index in the associated parsimony network was 0.95.

Table 3.

Distribution of mitochondrial haplotypes among Brazilian populations of D. hominis

Location

Clade I

Alfenas Aragoiaˆnia Braso´ polis Campinas Garç a Indaiatuba Paraibuna Piracaia Piracanjuba Pirassununga Ponta Grossa P. Prudente S. Carlos S. Sebastiaõ

1(11) 2(4), 3(6), 4(1), 5(1), 6(1) 18(1) 20(1) 38(1) 1(4), 3(1), 4(1), 13(1), 18(1), 26(1), 33(1), 41(1), 48(1) 1(8), 3(2), 17(1), 18(4), 27(1) 1(1), 4(1), 13(4), 20(1), 31(1) 1(9), 3(3), 4(1), 6(2), 40(2) 1(2), 2(1), 3(3), 11(3), 13(1) 1(5), 4(1), 6(1), 13(5), 14(1), 18(3), 24(1) 1(6), 3(2), 15(1) 1(5) 1(7), 2(1), 3(7), 25(1), 26(1), 31(2), 39(1) 1(2), 3(5), 4(4), 18(1) 1(1), 3(6), 4(5) 1(1), 3(3), 13(1), 18(4), 46(2), 47(2) 1(1), 23(2), 32(1), 33(2), 34(1)

Clade II

Clade III

9(1), 22(1) 42(1) 30(1), 45(1)

7(3), 8(1), 19(1), 21(2), 35(1) 8(1)

10(1) 9(3), 12(1), 16(1) 16(2), 29(1) 28(1) 36(3), 37(2), 44(1)

8(3) 7(1) 7(1) 7(1), 43(1) 7(3)

Numbers in parentheses refer to the absolute number of that haplotype in the population.

Table 4.

Nucleotide variability (␲) and haplotypic diversity (h) in Brazilian populations of D. hominis

Indices

Al

Ar

Br

Cp

Gr

Id

Pb

Pr

Ps

Pg

Pp

Sc

Ss

Avg

h ␲ (%)

0.86 1.44

0.91 1.00

0.83 1.24

0.71 0.77

0.72 0.94

0.86 1.16

0.89 2.38

0.78 2.65

0.79 1.13

0.87 2.85

0.60 1.75

0.86 1.55

0.90 2.80

0.81 1.67

Abbreviations of locations are the same as in Fig. 1.

September 2000


between phylogeographic categories are not clearly delimited, because of the degree of divergence of mitochondrial clades and/or their geographic distribution, it seems that D. hominis is in a transition zone among phylogeographic categories I and II. If the mitochondrial clades of D. hominis had allopatric origins, as suggested for species of phylogeographic categories I and II, this would imply the existence of at least three historically isolated populations. Assuming a constant evolutionary rate for mtDNA of 2% divergence per million years (Powell 1997), the 1.37% sequence divergence between clades I and II dates the split of these lineages to ⬇685,000 yr ago. The 0.94% sequence divergence between clades I and III suggests that isolation occurred 470,000 yr ago. Finally, the 2.0% sequence divergence among clades II and III equates to their isolation in the last 1 million years. Given the usual caution about large stochastic errors associated with these estimates, it is nevertheless apparent that the genetic subdivision of D. hominis occurred recently. Although there was no indication of the centers of origin for the mitochondrial clades found in D. hominis, the estimates of divergence coincide with those for divergence between North/Central American and South American clades of C. hominivorax, which correlated with the exchange of land fauna between the continents (Taylor et al. 1996). Whatever the origin of the three mtDNA clades, the shared presence of rare haplotypes strongly suggests that there has been recent gene ßow involving females, probably reinforced by the movement of infected cattle in the region. Not only were mitochondrial clades shared among locations, but rare mtDNA haplotypes also commonly appeared in multiple locations. For example, among four individuals exhibiting the rare mtDNA haplotype 6, one derived from Al, one from Pb, and two from Gr (Table 3). Only eight haplotypes were observed in more than one individual conÞned to a single population, but were insufÞcient for structuring the D. hominis population in the region (Fst ⫽ 0.05). The lack of genetic structure associated with the extensive genetic diversity of the populations reinforces the idea of allopatric evolution of these mitochondrial clades. Normally, species that do not exhibit a genetic structure in their mtDNA show little genetic diversity, with a sequence divergence of ⬍1% between haplotypes. This pattern very likely reßects an intense historical gene ßow or rapid expansion of the population (Avise et al. 1987, Rogers and Harpending 1992). Dermatobia hominis proved to be the most polymorphic species of myiasis ßies analyzed thus far. More studies involving samples from Mexico, Central America, and other South American countries are necessary to determine the origin and the distribution pattern of the mitochondrial clades of D. hominis throughout the speciesÕ range. Nevertheless, the lack of population structure and the shared presence of rare haplotypes suggest a recent contact among locations, probably facilitated via livestock movement,

1093

particularly because there was no relation between geographic distance and the similarity among populations. The results of the current study indicate that, in terms of parasite control, populations should not be considered independent units, because immigrations will restore local extinctions. Acknowledgments The authors thank R. A. Rodrigues, M.S.C. Campos, and M. Constantino F⬚ for technical assistance. This work was supported by Fundaç aõ de Amparo a` Pesquisa do Estado de Saõ Paulo (FAPESP, grant 1995/8856 Ð1) and Programa de Apoio ao Desenvolvimento Cientõ´Þco e Tecnolo´ gico-(PADCT) Subprograma de Biotecnologia-SBIO/CNPq (grant 62.0097/ 95-7) to A.M.L.E.; S.R.G. and M.E.I.V.M. were supported by fellowships from Conselho Nacional de Desenvolvimento Cientõ´Þco e Tecnolo´ gico,(CNPq), Brazil.

References Cited Avise, J. C. 1994. Molecular markers, natural history, and evolution. Chapman & Hall, New York. Avise, J. C., E. Bermingham, L. G. Kesler, and N. C. Saunders. 1984. Characterization of mitochondrial DNA variability in a hybrid swarm between subspecies of bluegill sunÞsh (Lepomis macrochirus). Evolution 38: 931Ð941. Avise, J. C., J. Arnold, R. M. Ball, E. Bermingham, T. Lamb, J. E. Neigel, C. A. Reeb, and N. C. Saunders. 1987. IntraspeciÞc phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annu. Rev. Ecol. Syst. 18: 489 Ð522. Avise, J. C., C. D. Ankey, and W. S. Nelson. 1990. Mitochondrial gene trees and the evolutionary relationship of mallard and black ducks. Evolution 44: 1109 Ð1119. Avise, J. C., R. T. Alisauskas, W. S. Nelson, and C. D. Ankney. 1992. Matriarchal population genetic structure in an avian species with female natal philopatry. Evolution 46: 1084 Ð1096. Azeredo-Espin, A.M.L., and N. G. Madeira. 1996. Primary myiasis in dog caused by Phaenicia eximia (Diptera: Calliphoridae) and preliminary mitochondrial analysis of the species in Brazil. J. Med. Entomol. 33: 839 Ð 843. Excoffier, L., P. E. Smouse, and J. M. Quattro. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479 Ð 491. Goldenthal, M. J., K. A. McKenna, and D. J. Joslyn. 1991. Mitochondrial DNA of the blowßy Phormia regina: restriction analysis and gene location. Biochem. Genet. 29: 1Ð11. Guimara˜ es, J. H., N. Papavero, and A. P. Prado. 1983. As miõáses na Regiaõ Neotropical (identiÞcaç aõ, biologia, bibliograÞa). Rev. Bras. Zool. 1: 239 Ð 416. Hall, M., and R. Wall. 1995. Myiasis of human and domestic animals. Adv. Parasitol. 35: 257Ð334. Infante, M. E., and A.M.L. Azeredo-Espin. 1995. Genetic Variability in Mitochondrial DNA of the Screwworm, Cochliomyia hominivorax (Diptera: Calliphoridae), from Brazil. Biochem. Genet. 33: 237Ð255. McElroy, D., P. Moran, E. Bermingham, and I. Kornfield. 1992. REAP: an integrated environment for the manipulation and phylogenetic analysis of restriction data. J. Hered. 83: 157Ð158. Nei, M. 1987. DNA Polymorphism within and between populations, pp. 254 Ð286. In Molecular evolutionary genetics. Columbia University Press, New York, NY.

1094


Nei, M., and J. C. Miller. 1990. A simple method for estimating average number of nucleotide substitutions within and between populations from restriction data. Genetics 125: 873Ð 879. Powell, J. R. 1997. Molecular evolution, pp. 381Ð395. In Progress and prospects in evolutionary biology: the drosophila model. Oxford University Press, Oxford, UK. Roehrdanz, R. L., and D. A. Johnson. 1988. Mitochondrial DNA variation among geographical populations of the screwworm ßy Cochliomyia hominivorax. J. Med. Entomol. 25: 136 Ð141. Rogers, A. R., and H. C. Harpending. 1992. Population growth makes waves in the distribution of pairwise genetic differences. Mol. Biol. Evol. 9: 552Ð569. Rohlf, F. J. 1988. NTSYS: numerical taxonomy and multivariate analysis system, version 1.4. Exeter Software, Setauket, NY. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Sancho, E. 1988. Dermatobia, the neotropical warble ßy. Parasitol. Today 4: 242Ð246. Schneider, S., J. M. Kueffer, D. Roessli, and L. Excoffier. 1997. ARLEQUIN version 1.0. A software for population genetic analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland.

Vol. 93, no. 5

Sneath, P.H.A., and R. R. Sokal. 1973. Numerical taxonomy. Freeman, San Francisco, CA. Stevens, J., and R. Wall. 1996. Species, subspecies and hybrid populations of the blowßies Lucilia cuprina and Lucilia sericata (Diptera: Calliphoridae). Proc. R. Soc. Lond. B 263: 1335Ð1341. Taylor, D. B., A. L. Szalanski, and R. D. Peterson. 1996. Mitochondrial DNA variation in screwworm. Med. Vet. Entomol. 10: 161Ð169. Thomas, D. B. 1988. The pattern of Dermatobia (Diptera: Cuterebridae) myiasis in cattle in tropical Mexico. J. Med. Entomol. 25: 131Ð135. Valle, J. S. 1997. Ana´lise da variabilidade gene´ tica de Cochliomyia macellaria (Diptera: Calliphoridae); Avaliaç aõ atrave´ s do DNA mitocondrial e ana´lise cariotõ´pica. M.S. thesis, Universidade Estadual de Campinas, S. P. Wayne, R. K., A. Meyer, N. Lehman, B. Van Valkenburgh, P. W. Kat, T. K. Fuller, D. Girman, and S. J. O’Brien. 1990. Large sequence divergence among mitochondrial genotypes within populations of eastern African blackbacked jackals. Proc. Natl. Acad. Sci. 87: 1772Ð1776. Received for publication 24 September 1999; accepted 1 March 2000.