Genetic profile and molecular resistance of Aedes ( Stegomyia ) aegypti ( Diptera : Culicidae ) in Foz do Iguaçu ( Brazil ) , at the border with Argentina and Paraguay

The genetic variability of populations of Aedes (Stegomyia) aegypti (Linnaeus, 1762) is often low due to the intense selection pressure caused by chemical control measures. In this study, we evaluated the susceptibility of larvae and adults of this mosquito to chemical insecticides, the frequency of the Val1016IIe mutation, and the genetic variability of the mitochondrial ND4 gene fragment in the urban area of Foz do Iguaçu, Paraná. The populations of A. aegypti in the southern and central regions of the city were resistant to the diagnostic dose of temephos 0.0162 ppm. Additionally, we detected difference in susceptibility in the northern population. The resistance ratios (RR95) were 3.8, 4.1, and 4.6 for the populations in the north, central, and south of the city, respectively. The entire population of A. aegypti in Foz do Iguaçu is resistant to pyrethroids. The mortality rates are as follows: 72.2% for cypermethrin (CD: 146 mg i.a./m2) and 57% for deltamethrin (CD: 18 mg i.a./m2). These results corroborate the hypothesis that the mutated 1016Ile allele is present in all analyzed strata. Out of the 234 samples genotyped for the Val1016Ile mutation, 15% were homozygous dominant for the wild allele (Val/Val), 62% were heterozygous (Va/Ile) and 23% were homozygous for the recessive mutation (Ile/Ile). With respect to the genetic variability of the mitochondrial ND4 gene fragment, 93% of the sequences analyzed belonged to haplotype 1, and 7% belonged to haplotype 2. The genetic diversity was low, the fixation index was not significant, and gene flow was high. The control of A. aegypti in Foz do Iguaçu using temephos and pyrethroids may be compromised because of the reduced vector susceptibility. Populations of A. aegypti, that undergoes a sudden reduction in effective population size and become resistant to pyrethroids may differ from the original population in vector capacity.

Mosquito control is still the only effective means to protect human populations from the Dengue virus, despite recent efforts to find a vaccine against the disease (TAUIL 2001).Attempts to control the vector Aedes (Stegomyia) aegypti (Linnaeus, 1762), usually involve a combination of chemical insecticides and environmental sanitation.Unfortunately, however, few insecticides are available for this purpose and because mosquito populations tend to become resistant to them, it is necessary to assess and continually monitor these populations to ensure the success of vector control programs.
In Brazil, two classes of insecticides -organophosphates (since 1967) and pyrethroids (since 1999) -have been used (BRAGA et al. 2004) in the control of A. aegypti.The first cases of resistance to temephos, an organophosphate, were recorded by MACORIS et al. (1995) in populations of A. aegypti, in Goiânia and Marilia.After that, a few other studies documented larval resistance to the same compound in several Brazilian populations (LIMA et al. 2003, 2011, MACORIS et al. 2003).In countries that share borders with southern states of Brazil, resistant populations have been documented only in Argentina (SECCACINI et al. 2008, LLINÁS et al. 2010).In Paraguay, no research to date was conducted on mosquito resistance to temethos, and mosquito control is limited to areas where A. aegypti is endemic (GÜRTLER et al. 2009).
Cases of resistance to pyrethroids were first detected in 2001 in some populations of A. aegypti in the states of Alagoas, Sergipe, and Rio de Janeiro (PEREIRA- DA-CUNHA et al. 2005).The assessment of vector susceptibility to this class of insecticides involves an evaluation of the biological response of adults using biological assays.Molecular assays are also used for that purpose, as mutations related to pyrethroid resistance occur at the voltage-gated sodium channel, the target site of the insecticide, which keeps the sodium channels of the axon membrane open, stimulating the nerve cells to produce repetitive signals, and resulting in the death of the insect (BRAGA & VALLE 2007a).A single amino acid substitution in this channel causing a reduction in the binding of the pesticide to the voltage-gated sodium channel may be associated with a resistance known as the knockdown resistance (kdr) (KAWADA et al. 2009, GARCÍA et al. 2009).
Mutations in the voltage-gated sodium channel have been detected in several insects, including A. aegypti, that might be associated with resistance (SODERLUND & KNIPPLE 2003).A substitution of adenine for guanine in the codon Val1016 substitutes the amino acid Valine with Isoleucine (Val1016Ile).This substitution is associated with a change in resistance to pyrethroids (SAAVEDRA-RODRIGUEZ et al. 2007, MARTINS et al. 2009).The allele associated with knockdown resistance is recessive and therefore is only expressed in homozygous individuals (DAVIES et al. 2008).This means that, under intense selection pressure, resistant alleles are selected quickly, changing the genetic structure of the population.
Studying the genetic structure of populations is crucial to understanding the population dynamics of A. aegypti (HIRAGI et al. 2009).The genetic variability in natural populations of this mosquito has been investigated using molecular markers such as RAPD (Random Amplified Polymorphic DNA) (JULIO et al. 2009), microsatellites (PAUPY et al. 2010), SNPs (single nucleotide polymorphism) (HEMME et al. 2010), and mitochondrial DNA (GORROCHOTEGUI-ESCALANTE et al. 2002).
Combining studies on the genetic variability of populations under selection pressure caused by insecticides is important, especially in areas prone to the input of new strains of A. aegypti.One such example is the city of Foz do Iguaçu, which shares borders with two Latin American countries and where there is intense trade and movement of people and products.Consequently, the present study evaluated the susceptibility of larvae and adults to chemical insecticides, the frequency of the mutation Val1016IIe associated with resistance to pyrethroids, and the genetic variability of the mitochondrial ND4 gene fragment in populations of A. aegypti in the urban area Foz do Iguaçu, Paraná.

Capture of A. aegypti for the temephos assays, Val1016Ile mutation and genetic diversity
Specimens were collected in the city of Foz do Iguaçu (25°32'52"S, 54°35'17"W) state of Paraná, Brazil.Foz do Iguaçu is located in the westernmost portion of the state and borders Puerto Iguazú, in Argentina, and Ciudad del Este, in Paraguay (IBGE 2011), with an estimated population of 325,137 inhabitants and an area of 617.70 km 2 .Foz do Iguaçu is a large city, with over 8,100 buildings.According to the LIRAa (Rapid Survey Index infestation by A. aegypti), it is necessary to divide the city area into strata, which are geographical regions with an equivalent number of buildings: each stratum having a minimum of 8,100 and a maximum of 12,000 buildings (MINISTÉRIO DA SAÚDE 2005) (Fig. 1).
The samples were collected by the staff members of the Center for Zoonosis Control of Foz do Iguaçu (CCZ) using ovitraps (FAY & ELIASON 1966) in December 2008, February 2009, May 2009and October 2009.According to LIRAa, in large municipalities it becomes necessary to divide the sampling into two stages: blocks and buildings.Blocks (primary sampling units) and the buildings (secondary sampling units) in each block are randomly selected (secondary sampling units) (MINISTÉRIO DA SAÚDE 2005).For logistical reasons, we chose to use the same number of ovitraps in each stratum (n = 15), totaling 165 traps per period.Ovitraps remained in peridomicile for five days, after which they were removed and taken to the laboratory.There, the blades containing eggs of A. aegypti were kept in an insectary under controlled temperature (25 ± 1°C), humidity (80 ± 10%) and photoperiod (12:12) until the adults emerged.
We used individuals of the F1 generation from the December 2008 sampling for the temephos bioassays.Strata one, two and four, in the northern region of the city, were combined; in the central region, strata five, six and seven; and in the southern region, strata nine, ten, and eleven (Fig. 1).Strata three and eight were excluded from the analysis on temephos susceptibility due to the absence of eggs for bioassays.
In analyzes of the Val1016Il mutation and the fragment of the mitochondrial gene ND4, newly emerged A. aegypti individuals were sampled in each of the eleven strata and the four collection periods.The mosquitoes were placed individually in absolute ethanol and stored at -20°C.

Biological assays
We performed tests using the insecticide Temephos pestanal 250 mg 97.5% (Sigma-Aldrich).The calibration and control of the bioassays were performed on the susceptible strain of A. aegypti Rockefeller strain (Centers of Diseasse Control, CDC).Bioassays were performed with a diagnostic concentration of 0.0162 ppm of temephos, which corresponded to twice the lethal concentration, 99% (LC99), for this susceptible strain.The tests were read after 24 hours, and an estimated larval mortality rate was calculated.Populations with mortality rates above 98% were considered susceptible, between 80% and 98%, with alterations in susceptibility, and below 80%, resistant (WHO 1998).Bioassays with multiple concentrations were performed according to the World Health Organization to detect and quantify resistance (WHO 1981a).
In each experiment, nine concentrations were tested, in four replicates.Each concentration had 20 larvae, late third instar, and fourth instar of A. aegypti, totaling 720 larvae per experiment.We added 80 ml of purified water and 20 ml of water containing the larvae in all glasses.As population controls, four replicates using ethanol as a solvent were tested.
According MAZZARI & GEORGHIOU (1995), the levels of resistance are defined as: resistance ratio (RR) <5, low; between 5 and 10, medium; above 10, high.The lethal concentrations were calculated in the basic program Probit (FINNEY 1971).The resistance ratios 50 and 95 were calculated by dividing the lethal concentration of the population examined by the lethal concentration of the Rockefeller population.
Bioassays with pyrethroids were performed at the Center for Drug Evaluation and Research Office of the Regional Superintendent of Marilia Endemic Disease Control (SUCEN, Marilia, São Paulo).Qualitative essays, using the diagnostic concentration (DC) on each trial, and paper impregnated with cypermethrin (DC: 146 mg ia/m 2 ) and deltamethrin (DC: 18 mg ia/m 2 ), followed the standard methodology of the World Health Organization (WHO 1981b(WHO , c, 1992(WHO , 1998)).The susceptible Rockefeller strain was used for comparison.The diagnostic concentration was measured after 24 hours, and the rate of adult mortality was calculated.Populations with a mortality rate above 98% were considered susceptible; between 80% and 98%, with changes in susceptibility, below 80%, resistant (WHO 1998).

DNA extraction
Each specimen was macerated in a microtube using a plastic pestle, followed by the addition of 160 µL of lysis buffer (TrisHCl 0.5 M, 4 M NaCl, 0.5 M EDTA) and 20 µL sodium dodecyl sulfate 10%.The tubes were incubated in serum bath for 30 minutes at 60°C.For deproteinization, 50 µL of chloroform: isoamyl alcohol (24:1) were added to each tube.The mixture was centrifuged for 15 minutes at 13,000 rpm in a refrigerated centrifuge at 15°C.The supernatant was transferred to a fresh 1.5 ml tube, to which 80 µL of ammonium acetate (7.5 M) and 300 µl 96% ethanol were added.The samples were homogenized by inversion and taken to a -20°C freezer for 30 minutes to precipitate the DNA.Samples were then centrifuged at 13,000 rpm for 15 minutes at 15°C.After centrifugation, the supernatant was discarded and 300 µL of 70% ethanol were added to the mixture to wash the DNA.The samples were centrifuged again at 13,000 rpm for five minutes at 15°C.The supernatant was discarded again.The tubes were then placed in an oven at 37°C to dry the samples.The DNA was re-suspended in 30 µL of TE (0.5 M TrisHCl, 0.05 M EDTA and MilliQ water) and stored at -20°C.DNA quantification was performed with a NanoDrop™ spectrophotometer

Voltage-gated sodium channel gene fragment -Genotyping 1016 site
In order to evaluate the presence of the mutation for pyrethroid resistance, three primers, one specific for the wild allele (5'-GCG GGC AGG GCG GCG GGG GCG GGG CCA CAA ATT GTT TCC CAC CCG CAC CGG-3'), one specific for the mutant allele (5'-GCG GGC ACA AAT TGT TTC CCA CCC GCA CTG A-3') and a third initiator, common to both (5'-GGA TGA ACC GAA ATT GGA CAA AAG C-3') (SAAVEDRA-RODRIGUEZ et al. 2007, MARTINS et al. 2009) were used.
The statistical analysis was performed using the program Bioestat 5.0 (AYRES et al. 2007), using a criterion analysis of variance and Tukey's test, with significance level of 0.05.The pairwise comparisons of the proportions of heterozygous and homozygous recessive genotypes followed ZAR (2010).Allele and genotype frequencies were calculated according to SALMAN (2007).The allele frequencies remained constant across generations, according to the Hardy-Weinberg model (HARTHL 2008).

Variability in the ND4 mitochondrial gene
The amplification of the gene segment expressing the subunit 4 of the mitochondrial enzyme NADH dehydrogenase was performed using 20 ng DNA, 12.5 pmoles of each primer (IDT, Coralville, IA, USA), dNTP 200 µM (Amresco, Solon, OH), 2.5 µl Buffer 10X (100 mM Tris-HCl pH 8.3, 500 mM KCl, 15 mM MgCl2), 1 U Taq DNA polymerase (Sigma-Aldrich) and Milli-Q water to a final volume of 25 µl.The primers used in the amplification and the amplification reactions followed COSTA DA SILVA et al. (2005).The success of the amplification was ascertained using 1% agarose gel.The reaction was performed using the Qiaquik PCR purification kit (Qiagen) according to the manufacturer's instructions.The samples were sent to Center for Human Genome Studies (University of São Paulo) for sequencing using the equipment ABI 3730 DNA Analyzer (Applied Biosystems).
Consensus sequences were obtained using the Staden Package version 1.5.Sequences were aligned using BioEdit version 7.0.0(HALL 2004) and ClustalW (THOMPSON et al. 1994).The sequences obtained were compared with sequences available on GenBank, using the tblastx tool for confirmation of the amplified fragment.Genetic diversity and neutrality tests were calculated using the program DnaSP, version 5 (LIBRADO & ROZAS 2009).Analysis of molecular variance (AMOVA) was performed using the program Arlequin version 3.5 (EXCOFFIER & LISCHER 2010).The number of migrants per generation (Nm) was obtained using the formula Fst = 1/2 +1 Nm (Fst = fixation index), considering haploid organisms.Geographical distances were obtained using Google Earth 6.

Biological assays
The populations of A. aegypti in the central (71% mortality) and southern (77% mortality) regions of Foz do Iguaçu were resistant to temephos.In addition, we detected a shift in the susceptibility levels of the northern population (85% mortality) to the organophosphate at the diagnostic concentration of 0.0162 ppm.The resistance ratios of the populations in the northern, central and southern regions were 3.79, 4.06 and 4.65 respectively (Tab.I).
In 2008, the population of A. aegypti in the urban area of Foz do Iguaçu was considered resistant to pyrethroids, with a 72.2% mortality for cypermethrin (LC: 146 mg i.a./m 2 ) and 57.0% for deltamethrin (LC: 18 mg i.a./m 2 ).

Voltage-gated sodium channel gene fragment -Genotyping 1016 site
A total of 234 specimens were surveyed for the Val1016I1 mutation.Of these, 15% (35) were homozygous dominant; 62% (146) were heterozygous and 23% (53) were homozygous recessive.The genotype frequency of the Val/Ile was higher than that of Val/Val and Ile/Ile (p = 0.0002, F = 12.7356, DF = 2).The frequencies of the genotypes bearing the recessive allele were not distributed randomly among the strata.(Val/Ile: 2 = 28.54,Ile/Ile: 2 = 28.16,critical value 18.30 for ten degrees of freedom).Nonetheless, the population is in Hardy-Weimberg equilibrium (Tab.II).The total frequencies of the Val and Ile alleles were 0.46 and 0.54, respectively.There were no significant differences in the frequencies of the 1016Val and 1016Ile alleles (p = 0.109, F = 2.757, DF = 1) in the 11 strata analyzed within the municipality.The mutant allele was more frequent in all strata, except for regions 2, 9 and 11 (Fig. 2).

Variability in the ND4 mitochondrial gene
The amplified fragment of the ND4 gene was 291 bp long.We observed 10 polymorphic sites and 281 monomorphic sites.The analysis of the amplified fragment of 157 individuals showed that there are two haplotypes in the population of Foz do Iguaçu, 93% of the H1 [GenBank: JN190355] and 7% of the H2 [GenBank: JN190356] (Tab.III).The haplotypes can be dis- The haplotype diversity in our data was 0.131 (± 0.035), nucleotide diversity was 0.00475 (± 0.00127) and the average number of nucleotide differences was 1.311.The results of the tests for neutral selectivity were not significant, according to the neutral evolution hypothesis test (KIMURA 1968) (p > 0.05;Tajima D Test: -0.62896;D Test of Fu and Li: 1.34621;F Test of Fu and Li: 0.76017;Fu Fs Test: 6470).
Heteroplasmy, the presence of more than one mitochondrial DNA in the same individual, was found in four analyzed individuals, at the polymorphic sites described in this study (sites 069, 084, 117, 162, 201).
For the analysis of molecular variance, the populations were grouped according to their strata.The results show that almost all the variation (99.78%) was found within populations, whereas 0.22% of the variation occurred among populations, with a value of Fst = 0.00217, not significant (p = 0.4164).The gene flow (Nm) for the total population was 230.In pairwise comparisons among the strata, all values of Fst were not significant.

Correlation between genetic diversity and KDR resistance
Seventy-nine individuals were analyzed concurrently for the ND4 fragment and the 1016Ile mutation, 73 individuals with haplotype H1 and six with haplotype H2.Both 1016Val and 1016Ile alleles were observed among individuals with haplotypes H1 and H2 (Figs 3-6).

DISCUSSION
Foz do Iguaçu is known for its tourist attractions and as an important commercial activities, yet the region had many cases of dengue in recent years (DUQUE et al. 2010).In this area, chemical insecticides are used to control the dengue vector, leading to the selection for insecticide resistance in these insects.Larval resistance to temephos in Foz do Iguaçu was reported by PROPHIRO et al. (2011PROPHIRO et al. ( ) in 2006PROPHIRO et al. ( /2007.Their study divided the city into a northern and a southern region.The mortality ratio (CD: 0.0060) for A. aegypti in the southern region (RR95: 3.62), was 77.5%, and 75.9% in the northern region (RR95: 3.13).According MAZZARI & GEORGHIOU (1995), resistance levels below five are considered low.However, when temephos use is interrupted in locations where the resistance ratio is high (greater than 10), resistance decreases only gradually, delaying the possibility of efficient use of temephos for several years (MONTELLA et al. 2007).
In 2006, at the technical meeting of the National Resistance Monitoring of A. aegypti (MoReNAa), it was determined that vector control programs in municipalities with resistance ratio of A. aegypti greater than three should use another insecticide with a different mechanism of action, in order to reduce the selection pressure of temephos on the exposed populations (MINISTÉRIO DA SAÚDE 2006).For instance, in 2011, the local vector control service in Foz do Iguaçu began to use the larvicide diflubenzuron, which interferes with the synthesis of chitin.In addition to the use of alternative insecticides, populations should continue to be monitored with respect to their resistance to temephos until this insecticide can be used again.Moreover, cultural changes need to take place in the community in order to effectively control A. aegypti.For instance, people should avoid disposing of containers that accumulate water and serve as oviposition sites (GUIRADO & BICUSO 2009).
Our bioassays detected resistance to pyrethroids in adult populations of A. aegypti in the urban area of Foz do Iguaçu.This was further confirmed after analyzing the frequency of the mutation in the gene fragment of the voltage-gated sodium channel, with the 1016Ile allele being found at high densities in the analyzed specimens in all strata.Heterozygotes were present in all but three strata in the urban area.Despite the selective advantage of heterozygous genotypes, mating between heterozygous individuals results in 25% of the progeny carrying the recessive mutation 1016Ile (SALMAN 2007).
Previous preliminary analysis of the population of A. aegypti in Foz do Iguaçu in hot spot areas with respect to insecticide use (MARTINS et al. 2009) had detected the allele 1016Ile at the frequency of 73.7%, which was higher than the frequency found in the present work for the same allele (54%).Our sampling, however, covers a larger area.Nevertheless, the frequency of the 1016Ile allele is high in the results of both studies, which is consistent with the high selective pressure caused by the pyrethroids used in mosquito control.
In our results, the spatial distribution of the genotype that carries the recessive allele is not randomly distributed in the urban area.The analysis of the 1016 mutation frequency should be based on a great number of individuals distributed across a wide geographical area.Consistent with the conclusion of GARCIA et al. (2009), we suggest that, in analyses of the frequency of the 1016 mutation, a large number of individuals over a large area should be used because populations in nearby locations may have very different allele frequencies.According to GARCIA et al. (2009), the frequency of the 1016Ile increased from the end of 1990 up to 2006-2009 in several Mexican states.In Nuevo Leon, the 1016Ile mutation was first reported in 1997.In 2003 and 2004, its frequency had increased by approximately 100%.Spatial heterogeneity was found between two states, and also in smaller geographic scales.Within the city of Chetumal, Quintana Roo, the frequencies of the 1016Ile mutation ranged from 38.3% to 88.3%.The frequency of the resistant genotype is likely to increase depending on the intensity and the persistence of control actions against the mosquito.
Genotyping the voltage-gated sodium channel gene fragment allows us to predict the effectiveness of pyrethroids.The detection of this mutation in field populations has serious consequences for the sustained use of pyrethroids in mosquito control (SODERLUND & KNIPPLE 2003, SAAVEDRA-RODRIGUEZ et al. 2007, DONNELLY et al. 2009).Certainly, the populations of A. aegypti in the field must be continuously monitored to maintain or to restore the susceptibility to pyrethroids, considering that there are few options for chemical control.
With regards to genetic variability, the H1 haplotype was found at a high frequency, being 100% similar to haplotype H6 [GenBank: EU650410] found by LIMA-JUNIOR & SCARPASSA (2009) in the states of Pará, Acre and the Brazilian Amazon, and haplotype H19 [GenBank: AY906853] found by PADUAN & RIBOLLA (2008) in the city of Barretos (São Paulo).The H2 haplotype was found less frequently and is 100% similar to haplotype H10 [GenBank: EU650414] of LIMA-JUNIOR & SCARPASSA (2009) in Belém (Pará) and H23 [GenBank: AF334863] of GORROCHOTEGUI-ESCALANTE et al. (2002) in Mexico.The distribution of genetic variants in natural populations is the result of several evolutionary forces and demographic factors (SLATKIN 1985).
Population genetic diversity is expected to be higher when samples collected in large scale are used in haplotype surveys.One example is the work of BRACCO et al. (2007), who found high values of genetic diversity (Hd = 0.8200 and = 0.0199) in the analysis of samples from 36 locations in the Americas, Asia and Africa.Conversely, when local surveys are conducted, genetic diversity tends to be lower.The values of genetic diversity (Hd = 0.1310 and = 0.0047) in our data were low for the ND4 analysis.However, we expected to find more haplotype diversity in Foz do Iguaçu, because of the elevated road traffic and the intense commercial trade with other countries in South America.The low values of genetic diversity in our results may suggest a reduction in the size of the population, which may have suffered a recent bottleneck.Bottlenecks lead to reduced genetic diversity with respect to the original gene pool.In the case of vector populations, bottlenecks are usually associated with control measures, which are more intense when infestation indexes are high and the local populations are being affected by the disease.Genetic polymorphisms may decline rapidly anywhere in the genome of the mosquito due to the use of insecticides (YAN et al. 1998).
We detected four individuals with more than one mitochondrial DNA genotype, a phenomenon called heteroplasmy (BUTLER & LEVIN.1998).In the same individual, up to five sites with two different nucleotides were observed.A cell can become heteroplasmic following a somatic or germ line point mutation at a specific site or an insertion/deletion mutation (CALLOWAY et al. 2000).A mechanism that explains the inherited heteroplasmy, and has been observed in Drosophila Fallén, 1823, is biparental mitochondrial inheritance.In such cases, the father contributes a small number of mtDNA molecules to the progeny (KONDO et al. 1990).Another explanation for the appearance of more than one genotype in the same individual is the sequencing of a nuclear pseudogene within the mito-ZOOLOGIA 29 (6): 540-548, December, 2012 chondrial DNA.Nuclear Mitochondrial pseudogenes, which have been reported for A. aegypti by HLAING et al. (2009) may generate mistakes in the interpretation of heteroplasmy (PARR et al. 2006).However, in the present contribution, we observed that all cases of heteroplasmy occurred in the polymorphic sites, which suggests that parental leakage might have taken place.According to PADUAN & RIBOLLA (2008), it is possible that parental leakage has been historically underestimated, because sequencing is not the best way to reveal this kind of heteroplasmy in the mitochondrial genome.Almost all of the genetic variation was found within populations (99.78%), whereas variation among populations was extremely low (0.22%).The value of the fixation index Fst = 0.00217 was extremely low and the average gene flow (Nm) was high (230) when compared with other studies using A. aegypti.AYRES et al. (2003), using RAPD, found values of Fst = 0.317 and Nm = 0.54 for the Brazilian populations.The values of molecular variance found in our study suggest that populations of A. aegypti in the urban area of Foz do Iguaçu are not genetically structured.
According YAN et al. (1998), the assumption that vector populations are in genetic equilibrium can be biased.This is because the interval between the applications of insecticides is usually three to four months, which is much less than the time required for achieving genetic equilibrium in a population.Despite the fact that we divided the urban area of Foz do Iguaçu into strata, we observed that there is a similarity in genetic populations of A. aegypti in the urban area as a whole, due to the short geographical distance among the sub-regions analyzed, and the extensive gene flow among them.The high gene flow may explain the rapid spread of the gene responsible for insecticide resistance.
The two mitochondrial strains found by us have the 1016Ile mutation allele, which is homozygous recessive and is associated with resistance.However, genetic variability differs substantially among genes in different regions of the genome (AQUADRO 1992), which means that we need to use different markers to gain a better understanding of the effects selection pressure has on the genetic variability of populations.
Many parameters presented here can change important aspects of the vector competence and capacity of A. aegypti.For instance, bottlenecks and/or genetic drift can dramatically change the phenotype of a population with respect to vector competence (BOSIO et al. 2005).The individual variation of A. aegypti with respect to susceptibility to the dengue virus derives in part from interactions between the genomes of the mosquito and that of the virus, with the potential for infection depending on the specific combination between the genotypes of the vector and the virus (LAMBRECHTS et al. 2009(LAMBRECHTS et al. , 2011)).Resistance to pyrethroids can also affect other aspects of vector ability, longevity for example, and the amount of blood that can be ingested, reducing the number and viability of the eggs (MARTINS et al. 2012).The local dynamics and information about population genetics over time and space should be con-sidered during entomological surveillance of A. aegypti, and used in the design of plans to control the vector population.

Figure 1 .
Figure 1.Map of South America with emphasis on the state of Paraná located in southern Brazil.The study area was the city of Foz do Iguaçu, located at the extreme West of the state of Paraná, divided into 11 strata according to number of buildings.(NR) northern region, strata 1, 2 and 4; (CR) central region, strata 5, 6 and 7; (SR) southern region, strata 9, 10 and 11.

Figure 2 .
Figure 2. Frequencies of the wild (1016Val, gray bars) and mutante (1016Ile, white bars) alleles for the mutation on the 1016 do fragmento codon of the voltage-gated sodium gene of Aedes aegypti in 11 strata of the urban area of Foz do Iguaçu.

Table I .
Generation (G), Chi-square, slope, lethal concentrations, resistance ratios to temephos by the larvae of Aedes aegypti in three urban areas of Foz do Iguaçu in 2008, using the cepa Rockefeller as a control.

Table II .
Number of individuals analyzed and genotypic frequencies for the mutation on the 1016 codon of the gene fragment of the sodium channel of Aedes aegypti, collected in 11 strata in the urban area of Foz do Iguaçu from Dec 2008 to Oct 2009.