• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Oct 2008; 180(2): 1137–1152.
PMCID: PMC2567363

Quantitative Trait Loci Mapping of Genome Regions Controlling Permethrin Resistance in the Mosquito Aedes aegypti


The mosquito Aedes aegypti is the principal vector of dengue and yellow fever flaviviruses. Permethrin is an insecticide used to suppress Ae. aegypti adult populations but metabolic and target site resistance to pyrethroids has evolved in many locations worldwide. Quantitative trait loci (QTL) controlling permethrin survival in Ae. aegypti were mapped in an F3 advanced intercross line. Parents came from a collection of mosquitoes from Isla Mujeres, México, that had been selected for permethrin resistance for two generations and a reference permethrin-susceptible strain originally from New Orleans. Following a 1-hr permethrin exposure, 439 F3 adult mosquitoes were phenotyped as knockdown resistant, knocked down/recovered, or dead. For QTL mapping, single nucleotide polymorphisms (SNPs) were identified at 22 loci with potential antixenobiotic activity including genes encoding cytochrome P450s (CYP), esterases (EST), or glutathione transferases (GST) and at 12 previously mapped loci. Seven antixenobiotic genes mapped to chromosome I, six to chromosome II, and nine to chromosome III. Two QTL of major effect were detected on chromosome III. One corresponds with a SNP previously associated with permethrin resistance in the para sodium channel gene and the second with the CCEunk7o esterase marker. Additional QTL but of relatively minor effect were also found. These included two sex-linked QTL on chromosome I affecting knockdown and recovery and a QTL affecting survival and recovery. On chromosome II, one QTL affecting survival and a second affecting recovery were detected. The patterns confirm that mutations in the para gene cause target-site insensitivity and are the major source of permethrin resistance but that other genes dispersed throughout the genome contribute to recovery and survival of mosquitoes following permethrin exposure.

DENGUE and yellow fever are caused by flaviviruses transmitted by mosquitoes. The principal vector of these flaviviruses on a worldwide basis is the mosquito Aedes aegypti. Dengue fever has become the most prevalent arboviral disease causing morbidity and mortality in most tropical regions (Gubler 2005) and dengue control campaigns rely on a small group of insecticides to prevent disease outbreaks. The most common insecticides for larval Ae. aegypti control are an organophosphate compound called temephos and a carbamate called propoxur. Both of these compounds bind noncompetitively to and inhibit the activity of the enzyme acetylcholine esterase at nerve synapses eventually leading to the buildup of the neurotransmitter acetylcholine and preventing nerve transmission. Adult Ae. aegypti control instead relies on pyrethroids and, a synergist, piperonyl butoxide which is a potent cytochrome P450 inhibitor that can act as the principal detoxification pathway for many insecticides. The pyrethroids are axonic poisons that bind to sodium-gated channels in neuronal membranes causing nerve cells to produce repetitive discharges and eventually paralysis. Both acetylcholine esterase inhibitors and axonic poisons are usually used to reduce selection for resistance associated with use of the same class of insecticide on both larvae and adults (Norma Oficial Mexicana 2003).

Nevertheless, as a result of constant insecticide pressure, Ae. aegypti populations have inevitably evolved resistance mechanisms that include target-site insensitivity and high levels of metabolic detoxification. In México, Ae. aegypti populations from Baja California have elevated metabolic resistance associated with esterases (Flores et al. 2005). In the Yucatán Peninsula, target-site resistance occurs as well as metabolic resistance associated with high levels of esterases and oxidases (Flores et al. 2006). Cuban and Venezuelan populations of Ae. aegypti have been well characterized for target-site insensitivity (Bisset et al. 2006) and various forms of metabolic resistance (Rodriguez et al. 2001, 2002, 2005).

Until recently, identification of metabolic detoxification mechanisms have relied upon synergist bioassays and broad-spectrum biochemical tests (Brogdon and McAllister 1998a). However, in Ae. aegypti there are 26 glutathione transferases, 160 cytochrome P450s, 49 carboxy/cholinesterases, and 67 genes encoding proteins with various types of oxidase activities (Strode et al. (2008). This high diversity of detoxification enzymes and difficulties associated with enzyme isolation and characterization has prevented an understanding of the involvement of individual genes in the overall detoxification pathway. Targeted microarray chips containing CYP, EST, GST, and various oxidase genes discovered in genome projects have been used to identify specific genes with elevated levels of transcription in resistant strains of the principal malaria vectors, Anopheles gambiae (David et al. 2005), An. stephensi (Vontas et al. 2007), and Ae. aegypti (Strode et al. (2008). These targeted arrays greatly simplified screening for the principal enzymes involved in insecticide detoxification. Limitations associated with the use of targeted microarray chips are that only those genes printed on the chip can be examined and only those genes whose function is related to changes in transcriptional activity can be identified. While proteomic approaches toward identifying resistant genes in Drosophila melanogaster have been developed (Pedra et al. 2005), similar tools have not been applied in mosquitoes.

Quantitative trait loci (QTL) mapping is a tool that can be used to identify genome regions associated with insecticide resistance. The power of QTL mapping is that it does not require the use of candidate genes and also makes no assumptions about resistance mechanisms. It can instead be used to test whether candidate genes are associated with resistance phenotypes in a controlled, common environment in the laboratory. QTL mapping may also eventually lead to the identification of novel transcription factors and regulatory elements that regulate the CYP, EST, and GSTs that directly metabolize insecticides. QTL mapping was used to identify genome regions conferring DDT and pyrethroid resistance in An. gambiae (Ranson et al. 2000, 2004) and An. funestus (Wondji et al. 2007a,b). Despite having an unassembled genome sequence (Nene et al. 2007), no similar QTL mapping study has been completed for resistance-associated genome regions in Ae. aegypti.

QTL mapping requires constructing families from parents with distinct phenotypes (e.g., permethrin susceptible or resistant). In an intercross design, F1 siblings are allowed to randomly mate and the resulting F2 offspring are phenotyped and genotypes are then determined at markers covering the genome at a density of ~1 marker/5 cM. A family that is continued through random intercrossing of F2 siblings is called an advanced intercross line (AIL). AILs allow for analysis of a larger pool of siblings thus increasing the statistical power of the QTL map and creating additional generations of recombination to allow more precise determination of QTL location. In Ae. aegypti a dense linkage map of RFLP, RAPD, and SSCP markers, covering the three chromosomes is available (Fulton et al. 2001; Black and Severson 2004).

Herein, we report on the selection of a permethrin-resistant Ae. aegypti strain from Isla Mujeres (IMU), Quintana Roo, México. An F3 AIL was constructed using parents from a reference-susceptible strain originally from New Orleans (Flores et al. 2006) and the IMU-F4 strain that had been subjected to two generations of selection with permethrin. A total of 439 F3 individuals were phenotyped as knockdown resistant, knocked down/recovered, or dead following permethrin exposure. We developed 34 SNP markers and mapped the locations of 21 antixenobiotic genes onto the Ae. aegypti linkage map. Genome regions affecting knockdown, recovery, and survival following permethrin exposure were then mapped using composite interval (Zeng 1994) and multiple interval (Kao et al. 1999) QTL mapping techniques.


Mosquito collection and bioassays:

Larvae from Isla Mujeres, Quintana Roo, México (latitude 21.2345, longitude 86.7316), were collected and transported to the Medical Entomology Laboratory at the Universidad Autonoma de Nuevo Leon in Monterrey, México. An F1 was reared, blood fed, and used to generate a large F2 generation designated IMU-F2. F2 eggs were sent to and hatched at Colorado State University. The concentration of permethrin that caused 50% mortality (LC50) was determined by releasing 40 3- to 4-day old adults into 250 ml Wheaton bottles with the inside walls coated with a known amount of permethrin (technical grade; Chem Services, West Chester, PA) following Brogdon and McAllister (1998b). The amounts used were 0.0, 0.7, 1.0, 2.5, 5.0, and 10.0 μg active ingredient (a.i.)/bottle. Following a 1-hr exposure, mosquitoes were transferred to cardboard containers and placed into a 28° incubator with 80% relative humidity, and a 14:10 photoperiod. After 24 hr, exposed mosquitoes were recorded as alive or dead. The cumulative number of dead mosquitoes was plotted against permethrin concentration and logistic regression on SAS 9.1 (Cary, NC) was used to estimate an LC50 and LC90. These were 2.7 and 7.0 μg a.i./bottle, respectively. The LC50 for the New Orleans (NO) standard susceptible strain (Flores et al. 2006) was consistently ~0.2 μg a.i./bottle.

Permethrin-resistant-strain selection:

A separate set of adults of the IMU-F2 generation was exposed to 5.0 μg a.i./bottle for 1 hr. Survivors were transferred to a cage and were blood fed with mice, allowing oviposition of the first selected generation designated IMU-F3. The IMU-F3 was hatched and adults were again exposed to a 5.0 μg a.i./bottle; survivors were allowed to mate to produce the next progeny IMU-F4 that were used as parents for QTL mapping. Table 1 shows the survivorship and the LC50 for the NO strain and for IMU-F2–IMU-F4 generations. Mortality in NO when exposed to a 5.0-μg a.i./bottle was consistently 100%.

Selection of the Isla Mujeres (IMU-F4) permethrin-resistant strain

Mapping family crosses:

For the P1 mapping family, we crossed IMU-F4 and NO adults. Twenty P1 ♀IMU-F4 × ♂NO and 20 reciprocal P1 ♀NO × ♂IMU-F4 crosses were made. Larvae from each line were hatched and at the pupal stage, a female (as judged by size) from one strain was transferred to plastic cups in cardboard containers with a male pupa from the other strain. After adults emerged, they were allowed to mate for 3 days and the P1 male was frozen and held at −80°. Females were blood fed three times with mice over the next 10 days and the P1 female was then frozen and held at −80°. Egg batches were maintained at room temperature for 7 days and then hatched by submersion in water followed by feeding them on Brewer's yeast. For the F1 intercross families, one female and one male pupae from the same P1 family were allowed to emerge, mate, and blood feed to obtain an F2 progeny. F2 eggs from the five largest F1 families were hatched and siblings were intercrossed in a single cage.

Resistance phenotyping of mapping families:

A portion of the adults from each of the five families were used to estimate the LC50. Among the five F3 families the LC50 ranged from 0.7 to 1.5 μg a.i./bottle. A family with an LC50 of 1.2 μg a.i./bottle was chosen for the mapping study. F3 adults that were 3–4 days old (791 total) were exposed to 1.2 μg a.i./bottle for 1 hr and adults were classified as knockdown resistant (kdr) or knocked down. Flying kdr mosquitoes were mechanically aspirated from the exposure bottle, transferred to a cardboard container, and frozen and held at −80°. Knocked-down mosquitoes were transferred to a different cardboard container and were maintained in an incubator at the conditions described above. Four hours later, the container was removed and flying and crawling mosquitoes were recorded as recovered and aspirated, frozen, and held at −80°. The remaining dead mosquitoes were then frozen and held at −80°. Normally mosquitoes are assayed for recovery after 24 hr, however in initial trials we obtained poor recovery of DNA from mosquitoes that had been dead for this length of time and furthermore we have routinely observed that few additional mosquitoes recover after 4 hr (our unpublished data).

DNA extraction:

The DNA of the P1 and F1 parents, and the 439 (226 ♀ + 213 ♂) F3 offspring was individually isolated following the DNA salt extraction method (Black and Duteau 1997) and suspended in 200 μl of TE buffer (10 mm Tris-HCl, 1 mm EDTA pH 8.0). The DNA was divided into 2- to 100-μl aliquots and stored at −80°. Many candidate genes were screened for segregation among the P1 and F1 adults. This required that total genomic DNA be amplified from the P1 and F1 DNA samples using multiple displacement amplification (Gorrochotegui-Escalante and Black 2003) with the TempliPhi 500 amplification kit (Amersham Biosciences). Amplified DNA was suspended in 90 μl ddH2O and stored at −20°.

PCR of cDNA-SSCP markers:

A total of 43 cDNA-SSCP markers (Fulton et al. 2001; Gomez-Machorro et al. 2004) and 91 CYP, EST, GST, and oxidase genes described by Strode et al. (2008) were tested for polymorphisms in the P1, F1, and ~20 F3 mosquitoes (Tables 2 and and3).3). PCR products between 170 and 370 bp from CYP, EST, and GST genes were designed using Primer Premier software. PCR reaction mixture sufficient to perform 25 50-μl reactions was made by mixing 1057 μl ddH2O, 125 μl 10 × Taq buffer (500 mm KCl, 100 mm Tris-HCL pH 9.0), 12.5 μl of 20 mm dNTPs, and 1250 pmol of each of the primers. This reaction mixture was set under a UV light source (302 nm) for 10 min, after which 10 μl of Taq DNA polymerase were added. The mixture was then dispensed into a 96-well plate. Template DNA (~100 ng) was then added to each well, followed by a drop of sterilized mineral oil. Each set of reactions was checked for contamination by the use of a negative control containing all reagents except template DNA. Samples were stored at 4° before electrophoresis. The contents of each well were tested for the presence of amplified products by loading 5 μl from each well onto a 1.5% (w/v) agarose gel made with Tris-Borate-EDTA buffer. DNA fragments were size fractionated by electrophoresis for 15–20 min at 112 V. Fragments were visualized by staining with Syber Green and viewing the gel over a UV transilluminator.

Genetic markers with known linkage locations (Black and Severson 2004) in the Aedes aegypti genome and associated primers for PCR amplification
Ninety-one selected cytochrome P450s (CYP), esterases (CCE), or glutathione transferases (GST) and various oxidase genes and associated primers and conditions for PCR amplification

SSCP analysis and silver staining procedures followed Black and Duteau (1997). Polymorphic SSCP markers were sequenced in the four P1 and F1 parents to test for SNPs and to determine the inheritance patterns of SNP alleles. Sequences were aligned using CLUSTALW (Thompson et al. 1994). Allele-specific primers were designed at those loci in which genotypes were fully or partially informative in the P1 and F1 parents. Design of primers for melting curve PCR is fully explained in Saavedra-Rodriguez et al. (2007). Allele-specific fragments were detected by melting curve PCR in an Opticon 2 DNA Engine (MJ Research, Waltham MA).

Linkage mapping:

Genotypes at each putative locus were entered into JoinMap 2.0 for a “recombinant inbreed RI3” cross (Stam 1993). These were tested for conformity to Mendelian ratios with a χ2 goodness-of-fit analysis using the JMSLA procedure in JoinMap. Loci at which Mendelian genotype ratios were observed were separated into individual linkage groups using the JMGRP by increasing the minimal LOD threshold from 0.0 up to 8.0 in increments of 0.1. After markers were assigned to linkage groups, the data set was split into three groups using JMSPL. Pairwise distances (Kosambi 1943) were then estimated among loci on each of the three linkage groups using JMREC and a maximum likelihood map was estimated using JMMAP.

QTL analysis:

Associations between genotypes at each marker locus and susceptibility phenotypes were initially assessed by a Fisher's exact test. The null hypothesis was that the numbers of kdr, recovered, and dead mosquitoes were equal in each genotype class. When the probability of the Fisher's exact test was <0.05, we examined the inheritance of the alleles at that locus. Our a priori hypothesis was that an excess of F3 individuals with an allele inherited from the IMU-F4 P1 parent would be kdr or recovered following permethrin exposure while an excess of F3 individuals with an allele inherited from the NO P1 parent would die following exposure.

Composite interval mapping (CIM) (Zeng 1994) was then performed using QTL Cartographer 2.5 (Wang et al. 2007). The number of separate regions (np) was set to the number of regions identified in the initial Fisher's exact test and walking speed (ws) was set to 2 cM. Permutations (n = 300) were run to establish a 95% experimentwise threshold. Three separate CIM were done. First, mosquitoes with kdr were scored as 2, recovered as 1, and dead as 0. Next, F3 mosquitoes with kdr were scored as 1, and recovered or dead were scored as 0 to test for kdr QTL. Lastly, surviving F3 mosquitoes (kdr + recovered) were scored as 1, and dead were scored as 0 to test for survival QTL.

Multiple interval mapping (MIM) (Kao et al. 1999) was also performed three separate times for kdr, recovered, or survival phenotypes using QTL Cartographer 2.5. In each case we (1) entered QTL map positions as detected by CIM, (2) estimated QTL effects, and (3) obtained and recorded a summary. The derived model was further refined in MIM by (1) searching for new QTL, (2) estimating QTL effects, (3) obtaining and recording a summary, (4) optimizing QTL position, (5) searching for new QTL interactions, (6) testing for existing QTL main effects, (7) testing for existing QTL interaction effects, and (8) obtaining and recording a final summary.


Susceptibility phenotype:

Of the 771 F3 mosquitoes assayed for permethrin susceptibility, 16% (127) exhibited kdr and were flying after 1-hr exposure to 1.2 μg a.i./bottle. The remainder were knocked down and immobile on the bottom of the bottle but 293 (35%) of these recovered and were flying 4 hr postexposure. These were collected and scored as recovered. The remaining 351 were scored as dead. Of these, 439 mosquitoes were used in QTL mapping. This included 226 females (75 kdr, 76 recovered, and 75 dead) and 213 males (68 kdr, 70 recovered, and 75 dead). The remainder of mosquitoes are stored in the freezer but were not used to reduce costs.

Marker generation:

Of the 55 previously mapped cDNA-SSCP markers, 29 were polymorphic by SSCP and 12 of these were used for mapping (Table 2). Of the 235 CYP, EST, or GST genes described in Strode et al. (2008) we biased our selection to include those genes that were overexpressed in a microarray analysis in the Isla Mujeres strain (Strode et al. 2008). Of the 91 selected CYP, EST, GST, and various oxidase genes selected, 61 were polymorphic by SSCP (Table 3). Sequences of 70 of the 88 variable markers and putative resistance genes were then analyzed for informative SNPs in the P1 and F1 parents. On the basis of this information, 32 allele-specific PCR systems were developed to detect one SNP per gene (22 putative resistance markers and 10 genomic markers) (Table 4). Transfer and sin3 were genotyped by SSCP. Sequences of the allele-specific and reverse primers, the SNP position with respect to Vectorbase (http://aaegypti.vectorbase.org/index.php) annotation, and the linkage position of the locus on the three chromosomes are shown in Table 4. Males were scored as heterozygous at the Sex locus while females were scored as homozygous (Gilchrist and Haldane 1947). The putative resistance markers included 16 CYP and 4 EST genes and 1 GST gene and the para gene (voltage-dependent sodium channel). The linkage location of para was previously mapped (Severson et al. 1997). The para SNP marker identifies the Val1016Iso substitution that we had previously shown to be associated with permethrin resistance in field populations of Ae. aegypti (Saavedra-Rodriguez et al. 2007).

Oligonucleotides used for allele specific PCR

Linkage mapping:

Genotypes at the 34 marker loci were analyzed in the P1 and F1 parents and in the 439 F3 offspring. The probability in the Fisher's exact test appears next to markers that had a test probability <0.05 (Figure 1). The linkage positions obtained with SNP markers were mostly consistent with those published by Black and Severson (2004). Three linkage groups were obtained with the JMGRP program in JoinMap 2.0 from an LOD threshold of 3.1–5.9. We used previously published centimorgan estimates from Black and Severson (2004) to fix gene order on chromosomes. Figure 1 shows the linkage positions of genomic and putative resistance markers.

Figure 1.
Linkage map of cytochrome P450s (CYP), esterases (EST), or glutathione transferases (GST) in the Aedes aegypti IMU-F4 × New Orleans advanced intercross family. Linkage positions of novel insecticide resistance marker loci appear with asterisks ...

QTL analysis:

Linkage positions derived above were entered into QTL Cartographer 2.5 along with the phenotype scores of all individuals. Scoring mosquitoes as kdr, recovered, or dead, CIM detected two QTL on chromosome I at map positions 38 cM and 60–65 cM and a QTL at 30 cM on chromosome II (Figure 2). At least two QTL of large effect were detected on chromosome III at map positions 24 and 31 cM. Note that the LOD scale for chromosome III in Figure 2 is ~10 times greater than those for chromosomes I and II.

Figure 2.
Plot of LOD values associated with kdr, recovery, or survival along chromosomes I–III in the Aedes aegypti IMU-F4 × New Orleans advanced intercross family. LOD values were estimated by composite interval mapping in QTL Cartographer 2.5 ...

Next, F3 mosquitoes with or without kdr were analyzed with CIM. CIM detected the 38-cM QTL again on chromosome I and a second smaller QTL at 36 cM (Figure 2). No QTL was detected on chromosome II and 3 QTL at 10, 24, and 31 cM were detected on chromosome III. Lastly, CIM identified QTL between surviving and dead F3 mosquitoes and detected the same 60–65 cM QTL detected earlier on chromosome I (Figure 2). A new QTL was detected on chromosome II at map position 49 cM. The two QTL of large effect were again detected on chromosome III at map positions 24 and 31 cM.

The QTL detected by CIM were entered into a MIM model to estimate the phenotypic variance (σp2) for the entire model and the broad sense genotypic variance (σg2) for the model and for individual QTL. MIM also calculated residual or environmental variance (σe2), the map position in centimorgans, the nearest marker, and additive and dominance effects for the entire model and for individual QTL (Table 5). Models were developed for the three phenotype comparisons (kdr, recovered, and dead), for kdr, and for survival. The σp2 was largest (0.6672) for the three phenotype comparisons and σg2 accounted for 70% of this variance. Most (59.3%) of σp2 was accounted for by chromosome III QTL at map positions 24.6 (16.4%) and 31 cM (42.9%). The nearest markers to these were CCEunk7o and para, respectively.

Multiple-interval mapping estimates of QTL position and associated genetic, environmental, and phenotypic variance and additive and dominance effects associated with knockdown, recovery and survival QTL in Aedes aegypti

The MIM model for kdr had the largest percentage (84.1%) of σp2 accounted for by σg2. Again, most (79.1%) of σp2 is accounted for by chromosome III QTL at map positions 24.7 (20.5%) and 31 cM (58.6%). The MIM model for survival had the smallest percentage (36.8%) of σp2 accounted for by σg2. The chromosome III QTL at map positions 24.7 (9.3%) and 31 cM (23.1%) accounted for most of σg2.

Individual QTL effects on each phenotype:

Clearly the chromosome III QTL at 24.7 and 31 cM had the largest effects on knockdown, recovery, and survival. The closest markers to these QTL were CCEunk7o at 24 cM and the Val1016Iso substitution in para at 31 cM. The knockdown, recovery, and survival rates for each of the three para and CCEunk7o genotypes are shown for females and males separately in Figure 3. At both QTL susceptibility alleles inherited from the New Orleans susceptible (S) parent were dominant in their effects on knockdown. Heterozygotes and S homozygotes had a 0.8–1.0 knockdown rate while homozygotes for alleles inherited from the Isla Mujeres resistant (R) parent had a 0.0–0.1 knockdown rate (Figure 3A). In contrast, genotypes at both QTL appeared to be overdominant on their effects on recovery (Figure 3B). Recovery was 0.5–0.6 in heterozygotes but 0.0–0.3 in either of the homozygote classes. However, note that the 95% confidence intervals surrounding R/R estimates were large because very few mosquitoes with this genotype were actually knocked down. A third pattern of QTL effects was seen with survival (Figure 3C). Among R homozygotes, the survival rate was 0.9–1.0, 0.6–0.7 among heterozygotes, and 0.0–0.4 for S homozygotes. Differences among females and males were not significant.

Figure 3.
Plot of knockdown, recovery, and survival rates as a function of F3 genotypes at the para and CCEunk7o loci. R/R indicates that both alleles were inherited from the IMU-F4 resistant P1 parent; S/S indicates that both alleles were inherited from the New ...

Two additional QTL on chromosome III were also detected and the closest associated markers were CCEae1o at 7 cM and CYP4H32 at 19 cM (Figure 2). At the 7-cM QTL, the effect of the S allele on knockdown was partially dominant (Figure 4A) with half of R homozygotes knocked down, 0.85–0.95 of heterozygotes knocked down and all S homozygotes knocked down. At the 19-cM QTL, the S allele was also partially dominant in affecting knockdown (Figure 4A) with 0.25 of R homozygotes knocked down, 0.7–0.8 of heterozygotes knocked down, and 0.80–0.85 of S homozygotes knocked down. At the 7-cM QTL, the R allele appears to be dominant to the S allele in conditioning recovery (Figure 4B) with half of R homozygotes and heterozygotes recovering but none of the S homozygotes recovering. Alleles at the 19-cM QTL appeared to be additive in conditioning the recovery rate in males with 0.80 recovery in R homozygotes, 0.55 in heterozygotes, and 0.15 in S homozygotes (Figure 4B). In contrast, recovery appeared to follow a pattern of overdominance in females at this QTL, with 0.4 of females homozygous for either alleles recovering but 0.6 of heterozygous females recovering. The R allele at the 7-cM QTL was also partially dominant in its effect on survival (Figure 4C) with 0.75 of R homozygotes surviving, 0.60 of heterozygotes surviving, and no S homozygotes surviving. Alleles at the 19-cM QTL appeared to be additive in conditioning survival and there were sex-specific differences. Male R homozygotes had 0.95 survival while 0.65 of heterozygotes and 0.25 of S homozygotes survived. In contrast, female R homozygotes had 0.85 survival while 0.7 of heterozygotes and 0.45 of S homozygotes survived.

Figure 4.
Plot of knockdown, recovery, and survival rates as a function of F3 genotypes at the CCEae1o and CYP4H32 loci. Labels along the abscissa and rates along the ordinate axis are as explained in Figure 3.

QTL of relatively minor effect were detected on chromosome I at positions 36, 38, and 66 cM and on chromosome II at 30 and 49 cM (Figure 2). The 36-cM (nearest marker, CYP6P12V1) and 38 cM (nearest marker, Sex) QTL were associated with knockdown. Because equal numbers of male and female mosquitoes from the three phenotypic classes were selected for genotyping, we could not analyze knockdown with regards to Sex. However, the 38-cM QTL probably reflects differences in knockdown rates between sexes (Figures 4A and and5A).5A). The 36-cM QTL near CYP6P12V1 is interesting in that, contrary to our a priori hypothesis the R homozygotes have a 0.05–0.20 greater knockdown rate than the heterozygotes or S homozygotes (Figure 5A).

Figure 5.
Plot of knockdown, recovery, and survival rates as a function of F3 genotypes at the CYP6BB2 and CYP6P12V1 loci on chromosome I and CYP12F5 and CYP9M8 on chromosome II. Labels along the abscissa and rates along the ordinate axis are as explained in Figure ...

The 38-cM (nearest marker, Sex) and 66-cM (nearest marker, CYP6BB2) QTL and the 31-cM QTL on chromosome II affected recovery rate (Figure 5B). Again, note that the recovery rate differed between the sexes at the QTL near CYP4H32 (Figure 4B) and at the QTL near CYP6BB2 and CYP6P12V1 markers (Figure 5B). Genotypes at the 66-cM QTL were overdominant with a recovery rate of 0.55–0.70 among R homozygotes, 0.65–0.90 among heterozygotes, and 0.2–0.30 among S/S homozygotes (Figure 5B).

QTL affecting survival (Figure 5C) were detected on chromosome I at 66 cM and on chromosome II at 49 cM. Survival among genotypes at the QTL near CYP6BB2 differed between sexes. Female R homozygotes had 0.75 survival while 0.55 of males survived. Among heterozygotes, 0.9 of females survived while 0.8 of males survived. Male and female S homozygotes had an equal survival rate of 0.55 (Figure 5C). Survival among genotypes at the 49-cM QTL on chromosome II followed a partially dominant pattern in females with 0.65 survival among R homozygotes and heterozygotes and 0.75 among S homozygotes (Figure 5C) but an additive pattern was noted in males with 0.45 survival among R homozygotes, 0.60 among heterozygotes, and all S homozygotes survived.


We used artificial selection to produce in just two generations a strain of Ae. aegypti with ~24-fold greater resistance to permethrin than the original, unselected Isla Mujeres collection and ~300-fold greater resistance than the New Orleans standard susceptible strain (Table 1). In this selected strain, we detected five kdr QTL near markers CYP6P12V1 and Sex on chromosome I and near markers CCEae1o, CYP4H32, CCEunk7o, and para on chromosome III. Three QTL that condition recovery were detected on chromosome I near CYP6BB2, on chromosome II near CYP12F5, and on chromosome III near CCEunk7o and para. One additional QTL conditioned survival and was located on chromosome II near CYP9M8.

The chromosome III 24- and 31-cM QTL accounted for 59.3% of σp2 for knockdown, recovery, or death, 79.1% of knockdown σp2, and 31.3% of σp2 for survival. The large contributions by the 31-cM QTL to kdr were expected and corresponded to a previously characterized ValIso replacement substitution at codon 1016 in hydrophobic segment 6 of domain II of para. We found this mutation to be associated with kdr in Ae. aegypti populations from throughout Latin America (Saavedra-Rodriguez et al. 2007). Generally mutations in this region of para in insects reduce permethrin binding and allow normal functioning of the sodium-gated channels in neuronal membranes (Soderlund and Knipple 2003).

What remains unclear is whether the 24-cM QTL nearest CCEunk7o represents an independent kdr QTL or is a marker that was swept along during selection for kdr associated with the nearby para locus. In the QTL analyses, genotypes at CCEunk7o and para were not independent. We had no knowledge of the Val1016Iso substitution when this mapping study was initiated and so we could not have selected out this mutation prior to performing the F1 intercrosses.

Despite the rapid response to selection, the QTL patterns that we have detected suggest that a diversity of loci and mechanisms in the Ae. aegypti genome respond to selection for pyrethroid resistance. Furthermore, these loci condition different phenotypes associated with resistance evolution. Some determine whether knockdown occurs (Figures 3A–5A),), while others affect recovery following knockdown (Figures 3B–5B)) and at least one QTL (Figure 5C) exclusively affected survival. The genes underlying these QTL probably act sequentially in determining the overall resistance response. QTL that prevent or reduce pyrethroid binding in the sodium-gated channels prevent knockdown. However, among knocked-down mosquitoes, other QTL may affect the subsequent metabolic degradation of the pesticide, ultimately removing pyrethroid from their systems and allowing these mosquitoes to recover. None of the three MIM models displayed in Table 5 accounted for all of the σp2. From 15.9 to 56.5% of σp2 was residual variance, cumulatively unaccounted for by the identified QTL. Thus “environmental,” uncontrolled factors in experimental design and execution also account for a substantial part of σp2 in resistance. Our QTL map contains 34 markers that cover 174 cM of the 204-cM Ae. aegypti linkage map (Black and Severson 2004). There was a 16-cM gap in marker coverage on the top of chromosome I, a 25-cM and 18-cM gap at the top and bottom of chromosome II, respectively, and a 15-cM gap at the bottom of chromosome III. The largest gap unbounded by a marker was at the top of chromosome II. Nine markers on the top of chromosome II were tested but none were informative. Thus it is possible that the top of chromosome II contained additional resistance QTL.

This study represents only two collections of Ae. aegypti. An obvious question is whether other geographic populations, or even replicate collections from Isla Mujeres, would respond in the same way to selection with pyrethroids. This is being addressed by exploring for other QTL using mosquitoes from different geographic locations. The polygenic, quantitative genetic patterns that we have observed in these experiments are supported by earlier studies that approached insecticide resistance evolution as a quantitative rather than discrete genetic character (Roush and McKenzie 1987; Firko and Hayes 1990; Ferrari and Georghiou 1991; Morton 1993; Roush 1993; Guillemaud et al. 1999; Paton et al. 2000; Oakeshott et al. 2003). Furthermore, of the few insecticide resistance QTL mapping studies completed to date, all have reported multiple regions that respond to selection with insecticides (Ranson et al. 2000, 2004; Hawthorne 2003; Jallow and Hoy 2006; Wondji et al. 2007c).


This work was supported by the Innovative Vector Control Consortium.


  • Bisset, J., M. M. Rodriguez and D. Fernandez, 2006. Selection of insensitive acetylcholinesterase as a resistance mechanism in Aedes aegypti (Diptera: Culicidae) from Santiago de Cuba. J. Med. Entomol. 43 1185–1189. [PubMed]
  • Black, W. C., and N. M. DuTeau, 1997. RAPD-PCR and SSCP analysis for insect population genetic studies, pp. 361–373 in The Molecular Biology of Insect Disease Vectors: A Methods Manual, edited by J. Crampton, C. B. Beard, and C. Louis. Chapman & Hall, New York.
  • Black, W. C., and D. W. Severson, 2004. Genetics of vector competence, pp. 415–448 in Biology of Disease Vectors, Ed. 2, edited by W. C. Marquardt. Elsevier, Amsterdam.
  • Brogdon, W. G., and J. C. McAllister, 1998. a Insecticide resistance and vector control. Emerging Infect. Dis. 4 605–613. [PMC free article] [PubMed]
  • Brogdon, W. G., and J. C. McAllister, 1998. b Simplification of adult mosquito bioassays through use of time-mortality determinations in glass bottles. J. Am. Mosq. Control Assoc. 14 159–164. [PubMed]
  • David, J. P., C. Strode, J. Vontas, D. Nikou, A. Vaughan et al., 2005. The Anopheles gambiae detoxification chip: a highly specific microarray to study metabolic-based insecticide resistance in malaria vectors. Proc. Natl. Acad. Sci. USA 102 4080–4084. [PMC free article] [PubMed]
  • Ferrari, J. A., and G. P. Georghiou, 1991. Quantitative genetic variation of esterase activity associated with a gene amplification in Culex quinquefasciatus. Heredity 66(Pt 2): 265–272. [PubMed]
  • Firko, M. J., and J. L. Hayes, 1990. Quantitative genetic tools for insecticide resistance risk assessment: estimating the heritability of resistance. J. Econ. Entomol. 83 647–654. [PubMed]
  • Flores, A. E., W. Albeldano-Vazquez, I. F. Salas, M. H. Badii, H. L. Becerra et al., 2005. Elevated alpha-esterase levels associated with permethrin tolerance in Aedes aegypti (L.) from Baja California, Mexico. Pest. Biochem. Physiol. 82 66–78.
  • Flores, A. E., J. S. Grajales, I. F. Salas, G. P. Garica, M. H. L. Becerra et al., 2006. Mechanisms of insecticide resistance in field populations of Aedes aegypti (L.) from Quintana Roo, Southern Mexico. J. Am. Mosq. Control Assoc. 22 672–677. [PubMed]
  • Fulton, R. E., M. L. Salasek, N. M. DuTeau and W. C. Black, 2001. SSCP analysis of cDNA markers provides a dense linkage map of the Aedes aegypti genome. Genetics 158 715–726. [PMC free article] [PubMed]
  • Gilchrist, B. M., and J. B. S. Haldane, 1947. Sex linkage and sex determination in a mosquito, Culex molestus. Hereditas 33 175–190.
  • Gomez-Machorro, C., K. E. Bennett, M. D. Munoz and W. C. Black, 2004. Quantitative trait loci affecting dengue midgut infection barriers in an advanced intercross line of Aedes aegypti. Insect Mol. Biol. 13 637–648. [PubMed]
  • Gorrochotegui-Escalante, N., and W. C. Black, 2003. Amplifying whole insect genomes with multiple displacement amplification. Insect Mol. Biol. 12 195–200. [PubMed]
  • Gubler, D., 2005. The emergence of epidemic dengue fever and dengue hemorrhagic fever in the Americas: a case of failed public health policy. Rev. Panam. Salud Publica-Panamer. J. Publ Health 17 221–224. [PubMed]
  • Guillemaud, T., M. Raymond, A. Tsagkarakou, C. Bernard, P. Rochard et al., 1999. Quantitative variation and selection of esterase gene amplification in Culex pipiens. Heredity 83 87–99. [PubMed]
  • Hawthorne, D. J., 2003. Quantitative trait locus mapping of pyrethroid resistance in Colorado potato beetle, Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae). J. Econ. Entomol. 96 1021–1030. [PubMed]
  • Jallow, M. F., and C. W. Hoy, 2006. Quantitative genetics of adult behavioral response and larval physiological tolerance to permethrin in diamondback moth (Lepidoptera: Plutellidae). J. Econ. Entomol. 99 1388–1395. [PubMed]
  • Kao, C. H., Z. B. Zeng and R. D. Teasdale, 1999. Multiple interval mapping for quantitative trait loci. Genetics 152 1203–1216. [PMC free article] [PubMed]
  • Kosambi, D. D., 1943. The estimation of map distances from recombination values. Ann. Eugen. 12 172–175.
  • Morton, R. A., 1993. Evolution of Drosophila insecticide resistance. Genome 36 1–7. [PubMed]
  • Nene, V., J. R. Wortman, D. Lawson, B. Haas, C. Kodira et al., 2007. Genome sequence of Aedes aegypti, a major arbovirus vector. Science 316 1718–1723. [PMC free article] [PubMed]
  • Norma Oficial Mexicana, 2003. NOM-032-SSA-2–2002 para la vigilancia epidemiologica, prevencion y control de enfermedades transmitidas por vector (D.O.F. 21 julio del 2003). Norma Oficial Mexicana, Mexico City.
  • Oakeshott, J. G., I. Home, T. D. Sutherland and R. J. Russell, 2003. The genomics of insecticide resistance. Genome Biol. 4 202. [PMC free article] [PubMed]
  • Paton, M. G., S. H. Karunaratne, E. Giakoumaki, N. Roberts and J. Hemingway, 2000. Quantitative analysis of gene amplification in insecticide-resistant Culex mosquitoes. Biochem. J. 346(Pt 1): 17–24. [PMC free article] [PubMed]
  • Pedra, J. H. F., R. A. Festucci-Buselli, W. L. Sun, W. M. Muir, M. E. Scharf et al., 2005. Profiling of abundant proteins associated with dichlorodiphenyltrichloroethane resistance in Drosophila melanogaster. Proteomics 5 258–269. [PubMed]
  • Ranson, H., B. Jensen, X. Wang, L. Prapanthadara, J. Hemingway et al., 2000. Genetic mapping of two loci affecting DDT resistance in the malaria vector Anopheles gambiae. Insect Mol. Biol. 9 499–507. [PubMed]
  • Ranson, H., M. G. Paton, B. Jensen, L. McCarroll, A. Vaughan et al., 2004. Genetic mapping of genes conferring permethrin resistance in the malaria vector, Anopheles gambiae. Insect Mol. Biol. 13 379–386. [PubMed]
  • Rodriguez, M. M., J. Bisset, D. M. De Fernandez, L. Lauzan and A. Soca, 2001. Detection of insecticide resistance in Aedes aegypti (Diptera: Culicidae) from Cuba and Venezuela. J. Med. Entomol. 38 623–628. [PubMed]
  • Rodriguez, M. M., J. Bisset, M. Ruiz and A. Soca, 2002. Cross-resistance to pyrethroid and organophosphorus insecticides induced by selection with temephos in Aedes aegypti (Diptera: Culicidae) from Cuba. J. Med. Entomol. 39 882–888. [PubMed]
  • Rodriguez, M. M., J. A. Bisset, Y. De Armas and F. Ramos, 2005. Pyrethroid insecticide-resistant strain of Aedes aegypti from Cuba induced by deltamethrin selection. J. Am. Mosq. Control Assoc. 21 437–445. [PubMed]
  • Roush, R. T., 1993. Occurrence, genetics and management of insecticide resistance. Parasitol. Today 9 174–179. [PubMed]
  • Roush, R. T., and J. A. McKenzie, 1987. Ecological genetics of insecticide and acaricide resistance. Annu. Rev. Entomol. 32 361–380. [PubMed]
  • Saavedra-Rodriguez, K., L. Urdaneta-Marquez, S. Rajatileka, M. Moulton, A. E. Flores et al., 2007. A mutation in the voltage-gated sodium channel gene associated with pyrethroid resistance in Latin American Aedes aegypti. Insect Mol. Biol. 16 785–798. [PubMed]
  • Severson, D. W., N. M. Anthony, O. Andreev and R. H. ffrench-Constant, 1997. Molecular mapping of insecticide resistance genes in the yellow fever mosquito (Aedes aegypti). J. Hered. 88 520–524. [PubMed]
  • Soderlund, D. M., and D. C. Knipple, 2003. The molecular biology of knockdown resistance to pyrethroid insecticides. Insect Biochem. Mol. Biol. 33 563–577. [PubMed]
  • Stam, P., 1993. Construction of integrated genetic-linkage maps by means of a new computer package: Joinmap. Plant J. 3 739–744.
  • Strode, C., C. S. Wondji, J. P. David, N. J. Hawkes, N. Lumjuan et al., 2008. Genomic analysis of detoxification genes in the mosquito Aedes aegypti. Insect Biochem. Mol. Biol. 38 113–123. [PubMed]
  • Thompson, J. D., D. G. Higgins and T. J. Gibson, 1994. Clustal-W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22 4673–4680. [PMC free article] [PubMed]
  • Vontas, J., J. P. David, D. Nikou, J. Hemingway, G. K. Christophides et al., 2007. Transcriptional analysis of insecticide resistance in Anopheles stephensi using cross-species microarray hybridization. Insect Mol. Biol. 16 315–324. [PubMed]
  • Wang, S., C. J. Basten and Z.-B. Zeng, 2007. Windows QTL Cartographer 2.5. North Carolina State University, Raleigh, NC.
  • Wondji, C. S., J. Hemingway and H. Ranson, 2007. a Identification and analysis of single nucleotide polymorphisms (SNPs) in the mosquito Anopheles funestus, malaria vector. BMC Genomics 8 5. [PMC free article] [PubMed]
  • Wondji, C. S., J. Morgan, M. Coetzee, R. H. Hunt, K. Steen et al., 2007. b Mapping a quantitative trait locus (QTL) conferring pyrethroid resistance in the African malaria vector Anopheles funestus. BMC Genomics 8 34. [PMC free article] [PubMed]
  • Zeng, Z. B., 1994. Precision mapping of quantitative trait loci. Genetics 136 1457–1468. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
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...