La Crosse virus is a leading cause of pediatric encephalitis in the United States. The mosquito Ochlerotatus triseriatus is an efficient vector for La Crosse virus, whereas the closely related O. hendersoni transmits only at very low rates. Quantitative trait loci (QTL) affecting the ability to orally transmit this virus and adult body size were identified in 164 F₂ female individuals from interspecific crosses of O. hendersoni females and O. triseriatus males using a combination of composite interval mapping (CIM), interval mapping (IM) for binary traits, and single-marker mapping. For oral transmission (OT), no genome locations exceeded the 95% experimentwise threshold for declaring a QTL using IM, but single-marker analysis identified four independent regions significantly associated with OT that we considered as tentative QTL. With two QTL, an increase in OT was associated with alleles from the refractory vector, O. hendersoni, and likely reflect epistatic interactions between genes that were uncovered by our interspecific crosses. For body size, two QTL were identified using CIM and a third tentative QTL was identified using single-marker analysis. The genome regions associated with body size also contain three QTL controlling OT, suggesting that these regions contain either single genes with pleiotropic effects or multiple linked genes independently determining each trait.

---

LA Crosse virus (LACV; family Bunyaviridae, genus Bunyavirus, species California encephalitis virus) is a leading cause of pediatric encephalitis in the United States, with 70 cases reported annually (CENTERS FOR DISEASE CONTROL AND PREVENTION 2003), and is distributed throughout the eastern United States (CALISHER 1983). Ochlerotatus triseriatus, the Eastern treehole mosquito, is an efficient vector of LACV (WATTS et al. 1972); however, the closely related O. hendersoni transmits only rarely, at frequencies of <10% (GRIMSTAD et al. 1985; PAULSON et al. 1989). Nontransmitting O. hendersoni develop a disseminated infection, and LACV readily infects and replicates in the salivary glands. This suggests that the failure to pass virus orally is due to inability of the virus to escape from the salivary glands. Concurrent infection with both LACV and the avian malaria parasite, Plasmodium gallinaceum, results in bypassing this barrier, and transmission rates are similar to those seen in O. triseriatus (PAULSON et al. 1992). Hence, the inability to transmit by O. hendersoni is likely due to a physical barrier that prevents escape of the virus; coinfection with the metazoan parasite results in disruption of this barrier and subsequent transmission of the virus.

Vectorial capacity refers to the ability of an intermediate (vector) host to transfer a pathogen between susceptible vertebrate hosts. After a mosquito takes a viremic blood meal, the virus must successfully passage several potential morphological and physiological barriers to be transmitted by the mosquito to another vertebrate host. Many factors, both intrinsic and extrinsic, can affect the success of this transmission. Significant extrinsic factors include temperature, humidity, and nutrition. At higher temperatures, dissemination is more rapid and viral titers rise more quickly with most mosquito/virus systems investigated to date (BATES and ROCA-GARCíA 1946; CHAMBERLAIN and SUDIA 1955; KAY et al. 1989; CORNEL et al. 1993; REISEN et al. 1993). However, the opposite effect is observed with some mosquito/virus systems (REISEN et al. 1993), and results in modulated or even cleared infections (KRAMER et al. 1983). Exposure to high humidity has been shown to both enhance dengue virus infections and increase daily probability of survival in Aedes aegypti (THU et al. 1998). Nutritional status of adult Anopheles stephensi (KOELLA and SORENSEN 2002) and larval O. triseriatus (GRIMSTAD and HARAMIS 1984) has been shown to significantly impact vectorial capacity.

Vector competence, as a critical component of vectorial capacity, is determined by intrinsic factors that predispose the innate immune response of a mosquito vector to a pathogen (BEERNTSEN et al. 2000). In some Anopheles/Plasmodium systems, malarial parasites are killed by melanotic encapsulation, thus preventing their transmission (COLLINS et al. 1986), as are larvae of the filarioid nematode parasite, Brugia malayi, by Armigeres subalbatus (FERDIG et al. 1993). Incompatibilities between a refractory strain of A. aegypti and P. gallinaceum result in the near-complete failure of parasite development (THATHY et al. 1994), likely due to an immunity-mediated lytic reponse that seems to also be activated in most strains of A. aegypti when exposed to filarioid nematode parasites (MACDONALD 1962). Finally, several studies have observed that vector competence for arbovirus transmission varies among mosquito populations, including LACV by O. triseriatus (GRIMSTAD et al. 1977).

Quantitative trait loci (QTL) that control vector competence have been mapped with several mosquito/pathogen systems. ZHENG et al. (1997) mapped QTL that control melanotic encapsulation of several Plasmodium spp. by Anopheles gambiae. With A. aegypti, QTL have been mapped that influence both P. gallinaceum (SEVERSON et al. 1995; MEECE 2002) and B. malayi development (SEVERSON et al. 1994; BEERNTSEN et al. 1995), as well as for midgut infection and midgut escape barriers for dengue viruses (BOSIO et al. 2000). GRAHAM et al. (2003) mapped QTL determining transovarial transmission of LACV by O. triseriatus. However, the genetic basis for oral transmission of LACV or other arboviruses has not been investigated. In this study, we used progeny from interspecific crosses between O. hendersoni and O. triseriatus to map QTL for oral transmission (OT) of LACV. In addition, because recent studies have shown that body size in A. aegypti may be correlated with vector competence for P. gallinaceum (MEECE 2002) and dengue virus (J. R. SCHNEIDER and D. W. SEVERSON, unpublished data), we examined whether genome regions containing QTL for oral transmission of LACV might also correspond to those containing QTL for body size.

Mosquito strains and rearing:

O. triseriatus WALTON strain was collected as eggs in 1969 from northern Indiana and has been maintained in the laboratory for over 100 generations. O. hendersoni PEORIA strain was established from eggs collected in Peoria, Illinois. It was initiated and maintained by forced copulation (MCDANIEL and HORSFALL 1957), and the F₃ generation was used for these studies. Mosquito colonies were reared in an environmental chamber at 21° and 80% relative humidity as previously described (MUNSTERMANN and WASMUTH 1985). Adults were provided honey-soaked cotton as a sugar source. Female mosquitoes were blood fed on anesthetized mice. Interspecific crosses were established by force copulating O. triseriatus WALTON males to O. hendersoni PEORIA females 1–2 days after the females had fed on anesthetized mice, as this crossing scheme was previously shown to produce viable, fertile offspring (TAYLOR 1987). The F₁ progeny also were force copulated to establish an F₂ generation.

Infection of mosquitoes with LACV:

Virus was obtained by inoculating a confluent layer of African green monkey kidney (Vero) cells (American Type Cell Culture, Manassas, VA) with 1 ml virus suspension from the previous passage and 1 ml M199 medium in 75-cm² tissue culture flasks at 37° and 5% CO₂ as previously described (GRIMSTAD and HARAMIS 1983, 1984). Adult female mosquitoes were orally infected with LACV (strain M18304A; 13 suckling mouse brain and 3 Vero cell passages) diluted to a titer of 4.97 log TCID₅₀ in defibrinated rabbit blood (Colorado Serum Company, Denver) via an artificial membrane feeder warmed to 37° by a circulating water system following procedures of GRIMSTAD et al. (1977). All females were allowed to feed on the same infectious blood meal and only those that fully engorged were retained and provided honey as a sugar source until tested for OT.

Determination of oral transmission:

Female mice [Hsd: ICR (CD-1)] with newborn litters were purchased from Harlan (Indianapolis) and maintained as family units in separate cages. Female mosquitoes were transferred individually to small cages (50-ml centrifuge tubes with fine mesh tops) and 2- to 3-day-old suckling mice were then gently hand held against the mesh until the mosquito had (1) fed to repletion, (2) probed three or more times, or (3) probed at least once, but showed no further interest. Because a single probing event is sufficient to transmit virus (GUBLER and ROSEN 1976; GRIMSTAD et al. 1977), even those females that probed only once but did not take a full blood meal were frozen at –80° until further assayed. In addition, a single wing from each female was removed and fixed to a glass slide to determine whether there was a correlation between body size and transmission. Wing lengths, as a proxy for body size, were measured from the apical notch to the axillary margin using an ocular micrometer. Because these species have previously been shown to differ significantly in OT (PAULSON et al. 1989) and wing length (O. triseriatus is larger than O. hendersoni; ZAVORTINK 1972), we did not obtain phenotype measurements from the parent strains. Also, due to the low success rate in interspecific matings and resulting small sample sizes (only a limited number of forced mating attempts are successful), phenotypes were not determined for F₁ females. Approximately one-half of the F₂ females were tested for OT 26 days postinfection (p.i.), while the remainder were tested on day 33 p.i. due to limited availability of suckling mice from the supplier. A single toe was clipped from each mouse as an identifier (DUBIN 1968). One mouse from each litter served as an uninfected control. Individual suckling mice fed on by individual mosquitoes were returned to cages with their mothers and observed for 14 days for signs of central nervous system disturbance; pups displaying signs were killed by rapid freezing at –80°.

Oral transmission was confirmed by plaque assay of aspirated brain material from individual frozen suckling mice in six-well tissue culture plates (Costar, Acton, MA) as described by GRIMSTAD et al. (1977). Following an initial 1-hr incubation of diluted brain tissue on Vero cells, the tissue homogenates were aspirated from the cell monolayer, and 2 ml M199, with 0.8% gum tragacanth, 25 µg/100 ml amphotericin B, and 15 µg/100 ml gentamycin, was added to each well. Plates were incubated for 3 days at 37° and 5% CO₂. Plaques were visualized by fixing and staining the plates for at least 1 hr in 0.1% crystal violet/10% formalin, rinsing in tap water, and air drying.

Midguts of mosquitoes were dissected and tested for the presence of LACV by plaque assay as described above. In addition, head squashes, followed by indirect immunofluorescence antibody tests (IFAs) were used to determine whether virus had disseminated (GUBLER and ROSEN 1976; GRIMSTAD and HARAMIS 1983). An uninfected head served as a negative control on each slide.

DNA extraction, PCR amplification, and genotyping:

Genomic DNA was extracted from individual mosquitoes as previously described (SEVERSON 1997). PCR-based genetic marker loci were developed from two sources. First, comparative anchor-tagged sequences (CATS) identified from A. aegypti cDNA sequences were tested for PCR amplification with five individuals each from O. triseriatus and O. hendersoni as described by CHAMBERS et al. (2003). In addition, several new CATS were developed for this study, again using A. aegypti cDNA sequences selected on the basis of their genome location in A. aegypti (SEVERSON et al. 2002). Second, primer sequences for several markers derived from O. triseriatus cDNAs were generously provided by William C. Black of Colorado State University prior to publication (GRAHAM et al. 2003) or were derived from O. triseriatus sequences taken from GenBank. Primers were designed using the PRIMER program (Whitehead Institute, Cambridge, MA). Details on markers used in constructing an interspecific linkage map are given in Table 1.

View this table: In this window In a new window   TABLE 1 Genetic markers used in O. hendersoni and O. triseriatus interspecific mapping

 
Individual markers were scored as species-specific single nucleotide polymorphism (SNP) loci. To identify informative SNPs, the target sequence was amplified from up to five individuals per species and subjected to direct cycle sequencing using the ABI Prism Big Dye Terminator kit (Applied Biosystems, Foster City, CA). Sequences were aligned using the default settings of ClustalW on the BCM Searchlauncher website (http://searchlauncher.bcm.tmc.edu) and species-specific SNPs were identified visually. Whenever possible, SNP analysis followed the McSNP protocol (AKEY et al. 2001). Species-specific SNPs located within restriction enzyme recognition sites were identified using the NEBcutter program (http://tools.neb.com/NEBcutter/index.php3). McSNP products from individual segregants were visualized and scored using a gel-free system (Hybaid DASH system; Thermo Hybaid, Franklin, MA). With one marker (AtL11), a diagnostic indel polymorphism was noted following sequence analysis, which allowed for visualization directly in 3% agarose gels. For markers where no restriction site was associated with the diagnostic SNP, the PCR products were visualized as single-strand conformation polymorphisms (SSCPs) on 5% polyacrylamide gels following standard protocols (FULTON et al. 2001; MEECE 2002).

Linkage and statistical analyses:

A composite linkage map was constructed from four parental crosses; females from three of these crosses were examined in this study. Linkage distances were estimated using the MAPMAKER computer program (LANDER et al. 1987) with a LOD of 3.0 as the threshold for significance and were compared to an existing linkage map for O. triseriatus (ANDERSON et al. 2001). Map distances are reported in Kosambi centimorgans (KOSAMBI 1944).

Identification of QTL controlling OT was performed using QTL-by-SAS (XU and XU 2003); this program identifies QTL associated with binary traits by assuming an underlying continuous trait that results in the phenotype only when a threshold is exceeded. Significance thresholds for declaring a QTL were estimated using the method of PIEPHO (2001). QTL locations were determined using the interval mapping (IM) for categorical traits module of QTL-by-SAS with a 1-cM step size. Linkage group (LG) III was evaluated as two linkage groups to remove any confounding effect of the large interval between markers ATE001 and LF334 (see Figure 1).

View larger version (16K): In this window In a new window Download PPT slide   FIGURE 1.— Linkage map for 502 progeny of four O. hendersoni female by O. triseriatus male crosses. Genetic markers are given to the left of the chromosome; KOSAMBI (1944) centimorgan distances are to the right. Markers identified with an asterisk were used to map oral transmission and body size in 164 female progeny.

 
QTL controlling body size were identified using the QTL Cartographer computer package (BASTEN et al. 2001). Markers with significant partial regression coefficients were identified with the SRmapqtl function by forward-backward stepwise regression. Quantitative trait loci affecting each trait were identified by composite interval mapping (CIM; ZENG 1993, 1994) using the Zmapqtl function with model 6. An experimentwise likelihood ratio (LR) threshold for identifying a QTL was determined by permutation test (CHURCHILL and DOERGE 1994). The data were permuted 1000 times, and a critical LR value was determined for the = 0.05 significance level. LGIII was again evaluated as two linkage groups.

In addition to interval mapping methods for QTL analysis, we also used simple, single-marker tests to assess relationships between phenotype and marker genotype. These procedures are rigorous for identifying significant marker associations, but are inadequate for accurately defining QTL locations (DOERGE 2002). We included these analyses as (1) interval mapping has low power across large (>30 cM) intervals, (2) the experimentwise thresholds for interval mapping are conservative and likely to exclude valid QTL (CHURCHILL and DOERGE 1994), and (3) we were unable to determine linkage associations for one marker (PABP) and thus could not include it for interval mapping. Fisher's exact test using the FREQ procedure in the SAS computer package (SAS INSTITUTE 1990, 1997) was performed for OT. Significant associations with body size were identified using proc GENMOD in SAS. Markers at P < 0.05 were considered to have a significant association with the phenotype, while markers with P < 0.10 were considered as suggestive of an association. Adjustments for multiple tests (e.g., Bonferroni corrections) were not performed for two reasons. First, linkage data, and, thus, QTL analyses, are likely not appropriate for these adjustments as markers on individual chromosomes are not independent, and multiple loci affecting a phenotype may not be independent even across chromosomes. Second, because of the relatively small number of unique marker tests involved in our study (n = 20), at the 5% significance level we expect only a single false positive.

Quantitative genetics of OT:

A total of 164 female F₁ intercross progeny were obtained from three interspecific crosses between an O. hendersoni female and an O. triseriatus male that provided a sufficiently large number of progeny for genetic analyses; these families consisted of 32, 33, and 99 individual females, respectively. The positions of the 24 genetic markers and the sex locus mapped in 502 progeny from four pairwise crosses (this includes both males and females) are shown in Figure 1. We were unable to determine linkage associations for one marker (PABP); this was not surprising, as other markers from the homologous genome region in A. aegypti also remain unlinked in O. triseriatus (ANDERSON et al. 2001). As observed previously (ANDERSON et al. 2001), segregation ratios deviated from the expected Mendelian ratio of 1:2:1 for all markers except AtIAP on LGII (data not shown). Markers identified with an asterisk in Figure 1 were used in this study.

Following OT trials, midguts were dissected to confirm LACV presence by plaque assay, while the heads were removed and tested for the presence of LACV antigen by IFA. Virus was present in 100% of the midguts, indicating that each female had taken an infective blood meal. Head squashes were inconclusive; there appeared to be very little specific fluorescence when compared to uninfected controls. Because OT was documented (see below), and because others have previously shown both O. hendersoni and O. triseriatus to readily disseminate virus upon infection (GRIMSTAD et al. 1985; PAULSON et al. 1989, 1992), we assumed that all females tested for OT had disseminated infections.

Due to difficulties obtaining sufficient suckling mice from the supplier, 66 female mosquitoes (approximately one-half from the two smallest families and the remainder from the largest) were tested 26 days p.i., and the remaining 98 were tested 33 days p.i. Transmission rates between families and between feeding dates were compared by logistic regression. No significant differences in transmission rates were found between families (family 4, 35% transmitted; family 14, 18%; family 16, 25%; P = 0.1479). However, significant differences were found between feeding groups (FG1, 18% transmitted; FG2, 38%; P = 0.0084). Therefore, for subsequent analyses each feeding group was tested separately for QTL determining OT. No significant differences in transmission rates were seen with respect to the number of times females probed (P = 0.54 for FG1 and 0.73 for FG2) or whether females engorged (P = 0.08 for FG1 and 0.76 for FG2).

Forty-nine of 164 (29.9%) females transmitted LACV by bite to suckling mice. Presence of viral plaques in brain material was confirmed for 47 of these mice. Two pups were not tested due to cannibalism by their mothers; these were assumed to be positive because this occurred during the period when other pups were exhibiting morbidity. No evidence of infectious virus was observed in 104 of the remaining 115 suckling mice that survived the experiments. Most of the mothers produced a second litter near the end of the test, and three began cannibalizing their first litter; therefore, 11 pups were not available to be tested for the presence of LACV in brain material. They were assumed to be negative because no morbidity or mortality was observed for >1 week prior to their loss. No nursing mothers seroconverted to LACV, indicating that virus could not have been passed during suckling.

To identify QTL influencing the ability to orally transmit LACV, the data were separated by feeding group and analyzed by IM for a binary trait. The results for FG1 and FG2 are given in Figure 2. No genome locations exceeded the experimentwise 95% threshold LR for declaring a QTL, although three genome regions with elevated LRs are evident. However, the stringency of the experimentwise threshold, while reducing the likelihood of declaring false positive QTL (type I error), also increases the potential for rejecting true positive QTL (type II error; CHURCHILL and DOERGE 1994). In addition, the power of our OT analyses was negatively impacted by the need to evaluate the two feeding groups independently, as this greatly reduced our sample sizes. Therefore, with the small cell values for most markers, individual markers within feeding groups were also analyzed for an association with OT by Fisher's exact test (Table 2). We considered markers with significant associations with OT as identifying tentative QTL. These analyses identified four tentative QTL regions. Of note, the parental alleles at each significant marker that contributed to an increased likelihood of OT did not always come from the O. triseriatus parent. With LGI, one marker (LF179) showed a significant association (P < 0.05) with OT and two markers (AtL07 and AtL14) were suggestive (P < 0.1) for FG1. With FG2, only AtL14 was suggestive of an association with OT. An increased likelihood of OT with LGI markers was associated with O. triseriatus alleles. LGIII likely contains at least two QTL associated with each end of the linkage group. At one end, LF99 was significant and LF296 and AtL72 were suggestive with FG1, while LF296 and AtL72 were suggestive with FG2. At the other end, CRALBP was significant and LF334 was suggestive with FG1, while both markers were significant with FG2. For both genome regions on LGIII, an increased likelihood of OT with LGI markers was associated with O. hendersoni alleles. Finally, a significant association was observed with an unlinked marker (PABP), but only with FG2; here the increased likelihood of OT was associated with O. triseriatus alleles.

View larger version (17K): In this window In a new window Download PPT slide   FIGURE 2.— Likelihood ratio profile for identifying quantitative trait loci for oral transmission of La Crosse virus. The solid line represents feeding group 1, while the dashed line represents feeding group 2. Horizontal lines are the experimentwise threshold value for identifying a QTL. Triangles represent the relative location of genetic markers (see Figure 1); two marker names are given for orientation of each chromosome.

 

View this table: In this window In a new window   TABLE 2 Associations of marker alleles with oral transmission of La Crosse virus

 

Quantitative genetics of body size:

The distribution of wing lengths as a proxy for body size was determined for 164 F₂ female progeny (Figure 3). The mean wing length was 3.49 ± 0.22 mm with a range from 2.85 to 3.95 mm. No significant differences in wing length were observed between FG1 (mean, 3.48 mm) and FG2 (mean, 3.50). Therefore, data for both groups were pooled for QTL analyses. As previously indicated (see MATERIALS AND METHODS), wing lengths of the parental species and their F₁ hybrids were not obtained.

View larger version (17K): In this window In a new window Download PPT slide   FIGURE 3.— Distribution of wing lengths among 164 females tested for LACV transmission. Values on the x-axis represent the midpoint of the interval.

 
LR estimates by composite interval mapping exceeded the 95% experimentwise threshold for most marker loci on LGI and one region on LGIII (Figure 4). QTL positions were best defined by the AtL14-LF159 interval on LGI and the AtL72-ATE001 interval on LGIII. Single-marker analyses by linear regression indicated that all markers on LGI and several markers representing both ends of LGIII were significant or suggestive for an association with body size, as well as one marker (AEGI8) on LGII (Table 3). The genome regions on LGI and LGIII showing significant associations with body size are, therefore, nearly identical to those identified for OT (Table 2). An association between body size and OT was further evidenced when we characterized individuals as above or below the mean body size and performed a comparison with OT using Fisher's exact test (Figure 5). Mosquitoes smaller than the mean were more likely to transmit than were those larger than the mean; observed results were highly significant with FG1 (P < 0.001) and suggestive with FG2 (P < 0.1). These results are contrary to those expected as the smaller body size phenotype is characteristic of the O. hendersoni parent.

View larger version (14K): In this window In a new window Download PPT slide   FIGURE 4.— Likelihood ratio profile for identifying quantitative trait loci for body size. The horizontal line represents the experimentwise threshold value for identifying a QTL. Triangles represent the relative location of genetic markers (see Figure 1); two marker names are given for orientation of each chromosome.

 

View this table: In this window In a new window   TABLE 3 Effect of substituting O. triseriatus alleles for O. hendersoni alleles at genetic markers with a significant association with body size

 

View larger version (14K): In this window In a new window Download PPT slide   FIGURE 5.— The effect of body size on oral transmission. Bars represent the percentage of individuals below or above the mean body size that transmitted LACV for each feeding group. n, total number of individuals that transmitted; parenthetic values are the total number tested in that size class. *P < 0.1; **P < 0.001.

 

Arbovirus transmission by mosquitoes remains a significant global threat to human health (GUBLER 2002). In this study, we examined the genetic basis for the ability to orally transmit LACV in hybrids of O. hendersoni (an inefficient vector) and O. triseriatus (the natural vector). Previous studies had shown a relationship between vector competence and vector size (GRIMSTAD and HARAMIS 1984; GRIMSTAD and WALKER 1991; PAULSON and HAWLEY 1991). These studies, however, examined only differences due to environmental effect, i.e., larval nutrition. Here, we identified genetic determinants of both ability to transmit a pathogen orally and adult female body size and then examined the relationship between the two phenotypes. Previous studies (BOSIO et al. 1998, 2000; GRAHAM et al. 2003) examined the genetic determinants for arbovirus development within the invertebrate vector, but not transmission to a vertebrate host.

We identified four genome regions containing tentative QTL determining virus transmission in the O. hendersoni/O. triseriatus/LACV system. We define QTL as tentative, on the basis of observed significant single-marker analyses because no genome regions exceeded the experimentwise 95% threshold for declaring a QTL using IM for a binary trait. Our justification is that experimentwise thresholds are conservative (CHURCHILL and DOERGE 1994) and the power of our OT analyses was reduced by the unanticipated need to evaluate our segregants as two independent feeding groups. However, results from these groups were generally consistent, as significant marker associations with LACV transmission were observed for genome regions on LGI (LF179, AtL07, AtL14) and two independent regions on LGIII (LF99, LF296, AtL72 and LF334, CRALBP). A fourth genome region, defined by a single marker showing no linkage associations (PABP) was associated with LACV transmission, but only with FG2. Of interest, the effect of these tentative QTL on OT did not mirror the parental phenotypes. That is, with two of the four QTL, an increase in OT was associated with alleles from the refractory vector, O. hendersoni. Such results are not unusual with studies of complex traits (TANKSLEY and MCCOUCH 1997) and likely reflect epistatic interactions between genes that were uncovered by our interspecific crosses. Although we did not directly determine phenotypes for the parental stocks used to prepare our segregating populations, the dissemination rate has previously been shown to be significantly higher for O. hendersoni than for O. triseriatus (GRIMSTAD et al. 1985). Therefore, one explanation for our results is that the increase in OT due to O. hendersoni alleles reflects the effects of genes determining a midgut escape barrier.

Our results indicate that at least three QTL influence body size. Composite interval mapping identified two QTL for body size defined by the AtL14-LF159 interval on LGI and the AtL72-ATE001 interval on LGIII. Single-marker analysis identified a third tentative QTL associated with markers LF334 and CRALBP on the opposite end of LGIII. For all loci except CRALBP, addition of O. triseriatus alleles resulted in larger mosquitoes than did O. hendersoni homozygotes.

The genome regions for three QTL determining OT also contain the QTL determining body size. This indicates that these regions contain either single QTL with pleiotropic effects affecting both traits or multiple QTL that independently act on each trait. Of particular note, mosquitoes smaller than the mean were more likely to transmit LACV than were mosquitoes larger than the mean. This was unexpected, as the highly competent vector for LACV transmission, O. triseriatus, is larger than the LACV refractory O. hendersoni (ZAVORTINK 1972). As previously discussed, this finding undoubtedly reflects the results of epistatic interactions between genes determining OT, wherein for two QTL the increase in OT was associated with alleles from O. hendersoni. Interestingly, previous studies had shown that smaller O. triseriatus females were more likely to transmit than were their larger counterparts, although the effect in those studies was due to different environmental conditions, namely being starved or well-fed as larvae (GRIMSTAD and HARAMIS 1984; GRIMSTAD and WALKER 1991).

The interspecific linkage map produced from O. hendersoni female by O. triseriatus male crosses spans 382.5 cM. This is 111.8 cM larger than the intraspecific linkage map generated for O. triseriatus (ANDERSON et al. 2001). Much of the difference in size is found on LGII, which is 2.5 times larger in the interspecific crosses. In addition, LGI is 50% larger than the intraspecific linkage map; however, the previously described linear orders of individual marker loci (ANDERSON et al. 2001) appear to be conserved in the interspecific map. Linkage group III is essentially the same size in both maps, and large recombinational distances across the likely centromere region (e.g., ATE001 and LF334 in Figure 1) are common to the two maps. Some of the observed increase in LGII size may be attributed to the use of genetic markers lying more distally on the chromosome (e.g., AEGI8) than those in the previously reported map. Lastly, marker PABP remained unlinked in this study. This marker is closely linked in A. aegypti to another marker (LF377) that was unlinked in the intraspecific map (ANDERSON et al. 2001) and may represent a distal chromosomal arm separated by sufficient recombination as to appear unlinked in both this and our previous study.

Significant deviations from the expected Mendelian ratios were evident for most of the markers. Such distortion may represent the presence of segregation distortion loci (VOGL and XU 2000) or lethal loci (e.g., SEVERSON et al. 1995) and may have an impact on identification and localization of QTL. However, distortion of segregation ratios may result in misplacement of the QTL, but not its presence (SCHäFER-PREGL et al. 1996; LU et al. 2002). Indeed, segregation distortion in mosquito crosses has been noted in identifying QTL for Plasmodium susceptibility (SEVERSON et al. 1995) and for DDT resistance (RANSON et al. 2000).

The relative inability of O. hendersoni to transmit LACV is not due to an incompatibility, since the virus readily replicates in all tissues, including the salivary glands (GRIMSTAD et al. 1985; PAULSON et al. 1989, 1992). Thus, there must be some barrier to its escape from the salivary glands, although whether this is physical or physiological is unknown. Nutritional deprivation of O. triseriatus larvae leads to a reduction in the thickness of the midgut basal lamina; virus subsequently passes through this potential barrier more easily (GRIMSTAD and WALKER 1991). This precedent for a physical barrier to virus transmission in this system suggests that a similar barrier may be present in the salivary glands of O. hendersoni. Three possible mechanisms to the inhibition of viral transmission have been suggested (HARDY et al. 1983): (1) sufficient time was not allowed for development of a salivary gland infection before transmission trials were attempted, (2) virus was shed from the salivary glands during a previous sugar meal(s) and had not reaccumulated prior to transmission studies, and (3) a physical barrier similar to the midgut escape barrier existed. This study and that of PAULSON and GRIMSTAD (1989) eliminated the first two as possibilities. Transmission trials were performed after a sufficient period of extrinsic incubation (here 26 or more days) such that salivary glands should have been fully infected and infective. Additionally, the sugar source was removed >24 hr before transmission was attempted, and the salivary glands should have been "restocked" with sufficient virus to infect suckling mice.

Two hypotheses for the inability to transmit virus in the saliva have been put forth following electron microscopic studies of virus-infected salivary glands. JANZEN et al. (1970) have shown that chikungunya virus replicates in large numbers in the proximal portions of the lateral salivary gland lobes of A. aegypti. However, in these areas, the virus may be too large to enter into the relatively narrow salivary duct openings. Additionally, filamentous areas of the duct could further act as a physical barrier by blocking passage through the channels into the ducts (JANZEN et al. 1970). Second, mature virions may accumulate in large masses in areas where there is effectively no means of entering the lumen. Such is the case in O. triseriatus salivary glands infected with EEEV (WHITFIELD et al. 1971), although large numbers of enveloped viruses were also present in the lumen.

Alternately, there may be physiological differences in the saliva of the two species. Previous studies have shown that mosquito saliva potentiates infections with some viruses, perhaps by inhibiting antiviral innate immune responses (EDWARDS et al. 1998; LIMESAND et al. 2000, 2003). It is possible that salivary proteins in O. hendersoni do not inhibit immune responses to the same level as those of O. triseriatus. However, this seems unlikely, as saliva from a diverse group of mosquitoes (O. triseriatus, A. aegypti, and Culex pipiens) were shown to enhance Cache Valley virus infection in mice (EDWARDS et al. 1998).

It is currently unknown whether any one of these hypotheses or a different explanation is responsible for the difference in oral transmission seen between O. hendersoni and O. triseriatus. A single study has examined the salivary glands of O. triseriatus, but in relation to infection with EEEV (WHITFIELD et al. 1971). Thus there is no literature describing the comparative morphologies of the salivary glands in the two species examined here, nor is there information concerning infection with and replication of LACV on an electron microscopic level in these species. Similarly, no studies have examined the effects of the saliva of these two species on LACV infection, although viremias have been shown to be higher in animals infected by O. triseriatus feeding than in those inoculated by syringe (PANTUWATANA et al. 1972).

In summary, we have identified four QTL conferring the ability to orally transmit LACV in interspecific hybrids of O. hendersoni and O. triseriatus. The genome regions associated with three of these QTL also contain QTL controlling the development of body size, suggesting that these regions contain either single genes with pleiotropic effects or multiple linked genes independently determining each trait. This study, together with those of GRAHAM et al. (1999)(2003), begins to address the impact that the genetic architecture of O. triseriatus and O. hendersoni has on the continued transmission of LACV in the United States. Further work is needed to identify and characterize the particular genes responsible for OT in the vector O. triseriatus and the nonvector O. hendersoni. Identification of such genes could ultimately lead to the identification of novel methods for interruption of the LACV transmission cycle and, potentially, those of other arboviruses.

We thank Diane Lovin and Becky deBruyn for technical support. Jack Swanson of the Illinois Department of Public Health provided O. hendersoni eggs. Chenwu Xu assisted in using QTL-by-SAS. Funding was provided by National Institutes of Health National Institute of Allergy and Infectious Disease grants AI02753 to P.R.G. and AI33127 to D.W.S.

Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession nos. BV096874, BV096875, BV096876, BV096877, BV096878, BV096879, BV096880, BV096881, BV096882, BV096883, BV096884, BV096885, BV096886, BV096887, BV096888, BV096889, BV096890, BV096891, BV096892, BV096893, BV096894, BV096895, BV096896, BV096897, BV096898, BV096899, BV096900, BV096901, BV096902, BV096903, BV096904, BV096905, BV096906, BV096907, BV096908, BV096909, BV096910, BV096911, BV096912, BV096913, BV096914, BV096915, BV096916, BV096917, BV096918, BV096919, BV096920, BV096921.

AKEY, J. M., D. SOSNOSKI, I. PARRA, S. DIOS, K. HIESTER et al., 2001 A melting curve analysis of SNPs (McSNP): a gel-free and inexpensive approach for SNP genotyping. Biotechniques 30: 358–367.[Medline]

ANDERSON, J. R., P. R. GRIMSTAD and D. W. SEVERSON, 2001 Chromosomal evolution among six mosquito species (Diptera: Culicidae) based on shared restriction fragment length polymorphisms. Mol. Phylogenet. Evol. 20: 316–321.[CrossRef][Medline]

BASTEN, C. J., B. S. WEIR and Z-B. ZENG, 2001 QTL Cartographer: A Reference Manual and Tutorial for QTL Mapping. Program in Statistical Genetics, Bioinformatics Research Center, Department of Statistics, North Carolina State University, Raleigh, NC.

BATES, M., and M. ROCA-GARCíA, 1946 The development of the virus of yellow fever in Haemagogus mosquitoes. Am. J. Trop. Med. 26: 585–605.

BEERNTSEN, B. T., D. W. SEVERSON, J. A. KLINKHAMMER, V. A. KASSNER and B. M. CHRISTENSEN, 1995 Aedes aegypti: a quantitative locus (QTL) influencing filarial worm intensity is linked to QTL for susceptibility to other mosquito-borne pathogens. Exper. Parasitol. 81: 355–362.

BEERNTSEN, B. T., A. A. JAMES and B. M. CHRISTENSEN, 2000 Genetics of mosquito vector competence. Microbiol. Mol. Biol. Rev. 64: 115–137.[Abstract/Free Full Text]

BOSIO, C. F., B. J. BEATY and W. C. BLACK, IV, 1998 Quantitative genetics of vector competence for dengue-2 virus in Aedes aegypti. Am. J. Trop. Med. Hyg. 59: 965–970.[Abstract]

BOSIO, C. F., R. E. FULTON, M. L. SALASEK, B. J. BEATY and W. C. BLACK, IV, 2000 Quantitative trait loci that control vector competence for dengue-2 virus in the mosquito Aedes aegypti. Genetics 156: 687–698.[Abstract/Free Full Text]

CALISHER, C. H., 1983 Taxonomy, classification, and geographic distribution of California serogroup Bunyaviruses, pp. 1–16 in California Serogroup Viruses, edited by C. H. CALISHER and W. H. THOMPSON. Alan R. Liss, New York.

CENTERS FOR DISEASE CONTROL AND PREVENTION, 2003 http://www.cdc.gov

CHAMBERLAIN, R. W., and W. D. SUDIA, 1955 The effects of temperature upon the extrinsic incubation of eastern equine encephalitis in mosquitoes. Am. J. Hyg. 62: 295–300.[Medline]

CHAMBERS, E. W., D. D. LOVIN and D. W. SEVERSON, 2003 Utility of comparative anchor-tagged sequences as physical anchors for comparative genome analysis among the Culicidae. Am. J. Trop. Med. Hyg. 69: 98–104.[Abstract/Free Full Text]

CHURCHILL, G. A., and R. W. DOERGE, 1994 Empirical threshold values for quantitative trait mapping. Genetics 138: 963–971.[Abstract]

COLLINS, F. H., R. K. SAKAI, K. D. VERNICK, S. PASKEWITZ, D. C. SEELEY et al., 1986 Genetic selection of a Plasmodium-refractory strain of the malaria vector Anopheles gambiae. Science 234: 607–610.[Abstract/Free Full Text]

CORNEL, A. J., P. G. JUPP and N. K. BLACKBURN, 1993 Environmental temperature on the vector competence of Culex univittatus (Diptera: Culicidae) for West Nile virus. J. Med. Entomol. 30: 449–456.[Medline]

DOERGE, R. W., 2002 Mapping and analysis of quantitative trait loci in experimental populations. Nat. Rev. Genet. 3: 43–52.[CrossRef][Medline]

DUBIN, S., 1968 A method for numbering laboratory animals using the binary system. Lab. Anim. Care 18: 574–577.[Medline]

EDWARDS, J. F., S. HIGGS and B. J. BEATY, 1998 Mosquito feeding-induced enhancement of Cache Valley virus (Bunyaviridae) infection in mice. J. Med. Entomol. 35: 261–265.[Medline]

FERDIG, M. T., B. T. BEERNTSEN, F. J. SPRAY, J. LI and B. M. CHRISTENSEN, 1993 Reproductive costs associated with resistance in a mosquito-filarial worm system. Am. J. Trop. Med. Hyg. 49: 756–762.

FULTON, R. E., M. L. SALASEK, N. M. DUTEAU and W. C. BLACK, IV, 2001 SSCP analysis of cDNA markers provides a dense linkage map of the Aedes aegypti genome. Genetics 158: 715–726.[Abstract/Free Full Text]

GRAHAM, D. H., J. L. HOLMES, S. HIGGS, B. J. BEATY and W. C. BLACK, IV, 1999 Selection of refractory and permissive strains of Aedes triseriatus (Diptera: Culicidae) for transovarial transmission of La Crosse virus. J. Med. Entomol. 36: 671–678.[Medline]

GRAHAM, D. H., J. L. HOLMES, B. J. BEATY and W. C. BLACK, IV, 2003 Linkage mapping of quantitative trait loci conditioning transovarial transmission of La Crosse virus in the eastern treehole mosquito, Ochlerotatus triseriatus. Insect Mol. Biol. 12: 307–318.[CrossRef][Medline]

GRIMSTAD, P. R., and L. D. HARAMIS, 1983 Evaluation of immunofluorescence and cell culture techniques for detecting La Crosse virus infection in individual Aedes triseriatus (Diptera: Culicidae). J. Med. Entomol. 20: 458–460.[Medline]

GRIMSTAD, P. R., and L. D. HARAMIS, 1984 Aedes triseriatus (Diptera: Culicidae) and La Crosse virus III. Enhanced oral transmission by nutrient deprived mosquitoes. J. Med. Entomol. 21: 249–265.[Medline]

GRIMSTAD, P. R., and E. D. WALKER, 1991 Aedes triseriatus (Diptera: Culicidae) and La Crosse virus IV. Nutritional deprivation of larvae affects the adult barriers to infection and transmission. J. Med. Entomol. 28: 378–386.[Medline]

GRIMSTAD, P. R., G. B. CRAIG, JR., Q. E. ROSS and T. M. YUILL, 1977 Aedes triseriatus and La Crosse virus: geographic variation in vector susceptibility and ability to transmit. Am. J. Trop. Med. Hyg. 26: 990–996.

GRIMSTAD, P. R., S. L. PAULSON and G. B. CRAIG, JR., 1985 Vector competence of Aedes hendersoni (Diptera: Culicidae) for La Crosse virus and evidence of a salivary-gland escape barrier. J. Med. Entomol. 22: 447–453.[Medline]

GUBLER, D. J., 2002 The global emergence/resurgence of arboviral diseases as public health problems. Arch. Med. Res. 33: 330–342.[CrossRef][Medline]

GUBLER, D. J., and L. ROSEN, 1976 A simple technique for demonstrating transmission of dengue virus by mosquitoes without the use of vertebrate hosts. Am. J. Trop. Med. Hyg. 25: 146–150.

HARDY, J. L., E. J. HOUK, L. D. KRAMER and W. C. REEVES, 1983 Intrinsic factors affecting vector competence of mosquitoes for arboviruses. Annu. Rev. Entomol. 28: 229–262.[CrossRef][Medline]

JANZEN, H. G., A. J. RHODES and F. W. DOANE, 1970 Chikungunya virus in salivary glands of Aedes aegypti (L.): an electron microscope study. Can. J. Microbiol. 16: 581–586.[Medline]

KAY, B. H., I. D. FANNING and P. MOTTRAM, 1989 Rearing temperature influences flavivirus vector competence of mosquitoes. Med. Vet. Entomol. 3: 415–422.[Medline]

KOELLA, J. C., and F. L. SORENSEN, 2002 Effect of adult nutrition on the melanization immune response of the malaria vector Anopheles stephensi. Med. Vet. Entomol. 16: 316–320.[CrossRef][Medline]

KOSAMBI, D. D., 1944 The estimation of map distance from recombination values. Ann. Eugen. 12: 172–175.

KRAMER, L. D., J. L. HARDY and S. B. PRESSER, 1983 Effect of temperature of extrinsic incubation on the vector competence of Culex tarsalis for western equine encephalomyelitis virus. Am. J. Trop. Med. Hyg. 32: 1130–1139.

LANDER, E. S., P. GREEN, J. ABRAHAMSON, A. BARLOW, M. DALY et al., 1987 MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1: 174–181.[CrossRef][Medline]

LIMESAND, K. H., S. HIGGS, L. D. PEARSON and B. J. BEATY, 2000 Potentiation of vesicular stomatitis New Jersey virus infection in mice by mosquito saliva. Parasite Immunol. 22: 461–467.[CrossRef][Medline]

LIMESAND, K. H., S. HIGGS, L. D. PEARSON and B. J. BEATY, 2003 Effect of mosquito salivary gland treatment on vesicular stomatitis New Jersey virus replication and interferon /ß expression in vitro. J. Med. Entomol. 40: 199–205.[Medline]

LU, H., J. ROMERO-SEVERSON and R. BERNARDO, 2002 Chromosomal regions associated with segregation distortion in maize. Theoret. Appl. Genet. 105: 622–628.

MACDONALD, W. W., 1962 The genetic basis of susceptibility to infection with semi-periodic Brugia malayi in Aedes aegypti. Ann. Trop. Med. Parasitol. 56: 373–382.

MCDANIEL, I. N., and W. R. HORSFALL, 1957 Induced copulation of aedine mosquitoes. Science 125: 745.[Free Full Text]

MEECE, J. K., 2002 Quantitative genetic analyses of Plasmodium susceptibility in Aedes aegypti and preliminary genetic investigations of Plasmodium susceptibility in Anopheles gambiae s.l. Ph.D. Dissertation, University of Notre Dame, Notre Dame, IN.

MUNSTERMANN, L. E., and L. M. WASMUTH, 1985 Aedes triseriatus, pp. 15–24 in Handbook of Insect Rearing, Vol. 2, edited by P. SINGH and R. F. MOORE. Elsevier Science, Amsterdam.

PANTUWATANA, S., W. H. THOMPSON, D. M. WATTS and R. P. HANSON, 1972 Experimental infection of chipmunks and squirrels with La Crosse and Trivittatus viruses and biological transmission of La Crosse virus by Aedes triseriatus. Am. J. Trop. Med. Hyg. 21: 476–481.

PAULSON, S. L., and P. R. GRIMSTAD, 1989 Replication and dissemination of La Crosse virus in the competent vector Aedes triseriatus and the incompetent vector Aedes hendersoni and evidence for transovarial transmission by Aedes hendersoni (Diptera: Culicidae). J. Med. Entomol. 26: 602–609.[Medline]

PAULSON, S. L., and W. A. HAWLEY, 1991 Effect of body size on the vector competence of field and laboratory populations of Aedes triseriatus for La Crosse virus. J. Amer. Mosq. Control Assoc. 7: 170–175.

PAULSON, S. L., P. R. GRIMSTAD and G. B. CRAIG, JR., 1989 Midgut and salivary gland barriers to La Crosse virus dissemination in mosquitoes of the Aedes triseriatus group. Med. Vet. Entomol. 3: 113–123.[Medline]

PAULSON, S. L., S. J. POIRIER, P. R. GRIMSTAD and G. B. CRAIG, JR., 1992 Vector competence of Aedes hendersoni (Diptera: Culicidae) for La Crosse virus: lack of impaired function in virus-infected salivary glands and enhanced virus transmission by sporozoite-infected mosquitoes. J. Med. Entomol. 29: 483–488.[Medline]

PIEPHO, H. P., 2001 A quick method for computing approximate thresholds for quantitative trait loci detection. Genetics 157: 425–432.[Abstract/Free Full Text]

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.[CrossRef][Medline]

REISEN, W. K., R. P. MEYER, S. B. PRESSER and J. L. HARDY, 1993 Effect of temperature on the transmission of western equine encephalomyelitis and St. Louis encephalitis viruses by Culex tarsalis. J. Med. Entomol. 30: 151–160.[Medline]

SAS INSTITUTE, 1990 SAS/STAT User's Guide, Version 6, Ed. 4. SAS Institute, Cary, NC.

SAS INSTITUTE, 1997 SAS/STAT Software: Changes and Enhancements Through Release 6.12. SAS Institute, Cary, NC.

SCHäFER-PREGL, R., F. SALAMINI and C. GEBHARDT, 1996 Models for mapping quantitative trait loci (QTL) in progeny of non-inbred parents and their behaviour in presence of distorted segregation ratios. Genet. Res. 67: 43–54.

SEVERSON, D. W., 1997 RFLP analysis of insect genomes, pp. 309–320 in The Molecular Biology of Insect Disease Vectors: A Methods Manual, edited by J. M. CRAMPTON, C. B. BEARD and C. LOUIS. Chapman & Hall, London.

SEVERSON, D. W., A. MORI, Y. ZHANG and B. M. CHRISTENSEN, 1994 Chromosomal mapping of two loci affecting filarial worm susceptibility in Aedes aegypti. Insect Mol. Biol. 3: 67–72.[Medline]

SEVERSON, D. W., V. THATHY, A. MORI, Y. ZHANG and B. M. CHRISTENSEN, 1995 Restriction fragment length polymorphism mapping of quantitative trait loci for malaria parasite susceptibility in the mosquito Aedes aegypti. Genetics 139: 1711–1717.[Abstract]

SEVERSON, D. W., J. K. MEECE, D. D. LOVIN, G. SAHA and I. MORLAIS, 2002 Linkage map organization of expressed sequence tags and sequence tagged sites in the mosquito, Aedes aegypti. Insect Mol. Biol. 11: 371–378.[CrossRef][Medline]

TANKSLEY, S. D., and S. R. MCCOUCH, 1997 Seed banks and molecular maps: unlocking genetic potential from the wild. Science 277: 1063–1066.[Abstract/Free Full Text]

TAYLOR, D. B., 1987 Genetic compatibility of Aedes (Protomacleaya) triseriatus with A. (P.) brelandi and A. (P.) hendersoni (Diptera: Culicidae). Ann. Entomol. Soc. Am. 80: 109–117.

THATHY, V., D. W. SEVERSON and B. M. CHRISTENSEN, 1994 Reinterpretation of the genetics of susceptibility of Aedes aegypti to Plasmodium gallinaceum. J. Parasitol. 80: 705–712.[CrossRef][Medline]

THU, H. M., K. M. AYE and S. THEIN, 1998 The effect of temperature and humidity on dengue virus propagation in Aedes aegypti mosquitos. Southeast Asian J. Trop. Med. Public Health 29: 280–284.[Medline]

VOGL, C., and S. XU, 2000 Multipoint mapping of viability and segregation distorting loci using molecular markers. Genetics 155: 1439–1447.[Abstract/Free Full Text]

WATTS, D. M., C. D. MORRIS, R. E. WRIGHT, G. R. DEFOLIART and R. P. HANSON, 1972 Transmission of La Crosse virus (California encephalitis group) by the mosquito Aedes triseriatus. J. Med. Entomol. 9: 125–127.[Medline]

WHITFIELD, S. G., F. A. MURPHY and W. D. SUDIA, 1971 Eastern equine encephalomyelitis virus: and electron microscopic study of Aedes triseriatus (Say) salivary gland infection. Virology 43: 110–122.[CrossRef][Medline]

XU, C., and S. XU, 2003 A SAS/IML program for mapping QTL in line crosses. Paper 235:1–6, SUGI 28 Proceedings, Seattle.

ZAVORTINK, T. J., 1972 Mosquito studies (Diptera, Culicidae) XXVIII. The New World species formerly placed in Aedes (Finlaya). Contrib. Am. Entomol. Inst. 8: 1–206.

ZENG, Z-B., 1993 Theoretical basis for separation of multiple linked gene effects in mapping quantitative trait loci. Genetics 90: 10972–10976.

ZENG, Z-B., 1994 Precision mapping of quantitative trait loci. Genetics 136: 1457–1468.[Abstract]

ZHENG, L., A. J. CORNEL, R. WANG, H. ERFLE, H. VOSS et al., 1997 Quantitative trait loci for refractoriness of Anopheles gambiae to Plasmodium cynomolgi B. Science 276: 425–428.[Abstract/Free Full Text]

Communicating editor: R. W. DOERGE

International Access Link

Online ISS 1091-6490
Copyright 2008 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu