|Home | About | Journals | Submit | Contact Us | Français|
The mating efficiencies of field-collected and laboratory-colonized Aedes triseriatus (Say) (Diptera: Culicidae) female mosquitoes transovarially-infected or uninfected with La Crosse virus (LACV) were compared. The females were placed in cages with age-matched males, and the insemination rates were determined daily by detection of sperm in the spermathecae. LACV-infected mosquitoes typically mated more quickly than uninfected mosquitoes. LACV load was not correlated with increased insemination.
La Crosse virus (LACV) is maintained in a cycle between Aedes triseriatus mosquitoes (Eastern tree hole mosquito) and chipmunks and tree squirrels (Calisher 1994, Watts et al. 1972). The virus can also be transmitted between mosquitoes horizontally (venereally) and vertically (transovarially), and LACV overwinters in the eggs of infected mosquitoes (Thompson and Beaty 1978, Watts et al. 1973, Watts et al. 1974). Transovarial transmission (TOT) is an extraordinary method of virus amplification since an infected female can produce many infected progeny in a lifetime. The newly hatched, infected females are capable of transmitting LACV upon emergence. Infection by TOT could have a significant effect on viral amplification and maintenance because any behavioral or physiological changes due to LACV infection will be evident immediately upon mosquito emergence. Mosquito behavior modification resulting from LACV infection could influence viral transmission and thereby increase the virus prevalence.
There is evidence of behavioral changes in mosquitoes due to arboviral infection. For example, dengue virus-infected Ae. aegypti mosquitoes require significantly longer times for probing and feeding than uninfected ones, which may increase the efficiency of transmission of dengue virus by Ae. aegypti (Platt et al. 1997). Longer probing and feeding periods are more likely to be interrupted by the host, which increases the chance that an infected mosquito will probe or feed on additional hosts (Platt et al. 1997).
LACV infection of Ae. triseriatus mosquitoes also produces behavioral changes that could promote virus amplification and maintenance in nature. For example, LACV-infected Ae. triseriatus females probe more and engorge less than uninfected mosquitoes (Grimstad et al. 1980). Increased probing and partial engorgement could result in increased viral transmission and amplification. Additionally, an increased insemination rate was observed in female Ae. triseriatus mosquitoes that were orally infected with LACV (Gabitzsch et al. 2006). Mosquitoes were given an infectious blood meal followed by a non-infectious blood meal 14 d later and allowed to mate. Of the mosquitoes ingesting the second blood meal, 85% of LACV-infected mosquitoes were inseminated compared to 51% of uninfected mosquitoes. Similar results were observed for the mosquitoes that did not ingest the second blood meal; 75% of the LACV-infected mosquitoes were inseminated compared to 38% of the uninfected mosquitoes.
Virus titer and tissue tropisms could condition the increased rate of insemination. Viral titer in a mosquito can affect transmission and possibly behavior (Watts et al. 1972). Virus infection of certain tissues or organs, may for example, influence pheromone expression by female mosquitoes, which in turn could condition mating efficiency (Anthony and Jallon 1982, Ferveur et al. 1996, Jallon 1984). Although the role of the female accessory sex gland in mating is currently unknown, virus infection could alter sex gland physiology, perhaps perturbing mating efficiency.
Because TOT is such a critical component of the cycle of LACV and indeed is likely the principal mechanism for LACV maintenance and amplification in nature (Beaty and Bishop 1988), studies were conducted to determine if Ae. triseriatus females that had been infected via TOT also exhibit enhanced mating efficiency. This would confer a fitness advantage for progeny of LACV-infected mosquitoes and would amplify LACV in nature.
Aedes triseriatus mosquitoes (AIDL strain) were originally collected as eggs near La Crosse, WI, in 1981 and have been continuously maintained in the Arthropod-borne and Infectious Diseases Laboratory at Colorado State University (Fort Collins, CO). Mosquitoes from a laboratory colony were selected for LACV TOT and have a FIR rate of ~ 60%. For this study, eggs were hatched and larvae were fed a 1:1 mixture of ground fish-food/mouse food. Pupae were separated by sex and allowed to emerge in 3.8-liter containers. Adults were maintained on sugar cubes and water at 20–23°C, 80% RH and a photoperiod of 16:8 (L:D) hours.
Aedes triseriatus eggs were collected from oviposition traps at various collection sites in Wisconsin and Minnesota in the summer of 2006. The eggs were transported to AIDL and promptly hatched and reared in the insectary as described above. Field-collected male mosquitoes were not used in these experiments. Female mosquitoes were collected, assayed for LACV infection, and compared for mating efficiency.
Only mosquitoes infected through TOT were used in the experiments. Female mosquitoes were assayed for LACV infection by immunofluorescence assay (IFA). One rear leg was removed from each mosquito, squashed onto a microscope slide, fixed in acetone, and assayed for LACV antigen by direct IFA using fluorescein isothiocyanate (FITC)-conjugated LACV-specific polyclonal antibody at a dilution of 1:100 (Beaty and Thompson 1975, Graham et al. 1999). Four groups of female mosquitoes (AIDL LACV+, AIDL LACV−, FC LACV+ and FC LACV−) were compared in regard to mating efficiency in the experiments outlined below.
In three separate trials, the AIDL LACV+ and LACV− mosquitoes were offered a blood meal containing defibrinated sheep blood through a membrane feeder (Rutledge et al. 1964). The mosquitoes were allowed to feed for 1.5 h, after which the engorged and non-engorged females were separated. The FC LACV+ and LACV− mosquitoes were also provided a blood meal. The engorged females were tested for LACV antigen by IFA of leg tissue, and infected and uninfected mosquitoes were separated.
In triplicate trials, eight different groups of females were placed separately into 0.27-m3 cages for mating: (1) AIDL LACV+ and (2) AIDL LACV− no blood meal, (3) AIDL LACV+ and (4) AIDL LACV− blood meal, (5) FC LACV+ and (6) FC LACV− no blood meal, and (7) FC LACV+ and (8) FC LACV− blood meal. Laboratory colonized male mosquitoes were added to each cage of AIDL and FC female mosquitoes at a 2:1 (male:female) ratio, which is optimal for insemination (Mather and Defoliart 1984). AIDL female mosquitoes were removed from the cage every 24 hours beginning at 1 d postmixing, and the experiments were carried out until there was no difference in the insemination rate observed (day 5 for blood-fed mosquitoes and day 7 for non-blood-fed mosquitoes). All FC females were removed 7 d postmixing to determine insemination rates. The spermathecae of all females were examined microscopically for the presence of motile sperm to determine if insemination had occurred (Gabitzsch et al. 2006, Mather and Defoliart 1984). The experiments were carried out for seven days.
Mosquito carcasses were triturated in 500 μl of Trizol (Invitrogen) using a pellet pestle (Fisher, Pittsburgh, PA) and placed at −70°C for >1 h. Samples were thawed to room temperature and RNA was extracted according to manufacturer’s instructions. Approximately 100 ng of total RNA was mixed with a reverse transcription primer with the sequence of nucleotides 71–95 of LACV S segment mRNA (5′-TCA AGA GTG TGA TGT CGG ATT TGG-3′) (Kempf et al. 2006). Superscript II (Invitrogen) was used to produce a cDNA with the sequence of LACV genomic RNA. In brief, total RNA, 10 μM dNTPs and 50 μM primer were incubated at 65°C for 10 minutes. Superscript II, 5X first-strand buffer, and 1mM dithiothreitol (DTT) were added and incubated at 42°C for 50 min. The reverse transcriptase was inactivated at 70°C for 15 min.
One-fourth of the cDNA prepared from each reverse transcription reaction was used for quantitative polymerase chain reaction (qPCR) analysis. The forward primer was SF (5′-GGT TAG CCT TCC TCT CTG GCT TA-3′), which has the sequence of nucleotides 268–246 of LACV S mRNA. The reverse primer, SR (5′-CCT TGC TGC AGT TAG GAT CTT CTT-3′) has the sequence of nucleotides 186–209 of the LACV S genomic RNA (Kempf et al. 2006). qPCR primers were purchased from Qiagen (Valencia, CA) and qPCR reagents were obtained from Stratagene (Brilliant Q-PCR reagents with SureStart TaqDNA polymerase) and used according to manufacturer’s instructions (with the exception that reactions contained a total volume of 20 μl instead of 50 μl). Serial 10-fold dilutions of known copy number control plasmid (1 × 101 to 1 × 108) were amplified simultaneously to generate standard curves.
Two different statistical tests were used to compare the mean LACV genome equivalents in the inseminated and noninseminated AIDL and FC mosquitoes. A Student’s t-test was used to compare the mean LACV genome equivalents in FC mosquitoes. The Wilcoxon signed rank test was used to determine if there was a significant difference in LACV genome equivalents between the AIDL LACV+ inseminated and noninseminated mosquitoes (Wilcoxon 1945). A χ2 test was used to determine if there was a significant difference between the proportion of inseminated and noninseminated mosquitoes for each group.
In non-blood-fed mosquitoes, the AIDL LACV+ females were inseminated more rapidly than the AIDL LACV− mosquitoes (Fig. 1). Motile sperm were not detected in the spermathecae of mosquitoes on days 1 and 2 after exposure to males. However, a significant difference was observed between AIDL LACV+ and LACV− mosquitoes on days 3 and 4. On day 3, the LACV+ mosquitoes had an insemination rate of 46.4% (51/110) compared with 31.8% (54/170) for the uninfected mosquitoes (χ2 = 6.07, P < 0.01). On day 4, the LACV+ mosquitoes had an insemination rate of 53.3% (48/90) compared with 37.3% (41/110) for the uninfected mosquitoes (χ2 = 5.17, P < 0.05). By day 7, the rates of insemination were 84.9% (73/86) for the AIDL LACV+ females and 85.6% (115/133) for the AIDL LACV− mosquitoes (χ2 = 0.108, p = 0.783; Fig. 1).
In the blood-fed mosquitoes, the AIDL LACV+ females also were inseminated earlier than LACV− females (Fig. 2). No insemination occurred in the first day of postmixing. A difference in insemination rates was observed on day 2 after exposure to males (Fig. 2), when LACV+ mosquitoes had an insemination rate of 46.7% (28/60) compared with 30.0% (18/60) of LACV− mosquitoes (χ2 = 3.53, P = 0.06); however, it was not a significant difference. On day 3, there was a significant difference between the LACV+ with an insemination rate of 80.0% (88/110) compared with 43.9% (58/132; χ2 = 79.6, P < 0.001). By day 5, the insemination rates for the infected and uninfected groups were similar at 87.3% (62/71) and 88.7% (89/97), respectively (χ2 = 0.70, P = 792) (Fig. 2).
In the non-blood-fed mosquitoes, the FC LACV+ females were inseminated more frequently than the FC LACV− females (Fig. 3). The FC mosquitoes were allowed to mate for 7 d because their insemination rates are much lower than those for laboratory colony mosquitoes. The FC LACV+ mosquitoes had an insemination rate of 33.3% (8/24) compared to 15.7% (18/115) for the uninfected mosquitoes (χ2 = 4.08, P = 0.043; Fig. 3).
In blood-fed mosquitoes, the females also were allowed to mate for 7 d. Again, more infected than non-infected mosquitoes were inseminated. However, the differences were not statistically different, perhaps because of a smaller sample size. The FC LACV+ mosquito insemination rate was 18.2% (6/33) compared to 15.5% (9/58) for the LACV− mosquitoes (χ2 = 0.108, P = 0.742; Fig. 3).
The carcasses of the infected female mosquitoes from the previous experiments were assayed by qRT-PCR to measure LACV genome equivalents to determine whether viral load conditioned insemination. Total RNA was extracted from the mosquitoes at a time point when the largest difference in insemination rates between each pair of groups was observed (day 3: AIDL LACV+, non-blood-fed and blood-fed groups; day 7: FC LACV+ blood fed and non-blood fed groups). Viral load as determined by genome equivalents did not differ significantly between inseminated and non-inseminated AIDL in blood-fed groups (χ2 = 3.73, P = 0.054) and non-blood-fed groups (χ2 = 0.920, P = 0.338). There was also no significant difference observed between inseminated and non-inseminated FC mosquitoes in both the blood-fed groups (χ2 = 1.02, P = 0.314) and non-blood-fed groups (χ2 = 1.80, P = 0.180; Table 1).
These experiments clearly showed that LACV infection promotes earlier insemination of laboratory colonized transovarially infected Ae. triseriatus mosquitoes (Figs. 1 and and2).2). This study expanded upon previous research that showed that oral LACV infection increases insemination rates in Ae. triseriatus mosquitoes (Gabitzsch et al. 2006). More rapid insemination rates were demonstrated in AIDL LACV+ mosquitoes. Earlier insemination rates in AIDL LACV+ mosquitoes were shown; however, the insemination rates for both AIDL LACV+ and LACV− mosquitoes were equal by day 7 (Figs. 1 and and2).2). Similarly, the non-blood-fed, FC LACV+ mosquitoes were more frequently inseminated than uninfected mosquitoes (Fig. 3). These experiments suggest that LACV infection is associated with earlier insemination.
The mating advantage of LACV+ females could be important epidemiologically. Higher early postemergence insemination rates in LACV+ females would increase the opportunity for TOT and venereal transmission of the virus and would thereby promote virus amplification and maintenance in nature by multiple mechanisms (Beaty et al. 2000, Gabitzsch et al. 2006). This could also compensate for the deleterious effects of LACV infection on Ae. triseriatus overwintering survival in natural conditions (McGaw et al. 1998).
Overall, AIDL female mosquitoes given a blood meal had mean insemination rates of 61.9%, compared to 39.1% of non-blood-fed mosquitoes by 3 d postmixing with males (Figs. 1 and and2).2). These results confirm previous studies that showed that blood-fed females are more receptive to insemination and at an earlier time point than unfed females (Mather and Defoliart 1984). In field collections, > 50% of mosquitoes seeking a bloodmeal were not inseminated (Porter and Defoliart 1985, Scholl et al. 1979). The earlier insemination of blood-fed, LACV+ mosquitoes seen in laboratory cage studies would likely be amplified in importance in forested natural areas where mate seeking is much more complex. Earlier insemination would provide a fitness advantage for progeny of LACV+ females.
When FC mosquitoes ingested a bloodmeal, the insemination rates of LACV+ and LACV− mosquitoes did not differ statistically. However, the trends were the same; infected females were inseminated (18.2%) more often than the uninfected (15.5%) (Fig. 3). This particular experiment was complicated by the low infection rate (3–5%)(Reese 2008) and consequently the small sample size of FC LACV+ mosquitoes. A greater sample size might provide statistically significant results.
There were multiple challenges encountered with the FC mosquitoes. The FC mosquitoes had low insemination rates compared with the AIDL mosquitoes, which could be because of a lack of laboratory adaptation or small sample size. Both the FC and AIDL females were mixed with AIDL males because the FC males would not mate in the laboratory setting. This suggests that the FC mosquitoes were not adapted to laboratory conditions resulting in minimal insemination. The use of AIDL male mosquitoes allowed consistency between the experiments using FC and AIDL females and also increased the insemination rates for the FC females.
The FC insemination rates could also have been affected by the small number of mosquitoes in the 0.27-m3 cage, which could reduce swarming and possibly mating efficiency. As mentioned previously, the number of FC LACV+ mosquitoes available for the experiments was limited. For each of the experiments with the AIDL mosquitoes, there were at least twice as many females in the cage, thereby four times as many males, allowing a much larger swarm than found in the FC mosquito cages. This may be a possible reason for lower insemination rates with the FC mosquitoes compared with the AIDL mosquitoes.
Another challenge encountered with the FC mosquitoes was the longer time period needed to observe insemination. The AIDL mosquitoes were examined for insemination every 24 h for 7 ds and the largest difference in rate of insemination was observed on day 3 for both non-blood-fed and blood-fed mosquitoes. However, there was minimal insemination observed before day 7 (data not shown). This may have been because of the lack of adaptation to laboratory settings and small sample size, as discussed above.
LACV load could possibly affect mating behavior and thus efficiency. Differences in individual mosquito virus titers could have an effect on behavior and therefore were investigated by qRT-PCR to measure genome equivalents (Kempf et al. 2006). These qRT-PCR analyses showed that earlier insemination was not virus dose dependent (Table 1). Interestingly, LACV load as determined by qRT-PCR was greater in the AIDL LACV+ mosquitoes. The reason for this remains to be determined. It is noteworthy that Ae. dorsalis stably infected with California encephalitis virus have lower virus titers than nonstably infected mosquitoes. Because their germinal tissues are infected, virtually all progeny of stably infected mosquitoes are infected (Turrell et al., 1982). Future studies will investigate the anatomical and potential epidemiological significance of this observation.
The earlier insemination could be caused by a physiological determinant. Stimulatory pheromones have an important role during the mating ritual of the order Diptera (Ferveur et al. 1996, Ferveur et al. 1997, Greenspan and Ferveur 2000, Jallon 1984). They are volatile and would be detected before the first physical contact, at a distance of less than a few centimeters (Greenspan and Ferveur 2000). Many dipterans have sex pheromones composed of cuticular hydrocarbons (CHC) for recognition and these substances often are altered with physiological state and age (Pomonis 1989, Trabalon et al. 1988). The female accessory sex gland could play a role in pheromone production, resulting in a more attractive female and thereby influencing mating efficiency. The role of the female accessory gland in the mating rituals of mosquitoes is not understood, but it is intriguing that the accessory glands of intrathoracically inoculated LACV+ Ae. triseriatus mosquitoes are infected (Reese 2008).
Numerous additional factors could be affected by LACV infection to result in earlier insemination. The female mosquito must undergo physiological changes to prepare for mating, including alteration in wing beat and CHC production. Female mosquitoes control the refractory period during which neither physical coupling nor cues that are necessary for male ejaculation can occur (Klowden and Zwiebel 2005). Posteclosion production of juvenile hormone (JH) appears to cause the development of mating competence. It is possible that the release of JH in a LACV+ mosquito occurs earlier than in an uninfected mosquito, allowing the LACV+ females to be receptive to early mating. A number of other factors influence mating of mosquitoes including acoustics, chemical cues and behavior. A pathological effect could result in decreased activity of the female, allowing easier access for mating. Any one of these factors or a combination of these factors could be affected by a LACV infection to alter the female mosquitoes to become either more attractive or receptive to mating.
The demonstration that insemination occurs more rapidly in LACV+ than in LACV− females could have major epidemiological implications. Transovarially infected females that emerge and immediately mate will then produce infected progeny to continue the cycle. If LACV+ females mate more rapidly than LACV− females, even if the overall rate of insemination is the same between the two groups, the number and percentage of infected mosquitoes will increase over a transmission season. Such a mating advantage could contribute to LACV persistence in nature and promote virus amplification and maintenance.
We thank C. Meredith for maintaining the Ae. triseriatus mosquito colonies used in this study and A. Winters for assisting with the statistical analyses performed in this study. This research was funded by grant AI 32543 from the National Institutes of Health. SR was supported by CDC FTP T01/CCT 822307.