INTRODUCTION
West Nile virus (WNV) has become the most prevalent arthropod-borne virus (arbovirus) in the United States, causing more than 16,000 reported cases and 600 deaths in humans since its introduction into New York in 1999.1 The virus is maintained in an enzootic cycle involving birds and mosquitoes (primarily Culex species).2 Vertebrates generally become infected with WNV during blood feeding by infected mosquitoes although alternative transmission routes do exist.2,3 Nonetheless, most laboratory studies of WNV pathogenesis, host competence, and vaccine efficacy use host animals that are infected with WNV via needle inoculation because of the added complications of using infected mosquitoes.
Although needle inoculation has been accepted as a practical alternative to mosquito transmission of viruses, it is an imperfect substitute4 due to the mechanics of mosquito blood feeding. Mosquitoes alight on a potential host and actively probe for blood with their mouthparts throughout the dermal layer of skin. After a blood vessel is pierced, the mosquito begins feeding either directly from the vessel or from the resulting hemorrhagic pool. During the probing and feeding process, the mosquito ejects saliva that contains virus (if the mosquito is infectious), along with pharmacologically active molecules that counteract the host hemostatic response, reduce inflammation, and alter host immunity.5–12
Vertebrates infected with arboviruses by mosquito bite or when associated with mosquito saliva have been shown to exhibit enhanced infection compared with those infected by needle inoculation. Mice inoculated intradermally with WNV in an area where ~11 uninfected mosquitoes had fed had lower survival rates, higher and longer-lasting viremia, and accelerated neuroinvasion.13 Mice infected with vesicular stomatitis virus (VSV) by mosquitoes had higher seroconversion rates than those infected by needle inoculation.14 Chipmunks and deer infected with La Crosse virus by mosquito bite had higher and longer-lasting viremia than those infected by intramuscular inoculation.15 Mice inoculated subcutaneously with Cache Valley virus in an area where uninfected mosquitoes had recently fed exhibited higher infection rates and antibody response than those inoculated without prior mosquito feeding.16
On the other hand, other studies have shown no effect of mosquito transmission on vertebrate infection. One-week-old chickens and adult house finches infected by needle or mosquito bite with St. Louis encephalitis (SLE) or western equine encephalomyelitis (WEE) viruses did not differ significantly in viremic level or antibody response.17 Additionally, no difference in viremia was observed in hamsters18 or chickens19 infected with WNV by needle or mosquito inoculation, but early time points were not included in these studies.
The goal of this study was to further explore mosquito enhancement of arboviral infections in vertebrates. Young chickens were used as an avian model for the WNV enzootic transmission cycle. Viremia and viral shedding were compared for chickens infected with WNV by mosquito bite or subcutaneous needle inoculation of 103 plaque-forming units (PFU). In addition, the effect of viral dose on infection was examined.
METHODS
Virus.
All experiments were conducted with a WNV strain isolated in 2000 from the brain of a crow collected in Staten Island, NY. To obtain a more consistent virus stock, this isolate was plaque-purified 3 times in African green monkey kidney (Vero, ATCC CCL-81) cell culture prior to use in our study and had a titer of 108.5 PFU/ml. We inoculated 5-day-old chickens (N = 5/group) with 103 PFU of the original WNV isolate or plaque purified stock to determine if virus passage impacted replication ability in chickens. There were no significant differences in viremia or oral swab titers between the two groups of chickens at 12 hours, or 1–7 days PI (data not shown).
Animals.
We used a Cx. pipiens colony established in 2000 from mosquitoes collected in Albany, NY and a Cx. tarsalis colony that was derived from the Bakersfield, CA colony established in 1953 (kindly provided by Dr. William Reisen). Specific pathogen-free chickens (Charles River SPAFAS, North Franklin, CT) were housed in a BSL-3 animal facility. The use of chickens in this experiment was approved and conducted in accordance with the Wadsworth Center Institutional Animal Care and Use Committee.
Infection of chickens with West Nile virus by needle or mosquito.
Mosquitoes were infected with WNV by intrathoracic inoculation of ~30 PFU WNV 7 days (Cx. pipiens) prior to feeding on 1-day-old chickens and 6 days (Cx. tarsalis) or 11 days (Cx. pipiens) prior to feeding on 5-day-old chickens. The amount of virus expelled by WNV-inoculated Cx. pipiens and Cx. tarsalis as they fed on hanging blood drops did not vary significantly between day 7–10 post-inoculation (PI) (data not shown). After inoculation, mosquitoes were held in 0.5L cardboard cartons with a mesh top at 27°C, and provided with 10% sucrose via a soaked cotton pad. Mosquitoes were starved by removing the sucrose pad from cartons for 24–48 hours prior to feeding on chickens.
One-day-old and 5-day-old chickens were infected with WNV by subcutaneous needle inoculation of ~103 PFU WNV in the lateral neck, or by allowing WNV-infected mosquitoes to feed on a restrained chicken for ~1 hour. A previous study showed that mosquitoes inoculate a median of 103–104 PFU and mean of 104.3 PFU WNV during in vitro transmission assays.20 Chickens were exposed to either a single or multiple (up to 12) infected mosquitoes. After the feeding period, mosquitoes exposed to each chicken were frozen, and the number of mosquitoes with blood in their abdomen was determined using a dissecting microscope. This procedure allowed us to ascertain the minimum number of mosquito bites that each chicken received. We were not able to detect mosquitoes that only had probed. Three replicates of this experiment were completed using the previously described protocol: two with WNV-infected Cx. pipiens and one with WNV-infected Cx. tarsalis.
Sample collection and processing.
At various times post-feeding (PF) or PI, blood, oral swab, and cloacal swab samples were collected from chickens. Blood samples were collected from the ulnar vein into a microtiter centrifuge tube, held at 4°C for up to 3 hours and centrifuged (8,000 rpm, 5 min). Serum was removed, diluted 1:10 with BA-1 diluent (M199H, 1% bovine serum albumin, 0.05 M Tris pH 7.6, 0.35 g/L sodium bicarbonate, 100 u/ml penicillin, 100 μg/ml streptomycin, 1 μg/ml fungizone), and stored at −80°C. Oral and cloacal swabs were obtained by swabbing the inside of the mouth or cloaca, respectively, with a cotton-tipped applicator stick, placing the swab into 500 μL BA-1 diluent, and storing at −80°C. All samples were titrated by plaque assay on Vero cells. After virus titers had been determined, serum samples were heat-inactivated at 56°C for 30 minutes, and tested for WNV-specific IgG and IgM antibodies by indirect IgG and IgM ELISA, respectively. We followed the IgG ELISA protocol of Ebel and coworkers,21 with a minor change in the blocking buffer from 2.0% casein to 5.0% skim milk. IgM ELISA followed the same protocol, except that horseradish peroxidase-conjugated goat anti-chicken IgM (Bethyl Laboratories, Montgomery, TX) was used. End point titrations were conducted by plaque reduction neutralization assay (PRNT90) on samples positive for IgG (positive antigen wells/negative antigen wells > 2.0).
Chickens inoculated with successively higher viral doses.
We investigated the effect of viral dose on viral replication in chickens. Groups of five 5-day-old chickens were inoculated subcutaneously in the lateral neck with successively higher doses of WNV from 101–107 PFU. Serum and oral swab samples were taken as described previously at various times PI and were tested for infectious virus by plaque assay.
Viral titers inoculated by mosquitoes.
Viral titers inoculated by mosquitoes were estimated by allowing mosquitoes to feed on hanging blood drops.22 Cx. tarsalis and Cx. pipiens were intrathoracically inoculated with ~30 PFU WNV, maintained in 0.5L cardboard cartons with a mesh top at 27°C for 6–8 days, and provided with 10% sucrose via a soaked cotton pad. Mosquitoes were starved by removing sucrose from the cartons at least 24 hours prior to blood feeding. On days 7–10 PI, mosquitoes were placed individually into 0.5L cardboard cartons, and a 30-uL drop of sweetened goose blood (1 part 50% sucrose, 24 parts defibrinated goose blood) was pipetted onto the mesh top of each carton. Mosquitoes were allowed 45–60 minutes to feed on the blood drop, after which the blood remaining in each drop was collected and placed into 500 μL of BA-1 diluent. Mosquitoes were observed throughout the feeding period; probing (mouthparts in contact with blood droplet) and engorgement status of each female was recorded. Blood drop samples were titrated by plaque assay on Vero cells.
Statistical analysis.
Serum, swab, and hanging drop titers were log transformed and checked for normality using Shapiro-Wilk and Kolmogorov-Smirnov statistics. Negative serum samples were given a mathematical value of 50 PFU/ml in statistical analyses due to the high limit of detection for virus in serum samples (100 PFU/ml). However, negative samples were set at 0 PFU/ml in the graphs for clarity. Because multiple serum and swab samples were taken from individual chickens, we used repeated measures ANOVA (PROC GLM)23 to determine whether viral titers varied by infection method (needle versus mosquito bite), mosquito exposure level (single or multiple mosquito bites), and inoculum dose for each age group of chickens. The Tukey-Kramer method was used to adjust for multiple comparisons. Because of the small sample size, Fisher’s exact test was used to determine significant differences between proportions of positive IgM and IgG samples.
RESULTS
Infection profile of young chickens infected with West Nile virus.
West Nile virus infection was assessed for two different ages of chickens, 1-day and 5-day, using viremia, shedding, and antibody profiles. All chickens inoculated with WNV by needle (N = 20) or fed upon by WNV-infected mosquitoes (N = 15) became infected, exhibiting significant viremia, viral shedding, and production of WNV-specific IgG and IgM antibodies. In addition, 5 of 18 chickens became infected after exposure to WNV-infected mosquitoes that did not blood feed. Of the 20 chickens infected after exposure to WNV-infected mosquitoes, 9 became infected after exposure to a single mosquito, 6 became infected after exposure to multiple mosquitoes, and 5 were infected by an unknown number of mosquitoes. (These 5 chickens were exposed to multiple mosquitoes, but blood was detected in ≤ 1 mosquito) (Table 1).
We did not detect blood in the midgut of mosquitoes exposed to 5 chickens (four 5-day-old chickens and one 1-day-old chicken) that became infected. The 5-day-old chickens showed similar viral replication kinetics to chickens infected by mosquitoes that imbibed blood; the absence of blood indicated that these chickens were most likely infected through mosquito probing. In contrast, serum and swab titers in the 1-day-old chicken were delayed 1–2 days compared with all other infected 1-day-old chickens in the study (Figure 1). Although it is possible that the delayed viremia was due to a low dose inoculated by a probing mosquito, a more likely scenario is that this chicken was infected via intracage transmission on day 1 PF. At that time, 3 cage mates of this chicken were shedding 103–104 PFU of virus orally and 101–103 PFU of virus via the cloaca.
Five infected chickens and one uninfected chicken died during the studies (see Table 1). The uninfected chicken, one of 12 uninfected chickens in the 1-day-old group, died on day 3 PF. This death was presumed to be due to failure to thrive. The infected chickens died on days 6–8 PF or PI; 4 were infected by mosquito bite, and 1 was infected by needle.
Age of the chicken had an impact on viral infection (see Figure 1). Peak titers in sera and swabs were ~10 fold lower in 5-day-old chickens than in 1-day-old chickens. Five-day-old chickens cleared virus ~2 days faster from sera and from oral and cloacal cavities. Age also had an impact on immune response. A greater percentage of 5-day-old chickens had detectable IgM and IgG antibodies at days 6–7 PF or PI compared with the younger chickens (IgM: 89% versus 17%, Fisher’s exact test, P = 0.01; IgG: 82% versus 38%, Fisher’s exact test, P = 0.07) (Figure 2A and 2B). Neutralizing antibody titers were also observed to rise later in 1-day-old chickens than in 5-day-old chickens (Figure 2C). Although a greater percentage of infected chickens from the 1-day-old group died (31%, N = 13) compared with the 5-day-old group (7%, N = 27), this difference was not statistically significant (Fisher’s exact test: P = 0.08).
Infection of chickens with West Nile virus by needle or mosquito.
The parameters of viral infection were analyzed for needle-inoculated and mosquito-inoculated chickens. One-day-old chickens infected with WNV by Cx. pipiens had significantly higher serum titers at 12 and 24 hours PF than chickens inoculated with 103 PFU by needle (12h: P = 0.013; 24h: P < 0.0001) (see Figure 1). Similarly, oral swab titers of chickens infected by Cx. pipiens were as much as 1,000 fold higher than titers in needle-inoculated chickens at 12, 24, and 48 hours PI (12h: P < 0.0001; 24h: P < 0.0001; 48h: P = 0.038). Cloacal swab titers of mosquito-infected chickens were significantly higher at 24 hours PF than those of needle-inoculated chickens (P < 0.0001).
A similar pattern of enhanced early infection was observed in older chickens. Serum titers in 5-day-old chickens infected by mosquito bite were significantly higher at 12 and 24 hours PF than in needle-inoculated chickens (Cx. pipiens versus needle 12h: P = 0.001, 24h: P < 0.0001; Cx. tarsalis versus needle 12h: P = 0.003; 24h: P < 0.0001) (see Figure 1). At 24 hours, oral and cloacal swab titers of 5-day-old chickens infected by mosquito bite were ~10 fold higher than in needle-inoculated chickens (oral: Cx. pipiens versus needle P = 0.0001; Cx. tarsalis versus needle P = 0.0007; cloacal: Cx. pipiens versus needle P = 0.0095).
There was some indication of more rapid viral clearance in chickens infected by mosquito bite. Viremia in 5-day-old chickens infected by Cx. pipiens was significantly lower at 48 and 72 hours post feeding than in needle-inoculated chickens (48h: P = 0.01, 72h: P = 0.02). Additionally, oral swab titers were significantly lower at 96 hours post feeding compared with 5-day-old chickens infected by needle (P = 0.0007). However, significantly faster viral clearance was not seen in 1-day-old chickens infected by Cx. pipiens or in 5-day-old chickens infected by Cx. tarsalis.
Five-day-old chickens infected by two different mosquito species had similar infection profiles (see Figure 1). There were no significant differences in serum titers between 5-day-old chickens infected by Cx. pipiens or by Cx. tarsalis at 12, 24, 72, and 96 hours PF. However, at 48 hrs PF, chickens infected by Cx. tarsalis had higher serum titers than chickens infected by Cx. pipiens (P = 0.0496). Oral swab titers were not significantly different between chickens infected by Cx. pipiens or by Cx. tarsalis, except at 96 hours PF when oral swab titers of chickens infected by Cx. tarsalis were higher than in those infected by Cx. pipiens (P = 0.0144). It is important to note that these differences in serum titers at 48 hours were confounded by the number of mosquitoes feeding (see analysis later in this article).
There were no significant differences in PRNT90 titers between chickens infected by mosquito and those infected by needle in each age group (see Figure 2C). However, a greater proportion of 5-day-old chickens infected by mosquito had detectable IgM antibody on day 4 PI (50%, N = 14) compared with 5-day-old chickens infected by needle (9%, N = 11) (Fisher’s exact test: P = 0.04) (see Figure 2A).
Chickens infected by a single mosquito or multiple mosquitoes.
We compared serum and swab titers of 5-day-old chickens infected by a single mosquito or multiple mosquitoes, as defined in Table 1. Chickens infected by multiple mosquitoes had serum titers that were ~25 times higher at 12 hours (P = 0.014) and ~3 times higher at 24 hours (P = 0.029), compared with chickens infected by a single mosquito (Figure 3). Serum titers of chickens infected by multiple mosquitoes were 100 fold lower at 72 hours (P = 0.042). When this restricted data set was used, differentiating single from multiple mosquito bites, there was no difference between serum titers of chickens infected by Cx. pipiens and by Cx. tarsalis. Similarly, oral swab titers of chickens infected by multiple mosquitoes were lower at 72 hours (P = 0.034) and higher at 24 hours than those infected by a single mosquito; however, the difference at 24 hours was not significant. Chickens infected by multiple mosquitoes also had significantly higher cloacal swab titers at 24 hours (P = 0.039) compared with chickens infected by single mosquitoes.
Chickens inoculated with successively higher doses by needle.
We determined whether enhanced early infection in mosquito-infected chickens could be due to higher inoculum dose delivered by mosquitoes. Chickens inoculated with successively higher doses of WNV by needle had serum and oral swab titers that increased in tandem at early time points PI (Figure 4). When compared with data from the previous experiment, serum and oral swab titers of chickens infected by a single mosquito were most similar to those from chickens inoculated with 107 PFU by needle. There was no statistical difference, however, between WNV titers in chickens infected by a single mosquito and chickens inoculated with ≥ 104 PFU at 6–24 hours PI. Serum titers in chickens infected by multiple mosquitoes were 25–50 times higher than all other groups at 6 and 12 hours PI (P < 0.04); there was no significant difference between chickens infected by multiple mosquitoes and those inoculated with ≥ 104 PFU at 24 hours PI (see Fig. 4).
Dose inoculated by mosquitoes while blood feeding.
We estimated the amount of virus injected by parenterally infected Cx. pipiens and Cx. tarsalis while probing and feeding on a hanging drop of blood. Mosquitoes that imbibed blood injected more virus than those that probed without blood feeding (t test: P = 0.03) (Figure 5). Geometric mean virus titer and range of titers expelled by blood fed mosquitoes did not differ significantly by species (Cx. pipiens: mean = 102.8, range = 100.7–104.3 PFU, Cx. tarsalis: mean = 102.0, range = 100.7–104.6 PFU). Median titer expelled by blood fed Cx. pipiens (103.1 PFU) was higher than that expelled by blood fed Cx. tarsalis (101.8 PFU).
DISCUSSION
In this study, young chickens infected with WNV by mosquito bite had enhanced early infection compared with those needle inoculated with 103 PFU. Mosquito-infected chickens exhibited viremia and viral shedding titers that were as much as 1,000 fold higher at early times PF than those of chickens infected by needle inoculation. Enhanced early infection in chickens infected by mosquitoes was seen in all experimental replicates, occurring in two age groups of chickens (1-day and 5-day old), as well as in chickens infected by two mosquito species (Cx. pipiens and Cx. tarsalis). Virus was cleared more quickly from the sera and oral swabs of 5-day-old chickens infected by Cx. pipiens, however this phenomenon was not seen in 5-day-old chickens infected by Cx. tarsalis or in 1-day-old chickens infected by Cx. pipiens.
These results are in contrast to two previous studies with WNV. In a hamster model, no difference in viremia was observed on days 1–3 in hamsters infected with WNV by needle or mosquito inoculation, but time points earlier than 24 hours were not evaluated.18 Another study reported no difference in viremia in older chickens (17–60 weeks) infected with WNV by needle or mosquito bite; however, viremic levels were low (< 104 PFU/ml) and were not measured at early time points post-inoculation (< 24 hr).19 Thus, the differences between these two studies and our current results are most likely explained by differences in sampling times and/or animal models.
Enhanced early WNV infection in mosquito-infected chickens may partially be explained by higher viral dose delivered by mosquitoes. Higher doses inoculated by needle resulted in more rapid development of viremia and oral shedding. On the other hand, a needle inoculation of 107 PFU was required to attain the high early viremia observed in chickens infected by single mosquitoes (although there was no statistical difference between chickens infected by a single mosquito and those inoculated with ≥ 104 PFU at 6–24 hours). This result suggests that mosquitoes need to inoculate at least 104 PFU consistently while probing and feeding if more rapid development of viremia is due only to high viral dose.
The amount of WNV inoculated by mosquitoes while probing or feeding on a live host is not known. Our data using the hanging drop method suggest that mosquitoes inoculate a wide range of viral titers (100.7–104.6, mean = 102.5 PFU) while blood feeding. Similar results were reported by Vanlandingham and colleagues,20 who observed that mosquitoes inoculate a range of 100.5–105.3 PFU (mean = 104.3) of WNV as measured by in vitro capillary tube transmission assay. Other studies with WEE virus,17 SLE virus,17 and VSV,14 using in vitro capillary tube transmission assays, have reported wide ranges and similar average salivary secretion titers of 102–104 PFU. Despite these consistent results for several arboviruses, in vitro assays may underestimate the inoculated dose and introduce variability because mosquitoes are not feeding normally in these assays. The hanging drop method that was used in this study may be more accurate because mosquitoes are able to probe and feed, but this method still does not mimic natural feeding completely because mosquitoes do not need to find a blood vessel or overcome the host hemostatic response. A previous study showed that significantly fewer Cx. tritaeniorhynchus transmitted WNV when feeding on a hanging blood drop than they did when feeding directly on a suckling mouse.24 In addition, mosquitoes probe for a shorter time (and presumably excrete less saliva) when feeding through a membrane than they do when feeding on an intact host.25 One study quantified the amount of Rift Valley fever virus inoculated by Cx. pipiens into a live host (median: 102.5, range: < 100.7–103.7 PFU); however, mosquitoes in this study were only allowed to feed or probe for 30 seconds and this study did not account for virus that may have bound to the cells prior to assay and thus was not detectable as infectious virus.26,27 We are currently conducting studies to determine the amount of WNV inoculated into live hosts. Preliminary results suggest that mosquitoes routinely inoculate between 104 and 105 PFU and can inoculate as much as 106.6 PFU while probing and feeding on a live host (unpublished data).
Overall our current data and studies by others indicate that mosquitoes routinely inoculate between 102 and 105 PFU and may inoculate as much as 106.6 PFU of various arboviruses, suggesting that the enhanced early infection in mosquito-infected chickens may be due to viral dose. However, two lines of evidence suggest that other factors are involved. First, viral dose cannot explain the high sera titers seen in chickens exposed to multiple infected mosquitoes (N = 3–11). Assuming that a single mosquito delivers a dose of ~106 PFU while feeding (a high estimate), the maximum dose inoculated by 11 mosquitoes is ~107 PFU. However, in our study chickens exposed to multiple mosquitoes developed sera titers that were 25–50 times higher than chickens inoculated with 107 PFU at 6 and 12 hours PI. Second, all chickens infected by mosquito feeding, or by mosquito probing without feeding, exhibited consistently high viremia and viral shedding titers at early time points. One would expect more variation in viremia and shedding titers if dose was the only cause for earlier development of viremia, especially considering that some mosquitoes only probed, whereas others both probed and fed during a 1-hour period.
Other differences, besides inoculum dose, exist between WNV infection via needle and mosquito bite that could explain the enhanced early infection observed in mosquito-infected chickens. First, mosquito saliva has been shown in previous studies to cause potentiation of viral infection,7,13,16 perhaps through its ability to modulate host immune response at the inoculation site. Mouse fibroblast cells treated with salivary gland homogenate had lower interferon α/β production and higher VSV growth kinetics than did untreated cells.7 Mosquito salivary gland extract and saliva also suppress murine splenocyte proliferation and alter cytokine production.8–12 A recent study demonstrated the important effect of mosquito feeding and salivary gland extracts on WNV infection. Mice fed upon by uninfected Aedes aegypti females prior to intradermal inoculation of WNV had higher WNV RNA titers at the inoculation site and draining lymph node, higher and longer lasting viremia, and more rapid neuroinvasion compared with mice inoculated with WNV without mosquito feeding.13 It is not known if potentiation due to mosquito saliva also occurs within the enzootic WNV cycle that involves Culex mosquitoes and avian hosts.
A second difference between mosquitoes and needles is viral inoculation site. Mosquitoes salivate (and, therefore, deposit virus) while probing and feeding throughout the dermal tissue, whereas subcutaneous needle inoculation deposits the entire inoculum in one location below the dermis. Although studies have suggested that mosquitoes inject most virus extravascularly,26,27 it is possible that some virus is injected intravascularly by mosquitoes while blood feeding. Faster introduction of virus into the blood stream by mosquitoes could result in earlier dissemination and higher early viremia in mosquito-infected chickens. In addition, multiple mosquitoes feeding on a single chicken may inoculate virus into multiple locations, resulting in more rapid initial viral replication. Increased viral replication at multiple sites could explain the high initial viremia that we observed in chickens bitten by multiple mosquitoes.
The final difference between viral infection by mosquito bite and needle inoculation is viral source. In this study, virus injected by needle was harvested from Vero cell culture, whereas virus inoculated by mosquitoes had replicated in mosquito cells. Differences in viral glycoprotein maturation have been observed between virus derived from invertebrate cells and from vertebrate cells.28,29 If these differences in viral glycoproteins alter viral entry into primary infection sites, it could affect viral amplification rates and thus early levels of viremia and viral shedding. In addition, mutations may occur during replication in the mosquito that can affect early viral replication.
West Nile virus infection in chickens is age dependent, perhaps due to age-related differences in the immune response. One-day-old chickens exhibited levels of viremia and viral shedding that peaked ~1 day later and ~10 fold higher than in 5-day-old chickens. Increased production of virus may have been due to slower antibody response in 1-day-old chickens; IgG, IgM, and neutralizing antibody titers of 1-day-old chickens were detectable 1–6 days later than when they were detectable in 5-day-old chickens. In studies by others, older chickens (7–60 weeks old) infected with WNV exhibited viral titers that peaked at a lower level (~105 PFU/ml).19,30 In addition, no significant difference was found between older chickens (17–60 weeks) infected by needle and those infected by mosquito bite; perhaps this lack of difference was due to lower overall viremic levels.19
One 1-day-old chicken, whose level of viral viremia and viral shedding rose 1–2 days later than other chickens, was presumably infected by intracage transmission. This chicken was housed with 3 chickens that were shedding 103–104 PFU of virus orally and 101–103 PFU of virus through the cloaca at the presumed time of infection. A previous study reported intracage transmission of WNV to a 20-week-old chicken; however, cloacal and oral swab titers of the infected cage mate were not recorded prior to the transmission event.19 Intracage transmission of WNV has also been reported in geese, ring-billed gulls, blue jays, black-billed magpies, and American crows.31,32 These species have been shown to shed up to 103.5–105.7 PFU orally and 102.4–106.0 PFU through the cloaca.32
Mosquitoes in our study were infected with WNV by intrathoracic inoculation to obtain mosquitoes that transmitted virus with ~100% efficiency. Because every mosquito was theoretically able to transmit virus, fewer animals were needed per treatment and mosquito-to-mosquito variability was reduced, leading to a more controlled study. Despite virus reaching the salivary glands more quickly in inoculated mosquitoes, no difference was found in the amount of virus expelled by perorally infected and inoculated mosquitoes into hanging drops or into live hosts (data not shown).
Previous studies showing mosquito enhancement of arbovirus infection used primarily negative-sense RNA viruses belonging to the virus families Rhabdoviridae and Bunyaviridae. Our study demonstrated mosquito bite enhancement of a positive-sense RNA virus (Flaviviridae) using a natural host and mosquito vector, and it is the first to demonstrate significant differences in viremia and viral shedding between needle-inoculated and mosquito-infected hosts. Although higher viral doses inoculated by mosquitoes may partially explain enhanced early WNV infection in chickens, other factors (mosquito saliva, differences in inoculum site, and viral source) could also play an important role. Future studies are planned to evaluate these possible factors.
Outcome of 1-day-old and 5-day-old chickens exposed to West Nile virus-infected mosquitoes*
| Age group | Mosquito species | Infected by single mosquito | Infected by multiple mosquitoes | Infected by unknown no. of mosquitoes | Uninfected |
|---|---|---|---|---|---|
| * Chickens either did not become infected or were infected by feeding/probing of single, multiple, or an unknown number of WNV-infected mosquitoes. Each fraction represents an individual chicken and shows the number of known bites (determined by presence of blood in mosquito abdomen)/total number of mosquitoes exposed to each chicken. | |||||
| † Presumed intracage transmission because sera and swab titers rose > 1 day later than all other infected chickens; ‡died day 3 PF; §died day 6 PF; ||died day 7 PF; ¶died day 8 PF. | |||||
| Note: One needle-inoculated chicken in the 1-day-old group died on day 8 PI (not included in table). | |||||
| 1 day old | Cx. pipiens | 0/1† | 3/10§ | 1/10¶, 1/12 | 0/1‡, 0/1, 0/1, 0/1, 0/1, 0/1, 0/1, 0/1, 0/1, 0/10, 0/10, 0/10 |
| 5 day old | Cx. pipiens | 0/1, 1/1§, 1/1 | 3/11, 5/10, 6/11 | 0/10, 1/10, 1/10 | 0/1, 0/1, 0/1, 0/1, 0/1 |
| 5 day old | Cx. tarsalis | 0/1||, 0/1, 1/1, 1/1, 1/1 | 5/6, 6/6 | – | – |
Geometric mean West Nile virus (WNV) titers in sera, oral swabs, and cloacal swabs of chickens infected with WNV. Chickens were infected at 1 or 5 days of age by subcutaneous needle inoculation of 103 PFU, the bite of WNV-infected Cx. pipiens or Cx. tarsalis, or (presumed) intracage transmission. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Geometric mean West Nile virus (WNV) titers in sera, oral swabs, and cloacal swabs of chickens infected with WNV. Chickens were infected at 1 or 5 days of age by subcutaneous needle inoculation of 103 PFU, the bite of WNV-infected Cx. pipiens or Cx. tarsalis, or (presumed) intracage transmission. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Geometric mean West Nile virus (WNV) titers in sera, oral swabs, and cloacal swabs of chickens infected with WNV. Chickens were infected at 1 or 5 days of age by subcutaneous needle inoculation of 103 PFU, the bite of WNV-infected Cx. pipiens or Cx. tarsalis, or (presumed) intracage transmission. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Antibody response of chickens infected with WNV by needle or mosquito bite (Cx. pipiens and Cx. tarsalis data combined) at 1 or 5 days of age. (A) Proportion of samples positive for WNV-specific IgM antibodies. (B) Proportion of samples positive for WNV-specific IgG antibodies. Samples were tested by indirect IgG and IgM ELISA at a serum dilution of 1:100. Positive samples had positive-to-negative ratio (P/N) ≥ 2.0. Results from 1-day-old chickens infected by mosquito bite were not included because of low sample size (n = 1). (C) Average WNV-specific neutralizing antibody (PRNT90) titers. PRNT titers are expressed as the reciprocal of the final dilution of serum that caused ≥ 90% reduction of plaques when tested against ~100 PFU of WNV. Only positive samples (PRNT90 > limit of detection) were used to calculate average.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Antibody response of chickens infected with WNV by needle or mosquito bite (Cx. pipiens and Cx. tarsalis data combined) at 1 or 5 days of age. (A) Proportion of samples positive for WNV-specific IgM antibodies. (B) Proportion of samples positive for WNV-specific IgG antibodies. Samples were tested by indirect IgG and IgM ELISA at a serum dilution of 1:100. Positive samples had positive-to-negative ratio (P/N) ≥ 2.0. Results from 1-day-old chickens infected by mosquito bite were not included because of low sample size (n = 1). (C) Average WNV-specific neutralizing antibody (PRNT90) titers. PRNT titers are expressed as the reciprocal of the final dilution of serum that caused ≥ 90% reduction of plaques when tested against ~100 PFU of WNV. Only positive samples (PRNT90 > limit of detection) were used to calculate average.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Antibody response of chickens infected with WNV by needle or mosquito bite (Cx. pipiens and Cx. tarsalis data combined) at 1 or 5 days of age. (A) Proportion of samples positive for WNV-specific IgM antibodies. (B) Proportion of samples positive for WNV-specific IgG antibodies. Samples were tested by indirect IgG and IgM ELISA at a serum dilution of 1:100. Positive samples had positive-to-negative ratio (P/N) ≥ 2.0. Results from 1-day-old chickens infected by mosquito bite were not included because of low sample size (n = 1). (C) Average WNV-specific neutralizing antibody (PRNT90) titers. PRNT titers are expressed as the reciprocal of the final dilution of serum that caused ≥ 90% reduction of plaques when tested against ~100 PFU of WNV. Only positive samples (PRNT90 > limit of detection) were used to calculate average.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Geometric mean WNV titers in sera, oral swabs, and cloacal swabs of 5-day-old chickens infected through the bite of a single or multiple WNV-infected Cx. pipiens or Cx. tarsalis. Cloacal swabs not done for chickens fed upon by Cx. tarsalis. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Geometric mean WNV titers in sera, oral swabs, and cloacal swabs of 5-day-old chickens infected through the bite of a single or multiple WNV-infected Cx. pipiens or Cx. tarsalis. Cloacal swabs not done for chickens fed upon by Cx. tarsalis. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Geometric mean WNV titers in sera, oral swabs, and cloacal swabs of 5-day-old chickens infected through the bite of a single or multiple WNV-infected Cx. pipiens or Cx. tarsalis. Cloacal swabs not done for chickens fed upon by Cx. tarsalis. Error bars represent standard error of the mean.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Average WNV titers in sera and oral swabs of 5-day-old chickens inoculated subcutaneously with successively higher doses of WNV (101–107 PFU). Five chickens inoculated per dose. Included for comparison are data from 5-day-old chickens infected by a single mosquito or multiple mosquitoes. Cx. pipiens and Cx. tarsalis data are combined. Groups contained within ellipses are not significantly different from chickens infected by single mosquitoes single mosquito group. Comparisons are by repeated measures ANOVA with adjustment for multiple comparisons (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Average WNV titers in sera and oral swabs of 5-day-old chickens inoculated subcutaneously with successively higher doses of WNV (101–107 PFU). Five chickens inoculated per dose. Included for comparison are data from 5-day-old chickens infected by a single mosquito or multiple mosquitoes. Cx. pipiens and Cx. tarsalis data are combined. Groups contained within ellipses are not significantly different from chickens infected by single mosquitoes single mosquito group. Comparisons are by repeated measures ANOVA with adjustment for multiple comparisons (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Average WNV titers in sera and oral swabs of 5-day-old chickens inoculated subcutaneously with successively higher doses of WNV (101–107 PFU). Five chickens inoculated per dose. Included for comparison are data from 5-day-old chickens infected by a single mosquito or multiple mosquitoes. Cx. pipiens and Cx. tarsalis data are combined. Groups contained within ellipses are not significantly different from chickens infected by single mosquitoes single mosquito group. Comparisons are by repeated measures ANOVA with adjustment for multiple comparisons (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Amount of WNV ejected by parenterally infected Cx. pipiens (PIP) and Cx. tarsalis (TAR) while probing (P) or blood feeding (BF) on a hanging drop of sweetened goose blood. Horizontal solid line indicates geometric mean titer of group. Horizontal dotted line indicates median titer of group. Thin solid line indicates limit of detection (LOD) of assay (5 PFU). The number of samples for each group that are below the LOD is shown.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Amount of WNV ejected by parenterally infected Cx. pipiens (PIP) and Cx. tarsalis (TAR) while probing (P) or blood feeding (BF) on a hanging drop of sweetened goose blood. Horizontal solid line indicates geometric mean titer of group. Horizontal dotted line indicates median titer of group. Thin solid line indicates limit of detection (LOD) of assay (5 PFU). The number of samples for each group that are below the LOD is shown.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Amount of WNV ejected by parenterally infected Cx. pipiens (PIP) and Cx. tarsalis (TAR) while probing (P) or blood feeding (BF) on a hanging drop of sweetened goose blood. Horizontal solid line indicates geometric mean titer of group. Horizontal dotted line indicates median titer of group. Thin solid line indicates limit of detection (LOD) of assay (5 PFU). The number of samples for each group that are below the LOD is shown.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 75, 2; 10.4269/ajtmh.2006.75.337
Address correspondence to Laura D. Kramer, Arbovirus Laboratories, Wadsworth Center, New York State Dept of Health, 5668 State Farm Road, Slingerlands, NY 12159. E-mail: Kramer@wadsworth.org
Authors’ addresses: Laura D. Kramer, Arbovirus Laboratories, Wadsworth Center, New York State Dept of Health, 5668 State Farm Road, Slingerlands, NY 12159, Telephone: (518) 869-4524, Fax: (518) 869-4530, E-mail: kramer@wadsworth.org. Linda M. Styer, Arbovirus Laboratories, Wadsworth Center, New York State Dept of Health, 5668 State Farm Road, Slingerlands, NY 12159, Telephone: (518) 862-4306, Fax: (518) 869-4530, E-mail: lstyer@wadsworth.org. Kristen A. Bernard, Arbovirus Laboratories, Wadsworth Center, New York State Dept of Health, 5668 State Farm Road, Slingerlands, NY 12159, Telephone: (518) 869-4519, Fax: (518) 869-4530, E-mail: kbernard@wadsworth.org.
Acknowledgments: The authors acknowledge the excellent technical assistance provided by Matthew Jones, Jennifer Longacker, Amy Lovelace, Christine Lussier, Sarah Sperry, and David Young. The authors thank the Wadsworth Center Tissue Culture Facility for providing cell culture support.
Financial support: This project has been funded in part with Federal funds from the National Institute of Allergy and Infectious Disease, National Institutes of Health, under Contract No. N01-AI-25490. The BSL-3 animal facility at the Wadsworth Center was used, which is funded in part by the animal core on the NIH/NIAID award U54A17158.
REFERENCES
- 1↑
Center for Disease Control. Division of vector borne and infectious disease: West Nile Virus. Available at: http://www.cdc.gov/ncidod/dvbid/westnile/index.htm.
- 2↑
Granwehr BP, Lillibridge KM, Higgs S, Mason PW, Aronson JF, Campbell GA, Barrett ADT, 2004. West Nile virus: where are we now? Lancet Infect Dis 4 :547–556.
- 3↑
Marra PP, Griffing S, Caffrey C, Kilpatrick AM, McLean R, Brand C, Saito E, Dupuis AP, Kramer L, Novak R, 2004. West Nile virus and wildlife. Bioscience 54 :393–402.
- 4↑
Randolph SE, Nuttall PA, 1994. Nearly right or precisely wrong? Natural versus laboratory studies of vector-borne diseases. Parasitol Today 10 :458–462.
- 5↑
Ribeiro JMC, Francischetti IMB, 2003. Role of arthropod saliva in blood feeding: sialome and post-sialome perspectives. Annu Rev Entomol 48 :73–88.
- 6
Ribeiro JM, 1987. Role of saliva in blood-feeding by arthropods. Annu Rev Entomol 32 :463–478.
- 7↑
Limesand KH, Higgs S, Pearson LD, Beaty BJ, 2003. Effect of mosquito salivary gland treatment on vesicular stomatitis New Jersey virus replication and interferon alpha/beta expression in vitro. J Med Entomol 40 :199–205.
- 8↑
Wasserman HA, Singh S, Champagne DE, 2004. Saliva of the Yellow Fever mosquito, Aedes aegypti, modulates murine lymphocyte function. Parasite Immunol 26 :295–306.
- 9
Cross ML, Cupp EW, Enriquez FJ, 1994. Differential modulation of murine cellular immune responses by salivary gland extract of Aedes aegypti. Am J Trop Med Hyg 51 :690–696.
- 10
Wanasen N, Nussenzveig RH, Champagne DE, Soong L, Higgs S, 2004. Differential modulation of murine host immune response by salivary gland extracts from the mosquitoes Aedes aegypti and Culex quinquefasciatus. Med Vet Entomol 18 :191–199.
- 11
Zeidner NS, Higgs S, Happ CM, Beaty BJ, Miller BR, 1999. Mosquito feeding modulates Th1 and Th2 cytokines in flavivirus susceptible mice: an effect mimicked by injection of sialo-kinins, but not demonstrated in flavivirus resistant mice. Parasite Immunol 21 :35–44.
- 12↑
Schneider BS, Soong L, Zeidner NS, Higgs S, 2004. Aedes aegypti salivary gland extracts modulate anti-viral and TH1/TH2 cytokine responses to sindbis virus infection. Viral Immunol 17 :565–573.
- 13↑
Schneider BS, Soong L, Girard YA, Campbell G, Mason P, Higgs S, 2006. Potentiation of west nile encephalitis by mosquito feeding. Viral Immunol 19 :74–82.
- 14↑
Limesand KH, Higgs S, Pearson LD, Beaty BJ, 2000. Potentiation of vesicular stomatitis New Jersey virus infection in mice by mosquito saliva. Parasite Immunol 22 :461–467.
- 15↑
Osorio JE, Godsey MS, DeFoliart GR, Yuill TM, 1996. La Crosse viremias in white-tailed deer and chipmunks exposed by injection or mosquito bite. Am J Trop Med Hyg 54 :338–342.
- 16↑
Edwards JF, Higgs S, Beaty BJ, 1998. Mosquito feeding-induced enhancement of Cache Valley Virus (Bunyaviridae) infection in mice. J Med Entomol 35 :261–265.
- 17↑
Reisen WK, Chiles RE, Kramer LD, Martinez VM, Eldridge BF, 2000. Method of infection does not alter response of chicks and house finches to western equine encephalomyelitis and St. Louis encephalitis viruses. J Med Entomol 37 :250–258.
- 18↑
Sbrana E, Tonry JH, Xiao SY, Darosa AP, Higgs S, Tesh RB, 2005. Oral transmission of West Nile virus in a hamster model. Am J Trop Med Hyg 72 :325–329.
- 19↑
Langevin SA, Bunning M, Davis B, Komar N, 2001. Experimental infection of chickens as candidate sentinels for West Nile virus. Emerg Infect Dis 7 :726–729.
- 20↑
Vanlandingham DL, Schneider BS, Klingler K, Fair J, Beasley D, Huang J, Hamilton P, Higgs S, 2004. Real-time reverse transcriptase-polymerase chain reaction quantification of West Nile virus transmitted by Culex pipiens quinquefasciatus. Am J Trop Med Hyg 71 :120–123.
- 21↑
Ebel GD, Dupuis AP, Nicholas D, Young D, Maffei J, Kramer LD, 2002. Detection by enzyme-linked immunosorbent assay of antibodies to West Nile virus in birds. Emerg Infect Dis 8 :979–982.
- 22↑
Gubler DJ, Rosen L, 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.
- 24↑
Akhter R, Hayes CG, Baqar S, Reisen WK, 1982. West Nile virus in Pakistan. III. Comparative vector capability of Culex tritaeniorhynchus and eight other species of mosquitoes. Trans R Soc Trop Med Hyg 76 :449–453.
- 25↑
Ribeiro JM, Rossignol PA, Spielman A, 1984. Role of mosquito saliva in blood vessel location. J Exp Biol 108 :1–7.
- 26↑
Turell MJ, Spielman A, 1992. Nonvascular delivery of Rift Valley fever virus by infected mosquitoes. Am J Trop Med Hyg 47 :190–194.
- 27↑
Turell MJ, Tammariello RF, Spielman A, 1995. Nonvascular delivery of St. Louis encephalitis and Venezuelan equine encephalitis viruses by infected mosquitoes (Diptera: Culicidae) feeding on a vertebrate host. J Med Entomol 32 :563–568.
- 28↑
Mason PW, 1989. Maturation of Japanese encephalitis virus glycoproteins produced by infected mammalian and mosquito cells. Virology 169 :354–364.
- 29↑
Heidner HW, Knott TA, Johnston RE, 1996. Differential processing of sindbis virus glycoprotein PE2 in cultured vertebrate and arthropod cells. J Virol 70 :2069–2073.
- 30↑
Senne DA, Pedersen JC, Hutto DL, Taylor WD, Schmitt BJ, Panigrahy B, 2000. Pathogenicity of West Nile virus in chickens. Avian Dis 44 :642–649.
- 31↑
Swayne DE, Beck JR, Smith CS, Shieh WJ, Zaki SR, 2001. Fatal encephalitis and myocarditis in young domestic geese (Anser anser domesticus) caused by West Nile virus. Emerg Infect Dis 7 :751–753.
- 32↑
Komar N, Langevin S, Hinten S, Nemeth N, Edwards E, Hettler D, Davis B, Bowen R, Bunning M, 2003. Experimental infection of North American birds with the New York 1999 strain of West Nile virus. Emerg Infect Dis 9 :311–322.