North American Culex pipiens mosquitoes are competent for Usutu virus transmission

TL;DR

North American Cx. pipiens mosquitoes can transmit Usutu virus, but they need a high viral dose to become infected.

Contribution

Demonstrates the vector competence of North American Cx. pipiens for Usutu virus under specific viral dose conditions.

Findings

Cx. pipiens showed competence for USUV transmission by day 10, while Cx. tarsalis had limited evidence.

Cx. pipiens transmission increased to 33% by day 21.

Infection threshold for Cx. pipiens was 5 log10 PFU/mL via artificial bloodmeal and 6 log10 PFU/mL via live avian host.

Abstract

Usutu virus (USUV, Flaviviridae) is a mosquito-borne, neuroinvasive flavivirus that is maintained in a transmission cycle between Culex spp. mosquitoes and passerine birds. USUV shares close ecological and antigenic similarity to West Nile virus (WNV). Our study focused on North American Cx. tarsalis and Cx. pipiens, both important in WNV transmission. At day 10, we found that Cx. pipiens were competent for USUV while Cx. tarsalis had limited evidence for transmission. Additionally, Cx. pipiens transmission significantly increased by day 21 to 33%. We found the minimum threshold for infection of Cx. pipiens to be 5 log10 PFU/mL when using an artificial bloodmeal and 6 log10 PFU/mL when using a live avian host. These results provide support that North American Cx. pipiens are competent vectors for USUV but require a high USUV dose for infection. Together, these data provide insight into the potential efficiency of North American Cx. pipiens for USUV transmission.

Figures

Competence of North American Cx. tarsalis mosquitoes for USUV.a Percentage of Cx. tarsalis mosquitoes positive for USUV in the bodies, legs and wings, and saliva out of the number bloodfed following exposure to a bloodmeal containing either USUV Netherlands 2016 (*n* = 28) or Uganda 2012 (*n* = 25). b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit

Competence of North American Cx. pipiens mosquitoes for USUV.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies, legs and wings, and saliva out of the number bloodfed following exposure to a bloodmeal containing either USUV Netherlands 2016 (*n* = 31) or Uganda 2012 (*n* = 40). b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of

Effect of incubation period on the vector competence of North American Cx. pipiens mosquitoes for USUV.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies, legs, and wings, and saliva at 10 (*n* = 31; data taken from Fig. [2](#Fig2)), 14 (*n* = 22), and 21 (*n* = 24) DPI out of number bloodfed following exposure to a bloodmeal containing USUV Netherlands 2016. b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, e

Effect of USUV dose on the vector competence of North American Cx. pipiens mosquitoes using an artificial bloodmeal.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies at 10 DPI out of number bloodfed following exposure to a bloodmeal containing USUV Netherlands 2016 at various doses (log_10_ PFU/mL). Data for the 7.9 log_10_ PFU/mL group is taken from Fig. [2](#Fig2). Infection rate: (# of positive mosquito bodies/total number tested) * 100. b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies at each dose. c USUV titer in the mosquito legs and wings at each dose. d U

Effect of USUV dose on the vector competence of North American Cx. pipiens mosquitoes using domestic canaries.a Viremia (log_10_ PFU/mL) of domestic canaries (whole-day timepoints: *n* = 10, half-day timepoints: *n* = 13) inoculated with 1500 PFU of USUV Netherlands 2016 and sampled every 12 h. b Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies at 10 DPI out of number bloodfed (*n* = 86) following exposure to canaries infected with USUV Netherlands 2016. c USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. d USUV titer in the mosquito legs and wings. e USUV titer in

Full text

Introduction

Usutu virus (USUV) is an emerging zoonotic flavivirus that is maintained in an enzootic cycle primarily between Culex spp. mosquitoes and passerine birds^1,2^. USUV causes multisystemic and neuroinvasive disease in birds. In humans, USUV infection most often results in mild febrile illness, but in some cases can progress to severe neuroinvasive illness (i.e., meningitis and encephalitis). However, humans and other mammals are considered “dead-end” hosts.

USUV was first isolated in 1959 from a Culex neavei mosquito in South Africa^3^. It then spread throughout sub-Saharan Africa and emerged in Europe in 1996 where it has undergone at least three separate introductions, likely by migratory birds^4,5^. USUV is now endemic in Europe where it has caused disease amongst bird populations; most notable is the mass mortality of the Eurasian blackbird (Turdus merula), a passerine^6–9^. As a member of the Japanese encephalitis virus (JEV) serocomplex, USUV shares antigenic and transmission cycle similarities with other important viruses, such as West Nile virus (WNV)^10^. WNV and USUV geographically overlap and currently co-circulate in Europe^11,12^.

For viruses in the JEV serocomplex, Culex spp. mosquitoes are considered to be the primary vectors responsible for transmission^10,13^. For USUV, several Culex spp. mosquito populations in Africa, including Cx. neavei, Cx. univitattus, Cx. quinquefasciatus, and Cx. perfuscus, have been found to be naturally infected^14–16^. Since the emergence of USUV in Europe, it has been detected in several Culex species, primarily Cx. pipiens, which is considered the main vector of both USUV and WNV in Europe^17–21^. Several studies have assessed vector competence, by measurement of infectious virus in the saliva of European populations of Culex spp. mosquitoes for USUV^22–26^. In particular, European colonies of Cx. pipiens were able to transmit infectious virus in their saliva, ranging from 4-69% of bloodfed mosquitoes, at DPI 14 when fed on bloodmeals containing European USUV isolates^22,23,25^.

As seen with WNV, which is now present on every continent except Antarctica^27^, the presence of competent hosts and vectors allows for rapid spread of a virus to non-endemic regions. In North America, one study found that Cx. pipiens and Cx. quinquefasciatus exposed to an artificial infectious bloodmeal spiked with USUV South Africa 1959 isolate had viral RNA in their saliva, but infectious virus was not measured^28^. Another study found that a small percentage of North American Cx. quinquefasciatus mosquitoes expectorated infectious USUV in their saliva following exposure to an artificial bloodmeal containing USUV Netherlands 2016 isolate^29^. In another study, North American Cx. tarsalis mosquitoes fed on a bloodmeal containing USUV Uganda 2012 isolate had limited transmission at a high dose; however, infectious virus in the saliva was not measured^30^. While these findings have shed light on the vector competence of North American Culex spp. mosquitoes for USUV, detection of infectious virus in the saliva was not considered for Cx. pipiens and Cx. tarsalis, two species important in WNV transmission.

While USUV has yet to be found in North America, it is important to understand if competent vectors exist that could aid in the maintenance of USUV should it emerge. The aim of this study was to assess the vector competence of two North American Culex spp. mosquitoes, Cx. tarsalis and Cx. pipiens, for USUV. We found limited evidence for transmission of USUV in Cx. tarsalis, while Cx. pipiens were competent and able to transmit two strains of USUV. Additionally, we found that extending the EIP (extrinsic incubation period), or the time from ingestion of an infectious bloodmeal to transmission of virus in the saliva, led to significant increases in dissemination and transmission rates of USUV in Cx. pipiens. We found that there is a minimum amount of USUV needed to infect North American Cx. pipiens mosquitoes, and Cx. pipiens were able to become infected from USUV Netherlands 2016-infected birds. The experiments in this study provide important information on North American Culex spp. mosquitoes that could vector USUV and the transmission dynamics within this species.

Results

To assess the vector competence of Cx. tarsalis for USUV, mosquitoes were fed on artificial infectious bloodmeals spiked with either 7.6 or 8.1 log_10_ PFU/mL of Netherlands 2016 or Uganda 2012, respectively, and viral titers in bodies, legs/wings, and saliva were measured. Bodies represent infection, legs/wings represent dissemination, and saliva represents transmission. At 10 DPI, 14/28 (50%, 95% confidence interval (CI): 32.6–67.4%) mosquitoes were infected for Netherlands 2016 and 7/25 (28%, 95% CI: 14.1–47.8%) mosquitoes were infected for Uganda 2012; no significant differences were observed between USUV isolates for infection rate or titers in the body homogenates (Fig. 1a, b). Both USUV isolates disseminated at similar rates and titers, with 9/28 (32.1%, 95% CI: 17.8–50.8%) of the Netherlands 2016 group and 6/25 (24%, 95% CI: 11.2–43.8%) of the Uganda 2012 group having infectious virus in their legs and wings (Fig. 1a, c). Infectious virus was detected in the saliva of 2/28 individuals (7.1%, 95% CI: 0.9–23.7%) for the Netherlands 2016 group; however, 0/25 (0%, 95% CI: 0–15.8%) of individuals had detectable infectious virus in their saliva for the Uganda 2012 group (Fig. 1a, d). Together, these data indicate that Cx. tarsalis mosquitoes are susceptible to USUV, with limited evidence for transmission.Fig. 1. Competence of North American Cx. tarsalis mosquitoes for USUV.a Percentage of Cx. tarsalis mosquitoes positive for USUV in the bodies, legs and wings, and saliva out of the number bloodfed following exposure to a bloodmeal containing either USUV Netherlands 2016 (n = 28) or Uganda 2012 (n = 25). b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of detection (bodies: 1.4 log_10_ PFU/mosquito, legs and wings: 1.2 log_10_ PFU/mosquito, saliva: 0.4 log_10_ PFU/mosquito). Negative samples are graphed at half the limit of detection.

Next, we evaluated the vector competence of Cx. pipiens for USUV Netherlands 2016 and Uganda 2012. Mosquitoes were fed on artificial infectious bloodmeals spiked with either 7.9 or 8.0 log_10_ PFU/mL of Netherlands 2016 or Uganda 2012, respectively. At 10 DPI, 29/31 (93.5%, 95% CI: 78.3–99.2%) of Netherlands 2016-exposed and 32/40 (80%, 95% CI: 65.0–89.8%) of Uganda 2012-exposed mosquitoes were infected (Fig. 2a). Similar titers were observed for both isolates in the body homogenates (Fig. 2b). Both USUV isolates disseminated at similar rates, with 18/31 (58.1%, 95% CI: 40.7–73.6%) of the Netherlands 2016 group and 17/40 (42.5%, 95% CI: 28.5–57.8%) of the Uganda 2012 group having infectious virus in their legs and wings at comparable titers (Fig. 2a, c). Additionally, for both USUV isolates, the transmission rates were very similar [Netherlands 2016: 3/31 (9.7%, 95% CI: 2.6–25.7%); Uganda 2012: 4/40 (10%, 95% CI: 3.4-23.6%)] with no significant differences in the amount of infectious virus in the saliva between the two isolates (Fig. 2a, d). From these data, we can conclude that Cx. pipiens mosquitoes are competent for USUV Netherlands 2016 and Uganda 2012. Due to the more robust transmission rates for Cx. pipiens, we moved forward with this species in additional studies.Fig. 2. Competence of North American Cx. pipiens mosquitoes for USUV.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies, legs and wings, and saliva out of the number bloodfed following exposure to a bloodmeal containing either USUV Netherlands 2016 (n = 31) or Uganda 2012 (n = 40). b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of detection (bodies: 1.4 log_10_ PFU/mosquito, legs and wings: 1.2 log_10_ PFU/mosquito, saliva: 0.4 log_10_ PFU/mosquito). Negative samples are graphed at half the limit of detection.

Cx. pipiens vector competence at 14 and 21 DPI

To further assess vector competence of Cx. pipiens over time, we extended the EIP to 14 and 21 DPI using artificial infectious bloodmeals containing 8.2–8.3 log_10_ PFU/mL of USUV Netherlands 2016. When compared to the infection rate of Cx. pipiens at 10 DPI for Netherlands 2016 (data from Fig. 2), 14 and 21 DPI had comparable infection rates [14 DPI: 22/22 (100%, 95% CI: 82.5–100%); 21 DPI: 23/24 (95.8%, 95% CI: 78.1− > 99.9%)] with significantly higher titers in the body homogenates at 21 DPI compared to 14 DPI (P = 0.0002, Fig. 3a, b). A trend towards higher titers in the body homogenates at 10 DPI compared to 14 DPI (P = 0.0556) and 21 DPI compared to 10 DPI (P = 0.0868) was observed (Fig. 3b). For dissemination, 22/24 (91.7%, 95% CI: 73–98.8%) mosquitoes had infectious virus in their legs and wings at 21 DPI which was significantly higher than the 18/31 (58.1%, 95% CI: 40.7–73.6%) at 10 (P = 0.0065) and 14/22 (63.6%, 95% CI: 42.9–80.4%) at 14 DPI (P = 0.0323, Fig. 3a). Significantly higher titers were observed in the legs and wings at 21 DPI compared to 10 (P = 0.0278) and 14 DPI (P = 0.0217) (Fig. 3c). For transmission, 3/22 (13.6%, 95% CI: 3.9–34.2%) and 8/24 (33.3%, 95% CI: 17.8–53.4%) mosquitoes at 14 and 21 DPI had infectious virus in their saliva, respectively (Fig. 3a). The transmission rate at 21 DPI was significantly higher compared to 3/31 (9.7%, 95% CI: 2.6–25.7%) at 10 DPI (P = 0.0428, Fig. 3a). A trend towards increased titers at 21 DPI compared to 10 DPI was observed for the saliva (P = 0.0704, Fig. 3d). Together, this data indicates that extending the EIP leads to increased rates of dissemination and transmission, as well as titer in the bodies and legs and wings, of USUV for Cx. pipiens.Fig. 3. Effect of incubation period on the vector competence of North American Cx. pipiens mosquitoes for USUV.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies, legs, and wings, and saliva at 10 (n = 31; data taken from Fig. 2), 14 (n = 22), and 21 (n = 24) DPI out of number bloodfed following exposure to a bloodmeal containing USUV Netherlands 2016. b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. c USUV titer in the mosquito legs and wings. d USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of detection (bodies: 1.4 log_10_ PFU/mosquito, legs and wings: 1.2 log_10_ PFU/mosquito, saliva: 0.4 log_10_ PFU/mosquito). Negative samples are graphed at half the limit of detection. ***P < 0.001, **P < 0.01, *P < 0.05.

USUV minimum dose required for infection of Cx. pipiens

To understand the minimum threshold of USUV needed to infect Cx. pipiens mosquitoes, we exposed mosquitoes to artificial infectious bloodmeals spiked with a range of doses of USUV Netherlands 2016 (Table 1). Netherlands 2016 data from Fig. 2 is repeated in Fig. 4 (7.9 log_10_ PFU/mL) for comparison. At 3.5, 4.2, and 5.4 log_10_ PFU/mL doses, 0% of mosquitoes were infected on 10 DPI (Fig. 4a). At a 5.3, 6.3, 6.4, 6.5, and 7.4 log_10_ PFU/mL dose, 1/17 (5.9%), 5/30 (16.7%), 11/30 (36.7%), 5/17 (29.4%), and 19/26 (73.1%) mosquitoes were infected, respectively (Fig. 4a). Linear regression analysis showed a positive relationship between USUV dose in the artificial bloodmeal and mosquito infection rate (P = 0.0020, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$y=20.98x-94.95$$\end{document} ) with a minimum dose of 5 log_10_ PFU/mL needed to infect Cx. pipiens (Fig. 4a). A trend towards increased titers in the body homogenates for the 7.9 dose compared to the 6.5 dose was observed (P = 0.0800, Fig. 4b). For dissemination, at a 5.3, 6.3, 6.4, 6.5, and 7.4 log_10_ PFU/mL dose, 1/17 (5.9%), 1/30 (3.3%), 4/30 (13.3%), 2/17 (11.8%), and 10/26 (38.5%) mosquitoes had infectious virus in the legs and wings, respectively. No significant differences were observed for USUV titer in the legs and wings between doses (Fig. 4c). For transmission, infectious virus was detected in the saliva of one mosquito for the 7.4 log_10_ PFU/mL dose (1/26, 3.8%) with a titer of 1.4 log_10_ PFU/mosquito; however, 0% of individuals were transmitting infectious virus in their saliva for the 5.3, 6.3, 6.4, and 6.5 log_10_ PFU/mL doses (Fig. 4d). Together, this data indicates that at least 5 log_10_ PFU/mL USUV is necessary to efficiently infect North American Cx. pipiens after exposure to an artificial bloodmeal.Fig. 4. Effect of USUV dose on the vector competence of North American Cx. pipiens mosquitoes using an artificial bloodmeal.a Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies at 10 DPI out of number bloodfed following exposure to a bloodmeal containing USUV Netherlands 2016 at various doses (log_10_ PFU/mL). Data for the 7.9 log_10_ PFU/mL group is taken from Fig. 2. Infection rate: (# of positive mosquito bodies/total number tested) * 100. b USUV titer (log_10_ PFU/mosquito) in the mosquito bodies at each dose. c USUV titer in the mosquito legs and wings at each dose. d USUV titer in the mosquito saliva at each dose. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of detection (bodies: 0.4 log_10_ PFU/mosquito, legs and wings: 1.2 log_10_ PFU/mosquito, saliva: 0.4 log_10_ PFU/mosquito). Negative samples are graphed at half the limit of detection.Table. 1Bloodmeal titers for Cx. pipiens dosing study using an artificial infectious bloodmealMosquito SpeciesVirus IsolateDay Post-InoculationTotal TestedBloodmeal Titer in log_10_ PFU/mLCx. pipiensNetherlands 201610n = 267.4n = 176.5n = 306.4n = 306.3n = 225.4n = 175.3n = 364.2n = 323.5

To assess the minimum USUV dose needed to infect Cx. pipiens after feeding on a passerine bird, we then fed uninfected Cx. pipiens mosquitoes on infected canaries on 2.5–3 DPI to encompass a range of viremias (Fig. 5a). The mean peak viremia in canaries was 6.5 log_10_ PFU/mL on DPI 3. The canaries used for the mosquito feed had viremias from 3.0 to 7.8 log_10_ PFU/mL. Of the mosquitoes that fed on canaries, 0–28% became infected across all canaries with a minimum dose of 6.0 log_10_ PFU/mL needed for infection and a mean USUV titer in the body homogenates of 5.6 log_10_ PFU/mosquito (Figs. 5b, c). However, linear regression analysis showed no correlation between Cx. pipiens infection rate and USUV dose in the avian host (Fig. 5b). Out of the bloodfed mosquitoes, 0–22% had infectious virus in the legs and wings (dissemination) with a mean titer of 2.7 log_10_ PFU/mosquito (Fig. 5d). Zero individuals had detectable infectious virus in their saliva (transmission) (Fig. 5e). Together, this data indicates that at least 6 log_10_ PFU/mL USUV is necessary to infect North American Cx. pipiens after feeding on an infected canary, though infection probability is non-linear.Fig. 5. Effect of USUV dose on the vector competence of North American Cx. pipiens mosquitoes using domestic canaries.a Viremia (log_10_ PFU/mL) of domestic canaries (whole-day timepoints: n = 10, half-day timepoints: n = 13) inoculated with 1500 PFU of USUV Netherlands 2016 and sampled every 12 h. b Percentage of Cx. pipiens mosquitoes positive for USUV in the bodies at 10 DPI out of number bloodfed (n = 86) following exposure to canaries infected with USUV Netherlands 2016. c USUV titer (log_10_ PFU/mosquito) in the mosquito bodies. d USUV titer in the mosquito legs and wings. e USUV titer in the mosquito saliva. Circles represent individual samples, horizontal lines represent the mean, error bars represent the standard deviation, and dashed lines represent the limit of detection (serum: 2.0 log_10_ PFU/mL, mosquito bodies: 1.4 log_10_ PFU/mosquito, mosquito legs and wings: 1.2 log_10_ PFU/mosquito, mosquito saliva: 0.4 log_10_ PFU/mosquito). Negative samples are graphed at half the limit of detection.

Discussion

Here, we showed that North American Cx. pipiens mosquitoes are competent vectors for USUV, with limited evidence for USUV transmission by Cx. tarsalis. Additionally, we found that USUV transmission rates of Cx. pipiens increased with a longer EIP. Finally, we showed that for infection of Cx. pipiens mosquitoes, a minimum dose of 5 log_10_ PFU/mL of USUV Netherlands 2016 is needed when fed on an artificial bloodmeal, while a minimum dose of 6 log_10_ PFU/mL is needed when fed on a USUV-infected bird.

Prior to our study, assessment of North American Cx. pipiens or Cx. tarsalis vector competence by measurement of infectious virus in the saliva had not been studied, which is essential to understand the potential for a vector to be able to transmit to a host. However, our results on the vector competence of North American Culex. spp. are similar to previous vector competence studies that measured USUV RNA in saliva. In a previous study, Cx. tarsalis mosquitoes exposed to a bloodmeal containing 7.2-7.5 log_10_ PFU/mL of USUV Uganda 2012 had low transmission rates of 8.9% [re-calculated with denominator as total exposed] on 13/14 DPI; however, viral RNA, not infectious virus, was measured^30^. Nonetheless, this study is in line with our results showing that 0-7% of Cx. tarsalis transmitted infectious virus in their saliva on 10 DPI. For Cx. pipiens, we found that 10% of mosquitoes are capable of transmitting infectious USUV on 10 DPI which is comparable to a previous study where 13.8% of mosquitoes [re-calculated with denominator as total exposed] had USUV RNA in their saliva on 14 DPI following exposure to a bloodmeal containing 7.4 log_10_ PFU/mL of the prototype isolate, South Africa 1959; however, infectious virus was not measured^28^. It is important to note that our studies used limited sample sizes, and experimental vector competence studies are performed under controlled conditions, without the complexity of real-world extrinsic factors, which may overestimate transmission potential. Nevertheless, our results suggest that North American Cx. pipiens mosquitoes can transmit USUV, with limited evidence for USUV transmission in Cx. tarsalis.

Vector competence data for European colonies of Cx. pipiens, the main vector for USUV and WNV in Europe^11,20,22^ suggests vector competence may be higher than in our North American colony. In these studies, 4–69% of Cx. pipiens transmitted infectious virus in their saliva after exposure to bloodmeals containing ~7 log_10_ PFU/mL of USUV^22,23,25^. The differences in vector competence between North American and European colonies support what is known about differences in WNV vector competence between populations of the same mosquito species^31–33^.

USUV isolates can be separated into 8 different lineages^34^. The isolates used in our study, Netherlands 2016 and Uganda 2012, belong to lineages Africa 3 and Europe 5, respectively, and represent contemporary isolates from two geographically distinct areas. Limited data exists on differences in vector competence for USUV isolates belonging to different lineages; however, one study found significant differences in transmission of a European colony of Cx. pipiens after exposure to bloodmeals containing either USUV from the Europe 2 lineage or the Europe 3 lineage^35^. While we did not find differences in vector competence between the lineages used in our studies, further experiments would be necessary to determine if other USUV lineages result in variability in vector competence.

The extrinsic incubation period (EIP), or the time from ingestion of an infectious bloodmeal to transmission of virus in the saliva, is an important variable in vectorial capacity^36,37^. In our study, we found that USUV dissemination and transmission rates, as well as titer in the bodies and legs and wings of North American Cx. pipiens, significantly increased with EIP, and detection of infectious virus in the saliva occurred as early as 10 DPI with a transmission rate of 10%. On average, female Cx. pipiens mosquitoes live for 23 days at 28°C^38^. Thus, North American Cx. pipiens have the potential to be effective vectors of USUV at this temperature and dose. In comparison, a Swedish colony of Cx. pipiens was found to have little to no change in dissemination and transmission rates across 7, 14, 21, and 28 DPI when fed on a bloodmeal containing 6.3 log_10_ PFU/mL of USUV Bologna 2009 isolate; however, these mosquitoes were housed at 25°C and infectious virus was not measured^39^. For WNV, several studies have investigated the EIP in Cx. pipiens mosquitoes. North American Cx. pipiens transmitted infectious virus in their saliva as early as 5–8 DPI, and transmission rates increased with the EIP after exposure to bloodmeals containing 6.4-7.3 log_10_ PFU/mL of WNV and held at 26–27°C^40,41^. Additionally, it has been shown that with an increase in temperature, EIP shortens for Culex mosquitoes and WNV^42,43^. Thus, we have identified that North American Cx. pipiens are competent vectors for USUV as early as 10 DPI and that transmission increases with time. Further studies investigating the effect of temperature on EIP for Culex spp. mosquitoes and USUV are warranted.

We found that the USUV minimum infectious threshold for North American Cx. pipiens is 5 log_10_ PFU/mL when fed on an artificial bloodmeal and 6 log_10_ PFU/mL when fed on live birds. Based on our results, we predict that house sparrows would not be competent reservoirs for transmission of USUV to Cx. pipiens as they reach mean peak titers less than 6 log_10_ PFU/mL^29^. While Agliani et al. did not quantify infectious virus in the blood of USUV-infected Eurasian blackbirds, we estimate, based on Ct values of <20, that these birds would reach sufficient titers to infect Cx. pipiens^44^. Here, we chose to use canaries, which are a passerine model previously developed for WNV and USUV^45,46^, due to practical reasons, as they are commercially available. Using a wild-caught avian species may result in different rates of transmission to mosquitoes. For example, a previous study using a North American colony of Cx. quinquefasciatus mosquitoes found the minimum threshold for USUV to be 3-4 log_10_ PFU/mL after feeding on house sparrows, which is lower than our results with Cx. pipiens^29^. Differences in the host species between these studies cannot be overlooked, as previous work with WNV demonstrated that infection rates differed for Cx. pipiens fed on different bird species even at equivalent viral titers^47^. We did observe this here as well, as the infection rates for our artificial dosing study using sheep blood were significantly higher than for Cx. pipiens fed on USUV-infected canaries at comparable doses, suggesting that differences in the blood source may be impacting mosquito infection rates in our study and thus explain the difference seen in the minimum infectious threshold between the two systems. Differences in mosquito infection rates when fed on different blood sources containing similar virus concentrations have been shown in several studies, through an artificial system or live avian host, for flaviviruses and Cx. pipiens^47,48^. It is unknown what the potential effects of host physiology are on virus availability, which is an area that warrants further investigation for fully understanding enzootic transmission.

Our results indicate that North American Cx. pipiens are competent vectors of USUV. However, the high viremia required for infection of Cx. pipiens may decrease the overall transmission potential of this species. Nevertheless, further research on the USUV transmission rates for different combinations of Culex spp. mosquitoes and bird species would be important to fully understand enzootic transmission dynamics of USUV.

Methods

Cell lines and virus stocks

Vero cells were maintained at 37°C and 5% CO_2_. Dulbecco’s Modified Eagle Medium (DMEM, Corning) supplemented with 5% fetal bovine serum (FBS, VWR International) and 1% penicillin-streptomycin (Gibco) was used to culture cells.

Throughout these studies, USUV isolates TMNetherlands (Netherlands 2016, Africa 3 lineage, GenBank MN813490, passage 5–6, isolated from Turdus merula)^49^ and UG09615 (Uganda 2012, Europe 5 lineage, GenBank MN813491, passage 4, isolated from Culex univittatus)^15^, both sequenced by our laboratory^50^, were used. WNV NY99ic was used for plaque reduction neutralization test (PRNT) assays^51^.

Mosquito rearing

The Cx. tarsalis mosquitoes used in these studies were obtained from BEI Resources and are derived from a colony isolated in Yolo County, California^52^. Adult mosquitoes and larvae were reared either in an environmental chamber (16:8 light:dark cycle, 26°C, and 60–70% relative humidity) or a warm room (14:10 light:dark cycle, 28°C, and 70–80% relative humidity). Larvae were maintained on Hikari Tropical First Bites fish food (Hikari Sales USA, Inc.), and adult mosquitoes were maintained on 10% sucrose ad libitum or defibrinated sheep blood (HemoStat Laboratories) for egg collection.

The Cx. pipiens biotype molestus mosquitoes used in these studies were obtained as eggs from the CDC and are derived from Chicago, IL. Adult mosquitoes and larvae were reared in a warm room at a 14:10 light:dark cycle, 28°C, and 70–80% relative humidity. The larvae were fed Tetramin tropical fish food (Tetra), and adult mosquitoes were maintained on 10% sucrose ad libitum or defibrinated sheep or chicken blood (Rockland Immunochemicals) for egg collection.

Cx. tarsalis artificial infectious feed

Adult Cx. tarsalis mosquitoes, aged 4–9 days post-eclosion, were sorted into cartons and starved of sucrose overnight. Mosquitoes were exposed to a 3 mL bloodmeal containing defibrinated sheep blood and 7.6 log_10_ PFU/mL of USUV Netherlands 2016 isolate or 8.1 log_10_ PFU/mL of USUV Uganda 2012 isolate supplemented with 10% sucrose using a Hemotek feeder with hog casing as the membrane. Blood fed females were sorted and kept at a 16:8 light:dark cycle, 26°C, and 60–70% relative humidity with 6% sucrose ad libitum. The temperature used is consistent with native temperatures reached in the summer in Yolo County, California^53^ where the colony is derived from. At 10 days post-inoculation (DPI), mosquitoes were cold-anesthetized and dissected. Legs and wings were collected in 300 µL of mosquito diluent (RPMI 1640 media with L-Glutamine [Gibco], 25 mM HEPES [Gibco], 2% FBS, 50 µg/mL gentamicin [Genesee Scientific], and 2.5 µg/mL amphotericin B [Gibco]). The mosquito proboscis was inserted into a trimmed pipette tip containing 10 µL of immersion oil. Mosquitoes were force salivated for 30–60 min upon which mosquito bodies were collected in 500 µL of mosquito diluent, and salivary secretions were expelled into 50 µL of mosquito diluent. Samples were stored at –80°C until further analysis. Two independent experiments were performed.

Cx. pipiens artificial infectious feeds

Adult Cx. pipiens mosquitoes, aged 5–8 days post-eclosion, were sorted into cartons and starved of sucrose overnight before being exposed to a 3 mL bloodmeal containing defibrinated sheep blood and 3.5–8.3 log_10_ PFU/mL of USUV Netherlands 2016 isolate or 8.0 log_10_ PFU/mL of USUV Uganda 2012 isolate supplemented with 10% sucrose. A Hemotek feeder with hog casing as the membrane was used for the feeds. Blood fed females were sorted and kept at a 12:12 light:dark cycle, 28°C, and 70–80% relative humidity with 6% sucrose ad libitum. The temperature used is consistent with native temperatures reached in the summer in Chicago, Illinois^53^ where the colony is derived from. On 10, 14, or 21 DPI, mosquitoes were cold-anesthetized and dissected. Legs and wings were collected in 300 µL of mosquito diluent, and the mosquito was force salivated as described above. Mosquito bodies were collected in 500 µL of mosquito diluent, and salivary secretions were expelled into 50 µL of mosquito diluent. Samples were stored at –80°C until further analysis. One to three independent experiments were performed for each bloodmeal dose.

Canary inoculations

Adult domestic canaries (Serinus canaria forma domestica) of mixed sex and age were obtained from a breeder. Birds were housed in flight cages (dimensions: 30”L × 18”W × 18”H) containing 4–7 individuals and provided commercial feed and fresh water ad libitum and acclimated for 3 days. An initial blood sample was collected prior to inoculation to assess for previous WNV exposure via PRNT assay. Birds were separated into two groups and inoculated 12 h apart with 1500 PFU of USUV Netherlands 2016 isolate to increase practical identifiability in downstream models^54^. A blood sample was collected every day post-inoculation for 6 days. Birds were monitored daily for signs of clinical disease such as lethargy, lack of responsiveness, or puffy feathers, and were euthanized when reaching humane endpoints. On 13 DPI, a final blood sample was collected, and the remaining birds were euthanized. Blood samples were centrifuged to separate serum from whole blood and then stored at -80°C until further testing. Two independent experiments were performed. Birds were monitored daily throughout the duration of the studies by research personnel and animal care staff.

Cx. pipiens feed on USUV-infected canaries

Cx. pipiens mosquitoes, 5–8 days post-eclosion, were cold-anesthetized and sorted into mesh-covered cartons. Mosquitoes were starved of sucrose overnight before exposure to USUV-infected canaries. On 3 DPI, a subset of infected canaries above (n = 7) was used for the mosquito feed. Each bird was assigned a carton of mosquitoes and gently restrained against the mesh for approximately 30 min to allow mosquitoes to feed^29^. Mosquitoes were cold-anesthetized, and blood-fed females were sorted out and returned to the carton. For the remainder of the study, mosquitoes were maintained under the conditions described above. On day 10 post-exposure to USUV-infected canaries, mosquitoes were cold-anesthetized and dissected. Legs and wings, bodies, and salivary secretions were collected as described above. Samples were stored at -80°C until further analysis.

Real-time RT-PCR

Mosquito body homogenates were tested by real-time quantitative reverse transcription PCR (RT-qPCR) targeting a conserved region of the USUV non-structural gene 5 (NS5) as described previously^55^. RT-qPCR was performed using the iTaq Universal Probes One-Step Kit (Bio-Rad Laboratories) and USUV primers (forward: 5′ CAAAGCTGGACAGACATCCCTTAC 3′, reverse: 5′ CGTAGATGTTTTCAGCCCACGT 3′) and probe (5′ FAM-AAGACATATGGTGTGGAAGCCTGATAGGCA 3′). To generate an RNA standard curve, a 2 kb fragment of USUV Uganda 2012 containing the primer binding regions in NS5 was cloned into the pCR4-TOPO TA vector. The plasmid containing USUV NS5 was linearized and in vitro transcribed using the Ampliscribe T7 kit. The RNA was quantified and serially diluted 1:10 to create a USUV RNA standard for use as a positive control. Mosquito body homogenates were diluted 1:10 in nuclease-free water, and 4 µL of each sample was added to the RT-qPCR reaction for a total volume of 20 µL. Cycling parameters were as follows: 50°C for 30 min, 95°C for 15 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min (Bio-Rad CFX Connect Real-Time System). The limit of detection for RNA samples was 10 RNA copies/sample.

Plaque assays

Bird serum and mosquito samples were titrated by Vero cell plaque assay to quantify infectious virus. Mosquito bodies, legs, and wings were homogenized at 50 oscillations per second for 3 min and clarified by centrifugation at 18,000 rpm for 3 min. Briefly, samples were serially diluted 1:10 in BA-1 diluent, and 200 µL of each dilution was added to confluent Vero cells. Inoculated plates were incubated at 37°C, 5% CO_2_ for 1 h and gently rocked every 15 min. Following incubation, an 0.8% agarose overlay containing Ye-Lah media and 3% sodium bicarbonate (Gibco) was added to each well. A second overlay, containing 4% neutral red (Millipore Sigma), was added two days later for staining. Plaques were counted the following day. The limits of detection were 2.0 log_10_ PFU/mL for serum samples, 0.4 or 1.4 log_10_ PFU/mosquito for mosquito bodies, 1.2 log_10_ PFU/mosquito for mosquito legs and wings, and 0.4 log_10_ PFU/mosquito for mosquito salivary secretions. Mosquito infection was defined as body homogenates positive by plaque assay. Mosquito dissemination was defined as legs and wings homogenates positive by plaque assay. Mosquito transmission was defined as salivary secretions positive by plaque assay.

Plaque reduction neutralization test

Blood samples collected from canaries prior to inoculation were screened for WNV neutralization by PRNT. Serum was diluted and heat-inactivated at 56°C for 30 min. Heat-inactivated samples were then mixed with ~100 PFU of WNV NY99ic and incubated at 37°C for 1 h. After incubation, samples were tested by Vero cell plaque assay to determine neutralization, which was defined as a reduction in plaque formation by at least 90%^56^.

Statistical analysis

All data were analyzed and graphed using GraphPad Prism 10 (GraphPad Software, San Diego, CA). Mosquito viral titers for bodies, legs, and wings, and saliva were compared using a Mann-Whitney test or an ordinary one-way ANOVA with Tukey’s multiple comparisons test. Mosquito infection, dissemination, and transmission rates were compared using Fisher’s exact test, and confidence intervals were calculated using the Wald method.

Ethics statement

All experiments were conducted in accordance with the Virginia Tech Institutional Animal Care and Use Committee (IACUC #21-048 and #24-063).

Supplementary information

Supplementary Tables