Transmission Incompetence of Culex quinquefasciatus and Culex pipiens pipiens from North America for Zika Virus

In late 2014, Zika virus (ZIKV; Flaviviridae, Flavivirus) emerged as a significant arboviral disease threat in the Western hemisphere. Aedes aegypti and Aedes albopictus have been considered the principal vectors of ZIKV in the New World due to viral isolation frequency and vector competence assessments. Limited reports of Culex transmission potential have highlighted the need for additional vector competence assessments of North American Culex species. Accordingly, North American Culex pipiens and Culex quinquefasciatus were orally exposed and intrathoracically inoculated with the African prototype ZIKV strain and currently circulating Asian lineage ZIKV strains to assess infection, dissemination, and transmission potential. Results indicated that these two North American Culex mosquito species were highly refractory to oral infection with no dissemination or transmission observed with any ZIKV strains assessed. Furthermore, both Culex mosquito species intrathoracically inoculated with either Asian or African lineage ZIKVs failed to expectorate virus in saliva. These in vivo results were further supported by the observation that multiple mosquito cell lines of Culex species origin demonstrated significant growth restriction of ZIKV strains compared with Aedes-derived cell lines. In summation, no evidence for the potential of Cx. pipiens or Cx. quinquefasciatus to serve as a competent vector for ZIKV transmission in North America was observed.


INTRODUCTION
Zika virus (ZIKV), a member of the genus Flavivirus, has recently surfaced as a significant threat to global public health despite only having been found in the Western hemisphere since late 2014. The first ZIKV isolates were made from a sentinel Rhesus monkey and subsequently from arboreal Aedes africanus mosquitoes, a suspected sylvatic cycle vector, 1 in Uganda in 1947 and 1948. 2 The first described clinical case was identified in Nigeria in 1954 in which the individual exhibited symptoms similar to many arthralgic arboviruses including fever, headache, diffuse joint pain, and slight jaundice. 3 Serological evidence of ZIKV infection was subsequently identified in Nigeria, Egypt, India, Malaysia, Indonesia, Philippines, Thailand, Vietnam, and Senegal. [4][5][6][7][8][9][10][11] Phylogenetic analysis demonstrates three distinct ZIKV genotypes corresponding to the initial emergence in east Africa and subsequent spread to west Africa and Asia. 12 ZIKV was associated with sporadic limited human disease prior to 2007 when ZIKV caused outbreaks in Yap and Gabon. 13,14 By 2013, ZIKV had spread to French Polynesia and other islands of the South Pacific including New Caledonia, the Cook Islands, and Easter Island prior to its first detection in the Western hemisphere in 2014. 15 First detected in the peridomestic vector, Aedes aegypti, in Malaysia in 1966, 16 subsequent vector competence and ecological studies have established ZIKV to be vectored primarily by members of the genus Aedes in both sylvatic and urban cycles through field studies. 2,[16][17][18][19][20][21][22][23][24] However, recent ZIKV detection in Culex mosquitoes 25,26 and reports of experimental evidence of transmission of ZIKV by Culex quinquefasciatus, 26,27 have raised public concerns that mosquitoes of the Culex pipiens complex could play a role in transmission as well. Vector competence by mosquitoes of alternative genera would have possible implications for many regions of North America where the known Aedes vectors, Ae. aegypti and Aedes albopictus, are absent and a significant number of Culex species are present. The utilization of both Aedes and Culex mosquitoes as vectors would alter the discourse and strategy for control efforts. Surveillance and control of Culex mosquitoes requires utilization of different tools and greatly increases the area of surveillance based on the wide geographic range of potential Culex vectors.
There have been a number of recent studies directly evaluating the vector competence of Culex mosquitoes for circulating strains of ZIKV and the results have been mixed. Two studies using Cx. pipiens and Cx. quinquefasciatus from North America showed a lack of ZIKV infection, dissemination, or saliva infection after oral exposure by artificial or viremic mice as blood meal sources. 28,29 A study from Brazil showed a low level of experimental infection in Cx. quinquefasciatus after artificial blood meal exposure, but no dissemination from the midgut or transmission. 30 An Italian study indicated a lack of infection in Cx. pipiens mosquitoes after oral exposure, 31 whereas a second study in Europe examining ZIKV oral infection of Cx. pipiens and Cx. quinquefasciatus found evidence of infection and low rates of dissemination but no transmission. 32 The findings from experiments described above are in direct contrast with two reports of highly efficient transmission potential from Brazil 26 and China. 27 In an attempt to rule out the role of Culex mosquitoes as possible ZIKV vectors in North America, we evaluated Cx. quinquefasciatus and Cx. pipiens mosquitoes for susceptibility to infection after oral and intrathoracic (IT) exposure to recent Asian genotype ZIKV isolates from Puerto Rico and Honduras as well as the original African lineage isolate from Uganda. Additionally, we compared the growth capacity of each virus in model Aedes and Culex cell lines to highlight the lack of replication competence of ZIKV for in vitro and in vivo Culex models.
*Address correspondence to Joan L. Kenney, Centers for Disease Control and Prevention, Fort Collins, CO 80521. E-mail: vwx1@ cdc.gov †These authors contributed equally to this work.

MATERIALS AND METHODS
Viruses and mosquitoes. ZIKV isolates MR766 (Uganda 1947), PRVABC59 (Puerto Rico 2015), and R103451 (Honduras 2016) were used for mosquito infections in this study. The MR766 strain had a history of 149 passages in unknown sources, including suckling mouse brain, and two known passages in African green monkey kidney (Vero) epithelial cells. PRVABC59 was isolated from the serum of a febrile traveler returning to the continental United States from Puerto Rico in 2015 and was passaged three times in Vero cells. 33 The R103451 isolate was derived from a human placenta from a patient who had traveled to Honduras in 2015 and passaged in C6/36 mosquito cells once. Colonized Cx. quinquefasciatus (Sebring) and Cx. pipiens pipiens (Chicago) were used for in vivo exposure experiments. The colony of Cx. quinquefasciatus was originally established in Florida in 1988 and has since been maintained at the Centers for Disease Control and Prevention in Fort Collins, CO, since 2004. 34 The original Cx. pipiens were collected in Chicago as mated, overwintering females in 2010 and distinguished from Cx. pipiens form molestus through microsatellite testing. 35 Oral mosquito infections. For each exposure dose of each virus, cohorts of 5-to 6-day-old female Cx. quinquefasciatus and Cx. pipiens were orally exposed to individual ZIKV strain containing blood meals through an artificial Hemotek membrane feeder (Discovery Workshops, Accrington, United Kingdom). Frozen stocks of known titers were thawed and used for blood meal exposure. For Cx. quinquefasciatus, two cohorts of 50 adult females were exposed to the same dose of MR766 (6.0 log 10 plaqueforming units [PFU]/mL); three cohorts were exposed to 4.0, 5.9, and 7.1 log 10 (PFU/mL) of PRVABC59; and two cohorts were exposed to 6.4 and 7.6 log 10 (PFU/mL) of R103451. For orally exposed Cx. pipiens, one cohort each was exposed to 6.0 log 10 (PFU/mL) of MR766 and PRVABC59. Female mosquitoes were allowed to feed for 45 minutes on the heated artificial blood meals containing 33% (v/v) defibrinated goose (Cx. pipiens) or calf (Cx. quinquefasciatus) washed erythrocytes (Colorado Serum Company, Denver, CO), 33% (v/v) heat-inactivated fetal bovine serum (FBS) (Omega Scientific, Inc., Tarzana, CA), 33% (v/v) of each individual virus in cell culture fluid (resulting in a final concentration ranging from 4.0 to 7.1 log 10 PFU/mL) and 1% (v/v) of 0.25 μM adenosine triphosphate (Table 1). After blood feeding, mosquitoes were cold-anesthetized for sorting, and engorged females were held at 28°C with a relative humidity of 70-75% for an extrinsic incubation period of 14 days.
Mosquito processing and testing. At the conclusion of the 14-day extrinsic incubation period, mosquitoes were anesthetized with triethylamine (Flynap ® ; Carolina Biological Supply company, Burlington, NC) as previously described. 36 Salivations were performed by insertion of the proboscis of each immobilized mosquito into a 10-μL capillary tube containing immersion oil (Cargille Laboratories, Cedar Grove, NJ) to induce salivation for approximately 45 minutes. Legs and wings were removed from anesthetized mosquitoes and placed in microcentrifuge tubes with 500 μL of complete media composed of Dulbecco's modified Eagle's essential medium (DMEM) complete with penicillin (100 U/mL), streptomycin (100 mg/mL), 10% FBS, amphotericin B (50μg/mL), and a sterile copper bead. After salivation, bodies and salivary capillary tubes were separated into 1.5-mL tubes with 500 μL and 350 μL of complete media, respectively. Samples with mosquito bodies and legs/wings were each triturated for 4 minutes at a frequency of 26 cycles per second using a Mixer Mill 300 (Retsch, Newton, PA). Samples containing salivary capillary tubes were centrifuged to clarify supernatants. Each sample was tested for cytopathic effects (CPEs) on Vero cells as previously described. 37 Collected salivary material for a given sample was not tested if dissemination was not detected.
IT mosquito infections. Cohorts of 50 adult female Cx. quinquefasciatus, Cx. pipiens, and Ae. aegypti colonized from Poza Rica, Mexico, 38 were subjected to IT inoculation of approximately 300 PFU (0.33 μL inoculation of 6 log 10 [PFU/mL]) of PRVABC59 or MR766 ZIKV. Mosquitoes were held for an extrinsic incubation period of 7 days at 28°C with a relative humidity of 70-75%. Mosquitoes were processed as described above. After processing, bodies and saliva samples were stored at −80°C until titration for virus by plaque assay or viral RNA testing. CPE-negative saliva from intrathoracically inoculated mosquitoes was subsequently analyzed by real-time reverse transcription polymerase chain reaction (qRT-PCR) 39 to confirm negative results. Viral RNA was extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's protocol.
In vitro growth on Culex cell lines. Replication profiles were assessed on Ae. aegypti (Aag-2), Ae. albopictus (C6/36), Cx. quinquefasciatus cells (Hsu), and Culex tarsalis (CT)-derived cell lines. All cell lines were propagated in DMEM supplemented with 5% FBS, penicillin (100 U/mL), and streptomycin (100 mg/mL). Cells were grown to 90% confluency in 6-well plates, and monolayers were infected in triplicate at a multiplicity of infection (MOI) of 0.1 PFU/ cell. 50-μL aliquots of supernatant were collected daily, followed by plaque assays to measure viral yield. Plaque assays used serial 10-fold dilutions of virus seeded onto 6-well plates containing monolayers of Vero cells, as previously described. 37 To compare the viral replication curves, a two-way analysis of variance (ANOVA) test with repeated measures and post hoc multiple comparisons test with a Tukey correction was performed, version 6.0h (Prism Graphpad Software, La Jolla, CA). P values < 0.05 were considered significant.

RESULTS
Oral infection, dissemination, and transmission in Culex mosquitoes. Oral infection rates were low for both Cx. quinquefasciatus (0-1%) and Cx. pipiens (1-10%) regardless of the virus used (Table 1). For example, only 1/95 Cx. quinquefasciatus was identified to be infected after oral exposure with 6.0 log 10 PFU/mL of MR766. Similarly, no infected Cx. quinquefasciatus mosquitoes were observed after oral exposure to PRVABC59 with doses ranging from 4.0 to 7.1 log 10 (PFU/mL) or to R103451 with doses of either 6.4 or 7.6 log 10 (PFU/mL) ( Table 1). There was no detectable dissemination of virus from the midgut of the single Cx. quinquefasciatus mosquito body identified to be ZIKV positive after oral exposure to MR766 as determined by CPE testing of legs/wings. For Cx. pipiens, 1/20 (5%) and 4/38 (11%) were found to have detectable virus in their bodies after exposure to 6.0 log 10 (PFU/mL) of MR766 or PRVABC59, respectively (Table 1). It should be noted that the four positive Cx. pipiens mosquitoes only had one plaque per well after undiluted inoculation of Vero cells, indicating an approximate concentration of 3 PFU/mosquito body. None of these Cx. pipiens were identified to have a disseminated infection as defined by the presence of virus in legs/wings.
Infection and transmission of inoculated mosquitoes. To determine whether Cx. pipiens and Cx. quinquefasciatus have the capability to transmit regardless of their apparent inability to establish infection of the midgut and disseminate virus, IT inoculations were performed to bypass the midgut infection and escape barriers. This allowed for direct assessment of ZIKV to infect the salivary glands and subsequently be transmitted during salivation. Seven days after IT inoculation, 16/23 (70%) and 5/33 (15%) Cx. quinquefasciatus bodies were identified to be positive for MR766 and PRVABC59 viruses, respectively ( Table 2). Plaque assay and subsequent qRT-PCR showed no virus or viral RNA-positive saliva samples. Experiments on Cx. pipiens provided similar results with 17/28 (61%) triturated bodies demonstrating virus 7 days after inoculation with PRVABC59 (Table 2), yet no evidence of transmissibility was observed by plaque assay or qRT-PCR assessments of expectorants. In contrast, evaluation of the IT susceptibility of Ae. aegypti showed that 12/12 (100%) of the surviving mosquitoes became infected after inoculation and 8/12 (67%) of infected mosquitoes demonstrated the presence of virus in salivary material indicating transmission potential.
In vitro growth of ZIKV strains in Aedes and Culex mosquito cells. Culex quinquefasciatus, Cx. tarsalis, Ae. aegypti, and Ae. albopictus cell lines were each inoculated at an MOI of 0.1 PFU/cell with PRVABC59 and MR766 and sampled daily for 7 days. A two-way repeated measures ANOVA for each virus demonstrated a significantly different level of growth between the three types of mosquito cells (P < 0.0001) (Figure 1). PRVABC59 reached a mean peak titer of 8.5 log 10 (PFU/mL) on day postinoculation (dpi)  7 and 7.5 log 10 (PFU/mL) on dpi 6 for Ae. albopictus and Ae. aegypti cells, respectively. The Puerto Rico virus titer in both Culex cell lines was highest on the initial sampling point at 0 dpi and dropped over the time course, indicating a growth rate below the loss of viral infectivity in the cultures over time. As such, mean titers of PRVABC59 growth in Ae. albopictus cells and Ae. aegypti were significantly higher (P < 0.0001) than either Culex cell line ( Figure 1A). Similarly, MR766 reached a mean peak titer of 7.8 log 10 (PFU/mL) on dpi 7 and 7.4 log 10 (PFU/mL) on dpi 6 in Ae. albopictus and Ae. aegypti cells. Subsequent multiple comparison testing of MR766 demonstrated growth reached significantly higher mean viral titers (P < 0.0001) on Ae. albopictus and Ae. aegypti cells as compared with each of the Culex cell lines, as viral titers continued to drop for these two cell lines from dpi 0-7 and never appeared to grow ( Figure 1B).

DISCUSSION
The potential vector competence of Cx. pipiens and Cx. quinquefasciatus was evaluated after oral and IT exposure of multiple ZIKV strains at a range of doses. Only a small percentage of mosquitoes were susceptible to infection orally (up to 10% for Cx. pipiens) and a range of 15-70% of intrathoracically inoculated mosquitoes demonstrated viruspositive bodies. However, unlike what was observed after IT inoculation of Ae. aegypti mosquitoes, there was no observed dissemination in any of the orally exposed Culex mosquitoes or salivary infection in any of the 38 intrathoracically inoculated mosquitoes showing body infection. The lack of detectable viral RNA or infectious virus in saliva, despite obviation of the midgut infection/escape barriers through IT inoculation, strongly suggests that ZIKV is incapable of infection of the salivary glands and/or entering the salivary fluids of Cx. quinquefasciatus and Cx. pipiens. Further, in vitro experiments demonstrate that ZIKV was unable to replicate in Culex cell lines.
The conclusions reported here regarding the inability of Culex mosquitoes to vector ZIKV correspond with a majority of recent studies addressing ZIKV transmission in Culex mosquitoes. [28][29][30][31][32] In contrast, two accounts report differing ZIKV transmission potential of Culex mosquitoes. The first case is a summary of findings presented during a ZIKV workshop in Brazil in 2016 in which the authors report detection of virus from the salivary glands of Cx. quinquefasciatus at 7 and 15 days postexposure. 26 To date, that presented data has not been formally published. However, a published study using several Brazilian ZIKV isolates with locally established Culex mosquito populations demonstrated incompetence of Cx. quinquefasciatus from Rio de Janeiro for infection, dissemination, and transmission. 30 Guo and others orally exposed colonized Cx. quinquefasciatus from the Hainan Province of China with a low passage strain isolated from a traveler returning from Samoa (GenBank: KU866423). In this report, 8/10 (80%) saliva samples demonstrated the presence of ZIKV RNA by RT-PCR 8 days after exposure, with the ZIKV RNA positivity rate in saliva dropping at later time points. 27 These results could indicate the possibility of differential susceptibility between Asian and American populations of Cx. quinquefasciatus. However, the incongruous comparison between detection of infectious virus, as traditionally measured by plaque formation or CPE assay, to RNA detection precludes drawing definitive conclusions between the two studies.
Phylogenetic analysis of mosquito-borne flaviviruses (MBFV) have long shown two distinct clades, one of which is primarily associated with Aedes mosquitoes with the other associated with Culex vectors. 40,41 For instance, dengue-1, 2, 3, and 4 viruses are transmitted by Ae. aegypti mosquitoes, but not Culex mosquitoes. 42 Considering the evolutionary precedent of MBFV steadfast adherence to being vectored by either Aedes or Culex mosquitoes but not both, it would be consistent that ZIKV would be vectored by Aedes mosquitoes. Similarly, Culex-borne flaviviruses typically use avian amplification hosts. Transmission of ZIKV to humans by Culex mosquitoes would require replication to high titers in avian hosts and potential epizootic spillover into humans as the Culex mosquitoes of concern primarily feed on avians and less frequently on humans. Blood meal analysis studies typically show humans to be a source of 0-2% of all blood meals taken by North American Cx. pipiens, Cx. quinquefasciatus, and Cx. tarsalis mosquitoes. [43][44][45][46][47][48][49][50] Lack of suitable ZIKV avian hosts combined with low human blood feeding rates for Culex mosquitoes diminish the potential role of Culex mosquitoes as vectors regardless of their vector competence for ZIKV.
Our use of oral and IT exposure in Cx. pipiens and Cx. quinquefasciatus to show the inability of ZIKV to disseminate to the salivary glands in addition to the lack of observed ZIKV replication in Culex cell lines, strongly indicates that these North American Culex mosquitoes are not competent vectors of ZIKV. In conjunction with previously mentioned ecological requirements of utilizing Culex mosquitoes as vectors, we conclude that there is no evidence to support Culex mosquito monitoring or management with regard to ZIKV control efforts.