Eastern equine encephalitis virus and identification of host bloodmeal sources from individual Culiseta melanura (Diptera: Culicidae) at an enzootic focus in central New York State

Abstract

Toward a more detailed understanding of the Eastern equine encephalitis virus (EEEV) cycle, individual blood-fed Culiseta melanura (Coquillett) mosquitoes were collected at an enzootic focus in New York State. Blooded females were tested for EEEV by multiplex real-time quantitative reverse transcription polymerase chain reaction, qRT-PCR. Host bloodmeals were identified using DNA amplification and sequencing of the cytochrome B gene. In 2018, 577 individual mosquitoes were tested. Virus was detected in the bodies of 16, of which two had the virus in their legs. In 2022, 606 individual mosquitoes were tested, and the virus was detected in the bodies of six, of which one had the virus in its legs. Virus in the legs suggested a disseminated infection. The qRT-PCR cycle threshold (Ct) values for individual bodies ranged from 17.6 to 38.3, and for legs, 22.8 to 28.1. Host sources for 17 of the 22 were: Passeriformes, Bombycillidae, cedar waxwings (Bombycilla cedrorum); Icteridae, red-winged blackbird (Agelaius phoeniceus); Turdidae, American robin (Turdus migratorius), veery (Catharus fuscescens), and wood thrush (Hylocichla mustelina); Vireonidae, yellow-throated vireo (Vireo flavifrons); Passerellidae, song sparrow (Melospiza melodia), and field sparrow (Spizella pusilla); Parulidae, American redstart (Setophaga ruticilla); and Gruiformes, Rallidae, American coot (Fulica americana). In both years, red-winged blackbirds were a source of blood for Cs. melanura in the earliest 2 d after EEEV was detected. Annually, the earliest detection of the virus was in blooded-gravid mosquitoes before non-blooded mosquitoes in 15 of 17 yr. These results support the thesis that birds contribute to the introduction and re-emergence of EEEV to this enzootic focus.

Introduction

The mosquito Culiseta melanura (Coquillett) is recognized as an enzootic vector of Alphavirus eastern, Eastern equine encephalitis virus (EEEV) (Wallis 1960, Morris et al. 1973). There are gaps in our knowledge regarding the epidemiology of EEEV (Hayes 1958), such as where and how it overwinters. Should the hardwood swamps that provide the unique oviposition habitat (Rupp et al. 2006) for Cs. melanura be referred to as endemic foci where EEEV is consistently present, or are they enzootic foci where EEEV amplification begins?

The application of DNA amplification to mosquito bloodmeals provided specificity to host identification (Apperson et al. 2004). Studies conducted in northeastern states have confirmed a wide range of bird species, primarily passerines, that were fed upon by field-collected Cs. melanura. These studies, regardless of geographic location, confirmed the ornithophilic preference of Cs. melanura often feeding on passerines, song and perching birds, which comprise over half of all bird species. Members of the family Turdidae (thrushes), such as American robin (Turdus migratorius L.), wood thrush (Hylocichla mustelina [Gmelin]), and veery (Catharus fuscescens [Stephens]) were fed on more frequently (Molaei et al. 2006, 2013, 2015, 2016), but these findings cannot be construed as Cs. melanura having a preference to feed on a particular bird species. Bloodmeal sources other than birds were a small percentage, generally <6%.

The first large-scale, site-specific study on molecular identification of bloodmeal sources of field-collected blooded Cs. melanura, concomitant with the presence of EEEV, was reported by Molaei et al. (2006). The setting was Oswego County, New York State (NYS), where EEEV is endemic (Morris et al. 1980a). The study included mosquito collections on the edge of Three Mile Bay Wildlife Management Area (TMBWMA), Town of West Monroe, and a distant site near the Village of Central Square (VCS), Town of Hastings, located 8 km from the edge site. Edge sites are sites where diurnal resting boxes (DRB) (Morris 1981) are set within sight of Cs. melanura oviposition locations (Morris et al. 1980a, Rupp et al. 2006) or in woods contiguous with the wet woods. Distant sites are DRB sites that are separated from the wet woods by non-sylvan habitats. The frequencies of host sources were the same at both sites.

Oswego County has recorded human EEE cases (Morris et al. 1973, Oliver et al. 2016) and has more equine EEE cases than any other county in NYS (Oliver et al. 2016, 2020). The Oswego County Health Department has participated in a mosquito and virus surveillance program since the program’s formation by the NYS Department of Health (NYSDOH) in 1984 (Howard et al. 1994). Much of the program’s history is focused on TMBWMA, a Cs. melanura oviposition habitat (Rupp et al. 2006). From 2006 to 2017, there were annual isolations of EEEV in Cs. melanura collected at TMBWMA, except for 2012 (Oliver et al. 2018, 2020). The annual presence of EEEV at a particular location, the ability to collect appreciable numbers of blooded Cs. melanura, and the availability of techniques to identify the source of a mosquito bloodmeal, to assay for the presence of virus, and to explore if an infection is disseminated or non-disseminated, led us to conduct a study to explore what bird species are fed upon by Cs. melanura when the initial EEEV detections are made at TMBWMA. Our study was designed to test the hypothesis that birds are responsible for the introduction or re-emergence of EEEV in central New York.

Methods and Materials

Oswego County EEEV Surveillance Program

Our study was conducted at the same time as the county participated in NYSDOH’s mosquito-borne virus surveillance. Each year, Oswego County began mosquito collections by early June and ended in September or October. Employees collected adult mosquitoes with DRB and miniature light traps (The John W. Hock Company, Gainesville, Florida, United States) baited with CO₂. Resting boxes collect Culiseta males and females that are unfed or contain blood and/or eggs (Morris 1981, 1984). Light traps collect unfed Culiseta females (Morris et al. 1980b). Up to 16 trap sites were located where EEEV has occurred (Howard and Grayson 1991, Howard et al. 1996, Oliver et al. 2016, 2018). A site was re-used each year, provided it remained productive for the collection of target mosquito species (Howard et al. 1996). Typically, 10 resting boxes were set at each resting box site. One light trap was set per light trap site, and sometimes in proximity to a resting box site. Generally, mosquitoes in resting boxes were collected on Monday and Wednesday mornings; light traps were collected on Tuesday and Thursday mornings. Collections were transported to the Shad C Slade Eastern Equine Encephalitis Field Station located within TMBWMA. Mosquitoes were killed using dry ice, females were identified to genus and species, and counts were recorded by site and collection date. Allocations of target species, primarily Cs. melanura, Culiseta morsitans (Theobald), Aedes canadensis (Theobald), and Coquillettidia perturbans (Walker) were pooled for the EEEV assay. Pools contained 10 to 60 specimens of the same species and physiological status and were collected by the same trap type, from the same site, within the same week. Culiseta were pooled separately as non-blooded (NB) or as blooded-gravid (BG). Species and physiologic status were confirmed using a dissecting microscope. Pools were stored on dry ice and delivered each Friday to the Arbovirus Laboratory, Wadsworth Center, NYSDOH, Albany.

Research Site

A county DRB site, named Field Station (FS), was selected for this study (Fig. 1). This edge site had been in use since 2012 and had been a source of EEEV-positive NB and BG Cs. melanura. The setting was a plantation of 40- to 50-yr-old Eastern white pine (Pinus strobus L.), 200 m from the Slade Field Station and 50 m from the wet woods edge of Toad Harbor Swamp in TMBWMA. Approximately 30 m from the county’s set of 10 resting boxes, a second set of 10 boxes was placed, for purposes of research, so that collections of mosquitoes could be made from this edge site even on a day when the county made a collection from their DRBs.

Fig. 1.

Map of townships within Oswego County showing a subset of locations of sites (rectangles) for collections of mosquitoes. Inset shows the location of Oswego County within New York State.

Collections of Specimens

In 2018 and in 2022, daily collections of Cs. melanura at the FS site began around 25 June and continued as often as 7 d per week when possible. The county collected from 10 resting boxes on Mondays and Wednesdays. Collections for research were made from the adjacent 10 resting boxes on those days. On days when the county was not collecting, a collection was made, for the purposes of research, from that site as well as from the adjacent research site, a total of 20 boxes. In 2018, for research, daily collections of Cs. melanura were sorted as NB, blood-fed, or gravid females. Pools of NB or gravid Cs. melanura were prepared. In 2022, for research, only blood-fed females were sorted; no NB or gravid specimens were pooled. In both years, the level of engorgement was assessed for each blooded mosquito using a dissecting microscope. Bloodmeal volumes (BMV) were recorded in increments of 10%, from 10% engorged to 100% engorged, by the same technical expert, for every collection, in both study years. In 2022, the presence or absence of eggs was recorded. Each blooded female was placed in a 0.6 ml snap-capped microcentrifuge tube. In 2018, a piece of Kimwipes was added to each tube, above the mosquito, to minimize its movement and to reduce the potential for the frozen specimen to break apart in transit. In 2018, the tube was sealed with Parafilm and labeled with an alphanumeric number. In 2022, neither Kimwipes nor Parafilm were considered necessary. Tubes were stored in tube racks on dry ice until transferred, on Fridays, to a −70 °C ultra-low temperature freezer at the NYSDOH laboratory at The State University of New York—Morrisville.

In both years, while research collections continued daily, NB and BG mosquito pools from the county surveillance program were being tested for viruses at the Arbovirus Laboratory. Test results for pools sent on a Friday would generally be available the following week. Collection of research specimens ended 2 wk after the first detection of EEEV in a surveillance pool at TMBWMA.

Dissections of Research Specimens

To determine the virus infection status of each individual blood-fed Cs. melanura, legs were carefully detached from the thorax, with sterile forceps, using a Zeiss microscope and subsequently processed in 500 µl mosquito diluent [MD, 20% heat-inactivated fetal bovine serum (FBS) in Dulbecco’s phosphate-buffered saline (PBS) plus 50 µg/ml penicillin/streptomycin, 50 µg/ml gentamicin, and 2.5 µg/ml Fungizone] with 1 steel bead (Daisy Outdoor Products, Rogers, Arkansas, United States). Forceps were sterilized in 70% ethanol (ETOH) after each mosquito dissection. Each individual, legless mosquito body, i.e. head, thorax, abdomen, was tested for EEEV.

Viral Assay

Mosquitoes, whether pooled or individual, were tested for arboviruses by multiplex real-time quantitative reverse transcription polymerase chain reaction, qRT-PCR (Lambert et al. 2003, Zink et al. 2013). The multiplex includes two sets of primers for EEEV targeting the three prime untranslated region (3′-UTR) and non-structural protein 1 (NSP1). Processing of pools and individuals consisted of homogenization for 3 min at 24 Hz, in a Mixer Mill MM301 (Retsch, Newtown, Pennsylvania, United States), followed by centrifugation at 12,700 rpm for 3 min at 4 °C. Total nucleic acid was extracted on the MagMax-96 Express Magnetic Particle Processor (Applied Biosystems, Foster City, California, United States) with a MagMAX Total Nucleic Acid Isolation Kit (Life Technologies, Waltham, Massachusetts, United States), according to the manufacturer’s specifications. Briefly, 50 µl of mosquito homogenate was added to 130 µl of lysis buffer containing 20 µl of RNA-binding beads diluted 1:1 with wash buffer 1. RNA was eluted in 90 µl of sterile water, and 5 µl was included in RT-PCR reactions as previously described (Zink et al. 2013). Interpretation of RT-PCR Ct values is based on previous validation. County surveillance pools were also tested for the presence of EEEV by assessment of cytopathic effect on Vero cell culture. Virus isolations from county surveillance pools are referred to as detections herein. Host source determination was made for individual Cs. melanura in which EEEV was detected by PCR, as well as for subsets of individual Cs. melanura in which EEEV was not detected.

PCR and Bloodmeal Host Identification

RT-PCR was done using a qScript one-step RT-PCR kit (Quantabio, Beverly, Massachusetts, United States), and samples were analyzed on a QuantStudio 5 Real-time PCR machine (Thermo Fisher, Waltham, Massachusetts, United States). EEEV-specific TaqMan primer pairs and probe were used to target the 3′-UTR (5′-CCCTAGTTCGATGTACTTCCG-3′, 5′-GCATTATGCACTGCCCTTAG-3′, 5′-6FAMCCGCCGATGCAGTGCATAAGGCTG/MGBNFQ/-3′) and the NSP1 gene (5′-ATGAAGAGCGCAGAAGACCC-3′, 5′-GCGTCGACATTACTGTTAGC-3′, 5′-Cy5/CAG ACT CTA/TAO/CCG CTA CGC AGA CAA G/31AbRQSp/-3′). Body samples that tested positive by RT-PCR were harvested and stored at −80 °C. Leg samples corresponding to positive bodies were processed, tested, and stored as above. A Taq PCR core kit (Qiagen, Germantown, Maryland, United States) was used to amplify the avian Cytochrome b gene using primer pairs, 5′-GACTGTGACAAAATCCCNTTCCA-3′, and 5′-GGTCTTCATCTYHGGYTTACAAGAC-3′ (Ngo and Kramer 2003). In 2022, PCR amplification of all samples was modified using Q5 Hot Start High-Fidelity 2× Master Mix (New England Biolabs, Ipswich, Massachusetts, United States) according to the manufacturer’s recommendations. Samples were amplified with the following thermocycler conditions: 98 °C for 1 min; 40 cycles at 98 °C for 10 s, 63 °C for 15 s, and 72 °C for 20 s; and one cycle at 72 °C for 2 min with a hold at 4 °C. Five microliters of each sample were visualized on a 2% agarose gel, and the remainder was purified using either a Zymo kit (Sigma-Aldrich, St Louis, Missouri, United States) or a QIAquick Gel Purification Kit (Qiagen, Germantown, Maryland, United States). Purified DNA was sequenced at the Wadsworth Center Advanced Genomics Technologies Core. Sequences were analyzed on Geneious version 11.1.5 (San Diego, California, United States) and identified using NCBI Blast software (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Results

Descriptive Analyses

Seasonal mosquito surveillance in Oswego County over a 17-yr period, 2006 to 2022, resulted in the testing of 9,343 pools containing 365,307 specimens, primarily in the genera Aedes, Coquillettidia, and Culiseta (Table 1). Pools of Cs. melanura represented 41% (n = 3,810) and 32% (n = 119,750) of the totals, respectively (Table 1). Of the 3,810 pools of Culiseta, 58% (n = 2,208) were NB females and 42% (n = 1,602) were BG (blood-fed and gravid) females (Table 1). Ninety-six percent (424) of isolations of EEEV were from Cs. melanura with pools of NB and BG representing 42% (n = 178) and 58% (n = 246), respectively (Table 1).

Table 1.

Mosquitoes collected and tested for the presence of Eastern equine encephalitis virus in Oswego County annually from 2006 through 2022

Surveillance			All mosquito species			Culiseta melanura blooded-gravid			Culiseta melanura non-blooded
Year	Beginning	Ending	Individuals (count)	Pools (count)	Pools containing virus (count)	Individuals (count)	Pools (count)	Pools containing virus (count)	Individuals (count)	Pools (count)	Pools containing virus (count)
2006	30 May	27 September	6,950	179	43	561	26	7	5,763	136	36
2007	4 June	13 September	13,228	356	20	1465	45	6	2,814	91	13
2008	27 May	9 October	17,095	458	17	2,212	79	11	4,976	158	5
2009	29 May	23 September	30,485	657	43	3,277	91	22	6,103	150	17
2010	24 May	23 September	21,045	484	47	1,762	67	19	5,098	146	25
2011	1 June	27 September	22,742	577	31	2,970	121	23	2,932	97	4
2012	29 May	18 September	23,382	642	1	2,174	86	1	1,492	71	0
2013	5 June	12 September	29,686	721	11	4,782	139	9	7,240	182	2
2014	2 June	18 September	29,192	743	50	4,191	145	30	7,609	206	18
2015	2 June	23 September	25,110	657	21	2,788	110	15	3,619	114	6
2016	1 June	15 September	19,313	478	5	1,341	70	5	1,726	73	0
2017	6 June	11 October	23,791	635	30	1,102	52	18	9,343	205	12
2018	4 June	20 September	17,077	511	17	1,952	108	12	1,846	79	5
2019	3 June	19 September	20,786	548	40	2,888	123	30	2,945	111	8
2020	9 June	17 September	18,806	488	19	2,501	90	11	1,549	72	8
2021	9 June	2 October	26,118	685	32	3,450	119	19	7,250	190	13
2022	6 June	15 September	20,501	524	15	4,175	131	8	3,854	127	6
Total			365,307	9,343	442a	43,591	1,602	246	76,159	2,208	178

^aBy frequency, detections of EEEV were: Culiseta melanura 424, Coquillettidia perturbans 15, Culiseta morsitans 2, Anopheles punctipennis 1.

Each year, the date of the first virus detection was between 25 June (2007) and 27 August (2012), and, in every year except 2006 and 2020, the first detection was in BG females (Table 2). In 8 of 15 yr, the first detection of each year was at an edge site. In 5 yr, the first detection was at a distant site (Table 2). In 2008 and 2015, pools of BG females from both edge and distant sites were virus-positive on the same date. The collection interval between the first detection from a pool of BG and a detection from a pool of NB females ranged between 9 and 23 d (Table 2). In each of the 5 yr, 2007, 2010, 2012, 2017, and 2022, the first detection of the year was followed by a week without a detection in Cs. melanura despite Cs. melanura being collected and tested. We use the term “skip week” to refer to a week without an EEEV detection from pooled specimens. Generally, subsequent detections in future weeks were also from pools of BG females (Table 2, Fig. 2) before the virus started to be detected from NB females.

Table 2.

First date of isolation, by year, of Eastern equine encephalitis virus in Culiseta melanura and number of days between first isolations from pools of blooded-gravid and non-blooded females in Oswego County from 2006 through 2022

Surveillance			Culiseta melanura blooded-gravid			Time interval	Culiseta melanura non-blooded
Year	Beginning (date)	Ending (date)	Earliest poola (date)	Site nameb	Habitatc	From earliest blooded-gravid pool to earliest non-blooded pool having EEEV (days)	Site nameb	Habitatc	Pools having EEEV from earliest blooded-gravid to earliest non-blooded (count)
2006	30 May	27 September	13 July	–	–	–	THR	Edge	–
2007	4 June	13 September	25 June	THR	Edge	14	THR	Edge	4
2007			–	–	–	14	VCS	Distant	4
2008	27 May	9 October	18 August	VCS	Distant	9	VCS	Distant	3
2008			18 August	THR	Edge		–	–	–
2009	29 May	23 September	1 July	THR	Edge	22	VCS	Distant	4
2010	24 May	23 September	30 June	THR	Edge	14	VCS	Distant	2
2011	1 June	27 September	18 July	THR	Edge	9	VCS	Distant	3
2012	29 May	18 September	27 August	VCS	Distant	n/ad	–	–	–
2013	5 June	12 September	7 August	VCS	Distant	21	FS	Edge	6
2014	2 June	18 September	1 July	VCS	Distant	23	VCS	Distant	5
2015	2 June	23 September	27 July	THR	Edge	16	FS	Edge	9
2015			27 July	VCS	Distant	16	VCS	Distant	9
2016	1 June	15 September	17 August	THR	Edge	n/ae	–	–	–
2017	6 June	11 October	26 July	AUR	Distant	19	FS	Edge	5
2018	4 June	20 September	30 July	FS	Edge	9	FS	Edge	4
2019	3 June	19 September	8 July	FS	Edge	23	FS	Edge	6
2020	9 June	17 September	15 July	–	–	–	FS	Edge	–
2021	13 June	7 October	21 July	DOW	Distant	19	AUR	Distant	3
2022	6 June	15 September	13 July	PAR	Distant	12	DOW	Distant	5

^aEach first isolation was in blooded-gravid Culiseta melanura except for 2006 and 2020.

^bSite name: THR = Toad Harbor Road, VCS = Village of Central Square, AUR = Auringer Road, FS = Field Station, DOW = Dowley Road, PAR = Paradise Road.

^cHabitat: Edge means at the edge of the swamp, Distant means at a distance from the edge of the swamp.

^dIn 2012, there was one detection of EEEV in blooded-gravid Culiseta melanura, but no detection of EEEV in non-blooded Culiseta melanura.

^eIn 2016, all five detections were blooded-gravid Culiseta melanura.

Fig. 2.

Culiseta melanura, by physiologic status, number of pools tested for, and number of pools positive for Eastern equine encephalitis virus, EEEV, Oswego County, New York State, 2018.

2018 Oswego County EEEV Surveillance

Surveillance for EEEV was conducted between 4 June and 20 September. The first detection of EEEV of the year was from a county surveillance pool of 14 BG Cs. melanura collected on 30 July from the FS DRB site. Prior to 30 July, 47 pools of BG and 44 pools of NB Cs. melanura tested negative. Of these, 22 BG pools (568 specimens) and 18 NB pools (554 specimens) were collected at TMBWMA. In 2018, throughout the county, there were 17 detections of EEEV from 511 pools containing 17,077 specimens. All 17 detections were from pools of NB (n = 5) or BG (n = 12) Cs. melanura (Table 1). There was no skip week (Tables 1 and 2, Fig. 2).

2018 Research

Cs. melanura were collected from 27 June to 17 August from the FS site. There were 1,562 female Cs. melanura (897 NB, 577 blood-fed, and 88 gravid). The 577 mosquitoes contained black blood, not red blood. All 44 pools of NB Cs. melanura, and 10 pools of gravid Cs. melanura were assayed. There were 23 detections of EEEV, 16 from blood-fed individuals and 7 from pooled Cs. melanura. The first detection in individual Cs. melanura was on 27 July. Prior to 27 July, 161 individual Cs. melanura were virus-negative. The first positive pool of NB females was collected on 31 July, and the first gravid pool on 1 August. Three of the first four EEEV-positive individual specimens were collected on 27 July and fed on cedar waxwings (Bombycilla cedrorum Vieillot) and red-winged blackbird (Agelaius phoeniceus [L.]) (Table 3). Three of these were females with non-disseminated infections, EEEV-positive bodies with legs that were either virus-negative or below the limit of detection. The fourth individual had a disseminated infection, EEEV-positive body and EEEV-positive legs (Table 3). Between 31 July and 16 August, EEEV was detected in 12 more individual specimens. Eleven had non-disseminated infections (Table 3).

Table 3.

Determinations of host species blood from 22 individual Culiseta melanura mosquitoes that tested positive for Eastern equine encephalitis virus in New York State during 2018 and 2022

Date of collection	Assay (no.)	Mosquito blood meal volume (%)a	Virus present in the body by PCR cycle threshold (Ct)	Virus present in legs by PCR cycle threshold (Ct)b	Order	Family	Bird species identified by PCR of blood meal (genus and species)		Bird species identified by PCR of blood meal (common name)
27 July 2018	162	100	22.1	–	Passeriformes	Bombycillidae	Bombycilla cedrorum	Cedar waxwing
27 July 2018	183	100	17.6	22.8			Not determined	Not determined
27 July 2018	200	100	20.3	–	Passeriformes	Bombycillidae	Bombycilla cedrorum	Cedar waxwing
27 July 2018	204	60	22.5	–	Passeriformes	Icteridae	Agelaius phoeniceus	Red-winged blackbird
31 July 2018	259	80	30.5	–			Not determined	Not determined
31 July 2018	266	80	27.4	–			Not determined	Not determined
1 August 2018	289	10	22.1	24.7			Not determined	Not determined
3 August 2018	315	100	21.5	–	Passeriformes	Turdidae	Turdus migratorius	American robin
3 August 2018	322	100	19.7	–	Passeriformes	Bombycillidae	Bombycilla cedrorum	Cedar waxwing
3 August 2018	324	80	28.1	–	Passeriformes	Bombycillidae	Bombycilla cedrorum	Cedar waxwing
7 August 2018	352	10	30.0	–			Not determined	Not determined
9 August 2018	367	90	29.0	–	Passeriformes	Passerellidae	Melospiza melodia	Song sparrow
10 August 2018	414	nd	19.9	–	Passeriformes	Turdidae	Catharus fuscescens	Veery
11 August 2018	422	100	23.7	–	Passeriformes	Passerellidae	Spizella pusilla	Field sparrow
11 August 2018	442	80	20.8	–	Passeriformes	Passerellidae	Melospiza melodia	Song sparrow
16 August 2018	525	80	21.1	–	Passeriformes	Parulidae	Setophaga ruticilla	American redstart
21 July 2022	61	90	24.9	28.1	Passeriformes	Vireonidae	Vireo flavifrons	Yellow-throated vireo
22 July 2022	198	80	29.1	–	Passeriformes	Icteridae	Agelaius phoeniceus	Red-winged blackbird
24 July 2022	332	70	38.3	–	Passeriformes	Turdidae	Turdus migratorius	American robin
25 July 2022	384	60	30.3	–	Gruiformes	Rallidae	Fulica americana	American coot
31 July 2022	570	80	36.7	–	Passeriformes	Turdidae	Turdus migratorius	American robin
31 July 2022	587	70	32.0	–	Passeriformes	Turdidae	Hylocichla mustelina	Wood thrush

^aBloodmeal volume as an estimate of bloodmeal fullness of the abdomen from 0% to 100% in increments of 10%, seen using a dissecting microscope.

^bA hyphen indicates legs were tested for EEEV, and EEEV was not detectable.

2022 Oswego County EEEV Surveillance

Surveillance for EEEV was conducted between 6 June and 15 September. The first detection of EEEV of the year was a county surveillance pool of BG Cs. melanura collected on 13 July at a DRB edge site, PAR, 30 km from the FS site (Fig. 1). This pool tested negative by qRT-PCR and tested positive by Vero cell culture. There was a skip week between the first and second detections from pools of blood-fed Cs. melanura (Table 2, Fig. 3), with the second detection being at that same site. The initial detection of EEEV closest to the FS site was a surveillance pool of BG Cs. melanura collected on 1 August at the THR edge surveillance site 1.6 km west and within TMBWMA (Fig. 1). Prior to 1 August, 74 pools of BG and 62 pools of NB Cs. melanura tested negative. Of these, 33 BG pools (1,409 specimens) and 35 NB pools (1,471 specimens) were collected in TMBWMA. In 2022, throughout the county, there were 15 EEEV-positive detections from 524 pools containing 20,501 specimens. Fourteen of the 15 detections were from pools of NB (n = 6) or BG (n = 8) Cs. melanura (Table 1, Fig. 3).

Fig. 3.

Culiseta melanura, by physiologic status, number of pools tested for, and number of pools positive for Eastern equine encephalitis virus, EEEV, Oswego County, New York State, 2022. Arrow at week 28 designates a “skip week” (a week with no virus isolations following a week when EEEV was first isolated), and isolations in weeks 29 and 30 were in only blooded-gravid Cs. melanura.

2022 Research

Cs. melanura were collected from 24 June to 9 August. Influenced by the first isolation of EEEV in Cs. melanura collected on 1 August at THR, we selected 606 individual blood-fed specimens collected from FS site between 17 July and 3 August. Each individual had BMV estimated at ≥50% to 100% because, in 2018, host determination was sometimes not successful for Cs. melanura having BMV <50%. The 606 Cs. melanura contained black blood, not red blood. The first detection of EEEV in individual blooded Cs. melanura was on 21 July. Prior to 21 July, 60 individual Cs. melanura were virus-negative. In total, there were six detections of EEEV in specimens collected through 31 July; five were non-infected females, and one had a disseminated infection (Table 3). Specimens 384 and 587 contained eggs, and specimens 61, 198, 332, and 570 did not contain eggs (Table 3). The host bloodmeal sources for these Cs. melanura were yellow-throated vireo (Vireo flavifrons Vieillot), red-winged blackbird, American robin, American coot (Fulica americana Gmelin), and wood thrush (Table 3).

Cycle Threshold (Ct) Values

The qRT-PCR Ct values for an individual blood-fed Cs. melanura body, ranged from 17.6 to 38.3, and for legs from individual Cs. melanura, 22.8 to 28.1 (Table 3). In 2018, the average body Ct value for the individual specimens was 23.5, compared to an average body Ct value of 31.8 for the individual specimens collected in 2022. The Ct values for the three females having EEEV in their legs were higher than the Ct values for their bodies (Table 3). Specimen 183 was estimated to be 100% full of blood. The Ct of the body, i.e. the abdomen, thorax, and head, was 17.6; the Ct of the legs was 22.8. Specimen 289 had a BMV = 10%, body Ct = 22.1, legs Ct = 24.7. Specimen 61 had a BMV = 90%, body Ct = 24.9, legs Ct = 28.1.

The Ct values of five research pools of NB Cs. melanura, collected from 31 July to 11 August 2018, ranged from 17.8 to 26.4, and of the two research pools of gravid Cs. melanura, collected on 3 August and 6 August 2018, were 23.2 and 22.2, respectively. The Ct value of the first county surveillance pool of Cs. melanura in 2018, collected on 30 July in TMBWMA, was 24.76. The Ct value of the first county surveillance pool of Cs. melanura in 2022, collected on 1 August in TMBWMA, was 30.1.

EEEV-Negative Cs. melanura Bloodmeal Sources

Bloodmeal sources were determined for 55 of 58 EEEV-negative individual Cs. melanura collected in the days surrounding the calendar dates, in 2018 and 2022, when the virus was known to be present: 34 Cs. melanura collected from 13 July to 1 August 2018 and 24 Cs. melanura collected from 21 to 31 July 2022. There were 19 bird species used as bloodmeal sources (Table 4).

Table 4.

Determinations of host species of blood from 55 individual Culiseta melanura mosquitoes that tested negative for Eastern equine encephalitis virus in New York State during 2018 and 2022

Order	Family	Genus and species	Common name	Count
Passeriformes	Turdidae	Hylocichla mustelina	Wood thrush	12
Passeriformes	Turdidae	Turdus migratorius	American robin	9
Passeriformes	Turdidae	Catharus fuscescens	Veery	2
Passeriformes	Cardinalidae	Cardinalis cardinalis	Northern cardinal	6
Passeriformes	Vireonidae	Vireo olivaceus	Red-eyed vireo	4
Passeriformes	Mimidae	Dumetella carolinensis	Gray catbird	3
Passeriformes	Parulidae	Setophaga ruticilla	American redstart	2
Passeriformes	Parulidae	Geothlypis trichas	Common yellowthroat	1
Passeriformes	Bombycillidae	Bombycilla cedrorum	Cedar waxwing	1
Passeriformes	Icteridae	Dolichonyx oryzivorus	Bobolink	1
Passeriformes	Icteridae	Molothrus ater	Brown-headed cowbird	1
Passeriformes	Icteridae	Agelaius phoeniceus	Red-winged blackbird	1
Passeriformes	Passerellidae	Melospiza melodia	Song sparrow	1
Passeriformes	Passerellidae	Ammodramus savannarum	Grasshopper sparrow	1
Passeriformes	Polioptilidae	Polioptila caerulea	Blue-gray gnatcatcher	1
Columbiformes	Columbidae	Zenaida macroura	Mourning dove	5
Gruiformes	Rallidae	Fulica americana	American coot	2
Pelecaniformes	Ardeidae	Butorides virescens	Green heron	1
Strigiformes	Strigidae	Strix varia	Barred owl	1

Discussion

Surveillance data spanning 17 years (Tables 1 and 2) supports a scenario where enzootic activity begins when host-seeking Cs. melanura imbibe viremic blood from birds infectious with EEEV and, after completion of oviposition, infectious female Cs. melanura may transmit the virus to a susceptible bird. In 15 of 17 years, the first EEEV isolations of each year were from pools of BG Cs. melanura. The number of days between the first BG isolations and the first NB isolations ranged between 9 and 23. In five of these 15 years, we observed a skip week, a 1-wk virus-negative period, between the first and second isolations of the year. The absence of virus in a week following a first detection is consistent with the gonotrophic cycle of Cs. melanura calculated to be 7 to 10 d at TMBWMA (Morris 1984). Adult Cs. melanura can be collected as early as mid-May (Morris and Srihongse 1978, Morris et al. 1980b, Oliver et al. 1996). All stages, except first instar, can overwinter, resulting in a protracted period of pupation and protracted appearance of first brood adults that extends into June when first instars of the first summer generation may be present (Andreadis et al. 2012).

In our study, EEEV was detected in the blood-fed bodies of 22 individual Cs. melanura collected at an edge site at TMBWMA that has been surveilled annually since 2012. Because we assayed the body minus the legs, we cannot say with certainty that the virus was in only the bloodmeal. Virtually all the blood-fed females collected at this site were black-blooded (Crans and McCuiston 1993), indicative of females that had fed 24 to 72 h prior to collection. For the mosquitoes without dissemination to the legs, Ct values ranged from 19.7 to 36.7 (Table 3). An inverse linear quantitative relationship between Ct value and number of PFU/ml of specimen tested has been demonstrated in a study of EEEV in pools of field-caught Cs. melanura (Armstrong and Andreadis 2010). The Ct values ranged from 15 to 35, and the PFU/ml ranged from 10¹ to 10^7.5, approximately (Armstrong and Andreadis 2010).

The three individual Cs. melanura with disseminated infections (positive legs) could be individuals that had ingested high host viremias or had fed days earlier. Two of these three individuals had bloodmeal fullness estimated at ≥90% and EEEV was likely obtained during a previous bloodmeal. The three individual Cs. melanura with positive legs had low Ct values, <30, as did 15 of the 19 with negative legs. In a laboratory-based study, virus-positive legs were observed in 89% (49 of 55) of “virus-infected” Cs. melanura 18 d after feeding on a virus-infected bird (Komar et al. 1999).

Specimens with Ct values >30 and legs without virus may have ingested a bloodmeal with a viremia below the infection threshold (Howard and Wallis 1974, Scott and Burrage 1984, Komar et al. 1999) or took a partial bloodmeal (Scott and Weaver 1989). These specimens may not be infected and would not contribute to the amplification of the virus.

In 2018, the finding, on the same date (27 July), of 4 Cs. melanura with EEEV suggests the virus may have been on-site prior to this date. In 2022, the finding of an individual blooded Cs. melanura with virus in both body and legs also suggests the virus may have already been on-site. In 2018 and 2022, the timing of detections of EEEV in individual Cs. melanura at TMBWMA preceded detections of EEEV in county surveillance pools by 3 and 11 d, respectively. In 2018, testing of pooled NB and pooled gravid Cs. melanura collected in research led to the detection of EEEV in a pool of NB Cs. melanura that were collected 4 d (31 July) after the first individual blooded Cs. melanura (27 July). In county surveillance, the first detection of EEEV in a pool of NB Cs. melanura was 9 d after the first BG pool (30 July). In 2022, in county surveillance, the first detection of EEEV in a pool of NB Cs. melanura was 21 d after the first BG pool (1 August). In 2022, the first detection of EEEV in county surveillance, a pool of 11 BG Cs. melanura, was negative when tested by qRT-PCR and positive when tested by Vero cell culture. Possible explanation for a PCR result of negative, undetectable, may be the inhibitory effect of the PCR reaction from the blood. A PCR result of negative, undetectable, for legs may reflect the limit of detection.

Culiseta melanura is the most competent EEEV vector: it has an ornithophilic feeding preference, more isolations of EEEV than all other mosquito species, and is found infected during enzootic, epizootic, and epidemic outbreaks (Armstrong and Andreadis 2022). In laboratory studies, with viremias of ≥10⁶ PFU/ml, Cs. melanura had a transmission rate of ≥90% (Howard and Wallis 1974, Komar et al. 1999). At viremias between 10⁶ and 10³ PFU/ml, transmission rates are 50% to 5%, and when viremias are <10³ PFU/ml, no or few Cs. melanura becomes virus-infected (Howard 1973, Howard and Wallis 1974, Komar et al. 1999). Viremias in birds sufficient to infect Cs. melanura are likely to be of shorter duration, 1 to 4 d (Komar et al. 1999). The contact time between the infectious host and host-seeking female is further shortened by the nocturnal feeding behavior of Cs. melanura (Hayes 1962). In central New York in mid-July, the light: dark ratio is approximately 14:10. Reactivated viremia in adult birds, if present, may not be sufficient to initiate detectable enzootic EEEV activity. The earliest detection of EEEV at TMBWMA is 25 June, but it more often appears in July (Oliver et al. 2018, 2020), coinciding with the appearance of the summer generation of adult Cs. melanura (Oliver et al. 1996) and fledged birds.

Rapid amplification of EEEV during epiornitics could be explained by dual or multiple feedings on birds by Cs. melanura (Scott and Weaver 1989). Studies have reported single bloodmeals from a wide variety of birds, with a small percentage of mixed avian or mixed avian and mammalian bloodmeal sources (Molaei et al. 2006, 2013, 2015, 2016). Host selection and abundance, specifically of the American robin and wood thrush, have been equated with importance for EEEV endemicity (Komar and Spielman 1994, Molaei et al. 2016, Armstrong and Andreadis 2022). We suspect amplification is more related to the abundance of Cs. melanura and susceptible hatching year (HY) birds (Howard et al. 2004). In 2017, in the 5 wk between 27 July and 21 August, including a skip week, EEEV increased from a single positive pool at a distant site (AUR, Table 2) to 14 positive pools, at five different surveillance sites, two edge and three distant sites.

The results support the recrudescence theory of Crans et al. (1994), which found EEEV in birds prior to finding the virus in infected Cs. melanura, though there is little laboratory support for this hypothesis (Brown and O’Brien 2011). In Crans et al. (1994), of 13 bird species found to be viremic for EEEV, a seropositive gray catbird (Dumetella carolinensis [L.]) was found viremic the following June. Gray catbirds are common within the range of EEEV in the United States (Lindsey et al. 2018, 2020), and are the only species from which EEEV has been isolated at coastal and inland EEEV foci (Brown and O’Brien 2011). The exception is the Michigan EEEV focus (McLean et al. 1985), with the caveat that there has been limited surveillance of vectors and enzootic hosts in Michigan (Stobierski et al. 2022). Gray catbirds were not represented in the EEEV RNA-positive bloodmeals in the 2 years of our study (Table 3).

Our study suggests the red-winged blackbird may be an important host for amplification or maintenance of EEEV in TMBWMA. It was one of the first viremic species in 2018 and 2022, and one of the species found EEEV-positive in New Jersey (Crans et al. 1994). Red-winged blackbirds are common in open fields, the preferred habitat of host-seeking Cs. melanura (Howard et al. 1983). This species was susceptible to infection with EEEV when used as a field sentinel (Williams et al. 1971) or laboratory host (Komar et al. 1999). Red-winged blackbirds captured at TMBWMA were found to have lower antibody rates than other species (Emord and Morris 1984, Howard et al. 2004). Four of our virus-positive Cs. melanura fed on red-winged blackbirds or song sparrows (Melospiza melodia [Wilson]) (Table 3). During our studies at TMBWMA, we have recorded more isolations of EEEV from song sparrows than in any other bird species (Howard et al. 2004). In a laboratory study, a red-winged blackbird and a song sparrow, each circulating 10³ PFU/ml, remained infectious to Cs. melanura for 1 d (Komar et al. 1999). A long-term study of birds by Main et al. (1988) at an enzootic focus in Massachusetts found that neutralizing antibodies from serial within- and between-year blood samples fluctuated between positive and negative during a period of minimal EEEV activity (Edman et al. 1993). A laboratory study with tricolored blackbird (Agelaius tricolor [Audubon]) reported that this species developed a chronic latent infection with the alphavirus Western equine encephalitis virus (Reeves et al. 1958).

Most birds are assumed to be susceptible to infection with EEEV, and EEEV has been directly isolated from at least 75 species (Emord 1983, Howard et al. 2004, Brown and O’Brien 2011, Oliver et al. 2016). In 2018 and 2022, EEEV was detected in the bodies of blooded Cs. melanura that fed on 10 bird species. If these birds were viremic, then these were the first detections in the United States of EEEV from cedar waxwing and American coot, and the first detections in NYS of EEEV from red-winged blackbird, American robin, American redstart (Setophaga ruticilla [L.]), yellow-throated vireo, and wood thrush. We would offer that it is the diversity of bird species and the indiscriminate avian host seeking of Cs. melanura that determines EEEV activity at individual foci.

At TMBWMA, 508 blooded Cs. melanura fed on 52 bird species in eight orders, and 80% were Passeriformes (Molaei et al. 2006). This was one of three studies by Molaei et al. (2006, 2013, 2016) that included assay of each individual Cs. melanura head with thorax for the presence of a virus, indicative of disseminated infection with the potential to transmit. There were three Cs. melanura that were EEEV-positive among 513 Cs. melanura tested. There was one Cs. morsitans EEEV-positive among 125 Cs. morsitans tested. The interpretation was these three Cs. melanura, collected on 25 August and 8 September, and the Cs. morsitans collected on 14 September were infected, and EEEV was either in the current bloodmeal or from a prior bloodmeal. Host sources of bloodmeals were wood thrush, song sparrow, and ovenbird (Seiurus aurocapilla) (Molaei et al. 2006).

In Massachusetts, 529 blooded Cs. melanura had fed on 55 bird species in eight orders, and 97% were Passeriformes (Molaei et al. 2013). One Cs. melanura collected on 3 September was EEEV-positive, and the host source of bloodmeal was not determined (Molaei et al. 2013).

In Vermont, 312 blooded Cs. melanura fed on 49 bird species in five orders, and 89% were Passeriformes (Molaei et al. 2015). Five Cs. melanura collected from 7 August to 18 September were EEEV-positive, and host sources of bloodmeals were American robin, common yellowthroat (Geothlypis trichas [L.]), and Savannah sparrow (Passerculus sandwichensis [Gmelin]) (Molaei et al. 2015). In the Vermont study, the green heron (Butorides virescens [L.], order Cuculiformes) was most often fed on, followed by the American robin. Whereas in the studies in New York, Massachusetts, and Connecticut, wood thrushes, American robin, and veery were the species most often fed on.

In Connecticut, 1,127 blooded Cs. melanura had fed on 65 bird species in 11 orders and, similar to Massachusetts, 97% were Passeriformes (Molaei et al. 2016). Pools, rather than individuals, were tested for virus, and four pools were EEEV-positive (Molaei et al. 2016).

Phylogenetic studies of EEEV support our hypothesis that multiple species of birds are involved in the transportation and introduction of EEEV into enzootic foci in the northeast. Analyses of isolates from central New York (Weaver et al. 1993, Armstrong et al. 2008, Young et al. 2008) determined that strains of EEEV persist on-site for 2 to 3 years and suggested that the virus overwintered on-site. A study of the complete EEEV genome supported the finding of virus persistence for 2 to 3 yr, and the routine introduction of new strains directly from Florida to central New York (Tan et al. 2018). The persistence of a virus strain onsite is consistent with the onsite, between-year recapture rates for birds banded at TMBWMA, demonstrating site affinity by resident nesting birds (Emord 1983, Emord and Morris 1984, Howard et al. 2004). Seven geographic areas have been identified in Florida where EEEV strains originate (Heberlein-Larson et al. 2019), and certain bird species have been proposed as possible transporters of EEEV from Florida to northeastern states (Burkett-Cadena et al. 2022).

In an analysis of EEEV strains present in NYS in 2013 and 2014, there were four EEEV clades identified, NY4, NY5, NY6, and NY7 (Tan et al. 2018). Strains from these four clades caused equine disease (Oliver et al. 2020). Clades 5, 6, and 7 were found in central New York and were isolated from Cs. melanura and Cq. perturbans collected from 29 July to 3 October 2014 at Cicero Swamp Wildlife Management Area, Onondaga County (Oliver et al. 2020), an enzootic EEE focus 30 km south of TMBWMA (Morris et al. 1980a, Howard et al. 1988).

The question remains, in its life cycle, how, where, and when EEEV appears at this enzootic focus? Over 17 years, we did not detect EEEV in first brood females (Table 2, Figs. 2 and 3). These results did not support the hypothesis that EEEV is maintained in mosquitoes by transovarial transmission. Others who have tested this hypothesis, in field or in laboratory, obtained results that were not consistent with transovarial transmission (Morris and Srihongse 1978, Sprance 1981, Scott et al. 1984, Watts et al. 1987, Crans et al. 1994). While our data suggest that enzootic virus activity begins with Cs. melanura feeding on infectious birds, the reason or mechanism by which birds become viremic remains elusive.

Crans et al. (1994) proposed that nest mites might be responsible for the transmission of EEEV from viremic adults to HY birds. In retrospect, the nest mite hypothesis is a more logical explanation for the 1990 epiornitic monitored at TMBWMA (Howard et al. 2004). This epiornitic began with three EEE viremic HY song sparrows captured and bled on 16 July 1990, 1 wk before the first positive mosquito pool from BG Cs. melanura (Howard et al. 2004). Between 16 July and 3 September, there were 25 isolations of EEEV from 12 bird species (Howard et al. 2004).

An alternative to nest mites is the hypothesis of Morris (1988) of a reservoir mechanism involving a multiphasic EEEV and a sub-population of Cs. melanura that is found only in the deep swamp. There is no evidence of EEEV having non-pathogenic strains (Weaver et al. 1991, Tan et al. 2018), and genome studies on Cs. melanura, including specimens from TMBWMA, have shown no variance in mosquito phenotypes (Soghigian et al. 2018). Additionally, for adult Cs. melanura, the use of the term “deep swamp” is a misnomer. Adult Cs. melanura are an edge species: they mate (Hayes 1958), congregate (Morris and Srihongse 1978, Morris 1984), and move around the edge (Howard et al. 1989). Females disperse from the edge during nocturnal host-seeking (Morris et al. 1980b, Howard et al. 1983).

Alternative hypotheses include viremic adult birds acting as transporters of EEEV during their northward migration (Burkett-Cadena et al. 2022), and the involvement of snakes in the maintenance and transmission of EEEV (Bingham et al. 2012).

During epiornitics in 1988 and 1990 at TMBWMA, when 30 EEEV isolations were made from birds, 15 of these isolations were from song sparrow and common yellowthroat (Howard et al. 2004). Present results suggest a role for cedar waxwing, red-winged blackbird, and yellow-throated vireo. Together, these three studies at TMBWMA suggest a diversity of bird species may be involved. In our experiment, the timing of the first detection of virus-infected individual Cs. melanura was consistent with the timing of the first detection of virus-infected pools of Cs. melanura obtained in the routine surveillance by the county health department. The results of this study present evidence that birds could be the source of the seasonal commencement of EEEV in TMBWMA, contrary to the conclusions in Brown and O’Brien (2011). Additional studies would be required for a more precise understanding of the relationship between birds containing the virus and host-seeking Cs. melanura. Future work might include focus on viremia in younger and older birds when EEEV first appears onsite and experimental studies assessing persistence in target species.

Author Contributions

John J. Howard (Conceptualization [lead], Investigation [equal], Methodology [equal], Visualization [supporting], Writing—original draft [lead], Writing—review & editing [equal]), JoAnne Oliver (Data curation [equal], Investigation [equal], Visualization [equal], Writing—review & editing [equal]), Alan P. Dupuis (Supervision [supporting], Writing—review & editing [supporting]), Kiet A. Ngo (Formal analysis [supporting], Investigation [supporting], Methodology [supporting]), Jessica Stout (Investigation [supporting]), Steven D. Zink (Methodology [supporting], Supervision [supporting]), Elyse Banker (Investigation [supporting], Writing—review & editing [supporting]), Joseph G. Maffei (Investigation [supporting]), Laura D. Kramer (Resources [supporting], Supervision [supporting]), James A. Sherwood (Visualization [lead], Writing—review & editing [equal]), and Alexander T. Ciota (Funding acquisition [lead], Project administration [lead], Resources [lead], Supervision [lead], Writing—review & editing [lead]).

Funding

This work was partially funded by the Cooperative Agreement Number U01CK000509, funded by the Centers for Disease Control and Prevention. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Centers for Disease Control and Prevention.

Conflicts of Interest

None declared.