Prevalence of Powassan Virus Seropositivity Among People with History of Lyme Disease and Non-Lyme Community Controls in the Northeastern United States

Abstract

Introduction:

Lyme disease (LD) affects ∼476,000 people each year in the United States. Symptoms are variable and include rash and flu-like symptoms. Reasons for the wide variation in disease outcomes are unknown. Powassan virus (POWV) is a tick-borne flavivirus that causes disease ranging from asymptomatic infection to encephalitis, neurologic damage, and death. POWV and LD geographic case distributions overlap, with Ixodes species ticks as the common vectors. Clinical ramifications of coinfection or sequential infection are unknown.

Objectives:

This study's primary objective was to determine the prevalence of POWV-reactive antibodies in sera samples collected from previously studied cohorts of individuals with self-reported LD history residing in the Northeastern United States. As a secondary objective, we studied clinical differences between people with self-reported LD history and low versus high POWV antibody levels.

Methods:

We used an enzyme-linked immunosorbent assay (ELISA) to quantify IgG directed at the POWV envelope (E) protein domain III in 538 samples from individuals with self-reported LD history and 16 community controls. The samples were also tested with an ELISA assay to quantify IgG directed at the POWV NS1 protein.

Results:

The percentage of individuals with LD history and possible evidence of POWV exposure varied depending on the assay utilized. We found no significant difference in clinical symptoms between those with low or high POWV IgG levels in the in-house assay. Congruence of the EDIII and NS1 assays was low with only 12% of those positive in the in-house EDIII ELISA testing positive in the POWV NS1 ELISA.

Conclusions:

The results highlight the difficulty in flavivirus diagnostic testing, particularly in the retrospective detection of flavivirus exposure. The findings suggest that a prospective study with symptomatic patients using approved clinical testing is necessary to address the incidence and clinical implications of LD and POWV co-infection or sequential infection.

Introduction

In the United States, Lyme disease (LD) is caused by the bite of an Ixodes species tick infected with Borrelia burgdorferi. LD affects ∼476,000 individuals per year in the United States alone (Kugeler et al., 2021). Recently, LD prevalence has been increasing at an alarming rate as the tick population grows (Kilpatrick et al., 2017). Symptoms of LD include fatigue, fever, headache, and a characteristic erythema migrans rash. If untreated, debilitating arthritis, cardiac conduction abnormalities, or neurologic sequelae may develop. Although antibiotic treatment often cures LD, 10–20% of patients experience residual symptoms. Borrelia persistence or postinfectious factors may contribute to these symptoms (Aucott, 2015; Marques, 2008). However, contributing factors to disease outcomes are poorly understood.

Viruses in the Flavivirus genus of the Flaviviridae family, which include yellow fever virus (YFV), dengue, West Nile virus (WNV), and Zika virus, also cause significant global human disease burden. Flaviviruses are maintained through enzootic cycling between vertebrate and arthropod hosts, typically mosquitoes or ticks (Pierson and Diamond, 2020).

Powassan virus (POWV) is an emerging tick-borne flavivirus threat appearing in the United States, Canada, and Russia (Fatmi et al., 2017). POWV was first identified in a 5-year-old boy from Powassan, Ontario, Canada who died of encephalitis (McLean and Donohue, 1959). Two genetically distinct but serologically indistinguishable lineages of POWV have been identified (Ebel et al., 2001). Although the clinical spectrum of POWV infection requires further study, human disease ranges from asymptomatic infection to neuroinvasive disease with case fatality rates of 10–15%. Fifty percent of survivors experience long-term neurologic sequelae, including headache, memory impairment, and hemiplegia (Fatmi et al., 2017; Kemenesi and Banyai, 2019). Cases in the United States have been increasing in recent years and occur in similar geographical locations as LD.

Since LD and POWV are transmitted by Ixodes species ticks, tick bites potentially expose individuals to both pathogens. In New York and Connecticut, POWV prevalence in ticks has been found to vary depending on the collection site and the study, with the prevalence typically in the 0 to ∼3–4% range. Coinfection of individual ticks with Borrelia species and POWV has also been documented (Aliota et al., 2014; Dupuis et al., 2013; Sanchez-Vicente et al., 2019; Tokarz et al., 2019; Tokarz et al., 2010; Yuan et al., 2020). It is likely that there is wide variability in the prevalence of POWV in ticks depending on the geographical location, as has previously been documented in NY state (Dupuis et al., 2013).

Human coinfection with multiple tick-borne pathogens has also been observed, although the frequency with which this occurs is uncertain. One study conducted in New York found that 11.5% of individuals with early LD were also infected with Babesia microti, but none with POWV (Wormser et al., 2019). A serological study of POWV exposure among Maine residents bitten by Ixodes ticks revealed just one equivocal POWV antibody assay result (0.4% of the samples) (Smith et al., 2019). In a study aimed at validating a serological test panel, 9.4% of samples from individuals with suspected LD tested positive for POWV (Thomm et al., 2018). Another study conducted in Wisconsin found evidence for POWV infection in 9.5% of participants with suspected tick-borne illness; of those with evidence of LD, 17.1% had evidence of POWV coinfection (Frost et al., 2017).

Clinical implications of LD and POWV coinfection are also unknown. Since both pathogens can cause neurological symptoms, the incidence and disease progression of LD and POWV coinfection are important areas of research.

Serologic responses to flavivirus infection result in production of antibodies that bind and neutralize the virus. Serologic testing and, less frequently, direct pathogen detection provide the basis for diagnosis of POWV, although few commercial tests are available in the United States. Diagnosis relies on the detection of IgM in serum or cerebral spinal fluid, confirmed by a plaque reduction neutralization test (PRNT), a fourfold rise in POWV antibody titer in a convalescent sample, or direct detection of POWV by virus culture or molecular detection method (Thomm et al., 2018). Most flavivirus-neutralizing antibodies are directed against the viral envelope (E) protein, which resides in the virion membrane as 90 pairs of head to tail dimers (reviewed in Pierson and Diamond, 2020). Although response to viral infection results in virus-specific antibodies, due to the genetic relatedness of the Flavivirus genus members, antibodies can cross-react with closely related viruses. Thus, Flavivirus infection can be difficult to diagnose, particularly in regions where multiple flaviviruses co-circulate (reviewed in Musso and Despres, 2020; Rathore and St John, 2020).

The flavivirus E protein contains three structural domains: E domain I (EDI), EDII, and EDIII, with EDIII mediating attachment to cells during infection (Barba-Spaeth et al., 2016; Dai et al., 2016; Kostyuchenko et al., 2016; Kuhn et al., 2002; Modis et al., 2003; Mukhopadhyay et al., 2005; Rey et al., 1995; Sirohi et al., 2016; Zhang et al., 2004). Of the antibodies recognizing flavivirus E, those against EDIII are among the most potent neutralizers (Beasley and Barrett, 2002; Crill and Roehrig, 2001; Screaton et al., 2015) and demonstrate greater virus specificity (Cabral-Miranda et al., 2019; Stettler et al., 2016; Zhang et al., 2019). Moreover, an EDIII-based enzyme-linked immunosorbent assay (ELISA) for the Zika flavivirus was recently shown to have high diagnostic sensitivity and specificity (Denis et al., 2019). However, another report suggested extensive cross-reactivity even for antibodies targeting this domain of E (Zaidi et al., 2020).

In this study, we investigated the frequency with which people with self-reported LD history in the Northeastern United States may have been exposed to POWV. Sera and plasma samples from individuals with self-reported LD history and community controls were assessed for serological responses to POWV. To enhance serologic testing specificity, we chose to quantify serum IgG capable of binding to the EDIII of POWV with an in-house ELISA test, using the EDIII from lineage 2 deer tick virus (DTV), whose endemic life cycle involves Ixodes scapularis. Most of the samples were also tested for binding to lineage 1 POWV (strain LB), which infects other Ixodes species ticks (McLean and Larke, 1963; McLean et al., 1964). As a secondary aim, we also compared clinical symptoms of people with self-reported LD history and high or low POWV IgG antibodies to investigate potential differences in disease severity.

Unexpected results for samples in the DTV and LB POWV in-house ELISAs led us to also test the samples using EUROIMMUN's POWV and WNV NS1-based research ELISA tests, as well as an FDA cleared WNV E protein-based IgG ELISA test. Samples that tested positive in the POWV ELISA assays were additionally tested by indirect immunofluorescence assay (IFA) for different flavivirus antibodies to investigate for additional potential cross-reactivities.

Materials and Methods

Ethical approval and registration

This study was approved by the New York State Psychiatric Institutional Review Board at Columbia University Irving Medical Center (protocols #5847, #6805, and #7683) and The Rockefeller University (MMD-0973).

Participants

554 adults ages 18–75 were recruited from multiple sites as part of three different research protocols in the Northeastern United States (Table 1). Participants were categorized as follows:

Table 1.

Overview of Participant Recruitment Sites

Recruitment protocol	Recruitment sites	Year(s) of sample collection	No. of self-reported LD history samples	No. of self-reported community control samples	Total No. of samples
#5847	Norwalk, CT Kinderhook, NY Wappingers Falls, NY Livingston, NY Red Hook, NY	2009	398	0	398
#6805	New York City, NY	2014–2018	71	1	72
#6805	Islesboro, ME	2015–2016	10	0	10
#7683	Pleasant Valley, NY	2018	14	1	15
#7683	Hyde Park, NY	2019	45	14	59
Total			538	16	554

LD, Lyme disease.

LD history: Reported current symptoms and had either acute LD or a history of clinician-prescribed treatment for LD. Participants in protocol #6805 were required to submit prior medical records for enrollment, but participants in protocol #7683 were not. In protocol #5847, participants with recent erythema migrans or disseminated LD were asked to provide supporting documentation after enrollment, while those with more distant or less well-categorized LD were not asked to provide medical records.

Community controls: Reported no diagnosis or treatment for LD or other tick-borne diseases within the past 5 years, no tick bites within the past 6 months, no unstable medical illness or severe flu-like symptoms within the past 6 months, and no immunocompromising condition.

All participants completed self-report questionnaires and blood collection. Most of the sera and plasma samples were aliquoted and stored at −80°C within 24 h of collection. Ten samples collected in Islesboro, ME (protocol #6805) were stored in a −20°C freezer within 12 h of collection and months later transferred to a −80°C freezer.

Clinical measures

All participants at the time of blood collection completed self-report questionnaires about their clinical history and symptoms. Among other measures, participants indicated “yes,” “no,” or “not sure” to experiencing symptoms they believed were due to LD: muscle pain, joint pain, fatigue, memory or other cognitive problems, numbness and/or tingling, headaches, shooting or stabbing pains, sleep disturbance, and mood changes/irritability.

LD testing

Almost all sera were tested for exposure to B. burgdorferi using C6 Peptide ELISA conducted at Medical Diagnostic Laboratories (MDL) in Hamilton Township, NJ. This assay detects IgG and IgM antibodies that react with a peptide corresponding to invariable region 6 of B. burgdorferi s.s. strain B31 (Branda et al., 2011). The cutoff value for a positive C6 ELISA result was 1.1.

POWV testing using the in-house assays

Powassan EDIII was expressed in Escherichia coli and refolded from inclusion bodies as previously described (Robbiani et al., 2017; Sapparapu et al., 2016). Briefly, BL21(DE3) E. coli were transformed with an expression vector encoding residues 301–398 (DTV or LB strain) and a C-terminal 6x-histidine and Avitag. The sequences expressed were MGTTYSMCDKTKFKWKRVPVDSGHDTVVMEVSYTGSDKPCRIPVRAVAHGVPTINVAMLITPNPTIETSGGGFIEMQLPPGDNIIYVGDLSQQWFQKGSHHHHHHGLNDIFEAQKIEWHE (DTV) and MGTTYSMCDKAKFKWKRVPVDSGHDTVVMEVSYTGSDKPCRIPVRAVAHGVPAVNVAMLITPNPTIETNGGGFIEMQLPPGDNIIYVGDLSQQWFQKGSHHHHHHGLNDIFEAQKIEWHE (LB).

Cells were grown to mid-log phase and induced with isopropyl β-D-1-thiogalactopyranoside (IPTG) for 4 h. The cells were lysed, and the insoluble fraction containing inclusion bodies was solubilized in buffer containing 6 M guanidine hydrochloride and 20 mM β-mercaptoethanol, then clarified by centrifugation. The solubilized inclusion bodies were refolded using a quick dilution method into 400 mM L-arginine, 100 mM Tris-HCl (pH 8.0), 2 mM EDTA, 5 and 0.5 mM reduced and oxidized glutathione, and 10% glycerol at 4°C. The refolded protein was filtered and concentrated, then purified by size exclusion chromatography (Superdex 75; Cytiva/GE Healthcare) in 20 mM Tris pH 8.0, 150 mM NaCl, 0.02% NaN₃.

Before use, sera and plasma were thawed, heat inactivated (56°C) for 1 h, and subsequently stored at 4°C. Ninety-six-well ELISA plates were coated with 250 ng EDIII protein per well in phosphate buffered saline (PBS) overnight at room temperature (RT), followed by blocking for 2 h at RT with PBS-T (PBS with 0.05% Tween20) containing 1% bovine serum albumin, 0.1 mM EDTA. PBS-T was used to wash between each step. Serum samples were diluted 1:500 in PBS-T, and 50 μL was added to wells for 1 h at RT. Bound IgG was detected by addition of secondary HRP-conjugated goat anti-human IgG (0.16 mg/mL; Jackson ImmunoResearch) for 1 h at RT followed by development using TMB substrate and reading at 450 nm. The raw values were normalized to no serum controls consisting of only PBS-T. Based on results from samples from control subjects without suspected LD or recent tick exposure, samples with relative optical density (OD) greater than or equal to a cutoff of 1.3 were considered positive.

Anti-POWV and anti-WNV IgG ELISAs

For all samples, we performed NS1-based POWV, NS1-based WNV, and E-based WNV IgG ELISAs according to the manufacturer's instructions (EUROIMMUN, Luebeck, Germany). In addition to the 16 controls from the cohorts mentioned above, 22 additional community control samples from the New York site of recruitment protocol 6805 (Table 1) were tested with the assays.

Briefly, serum or plasma samples were diluted 1:101 and incubated in the wells of the coated microtiter plate. The bound antibody was detected using an enzyme-labeled specific antibody (enzyme conjugate) that catalyzes a color reaction. All necessary reagents were included in the kit, and the required equipment included an incubator set to 37°C and an ELISA plate reader with a wavelength of 450 nm and a reference wavelength between 620 and 650 nm. Semiquantitative results were reported based on a standard curve created with three calibrators. Samples with a concentration <16 relative units (RU)/mL were considered IgG negative, those with a concentration ≥22 RU/mL were considered positive, and those with a concentration ≥16 to <22 RU/mL were recorded as equivocal.

Anti-POWV NS1 IgG and anti-WNV NS1 IgG ELISAs are for research use only in the United States. The anti-WNV E IgG ELISA is FDA cleared.

Immunofluorescence assay

IFA analysis was performed using the Flavivirus Mosaic 1 IIFT Kit (EUROIMMUN), including microscope slides with ten reaction fields each containing four biochips with tick-borne encephalitis virus (TBEV)-, WNV-, Japanese encephalitis virus (JEV)-, and YFV-infected cells. Samples were tested and diluted 1:10 and 1:100 and according to the manufacturer's instructions, followed by assessment of the slides through fluorescence microscopy. Samples testing positive at 1:100 were further diluted 1:1000 and tested. The positive immunofluorescent signals were graded as weak, moderate, or strong. The titer was recorded as the reciprocal of the highest dilution showing a detectable signal. In cases where this signal was still moderate or strong, but the next higher dilution was negative, the titer was estimated to be between that dilution and the next higher dilution according to the manufacturer's instructions. Based on the cutoff validated by the manufacturer, reactivity at a dilution of 1:10 was interpreted as a positive result for TBEV, WNV, and JEV, while for YFV reactivity at a dilution of 1:100 was interpreted as positive.

Statistical analysis

For laboratory measures, Cohen's kappa coefficient was calculated to measure the agreement beyond chance between C6 assay and lineage 1 or lineage 2 EDIII assay. Spearman correlation was used to evaluate the monotonic relationship between lineage 1 and lineage 2 EDIII ELISA results. For clinical measures, Fisher's exact test for categorical data was used. Data analysis was performed using R Studio v.4.0.4, and p values <0.05 were considered statistically significant.

Results

We studied serum or plasma samples from 538 subjects with self-reported LD history and 16 community controls (see Table 2 for demographic information). Of the participants in the LD group, 35.9% tested positive on the C6 Peptide ELISA.

Table 2.

Participant Demographics from All Study Protocols

	Protocol #5847 (n = 398)	Protocol #6805 (n = 82) ^a	Protocol #7683 (n = 74) ^a
Sex = female, n (%)	238 (59.8)	46 (56.1)	47 (63.5)
Ethnicity, n (%)
Hispanic	11 (2.8)	3 (3.7)	3 (4.1)
Non-Hispanic	259 (65.1)	79 (96.3)	71 (95.9)
Not answered	128 (32.2)	—	—
Race, n (%)
White (non-Hispanic)	371 (93.2)	71 (86.6)	67 (90.5)
Black (non-Hispanic)	1 (0.3)	2 (2.4)	2 (2.7)
Asian/Pacific Islander/American Indian/Alaskan Native	9 (2.3)	2 (2.4)	1 (1.4)
Other	2 (0.5)	3 (3.7)	0 (0.0)
Not reported	15 (3.8)	1 (1.2)	1 (1.4)
Age, mean (SD)	55.5 (12.6)	44.0 (16.5)	53.4 (16.4)

Demographic data missing for some participants.

SD, standard deviation.

To assess whether participants may have been exposed to POWV, we initially performed in-house ELISAs on the serum or plasma samples to detect IgG antibodies capable of binding to POWV EDIII, a common target of flavivirus neutralizing antibody responses (Fig. 1). All 554 samples were tested for the presence of antibodies binding to lineage 2 POWV EDIII using bacterially expressed and purified DTV Spooner strain EDIII (Fig. 1A). DTV is carried by the I. scapularis tick and thus of concern for cotransmission during exposure to the LD pathogen. A subset (495 samples) was also tested for antibodies that bind to lineage 1 POWV (LB strain), which is carried by other Ixodes species (Fig. 1B).

FIG. 1.

ELISA quantification of POWV EDIII-binding IgG. Sera or plasma samples diluted 1:500 were tested in ELISA for binding to lineage 2 (A) or lineage 1 (B) POWV EDIII. Based on results from control subjects, samples with relative O.D. values 1.3 were considered positive. Red symbols indicate samples from subjects with a positive C6 LD test, while black symbols are from untested individuals or those with a negative test. Blue filled symbols indicate samples from control subjects. (C) Relative O.D. values for each sample tested in both POWV lineage 1 and lineage 2 EDIII ELISA. Each symbol represents a different sample; dotted lines indicate the positivity cutoff. EDIII, envelope protein domain III; ELISA, enzyme-linked immunosorbent assay; LD, Lyme disease; OD, optical density; POWV, Powassan virus.

Given that the two lineages are serologically indistinguishable, we expected the results to be similar with the two assays and did not attempt to use assay results as an indication of the lineage of POWV to which a participant may have been exposed. Using an OD cutoff of ≥1.3 to define ELISA positivity, 13.2% of the 538 LD cohort participants tested positive for antibodies to lineage 2 POWV (Table 3 and Supplementary Table S1). Of the 493 samples from the LD cohort tested for antibodies to both lineages, 14.4% and 15.0% of individuals tested positive for IgG binding to lineage 2 and lineage 1 EDIII, respectively (Table 3). Based on the LD cohort samples tested with both in-house assays, 113 of 493 (22.9%) had detectable antibodies in either of our EDIII ELISA assays, suggesting possible POWV exposure (Table 3).

Table 3.

Summary of In-House Envelope Protein Domain III Enzyme-Linked Immunosorbent Assay Results

Protocol #	Sample type	Sample No.	DTV EDIII ELISA positive (OD ≥1.3)	LB EDIII ELISA positive (OD ≥1.3)	Either DTV or LB EDIII ELISA positive (OD ≥1.3)
#5847, #6805, #7683 (Pleasant Valley)	LD	493	71 (14.4%)	74 (15.0%)	113 (22.9%)
#5847, #6805, #7683 (Pleasant Valley)	Control	2	0	0	0
#7683 (Hyde Park)	LD	45	0	NT	N/A
#7683 (Hyde Park)	Control	14	0	NT	N/A
ALL	All LD	538	71 (13.2%)	N/A	N/A

DTV, deer tick virus; EDIII, envelope protein domain III; ELISA, enzyme-linked immunosorbent assay; N/A, not applicable; NT, not tested; OD, optical density.

We wondered if samples from LD participants with positive C6 peptide ELISA titers would be more likely to evidence POWV antibodies than those with negative C6 peptide ELISA titers. While there was ∼60% simple agreement between positivity in the C6 assay and either the lineage 1 or lineage 2 EDIII assay, the Cohen's kappa coefficient to account for chance was zero, indicating no agreement. Since lineage 1 and 2 POWV have been considered serologically indistinguishable, we looked at the correlation in the magnitude of the ELISA results for the two lineages (Fig. 1C) but found no correlation (Spearman's correlation r = −0.08, p = 0.40) in positive samples with OD values ≥1.3.

To assess if there were unique clinical features of participants who may have experienced both LD and POWV infections, we compared LD participants from protocol #5847 with positive (OD ≥1.3) results in either lineage in-house POWV EDIII ELISA (n = 109, “LD⁺POWV^high”) versus the same number of LD participants from protocol #5847 with the lowest negative (OD <1.3) POWV ELISA OD values (“LD⁺POWV^low”). We only studied LD participants from protocol #5847 because it enrolled the largest number of participants with consistent clinical measures to allow comparison. Table 4 illustrates participants' responses to the question, “Do you suffer from any of the symptoms listed below that you believe are due to Lyme disease?” We found no significant difference in symptom burden of any of the nine common symptoms of tick-borne illness between the LD⁺POWV^high and LD⁺POWV^low groups.

Table 4.

Symptom Burden Among Participants with the Highest and Lowest In-House Powassan Virus Envelope Protein Domain III Enzyme-Linked Immunosorbent Assay Titers from Protocol #5847

	POWV EDIII ELISA positive (OD ≥ 1.3, LD⁺POWV^high, n = 109)	Lowest OD values of POWV EDIII ELISA negative (OD <1.3, LD⁺POWV^low, n = 109)
Muscle pain
Yes	71 (65.1)	74 (67.9)
No	13 (11.9)	9 (8.3)
Not sure	7 (6.4)	6 (5.5)
Missing	18 (16.5)	20 (18.3)
Joint pain
Yes	80 (73.4)	78 (71.6)
No	11 (10.1)	8 (7.3)
Not sure	2 (1.8)	3 (2.8)
Missing	16 (14.7)	20 (18.3)
Fatigue
Yes	89 (81.7)	91 (83.5)
No	5 (4.6)	2 (1.8)
Not sure	6 (5.5)	5 (4.6)
Missing	9 (8.3)	11 (10.1)
Memory or other cognitive problems
Yes	66 (60.6)	66 (60.6)
No	12 (11.0)	9 (8.3)
Not sure	8 (7.3)	8 (7.3)
Missing	22 (20.2)	26 (23.9)
Numbness and/or tingling
Yes	57 (52.3)	67 (61.5)
No	18 (16.5)	14 (12.8)
Not sure	8 (7.3)	3 (2.8)
Missing	26 (23.9)	25 (22.9)
Headaches
Yes	58 (53.2)	56 (51.4)
No	18 (16.5)	14 (12.8)
Not sure	7 (6.4)	7 (6.4)
Missing	26 (23.9)	32 (29.4)
Shooting or stabbing pains
Yes	42 (38.5)	41 (37.6)
No	27 (24.8)	20 (18.3)
Not sure	5 (4.6)	6 (5.5)
Missing	35 (32.1)	42 (38.5)
Sleep disturbance
Yes	74 (67.9)	63 (57.8)
No	12 (11.0)	10 (9.2)
Not sure	7 (6.4)	11 (10.1)
Missing	16 (14.7)	25 (22.9)
Mood changes/irritability
Yes	65 (59.6)	55 (50.5)
No	13 (11.9)	12 (11.0)
Not sure	8 (7.3)	13 (11.9)
Missing	23 (21.1)	29 (26.6)

POWV, Powassan Virus.

Given that lineage 1 and 2 POWVs are considered to be serologically indistinguishable, it was surprising that we found samples that contained IgG that only bound lineage 1 EDIII, or only lineage 2 EDIII. The two proteins differ by only four amino acids; it is theoretically possible, but unlikely, that this accounts for the results. Similarly, it is possible that exposure to another flavivirus, such as WNV, results in the production of cross-reactive antibodies that differ in different individuals with respect to their binding activity to lineage 1 or 2 EDIII. To investigate this further we tested the samples, along with 22 additional community controls from recruitment protocol 6805, with an NS1-based anti-POWV IgG ELISA (EUROIMMUN, Luebeck, Germany). All samples were also tested for anti-WNV IgG using E and NS1-based ELISA assays. Those samples that were positive in either the in-house EDIII ELISA or the NS1-based anti-POWV IgG ELISA were then tested for IgG antibodies against other flaviviruses, including WNV, TBEV, JEV, and YFV using an IFA. The results are available in Supplementary Table S1.

Of the 113 samples that were positive in either of the in-house ELISA assays, 14 (12%) were positive, considering borderline as positive, with the anti-POWV NS1-based ELISA (Table 5). Of these, 1 (<1%) was also positive in both WNV E-based and WNV NS1-based ELISA assays, suggesting potential cross-reactivity from a previous WNV infection (Table 5).

Table 5.

Anti-Powassan Virus NS1 IgG Enzyme-Linked Immunosorbent Assay Results in Self-Reported Lyme Disease Samples Positive in Either the In-House LB and/or Deer Tick Virus IgG Enzyme-Linked Immunosorbent Assays (n = 113)

		Anti-WNV E or NS1 IgG ELISA
	No. of samples	WNV E+, WNV NS1+	WNV E−, WNV NS1+	WNV E+, WNV NS1−	WNV E+, NS1 BL	WNV E−, WNV NS1 BL	WNV E BL, WNV NS1−	WNV E−, WNV NS1-
Anti-POWV NS1 IgG ELISA
Positive	11	1	3	0	1	4	0	2
BL	3	0	0	0	0	1	0	2
Negative	99	2	0	5	0	0	2	90

BL, borderline; E, envelope; WNV, West Nile virus.

The IFA results showed that out of the 113 in-house POWV EDIII ELISA positive samples (Tables 3 and 5), a total of 22 (19%) were positive in the IFA for other flaviviruses (Supplementary Table S1). Of the 14 that were also positive in the anti-POWV NS1 ELISA assay (Table 5), 3 (21%) were positive in IFA, with IgG against only the closely related TBEV detected by IFA in 2 samples, both of which were negative in the anti-WNV E IgG assay, suggesting that they likely represent true POWV exposure in these individuals. The other sample was positive for IgG reacting with both WNV and JEV; the titer for JEV was higher than for WNV suggesting potential prior JEV infection or vaccination.

Of the 99 samples that were positive in the in-house anti-POWV EDIII ELISA assay but negative in the anti-POWV NS1 ELISA, 19 (19%) were positive in the IFA for IgG against other flaviviruses. Of these, 1 contained IgG reacting with WNV, 1 with JEV, 5 with TBEV, 6 with YFV, 1 with both TBEV and YFV, and 5 with all 4 flaviviruses tested. Of the samples reacting with TBEV, all had negative or borderline reactivity in the anti-WNV E protein ELISA assay. Of the five samples reacting with all four flaviviruses in IFA, four were also positive with the anti-WNV E-based IgG ELISA, and for two of them, the anti-WNV NS1-based IgG ELISA was positive as well. These findings suggest the potential of previous infection or vaccination against some of these flaviviruses leading to a broadly cross-reactive immune response.

Of the 425 samples from participants with a history of LD that were negative in the in-house anti-lineage 2 EDIII IgG ELISA and negative or not tested in the in-house anti-lineage 1 EDIII IgG ELISA (Table 6), 7 (1.6%) tested positive in the anti-POWV NS1 IgG ELISA; 1 of these also tested positive in the anti-WNV E and NS1 IgG ELISA assays (Table 6). None of the seven tested positive in the flavivirus IFA. One (2.6%) of 38 community control samples tested positive in the anti-POWV NS1 ELISA (IgG) (Table 7) but was negative in the IFA against other flaviviruses (Supplementary Table S1).

Table 6.

Anti-Powassan Virus NS1 IgG Enzyme-Linked Immunosorbent Assay Results in Self-Reported Lyme Disease Samples Negative in the In-House LB and Deer Tick Virus IgG Enzyme-Linked Immunosorbent Assays (n = 425)

		Anti-WNV E or NS1 IgG ELISA
	No. of samples	WNV E+, WNV NS1+	WNV E−, WNV NS1+	WNV E+, WNV NS1−	WNV E+, WNV NS1 BL	WNV E−, WNV NS1 BL	WNV E BL, WNV NS1−	WNV E−, WNV NS1−
Anti-POWV NS1 IgG ELISA
Positive	4	1	1	0	0	0	0	2
BL	3	0	0	0	0	0	0	3
Negative	418	1	2	13	1	2	5	394

Forty-five of the samples were not tested in the in-house LB IgG ELISA.

Table 7.

Anti-Powassan Virus NS1 IgG Enzyme-Linked Immunosorbent Assay Results in Controls (n = 38)

		Anti-WNV E or NS1 IgG ELISA
	No. of samples	WNV E+, WNV NS1+	WNV E−, WNV NS1+	WNV E+, WNV NS1−	WNV E+, WNV NS1 BL	WNV E−, WNV NS1 BL	WNV E BL, WNV NS1−	WNV E−, WNV NS1−
Anti-POWV NS1 IgG ELISA
Positive	1	0	1	0	0	0	0	0
BL	0	0	0	0	0	0	0	0
Negative	37	0	1	3	0	0	1	32

Overall, depending on the antigen used in the assay to determine POWV IgG seropositivity, the prevalence of POWV antibodies in the cohort of participants with a self-reported history of LD ranged from 3.9% (21/538) to 22.9% (Tables 3 and 8).

Table 8.

Powassan Virus Seroprevalence Based on EUROIMMUN NS1-Based IgG Enzyme-Linked Immunosorbent Assay

	Self-reported LD patients	Controls
Anti-POWV NS1 IgG ELISA	21/538 = 3.9%	1/38 = 2.6%

Discussion

In our study assessing potential POWV exposure among people with self-reported LD history from the Northeastern United States, we found that 13.2% had detectable IgG recognizing the lineage 2 (DTV) POWV EDIII by ELISA. Although lineage 2 and lineage 1 POWVs are considered serologically indistinguishable, for a subset of individuals tested with both lineage 1 and lineage 2 EDIII ELISAs we found individuals with detectable IgG recognizing POWV lineage 1 or lineage 2 only, or both. We were unable to determine a reason for positive results against one lineage and not the other. That the EDIII region between the two POWV lineages is immunologically distinct is supported by the finding that some human monoclonal antibodies from individuals infected with TBEV have been found to bind POWV lineage 1 or lineage 2 EDIII, but not both (Agudelo et al., 2021). However, antibody responses in infected individuals would be expected to be polyclonal in nature. Without more data, any potential explanation would be speculation; understanding why some individuals had detectable IgG recognizing only one of the lineages requires further study.

Taking results from both in-house assays into account, 22.9% of these individuals with self-reported LD history had results suggestive of prior infection with POWV (Fig. 1 and Table 3). Regarding potential clinical implications of LD and POWV coinfection, we found no difference in disease severity between our LD⁺POWV^high and LD⁺POWV^low groups.

Our finding of 22.9% potential POWV infection rate among patients with a history of LD is similar to that of a previously published study in which 17.1% of 41 patients with evidence of LD from the upper Midwest United States had serological evidence of acute POWV infection (Frost et al., 2017). In the Frost et al. study, samples were derived from individuals with suspected tick-borne disease (samples for which a B. burgdorferi serological test was ordered, n = 95) and individuals undergoing routine chemistry screening (n = 50) over a 2-month period in 2015. Assessment included screening for TBEV complex by IgM and IgG enzyme immunoassays, which were followed up with commercially available POWV-specific IgG and IgM immunofluorescence antibody assays. Positive samples were further screened to assess cross-reactivity in an IFA, including a panel of eight flaviviruses.

In comparison, our study examined a larger number of samples from people with self-reported LD history from a different geographical area. Unlike the Frost et al. study, which utilized laboratory samples for which a serologic test for B. burgdorferi had been ordered (categorized for the study as “suspected tick-borne disease”), our study enrolled individuals with either acute LD or a history of having been treated for LD months to years before sample collection. Moreover, our participants may not have been exposed to POWV at the same time as B. burgdorferi; concurrent active coinfection cannot be determined from our study. Like the Frost et al. study, our study was limited by the lack of paired serum samples to assess change in anti-POWV IgG titers.

Flaviviruses are known for their high degree of serological cross-reactivity, meaning that antibodies produced against one flavivirus can recognize and react with other flaviviruses. For instance, POWV has been found to exhibit cross-reactivity with other tick-borne flaviviruses, as well as mosquito-borne flaviviruses (VanBlargan et al., 2021). To investigate the potential for cross-reactivity with other flaviviruses endemic to the Northeast region, we used an NS1-based anti-POWV IgG ELISA (EUROIMMUN) to test our samples. The NS1 antigen is often more specific than the E protein (Fisher et al., 2023; Mora-Cardenas et al., 2020). Furthermore, samples were further tested using assays from EUROIMMUN for IgG antibodies against WNV using two different ELISAs (one based on the E protein and the other based on NS1). Those that tested positive in either the in-house EDIII ELISA or the anti-POWV IgG NS1-based ELISA were tested against a panel of flaviviruses using an IFA based on infected cells (TBEV, WNV, JEV, and YFV). TBEV is the most closely related to POWV with the highest degree of cross-reactivity (VanBlargan et al., 2021).

Of the samples tested, there was only a 12% agreement between those that tested positive in either the in-house EDIII ELISA or the anti-POWV NS1-based IgG ELISA. However, two samples were found to be positive across multiple methods, including both the in-house ELISAs and the anti-POWV NS1-based IgG ELISA and the anti-TBEV IgG IFA. These samples could potentially represent true positive results, because different methods detected the presence of antibodies in the same samples.

The reasons for incongruent results between the in-house EDIII-based IgG ELISA and the NS1-based IgG ELISA are uncertain and likely multifactorial. The most obvious difference is that the in-house assay used EDIII of the virion-associated E protein as the target antigen, while the EUROIMMUN target antigen was NS1, which is not found in the virion, but has been described as being more specific than E protein-based assays (Fisher et al., 2023, Mora-Cardenas et al., 2020). The two assays differed in other aspects. The EUROIMMUN NS1-based assay used lineage 1 POWV-based antigen expressed in mammalian cells, while the in-house assay utilized bacterially expressed and purified DTV or lineage 1 POWV EDIII proteins. How the ELISA plates were prepared for each of the assays also differed in blocking, stabilization, sample dilution, and incubation schemes. Moreover, the assays themselves were conducted using different reagents, incubation times, and temperatures, all of which could influence specificity and sensitivity. Cutoffs for positivity for each of the assays were also not determined using the same set of samples.

If possible, future studies should utilize tests that have been characterized using clearly established positive and negative control samples to characterize sensitivity and specificity for antibodies directed against the flavivirus of interest.

We also observed that a few samples positive for anti-POWV IgG in the NS1-based ELISA also were positive for the anti-WNV NS1-based IgG ELISA but were negative in the anti-WNV E protein-based IgG ELISA and negative in the IFA for all flaviviruses tested. These results are not plausible in the context of a flavivirus infection. The cause of these results could not be further clarified in the context of this publication.

Since the samples were collected from patients living in a WNV-endemic area, it is likely that some of the patients had prior exposure to WNV. As expected, we detected several samples that were positive for POWV using either the in-house EDIII ELISA or the anti-POWV NS1-based IgG ELISA from EUROIMMUN, as well as both anti-WNV NS1- and E-based IgG EUROIMMUN ELISAs. In addition, the IFA results showed a few patients with high titers for multiple flaviviruses, which may indicate previous infections or vaccination, such as for JEV, TBEV, or YFV. We do not have information about the study participants regarding travel or vaccinations. These findings highlight the importance of considering the possibility of cross-reactivity when interpreting serological test results in vaccinated individuals or those living in regions endemic for multiple flaviviruses.

Interpretation of our clinical data, too, is limited by aforementioned factors—notably, potential cross-reactivity of POWV assays with other flaviviruses and variable duration from diagnosis of LD to sample acquisition. Moreover, only a subset of our participants with self-reported LD history provided prior medical records for corroboration. Because the disease spectrum of both B. burgdorferi and POWV infection is broad (reviewed in Corrin et al., 2018; Schoen, 2020), it is possible that clinical manifestations of LD and POWV coinfection are also highly variable. This question awaits the results of an appropriately powered prospective study.

Despite these caveats, our findings may inform larger studies aimed at determining the incidence and ramifications of LD and POWV coinfection. Due to likely cross-reactivity of antibody responses to other flaviviruses, the real incidence of POWV infection is expected to be lower than our estimate based on in-house assay results of 22.9%. Using a likely less sensitive but more specific assay for detecting anti-POWV NS1 IgG, we found evidence for a 3.9% incidence of POWV infection in individuals with self-reported LD history. The true incidence likely is somewhere in between. Enrollment in future studies should be adjusted accordingly to ensure adequate statistical power. Moreover, future studies should consider a prospective design, where participants are enrolled at the time of LD infection and are followed longitudinally, using clinically approved diagnostic tests and standardized symptom measures. Finally, future studies should consider more stringent inclusion criteria to limit enrollment to participants with definitive LD.

Footnotes

Acknowledgments

The authors thank Jennifer R. Keefe and Yu E. Lee (CalTech) for providing recombinant EDIII proteins for use in ELISA. The authors thank Aileen O'Connell, Anesta Webson, Sonja Shirley, and Mary Ellen Castillo for logistical and technical assistance.

Authors' Contributions

The authors made the following contributions to warrant authorship: study conceptualization and design: M.R.M., B.A.F., and D.F.R.; data acquisition: T.K., A.A.P., A.J., L.M., M.S., O.K., E.L.; data analysis: T.K., A.A.P., L.M., C.S.J., M.A., M.K., M.S., O.K., E.L.; data interpretation: L.M., C.S.J., M.K., D.F.R., B.A.F., M.R.M., M.S., O.K., E.L., L.M. and M.R.M. wrote the manuscript. All authors contributed to editing the manuscript and approving the final submitted version and agree to be accountable for all aspects of the work.

Author Disclosure Statement

O.K., M.S., and E.L. are EUROIMMUN employees. Otherwise, the authors have no competing or financial interests to declare.

Funding Information

This work was supported, in part, by the National Institutes of Health (NIH) through National Institute of Allergy and Infectious Disease grants R21AI142010 and P01AI138938 and by grant UL1 TR001866 from the National Center for Advancing Translational Sciences NIH Clinical and Translational Science Award (CTSA) program. This work was also supported by the Lyme and Tick-borne Diseases Research Center at Columbia University Irving Medical Center established by the Global Lyme Alliance, Inc., and the Lyme Disease Association, Inc. The funders had no role in study design, data interpretation, or decision to publish.

Supplementary Material

Supplementary Table S1

References

Aliota

, Dupuis

, 2nd, Wilczek

, et al. The prevalence of zoonotic tick-borne pathogens in Ixodes scapularis collected in the Hudson Valley, New York State. Vector Borne Zoonotic Dis, 2014; 14(4):245–250; doi: 10.1089/vbz.2013.1475

Aucott

. Posttreatment Lyme disease syndrome. Infect Dis Clin North Am, 2015; 29(2):309–323; doi: 10.1016/j.idc.2015.02.012

Barba-Spaeth

, Dejnirattisai

, Rouvinski

, et al. Structural basis of potent Zika-dengue virus antibody cross-neutralization. Nature, 2016; 536(7614):48–53; doi: 10.1038/nature18938

Beasley

, Barrett

. Identification of neutralizing epitopes within structural domain III of the West Nile virus envelope protein. J Virol, 2002; 76(24):13097–13100.

Branda

, Linskey

, Kim

, et al. Two-tiered antibody testing for Lyme disease with use of 2 enzyme immunoassays, a whole-cell sonicate enzyme immunoassay followed by a VlsE C6 peptide enzyme immunoassay. Clin Infect Dis, 2011; 53(6):541–547; doi: 10.1093/cid/cir464

Cabral-Miranda

, Lim

, Mohsen

, et al. Zika virus-derived E-DIII protein displayed on immunologically optimized VLPs induces neutralizing antibodies without causing enhancement of dengue virus infection. Vaccines (Basel), 2019; 7(3):72; doi: 10.3390/vaccines7030072

Corrin

, Greig

, Harding

, et al. Powassan virus, a scoping review of the global evidence. Zoonoses Public Health, 2018; 65(6):595–624; doi: 10.1111/zph.12485

Crill

, Roehrig

. Monoclonal antibodies that bind to domain III of dengue virus E glycoprotein are the most efficient blockers of virus adsorption to Vero cells. J Virol, 2001; 75(16):7769–7773; doi: 10.1128/JVI.75.16.7769-7773.2001

Dai

, Song

, Lu

, et al. Structures of the Zika virus envelope protein and its complex with a flavivirus broadly protective antibody. Cell Host Microbe, 2016; 19(5):696–704; doi: 10.1016/j.chom.2016.04.013

10.

Denis

, Attoumani

, Gravier

, et al. High specificity and sensitivity of Zika EDIII-based ELISA diagnosis highlighted by a large human reference panel. PLoS Negl Trop Dis, 2019; 13(9):e0007747; doi: 10.1371/journal.pntd.0007747

11.

Dupuis

, 2nd, Peters

, Prusinski

, et al. Isolation of deer tick virus (Powassan virus, lineage II) from Ixodes scapularis and detection of antibody in vertebrate hosts sampled in the Hudson Valley, New York State. Parasit Vectors, 2013; 6:185; doi: 10.1186/1756-3305-6-185

12.

Ebel

, Spielman

, Telford

. Phylogeny of North American Powassan virus. J Gen Virol, 2001; 82(Pt 7):1657–1665; doi: 10.1099/0022-1317-82-7-1657

13.

Fatmi

, Zehra

, Carpenter

. Powassan virus—A new reemerging tick-borne disease. Front Public Health, 2017; 5:342; doi: 10.3389/fpubh.2017.00342

14.

Fisher

, Lustig

, Sklan

, et al. The role of NS1 protein in the diagnosis of flavivirus infections. Viruses, 2023; 15(2):572; doi: 10.3390/v15020572

15.

Frost

, Schotthoefer

, Thomm

, et al. Serologic evidence of Powassan virus infection in patients with suspected Lyme disease. Emerg Infect Dis, 2017; 23(8):1384–1388; doi: 10.3201/eid2308.161971

16.

Kemenesi

, Banyai

. Tick-borne flaviviruses, with a focus on Powassan virus. Clin Microbiol Rev, 2019; 32(1):e00106-17; doi: 10.1128/CMR.00106-17

17.

Kilpatrick

, Dobson

ADM

, Levi

, et al. Lyme disease ecology in a changing world: Consensus, uncertainty and critical gaps for improving control. Philos Trans R Soc Lond B Biol Sci, 2017; 372(1722):20160117; doi: 10.1098/rstb.2016.0117

18.

Kostyuchenko

, Lim

, Zhang

, et al. Structure of the thermally stable Zika virus. Nature, 2016; 533(7603):425–428; doi: 10.1038/nature17994

19.

Kugeler

, Schwartz

, Delorey

, et al. Estimating the frequency of Lyme disease diagnoses, United States, 2010–2018. Emerg Infect Dis, 2021; 27(2):616–619; doi: 10.3201/eid2702.202731

20.

Kuhn

, Zhang

, Rossmann

, et al. Structure of dengue virus: Implications for flavivirus organization, maturation, and fusion. Cell, 2002; 108(5):717–725; doi: 10.1016/s0092-8674(02)00660-8

21.

Marques

. Chronic Lyme disease: A review. Infect Dis Clin North Am, 2008; 22(2):341–360, vii–viii; doi: 10.1016/j.idc.2007.12.011

22.

McLean

, Best

, Mahalingam

, et al. Powassan virus: Summer infection cycle, 1964. Can Med Assoc J, 1964; 91:1360–1362.

23.

McLean

, Donohue

. Powassan virus: Isolation of virus from a fatal case of encephalitis. Can Med Assoc J, 1959; 80(9):708–711.

24.

McLean

, Larke

. Powassan and Silverwater viruses: Ecology of two Ontario arboviruses. Can Med Assoc J, 1963; 88:182–185.

25.

Modis

, Ogata

, Clements

, et al. A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc Natl Acad Sci U S A, 2003; 100(12):6986–6991; doi: 10.1073/pnas.0832193100

26.

Mora-Cardenas

, Aloise

, Faoro

, et al. Comparative specificity and sensitivity of NS1-based serological assays for the detection of flavivirus immune response. PLoS Negl Trop Dis, 2020; 14(1):e0008039; doi: 10.1371/journal.pntd.0008039

27.

Mukhopadhyay

, Kuhn

, Rossmann

. A structural perspective of the flavivirus life cycle. Nat Rev Microbiol, 2005; 3(1):13–22; doi: 10.1038/nrmicro1067

28.

Musso

, Despres

. Serological diagnosis of flavivirus-associated human infections. Diagnostics (Basel), 2020; 10(5):302; doi: 10.3390/diagnostics10050302

29.

Pierson

, Diamond

. The continued threat of emerging flaviviruses. Nat Microbiol, 2020; 5(6):796–812; doi: 10.1038/s41564-020-0714-0

30.

Rathore

APS

, St John

. Cross-reactive immunity among flaviviruses. Front Immunol, 2020; 11:334; doi: 10.3389/fimmu.2020.00334

31.

Rey

, Heinz

, Mandl

, et al. The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature, 1995; 375(6529):291–298; doi: 10.1038/375291a0

32.

Robbiani

, Bozzacco

, Keeffe

, et al. Recurrent potent human neutralizing antibodies to Zika virus in Brazil and Mexico. Cell, 2017; 169(4):597.e11–609.e11; doi: 10.1016/j.cell.2017.04.024

33.

Sanchez-Vicente

, Tagliafierro

, Coleman

, et al. Polymicrobial nature of tick-borne diseases. mBio, 2019; 10(5):e02055-19; doi: 10.1128/mBio.02055-19

34.

Sapparapu

, Fernandez

, Kose

, et al. Neutralizing human antibodies prevent Zika virus replication and fetal disease in mice. Nature, 2016; 540(7633):443–447; doi: 10.1038/nature20564

35.

Schoen

. Challenges in the diagnosis and treatment of Lyme disease. Curr Rheumatol Rep, 2020; 22(1):3; doi: 10.1007/s11926-019-0857-2

36.

Screaton

, Mongkolsapaya

, Yacoub

, et al. New insights into the immunopathology and control of dengue virus infection. Nat Rev Immunol, 2015; 15(12):745–759; doi: 10.1038/nri3916

37.

Sirohi

, Chen

, Sun

, et al. The 3.8 A resolution cryo-EM structure of Zika virus. Science, 2016; 352(6284):467–470; doi: 10.1126/science.aaf5316

38.

Smith

Jr. , Elias

, Cavanaugh

, et al. Seroprevalence of Borrelia burgdorferi, B. miyamotoi, and Powassan virus in residents bitten by Ixodes ticks, Maine, USA. Emerg Infect Dis, 2019; 25(4):804–807; doi: 10.3201/eid2504.180202

39.

Stettler

, Beltramello

, Espinosa

, et al. Specificity, cross-reactivity, and function of antibodies elicited by Zika virus infection. Science, 2016; 353(6301):823–826; doi: 10.1126/science.aaf8505

40.

Thomm

, Schotthoefer

, Dupuis

, 2nd, et al. Development and validation of a serologic test panel for detection of Powassan virus infection in U.S. patients residing in regions where Lyme disease is endemic. mSphere, 2018; 3(1):e00467-17; doi: 10.1128/mSphere.00467-17

41.

Tokarz

, Jain

, Bennett

, et al. Assessment of polymicrobial infections in ticks in New York state. Vector Borne Zoonotic Dis, 2010; 10(3):217–221; doi: 10.1089/vbz.2009.0036

42.

Tokarz

, Tagliafierro

, Sameroff

, et al. Microbiome analysis of Ixodes scapularis ticks from New York and Connecticut. Ticks Tick Borne Dis, 2019; 10(4):894–900; doi: 10.1016/j.ttbdis.2019.04.011

43.

VanBlargan

, Errico

, Kafai

, et al. Broadly neutralizing monoclonal antibodies protect against multiple tick-borne flaviviruses. J Exp Med, 2021; 218(5):e20210174; doi: 10.1084/jem.20210174

44.

Wormser

, McKenna

, Scavarda

, et al. Co-infections in persons with early Lyme disease, New York, USA. Emerg Infect Dis, 2019; 25(4):748–752; doi: 10.3201/eid2504.181509

45.

Yuan

, Llanos-Soto

, Gangloff-Kaufmann

, et al. Active surveillance of pathogens from ticks collected in New York State suburban parks and schoolyards. Zoonoses Public Health, 2020; 67(6):684–696; doi: 10.1111/zph.12749

46.

Zaidi

, Cedillo-Barron

, Gonzalez

YAME

, et al. Serological tests reveal significant cross-reactive human antibody responses to Zika and Dengue viruses in the Mexican population. Acta Trop, 2020; 201:105201; doi: 10.1016/j.actatropica.2019.105201

47.

Zhang

, Qu

, Li

, et al. Yeast-produced subunit protein vaccine elicits broadly neutralizing antibodies that protect mice against Zika virus lethal infection. Antiviral Res, 2019; 170:104578; doi: 10.1016/j.antiviral.2019.104578

48.

Zhang

, Zhang

, Ogata

, et al. Conformational changes of the flavivirus E glycoprotein. Structure, 2004; 12(9):1607–1618; doi: 10.1016/j.str.2004.06.019

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.10 MB

0.00 MB