A West Nile Virus DNA Vaccine Induces Neutralizing Antibody in Healthy Adults during a Phase 1 Clinical Trial

Abstract

Background

West Nile virus (WNV) is a mosquitoborne flavivirus that can cause severe meningitis and encephalitis in infected individuals. We report the safety and immunogenicity of a WNV DNA vaccine in its first phase 1 human study.

Methods

A single-plasmid DNA vaccine encoding the premembrane and the envelope glycoproteins of the NY99 strain of WNV was evaluated in an open-label study in 15 healthy adults. Twelve subjects completed the 3-dose vaccination schedule, and all subjects completed 32 weeks of evaluation for safety and immunogenicity. The development of a vaccine-induced immune response was assessed by enzyme-linked immunosorbant assay, neutralization assays, intracelluar cytokine staining, and enzyme-linked immunospot assay.

Results

The vaccine was safe and well tolerated, with no significant adverse events. Vaccine-induced T cell and antibody responses were detected in the majority of subjects. Neutralizing antibody to WNV was detected in all subjects who completed the 3-dose vaccination schedule, at levels shown to be protective in studies of horses, an incidental natural host for WNV.

Conclusions

Further assessment of this DNA platform for human immunization against WNV is warranted.

Trial registration

ClinicalTrials.gov identifier: NCT00106769.

West Nile virus (WNV) is a vectorborne member of the Flavivirus genus, which includes several clinically and economically important human pathogens, such as yellow fever virus, 4 serotypes of dengue virus, and Japanese encephalitis virus (JEV). WNV was initially isolated in Uganda in 1937 and was first recognized in the United States in 1999, when it caused an epidemic of encephalitis and meningitis in New York City. WNV has since spread across North America [1] and into portions of Central and South America [2, 3]. In 2005, there were 3000 cases of WNV infection in humans reported in the United States, and, as of 11 December 2006, 4052 cases had been reported for the year in 42 states [4, 5].

WNV naturally exists as an enzootic infection in mosquitoes and birds, although a large number of incidental hosts have been identified, including humans, horses, and alligators [6]. The principal form of transmission to humans is from the bite of an infected mosquito. There is no evidence of person-to-person spread, but transmission of WNV has occurred by blood transfusion, by organ transplantation, by breast-feeding, transplacentally, and in the laboratory [1, 7].

WNV infections in humans can be severe but are often subclinical or may present as a mild to moderate febrile illness. Approximately 1 in 150 infected persons have a serious illness with involvement of the central nervous system [8], and, although cases of severe WNV infection (including meningitis and encephalitis) have been reported in otherwise healthy young adults [9], the risk of severe disease and death increases in elderly persons and in immunocompromised individuals [2]. The first known genetic risk factor for severe and fatal WNV infection has recently been described in patients with the defective CCR5 allele, CCR5Δ32, and is possibly related to a lack of CCR5 regulation of WNV-infected leukocytes [10, 11]. Although intravenous immunoglobulin has been investigated as a therapeutic intervention for severe cases of WNV [12], the standard of care for WNV infection is supportive.

Several lines of evidence suggest an important role for antibody in protection from and clearance of flavivirus infections [13]. Surface envelope (E) proteins are the primary target for the humoral response against flavivirus infection. The mature WNV virion is composed of 180 copies of the E protein, arranged with an unusual herringbone icosahedral symmetry. The E protein is thought to mediate interactions with the cell surface and promotes fusion between viral and cellular membranes. In addition, virions incorporate a second protein, the premembrane (prM) protein, which is cleaved during virion maturation into a smaller virion-associated membrane (M) peptide. Of interest, expression of the prM and E proteins in cells results in the formation and release of a virus-like subviral particle that shares many of the structural, antigenic, and functional properties of mature infectious virus.

Since its introduction into North America, WNV infection of horses has become a significant problem, with a 30%−40% mortality rate and as many as 5000 cases of sick horses per year since 1999 [14, 15]. A formalin-inactivated whole-virus WNV vaccine for the prevention of WNV infection in horses (Innovator; Fort Dodge Animal Health) has been available since 2002 and is being evaluated in other species [7, 16]. A recombinant canarypox-based vaccine (Recombitek; Merial) expressing the prM and E genes has been shown to induce neutralizing antibodies and protect horses against experimental challenge with WNV [17] and was approved for veterinary use in the United States in 2004. A yellow fever–WNV chimeric vaccine was also licensed by the US Department of Agriculture (USDA) in 2007 for use in horses [18]. The equine DNA vaccine, pCBWN (Fort Dodge Animal Health in collaboration with the Centers for Disease Control and Prevention [CDC]), also encodes for the prM and E proteins from WNV and, in WNV challenge studies in both mice and horses, elicits neutralizing antibody and confers protection against viremia [19]. In July 2005, this DNA vaccine was licensed by the USDA for the prevention of WNV infection in horses and represents the first license granted for a DNA vaccine for use in animals [14]. The DNA vaccine licensed for use in horses has recently been shown to protect endangered California condors from WNV infection [20].

Several approaches for the development of a vaccine in humans are also being investigated. A live attenuated WNV vaccine based on chimeric flavivirus technology (ChimeriVax–West Nile; Acambis) has completed phase 1 and is currently in phase 2 trials [21]. Several other vaccine approaches have shown promise in animal models, including a recombinant canarypox vector expressing the prM and E genes [17], a recombinant subunit vaccine expressing E and nonstructural protein 1, and a live attenuated recombinant measles vaccine expressing the secreted form of the E glycoprotein of WNV [22].

Candidate DNA vaccines have been evaluated in phase 1 studies of healthy adults and have been shown to be safe and immunogenic for other pathogens, including HIV and Ebola virus [23, 24]. A promising vaccine approach for the prevention of WNV infection is a single-plasmid recombinant DNA vaccine encoding the WNV prM and E proteins described in the present report of a phase 1 clinical trial. This approach is unique among the potential WNV candidate vaccines being investigated for use in humans in that it eliminates the concern typically associated with live attenuated vaccines. It neither induces nor is susceptible to antivector immunity, which is a concern in vector-based gene-delivery vaccine candidates. In addition, it encodes for proteins that have the potential to form a multivalent antigen in vivo, has the ability to induce T cell and neutralizing antibody responses, and it is very similar to the DNA vaccine licensed for the prevention of WNV infection in horses [19].

SUBJECTS, MATERIALS, AND METHODS

Study design

The Vaccine Research Center (VRC) 302 protocol was a single-site, open-label, phase 1 study to examine the safety, tolerability, and immune response to an investigational recombinant DNA WNV vaccine. Healthy adult subjects 18−50 years of age who were negative for WNV IgG were recruited at the VRC, National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH). The experimental guidelines of the US Department of Health and Human Services were followed in the conduct of the clinical research, and the protocol was approved by the NIAID Institutional Review Board. Fifteen subjects were enrolled between 18 April and 7 October 2005. Vaccine was administered on study days 0, 28 (±7 days), and 56 (±7 days), with at least 21 days between injections.

Vaccine (VRC-WNVDNA017−00-VP) was administered at a 4-mg dose via intramuscular injection in the lateral deltoid by use of the Biojector 2000 Needle-Free Injection Management System. The dose and route studied in this trial were based on clinical trials of VRC DNA vaccines for other pathogens [23-25]. Adverse reactions and subject safety were evaluated by laboratory and clinical evaluations at scheduled study visits. Adverse events were coded using the Medical Dictionary for Regulatory Activities (MedDRA), and the severity of adverse events was graded using a scale of 0 to 5 (based on the NIH Division of AIDS table for grading the severity of adverse events; version 1.0; December 2004). Local and systemic solicited reactogenicity—including pain, erythema, swelling, myalgia, malaise, headache, chills, nausea, and temperature—was collected by subject self-report on 5-day diary cards after each injection. Subjects were followed for a total of 32 weeks, and the study was completed in July 2006.

Vaccine

VRC-WNVDNA017−00-VP is composed of a single, closed, circular plasmid DNA macromolecule (VCL-6428) constructed to produce the prM and E proteins of the WNV E glycoprotein. The plasmid was based on an analogous construct shown to protect mice and horses from virus challenge [19]. The plasmid encodes for a single polypeptide encompassing a modified signal sequence from JEV fused upstream of the WNV prM and E coding sequences cloned into the expression vector VR-1012 (CMV backbone). This backbone encoding the CMV promoter and 5' intron A was also used to construct a 4-plasmid HIV multiclade DNA vaccine previously shown to be safe and immunogenic in healthy adults [23]. The coding sequence for the first 6 aa was optimized for expression in humans by using commonly used codons. The WNV prM and E sequence in VCL-6428 is identical to that in the NY99 human WNV isolate. In vitro expression of this gene product results in the formation of noninfectious virus-like particles called “subviral particles” (SVPs). The plasmid in this vaccine is incapable of replication in animal cells and does not permit the generation of an infectious virion even if recombination or gene duplication were to occur.

This DNA vaccine has been produced in bacterial cell cultures containing a kanamycin-selection medium. The process involved Escherichia coli fermentation, purification, and formulation as a sterile, liquid, injectable dosage form for intramuscular injection. After growth of bacterial cells harboring the plasmid, the plasmid DNA was purified from cellular components.

Vaccine was prepared under current good manufacturing practice conditions by Vical. The vaccine was manufactured at a 4-mg dose in PBS and met lot-release specifications before use in the clinical trial.

Measurement of antibody responses by ELISA

Duplicate wells of serial dilutions of volunteer serum samples were incubated for 1 h at 37°C on commercially available WNV recombinant antigen–coated plates (Focus Technologies), followed by horseradish peroxidase–conjugated antibody (30 min at room temperature) and tetramethylbenzidine substrate (10 min at room temperature). Color development was stopped by the addition of 0.9 mol/L sulfuric acid, and plates were read within 30 min at 450 nm on the Molecular Devices Spectramax 384-plus ELISA plate reader. The mean optical density for each dilution was corrected for the mean optical density of the same dilution of the preimmunization sample. End-point titers for each volunteer were established as the last dilution with a corrected preimmunization OD >0.2.

Measurement of neutralizing antibody responses by plaque reduction neutralization (PRNT)

PRNT was performed with Vero cells, as previously described [26], using NY99−6480 virus. End-point titers were determined for 50% and 75% plaque reduction.

Measurement of antibody-mediated neutralization by use of reporter-virus particles (RVPs)

WNV RVPs composed of the structural proteins of the NY99 strain of WNV and a subgenomic replicon were produced by complementation in BHK-21 cells, as described elsewhere [27]. Antibody-mediated neutralization was measured using a Raji B lymphoblastoid cell line that expresses the WNV attachment factor CD209L (DC-SIGNR), as described elsewhere [28]. WNV RVPs were incubated with serial 3-fold dilutions of volunteer serum at room temperature for 2 h and then added to 5 × 10⁴ cells plated on the day of the assay. The concentration of RVPs used in each experiment was validated empirically to establish assay conditions of antibody access at all informative points of the dose-response profile for each volunteer. Infection was monitored as a function of green fluorescent protein expression (encoded by the replicon) at 48 h after infection by flow cytometry. Data were analyzed by curve fitting and nonlinear regression in order to describe the shape of the dose response and calculate an EC₅₀. Data are presented as the reciprocal dilution of serum corresponding to the EC₅₀ and are adjusted to consider the final 300-μL volume of the neutralization reaction.

Measurement of T cell responses by enzyme-linked immunospot (ELISpot) assay

ELISpot assay was performed on subject samples at baseline and at weeks 2, 6, 8, 10, 12, and 32, as described elsewhere [24]. Cells were stimulated overnight with vaccine insert–specific peptide pools (WNV-E and WNV-M) at 2 × 10⁵ cells/well. Results are expressed as mean spot-forming cells per million peripheral blood mononuclear cells.

Measurement of T cell responses by intracelluar cytokine staining (ICS)

CD4⁺ and CD8⁺ T cell responses were measured by ICS at baseline and at weeks 2, 6, 8, 10, 12, and 32, as described elsewhere [24]. Cells were stimulated with vaccine insert–specific peptide pools (WNV-E and WNV-M). Cells were permeabilized, washed, and stained with directly conjugated anti-human CD3, CD4, CD8, interferon (IFN)–γ, and interleukin (IL)–2 antibodies and were assessed for CD3, CD8, CD4, and IFN-γ/IL-2 expression on a FACSCalibur flow cytometer (BD Immunocytometry Systems).

Statistical methods

All assays are treated as binary (responders/nonresponders). We use the usual central, exact 95% confidence intervals (CIs) for binomial rates. We are 97.5% confident that the true response rates in the antibody assays are higher than the lower limit. Calculations were done using R software (version 2.3.1; R Foundation). A positive T cell response for ICS and ELISpot data was based on composite criteria as previously described in 4 published studies of candidate vaccines [23, 24, 29]. SAS (version 9.0; SAS Institute) and S-plus (version 6.2; Insightful) were used for analyses.

RESULTS

Study population demographics

A total of 15 healthy adults were enrolled. Table 1 includes demographic data on subject sex, age, race/ethnicity, body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters), and educational level at the time of enrollment. The subject population was 67% male, with a mean age of 28.5 years. Subjects were predominantly white (87%) and non-Hispanic/Latino (93%). The mean BMI was 26. All subjects had an educational level of high school or higher, with 60% having college-level degrees.

Table 1.

Summary of demographic characteristics at enrollment.

Category, parameter	All vaccinees (n = 15)
Sex
Male	10 (67)
Female	5 (33)
Age
18−20 years	2 (13)
21−30 years	9 (60)
31−40 years	2 (13)
41−50 years	2 (13)
Mean ± SD	28.5 ± 8.2
Range	19−44
Race
American Indian/Alaska Native	0
Asian	0
Black or African American	2 (13)
Native Hawaiian or other Pacific Islander	0
White	13 (87)
Multiracial	0
Other/unknown	0
Ethnicity
Non-Hispanic/Latino	14 (93)
Hispanic/Latino	1 (7)
Body mass index
<18.5	0
18.5−24.9	6 (40)
25.0−29.9	7 (47)
≥30.0	2 (13)
Mean ± SD	26 ± 4
Range	21.0−34.8
Education
Less than high school graduate	0
High school graduate/GED	5 (33)
College/university	9 (60)
Advanced degree	1 (7)

NOTE. Data are no. (%) of subjects, unless otherwise indicated. Age represents age at enrollment day. Height and weight (used for calculation of body mass index) were from screening evaluation before enrollment. GED, General Educational Development.

Vaccine safety

The diary cards showed that 93% (14/15) of subjects reported at least 1 local injection site symptom (mild to moderate pain/tenderness, mild induration, or mild skin erythema) after a vaccination. When systemic symptoms were recorded on diary cards, they were primarily mild and included malaise, myalgia, headache, and nausea. No subjects reported fever or severe reactogenicity (table 2). A superficial, self-limited cutaneous erosion of ∼1 cm in diameter, which healed without treatment or sequelae, was noted after 1 vaccination; this was similar to superficial skin erosions observed with a candidate HIV DNA vaccine also administered by Biojector [29].

Table 2.

Summary of maximal local and systemic reactogenicity.

Symptom, intensity	All vaccines (n = 15)
Local reactogenicity
Pain/tenderness
None	2 (13.3)
Mild	11 (73.3)
Moderate	2 (13.3)
Swelling
None	6 (40.0)
Mild	9 (60.0)
Moderate	0
Redness
None	5 (33.3)
Mild	10 (66.7)
Moderate	0
Any local symptom
None	1 (6.7)
Mild	12 (80.0)
Moderate	2 (13.3)
Systemic reactogenicity
Malaise
None	11 (73.3)
Mild	3 (20.0)
Moderate	1 (6.7)
Myalgia
None	13 (86.7)
Mild	0
Moderate	2 (13.3)
Headache
None	11 (73.3)
Mild	3 (20.0)
Moderate	1 (6.7)
Chills
None	14 (93.3)
Mild	0
Moderate	1 (6.7)
Nausea
None	14 (93.3)
Mild	0
Moderate	1 (6.7)
Temperature
None	15 (100)
Mild	0
Moderate	0
Any systemic symptom
None	9 (60.0)
Mild	4 (26.7)
Moderate	2 (13.3)

NOTE. Data are no. (%) of subjects, unless otherwise indicated. Local injection site reactions were recorded by clinicians 30−45 min after an injection and were then recorded as self-assessments at home by subjects on a 5-day diary card. Systemic reactions were recorded as self-assessments at home by subjects on a 5-day diary card after each injection. There were no reports of severe local or systemic symptoms after vaccination.

There were no serious adverse events. Twelve of 15 subjects completed the vaccination schedule. Three subjects were withdrawn from the vaccination schedule because of concomitant illness; 2 after the second vaccination and 1 after the first vaccination. One subject developed a benign grade 1 condition, aquagenic wrinkling of the palms, which occurred 12 days after the second vaccination and continued intermittently through the follow-up period. This condition has been described after administration of COX-2 inhibitors [30] and in people with cystic fibrosis (CF) gene mutations [31]. This subject had not taken concomitant medications, was healthy, and had a negative evaluation for CF gene mutations after the onset of this benign condition. One subject experienced a grade 2 methicillin-resistant Staphylococcus aureus skin infection distant to the vaccination site, 2 weeks after the second vaccination. One subject was given a diagnosis of streptococcal pharyngitis and vaginal candidiasis, which occurred 25 days after the first vaccination. These events resolved without residual sequelae.

Safety and immunogenicity data for all subjects (n = 15) are included in this intent-to-treat analysis through the time points available. All subjects were followed for 32 weeks for safety and immune response; 1 subject who completed the 3-dose vaccination schedule did not present for the study visit and blood collection at week 12; therefore, that subject's week 12 immune response could not be assessed. Overall, study vaccinations were well tolerated and safe in the healthy subjects (age range, 19−44 years).

Antibody responses

Vaccine-induced humoral immune responses were assessed at weeks 8, 12, and 32. Subjects are represented by an identifier on the basis of age (designated by the letters A–O). Vaccine-induced antibody, assessed by ELISA at week 12 relative to baseline, was present in all subjects who received 2 or 3 doses of vaccine (table 3).

Table 3.

Antibody titers.

			Reciprocal PRNT titer
	Reciprocal ELISA titer			75%
Subject	Week 12	Week 32	50%, week 12	Week 12	Week 32
A	480	30	32	16	16
B	Not done	90	Not done	Not done	8
Ca	Negative	Negative	Negative	Negative	Negative
D	1920	30	64	32	8
E	480	Negative	64	32	16
F	1920	30	32	4	8
Gb	120	30	4	Negative	8
Hb	120	Negative	2	Negative	2
I	480	30	64	32	16
J	1920	90	128	16	8
K	480	90	64	16	32
L	1920	90	128	32	16
M	480	30	64	8	8
N	7680	810	16	4	4
O	480	90	16	16	2

NOTE. Reciprocal titers are shown for each subject as assessed by ELISA and plaque-reduction neutralization (PRNT) at weeks 12 and 32. Positive responses are reported as a reciprocal titer.

^aReceived 1 dose of vaccine.

^bReceived 2 doses of vaccine.

To determine whether vaccine-induced WNV-specific antibody could neutralize virus, 2 different methods were used to detect antibody-mediated neutralization. First, neutralizing titers for each subject at weeks 12 and 32 were estimated using PRNT on Vero cells. Vaccine-induced neutralizing antibody was detected in all vaccinees who completed the 3-dose vaccination schedule and were assessed at week 12 (11/11 [100%; 95% CI, 71.5%−100%]) and week 32 (12/12 [100%; 95% CI, 73.5%−100%]). A response was also seen at week 32 in both subjects who received 2 doses of vaccine (table 3).

To complement the PRNT, neutralization profiles of serum from each volunteer at weeks 8, 12, and 32 was studied using a recently described quantitative approach. WNV RVPs are pseudoinfectious virions that are capable of a single round of infection and that allow WNV entry to be measured as a function of reporter gene expression. For each vaccinee, a profile of the neutralization activity in serum at each collection point was measured using 7 or 8 dilutions of serum. In agreement with data from the PRNT, neutralizing antibody was demonstrated in all subjects who received 3 doses of vaccine and in 1 subject who received 2 doses of vaccine. Subjects developed neutralizing antibody after the second vaccination by week 8; antibody titers typically peaked at week 8 or 12 and remained positive at week 32, albeit at a lower level (figure 1).

Figure 1.

Serum samples from vaccinees at week 8 (before the third vaccination), week 12 (4 weeks after the third vaccination), and week 32, assessed for the presence of neutralizing antibody by a West Nile virus reporter-virus particle (RVP) neutralization assay. The X-axis shows individual vaccine clinical trial subjects, and the Y-axis shows the log₁₀ reciprocal EC₅₀ neutralizing antibody titer. Subject B was not assessed at the week 12 time point because of visit noncompliance. Subjects G and H received 2 doses of vaccine. Subject C received 1 dose of vaccine. ND, not done.

To compare the magnitude of the neutralization response elicited by this DNA vaccine in humans to a response that confers protection to challenge by WNV, we evaluated the neutralizing titers in serum obtained from horses that received an equine WNV DNA vaccine (Fort Dodge Animal Health and the CDC) encoding for the same construct as the vaccine used this clinical trial. That vaccine is USDA approved for use in horses and has been shown to elicit neutralizing antibody by PRNT [19]. We evaluated serum from a sample of these horses by the RVP neutralizing antibody assay. The postvaccination titers 3 weeks after a single 1-mg dose of the WNV DNA vaccine in horses was similar to the titers elicited in humans in this clinical trial after 2 or 3 doses of vaccine (figure 2).

Figure 2.

Serum samples from vaccinees at week 12 (4 weeks after the third vaccination) and serum samples from horses 3 weeks after receipt of a 1-mg dose of pCBWN West Nile virus (WNV) DNA vaccine, assessed by WNV reporter-virus particles (RVP) neutralization assay and by plaque reduction neutralization (PRNT). The X-axis shows individual vaccine clinical trial subjects or horse samples, and the Y-axis shows the log₁₀ reciprocal antibody titer. Subject B was not assessed at the week 12 time point because of visit noncompliance. Subjects G and H received 2 doses of vaccine. Subject C received 1 dose of vaccine. VRC, Vaccine Research Center.

T cell responses

The majority of vaccine-specific T cell responses were elicited against WNV-E, and CD4⁺ T cell responses were more frequent than CD8⁺ T cell responses. The peak frequency and magnitude of the response occurred between weeks 8 and 12 (figure 3). Vaccine-specific T cell responses to WNV-E were detected by ELISpot assay in 8 (53%) of 15 subjects. CD4⁺ T cell responses to WNV-E as measured by ICS were detected in 14 (93%) of 15 subjects. A T cell response was not induced in the subject who received only 1 injection. A T cell response to WNV-M was detected in 4 (27%) of 15 subjects, and a CD8⁺ T cell response as measured by ICS was detected in 1 subject and was present against both WNV-E and WNV-M.

Figure 3.

T cell responses specific for West Nile virus (WNV) DNA vaccine insert–specific peptide pools (WNV-E [left] and WNV-M [right]), assessed by enzyme-linked immunospot (ELISpot) assay and intracelluar cytokine staining (ICS). Data are shown for the entire study (weeks 0−32). The magnitude of ELISpot and ICS responses are shown in the top graphs, and the frequency of responses among all subjects is shown in the bottom graphs. The threshold for positivity by ELISpot assay is a mean of 59 sfc per million peripheral blood mononuclear cells. For ICS responses, a positive response required that the P value from a Fisher's exact test of the difference between the antigen-specific response and the negative control response be <.01 and that the antigen-specific response be at least 0.0241% (for CD3⁺CD4⁺ responses) or at least 0.0445% (for CD3⁺CD8⁺ responses) higher than the negative control response.

DISCUSSION

WNV represents an economically and medically important reemerging flavivirus that continues to cause morbidity and mortality in North America. The virus has spread across North and Central America and into South America since its emergence in New York in 1999. Neutralizing antibody is known to provide protection against flaviviruses in animal models and humans. A DNA vaccine encoding for WNV proteins prM and E protects horses from WNV infection in a challenge model [19] and is currently USDA approved for veterinary use.

Production of bacteria-derived plasmid DNA vaccines has many favorable features, including (1) ease and flexibility of construction, (2) scalable manufacturing capacity, (3) stability, (4) intracellular production of vaccine antigen, (5) transient expression, (6) no induction of antivector immunity, (7) induction of both T cell and antibody responses, and (8) low occurrence of systemic reactogenicity.

This WNV candidate vaccine was well tolerated in 15 healthy subjects. The rate and severity of reactogenicity reported is similar to that reported in previous studies of an Ebola virus DNA vaccine [24] and a multiclade HIV DNA vaccine [23].

DNA vaccines have been shown to be safe and to elicit vaccine-induced humoral and cellular immune responses for HIV and Ebola virus [23-25]. Although these DNA vaccines are able to elicit virus-specific antibodies detected by immunoprecipitation and Western blotting or by ELISA [23, 32], they have not been highly efficient at eliciting neutralizing antibodies, possibly because these viruses are generally less sensitive to antibody neutralization. The ability of this WNV DNA vaccine to elicit neutralizing antibody could possibly be due to the formation of virus-like particles produced by the vaccine-encoded proteins. E and prM are known targets of WNV-neutralizing antibody, and SVP formation may allow for the authentic antigenic sites constrained by presentation in the icosahedral structure to efficiently induce the relevant antibody specificity.

The T cell immune response and neutralizing antibody elicited by this WNV DNA vaccine further suggests that needle-free immunization of DNA plasmids can reproducibly elicit significant cellular and humoral immune responses and merits further clinical investigation. The consistent humoral immunity elicited by the WNV DNA vaccine described here likely reflects a combination of factors, including in vivo SVP formation, optimization of vector design, the purity of the DNA product, delivery, and sensitive immunological assays. This clinical trial represents the first report of neutralizing antibody activity elicited by a DNA vaccine in humans. Although further studies are needed, including investigation of the possibility of dose reduction, this level of neutralizing antibody activity is known to be protective in animal models of infection [19] and suggests that DNA plasmid immunization is a promising vaccine approach for the prevention of WNV infection. Correlates of protection from animal models of infection are essential to determine the efficacy of candidate vaccines designed to prevent emerging infectious diseases. Established correlates of protection applied to the US Food and Drug Administration Animal Rule could potentially be used to license a vaccine for the prevention of WNV infection in humans and to establish efficacy for licensure of vaccines against other emerging and reemerging infectious diseases.

VACCINE RESEARCH CENTER (VRC) 302 STUDY TEAM

The VRC 302 Study Team includes Brenda Larkin, Margaret McCluskey, LaSonji Holman, Laura Novik, Pamela Edmonds, LaChonne Stanford, Woody Dubois, Tiffany Alley, Erica Eaton, Sandra Sitar, Ericka Thompson, Andrew Catanzaro, Joseph Casazza, Laurence Lemiale, and Rebecca Sheets.