Demonstration of Efficacy of a West Nile Virus DNA Vaccine in Foals

The West Nile Virus (WNV) DNA vaccine developed by Fort Dodge Animal Health (FDAH) induced protective immunity in young horses. The induction of the immunity was not interfered by maternally derived antibodies.

1. Introduction

The West Nile Virus (WNV) was first isolated and identified from birds, mosquitoes, and mammals including horses in three states of the northeastern United States in 1999. Since then, WNV infection in horses has been spread westward and completed its coast-to-coast spread in 2002 when two cases in the state of Washington were confirmed. In 1999, there were 25 confirmed cases of WNV in horses. WNV reached its peak in 2002 (14717 cases) and started to decline in 2003 (5181 cases). By December 10, 2004, there were 1341 confirms cases in horses [1].

A killed WNV vaccine developed by Fort Dodge Animal Health (FDAH) was introduced in late summer of 2001 and was the only WNV vaccine available until the end of the mosquito season in 2003. The beneficial effects of this vaccine against WNV infection and death have been reported recently [2,3] .

Concurrent with the development of the aforementioned killed WNV vaccine, we have also developed a DNA vaccine for horses using the DNA plasmid developed by the scientists at Centers for Disease Control and Prevention [4]. This DNA plasmid encodes the genes of pre-membrane and envelope proteins of WNV. Unlike conventional vaccines, the DNA vaccine represents a unique way of vaccination in which host animals become the source of the antigens of interest [5] and therefore, may override maternally derived antibodies (MDA) in different ways than conventional vaccines.

In the studies reported here, we evaluated the efficacy of a WNV DNA vaccine in sero-negative foals and the ability of this DNA vaccine to induce immune responses in foals with high levels of anti-WNV MDA. Both studies were sponsored and conducted by FDAH.

2. Materials and Methods

Test Animals

In study I, 34 foals (16 males and 18 females) that were sero-negative (<5) to WNV by plaque reduction neutralization test (PRNT) were enrolled. In study II, 29 pregnant mares with recent vaccinations against WNV were purchased, and the foals (13 males and 16 females) with anti-WNV MDA titers measured by PRNT from the mares were enrolled. The horses were housed communally at off-site facilities in Iowa. Shortly before challenge, the horses were transported back to the FDAH challenge facility and individually housed throughout the challenge phase of the study. The horses were fed a standard commercial feed with hay and water ad libitum.

Test Vaccine

The WNV DNA vaccine contains DNA plasmid that encodes the genes of pre-membrane and envelope proteins of WNV and was formulated with Metastim. Each dose of vaccine contains 2 ml. The making of a DNA vaccine is depicted in Figure 1.

Figure 1. The making of a DNA vaccine. Reprinted with the permission of the National Institute of Allergy and Infectious Diseases [8].

Experimental Design

In study I, 34 foals (~8 - 9 mo of age) were randomized into one group of 17 vaccinates and one group of 17 controls. Because of the limited space in the challenge facility, 15 horses from each group were randomly selected and were subjected to the experimental challenge.

In study II, 29 foals (~3 - 13 wk of age) were randomized into one vaccinated group (15 animals) and one control group (14 animals).

Vaccination

Animals in the vaccinated group in either study were administered the WNV DNA vaccine IM twice, 3 wk apart. The control animals in study I did not receive placebo vaccine, but control foals in study II did receive a placebo vaccine.

Experimental Challenge

WNV crow isolate [a] (lot V76-2), obtained from Dr. Eileen Ostlund of National Veterinary Services Laboratories (NVSL), was used as the challenge virus. The isolate is a 1999 North American avian isolate that was passaged twice in Vero76 cells at NVSL. At FDAH, the cell culture material received from NVSL was subcultured one time in Vero cell culture.

Foals were challenged with 1.0 × 106 TCID50 of live WNV administered SC ~4 mo after the second vaccination (study I) or ~5 mo after the second vaccination (study II).

Clinical Observations

Rectal temperatures and clinical signs were monitored twice daily for 2 days before challenge to establish baselines. Clinical signs that were monitored included, but were not limited to, lethargy/depression, ataxia, incoordination, muscle weakness, recumbency, difficulty rising, muscle fasciculation/twitching, and paresis. After challenge, rectal temperatures and clinical signs were monitored twice daily for 10 days and once daily until 14 days post-challenge (DPC). Foals in study I were observed for clinical signs daily for 1 additional wk.

Sample Collection

Foals were bled for serum periodically for detection of antibody responses starting with the first vaccination and ending with the termination of the study. In addition, serum samples were collected from each animal for detection of viremia twice daily for 10 days starting on the day of challenge, once daily at 14 DPC (both study I and study II), and at 21 DPC (study I only).

PRNT

Serum samples collected from both studies were tested for neutralizing antibody response using PRNT. Briefly, two-fold serial dilutions of serum samples were allowed to incubate overnight at 4°C with reference WNV at 200 PFU/0.1 ml. The virus-serum mixture was then transferred to Vero cell monolayers, overlayed, and allowed to incubate at 37 ± 2°C for 48 - 60 h. The wells were stained for 24-48 h with neutral red overlay, and plaques were counted using a light box. Endpoints were determined at a 90% plaque-reduction level.

WNV IgG Enzyme-Linked Immunosorbent Assay

Serum samples were assayed by WNV IgG antibody enzyme-linked immunosorbent assay (ELISA). Briefly, 96-well ELISA plates [b] were coated with a monoclonal antibody (4G2) [c] and stored overnight at 4°C. The plates were washed four times using a 0.3% Tween phosphate buffered saline (PBS) wash buffer and blocked using 5% non-fat dry milk in 0.3% Tween/PBS for 1 h at 37°C with agitation. The plates were then washed four times as indicated above. Positive (mixture of recombinant pre-membrane and envelope proteins) and negative (Cos-1 cells) antigens were added and incubated at 37°C for 1 h with shaking. Plates were again washed as above, and diluted equine serum samples (1:1000), including known positive and negative control equine sera, were added and allowed to incubate 1 h at 37°C with shaking. After incubation, the plates were washed. Secondary antibody (GαH IgGγ-horseradish peroxidase [HRP]) [d] was added and allowed to incubate 1 h at 37°C with agitation. 3,3',5,5'-tetramethylbenzidine (TMB) solutions (1:1) were added and allowed to incubate 20 - 30 min at 37°C. The plates were read at 650 - 490 nm.

The mean optical density (OD) of the duplicate wells of each serum sample with negative antigen was subtracted from the mean OD of each serum sample with positive antigen. This value was divided by the mean OD of the positive control sera with positive antigen to standardize each plate.

WNV IgM-Capture ELISA

Serum samples were assayed by WNV IgM-capture antibody ELISA, which was developed at NVSL. Briefly, ELISA plates were coated with an anti-equine IgM antibody [d] at a dilution of 1:400 in coating buffer and placed at 4°C overnight. After incubation, plates were washed in a wash buffer (0.05% Tween/PBS) and blocked using 5% non-fat dry milk. Plates were incubated at room temperature for 60 min. Equine serum samples were diluted 1:400 in dilution buffer and added to the blocked plate. Positive and negative control sera were included on each plate. Plates were incubated at 37°C for 75 min. Normal and viral antigens were added to the appropriate wells of each plate. After overnight incubation at 4°C, HRP-conjugated Flavivirus monoclonal antibody 6B6C-1 (CDC) was added to each plate and incubated for 60 min at 37°C. Peroxidase substrate 2,2'-azino-di(3-ethylbenzthiazoline-6-sulfonate) (ABTS) [d] was added to each plate after incubation, and color development was quantified at 405 nm after 30 min.

The interpretation of the results was performed by first determining the positive threshold (PT). The PT was the mean OD of the negative control serum with viral antigen multiplied by two. The mean OD of the positive control serum with viral antigen was divided by the mean OD of the negative control serum with viral antigen. This ratio (positive/negative [P/N]) must be ≥2 for a test to be considered valid.

The result of each individual sample was determined by first comparing the mean OD of the sample with viral antigen to the PT and then calculating the P/N of each sample (mean OD with viral antigen divided by mean OD with normal antigen). Samples with ODs greater than the PT and with a P/N >2 were considered positive. Results from samples with ODs greater than the PT and with P/N <2 were considered non-specific.

Virus Isolation

Virus isolation in Vero cell cultures was conducted on serum samples collected during the challenge phase of the study. Each sample was passaged twice in Vero cell culture and then read by indirect fluorescent antibody (IFA) assay specific to WNV. A sample was considered positive when the second passage was determined to be WNV positive by IFA.

Statistical Analysis

For the evaluation of viremia, the number of viremic animals in the vaccinated group was compared with the number of viremic animals in the control group using Fisher's exact test. The vaccine efficacy (preventable fraction) and its 95% confidence interval were computed. The level of significance was set at p < 0.05.

3. Results

Study I

The efficacy of the WNV DNA vaccine in sero-negative foals was evaluated in study I. No clinical signs associated with WNV infection were detected in any of the challenged horses in study I. Viremia was detected in 12 of 15 non-vaccinated control horses (80%), and 2 of 15 vaccinates (13%) had viremia. This difference was statistically significant (p = 0.0007). The efficacy against viremia expressed as a preventable fraction was 83.3% (95% CI = 48.6, 98.1) in the vaccinated horses.

Serum-antibody responses against WNV were measured by PRNT and/or IgG ELISA and are shown in Figure 2 and Figure 3, respectively. Two doses of vaccination increased neutralization and specific IgG antibody levels in the vaccinates, whereas controls remained negative until 0 DPC. After challenge, an apparent anamnestic IgG response was detected in vaccinates (Fig. 3).

Figure 2. WNV-neutralizing antibody response measured by PRNT in foals vaccinated with a WNV DNA vaccine and in control foals (study I). For the ease of calculation, titer <5 was given a value of 3. DPV, day post-vaccination; MPV, month post-vaccination; DPC, day post-challenge.

Figure 3.Anti-WNV IgG response measured by IgG ELISA in foals vaccinated with a WNV DNA vaccine and in control foals (study I).

Post-challenge serum IgM response in horses was measured by IgM-Capture ELISA and is shown in Figure 4. Positive detection of anti-WNV IgM response was observed in more controls than vaccinated horses.

Figure 4. Post-challenge anti-WNV IgM response measured by Capture-IgM ELISA in foals vaccinated with a WNV DNA vaccine and in control foals (study I).

Study II

The immunity induced by the WNV DNA vaccine in the presence of anti-WNV MDA was evaluated in study II. A majority of foals enrolled had PRNT titers ≥160 (87% of vaccinates and 93% of controls) at the day of first vaccination (0 DPV; data not shown). Similar to study I, no clinical signs associated with WNV infection were detected in any of the challenged foals. However, in a complete contrast to study I, none of the challenged foals developed viremia after challenge with a similar challenge dose as in study I.

This study was designed to be terminated at 14 DPC. However, because of the unexpected viremic results, additional serum samples were collected to obtain additional data in antibody responses. As shown in Figure 5 and Figure 6, neutralization and IgG responses in both vaccinated and control foals followed a downward trend. However, an anamnestic IgG response similar to the one observed in study I was detected in vaccinated foals (Fig. 6). Post-challenge, IgM response was observed in both vaccinates and controls, although in a smaller proportion of control foals compared with study I (Fig. 7).

Figure 5. WNV-neutralizing antibody response measured by PRNT in foals vaccinated with a WNV DNA vaccine in the presence of MDA and in control foals (study II). For the ease of calculation, titer ≥160 was given a value of 160, whereas titer <5 was given a value of 3.

Figure 6. Anti-WNV IgG response measured by IgG ELISA in foals vaccinated with a WNV DNA vaccine in the presence of MDA and in control foals (study II).

Figure 7. Post-challenge anti-WNV IgM response measured by Capture-IgM ELISA in foals vaccinated with a WNV DNA vaccine in the presence of MDA and in control foals (study II).

4. Discussion

The efficacy of the WNV DNA vaccine against viremia associated with WNV infection was shown in study I. As seen in the previous studies [4,6], the detection of viremia was transient and occurred primarily in the first 6 days after challenge.

The specific neutralization or IgG antibody levels induced in foals by vaccination with two doses of WNV DNA vaccine did not have similar magnitude as those induced by conventional killed vaccine [7]. Nevertheless, the kinetic of the post-challenge anamnestic IgG response detected in foals vaccinated with DNA vaccine in study I was very similar to the one observed in horses vaccinated with the killed vaccine and challenged 1 yr after vaccination [7]. In addition, the post-challenge IgM response in study I was very similar to the aforementioned study. The detection of anti-WNV IgM response has been used as a diagnostic tool for detecting WNV infection. However, results from the present studies and those from previous studies conducted by us indicate that there is no association between the detection of IgM response and the detection of viremia after experimental challenge (data not shown).

The induction of immunity by DNA vaccine, in theory, should not be interfered by the presence of MDA. Study II was designed to evaluate the immunity induced by the WNV DNA vaccine in foals with substantially high anti-WNV MDA.

The lack of viremia after experimental challenge was completely unexpected. The experimental challenge model used consistently resulted in at least 80% of control animals with viremia as evidenced in study I and the previous study [6]. Fifty percent of control foals at the day of challenge had low PRNT titers (data not shown). One foal had a titer of 40, and the other 6 foals had titers ranging from 5 to 10. However, this observation alone is not sufficient enough to support the cause for the lack of viremia in any of the control foals. Repeated culture isolation yielded the same results. The polymerase chain reaction (PCR) assay that was available to us was not sensitive enough to detect <100 viral particles.

One potential cause of the lack of viremia in the control foals was that the challenge virus lost its virulence. In a subsequent confirmatory study, a freshly prepared challenge inoculum was tested side by side with the existing challenge inoculum. All the horses (100%) challenged with either preparations developed viremia. This would indicate that the virulence of the existing challenge preparation remained the same and that the experimental model used was valid. It is speculated that the residual MDA in the control foals was not detectable by the serological assays used, but it was sufficient enough to protect the control foals against the development of viremia.

Serological data from study II supports the contention that the immunity induced by the WNV DNA vaccine was not affected by the presence of specific MDA. The kinetic of the anamnestic IgG response was very similar to the one observed in study I and the one in the 1-yr duration of immunity study using the killed conventional vaccine [7]. Accordingly, we would speculate that the WNV DNA vaccine could induce a protective immunity in the presence of MDA.

In summary, results from these studies show that the WNV DNA vaccine developed by FDAH induced protective immunity in young horses and that MDA did not interfere with the induction of the immunity.

Footnotes

[a] Lot V76-2, NVSL, Ames, IA 50010.
[b] Maxi-Sorp F96, Nalge Nunc International, Rochester, NY 14625-2385.
[c] Produced from hyridoma D1-4G2-4-15 obtained from ATCC, Manassas, VA 20110-2209.
[d] Kirkegaard and Perry Laboratory (KPL), Gaithersburg, MD 20879.