AJTMH ASTMH Job Mart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am. J. Trop. Med. Hyg., 78(3), 2008, pp. 434-441
Copyright © 2008 by The American Society of Tropical Medicine and Hygiene

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Patiris, P. J.
Right arrow Articles by Hanson, C. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Patiris, P. J.
Right arrow Articles by Hanson, C. V.
Related Collections
Right arrow Flaviviruses
Right arrow Arboviruses

Serologic Diagnosis of West Nile and St. Louis Encephalitis Virus Infections in Domestic Chickens

Peter J. Patiris*, Leopoldo F. Oceguera, III, George W. Peck, Robert E. Chiles, William K. Reisen, AND Carl V. Hanson
California Department of Public Health, Viral and Rickettsial Disease Laboratory, Richmond, California; Center for Vectorborne Diseases, University of California, Davis, California


ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Adult domestic chickens were infected with West Nile virus (WNV) or St. Louis encephalitis virus (SLEV) and challenged with homologous or heterologous virus at 21 or 56 days postinfection (dpi). Sera were collected at selected time points after infection and assayed by enzyme immunoassay (EIA), plaque reduction neutralization test (PRNT), and a Western blot (WB) alternative to PRNT. EIA results were sensitive and accurate (few false positives) but not specific, requiring a confirmatory test to determine virus infection history. PRNT results generally were specific until challenge, after which test results were frequently equivocal and inadequate to determine first or second infecting virus. WB results confirmed the serologic cross-reactivity between WNV and SLEV envelope protein. Non-structural protein 1 and pre-membrane protein reactivities were highly specific for WNV during SLEV infection, but less specific for SLEV during WNV infection. WB and PRNT specificities were similar for both viruses from 6 to 14 dpi, and sensitivities to WNV were virtually identical.


INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Adult domestic chickens, Gallus gallus, are ideal sentinel animals for monitoring the enzootic transmission of mosquito-borne encephalitis viruses by Culex mosquitoes among wild birds. Genetically similar birds of the same age can be purchased in large numbers at a low cost from brooding services and maintained easily on commercial diets.1 In addition, experimental infections have shown that birds > 18 weeks of age do not produce viremias that can serve as a source of infection for host-seeking mosquitoes.2–5 This contrasts with other birds such as Rock doves, Columba livia, that occasionally produce elevated viremias.6 Adult chickens are large and robust and, therefore, blood samples can be taken easily using a variety of methods, including a lancet prick of the comb with the sample adsorbed into filter paper strips.7 These strips can be dried and stored under ambient conditions and screened rapidly using an enzyme immunoassay.4 In California, sentinel chickens are a standard surveillance measure of enzootic transmission8 and form an integral factor in the California Encephalitis Virus Surveillance Program risk model.9 More than 30,000 chicken serum samples are tested for three viruses annually by the Viral and Rickettsial Diseases Laboratory (VRDL) of the California Department of Health Services.10

Historically, laboratory testing of serologic specimens was simple, because St. Louis encephalitis virus (SLEV) was the only Flavivirus transmitted to chickens by mosquitoes in North America. The invasion by West Nile virus (WNV), a member of the same Japanese encephalitis serocomplex as SLEV, complicated diagnostics, because these two viruses cross-react in most enzyme immunoassays (EIAs).11,12 This led to evaluations of various diagnostic procedures to separate antibodies raised against these two antigenically similar viruses13–17; however, separation has been complicated by antibody decay as a function of time after infection and by heterologous serial infections.18,19

The objective of our current research was to compare three serologic assays for differentiation of WNV and SLEV infections in adult hens similar to those used for the California surveillance program after primary infection with WNV and SLEV and secondary infection with homologous and heterologous viruses. The antibody response was measured by EIA, plaque reduction neutralization test (PRNT), and a novel Western blot (WB) assay developed as an alternative to PRNT.20 The effects of repeat infections and co-infections on the results of these serologic assays were determined.


MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Chickens. Thirty white Leghorn x Rhode Island Red laying hens were purchased from Gemperle Farms and were 20 weeks of age at the start of experimentation. Birds were fed Layena Egg Maker Crumbles (Purina, St. Louis, MO) and maintained within mosquito-proof cages. Birds were wing-banded, prebled, and tested by EIA to determine previous exposure to WNV, SLEV, and western equine encephalomyelitis viruses, with negative findings. Maintenance, infection, and blood sampling from chickens were done in accordance with protocols approved by the University of California Institutional Animal Care and Use Committee.

Viruses. We used the New York strain of WNV isolated from a flamingo that died in the Bronx Zoo (strain 35211 AAF) and that previously was passaged twice in Vero cells. The Kern217 strain of SLEV isolated from Cx. tarsalis collected in Bakersfield in 1989 also had been passaged twice in Vero cells before use. Both virus strains have been used extensively in host competence studies in our laboratory.5,21 Serologic assays used the Ruis strain of SLEV isolated from human brain tissue sent to the VRDL by a hospital in Fresno, CA, and most recently passaged three times in Vero cells.

Viremia. Six chickens each inoculated with either WNV or SLEV were monitored for their viremia response on 1–5 days postinfection (dpi). A sample of 0.1 mL blood was collected by jugular puncture with a 28-gauge syringe, expressed into 0.4 mL of diluent (phosphate-buffered saline, 15% fetal bovine serum, antibiotics), held at ambient temperature for > 15 minutes, clarified by centrifugation, and frozen at –80°C until tested. Viremia levels were measured by plaque assay using Vero cell cultures in 6-well plates.22

Experimental design. Infection protocols attempted to accommodate varied infection and re-infection scenarios, including infection with a single virus, dual infection, and immediate or delayed re-infection with homologous and heterologous viruses, to describe the ability of the three assays to determine these infection histories. Chickens were infected by subcutaneous needle inoculation in the cervical region with 5.3 log10 plaque forming units (PFUs) of WNV and 2.2 log10 PFUs of SLEV. Twelve hens each were inoculated with either WNV or SLEV. At 3 or 8 weeks after infection, three hens from each group were challenged with either homologous or heterologous virus. An additional three hens were co-infected concurrently with WNV and SLEV, and three hens were inoculated with virus diluent as negative controls. Blood samples were collected at 6, 8, 10, 12, and 14 dpi or after challenge to detect the time from infection until initial antibody detection and then weekly from 3 to 8 weeks after infection or challenge to detect any decay or increase in non-specificity. Approximately 7–8 mL of whole blood was collected by vacutainer with 22-gauge needles, allowed to clot at room temperature, and centrifuged, and undiluted serum was transferred to cryovials that were frozen at –80°C until tested.

Serology. Each serum sample was tested for IgG by EIA4 and for neutralizing antibody by PRNT.22 Sera were also tested for both IgM and IgG using a novel WB assay.20

Enzyme immunoassay. The New York 99 (NY-99) strain of WNV and the Ruis strain of SLEV were used to prepare the crude tissue culture antigen used to coat 96-well tissue culture plates (Immulon 1; Dynatech Laboratories, Chantilly, VA) for the EIA. Control antigen was prepared from uninfected cell culture. For each sample, dilute chicken sera were reacted in antigen-coated 96-well microplates, with two wells with viral antigen and one well with control antigen. After washing, mouse anti-chicken immunoglobulin G antibody conjugated with alkaline phosphatase was added and incubated. After removal of the unreacted conjugate by washing, substrate was added (ABTS; KPL, Gaithersburg, MD), and optical density was measured at 405 nm. Results were expressed as the ratio of the mean optical density of the two antigen-containing wells over the optical density of the corresponding control well. A ratio > 2 was considered positive.

Plaque reduction neutralization test. Sera were assayed for neutralizing antibodies to the NY-99 strain of WNV and the Ruis strain of SLEV by PRNT as previously described.23 Briefly, 100 plaque forming units (PFUs) of viral inoculum was pre-incubated with test sera and adsorbed onto confluent Vero cell monolayers and cultured in 6-well tissue culture plates (Corning Inc., Lowell, MA). After adsorption, a single agarose overlay system was used for WNV and a double overlay system for SLEV. Plaques were counted at 4 dpi for WNV and 7–10 dpi for SLEV. A plaque reduction of ≥ 80% was considered positive, with the titer measured as the highest serum dilution showing > 80% reduction of plaque relative to a serum-free control. A 4-fold difference in titer between WNV and SLEV was required for virus identification.

Western blot. Antigens for WB were prepared as previously described.20 Briefly, viruses were grown in Vero cells until the cytopathic effect (CPE) could be observed in ~75% of the cells (3+), at roughly 3–4 dpi. Infected cells were harvested in lysis buffer containing protease inhibitors. After centrifugation, lysates were aliquoted and stored at –80°C.

Thawed viral lysates were electrophoresed, without reducing or heating, on 10% discontinuous polyacrylamide gels and transblotted to nitrocellulose membranes. The nitrocellulose membranes were washed and cut into 3-mm strips and stored at –80°C. When needed, nitrocellulose strips were placed in individual plastic troughs and incubated overnight at room temperature with 10 µL test serum in 1 mL of blot buffer consisting of 5% blotting grade blocker (Bio-Rad, Hercules, CA) in phosphate-buffered saline, on a platform rocker. After washing, the strips were incubated for 60 minutes with the appropriate peroxidase labeled, goat anti-chicken IgG or IgM conjugate (Bethyl Laboratories, Montgomery, TX) and developed for 10 minutes with diaminobenzidine (DAB) substrate, washed, and air dried before reading.

Analysis. Reciprocal titers of PRNT endpoints were transformed by loge for analyses; negative titers < 1:20 were coded as 1:10 for analyses and visualization. Geometric means were presented as reciprocal titers. For visualization, arithmetic means were plotted using a log scale for viral titers. The complexity of our experimental design (17–22 tests per hen, 3 virus treatments, and 2 test analytes, antigens, or viruses) coupled with limited replication (N = 3 hens per treatment) forced us to compartmentalize our analyses into 16 separate tests (Table 1Go). During the time points after infection (21 or 56 days) but before pre-challenge, we compared the means (N = 6) of the assay results for SLEV and WNV analytes to determine the extent of original cross-reactivity using a two-way ANOVA with virus test and time as main effects. After challenge, we compared means (N = 3) of the assay results for SLEV and WNV analytes to determine the extent of cross-reactivity after homologous or heterologous challenge using a three-way ANOVA with virus test (SLEV or WNV), time (11 bleeds over 56 days), and challenge (heterologous or homologous) as main effects.


View this table:
[in this window]
[in a new window]

 
TABLE 1
Summary of EIA and PRNT ANOVA F values
 
WB was scored positive or negative for IgM and IgG reactivity with three protein bands on WNV and SLEV strips: non-structural protein 1 (NS1), envelope protein (E), and pre-membrane protein (prM). The presence of the cross-reactive E band alone on one or both strips was interpreted as Flavivirus positive. Detection of prM alone or combined with E was interpreted as a positive identification only if prM was negative for the alternate virus. If prM and typically E were positive on both strips, only a positive NS1 band could identify the virus. Pre-challenge sera were analyzed for sensitivity and specificity over time. Sensitivities over time were compared by fitting a nonlinear regression, exponential association model to each data set (GraphPad Software, San Diego, CA), allowing a comparison of "rate constants" (K) to evaluate relative assay sensitivities.


RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Viremia. Twenty-week-old chickens exhibited a low-grade viremia response after infection with WNV or SLEV (Figure 1Go). Response was variable among replicate hens as exhibited by the wide SE bars. Five of six hens had a detectable viremia after inoculation with WNV that ranged from 2.0 to 4.5 log10 PFU/mL. Response to SLEV was significantly lower; three of six hens had a detectable viremia ranging from 1.9 to 2.5 log10 PFU/mL. Threshold of detection was 1.7 log10 PFU/mL. Viremia of both viruses subsided after 5 days.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		    FIGURE 1. Mean ± SE viremia responses of adult chickens infected with WNV or SLEV (N = 6 hens each).

 
Serology. Enzyme immunoassay. Field samples from sentinel chickens typically are screened for SLEV and WNV infection using the same EIA. In this experiment, there were few false positives, and control values remained near or below our threshold P/N value of 2.0. Means for control chickens were always significantly less that those infected with SLEV or WNV. The immune response of chickens to original infection and to homologous and heterologous challenge at 21 and 56 days were measured by EIA using SLEV or WNV antigen (Figure 2Go) and analyzed by ANOVA (Table 1Go, Tests 1–8). Mean P/N ratios were always significantly greater when sera were tested by antigen homologous to the infecting virus; e.g., initial SLEV P/N ratios were 5.59 when tested by SLEV antigen and significantly > 2.98 when the same samples were tested by EIA using WNV antigen (Table 1Go, Analysis 1). Over the course of 56 days, P/N ratios increased as a function of time both before and after challenge, and the differences between tests using different antigens changed significantly over time as shown by the interaction terms in the ANOVAs. Over 21 days, there was a significant increase in P/N ratios over time for WNV but not for SLEV, which may be related to differences in the intensity of the original infection and resulting viremias (Figure 1Go).


Figure 2
View larger version (27K):
[in this window]
[in a new window]

  		    FIGURE 2. Antibody response measured by EIA with homologous and heterologous antigens after infection with either SLEV or WNV and expressed as mean P/N ratios (N = 3 hens per group) after challenge at 56 (A and B) or 21 dpi (C and D). Cut-off P/N value was 2. Legend shows the viral antigen used (W or S) followed by the sequence of virus infection and challenge; no challenge was done if a second virus not indicated. Homologous antigen shown with a solid line. Control hens were inoculated with virus diluent only.

 
After both homologous and heterologous challenge, all P/N ratios, regardless of antigen, were well above our cut-off value of 2 and would be considered presumptively positive. P/N values after homologous or heterologous viral challenge were not significantly different (Table 1Go), perhaps indicating that second inoculations with these closely related Flaviviruses actually boosted immunity relating to the original infecting virus rather than the challenge virus (so-called original antigenic sin).

When infected concurrently with both WNV and SLEV, EIA results were not statistically different (P > 0.05) using either SLEV (mean = 6.3) or WNV (mean = 7.4) antigen, and EIA P/N values did increase significantly as a function of dpi (data not shown).

Plaque reduction neutralization test. Field samples from sentinel chickens positive by EIA are routinely confirmed and the infecting virus identified using end point PRNTs. In this study, there were no false positives, and all uninfected chickens had titers < 1:20 cut-off value. Responses of chickens to original infection and homologous and heterologous challenge at 21 and 56 days were measured by PRNT using SLEV or WNV (Figure 3Go) and analyzed by ANOVA (Table 1Go, Tests 9–16). During both 56 and 21 dpi periods, the reciprocal of the geometric mean titers of PRNTs using test virus homologous to the infecting virus were always significantly greater than using heterologous test virus. Titers for the heterologous test virus initially remained negative but began to rise as early as 14 dpi for WNV and 21 dpi for SLEV (Figure 3, A and DGo).


Figure 3
View larger version (28K):
[in this window]
[in a new window]

  		    FIGURE 3. Antibody response measured by PRNT using homologous and heterologous viruses after infection with either SLEV or WNV and expressed as the inverse of the geometric mean PRNT titer (N = 3 hens per group) after challenge at 56 (A and B) or 21 dpi (C and D). Cut-off value was 1:20. Legend shows the test virus (W or S) followed by the sequence of virus infection and challenge; no challenge was done if a second virus not indicated. Homologous virus was shown with a solid line. Control hens were inoculated with virus diluent only.

 
When tested using virus homologous to the original infecting virus, it was not possible to tell if challenge had been done with homologous or heterologous virus. There were interesting differences in post-challenge titers, depending on whether challenge was at 56 or 21 dpi or if testing was done using virus heterologous or homologous to the infecting virus. If challenge was done on 56 dpi, the titers for the heterologous virus usually remained below titer for the homologous virus (Figure 3, A and BGo). However, if challenge was done at 21 dpi, after Day 40–50, challenge titers using the virus heterologous to the original infecting virus usually produced titers comparable to those sera tested with the virus homologous to the infecting virus (Figure 3, C and DGo); this was not true for the homologous virus, which remained markedly lower throughout. This difference indicated that heterologous challenge produced cross-reacting antibody if challenge was done soon after the original infection but that the anamnestic response was more specific if challenge occurred at 56 dpi. With the exception of WNV challenged at 56 dpi, all re-infections resulted in sufficient heterogeneity to confound diagnosis of original and subsequent infections.

When infected concurrently by WNV and SLEV, PRNT titers against WNV (reciprocal geometric mean = 74) were significantly greater (ANOVA F value = 7.4, df = 1,62; P < 0.01) than against SLEV (mean = 27). These values did not change significantly as a function of time, and there was no change in the difference between means over time as indicated by the negative interaction term in the two-way ANOVA (Figure 4Go).


Figure 4
View larger version (12K):
[in this window]
[in a new window]

  		    FIGURE 4. Antibody response measured by PRNT after dual infection with SLEV and WNV and expressed as the inverse of the geometric mean PRNT titer (N = 3 hens per group). Control hens were inoculated with virus diluent only.

 
Western blot. Single infection.
Antibody reactivity with NS1, E, and prM viral proteins by WB analysis showed varying degrees of cross-reactivity for singly infected chickens before challenge (N = 6 for 6–21 days and N = 6 for 6–56 days). Envelope protein was consistently the least specific antigen and was completely cross-reactive from 28 to 56 dpi for both viruses against IgM and IgG (Figure 5, C and DGo). In contrast, WNV NS1 and prM were highly specific and nearly identical for IgG sensitivity. SLEV NS1 protein showed greater specificity than SLEV prM but was less specific than WNV NS1 (Figure 5Go). WNV IgM testing offered no sensitivity advantage over IgG testing; NS1 IgM results were distinctly less sensitive than NS1 IgG results. SLEV IgM and IgG results for individual proteins were more complicated to decipher, but optimum sensitivity resulted from combining prM IgM and IgG results (Figure 5, A and CGo). IgM response and persistence was variable. NS1 IgM was not always detected and was the least persistent antibody response (Figure 5, A and BGo). The 56-day time course for single infection was not sufficient to evaluate IgM persistence against E and prM proteins.


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		    FIGURE 5. IgM and IgG WB results by protein: non-structural protein 1 (NS1), envelope protein (E), and pre-membrane protein (prM).

 
For the purpose of comparison, PRNT and WB shared three positive interpretations: SLEV, WNV, and undifferentiated Flavivirus. The general Flavivirus interpretation for WB was associated only with weakly positive specimens at early time points. When WB strips for WNV and SLEV were both positive, the interpretation was dual positive. PRNT requires a 4-fold difference in titer between SLEV and WNV assays to make the identification; therefore, the Flavivirus interpretation for PRNT sometimes occurred when titers were high and was considered a false-positive result. Sensitivity and specificity statistics based on these interpretations for PRNT and WB were generated for both viruses by day postinfection (Figure 6Go). EIA sensitivity was included for comparison. Although EIA results were generated for both antigens, the test is generally understood to be a sensitive but non-specific screening assay and a positive result for either antigen should only be interpreted as Flavivirus positive.


Figure 6
View larger version (25K):
[in this window]
[in a new window]

  		    FIGURE 6. Sensitivity and specificity comparison by assay. Non-linear regression model (dashed lines) applied to sensitivity data.

 
Sensitivity.
Comparing the nonlinear regression rate constants, 0.06 and 0.12 for PRNT and WB respectively, PRNT was significantly less sensitive than WB for identifying SLEV antibodies (Figure 6AGo; F = 4.972; df = 1,190; P < 0.04). Compared with EIA (K = 0.08), SLEV WB seemed to be marginally more sensitive but not significantly so (P = 0.08). When detecting WNV, sensitivities were mixed but closely matched between 6 and 14 dpi, with both assays reaching 100% from 28 to 56 dpi (Figure 6BGo). WNV sensitivity differences over the entire time course were not significant; rate constants were 0.14 and 0.13 for PRNT and WB, respectively (P > 0.60). Compared with EIA (K = 0.22), WNV WB was significantly less sensitive (P < 0.003), as was PRNT (P < 0.03).

Specificity.
Neither PRNT nor WB produced a false-positive result for any of the three negative control chickens that were each bled 11 times over the course of 56 days. EIA tests of negative control chickens produced some false positives, with specificities of 73% and 91% for WNV and SLEV antigens, respectively. WNV WB during SLEV infection was 100% specific from 6 to 56 dpi, but PRNT produced some false positives (Figure 6DGo). SLEV test specificities during WNV infection were mixed, with both assays averaging equivalent specificities from 6 to 14 dpi (~93%). SLEV PRNT was 100% specific from 28 to 56 dpi compared with ~85% specificity for WB during the same period.

Dual infection.
Although infrequent, SLEV/WNV dual infections can sometimes occur in sentinel chicken flocks. For each primary viral infection, WNV or SLEV, three chickens were challenged with heterologous virus at 21 dpi and three were challenged at 56 dpi. A control group of six chickens were challenged with homologous virus: three at 21 dpi and three at 56 dpi. When SLEV was the primary infection, WB correctly identified, as dual positive, a subsequent WNV infection at an average of ~8 dpi, regardless of when it occurred, for all six chickens that received WNV after initial SLEV inoculation. When WNV was the primary infection followed by SLEV, WB interpretation after challenge was confounded by weak cross-reactivity with SLEV prM and/or NS1 before challenge in three of six chickens, although the six control chickens that received WNV homologous challenge showed no SLEV cross-reactivity before or after challenge. The two chickens that received SLEV at Day 56 after WNV infection and did not cross-react before challenge remained positive for WNV only. The one chicken that received SLEV at Day 21 after WNV infection and did not cross-react before challenge responded with a dual positive result 6 days after challenge. Simultaneous WNV/SLEV infection (N = 3) resulted in dual positive WB interpretations for all three chickens at various days postinfection, with one chicken maintaining the dual positive result from 10 to 21 dpi before reverting back to a WNV-positive interpretation (Table 2Go).


View this table:
[in this window]
[in a new window]

 
TABLE 2
WNV/SLEV simultaneous infection interpretations by day postinfection: SLEV (S), WNV (W), dual positive (SW)*, undifferentiated Flavivirus (F), EIA negative (N), and EIA positive (P)
 

DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Co-circulation of WNV and SLEV has complicated diagnostics in sentinel chicken surveillance programs and prompted the routine use of the PRNT for confirmation and Flavivirus antibody differentiation.17 EIA was found to be a sensitive but non-specific screening assay. During WNV primary infection, mean WNV and SLEV P/N ratios were significantly different and differentiated infections by these viruses; however, in practice, it would not be possible to differentiate a weak positive from a new infection, and although means were different, some hens within different virus groups had similar values. In addition, during SLEV primary infection, a simple comparison of P/N ratios was not adequate for differentiation especially during the first 3 weeks after infection critical in surveillance diagnostics. Although specificity can be increased by manipulating cut-off values for WNV and SLEV P/N ratios, as others have done for humans using IgM antibody capture-enzyme linked immunosorbent assay (MAC-ELISA),24 sensitivity is adversely affected by increased stringency, delaying detection and reporting for some hens. The seasonal nature of arbovirus transmission activity and the time-consuming nature of PRNT slow reporting of results to local control agencies, possibly delay intervention efforts, and thereby increase the risk of human involvement. The Flavivirus WB was developed as a rapid alternative to PRNT, and these results support its use in this regard. During WNV single infection, WB was equally as sensitive as PRNT and averaged higher specificity from 6 to 56 dpi. During SLEV single infection, WB was more sensitive than PRNT from 6 to 56 dpi and equally specific from 6 to 14 dpi. Additionally, WB offers the possibility of a dual positive result that was consistently accurate when SLEV infection preceded WNV infection; however, when WNV preceded SLEV challenge, weakly cross-reactive NS1 and/or prM bands on SLEV strips before challenge invalidated the post-challenge dual positive interpretation for three of six chickens. In practice, the three diagnostic tests EIA, PRNT, and WB can be implemented as a diagnostic algorithm, in which WB routinely confirms EIA positive results and PRNT is used sparingly for the occasional equivocal interpretation. The PRNT assay and preparation of WB strips both require technical skill and hours of labor, but WB strips may be prepared in the off-season and stored frozen until needed. Specimen testing by WB is a simple procedure and could conveniently supplement EIA as a more rapid confirmatory assay, requiring only 1 day compared with the 7- to 10-day PRNT assay. In California, chickens are bled by lancet prick of the comb with blood collected on filter paper strips.7 Samples are screened by testing eluates of these strips by EIA; confirmation by PRNT requires rebleeding of the hens by jugular puncture to collect whole sera, delaying confirmation. WB may be used on the original eluates, thereby precluding repeat travel and the collection of additional whole blood samples for confirmation.

As revealed by back titration, the inocula of WNV and SLEV were at doses of 5.3 and 2.2 log10 PFU/mL, respectively. Although it could be argued that equal inocula would have been preferable, it should be noted that, in nature, it is likely that mosquitoes infect chickens with variable doses of virus and that the stimulus to the immune response would be expected to be dominated by the titer eventually reached in vivo.25,26 For example, although in our study West Nile virus produced a more elevated viremia in adult hens than did SLEV, it has been shown that, in highly susceptible house finches and moderately susceptible mourning doves, the WNV viremic response after 3 dpi is not significantly dose-dependent when the infectious dose ranges from < 1.0 to 4.0 log10 PFU/mL.25 Viremias elicited by both viruses were low and were considered to be insufficient to effectively infect Culex vector mosquitoes,5 an important finding if these birds are to be intentionally placed near human residences to measure transmission activity. Elevated viremia response after WNV infection seemed to also result in a more pronounced antibody response after initial or challenge infection. Interestingly, an initial SLEV infection followed by a WNV infection greatly amplified antibody levels against SLEV, precluding detection of the subsequent WNV infection using a 4-fold difference in titer to separate end point titers. However, the reverse order of infection produced a limited response, and titers to both viruses were fairly similar. A similar immunologic response of original antigenic sin was shown recently in house finches,19 a highly susceptible and competent host for both SLEV and WNV.5,21 Collectively, our data clearly show the difficulties in confirming the identity of the infecting virus soon after Flavivirus seroconversions are detected in sentinel hens and show the importance and use of our new WB assay.

Received May 1, 2007. Accepted for publication December 10, 2007.

Acknowledgments: The authors thank Vincent Martinez and Scott Halam, Center for Vectorborne Diseases, UC Davis, for assistance with the care and blood sampling of the chickens, and Dr. Harvey Motulsky, President, GraphPad Software, for assistance with sensitivity analysis.

Financial support: This research was funded, in part, by NIH Grant AI-65359 and R01-AI55607 to WKR and CDC Grant U50/ CCU923677.

* Address correspondence to Peter J. Patiris, California Department of Public Health, Viral and Rickettsial Disease Laboratory, 850 Marina Bay Parkway, Richmond, CA 94804. E-mail: ppatiris{at}dhs.ca.gov Back

Authors’ addresses: Peter J. Patiris, Leopoldo F. Oceguera III, Robert E. Chiles, and Carl V. Hanson, California Department of Public Health, Viral and Rickettsial Disease Laboratory, 850 Marina Bay Parkway, Richmond, CA 94804, Telephone: 510-307-8555, Fax: 510-307-8955, E-mail: ppatiris{at}dhs.ca.gov. George W. Peck, Northwest Mosquito and Vector Control District, 1966 Compton Avenue, Corona, CA 92881, Telephone: 951-340-9792. William K. Reisen, Center for Vectorborne Diseases, University of California, Old Davis Rd., Davis, CA 95616, Telephone: 530-752-0124.

Reprint requests: Peter J. Patiris, California Department of Public Health, Viral and Rickettsial Disease Laboratory, 850 Marina Bay Parkway, Richmond, CA 94804, Telephone: 510-307-8556, Fax: 510-307-8955, E-mail: ppatiris{at}dhs.ca.gov.


REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Scott TW, Wright SA, Eldridge BF, Brown DA, 2001. Cost effectiveness of three arbovirus surveillance methods in northern California. J Am Mosq Control Assoc 17: 118–123.[Web of Science][Medline]
  2. LaMotte LC, Crane GT, Shriner RB, Kirk LJ, 1967. Use of adult chickens as arbovirus sentinels. I. Viremia and persistence of antibody in experimentally inoculated adult chickens. Am J Trop Med Hyg 16: 348–356.[Abstract/Free Full Text]
  3. Langevin SA, Bunning M, Davis B, Komar N, 2001. Experimental infection of chickens as candidate sentinels for West Nile virus. Emerg Infect Dis 7: 726–729.[Web of Science][Medline]
  4. Reisen WK, Presser SB, Lin J, Enge B, Hardy JL, Emmons RW, 1994. Viremia and serological responses in adult chickens infected with western equine encephalomyelitis and St. Louis encephalitis viruses. J Am Mosq Control Assoc 10: 549–555.[Web of Science][Medline]
  5. Reisen WK, Fang Y, Martinez VM, 2005. Avian host and mosquito (Diptera: Culicidae) vector competence determine the efficiency of West Nile and St. Louis encephalitis virus transmission. J Med Entomol 42: 367–375.[Web of Science][Medline]
  6. Allison AB, Mead DG, Gibbs SE, Hoffman DM, Stallknecht DE, 2004. West Nile virus viremia in wild rock pigeons. Emerg Infect Dis 10: 2252–2255.[Web of Science][Medline]
  7. Reisen WK, Lin J, Presser SB, Enge B, Hardy JL, 1993. Evaluation of new methods for sampling sentinel chickens for antibodies to western equine encephalomyelitis and St. Louis encephalitis viruses. Proc Calif Mosq Vector Control Assoc 61: 33–36.
  8. Reeves WC, Milby MM, Reisen WK, 1990. Development of a statewide arbovirus surveillance program and models of vector populations and virus transmission. Reeves WC, ed. Epidemiology and Control of Mosquito-Borne Arboviruses in California, 1983–1987. Sacramento: Calif. Mosq. Vector Control Assoc, 431–458.
  9. Kramer VL, 2005. California State Mosquito-Borne Virus Surveillance and Response Plan. Richmond: California Department of Health Services.
  10. Hom A, Marcus L, Kramer VL, Cahoon B, Glaser C, Cossen C, Baylis E, Jean C, Tu E, Eldridge BF, Carney R, Padgett K, Sun B, Reisen WK, Woods L, Husted S, 2006. Surveillance for Mosquito-borne Encephalitis Virus Activity and Human Disease, Including West Nile Virus in California, 2005. Sacramento: Proceedings and Papers of the Mosquito and Vector Control Association of California, 43–54.
  11. Hogrefe WR, Moore R, Lape-Nixon M, Wagner M, Prince HE, 2004. Performance of immunoglobulin G (IgG) and IgM enzyme-linked immunosorbent assays using a West Nile virus recombinant antigen (preM/E) for detection of West Nile virus- and other flavivirus-specific antibodies. J Clin Microbiol 42: 4641–4648.[Abstract/Free Full Text]
  12. Wong SJ, Boyle RH, Demarest VL, Woodmansee AN, Kramer LD, Li H, Drebot M, Koski RA, Fikrig E, Martin DA, Shi PY, 2003. Immunoassay targeting nonstructural protein 5 to differentiate West Nile virus infection from dengue and St. Louis encephalitis virus infections and from Flavivirus vaccination. J Clin Microbiol 41: 4217–4223.[Abstract/Free Full Text]
  13. Johnson AJ, Langevin S, Wolff KL, Komar N, 2003. Detection of anti-west nile virus immunoglobulin M in chicken serum by an enzyme-linked immunosorbent assay. J Clin Microbiol 41: 2002–2007.[Abstract/Free Full Text]
  14. Malan AK, Martins TB, Hill HR, Litwin CM, 2004. Evaluations of commercial West Nile virus immunoglobulin G (IgG) and IgM enzyme immunoassays show the value of continuous validation. J Clin Microbiol 42: 727–733.[Abstract/Free Full Text]
  15. Tilley PA, Zachary GA, Walle R, Schnee PF, 2005. West Nile virus detection and commercial assays. Emerg Infect Dis 11: 1154–1155.[Web of Science][Medline]
  16. Wang T, Anderson JF, Magnarelli LA, Bushmich S, Wong S, Koski RA, Fikrig E, 2001. West Nile virus envelope protein: role in diagnosis and immunity. Ann N Y Acad Sci 951: 325–327.[Web of Science][Medline]
  17. Weingartl HM, Drebot MA, Hubalek Z, Halouzka J, Andonova M, Dibernardo A, Cottam-Birt C, Larence J, Marszal P, 2003. Comparison of assays for the detection of West Nile virus antibodies in chicken serum. Can J Vet Res 67: 128–132.[Web of Science][Medline]
  18. Inouye S, Matsuno S, Tsurukubo Y, 1984. "Original antigenic sin" phenomenon in experimental Flavivirus infections of guinea pigs: studies by enzyme-linked immunosorbent assay. Microbiol Immunol 28: 569–574.[Web of Science][Medline]
  19. Fang Y, Reisen WK, 2006. Previous infection with West Nile or St. Louis encephalitis viruses provides cross protection during reinfection in house finches. Am J Trop Med Hyg 75: 480–485.[Abstract/Free Full Text]
  20. Oceguera LF 3rd, Patiris PJ, Chiles RE, Busch MP, Tobler LH, Hanson CV, 2007. Flavivirus serology by Western blot analysis. Am J Trop Med Hyg 77: 159–163.[Abstract/Free Full Text]
  21. Reisen WK, Chiles RE, Martinez VM, Fang Y, Green EN, 2003. Experimental infection of California birds with western equine encephalomyelitis and St. Louis encephalitis viruses. J Med Entomol 40: 968–982.[Web of Science][Medline]
  22. Kramer LD, Wolfe TM, Green EN, Chiles RE, Fallah H, Fang Y, Reisen WK, 2002. Detection of encephalitis viruses in mosquitoes (Diptera: Culicidae) and avian tissues. J Med Entomol 39: 312–323.[Web of Science][Medline]
  23. Lindsey HS, Calisher CH, Mathews JH, 1976. Serum dilution neutralization test for California group virus identification and serology. J Clin Microbiol 4: 503–510.[Abstract/Free Full Text]
  24. Martin DA, Noga A, Kosoy O, Johnson AJ, Peterson LR, Lanciotti RS, 2004. Evaluation of a diagnostic algorithm using immunoglobulin M enzyme–linked immunosorbent assay to differentiate human West Nile virus and St. Louis encephalitis virus infections during the 2002 West Nile epidemic in the United States. Clin Diagn Lab Immunol 11: 1130–1133.[Medline]
  25. Reisen WK, Fang Y, Martinez VM, 2005. Avian host and mosquito (Diptera: Culicidae) vector competence determine the efficiency of West Nile and St. Louis encephalitis virus transmission. J Med Entomol 42: 367–375.[Web of Science][Medline]
  26. Styer LM, Bernard KA, Kramer LD, 2006. Enhanced early West Nile virus infection in young chickens infected by mosquito bite: effect of viral dose. Am J Trop Med Hyg 75: 337–345.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Patiris, P. J.
Right arrow Articles by Hanson, C. V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Patiris, P. J.
Right arrow Articles by Hanson, C. V.
Related Collections
Right arrow Flaviviruses
Right arrow Arboviruses

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS