Diagnostic Approach for Differentiating Infected f
http://www.100md.com
微生物临床杂志 2005年第2期
Southeast Poultry Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture
Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, Georgia
ABSTRACT
Vaccination programs for the control of avian influenza (AI) in poultry have limitations due to the problem of differentiating between vaccinated and virus-infected birds. We have used NS1, the conserved nonstructural protein of influenza A virus, as a differential diagnostic marker for influenza virus infection. Experimentally infected poultry were evaluated for the ability to induce antibodies reactive to NS1 recombinant protein produced in Escherichia coli or to chemically synthesized NS1 peptides. Immune sera were obtained from chickens and turkeys inoculated with live AI virus, inactivated purified vaccines, or inactivated commercial vaccines. Seroconversion to positivity for antibodies to the NS1 protein was achieved in birds experimentally infected with multiple subtypes of influenza A virus, as determined by enzyme-linked immunosorbent assay (ELISA) and Western blot analysis. In contrast, animals inoculated with inactivated gradient-purified vaccines had no seroconversion to positivity for antibodies to the NS1 protein, and animals vaccinated with commercial vaccines had low, but detectable, levels of NS1 antibodies. The use of a second ELISA with diluted sera identified a diagnostic test that results in seropositivity for antibodies to the NS1 protein only in infected birds. For the field application phase of this study, serum samples were collected from vaccinated and infected poultry, diluted, and screened for anti-NS1 antibodies. Field sera from poultry that received commercial AI vaccines were found to possess antibodies against AI virus, as measured by the standard agar gel precipitin (AGP) test, but they were negative by the NS1 ELISA. Conversely, diluted field sera from AI-infected poultry were positive for both AGP and NS1 antibodies. These results demonstrate the potential benefit of a simple, specific ELISA for anti-NS1 antibodies that may have diagnostic value for the poultry industries.
INTRODUCTION
Avian influenza (AI) is a viral disease of poultry that causes a wide range of disease signs. There are two main AI pathotypes: highly pathogenic AI (HPAI) and low pathogenic AI (LPAI) virus infections. HPAI virus infections, which are caused by some viruses of the H5 and H7 subtypes, usually result in multiorgan systemic disease, with high rates of morbidity and mortality. LPAI viruses are more commonly isolated from domestic poultry and may be associated with mild respiratory disease and reductions in egg production (1, 10, 34, 41). When an AI virus infects poultry, the virus is excreted from both the respiratory and the digestive tracts, resulting in the rapid spread through a population of susceptible hosts (37). Wild waterfowl and shorebirds provide a reservoir for all 15 influenza A virus hemagglutinin (HA) subtypes, but infections in these species generally do not produce clinical signs (1, 7, 26, 27, 29). Although chickens and turkeys represent abnormal hosts for AI virus infections (32), the viruses have crossed over from the wild bird reservoir to infect domestic poultry (13, 43, 44). Over the last decade, AI viruses circulating in live-bird markets have provided a secondary reservoir from which influenza viruses have crossed over to infect commercial chicken and turkey operations (23, 28, 31).
Virological surveillance over the last several years has shown that multiple subtypes of influenza A virus have been isolated from poultry (41, 42). Notably, in 2002, an LPAI virus caused a major outbreak in commercial poultry in Virginia and adjacent states. During that outbreak, mild respiratory disease, decreased egg production, and increased mortality were associated with the presence of an H7N2 virus, detected primarily in turkey flocks (41). More than 4.7 million turkeys and chickens were depopulated, and the outbreak was estimated to have cost the poultry industry more than $120 million (41). The recent H7N2 North American isolates appear to be well adapted for poultry and replicate to high titers in the respiratory tract for up to 7 days after infection (40). In 2003, another outbreak of H7N2 LPAI virus infection occurred in a large egg layer complex in Connecticut and in a single layer operation in Rhode Island (42). In 2004, H7N2 LPAI virus was detected on two farms in Delaware and one farm in Maryland and resulted in the quarantine of more than 100 commercial and noncommercial flocks. Many other subtypes of AI viruses have been recovered from poultry and other gallinaceous birds in recent years, including the H1N1, H5N2, H5N3, H6N2, and H7N3 subtypes (41, 42). The H5 and H7 LPAI virus subtypes present a much greater concern because some of these viruses have undergone phenotypic shifts to HPAI virus variants both in the field and in laboratory settings (1, 16, 24, 43).
The traditional method of AI vaccination of poultry includes the use of inactivated whole-virus vaccine licensed for parenteral (subcutaneous [s.c.] or intramuscular) administration. These inexpensive commercial AI virus vaccines are produced from unpurified allantoic fluid emulsified in an oil-based proprietary adjuvant. Although AI vaccines have been successful at providing protection against clinical signs and death among poultry (11, 14, 35), they are not commonly used as part of programs for the control and eradication of LPAI or HPAI virus infections (14). For a number of reasons, including the concerns of trade embargoes on poultry and poultry products, quarantine and elimination of infected flocks (i.e., stamping-out or controlled marketing) are often preferred measures for controlling AI. An additional limitation is that traditional AI virus vaccines can interfere with serologic surveillance; hence, these vaccines induce antibodies that are indistinguishable from live virus infection, as determined by the traditional agar gel precipitin (AGP) test, commercial enzyme-linked immunosorbent assays (ELISAs), and hemagglutination inhibition (HI) tests (36). The AGP test is considered the international standard for the serologic diagnosis of AI among poultry (2, 3).
The NS1 protein of influenza A is a nonstructural protein expressed in large amounts in virus-infected cells, but it has not been detected in virions (17). The phylogenetic relationships of the NS genes have revealed two different gene lineages, referred to as groups or alleles A and B. AI viruses as well as mammalian influenza viruses separate into the A group, whereas the B group contains virtually all AI virus strains (19, 30, 39). This protein is synthesized early in infection and has been implicated in inhibition of the host antiviral defense mediated by alpha and beta interferons (12). The interferon antagonist properties of the NS1 protein depend on its ability to bind double-stranded RNA, a known potent inducer of interferon. Selected amino acid residues within the NS1 protein play a critical role in the ability of this protein to bind to double-stranded RNA, and mutations of just two amino acids can result in an attenuated virus (9). This may partially explain the high degree of NS1 conservation among influenza A virus strains (5, 19, 30) and provides an antigenic marker for influenza virus infections (4, 22). Nonstructural protein has also been used as a marker for hepatitis C virus (15) and foot-and-mouth disease virus infections (20).
The main goal of the present study was to establish a diagnostic test for the differentiation of AI-vaccinated poultry and AI virus-infected poultry. We describe a differential test, an ELISA for the NS1 virus protein, based on the production of antibodies to this protein that are readily detectable in sera from infected poultry. Such a diagnostic test based on the detection of conserved NS1 antigens could be useful for differentiating multiple influenza A virus subtypes.
MATERIALS AND METHODS
Viruses and NS1 reagents. The AI viruses used in this study, A/Turkey/Wisconsin/68 (H5N9, TW/68), A/Turkey/Virginia/15851/02 (H7N2, TV/02), A/Chicken/Pennslyvania/21342/97 (H7N2, CP/97), A/Chicken/New York/13142-5/94 (H7N2, CNY/94), A/Rhea/North Carolina/39482/93 (H7N1, RNC/93), and A/Turkey/Ontario/7732/66 (H5N2, TO/66, HPAI), were received in allantoic fluid from the National Veterinary Services Laboratories (courtesy of Dennis Senne, Ames, Iowa). Human influenza A virus A/PR/8/34 (H1N1, PR/8) was obtained from the Centers for Disease Control and Prevention (Atlanta, Ga.), and A/Duck/Alberta/35/76 (DK/Ab/76, H1N1) was obtained from the American Type Culture Collection (Manassas, Va.). RNC/93 and H1N1 viruses are representatives of the NS1(A) gene lineage, whereas the remaining avian viruses used in these studies are NS1(B) influenza viruses.
Virus stocks were propagated for 30 to 37 h in the allantoic cavities of 10-day-old embryonated chickens' eggs at 37°C. Infectious allantoic fluid was aliquoted and stored at –70°C. Fifty percent egg infectious dose (EID50) titers were determined by serial titration of viruses in eggs, and endpoints were calculated by the method of Reed and Muench (25). All virus stocks had high infectivity titers in eggs (7.7 to 9.2 log10 EID50s/ml). To select the peptides to be synthesized for this study, amino acid sequences of multiple influenza A virus isolates, including both NS1 allele (group) A and B viruses, were aligned in the Megalign computer analysis program (DNASTAR, Madison, Wis.). Specific sequences from conserved areas of the NS1 gene were identified by using the Protean computer program (DNASTAR) to identify areas that were predicted to be hydrophilic and antigenic. Two ELISA peptides with NS1(B) sequences 28-GDAPFLDRLRRDQK-42 (NS129-41) and 35-LRRDQKALKGRGS-49 (NS136-48) were selected on the basis of their high degrees of antigenicity and were synthesized by Biopeptide Co. (San Diego, Calif.). The maltose binding protein (MBP)-NS1(B) fusion protein was prepared from A/Mallard/WI/458/75 viruses, as described below. The glutathione S-transferase (GST)-NS1 fusion protein (generously provided by Adolfo García-Sastre, Mount Sinai School of Medicine, New York, N.Y.) encodes a protein consisting of GST fused to the NS1(A) protein of A/WSN/33 virus (9).
Cloning of the NS1 gene. For plasmid construction, the NS1-coding sequence was amplified by PCR from a previously cloned NS gene segment from A/Mallard/WI/458/75 virus. The PCR primers contained the restriction sites XmnI and XbaI. The primer used for Mallard/WI/428/75 was GCGAGAAGGATTTCAATGGACTCCAACACGATAAC [NS(B)-XmnI]. The PCR products were run on a 1.5% agarose gel in TBE (Tris-borate-EDTA) buffer, and the 700-bp bands were excised and extracted by use of a QIAquick gel extraction kit (Qiagen, Valencia, Calif.). The purified PCR product and the pMAL-c2 vector were digested separately with 0.5 U of XmnI and XbaI per μl in Promega (Madison, Wis.) buffer B in the presence of 0.5 μg of bovine serum albumin per μl at 37°C overnight. pMAL-c2 was dephosphorylated with shrimp alkaline phosphatase (Promega) according to the recommendations of the manufacturer. The inserts and the vector were ligated with a Ligafast Rapid DNA ligation system (Promega). The ligation products were transformed into competent Escherichia coli cells (Invitrogen, Carlsbad, Calif.) and plated on Luria broth plates containing 100 μg of ampicillin per ml.
For protein expression, E. coli clones of pMAL-c2 with the NS1 inserts were grown in Luria broth, Miller (Sigma, St. Louis, Mo.), supplemented with 0.2% glucose and 100 μg of ampicillin per ml at 37°C in a shaking incubator at 250 rpm. Isopropyl--D-thiogalactopyranoside (0.3 mM) was added to cultures with an absorbance at 600 nm of ca. 0.5, and the cultures were incubated for an additional 2 h. The culture fluid was centrifuged at 4,000 x g for 20 min at 4°C in an SLA-3000 rotor in a Sorvall RC 5B plus centrifuge. The supernatants were discarded, and the cells were resuspended in column buffer containing 20 mM Tris-HCl (pH 7.4), 200 mM NaCl, and 1 mM EDTA. After overnight incubation at –20°C, the cell pellet was thawed and the cells were lysed by sonication on ice at 15 W in 20-s pulses. The lysate was centrifuged in a Sorvall SLA-1500 rotor at 9,000 x g for 40 min at 4°C.
Chromatography. The lysate supernatant was loaded on an amylose affinity resin (New England Biolabs, Beverly, Mass.) in a Kontes column (10 by 1.75 cm), and protein elution was monitored by measurement of the absorbance at 280 on a BioLogic LP system (Bio-Rad Laboratories, Hercules, Calif.). The column was washed with 2 volumes of column buffer, and the MSB-NS1 fusion protein was eluted with column buffer plus 10 mM maltose. The fractions that eluted with maltose were concentrated by ultrafiltration in an Amicon stirred cell by using nominal molecular weight limit (NMWL) 10,000 membranes (Millipore, Billerica, Mass.), and the buffer was exchanged for 20 mM Tris-HCl and 25 mM NaCl (pH 8.0). The desalted MBP-NS1 fractions were loaded on a DEAE Sepharose FF column (10 by 1 cm; Sigma). After two washes with 1 volume of column buffer, proteins were eluted with a 0 to 16% gradient of 500 mM NaCl in 20 mM Tris-HCl buffer (pH 8.0). Fractions containing the MBP-NS1 protein were desalted and concentrated by ultrafiltration and then stored in 20 mM Tris-HCl (pH 7.4) plus 5% glycerol.
Vaccine preparation. The commercial AI vaccines, subtypes H1N1 and H7N2, were prepared by Lohmann Animal Health (Gainesville, Ga.). The seed stocks, A/Swine/North Carolina/17366/01 (H1N1) and A/Chicken/Pennslyvania/21342/97 (CP/97, H7N2) viruses, were propagated in embryonated chicken eggs, inactivated with formalin, and emulsified in a proprietary oil-based adjuvant. The purified vaccines (TW/68 and CP/97) were concentrated from allantoic fluid and purified by equilibrium density centrifugation through a 30 to 60% linear sucrose gradient, as described previously (8). This purification strategy removes allantoic fluid and egg proteins and leaves morphologically intact virions. For inactivation, purified whole viruses were adjusted to a protein concentration of 1 mg/ml and treated with 0.025% formalin at 4°C for 3 days. The treatment resulted in the complete loss of infectivity, as determined by titration of the vaccine preparations in eggs. The purified vaccine doses given throughout are expressed as amounts of total protein measured by the Bradford assay (Bio-Rad Laboratories) at 595 nm.
Inoculations. Four-week-old specific-pathogen-free white Plymouth Rock (WPR) chickens (Southeast Poultry Research Laboratory [SEPRL], Athens, Ga.) and medium white toms (MWTs; obtained from the British United Turkeys of America, Lewisburg, W.Va.) were used to generate experimental immune sera. The chickens were housed in stainless steel isolation cabinets ventilated under negative pressure with HEPA-filtered air. Turkeys were maintained in self-contained isolation units (Mark 4; Controlled Isolation Systems, San Diego, Calif.) ventilated under negative pressure. Animal care was provided as required by the Institutional Animal Care and Use Committee. Feed and water were provided ad libitum. Groups of four to five chickens and turkeys were immunized with the commercial inactivated vaccine (undiluted) or gradient-purified vaccine (10 μg) one to three times by s.c. inoculation of 0.5 ml of vaccine in the nape of the neck. Boosted poultry received a second inoculation 14 days after the initial vaccination, and birds receiving a third vaccination were inoculated 16 days after the second vaccination. Control birds received the same volume of normal allantoic fluid emulsified in the same adjuvant (Lohmann Animal Health). Controls were identified as sham vaccinated. Some vaccinated turkeys were subsequently challenged with live virus, as described previously (40). In brief, at 29 days after the initial vaccination, turkeys were challenged intranasally (i.n.) with 107.0 EID50s of TV/02 virus in a 200-μl volume. For virus infection of nave birds, groups of four to five chickens and turkeys were infected i.n. or intravenously (i.v.) with 106.0 EID50s (unless otherwise indicated) of infectious allantoic fluid in a 200-μl volume of brain heart infusion medium (Difco, Detroit, Mich.).
AGP and HI antibody tests. Serum samples were collected from individual chickens or turkeys on various days before and after vaccination or infection. AGP and HI assays were performed by standard methods (2, 3). Each HI test was performed in V-bottom, 96-well microtiter plates (Corning-Costar Co., Cambridge, Mass.) with 4 hemagglutinating units (HAU) of the indicated H7N2 virus antigen and 0.5% chicken erythrocytes. For the AGP tests, precipitating antibodies were detected by using A/Turkey/Wisconsin/129/66 (H9N2) virus as the test antigen and chicken antiserum with antibodies against this virus as the positive control serum.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot assays. The protein concentrations of the recombinant MBP-NS1 or GST-NS1 fusion protein were determined with Bradford's reagent (Bio-Rad Laboratories). Protein samples were denatured in Laemmli's sample buffer (Bio-Rad Laboratories) and boiled for 5 min. Denatured proteins (6 μg of protein/lane) were separated in a sodium dodecyl sulfate-4 to 20% polyacrylamide gel gradient (Criterion; Bio-Rad Laboratories) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 120 V for 2 h. The NS1 protein preparations were then transferred to nitrocellulose, and the blots were blocked with 20% (wt/vol) nonfat skim milk overnight at 4°C and washed three times in phosphate-buffered saline (PBS). Poultry sera were diluted 1:10, incubated for 1 h at 37°C, and then washed three times in PBS. The secondary antibody goat anti-turkey immunoglobulin G (IgG) or goat anti-chicken IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Md.) was added (1:500), and the mixture was incubated for 1 h at 37°C. The preparations were then developed by using an alkaline phosphatase substrate kit (Vector, Burlingame, Calif.). Anti-GST antibody (Bethyl Laboratories, Montgomery, Tex.) was used at a 1:1,000 dilution.
ELISA for NS1 detection. Serum samples were diluted in titer tubes 1 day before testing by ELISA. The serum diluent for the NS1 fusion protein ELISAs consisted of 5% (wt/vol) nonfat skim milk containing 0.5% normal E. coli cell lysate in PBS. The E. coli cell lysate was omitted for the serum diluent used on NS1 peptide-coated ELISA plates. Serum samples were heat inactivated at 56°C for 30 min, cooled, and then placed on ELISA plates. Immunolon II ELISA plates (Dynatech Laboratories, Chantilly, Va.) were coated with NS1 fusion protein (1 μg/well) in PBS and incubated at room temperature overnight. The NS1 peptides (1 μg/well) were diluted in carbonate-bicarbonate buffer (pH 9.6) and incubated overnight at 4°C. The bound antibody was detected by the addition of goat anti-turkey IgG or goat anti-chicken IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, Inc.). The absorbance at 405 nm was measured for 30 min following the addition of 3-ethylbenzthiazoline-6-sulfonic acid (Kirkegaard & Perry Laboratories, Inc.). Some ELISA plates were coated with purified whole influenza virus (50 HAU of gradient-purified H5N9 [TW/68] or H7N2 [CP/97] per well) to show the overall influenza virus antibody response in vaccinated poultry compared with that in infected poultry. This protocol detected antibodies directed against internal NP or M1 proteins as well as antibodies directed against the surface glycoproteins. For the initial ELISA (Table 1), the titers were expressed as the highest dilution that yielded an optical density (OD) greater than the mean plus 3 standard deviations for similarly diluted control sera. For subsequent ELISAs, the sera were prediluted and samples were considered positive if they yielded an OD greater than the mean plus 4 standard deviations for the control sera.
Field samples. For the first set of field serum samples, 20 samples (generously provided by Eric Gonder, Goldsboro Milling Company, Goldsboro, Mass.) from two different turkey breeder flocks were collected after the birds received two or three vaccinations with a commercial H1N1 subtype virus vaccine. At 23 weeks of age, these birds received an initial s.c. vaccination with a killed autogenous preparation derived from an H1N1 virus of turkey origin in North Carolina. A vaccine boost was given at 29 weeks of age; and a third vaccination, which consisted of a killed autogenous preparation derived from an H1N1 virus of swine origin in North Carolina in 2001, was given at 34 weeks of age. For the second set of field serum samples, samples from 25 randomly selected, clinically recovered birds (generously provided by William Smith and Robert Brady, Animal and Plant Health Inspection Service, U.S. Department of Agriculture, Sutton, Mass.) were collected and tested by the AGP test and the NS1 ELISA. The field sera were collected from a single red layer operation located in Providence, R.I., that was found to be infected on 16 April 2003 with an H7N2 LPAI virus (42). This layer operation consisted of a single house with 30,000 multiage hens.
RESULTS
Anti-NS1 antibody response as a measure of influenza virus infection. Sera from WPR chickens that received live virus, commercial AI vaccine, or gradient-purified AI vaccines were tested for the presence or absence of specific antibodies to the NS1 protein. Chickens in the last two groups were immunized one to three times s.c., and sera were collected 1 day before administration of a vaccine boost and 2 weeks after the final boost. Chickens infected with TW/68, RNC/93, CK/NY/94, CP/97, or TV/02 LPAI virus, representatives of the H5 or H7 subtype, were inoculated i.n. or i.v. with infectious allantoic fluid (Table 1). Two different NS1 fusion proteins, GST-NS1 and MBP-NS1, were used as representatives of the NS(A) and the NS(B) alleles, respectively, and were used to coat the ELISA plates. As shown in Table 1, immunization by any method resulted in a substantial induction of IgG antibody responses to whole influenza virus. A boost(s) with inactivated vaccines enhanced the response by 16- to 250-fold. The gradient-purified vaccine given three times induced the highest titers of antibody to whole influenza virus antigen among all groups; however, anti-NS1 antibodies were undetectable in these birds. In contrast, live H5 or H7 virus induced the highest anti-NS1 antibody responses among the three groups, with titers ranging from 100 to 400. Antisera from poultry infected with viruses representative of groups NS1(A) and NS1(B) reacted similarly to both fusion proteins. Chickens that received one to three inoculations of commercial vaccine had low but detectable antibody responses (titers, 25 to 100) to the NS1 protein, indicating the presence of low levels of NS1 protein in the commercial AI (H7N2) vaccine (Table 1). No ELISA responses (titer, <25) were detected against MBP or the GST protein (data not shown).
Western immunoblotting was performed as a second serologic assay to detect the presence of serum IgG antibodies to the NS1 protein. The source of the antiserum used for this assay was a pool of four to five chickens that were either infected or vaccinated one to three times with H5 or H7 LPAI virus. Regardless of the fusion protein used, Western blot analysis confirmed the ELISA results by detecting anti-NS1 antibodies in birds that received live H5 (TW/68) or H7 (RNC/93) virus but not in chickens that received purified vaccine (Fig. 1). Faint bands of reactivity to the NS1 protein were observed in sera from chickens that received two vaccinations (data not shown) or three vaccinations (Fig. 1A and B, lanes 5) with the commercial H7N2 vaccine. No NS1 antibodies were observed in normal chicken sera or sera from chickens that received normal allantoic fluid emulsified in commercial adjuvant (sham-vaccinated chickens). Furthermore, no reactions against MBP or the GST protein itself were observed; however, the anti-GST antibody reacted strongly to the GST-NS1 fusion protein (Fig. 1A, lane 1). Taken together, these results confirm that anti-NS1 antibodies were induced to the greatest levels in birds infected with live AI virus in comparison to the levels in birds inoculated with inactivated AI vaccines.
Because sera from birds vaccinated with the commercial vaccine had low reactivities to the NS1 protein, we next evaluated the ability of diluted serum from immunized chickens to react with the NS1 protein. Pooled serum samples (n = 5) were tested by the MBP-NS1 protein ELISA at dilutions of 1:10, 1:50, 1:100, and 1:200. The mean OD values for sham-vaccinated and normal serum samples (five of each) plus 4 standard deviations set the cutoff OD value used to flag samples for infection. All serum samples from chickens vaccinated with the commercial vaccine and virus-infected chickens were positive at dilutions of 1:10 and 1:50 (data not shown). At a 1:100 dilution, pooled antisera from chickens that received the commercial AI vaccine plus a single boost did not react with the NS1 protein; however, pooled sera from the group of birds that received two boosts were borderline positive for the NS1 protein (Table 2). The ELISA revealed that a 1:200 dilution of immune serum from birds that received the commercial H7 vaccine did not react to the NS1 protein (Table 2). In contrast, 100% of infected birds tested positive for NS1 antibody when their sera were tested at a 1:200 dilution. For the MBP-NS1-coated ELISA plates, OD values greater than or equal to 0.177 (the cutoff value) were considered positive. Individual serum samples (five per group) were also run at the dilutions indicated above and confirmed the results for the pooled sera. These results demonstrate that by diluting immune sera 1:200, a specific NS1 ELISA can provide an antigenic marker for influenza virus infections among poultry.
Application of NS1 ELISA for diagnostic testing of field samples. Field validation of the NS1 ELISA method was undertaken by testing immune sera from domestic poultry that had been vaccinated or infected with AI virus. Sera from 20 H1N1-vaccinated turkeys along with matching preimmune samples collected before vaccination were tested for AGP and NS1 antibodies. The commercial influenza H1N1 vaccine was a killed autogenous preparation that uses a 2001 North Carolina H1N1 subtype virus of swine or turkey origin. For comparison, sera from five experimentally H1N1-infected MWT turkeys collected 4 weeks after i.n. infection were also tested. Sera from 20 turkeys that had been vaccinated two to three times and that had been found to be AGP positive reacted negatively in the MBP-NS1 ELISA (Table 3) and the GST-NS1 ELISA (data not shown). In contrast, NS1 antibody could be measured in diluted sera from all five turkeys that received live H1N1 virus. Positive OD values for the sera from H1N1-infected turkeys ranged from 0.322 to 0.569 and from 0.237 to 0.489 on ELISA plates coated with the MBP-NS1 and the GST-NS1 fusion proteins, respectively.
Field validation of the NS1 ELISA was also performed with sera from 25 H7N2-infected red layers that had clinically recovered. This LPAI virus was isolated in April 2003 from a single layer operation located in Rhode Island. For comparison, birds in an experimental group of five WPR chickens were immunized three times with the commercial H7N2 vaccine and tested for AGP and NS1 antibodies. Sera from these birds were collected 2 weeks after the final vaccine boost and were found to be positive for AGP antibodies but negative for antibodies to the NS1 protein when the sera were tested at a dilution of 1:200 (Table 3). In contrast, 19 of 25 serum samples from the H7N2-infected chickens, which were positive for antibody by the standard AGP test, were also positive for antibodies to the MPB-NS1 fusion protein. To determine if two synthetic NS1 peptides could serve as antigens capable of detecting antibodies in the sera of AI virus-infected poultry, peptides NS129-41 and NS136-48 were synthesized on the basis of the predicted antigenicity of NS1(B) consensus sequences. Chicken anti-H7 sera were reactive against the MBP-NS1 fusion protein and the NS136-48 peptide but reacted poorly with the NS129-41 peptide. Positive OD values for the sera from H7N2-infected turkeys ranged from 0.322 to 0.569, 0.106 to 0.338, and 0.08 to 0.141 on ELISA plates coated with the MBP-NS1 fusion protein, the NS136-48 peptide, and NS129-41 peptide, respectively.
In field situations, vaccinated chickens and turkeys may subsequently be exposed to live virus; and although infectious virus can be detected among vaccinated poultry (33, 35, 40), the low level of virus replication may result in a low NS1 antibody response. Therefore, the MBP-NS1 and peptide ELISAs were used to test sera from 20 MWT turkeys that were vaccinated two to three times with the commercially available H7N2 vaccine and subsequently challenged with 107 EID50s of TV/02 virus. As shown in Table 3, sera collected from all 20 turkeys 2 weeks after virus challenge were positive for AGP and NS1 antibodies. Turkey anti-H7 sera were reactive against the fusion protein and the NS136-48 peptide, but immune sera from infected poultry reacted poorly to the NS129-41 peptides, with only 4 of 20 field serum samples being positive by the peptide ELISA. Taken together, these results suggest that an ELISA for the measurement of antibodies against NS1 recombinant protein or an NS1 peptide may be useful for the diagnosis of AI virus infections among poultry flocks.
DISCUSSION
Because LPAI virus infections among chickens and turkeys are often asymptomatic, the diagnosis requires serologic monitoring. Most poultry diagnostic laboratories prefer the serologic AGP test because of its simplicity and broad specificity for the detection of type A influenza virus infections among poultry. AGP antibodies recognize conserved nucleoprotein (NP) and matrix (M1) influenza virus proteins present in the sera of poultry exposed to AI virus and therefore can detect multiple subtypes of influenza A virus (3). Other serologic tests for AI include an ELISA, which also measures antibody responses to conserved internal proteins, and the subtype-specific HI assay (21). However, these standard serologic tests for AI virus exposure do not differentiate between vaccinated and infected poultry when traditional vaccines are used (36). An improved serologic assay for AI virus infection would allow the identification of poultry that have been infected with AI virus, while it would correctly exclude animals that have been vaccinated with any subtype. The diagnostic test described in this study used a conserved nonstructural protein that is not associated with virions but that is expressed in large amounts in influenza virus-infected cells (17). Testing by sensitive immunoassays for antibodies against the NS1 antigen, we compared the NS1 antibody responses in vaccinated and infected poultry and found that live virus infection induced the highest levels of NS1 antibody compared with the amount induced by commercial vaccines, which induced little or no detectable antibody to NS1 protein. Following dilution of immune sera, the NS1 ELISA was an effective tool for the screening of poultry sera for evidence of AI virus infection.
The NS1 protein, although a weak antigen in comparison to other influenza virus proteins, is remarkably conserved among type A influenza viruses (5, 19, 30, 39). Previous studies demonstrated that the NS1 protein could be detected in the sera of horses experimentally infected with the H3 subtype of influenza virus but not in the immune sera of animals immunized with inactivated viruses (4, 22). The inactivated whole-virus equine vaccine used in the aforementioned studies was partially purified, thus removing residual NS1 protein that would be present in the infectious medium. This allowed the clear distinction between H3-vaccinated and H3-infected horses by using the NS1 protein as a differential marker (4, 22). In our study, antiserum from poultry that received commercial vaccines had some weak reactivity to the NS1 protein, as detected by the NS1 ELISA and Western blot analysis. This reactivity was not observed in poultry that received normal allantoic fluid emulsified in the same commercial adjuvant. Thus, it seems likely that the NS1 protein is present at low levels in the unpurified, infectious allantoic fluid of embryonated chicken eggs used for the production of commercial AI vaccines. AI virus replicates and causes the death of the endothelial cells lining the allantoic sac. This results in the release of the cell contents, which potentially could include some NS1 protein. However, we were unable to detect NS1 protein in infectious allantoic fluid by using Western blot analysis and a number of different (monoclonal and polyclonal) antibodies to NS1 protein (data not shown). In contrast to commercially vaccinated poultry, immune sera from birds that received gradient-purified vaccines had no detectable levels of NS1 antibody, even though these animals displayed a high-level antibody response to other influenza virus proteins (Fig. 1B and Table 1). Due to the lack of NS1 protein in purified virions, it was not surprising that these animals were unable to generate an antibody response to NS1 protein, even after three immunizations. We then sought to determine the serum dilutions that would allow the differentiation of vaccinated and infected poultry. At dilutions of 1:10 and 1:50, most sera from poultry that received the traditional commercial AI vaccine still reacted weakly to the NS1 protein by the NS1 ELISA. However, optimal discrimination between vaccinated and infected poultry was achieved with sera diluted 1:200, a dilution at which sera from infected poultry yielded positive ELISA reactions but sera from vaccinated poultry gave consistently negative results. Using this value, we were able to provide a test that identifies poultry as positive or negative for influenza virus infection.
Concerning the two different NS gene lineages described (19, 30, 39), the selection of an NS1 protein from NS1(A) or NS1(B) may have to be considered to optimize the reactivity of immune sera. However, there are four highly conserved regions among both groups (19), and this conservation would allow the NS1 ELISA to work for viruses found in the NS1(A) and NS1(B) groups. Although extensive testing of the antigenic cross-reactivities of the two different groups of NS genes was beyond the scope of the present study, the diagnostic specificity of the NS1 antibody response did not appear to be confined to the NS1(A) and the NS1(B) groups. Our data indicate that there are antigenic cross-reactions between the gene lineages because antisera from poultry infected with the NS1(A) or NS1(B) type of virus reacted equally well to the GST-NS1 and the MBP-NS1 fusion proteins (Table 1).
Confirmation of the anti-NS1 test result required verification in tests with field sera from animals that represented populations that might be targeted. Therefore, field sera from infected or vaccinated poultry, which possessed AGP antibodies, were chosen for testing; hence, it was unlikely that these poultry were not properly infected or vaccinated since they demonstrated AI virus exposure when their sera were tested by the standard serologic test. A second serologic test was performed with sera from experimental birds to demonstrate that vaccinated poultry responded to the H5 and H7 vaccines with increases in serum IgG antibody levels. The highest serum antibody titers to whole influenza virus detected by ELISA were observed in poultry sera collected after three vaccinations. Although each vaccine (commercial or purified) boost significantly enhanced the IgG antibody responses to whole influenza proteins, the specific NS1 antibody response never displayed an endpoint ELISA titer of greater than 100. Particular emphasis was placed on NS1 antibody responses to influenza H5 and H7 virus strains, because these are the subtypes most frequently found in AI outbreaks worldwide (34). We also found that the NS1 ELISA could be a diagnostic tool for the determination of H1N1 virus infection of poultry (Table 3). Avian influenza H1N1 virus has been a problem in U.S. commercial turkey flocks in areas with high densities of swine, and some states routinely vaccinate flocks against the H1N1 AI virus (41, 42).
The influenza virus NS1 proteins used in these studies were expressed in bacterial cells, and purification of the protein was carried out to remove the bulk of the bacterial proteins. Initial preparations of unpurified NS1 proteins expressed in E. coli resulted in high ELISA background levels (OD value range, 0.235 to 0.588) in sera from older (age, >4 weeks) uninfected chickens and turkeys (data not shown). This nonspecific reactivity to unpurified E. coli-derived NS1 proteins may be due to cross-reactive antibodies to common bacterial antigens, particularly those present in the serum of older poultry. To overcome this problem, the MBP-NS1 and GST-NS1 purified expression systems were used to remove the bulk of the E. coli proteins, which resulted in significant reductions (OD value range, 0.09 to 0.149) in the reactivity of uninfected poultry sera to bacterial proteins. In addition, serum samples were diluted with 0.5% normal E. coli cell lysate. To further reduce the background cross-reactivity to negligible levels and to avoid the addition of E. coli cell lysate, we tested the immune reactivity of N-terminal NS1-specific peptides synthesized on the basis of the predicted antigenicity of consensus sequences. The two different peptides used in these studies encompass a 13-amino-acid region within the NS1 protein. Our data demonstrated that the NS129-41 peptide was weakly antigenic compared with the antigenicities of the MBP-NS1 fusion protein and the NS136-48 peptide. Although the NS136-48 peptide was less antigenic than the full-length MBP-NS1 fusion protein, the former was still able to detect anti-NS1 antibodies exclusively in the sera of poultry infected with influenza virus and gave consistently lower background levels by the NS1 ELISA. In regards to the diagnostic sensitivity of the NS1 antibody response, we found that it was at least equal to that of the AGP test. Thus, sera from infected poultry that possessed AGP antibodies also displayed NS1 antibodies, whereas the field chickens that did not mount an AGP antibody response also failed to show anti-NS1 antibodies (Table 3).
AI is a disease of economic importance and remains a continuous threat to commercial poultry operations. During the initial stages of an AI virus outbreak, the feasibility of using an AI vaccine would depend on a number of factors, including the ability to differentiate between vaccinated and infected poultry. A diagnostic tool that differentiates vaccinated from infected poultry will provide valuable information to state veterinarians and poultry industries when they are considering vaccination as an option to control LPAI. A recently developed test has allowed the differentiation of infected from vaccinated animals by using a heterologous neuraminidase (NA) subtype and an anti-NA antibody detection test (6). This strategy has provided some success but has limitations because of the availability of vaccine strains that require proper combinations of the HA and the NA subtypes. Thus, the HA subtype of the vaccine strain must have a high degree of homology to the HA subtype of the circulating strain and, in addition, must possess a different NA subtype; however, the use of the reverse genetics system would allow a more flexible strategy for the differentiation of infected from vaccinated poultry (18). The results from this study indicate that the use of a simple and specific ELISA for the serodiagnosis of infection with multiple LPAI virus subtypes can distinguish infected poultry from vaccinated poultry. This diagnostic test may lay the groundwork for additional strategies that will yield optimal results. One such strategy is the construction of recombinant influenza viruses with a truncated or deleted NS1 protein that could be used for the generation of influenza vaccines (38). Such recombinant vaccines would lack NS1 epitopes and would provide a means for the clear distinction between vaccinated and virus-infected poultry.
ACKNOWLEDGMENTS
We thank Karen Burns from Lohmann Animal Health for providing vaccine material. We also thank Suzanne Deblois, Joan Beck, Marinda Thomas, Laura Kelly, and Rashida Horton for technical assistance and Roger Brock for animal care assistance. We thank Chang-Won Lee and Claudia Chesley for critical review of the manuscript.
This research was supported by USDA, ARS, CRIS project 6612-32000-040-00D.
REFERENCES
Alexander, D. J. 2000. A review of avian influenza in different bird species. Vet. Microbiol. 74:3-13.
Beard, C. W. 1970. Avian influenza antibody detection by immunodiffusion. Avian Dis. 14:337-341.
Beard, C. W. 1989. Serological procedures, p. 192-200. In H. Friedman (ed.) A laboratory manual for the isolation and identification of avian pathotypes. Kendall-Hunt Publishing, Dubuque, Iowa.
Birch-Machin, I., A. Rowan, J. Pick, J. Mumford, and M. Binns. 1997. Expression of the nonstructural protein NS1 of equine influenza A virus: detection of anti-NS1 antibody in post infection equine sera. J. Virol. Methods 65:255-263.
Buonagurio, D. A., S. Nakada, J. D. Parvin, M. Krystal, P. Palese, and W. M. Fitch. 1986. Evolution of human influenza A viruses over 50 years: rapid, uniform rate of change in NS gene. Science 232:980-982.
Capua, I., C. Terregino, G. Cattoli, F. Mutinelli, and J. F. Rodriguez. 2003. Development of a DIVA (differentiating infected from vaccinated animals) strategy using a vaccine containing a heterologous neuraminidase for the control of avian influenza. Avian Pathol. 32:47-55.
Condoberry, P. K., and R. D. Slemons. 1992. Biological properties of waterfowl-origin type A influenza viruses in chickens. Avian Dis. 36:17-23.
Cox, N. J., and A. P. Kendal. 1984. Genetic stability of A/Ann/6/60 cold-mutant (temperature-sensitive) live influenza virus genes; analysis by oligonucleotide mapping of recombinant vaccine strains before and after replication in volunteers. J. Infect. Dis. 149:194-200.
Donelan, N. R., C. F. Basler, and A. García-Sastre. 2003. A recombinant influenza A virus expressing an RNA-binding-defective NS1 protein induces high levels of beta interferon and is attenuated in mice. J. Virol. 77:13257-13266.
Easterday B. C., V. S. Hinshaw and D. A. Halvorson. 1997. Influenza, p. 583-605. In B. W. Calnek, H. J. Barnes, C. W. Beard, L. R. McDougald, and Y. M. Saif (ed.), Diseases of poultry. Iowa State University Press, Ames.
Garcia, A., J. Hector, D. K. Srivastava, D. A. Jayawardene, D. R. Wehr, and R. G. Webster. 1998. Efficacy of inactivated H5N2 influenza vaccines against lethal A/Chicken/Queretaro/19/95 infection. Avian Dis. 42:248-256.
García-Sastre, A., A. Egorov, D. Matassov, S. Brandt, D. E. Levy, J. E. Durbin, P. Palese, and T. Muster. 1998. Influenza A virus lacking the NS1 gene replicates in interferon-deficient systems. Virology 252:324-330.
Halvorson, D. A., C. J. Kelleher, and D. A. Senne. 1985. Epizootiology of avian influenza: effect of season on incidence in sentinel ducks and domestic turkeys in Minnesota. Appl. Environ. Microbiol. 49:914-919.
Halvorson, D. A. 2002. The control of H5 or H7 mildly pathogenic avian influenza—a role for inactivated vaccine. Avian Pathol. 31:5-12.
Inoue, Y., R. Suzuki, Y. Matsuura, S. Harada, J. Chiba, Y. Watanabe, I. Saito, and T. Miyamura. 1992. Expression of the amino-terminal half of the NS1 region of the hepatitis C virus genome and detection of an antibody to the expressed protein in patients with liver disease. J. Gen. Virol. 73:2151-2154.
Kawaoka, Y., C. W. Naeve, and R. G. Webster. 1984. Is virulence of H5N2 influenza viruses in chickens associated with loss of carbohydrate from the hemagglutinin Virology 139:303-316.
Krug, R. M., and P. R. Etkind. 1973. Cytoplasmic and nuclear virus-specific proteins in influenza virus-infected MDCK cells. Virology 56:334-348.
Lee, C.-W., D. A. Senne, and D. L. Suarez. 2004. Generation of reassortant influenza vaccines by reverse genetics that allows utilization of a DIVA (differentiating infected from vaccinated animals) strategy for the control of avian influenza. Vaccine 22:3175-3181.
Ludwig, S., U. Schultz, J. Mandler, W. M. Fitch, and C. Scholtissek. 1991. 1991. Phylogenic relationship of the nonstructural (NS) genes of influenza A viruses. Virology 183:566-577.
Niezert, E., E. Beck, P. A. De Mello, I. Gomes, and I. E. Bergmann. 1991. Expression of the aphtovirus RNA polymerase gene in Escherichia coli and its use with other bioengineered nonstructural antigens in detection of late persistent infections. Virology 184:799-804.
Office International des Epizooties. 1996. Highly pathogenic avian influenza (fowl plague), p. 155-160. In Manual of standards for diagnostic tests and vaccines, vol. 3. Office International des Epizooties, Paris, France.
Ozaki, H., T. Sugiura, S. Sugita, H. Imagawa, and H. Kida. 2001. Detection of antibodies to the nonstructural protein (NS1) of influenza A virus allows distinction between vaccinated and infected horses. Vet. Microbiol. 82:111-119.
Panigrahy, B., D. A. Senne, and J. C. Pedersen. 2002. Avian influenza virus subtypes inside and outside the live bird markets, 1993-2000; a spatial and temporal relationship. Avian Dis. 46:298-307.
Perdue, M. L., M. Garcia, D. Senne, and M. Fraire. 1997. Virulence-associated sequence duplication at the hemagglutinin cleavage site of avian influenza viruses. Virus Res. 49:173-186.
Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty per cent endpoints. Am. J. Hyg. 27:493-497.
Slemons, R. D., J. S. Osborn, and F. Hayes. 1974. Viruses isolated from wild free-flying ducks in California. Avian Dis. 18:119-124.
Slemons, R. D., and B. C. Easterday. 1977. Type-A influenza viruses in the feces of migratory waterfowl. J. Am. Vet. Med. Assoc. 171:947-948.
Spackman, E., D. A. Senne, S. Davison, and D. L. Suarez. 2003. Sequence analysis of recent H7 avian influenza viruses associated with three different outbreaks in commercial poultry in the United States. J. Virol. 77:13399-13402.
Stallknecht D. E. 1998. Ecology and epidemiology of avian influenza viruses in wild bird populations: waterfowl, shorebirds, pelicans, cormorants, etc., p. 61-69. In D. E. Swayne and R. D. Slemons (ed.), Proceedings of the 4th International Symposium on Avian Influenza. U.S. Animal Health Association, Richmond, Va.
Suarez, D. L., and M. L. Perdue. 1999. Multiple alignment comparison of the non-structural genes of influenza A viruses. Virus Res. 54:59-69.
Suarez, D. L., M. Garcia, J. Latimer, D. A. Senne, and M. L. Perdue. 1999. Phylogenetic analysis of H7 avian influenza viruses isolated from live bird markets of the Northeast United States. J. Virol. 73:3567-3573.
Suarez, D. L. 2000. Evolution of avian influenza viruses. Vet. Microbiol. 74:15-27.
Swayne, D. E., J. R. Beck, M. Garcia, and H. D. Stone. 1999. Influence of virus strain and antigen mass on efficacy of H5 avian influenza inactivated vaccines. Avian Pathol. 28:245-255.
Swayne, D. E., and D. L. Suarez. 2000. Highly pathogenic avian influenza. Rev. Sci. Tech. Off. Int. Epizoot. 19:463-482.
Swayne, D. E., J. R. Beck, M. L. Perdue, and C. W. Beard. 2001. Efficacy of vaccines in chickens against highly pathogenic Hong Kong H5N1 avian influenza. Avian Dis. 45:355-365.
Swayne, D. E. 2003. Vaccines for list A poultry diseases: emphasis on avian influenza. Dev. Biologics (Basel) 114:201-212.
Swayne, D. E., and D. A. Halvorson. 2003. Influenza, p. 135-160. In Y. M. Saif, H. J. Barns, J. R. Glisson, A. M. Fadly, L. R. McDougald, and D. E. Swayne (ed.), Diseases of poultry. Iowa State University Press, Ames.
Talon, J., M. Salvatore, R. E. O'Neill, Y. Nakaya, H. Zheng, T. Muster, A. García-Sastre, and P. Palese. 2000. Influenza A and B viruses expressing altered NS1 proteins: a vaccine approach. Proc. Natl. Acad. Sci. USA 97:4309-4314.
Treanor, J. J., M. H. Synder, W. T. London, and B. R. Murphy. 1989. The B allele of the NS gene of avian influenza viruses, but not the A allele, attenuate a human influenza A virus for squirrel monkeys. Virology 171:1-9.
Tumpey, T. M., D. R. Kapczynski, and D. E. Swayne. 2004. Comparative susceptibility of chickens and turkeys to avian influenza A H7N2 virus infection and protective efficacy of a commercial avian influenza H7N2 virus vaccine. Avian Dis. 48:167-176.
U.S. Animal Health Association. 2002. Report of the Committee on Transmissible Diseases of Poultry and Other Avian Species. Criteria for determining that an AI virus isolation causing an outbreak must be considered for eradication, p. 524-586. In Proceedings of the 106th Annual Meeting of the U.S. Animal Health Association. U.S. Animal Health Association, Richmond, Va.
U.S. Animal Health Association. 2003. Report of the Committee on Transmissible Diseases of Poultry and Other Avian Species. Avian Influenza Workshop—Recent Outbreaks and Developments, p. 312-337. In Proceedings of the 107th Annual Meeting of the U.S. Animal Health Association. U.S. Animal Health Association, Richmond, Va.
Webster, R. G., W. J. Bean, O. T. Gorman, T. M. Chambers, and Y. Kawaoka. 1992. Evolution and ecology of influenza A viruses. Microbiol. Rev. 56:152-179.
Webster, R. G., and Y. Kawaoka. 1994. Influenza: an emerging and reemerging disease. Semin. Virol. 5:103-111.(Terrence M. Tumpey, Rene )
Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, Georgia
ABSTRACT
Vaccination programs for the control of avian influenza (AI) in poultry have limitations due to the problem of differentiating between vaccinated and virus-infected birds. We have used NS1, the conserved nonstructural protein of influenza A virus, as a differential diagnostic marker for influenza virus infection. Experimentally infected poultry were evaluated for the ability to induce antibodies reactive to NS1 recombinant protein produced in Escherichia coli or to chemically synthesized NS1 peptides. Immune sera were obtained from chickens and turkeys inoculated with live AI virus, inactivated purified vaccines, or inactivated commercial vaccines. Seroconversion to positivity for antibodies to the NS1 protein was achieved in birds experimentally infected with multiple subtypes of influenza A virus, as determined by enzyme-linked immunosorbent assay (ELISA) and Western blot analysis. In contrast, animals inoculated with inactivated gradient-purified vaccines had no seroconversion to positivity for antibodies to the NS1 protein, and animals vaccinated with commercial vaccines had low, but detectable, levels of NS1 antibodies. The use of a second ELISA with diluted sera identified a diagnostic test that results in seropositivity for antibodies to the NS1 protein only in infected birds. For the field application phase of this study, serum samples were collected from vaccinated and infected poultry, diluted, and screened for anti-NS1 antibodies. Field sera from poultry that received commercial AI vaccines were found to possess antibodies against AI virus, as measured by the standard agar gel precipitin (AGP) test, but they were negative by the NS1 ELISA. Conversely, diluted field sera from AI-infected poultry were positive for both AGP and NS1 antibodies. These results demonstrate the potential benefit of a simple, specific ELISA for anti-NS1 antibodies that may have diagnostic value for the poultry industries.
INTRODUCTION
Avian influenza (AI) is a viral disease of poultry that causes a wide range of disease signs. There are two main AI pathotypes: highly pathogenic AI (HPAI) and low pathogenic AI (LPAI) virus infections. HPAI virus infections, which are caused by some viruses of the H5 and H7 subtypes, usually result in multiorgan systemic disease, with high rates of morbidity and mortality. LPAI viruses are more commonly isolated from domestic poultry and may be associated with mild respiratory disease and reductions in egg production (1, 10, 34, 41). When an AI virus infects poultry, the virus is excreted from both the respiratory and the digestive tracts, resulting in the rapid spread through a population of susceptible hosts (37). Wild waterfowl and shorebirds provide a reservoir for all 15 influenza A virus hemagglutinin (HA) subtypes, but infections in these species generally do not produce clinical signs (1, 7, 26, 27, 29). Although chickens and turkeys represent abnormal hosts for AI virus infections (32), the viruses have crossed over from the wild bird reservoir to infect domestic poultry (13, 43, 44). Over the last decade, AI viruses circulating in live-bird markets have provided a secondary reservoir from which influenza viruses have crossed over to infect commercial chicken and turkey operations (23, 28, 31).
Virological surveillance over the last several years has shown that multiple subtypes of influenza A virus have been isolated from poultry (41, 42). Notably, in 2002, an LPAI virus caused a major outbreak in commercial poultry in Virginia and adjacent states. During that outbreak, mild respiratory disease, decreased egg production, and increased mortality were associated with the presence of an H7N2 virus, detected primarily in turkey flocks (41). More than 4.7 million turkeys and chickens were depopulated, and the outbreak was estimated to have cost the poultry industry more than $120 million (41). The recent H7N2 North American isolates appear to be well adapted for poultry and replicate to high titers in the respiratory tract for up to 7 days after infection (40). In 2003, another outbreak of H7N2 LPAI virus infection occurred in a large egg layer complex in Connecticut and in a single layer operation in Rhode Island (42). In 2004, H7N2 LPAI virus was detected on two farms in Delaware and one farm in Maryland and resulted in the quarantine of more than 100 commercial and noncommercial flocks. Many other subtypes of AI viruses have been recovered from poultry and other gallinaceous birds in recent years, including the H1N1, H5N2, H5N3, H6N2, and H7N3 subtypes (41, 42). The H5 and H7 LPAI virus subtypes present a much greater concern because some of these viruses have undergone phenotypic shifts to HPAI virus variants both in the field and in laboratory settings (1, 16, 24, 43).
The traditional method of AI vaccination of poultry includes the use of inactivated whole-virus vaccine licensed for parenteral (subcutaneous [s.c.] or intramuscular) administration. These inexpensive commercial AI virus vaccines are produced from unpurified allantoic fluid emulsified in an oil-based proprietary adjuvant. Although AI vaccines have been successful at providing protection against clinical signs and death among poultry (11, 14, 35), they are not commonly used as part of programs for the control and eradication of LPAI or HPAI virus infections (14). For a number of reasons, including the concerns of trade embargoes on poultry and poultry products, quarantine and elimination of infected flocks (i.e., stamping-out or controlled marketing) are often preferred measures for controlling AI. An additional limitation is that traditional AI virus vaccines can interfere with serologic surveillance; hence, these vaccines induce antibodies that are indistinguishable from live virus infection, as determined by the traditional agar gel precipitin (AGP) test, commercial enzyme-linked immunosorbent assays (ELISAs), and hemagglutination inhibition (HI) tests (36). The AGP test is considered the international standard for the serologic diagnosis of AI among poultry (2, 3).
The NS1 protein of influenza A is a nonstructural protein expressed in large amounts in virus-infected cells, but it has not been detected in virions (17). The phylogenetic relationships of the NS genes have revealed two different gene lineages, referred to as groups or alleles A and B. AI viruses as well as mammalian influenza viruses separate into the A group, whereas the B group contains virtually all AI virus strains (19, 30, 39). This protein is synthesized early in infection and has been implicated in inhibition of the host antiviral defense mediated by alpha and beta interferons (12). The interferon antagonist properties of the NS1 protein depend on its ability to bind double-stranded RNA, a known potent inducer of interferon. Selected amino acid residues within the NS1 protein play a critical role in the ability of this protein to bind to double-stranded RNA, and mutations of just two amino acids can result in an attenuated virus (9). This may partially explain the high degree of NS1 conservation among influenza A virus strains (5, 19, 30) and provides an antigenic marker for influenza virus infections (4, 22). Nonstructural protein has also been used as a marker for hepatitis C virus (15) and foot-and-mouth disease virus infections (20).
The main goal of the present study was to establish a diagnostic test for the differentiation of AI-vaccinated poultry and AI virus-infected poultry. We describe a differential test, an ELISA for the NS1 virus protein, based on the production of antibodies to this protein that are readily detectable in sera from infected poultry. Such a diagnostic test based on the detection of conserved NS1 antigens could be useful for differentiating multiple influenza A virus subtypes.
MATERIALS AND METHODS
Viruses and NS1 reagents. The AI viruses used in this study, A/Turkey/Wisconsin/68 (H5N9, TW/68), A/Turkey/Virginia/15851/02 (H7N2, TV/02), A/Chicken/Pennslyvania/21342/97 (H7N2, CP/97), A/Chicken/New York/13142-5/94 (H7N2, CNY/94), A/Rhea/North Carolina/39482/93 (H7N1, RNC/93), and A/Turkey/Ontario/7732/66 (H5N2, TO/66, HPAI), were received in allantoic fluid from the National Veterinary Services Laboratories (courtesy of Dennis Senne, Ames, Iowa). Human influenza A virus A/PR/8/34 (H1N1, PR/8) was obtained from the Centers for Disease Control and Prevention (Atlanta, Ga.), and A/Duck/Alberta/35/76 (DK/Ab/76, H1N1) was obtained from the American Type Culture Collection (Manassas, Va.). RNC/93 and H1N1 viruses are representatives of the NS1(A) gene lineage, whereas the remaining avian viruses used in these studies are NS1(B) influenza viruses.
Virus stocks were propagated for 30 to 37 h in the allantoic cavities of 10-day-old embryonated chickens' eggs at 37°C. Infectious allantoic fluid was aliquoted and stored at –70°C. Fifty percent egg infectious dose (EID50) titers were determined by serial titration of viruses in eggs, and endpoints were calculated by the method of Reed and Muench (25). All virus stocks had high infectivity titers in eggs (7.7 to 9.2 log10 EID50s/ml). To select the peptides to be synthesized for this study, amino acid sequences of multiple influenza A virus isolates, including both NS1 allele (group) A and B viruses, were aligned in the Megalign computer analysis program (DNASTAR, Madison, Wis.). Specific sequences from conserved areas of the NS1 gene were identified by using the Protean computer program (DNASTAR) to identify areas that were predicted to be hydrophilic and antigenic. Two ELISA peptides with NS1(B) sequences 28-GDAPFLDRLRRDQK-42 (NS129-41) and 35-LRRDQKALKGRGS-49 (NS136-48) were selected on the basis of their high degrees of antigenicity and were synthesized by Biopeptide Co. (San Diego, Calif.). The maltose binding protein (MBP)-NS1(B) fusion protein was prepared from A/Mallard/WI/458/75 viruses, as described below. The glutathione S-transferase (GST)-NS1 fusion protein (generously provided by Adolfo García-Sastre, Mount Sinai School of Medicine, New York, N.Y.) encodes a protein consisting of GST fused to the NS1(A) protein of A/WSN/33 virus (9).
Cloning of the NS1 gene. For plasmid construction, the NS1-coding sequence was amplified by PCR from a previously cloned NS gene segment from A/Mallard/WI/458/75 virus. The PCR primers contained the restriction sites XmnI and XbaI. The primer used for Mallard/WI/428/75 was GCGAGAAGGATTTCAATGGACTCCAACACGATAAC [NS(B)-XmnI]. The PCR products were run on a 1.5% agarose gel in TBE (Tris-borate-EDTA) buffer, and the 700-bp bands were excised and extracted by use of a QIAquick gel extraction kit (Qiagen, Valencia, Calif.). The purified PCR product and the pMAL-c2 vector were digested separately with 0.5 U of XmnI and XbaI per μl in Promega (Madison, Wis.) buffer B in the presence of 0.5 μg of bovine serum albumin per μl at 37°C overnight. pMAL-c2 was dephosphorylated with shrimp alkaline phosphatase (Promega) according to the recommendations of the manufacturer. The inserts and the vector were ligated with a Ligafast Rapid DNA ligation system (Promega). The ligation products were transformed into competent Escherichia coli cells (Invitrogen, Carlsbad, Calif.) and plated on Luria broth plates containing 100 μg of ampicillin per ml.
For protein expression, E. coli clones of pMAL-c2 with the NS1 inserts were grown in Luria broth, Miller (Sigma, St. Louis, Mo.), supplemented with 0.2% glucose and 100 μg of ampicillin per ml at 37°C in a shaking incubator at 250 rpm. Isopropyl--D-thiogalactopyranoside (0.3 mM) was added to cultures with an absorbance at 600 nm of ca. 0.5, and the cultures were incubated for an additional 2 h. The culture fluid was centrifuged at 4,000 x g for 20 min at 4°C in an SLA-3000 rotor in a Sorvall RC 5B plus centrifuge. The supernatants were discarded, and the cells were resuspended in column buffer containing 20 mM Tris-HCl (pH 7.4), 200 mM NaCl, and 1 mM EDTA. After overnight incubation at –20°C, the cell pellet was thawed and the cells were lysed by sonication on ice at 15 W in 20-s pulses. The lysate was centrifuged in a Sorvall SLA-1500 rotor at 9,000 x g for 40 min at 4°C.
Chromatography. The lysate supernatant was loaded on an amylose affinity resin (New England Biolabs, Beverly, Mass.) in a Kontes column (10 by 1.75 cm), and protein elution was monitored by measurement of the absorbance at 280 on a BioLogic LP system (Bio-Rad Laboratories, Hercules, Calif.). The column was washed with 2 volumes of column buffer, and the MSB-NS1 fusion protein was eluted with column buffer plus 10 mM maltose. The fractions that eluted with maltose were concentrated by ultrafiltration in an Amicon stirred cell by using nominal molecular weight limit (NMWL) 10,000 membranes (Millipore, Billerica, Mass.), and the buffer was exchanged for 20 mM Tris-HCl and 25 mM NaCl (pH 8.0). The desalted MBP-NS1 fractions were loaded on a DEAE Sepharose FF column (10 by 1 cm; Sigma). After two washes with 1 volume of column buffer, proteins were eluted with a 0 to 16% gradient of 500 mM NaCl in 20 mM Tris-HCl buffer (pH 8.0). Fractions containing the MBP-NS1 protein were desalted and concentrated by ultrafiltration and then stored in 20 mM Tris-HCl (pH 7.4) plus 5% glycerol.
Vaccine preparation. The commercial AI vaccines, subtypes H1N1 and H7N2, were prepared by Lohmann Animal Health (Gainesville, Ga.). The seed stocks, A/Swine/North Carolina/17366/01 (H1N1) and A/Chicken/Pennslyvania/21342/97 (CP/97, H7N2) viruses, were propagated in embryonated chicken eggs, inactivated with formalin, and emulsified in a proprietary oil-based adjuvant. The purified vaccines (TW/68 and CP/97) were concentrated from allantoic fluid and purified by equilibrium density centrifugation through a 30 to 60% linear sucrose gradient, as described previously (8). This purification strategy removes allantoic fluid and egg proteins and leaves morphologically intact virions. For inactivation, purified whole viruses were adjusted to a protein concentration of 1 mg/ml and treated with 0.025% formalin at 4°C for 3 days. The treatment resulted in the complete loss of infectivity, as determined by titration of the vaccine preparations in eggs. The purified vaccine doses given throughout are expressed as amounts of total protein measured by the Bradford assay (Bio-Rad Laboratories) at 595 nm.
Inoculations. Four-week-old specific-pathogen-free white Plymouth Rock (WPR) chickens (Southeast Poultry Research Laboratory [SEPRL], Athens, Ga.) and medium white toms (MWTs; obtained from the British United Turkeys of America, Lewisburg, W.Va.) were used to generate experimental immune sera. The chickens were housed in stainless steel isolation cabinets ventilated under negative pressure with HEPA-filtered air. Turkeys were maintained in self-contained isolation units (Mark 4; Controlled Isolation Systems, San Diego, Calif.) ventilated under negative pressure. Animal care was provided as required by the Institutional Animal Care and Use Committee. Feed and water were provided ad libitum. Groups of four to five chickens and turkeys were immunized with the commercial inactivated vaccine (undiluted) or gradient-purified vaccine (10 μg) one to three times by s.c. inoculation of 0.5 ml of vaccine in the nape of the neck. Boosted poultry received a second inoculation 14 days after the initial vaccination, and birds receiving a third vaccination were inoculated 16 days after the second vaccination. Control birds received the same volume of normal allantoic fluid emulsified in the same adjuvant (Lohmann Animal Health). Controls were identified as sham vaccinated. Some vaccinated turkeys were subsequently challenged with live virus, as described previously (40). In brief, at 29 days after the initial vaccination, turkeys were challenged intranasally (i.n.) with 107.0 EID50s of TV/02 virus in a 200-μl volume. For virus infection of nave birds, groups of four to five chickens and turkeys were infected i.n. or intravenously (i.v.) with 106.0 EID50s (unless otherwise indicated) of infectious allantoic fluid in a 200-μl volume of brain heart infusion medium (Difco, Detroit, Mich.).
AGP and HI antibody tests. Serum samples were collected from individual chickens or turkeys on various days before and after vaccination or infection. AGP and HI assays were performed by standard methods (2, 3). Each HI test was performed in V-bottom, 96-well microtiter plates (Corning-Costar Co., Cambridge, Mass.) with 4 hemagglutinating units (HAU) of the indicated H7N2 virus antigen and 0.5% chicken erythrocytes. For the AGP tests, precipitating antibodies were detected by using A/Turkey/Wisconsin/129/66 (H9N2) virus as the test antigen and chicken antiserum with antibodies against this virus as the positive control serum.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot assays. The protein concentrations of the recombinant MBP-NS1 or GST-NS1 fusion protein were determined with Bradford's reagent (Bio-Rad Laboratories). Protein samples were denatured in Laemmli's sample buffer (Bio-Rad Laboratories) and boiled for 5 min. Denatured proteins (6 μg of protein/lane) were separated in a sodium dodecyl sulfate-4 to 20% polyacrylamide gel gradient (Criterion; Bio-Rad Laboratories) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 120 V for 2 h. The NS1 protein preparations were then transferred to nitrocellulose, and the blots were blocked with 20% (wt/vol) nonfat skim milk overnight at 4°C and washed three times in phosphate-buffered saline (PBS). Poultry sera were diluted 1:10, incubated for 1 h at 37°C, and then washed three times in PBS. The secondary antibody goat anti-turkey immunoglobulin G (IgG) or goat anti-chicken IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Md.) was added (1:500), and the mixture was incubated for 1 h at 37°C. The preparations were then developed by using an alkaline phosphatase substrate kit (Vector, Burlingame, Calif.). Anti-GST antibody (Bethyl Laboratories, Montgomery, Tex.) was used at a 1:1,000 dilution.
ELISA for NS1 detection. Serum samples were diluted in titer tubes 1 day before testing by ELISA. The serum diluent for the NS1 fusion protein ELISAs consisted of 5% (wt/vol) nonfat skim milk containing 0.5% normal E. coli cell lysate in PBS. The E. coli cell lysate was omitted for the serum diluent used on NS1 peptide-coated ELISA plates. Serum samples were heat inactivated at 56°C for 30 min, cooled, and then placed on ELISA plates. Immunolon II ELISA plates (Dynatech Laboratories, Chantilly, Va.) were coated with NS1 fusion protein (1 μg/well) in PBS and incubated at room temperature overnight. The NS1 peptides (1 μg/well) were diluted in carbonate-bicarbonate buffer (pH 9.6) and incubated overnight at 4°C. The bound antibody was detected by the addition of goat anti-turkey IgG or goat anti-chicken IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, Inc.). The absorbance at 405 nm was measured for 30 min following the addition of 3-ethylbenzthiazoline-6-sulfonic acid (Kirkegaard & Perry Laboratories, Inc.). Some ELISA plates were coated with purified whole influenza virus (50 HAU of gradient-purified H5N9 [TW/68] or H7N2 [CP/97] per well) to show the overall influenza virus antibody response in vaccinated poultry compared with that in infected poultry. This protocol detected antibodies directed against internal NP or M1 proteins as well as antibodies directed against the surface glycoproteins. For the initial ELISA (Table 1), the titers were expressed as the highest dilution that yielded an optical density (OD) greater than the mean plus 3 standard deviations for similarly diluted control sera. For subsequent ELISAs, the sera were prediluted and samples were considered positive if they yielded an OD greater than the mean plus 4 standard deviations for the control sera.
Field samples. For the first set of field serum samples, 20 samples (generously provided by Eric Gonder, Goldsboro Milling Company, Goldsboro, Mass.) from two different turkey breeder flocks were collected after the birds received two or three vaccinations with a commercial H1N1 subtype virus vaccine. At 23 weeks of age, these birds received an initial s.c. vaccination with a killed autogenous preparation derived from an H1N1 virus of turkey origin in North Carolina. A vaccine boost was given at 29 weeks of age; and a third vaccination, which consisted of a killed autogenous preparation derived from an H1N1 virus of swine origin in North Carolina in 2001, was given at 34 weeks of age. For the second set of field serum samples, samples from 25 randomly selected, clinically recovered birds (generously provided by William Smith and Robert Brady, Animal and Plant Health Inspection Service, U.S. Department of Agriculture, Sutton, Mass.) were collected and tested by the AGP test and the NS1 ELISA. The field sera were collected from a single red layer operation located in Providence, R.I., that was found to be infected on 16 April 2003 with an H7N2 LPAI virus (42). This layer operation consisted of a single house with 30,000 multiage hens.
RESULTS
Anti-NS1 antibody response as a measure of influenza virus infection. Sera from WPR chickens that received live virus, commercial AI vaccine, or gradient-purified AI vaccines were tested for the presence or absence of specific antibodies to the NS1 protein. Chickens in the last two groups were immunized one to three times s.c., and sera were collected 1 day before administration of a vaccine boost and 2 weeks after the final boost. Chickens infected with TW/68, RNC/93, CK/NY/94, CP/97, or TV/02 LPAI virus, representatives of the H5 or H7 subtype, were inoculated i.n. or i.v. with infectious allantoic fluid (Table 1). Two different NS1 fusion proteins, GST-NS1 and MBP-NS1, were used as representatives of the NS(A) and the NS(B) alleles, respectively, and were used to coat the ELISA plates. As shown in Table 1, immunization by any method resulted in a substantial induction of IgG antibody responses to whole influenza virus. A boost(s) with inactivated vaccines enhanced the response by 16- to 250-fold. The gradient-purified vaccine given three times induced the highest titers of antibody to whole influenza virus antigen among all groups; however, anti-NS1 antibodies were undetectable in these birds. In contrast, live H5 or H7 virus induced the highest anti-NS1 antibody responses among the three groups, with titers ranging from 100 to 400. Antisera from poultry infected with viruses representative of groups NS1(A) and NS1(B) reacted similarly to both fusion proteins. Chickens that received one to three inoculations of commercial vaccine had low but detectable antibody responses (titers, 25 to 100) to the NS1 protein, indicating the presence of low levels of NS1 protein in the commercial AI (H7N2) vaccine (Table 1). No ELISA responses (titer, <25) were detected against MBP or the GST protein (data not shown).
Western immunoblotting was performed as a second serologic assay to detect the presence of serum IgG antibodies to the NS1 protein. The source of the antiserum used for this assay was a pool of four to five chickens that were either infected or vaccinated one to three times with H5 or H7 LPAI virus. Regardless of the fusion protein used, Western blot analysis confirmed the ELISA results by detecting anti-NS1 antibodies in birds that received live H5 (TW/68) or H7 (RNC/93) virus but not in chickens that received purified vaccine (Fig. 1). Faint bands of reactivity to the NS1 protein were observed in sera from chickens that received two vaccinations (data not shown) or three vaccinations (Fig. 1A and B, lanes 5) with the commercial H7N2 vaccine. No NS1 antibodies were observed in normal chicken sera or sera from chickens that received normal allantoic fluid emulsified in commercial adjuvant (sham-vaccinated chickens). Furthermore, no reactions against MBP or the GST protein itself were observed; however, the anti-GST antibody reacted strongly to the GST-NS1 fusion protein (Fig. 1A, lane 1). Taken together, these results confirm that anti-NS1 antibodies were induced to the greatest levels in birds infected with live AI virus in comparison to the levels in birds inoculated with inactivated AI vaccines.
Because sera from birds vaccinated with the commercial vaccine had low reactivities to the NS1 protein, we next evaluated the ability of diluted serum from immunized chickens to react with the NS1 protein. Pooled serum samples (n = 5) were tested by the MBP-NS1 protein ELISA at dilutions of 1:10, 1:50, 1:100, and 1:200. The mean OD values for sham-vaccinated and normal serum samples (five of each) plus 4 standard deviations set the cutoff OD value used to flag samples for infection. All serum samples from chickens vaccinated with the commercial vaccine and virus-infected chickens were positive at dilutions of 1:10 and 1:50 (data not shown). At a 1:100 dilution, pooled antisera from chickens that received the commercial AI vaccine plus a single boost did not react with the NS1 protein; however, pooled sera from the group of birds that received two boosts were borderline positive for the NS1 protein (Table 2). The ELISA revealed that a 1:200 dilution of immune serum from birds that received the commercial H7 vaccine did not react to the NS1 protein (Table 2). In contrast, 100% of infected birds tested positive for NS1 antibody when their sera were tested at a 1:200 dilution. For the MBP-NS1-coated ELISA plates, OD values greater than or equal to 0.177 (the cutoff value) were considered positive. Individual serum samples (five per group) were also run at the dilutions indicated above and confirmed the results for the pooled sera. These results demonstrate that by diluting immune sera 1:200, a specific NS1 ELISA can provide an antigenic marker for influenza virus infections among poultry.
Application of NS1 ELISA for diagnostic testing of field samples. Field validation of the NS1 ELISA method was undertaken by testing immune sera from domestic poultry that had been vaccinated or infected with AI virus. Sera from 20 H1N1-vaccinated turkeys along with matching preimmune samples collected before vaccination were tested for AGP and NS1 antibodies. The commercial influenza H1N1 vaccine was a killed autogenous preparation that uses a 2001 North Carolina H1N1 subtype virus of swine or turkey origin. For comparison, sera from five experimentally H1N1-infected MWT turkeys collected 4 weeks after i.n. infection were also tested. Sera from 20 turkeys that had been vaccinated two to three times and that had been found to be AGP positive reacted negatively in the MBP-NS1 ELISA (Table 3) and the GST-NS1 ELISA (data not shown). In contrast, NS1 antibody could be measured in diluted sera from all five turkeys that received live H1N1 virus. Positive OD values for the sera from H1N1-infected turkeys ranged from 0.322 to 0.569 and from 0.237 to 0.489 on ELISA plates coated with the MBP-NS1 and the GST-NS1 fusion proteins, respectively.
Field validation of the NS1 ELISA was also performed with sera from 25 H7N2-infected red layers that had clinically recovered. This LPAI virus was isolated in April 2003 from a single layer operation located in Rhode Island. For comparison, birds in an experimental group of five WPR chickens were immunized three times with the commercial H7N2 vaccine and tested for AGP and NS1 antibodies. Sera from these birds were collected 2 weeks after the final vaccine boost and were found to be positive for AGP antibodies but negative for antibodies to the NS1 protein when the sera were tested at a dilution of 1:200 (Table 3). In contrast, 19 of 25 serum samples from the H7N2-infected chickens, which were positive for antibody by the standard AGP test, were also positive for antibodies to the MPB-NS1 fusion protein. To determine if two synthetic NS1 peptides could serve as antigens capable of detecting antibodies in the sera of AI virus-infected poultry, peptides NS129-41 and NS136-48 were synthesized on the basis of the predicted antigenicity of NS1(B) consensus sequences. Chicken anti-H7 sera were reactive against the MBP-NS1 fusion protein and the NS136-48 peptide but reacted poorly with the NS129-41 peptide. Positive OD values for the sera from H7N2-infected turkeys ranged from 0.322 to 0.569, 0.106 to 0.338, and 0.08 to 0.141 on ELISA plates coated with the MBP-NS1 fusion protein, the NS136-48 peptide, and NS129-41 peptide, respectively.
In field situations, vaccinated chickens and turkeys may subsequently be exposed to live virus; and although infectious virus can be detected among vaccinated poultry (33, 35, 40), the low level of virus replication may result in a low NS1 antibody response. Therefore, the MBP-NS1 and peptide ELISAs were used to test sera from 20 MWT turkeys that were vaccinated two to three times with the commercially available H7N2 vaccine and subsequently challenged with 107 EID50s of TV/02 virus. As shown in Table 3, sera collected from all 20 turkeys 2 weeks after virus challenge were positive for AGP and NS1 antibodies. Turkey anti-H7 sera were reactive against the fusion protein and the NS136-48 peptide, but immune sera from infected poultry reacted poorly to the NS129-41 peptides, with only 4 of 20 field serum samples being positive by the peptide ELISA. Taken together, these results suggest that an ELISA for the measurement of antibodies against NS1 recombinant protein or an NS1 peptide may be useful for the diagnosis of AI virus infections among poultry flocks.
DISCUSSION
Because LPAI virus infections among chickens and turkeys are often asymptomatic, the diagnosis requires serologic monitoring. Most poultry diagnostic laboratories prefer the serologic AGP test because of its simplicity and broad specificity for the detection of type A influenza virus infections among poultry. AGP antibodies recognize conserved nucleoprotein (NP) and matrix (M1) influenza virus proteins present in the sera of poultry exposed to AI virus and therefore can detect multiple subtypes of influenza A virus (3). Other serologic tests for AI include an ELISA, which also measures antibody responses to conserved internal proteins, and the subtype-specific HI assay (21). However, these standard serologic tests for AI virus exposure do not differentiate between vaccinated and infected poultry when traditional vaccines are used (36). An improved serologic assay for AI virus infection would allow the identification of poultry that have been infected with AI virus, while it would correctly exclude animals that have been vaccinated with any subtype. The diagnostic test described in this study used a conserved nonstructural protein that is not associated with virions but that is expressed in large amounts in influenza virus-infected cells (17). Testing by sensitive immunoassays for antibodies against the NS1 antigen, we compared the NS1 antibody responses in vaccinated and infected poultry and found that live virus infection induced the highest levels of NS1 antibody compared with the amount induced by commercial vaccines, which induced little or no detectable antibody to NS1 protein. Following dilution of immune sera, the NS1 ELISA was an effective tool for the screening of poultry sera for evidence of AI virus infection.
The NS1 protein, although a weak antigen in comparison to other influenza virus proteins, is remarkably conserved among type A influenza viruses (5, 19, 30, 39). Previous studies demonstrated that the NS1 protein could be detected in the sera of horses experimentally infected with the H3 subtype of influenza virus but not in the immune sera of animals immunized with inactivated viruses (4, 22). The inactivated whole-virus equine vaccine used in the aforementioned studies was partially purified, thus removing residual NS1 protein that would be present in the infectious medium. This allowed the clear distinction between H3-vaccinated and H3-infected horses by using the NS1 protein as a differential marker (4, 22). In our study, antiserum from poultry that received commercial vaccines had some weak reactivity to the NS1 protein, as detected by the NS1 ELISA and Western blot analysis. This reactivity was not observed in poultry that received normal allantoic fluid emulsified in the same commercial adjuvant. Thus, it seems likely that the NS1 protein is present at low levels in the unpurified, infectious allantoic fluid of embryonated chicken eggs used for the production of commercial AI vaccines. AI virus replicates and causes the death of the endothelial cells lining the allantoic sac. This results in the release of the cell contents, which potentially could include some NS1 protein. However, we were unable to detect NS1 protein in infectious allantoic fluid by using Western blot analysis and a number of different (monoclonal and polyclonal) antibodies to NS1 protein (data not shown). In contrast to commercially vaccinated poultry, immune sera from birds that received gradient-purified vaccines had no detectable levels of NS1 antibody, even though these animals displayed a high-level antibody response to other influenza virus proteins (Fig. 1B and Table 1). Due to the lack of NS1 protein in purified virions, it was not surprising that these animals were unable to generate an antibody response to NS1 protein, even after three immunizations. We then sought to determine the serum dilutions that would allow the differentiation of vaccinated and infected poultry. At dilutions of 1:10 and 1:50, most sera from poultry that received the traditional commercial AI vaccine still reacted weakly to the NS1 protein by the NS1 ELISA. However, optimal discrimination between vaccinated and infected poultry was achieved with sera diluted 1:200, a dilution at which sera from infected poultry yielded positive ELISA reactions but sera from vaccinated poultry gave consistently negative results. Using this value, we were able to provide a test that identifies poultry as positive or negative for influenza virus infection.
Concerning the two different NS gene lineages described (19, 30, 39), the selection of an NS1 protein from NS1(A) or NS1(B) may have to be considered to optimize the reactivity of immune sera. However, there are four highly conserved regions among both groups (19), and this conservation would allow the NS1 ELISA to work for viruses found in the NS1(A) and NS1(B) groups. Although extensive testing of the antigenic cross-reactivities of the two different groups of NS genes was beyond the scope of the present study, the diagnostic specificity of the NS1 antibody response did not appear to be confined to the NS1(A) and the NS1(B) groups. Our data indicate that there are antigenic cross-reactions between the gene lineages because antisera from poultry infected with the NS1(A) or NS1(B) type of virus reacted equally well to the GST-NS1 and the MBP-NS1 fusion proteins (Table 1).
Confirmation of the anti-NS1 test result required verification in tests with field sera from animals that represented populations that might be targeted. Therefore, field sera from infected or vaccinated poultry, which possessed AGP antibodies, were chosen for testing; hence, it was unlikely that these poultry were not properly infected or vaccinated since they demonstrated AI virus exposure when their sera were tested by the standard serologic test. A second serologic test was performed with sera from experimental birds to demonstrate that vaccinated poultry responded to the H5 and H7 vaccines with increases in serum IgG antibody levels. The highest serum antibody titers to whole influenza virus detected by ELISA were observed in poultry sera collected after three vaccinations. Although each vaccine (commercial or purified) boost significantly enhanced the IgG antibody responses to whole influenza proteins, the specific NS1 antibody response never displayed an endpoint ELISA titer of greater than 100. Particular emphasis was placed on NS1 antibody responses to influenza H5 and H7 virus strains, because these are the subtypes most frequently found in AI outbreaks worldwide (34). We also found that the NS1 ELISA could be a diagnostic tool for the determination of H1N1 virus infection of poultry (Table 3). Avian influenza H1N1 virus has been a problem in U.S. commercial turkey flocks in areas with high densities of swine, and some states routinely vaccinate flocks against the H1N1 AI virus (41, 42).
The influenza virus NS1 proteins used in these studies were expressed in bacterial cells, and purification of the protein was carried out to remove the bulk of the bacterial proteins. Initial preparations of unpurified NS1 proteins expressed in E. coli resulted in high ELISA background levels (OD value range, 0.235 to 0.588) in sera from older (age, >4 weeks) uninfected chickens and turkeys (data not shown). This nonspecific reactivity to unpurified E. coli-derived NS1 proteins may be due to cross-reactive antibodies to common bacterial antigens, particularly those present in the serum of older poultry. To overcome this problem, the MBP-NS1 and GST-NS1 purified expression systems were used to remove the bulk of the E. coli proteins, which resulted in significant reductions (OD value range, 0.09 to 0.149) in the reactivity of uninfected poultry sera to bacterial proteins. In addition, serum samples were diluted with 0.5% normal E. coli cell lysate. To further reduce the background cross-reactivity to negligible levels and to avoid the addition of E. coli cell lysate, we tested the immune reactivity of N-terminal NS1-specific peptides synthesized on the basis of the predicted antigenicity of consensus sequences. The two different peptides used in these studies encompass a 13-amino-acid region within the NS1 protein. Our data demonstrated that the NS129-41 peptide was weakly antigenic compared with the antigenicities of the MBP-NS1 fusion protein and the NS136-48 peptide. Although the NS136-48 peptide was less antigenic than the full-length MBP-NS1 fusion protein, the former was still able to detect anti-NS1 antibodies exclusively in the sera of poultry infected with influenza virus and gave consistently lower background levels by the NS1 ELISA. In regards to the diagnostic sensitivity of the NS1 antibody response, we found that it was at least equal to that of the AGP test. Thus, sera from infected poultry that possessed AGP antibodies also displayed NS1 antibodies, whereas the field chickens that did not mount an AGP antibody response also failed to show anti-NS1 antibodies (Table 3).
AI is a disease of economic importance and remains a continuous threat to commercial poultry operations. During the initial stages of an AI virus outbreak, the feasibility of using an AI vaccine would depend on a number of factors, including the ability to differentiate between vaccinated and infected poultry. A diagnostic tool that differentiates vaccinated from infected poultry will provide valuable information to state veterinarians and poultry industries when they are considering vaccination as an option to control LPAI. A recently developed test has allowed the differentiation of infected from vaccinated animals by using a heterologous neuraminidase (NA) subtype and an anti-NA antibody detection test (6). This strategy has provided some success but has limitations because of the availability of vaccine strains that require proper combinations of the HA and the NA subtypes. Thus, the HA subtype of the vaccine strain must have a high degree of homology to the HA subtype of the circulating strain and, in addition, must possess a different NA subtype; however, the use of the reverse genetics system would allow a more flexible strategy for the differentiation of infected from vaccinated poultry (18). The results from this study indicate that the use of a simple and specific ELISA for the serodiagnosis of infection with multiple LPAI virus subtypes can distinguish infected poultry from vaccinated poultry. This diagnostic test may lay the groundwork for additional strategies that will yield optimal results. One such strategy is the construction of recombinant influenza viruses with a truncated or deleted NS1 protein that could be used for the generation of influenza vaccines (38). Such recombinant vaccines would lack NS1 epitopes and would provide a means for the clear distinction between vaccinated and virus-infected poultry.
ACKNOWLEDGMENTS
We thank Karen Burns from Lohmann Animal Health for providing vaccine material. We also thank Suzanne Deblois, Joan Beck, Marinda Thomas, Laura Kelly, and Rashida Horton for technical assistance and Roger Brock for animal care assistance. We thank Chang-Won Lee and Claudia Chesley for critical review of the manuscript.
This research was supported by USDA, ARS, CRIS project 6612-32000-040-00D.
REFERENCES
Alexander, D. J. 2000. A review of avian influenza in different bird species. Vet. Microbiol. 74:3-13.
Beard, C. W. 1970. Avian influenza antibody detection by immunodiffusion. Avian Dis. 14:337-341.
Beard, C. W. 1989. Serological procedures, p. 192-200. In H. Friedman (ed.) A laboratory manual for the isolation and identification of avian pathotypes. Kendall-Hunt Publishing, Dubuque, Iowa.
Birch-Machin, I., A. Rowan, J. Pick, J. Mumford, and M. Binns. 1997. Expression of the nonstructural protein NS1 of equine influenza A virus: detection of anti-NS1 antibody in post infection equine sera. J. Virol. Methods 65:255-263.
Buonagurio, D. A., S. Nakada, J. D. Parvin, M. Krystal, P. Palese, and W. M. Fitch. 1986. Evolution of human influenza A viruses over 50 years: rapid, uniform rate of change in NS gene. Science 232:980-982.
Capua, I., C. Terregino, G. Cattoli, F. Mutinelli, and J. F. Rodriguez. 2003. Development of a DIVA (differentiating infected from vaccinated animals) strategy using a vaccine containing a heterologous neuraminidase for the control of avian influenza. Avian Pathol. 32:47-55.
Condoberry, P. K., and R. D. Slemons. 1992. Biological properties of waterfowl-origin type A influenza viruses in chickens. Avian Dis. 36:17-23.
Cox, N. J., and A. P. Kendal. 1984. Genetic stability of A/Ann/6/60 cold-mutant (temperature-sensitive) live influenza virus genes; analysis by oligonucleotide mapping of recombinant vaccine strains before and after replication in volunteers. J. Infect. Dis. 149:194-200.
Donelan, N. R., C. F. Basler, and A. García-Sastre. 2003. A recombinant influenza A virus expressing an RNA-binding-defective NS1 protein induces high levels of beta interferon and is attenuated in mice. J. Virol. 77:13257-13266.
Easterday B. C., V. S. Hinshaw and D. A. Halvorson. 1997. Influenza, p. 583-605. In B. W. Calnek, H. J. Barnes, C. W. Beard, L. R. McDougald, and Y. M. Saif (ed.), Diseases of poultry. Iowa State University Press, Ames.
Garcia, A., J. Hector, D. K. Srivastava, D. A. Jayawardene, D. R. Wehr, and R. G. Webster. 1998. Efficacy of inactivated H5N2 influenza vaccines against lethal A/Chicken/Queretaro/19/95 infection. Avian Dis. 42:248-256.
García-Sastre, A., A. Egorov, D. Matassov, S. Brandt, D. E. Levy, J. E. Durbin, P. Palese, and T. Muster. 1998. Influenza A virus lacking the NS1 gene replicates in interferon-deficient systems. Virology 252:324-330.
Halvorson, D. A., C. J. Kelleher, and D. A. Senne. 1985. Epizootiology of avian influenza: effect of season on incidence in sentinel ducks and domestic turkeys in Minnesota. Appl. Environ. Microbiol. 49:914-919.
Halvorson, D. A. 2002. The control of H5 or H7 mildly pathogenic avian influenza—a role for inactivated vaccine. Avian Pathol. 31:5-12.
Inoue, Y., R. Suzuki, Y. Matsuura, S. Harada, J. Chiba, Y. Watanabe, I. Saito, and T. Miyamura. 1992. Expression of the amino-terminal half of the NS1 region of the hepatitis C virus genome and detection of an antibody to the expressed protein in patients with liver disease. J. Gen. Virol. 73:2151-2154.
Kawaoka, Y., C. W. Naeve, and R. G. Webster. 1984. Is virulence of H5N2 influenza viruses in chickens associated with loss of carbohydrate from the hemagglutinin Virology 139:303-316.
Krug, R. M., and P. R. Etkind. 1973. Cytoplasmic and nuclear virus-specific proteins in influenza virus-infected MDCK cells. Virology 56:334-348.
Lee, C.-W., D. A. Senne, and D. L. Suarez. 2004. Generation of reassortant influenza vaccines by reverse genetics that allows utilization of a DIVA (differentiating infected from vaccinated animals) strategy for the control of avian influenza. Vaccine 22:3175-3181.
Ludwig, S., U. Schultz, J. Mandler, W. M. Fitch, and C. Scholtissek. 1991. 1991. Phylogenic relationship of the nonstructural (NS) genes of influenza A viruses. Virology 183:566-577.
Niezert, E., E. Beck, P. A. De Mello, I. Gomes, and I. E. Bergmann. 1991. Expression of the aphtovirus RNA polymerase gene in Escherichia coli and its use with other bioengineered nonstructural antigens in detection of late persistent infections. Virology 184:799-804.
Office International des Epizooties. 1996. Highly pathogenic avian influenza (fowl plague), p. 155-160. In Manual of standards for diagnostic tests and vaccines, vol. 3. Office International des Epizooties, Paris, France.
Ozaki, H., T. Sugiura, S. Sugita, H. Imagawa, and H. Kida. 2001. Detection of antibodies to the nonstructural protein (NS1) of influenza A virus allows distinction between vaccinated and infected horses. Vet. Microbiol. 82:111-119.
Panigrahy, B., D. A. Senne, and J. C. Pedersen. 2002. Avian influenza virus subtypes inside and outside the live bird markets, 1993-2000; a spatial and temporal relationship. Avian Dis. 46:298-307.
Perdue, M. L., M. Garcia, D. Senne, and M. Fraire. 1997. Virulence-associated sequence duplication at the hemagglutinin cleavage site of avian influenza viruses. Virus Res. 49:173-186.
Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty per cent endpoints. Am. J. Hyg. 27:493-497.
Slemons, R. D., J. S. Osborn, and F. Hayes. 1974. Viruses isolated from wild free-flying ducks in California. Avian Dis. 18:119-124.
Slemons, R. D., and B. C. Easterday. 1977. Type-A influenza viruses in the feces of migratory waterfowl. J. Am. Vet. Med. Assoc. 171:947-948.
Spackman, E., D. A. Senne, S. Davison, and D. L. Suarez. 2003. Sequence analysis of recent H7 avian influenza viruses associated with three different outbreaks in commercial poultry in the United States. J. Virol. 77:13399-13402.
Stallknecht D. E. 1998. Ecology and epidemiology of avian influenza viruses in wild bird populations: waterfowl, shorebirds, pelicans, cormorants, etc., p. 61-69. In D. E. Swayne and R. D. Slemons (ed.), Proceedings of the 4th International Symposium on Avian Influenza. U.S. Animal Health Association, Richmond, Va.
Suarez, D. L., and M. L. Perdue. 1999. Multiple alignment comparison of the non-structural genes of influenza A viruses. Virus Res. 54:59-69.
Suarez, D. L., M. Garcia, J. Latimer, D. A. Senne, and M. L. Perdue. 1999. Phylogenetic analysis of H7 avian influenza viruses isolated from live bird markets of the Northeast United States. J. Virol. 73:3567-3573.
Suarez, D. L. 2000. Evolution of avian influenza viruses. Vet. Microbiol. 74:15-27.
Swayne, D. E., J. R. Beck, M. Garcia, and H. D. Stone. 1999. Influence of virus strain and antigen mass on efficacy of H5 avian influenza inactivated vaccines. Avian Pathol. 28:245-255.
Swayne, D. E., and D. L. Suarez. 2000. Highly pathogenic avian influenza. Rev. Sci. Tech. Off. Int. Epizoot. 19:463-482.
Swayne, D. E., J. R. Beck, M. L. Perdue, and C. W. Beard. 2001. Efficacy of vaccines in chickens against highly pathogenic Hong Kong H5N1 avian influenza. Avian Dis. 45:355-365.
Swayne, D. E. 2003. Vaccines for list A poultry diseases: emphasis on avian influenza. Dev. Biologics (Basel) 114:201-212.
Swayne, D. E., and D. A. Halvorson. 2003. Influenza, p. 135-160. In Y. M. Saif, H. J. Barns, J. R. Glisson, A. M. Fadly, L. R. McDougald, and D. E. Swayne (ed.), Diseases of poultry. Iowa State University Press, Ames.
Talon, J., M. Salvatore, R. E. O'Neill, Y. Nakaya, H. Zheng, T. Muster, A. García-Sastre, and P. Palese. 2000. Influenza A and B viruses expressing altered NS1 proteins: a vaccine approach. Proc. Natl. Acad. Sci. USA 97:4309-4314.
Treanor, J. J., M. H. Synder, W. T. London, and B. R. Murphy. 1989. The B allele of the NS gene of avian influenza viruses, but not the A allele, attenuate a human influenza A virus for squirrel monkeys. Virology 171:1-9.
Tumpey, T. M., D. R. Kapczynski, and D. E. Swayne. 2004. Comparative susceptibility of chickens and turkeys to avian influenza A H7N2 virus infection and protective efficacy of a commercial avian influenza H7N2 virus vaccine. Avian Dis. 48:167-176.
U.S. Animal Health Association. 2002. Report of the Committee on Transmissible Diseases of Poultry and Other Avian Species. Criteria for determining that an AI virus isolation causing an outbreak must be considered for eradication, p. 524-586. In Proceedings of the 106th Annual Meeting of the U.S. Animal Health Association. U.S. Animal Health Association, Richmond, Va.
U.S. Animal Health Association. 2003. Report of the Committee on Transmissible Diseases of Poultry and Other Avian Species. Avian Influenza Workshop—Recent Outbreaks and Developments, p. 312-337. In Proceedings of the 107th Annual Meeting of the U.S. Animal Health Association. U.S. Animal Health Association, Richmond, Va.
Webster, R. G., W. J. Bean, O. T. Gorman, T. M. Chambers, and Y. Kawaoka. 1992. Evolution and ecology of influenza A viruses. Microbiol. Rev. 56:152-179.
Webster, R. G., and Y. Kawaoka. 1994. Influenza: an emerging and reemerging disease. Semin. Virol. 5:103-111.(Terrence M. Tumpey, Rene )