Glycoprotein-Based Enzyme-Linked Immunosorbent Assays for Serodiagnosis of Infectious Laryngotracheitis

TEXT

Infectious laryngotracheitis virus (ILTV), an alphaherpesvirus, causes an acute respiratory disease in chickens known as infectious laryngotracheitis (ILT) that leads to significant economic losses to the poultry industry worldwide (1, 2). Currently, live attenuated vaccines are used to control ILTV infections. However, these vaccines are not satisfactory since the vaccine virus can revert to virulence after bird-to-bird passage (3) and can induce latent infections (4). Due to these limitations, live virus-vectored vaccines expressing the surface glycoproteins of ILTV have been developed as a safer alternative to attenuated live vaccines. A recombinant herpesvirus of turkey (HVT-LT) expressing ILTV glycoproteins D (gD) and I (gI) and Fowl pox virus (FPV-LT) expressing ILTV glycoprotein B (gB) and UL-34 genes are commercially used in the United States (5, 6). Recently, we have evaluated the role of ILTV gB, gC, and gD in immunogenicity and protection against a virulent ILTV challenge using recombinant Newcastle disease virus (rNDV) as a vaccine vector. Our results indicated that rNDV expressing ILTV gD provided complete protection against a virulent ILTV challenge in chickens (7). A vectored vaccine against ILTV infection will be safe without reversion of vaccine virus to be virulent or establishment of latency and also allow differentiation of vaccinated birds from the infected birds.

The detection of the humoral immune response is critical for the rapid identification of ILTV-infected birds (5). The enzyme-linked immunosorbent assay (ELISA) has been used for the detection of the humoral immune response. Despite its simplicity and rapidity, the commercially available ELISA using whole virus as an antigen is inefficient for detecting seroconversion with virus-vectored vaccines (6). Recently, individual ILTV surface glycoproteins have been used for ELISAs to detect ILTV antibodies in sera from vaccinated birds with attenuated and vectored vaccines against ILT (8–10). However, the specific glycoprotein-based ELISA has not been commercially available for rapid detection of seroconversion by vectored vaccines against ILTV. Therefore, in the present study, we developed rapid diagnostic ELISAs by using ILTV gB, gC, and gD (B-, C-, and D-ELISAs, respectively) as antigens, since these glycoproteins can induce humoral and cell-mediated immune responses against ILTV and other herpesviruses (7, 11–14). Each glycoprotein was eukaryotically expressed in insect cells. The diagnostic potential of these ELISAs was assayed with sera collected from chickens vaccinated with various virus-vectored vaccines. Furthermore, the efficacy of our ELISAs was validated by testing field serum samples and compared to that of a commercially available ELISA as a reference (fowl laryngotracheitis virus antibody test kit; Zoetis, San Diego, CA) (15–17).

The ILTV gB, gC, and gD genes were cloned into the pCR 4 TOPO vector (Invitrogen) as described previously (7), and these genes were amplified from the TOPO vectors with concurrent introduction of a C-terminal His₆ tag and the NotI cloning site at their reverse primers and the EcoRI cloning site at their forward primers (the His₆ tag sequence is underlined and the cloning sites are italicized in the following primer sequences). The forward and reversed primers used were 5′-GATCGAATTCATGCAATCCTACATCGCCGT-3′ and 5′-GATCGCGGCCGCTTAGTGATGGTGATGGTGATGTTCGTCTTCGCTTTCTTCT-3′ for gB, 5′-GATCGAATTCATGCAGCATCAGAGTACTGC-3′ and 5′-GATCGCGGCCGCTTAGTGATGGTGATGGTGATGTGTTGTCTTCCAGCACCAT-3′ for gC, and 5′-GATCGAATTCATGGACCGCCATTTATTTTTGAGG-3′ and 5′-GATCGCGGCCGCTTAGTGATGGTGATGGTGATGGCTACGCGCGCATTTTACG-3′ for gD. The amplified products were sequenced, digested with restriction enzymes, and individually cloned into a pFastBac1 vector (Invitrogen, Carlsbad, CA). The recombinant bacmid DNAs and pFastBac DNA (control) were used to transfect Sf9 cells according to the manufacturer's instructions (Bac-to-Bac baculovirus expression system; Invitrogen).

The gB and gC fusion proteins in the cell lysates and the gD fusion protein in the cell lysate and supernatant were purified using His Trap HP (GE Healthcare). Purified glycoproteins were confirmed by Western blotting using rabbit anti-ILTV gB, gC, and gD antisera (5). We detected two bands representing the uncleaved monomeric precursor form (>100 kDa) and C-terminal cleavage product (58 kDa) of gB, the 60 -kDa band of gC, and the 42-kDa band of gD, as shown with ILTV gB, gC, and gD in cell lysates infected with the ILTV virus (Fig. 1).

• Open in new tab
• Download powerpoint

FIG 1

Western blot analysis of the recombinant ILTV antigens gB, gC, and gD. Western blot analysis was performed on purified gB (A), gC (B), and gD (C) using rabbit anti-ILTV gB, gC, and gD antisera. CEK, chicken embryo kidney.

We first optimized the concentrations (producing the highest optical density [OD] and positive/negative ratio) of gB, gC, and gD to coat the ELISA plates and the dilution of the test sera by checkerboard titration using ILTV-positive and -negative sera. The optimal concentrations were 200 ng/well for gB or gD and 500 ng/well for gC. The optimum dilutions of the test sera and the conjugate were found to be 1:100. Thus, in our standardized ELISA procedure, ELISA plates were coated with the optimum concentration of each antigen in sodium carbonate-bicarbonate buffer (pH 9.8) at 4°C overnight. The plates were blocked first with 3% skim milk in water for 30 s followed by 2% sucrose in water for 30 s and subsequently dried for 2 h at 37°C. The assay was conducted following the instructions for the commercial ELISA (Zoetis). Briefly, chicken sera (1:100 dilution) followed by goat anti-chicken IgG(H+L) peroxidase conjugate (1:100 dilution) were added into the wells for 30 min each at room temperature and extensively washed with ELISA washing buffer. The reaction color was developed by adding an 2,2′-azinobis(3-ethylbenzthiazolinesulfonic acid) (ABTS)-hydrogen peroxide substrate solution for 15 min at room temperature and stopped by addition of a peroxidase stop solution. The OD of each sample was read at 405 nm using an ELx800 ELISA plate reader (BioTek, Winooski, VT), corrected using both positive and negative samples in each plate, and expressed as the sample-to-positive ratio (Sp) [Sp = (sample absorbance − average negative-control absorbance)/(average positive-control absorbance − average negative-control absorbance)]. Our result showed that the average Sp values of the ILTV-negative sera (mean ± standard deviation) were 0.049 ± 0.025, 0.085 ± 0.02, and 0.045 ± 0.022 by the B-, C-, and D-ELISAs, respectively. Therefore, ILTV-positive sera (1:100 dilution) were defined at an Sp value of >0.1 for the B- and D-ELISAs and an Sp value of >0.125 for the C-ELISA in our standardized test.

To evaluate the efficacy of the B-, C-, and D-ELISAs, we tested the sera collected from 45 chickens after the ILTV challenge and the negative sera of 45 specific pathogen-free chickens prior to vaccination in our previous study (7). All of the test sera collected from the ILTV-challenged chickens were positive with average Sp values of 0.33 ± 0.24, 0.58 ± 0.32, and 0.91 ± 0.91 by the B-, C-, and D-ELISAs, respectively. In addition, all 45 ILTV-negative sera were confirmed as negative by all ELISAs. We further evaluated the specificity of our ELISAs by determining their reactivity with hyperimmune sera distinct for other well-known chicken pathogens (Mycoplasma gallisepticum, Mycoplasma synoviae, avian influenza virus, infectious bronchitis virus, and Newcastle disease virus [NDV]). In general, the B- and D-ELISAs did not react to these antisera, thus confirming 100% specificity (n = 10). However, the C-ELISA showed cross-reactivity to antisera raised against Mycoplasma gallisepticum, NDV, and avian influenza virus, showing only 40% specificity (n = 10). This suggests that gC is unsuitable to be used for an ELISA, at least in the current form. Therefore, only the B- and D-ELISAs were further evaluated in our subsequent experiments.

We next compared the efficacies of the B- and D-ELISAs to that of commercial ELISAs for detecting anti-ILTV antibodies in sera from chickens immunized with various virus-vectored and attenuated vaccines against ILTV, such as rNDV vectored ILTV gB and gD, FPV-LT, and HVT-LT (Table 1). The percentage of agreement between the B- and D-ELISAs and the commercial ELISA was calculated as the portion of samples with similar results by the two tests out of total number of samples tested. Our results indicated that the B-ELISA and commercial ELISA showed similar detection rates for seroconversion resulting from the vaccination of birds with FPV-LT and rNDV expressing ILTV gB. More importantly, the D-ELISA alone had detection rates superior to that of the commercial ELISA for detecting seroconversion with the rNDV gD vaccine, indicating a potential for gD in diagnostic applications. For specific detection of gB and gD, the efficacies of the B- and D-ELISAs were cross-confirmed with that of Western blotting (Fig. 2A, lanes 1 and 2, and B, lane 1). The results obtained with the D-ELISA against sera from groups rNDV gD prechallenge and HVT-LT prechallenge and from the B-ELISA against sera from group rNDV gB+ gD prechallenge were in agreement with the results of our previous study (5) where all of the birds immunized with rNDV gD vaccine possessed neutralizing antibodies (NAb) against ILTV. However, the HVT-LT sera lacked the NAb, indicating that the HVT-LT vaccine induced a cell-mediated immune response against ILTV rather than the humoral response, and poor immunogenicity of the rNDV gB vaccine, which presumably might be the reason for the absence of ELISA antibodies in rNDV gB + gD multivalent vaccinated birds, respectively. As expected, seroconversion to ILTV was detected in all of the sera collected postchallenge with ILTV. As shown in Table 1, the B- and D-ELISAs showed 100% sensitivity compared to that of the commercial ELISA (n = 32 for postchallenged serum samples). Further, the diagnostic inabilities of the B- and D-ELISAs compared to that of commercial ELISA for detect ILTV antibodies in prechallenge sera from birds vaccinated with whole virus attenuated chicken embryo origin (CEO) Trachivax vaccine imply that the vaccine did not direct the humoral immune response against gB and gD. Alternatively, the purified gB and gD used in the B- and D-ELISAs may not be suitable in the current form to detect seroconversion with attenuated ILTV vaccine.

• Open in new tab
• Download powerpoint

FIG 2

Western blot analysis of chicken sera that were positive by the B- and D-ELISAs but negative by a routine commercial whole-antigen-based ELISA. (A) Reactivity with purified gB. Eleven positive samples by the B-ELISA were further confirmed by Western blotting (lanes 3 to 11). The FPV-LT prechallenge sera (lane 1), the rNDV gB postchallenge sera (lane 2), and the rabbit anti-ILTV gB sera (lane 14) were used as positive controls. All positive samples reacted with the 58-kDa band of gB. (B) Reactivity with purified gD. Six out of nine samples that were positive by the D-ELISA but negative by the commercial ELISA reacted with gD (lanes 2, 3, 5, 7, 9, and 10). The rNDV gD prechallenge (lane 1) and rabbit anti-ILTV gD (lane 11) sera were used as positive controls. All positive samples reacted with the 42-kDa band of gD.

We further compared the efficacy of the B- and D-ELISAs to that of the commercial ELISA by analyzing 105 clinical serum samples (Table 2). These serum samples were collected from sick chickens with ILTV infection at the Georgia Poultry Lab (GA, USA). A total of 102 (97.15%) samples were positive by the B-ELISA, whereas only 35 (33.33%) and 57 (54.28%) samples were positive by the D-ELISA and commercial ELISA, respectively. Forty-five and nine samples were positive by the B- and D-ELISAs, respectively, but negative by the commercial ELISA. To determine whether these samples had given false-positive results by the B- and D-ELISAs, Western blot analysis was performed using partially purified recombinant gB and gD. All 45 positive samples by the B-ELISA were confirmed as positive, and 6 out of 9 positive samples by the D-ELISA reacted with their respective antigens (Fig. 2). From the Western blot results, we concluded that the detection rates of the commercial and D-ELISAs were only 55.8% and 34.3%, respectively, compared to that of the B-ELISA. The concordances between the B-ELISA and commercial ELISA, the D-ELISA and commercial ELISA, and the B-ELISA and D-ELISA were 57.14%, 61.9%, and 31.37%, respectively.

In summary, we have shown that the recombinant ILTV antigens gB and gD are valuable diagnostic reagents for the detection of ILTV infection in chickens. We further demonstrated that the D- and B-ELISAs showed enhanced detection of the seroconversion by the rNDV gD vaccine and the serodetection of ILTV infection in chickens, respectively, compared to that of a commercial ELISA. Therefore, a mixture of the gB and gD recombinant antigens might improve the ILTV diagnostic tests. This can also make it feasible to differentiate the birds vaccinated with the rNDV gD vaccine from naturally infected birds. Furthermore, our ELISAs can easily be set up and standardized for rapid diagnostic tests of serum samples, which can also be beneficial for control and seroepidemiological studies of ILTV infection in chickens.