Classification of infectious bursal disease virus into genogroups

Bursas from domestic and international flocks suspected of having IBDV were collected during the years 2013 to 2017. Bursas were cut in half, and the cut face was pressed onto Whatman FTA cards (GE Healthcare Life Sciences, Pittsburgh, PA). Foreign samples were imported into our laboratory under import permit #44226 from the U.S. Department of Agriculture Animal and Plant Health Inspection Service (Riverdale, Maryland, USA). During this study, we examined 90 samples from Algeria, Colombia, Ecuador, Egypt, Fiji, France, Guatemala, India, Indonesia, Iraq, Jordan, Kazakhstan, the Kingdom of Saudi Arabia (KSA), Kuwait, Malaysia, Mexico, Morocco, Philippines, Russia, United Arab Emirates (UAE), the United Kingdom (UK), the United States, and Vietnam.

Viral RNA was extracted from bursa samples on FTA cards as follows: Eight FTA card punches per sample were vortexed with 300 µl of RNase-free TE buffer pH 8.0 (Invitrogen, Carlsbad, CA) containing 1 mg of proteinase K (Invitrogen) per ml and 0.5% SDS (Sigma-Aldrich, St. Louis, MO) and incubated at 56 °C for 60 min. A 300-µl volume of RNA Lysis Buffer (Zymo Research, Irvine, CA) was then added and the sample was vortexed and incubated at room temperature for 9 min. RNA was purified using a Quick-RNA ^™  MiniPrep (Zymo Research) according to the manufacturer’s instructions. RNA was then precipitated from samples with 0.1 volume of 3 M sodium acetate buffer solution (Sigma-Aldrich) and 2.5 volumes of ethanol (Sigma-Aldrich) and resuspended in 25 µl of 90% dimethyl sulfoxide (Sigma-Aldrich).

RT-PCR was conducted using an AgPath-ID™ One-Step RT-PCR Reagents Kit (Applied Biosystems). Primers 743-F (5´-GCCCAGAGTCTACACCAT-3´) and 1331-R (5´-ATGGCTCCTGGGTCAAATCG-3´) were used to amplify a 579-bp fragment of the hypervariable region of the VP2 gene (hvVP2). RT was performed at 48 °C for 30 min and terminated by incubation at 95 °C for 10 min, and this was followed by 35 cycles of PCR at 95 °C for 30 s, 57 °C for 1.5 min and 72 °C for 1.5 min. A final extension at 72 °C for 5 min followed the PCR. Primers B-Univ-F (5’- AATGAGGAGTATGAGACCGA-3’) and B-Univ-R (5’-CCTTCTCTAGGTCAATTGAGTACC-3’) [18] were used to amplify a 1051-bp fragment (nt 319-1369) of the VP1 gene. RT was performed at 45 °C for 30 min, followed by 35 cycles of PCR at 95 °C for 30 s, 58 °C for 1.5 min and 72 °C for 1.5 min. A final extension at 72 °C for 7 min followed the PCR.

The RT-PCR products were prepared for sequencing using a Wizard SV Gel and PCR Clean-Up System (Promega, Madison, WI). Cycle sequencing was conducted at the University of Wisconsin Biotechnology Center DNA Sequencing Facility (Madison, WI). DNA sequences were submitted to GenBank (accession numbers MF142502-MF142591, PopSet 1229404744 for VP2 and MF142461-MF142501, PopSet 1229404662 for VP1). Nucleotide and predicted amino acid sequences were aligned with those of reference strains from GenBank (Table 2) using Geneious® 8.1.8 [19]. The evolutionary history was inferred using the neighbor-joining method [20] with 1000 bootstrap replicates [21]. The evolutionary distances were computed using the maximum composite likelihood method (MCL) [22] for nucleotide sequences and the Poisson correction method [23] for deduced amino acid sequences. Molecular phylogenetic analysis was also performed using the maximum-likelihood method based on the Kimura 2-parameter model [24]. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the MCL approach, and then selecting the topology with superior log likelihood value. A discrete gamma distribution was used to model evolutionary rate differences among sites (five categories (+G, parameter = 1.2420)). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 25.14% sites). All positions containing gaps and missing data were eliminated. Evolutionary analysis was conducted in MEGA7 [25].

The hvVP2 sequences were aligned, and a phylogenetic tree was constructed using the neighbor joining method (Fig. 1). The isolates clustered into seven major genogroups, which generally corresponded to their serotype or pathotype classification (Table 2). Genogroup 1 (generally classical viruses) were found worldwide, genogroup 2 (primarily antigenic variant viruses) are still predominately distributed in the Americas, and genogroup 3 (vvIBDV pathotype and vvIBDV reassortants) were found worldwide but most often identified outside North America. Some viral isolates, however, did not clearly fit into any of the three major genogroups and were classified separately. A genogroup 4 sample from United Arab Emirates (741_UAE) was most closely related to the distinct IBDV lineage that is prevalent in South America [26]. Genogroup 5 strains consisted of viruses from Mexico that were postulated by Jackwood [27] to be the result of a recombination between classical and variant viruses. These viruses have variant-type amino acid sequences in the P_BC loop, whereas the P_FG loop is more similar to the classical viruses. The P_DE loop sequence is also similar to classical viruses in that it has 249Q; however, it also has N251 and N254, a pattern we have also seen in variant strains from Guatemala and the southern US (data not shown). The P_HI loop of these viruses shows two unique substitutions, S317K and A321P. Genogroup 6 consisted of samples from the Kingdom of Saudi Arabia (751_KSA and 772_KSA) that did not have identical matches in GenBank but showed 92.26-93.64% identity to the ITA genotype observed in Italy (characterized by 220H, 222Q, 253E, 254S and 321V) [28] and 94.02-95.40% identity to isolate IBDVRF-5/94 from Russia [29]. Genogroup 7 was composed of viruses from Australia as well as a sample from Russia (429_Russia). A phylogenetic tree of the isolates was also constructed using the maximum-likelihood method and presented similar branching (Online Resource 1). The deduced amino acid sequences of representatives of each genogroup were also determined and are aligned in Online Resource 2.

While the majority of our genogroup 3 isolates contained the conserved residues that define the vvIBDV pathotype (A222, I242, I256, and I294), we noted some genetic heterogeneity among the strains grouping with vvIBDV reference strains HK46, Henan and UK661 in this study. Therefore, a phylogenetic tree of nucleic acid sequences of genogroup 3 samples and reference strains was prepared and is shown in Figure 2a. Several distinct branches contained viruses from specific geographic regions such as Egypt, India and Malaysia/Indonesia. Viruses isolated from Russia were found to form three distinct branches (Fig. 2a, groups 3-1, 3-2, and 3-3).

In order to determine if the nucleic acid substitutions were altering VP2 amino acid composition and possibly affecting the antigenicity of the genogroup 3 strains, deduced amino acid sequences were determined, and a phylogenetic tree was created (Fig. 2b). While many samples were closely related to UK661, distinct groups of viruses from Egypt, Malaysia/Indonesia, and the 3-1 to 3-3 groups from Russia were again evident. The specific amino acid changes in the hvVP2 region can be seen in Online Resource 3. Of the differences noted, we focused on ones that occurred on the tips of the projection domain loops of VP2. Significant amino acid differences noted in the P_BC loop include Y220F in Egyptian samples, A222T in 866_Malaysia and Group 3-2 of the Russian samples, and A222S in 616_Indonesia and 767_Indonesia. In the P_DE loop, G254D (found in Egyptian samples and groups 3-2 and 3-3 from Russia) or G254S (in the Egyptian samples) were identified. Additionally, samples from Malaysia and Indonesia also had amino acid changes in the P_HI loop, including A321E and, in 739_Malaysia and 866_Malaysia, S317R and D323N. A T359K substitution was found in samples in genogroups 2, 4, 5 and 6 but in only one sample (866_Malaysia) in genogroup 3. This 866_Malaysia virus also had an A222T substitution, but otherwise it was nearly identical to 739_Malaysia. Other substitutions were noted outside the loops. For example, a number of samples, including every Russian sample tested, were found to contain the mutation D279N. Additionally, groups 3-2 and 3-3 of the Russian samples were found to have a pair of mutations: S328K and S332N. This was not observed in any other samples outside Russia, although a sample from Kazakhstan was found which had only the S332N mutation.

The nucleotide sequences of a portion of the VP1 gene were determined for the genogroup 3 strains, and a phylogenetic tree was created from the deduced amino acid sequences (Fig. 3). The samples were grouped on the basis of the presence or absence of the conserved TDN tripeptide (aa 145–147) found in the vvIBDV pathotype. Roughly 1/3 of the samples possessed this tripeptide and are highlighted in Figure 3. Although the TDN triplet was conserved in all of the vvIBDV sequences, other changes were noted, including V141I in Egyptian and Indonesian samples and D161A in Russian samples. The remaining non-vvIBDV VP1 gene sequences had D146E and N147G with some variation at the first position (T145N, T145D or T145S), indicating that these were likely reassortants between vvIBDV and non-vvIBDV strains. Some regional changes were also observed among the non-vvIBDV segment B sequences, including V141I, E143D and A163V in Indian samples, V141I in samples from Kuwait and Iraq, and T153A in some Russian samples.

As indicated above, the 866_Malaysia sample in genogroup 3 has an A222T substitution. This was the only strain we found with that mutation that also contained the TDN tripeptide in the VP1 gene, which is typical of the vvIBDV pathotype.