Genomic recombination between infectious laryngotracheitis vaccine strains occurs under a broad range of infection conditions in vitro and in ovo

Abstract

Gallid alphaherpesvirus 1 causes infectious laryngotracheitis (ILT) in farmed poultry worldwide. Intertypic recombination between vaccine strains of this virus has generated novel and virulent isolates in field conditions. In this study, in vitro and in ovo systems were co-infected and superinfected under different conditions with two genomically distinct and commonly used ILTV vaccines. The progeny virus populations were examined for the frequency and pattern of recombination events using multi-locus high-resolution melting curve analysis of polymerase chain reaction products. A varied level of recombination (0 to 58.9%) was detected, depending on the infection system (in ovo or in vitro), viral load, the composition of the inoculum mixture, and the timing and order of infection. Full genome analysis of selected recombinants with different in vitro phenotypes identified alterations in coding and non-coding regions. The ability of ILTV vaccines to maintain their capacity to recombine under such varied conditions highlights the significance of recombination in the evolution of this virus and demonstrates the capacity of ILTV vaccines to play a role in the emergence of recombinant viruses.

Fig 1. Growth kinetics of Serva (closed circles) and A20 (open circles) strains of ILTV in CEK cells.

(a) Entry kinetics of Serva and A20 in CEK cells. The mean and standard deviation from 3 independent experiments are shown. No significant difference between the two different viruses was detected at any time point (b) Multi-step growth curve of Serva and A20 in CEK cells infected with a MOI of 0.001. The mean and standard deviation of triplicate measurements performed in each of 3 independent experiments are shown; * indicates p < 0.05 (Student’s t-test). (c) Kinetics of plaque development by the Serva and A20 strains of ILTV in CEK cells. The mean and standard deviation of the plaque area (mm²) of 13 to 22 individual plaques at each time point are shown; * indicates p < 0.05 (Student’s t-test).

Fig 2. Summary of results from co-infection and superinfections in ovo and in CEK cultures.

Proportions of parental strains or recombinants in both the inoculum and progeny viruses are shown as red (Serva), green (A20), or yellow (recombinant) blocks. The isolates from co-infection/superinfection experiments on CEK cells were obtained from a combination of the cell lysate and the supernatant fluid (Cell + SNF) from the infected CEK monolayers. The isolates from the co-infection experiment in embryonated eggs were obtained from CAM homogenates or AF, which were collected separately. The overall titre of virus recovered from each infection condition is also shown.

Fig 3. Schematic view of SNP distribution, showing the composition of the progeny population collected from the CAM (left column = A, B, C, D and E) and AF (right column = F, G, H, I and J) of co-infected chicken embryos 72 hours after inoculation.

Five eggs were inoculated per group and the viruses were isolated from 3 or 4 eggs (pooled samples) that did not show any signs of embryonic death. The panels represent each ratio of infecting parent viruses (Serva:A20) as 1:1 (A, F), 1:2 (B, G), 2:1 (C, H), 1:10 (D, I) or 10:1 (E, J). The closed circles indicate Serva-like SNPs and open circles indicate A20-like SNPs. Each row represents a virus that was genotyped as A20, Serva, or a recombinant with a unique genotype code (1 to 62) for each genotype pattern code.

Fig 4. The composition of the progeny population collected from CEK cells co-infected with A20 and Serva ILTV at a total MOI of 10.

The distribution of Serva-like (closed circles) and A20-like (open circles) SNPs in each isolate is shown. The isolates were collected from (A) 1:1 co-infection, (B) 1:4 (1 part Serva, 4 parts A20) co-infection, and (C) 4:1 (4 parts Serva, 1 part A20) co-infection.

Fig 5. Schematic view of SNP distribution showing the composition of the progeny population collected from CEK cells (A) co-infected at a MOI of 5 with Serva and A20, or (B-I) superinfected with either Serva or A20.

The left panels (B, C, D, E) show primary infection with Serva, while the right panels (F, G, H, I) show primary infection with A20. The incubation period between primary and secondary infection was 2 (B, F), 4 (C, G), 6 (D, H) or 8 (E, I) hours. The samples were collected at 72 hpi. The closed circles indicate Serva-like SNPs and open circles indicate A20-like SNPs. Each row represents a virus that was genotyped as A20, Serva, or a recombinant, with a unique genotype code number (1 to 62) for each genotype pattern code.

Fig 6. Number of isolates with each genotype pattern code (left y axis) and the origin of each isolate, together with the number of recombination breakpoints on the right y axis.

Isolates resulting from multiple recombination events were obtained less frequently. The negative correlation between the number of recombination events in a recombinant virus and the abundance of those recombinants in samples was statistically significant (Spearman’s correlation, r_s = -0.55, 95% confidence interval = -0.7579 to -0.2490, two-tailed p value = 0.0008).

Fig 7. Schematic view of the genomic composition of the isolates sequenced in this study (A) and sequence similarity plots (B) of isolate #29, #109, #138, #157, #237, #238, #319 and #358 with those of Serva and A20.

The plots were generated using SimPlot v3.5.1 set to scan the sequences with 6000 bp sliding windows, which were moved across the alignments in 200 bp steps.