To determine the role of LORF5 gene in virus replication, a DPV ΔLORF5 mutant was generated on the basis of DPV CHv-BAC, following the method of two-step Red recombination for constructing point mutations reported by Tischer et al. (2010; Figure 1). After transfection of the constructed infectious clone plasmids into DEFs, green fluorescent spots with matching cytopathic lesions were observed in DEF cells on the third day, and they could be passaged stably in new DEF cells, indicating that the LORF5 deletion mutant (CHv-BAC-ΔLORF5) was successfully rescued. Similarly, the revertant virus (CHv-BAC-RΔLORF5) was successfully rescued and harvested (Figure 2A). To confirm the abrogation of LORF5 gene, viral DNA was extracted for PCR analysis (Figure 2B) and positive bacterial artificial chromosome (BAC) clones were confirmed by restriction fragment length polymorphism (RFLP) analysis (Figure 2C) and Sanger sequencing. The plasmid digestion map showed that the bands (right) produced after ΔLORF5, RΔLORF5, and CHv-BAC were digested with XhoI or EcoRI were consistent with the expected bands (left) of the digestion map. Specifically, the corresponding LORF5 gene-deleted strain has a band around 4 kb after EcoRI digestion, and a band around 3 kb has been added (as shown by the arrows). There is no difference in the bands of ΔLORF5, RΔLORF5, and CHv-BAC after XhoI digestion. Then, we analyzed ΔLORF5, RΔLORF5, and CHv-infected DEF by quantitative real-time PCR (q-PCR) (Figure 2D), production of the LORF5 mRNA was completely abrogated in ΔLORF5-infected cells, and the transcription of neighboring genes UL55 and LORF4 was not affected. We have identified the extracellular virion protein content by mass spectrometry previously. The results showed that the LORF5 protein was present in mature extracellular virions. Five unique DPV LORF5 peptides were detected, while one unique peptide matched DPV gE (P < 0.05). The relative abundance of LORF5 was low based on the exponentially modified protein abundance index (emPAI) (Table 1).

To study whether LORF5 affects the production of mature DPV particles, we investigated if LORF5 plays a role in DPV replication through multistep growth kinetics. Growth kinetics revealed that abrogation of LORF5 slightly inhibits virus replication in vitro compared to the wild-type (WT, CHv-BAC) and revertant viruses (Figures 3A,B). Infectious particles were not detected at 12 hpi, which may be related to the latent infection characteristics of DPV. A rapid increase in virus particles was observed during the period of 24–48 hpi, indicating that the replication efficiency of the virus was the highest during this period, entering a slow stage at 48–96 hpi, but the virus titer reached a peak and entered a plateau. Next, we analyzed the difference of the data using GraphPad Prism version 8 (San Diego, CA, United States) and found that regardless of the supernatant (Figure 3C) or intracellular (Figure 3D) source, the viral titers of the LORF5-deleted mutant were lower than those of the parental virus from 24 hpi, especially during the 48–96-h period; the virus titer in the cell decreases more obviously with the difference of two asterisks (^∗∗P < 0.01), and the growth efficiency was restored to the level of the parental virus by reintroducing LORF5 in the revertant virus. We concluded from the results that abrogation of the LORF5 gene slightly but significantly impaired viral proliferation in vitro compared to the WT.

To explore how the LORF5 gene affects the proliferation of the virus in vitro, we tested the ability of the LORF5-deleted virus to adsorb cells, invade cells, replicate in the nucleus, and assemble and release virus particles and the infectivity of the virus to spread from cell to cell.

First, we investigated whether LORF5 acts on virus-adsorbed cells, and the results showed that the number of plaques of the three viruses was almost the same, and there was no significant difference (Figure 4A). On the other hand, the virus copies tested also obtained the same result (Figure 4B). It was concluded that the presence or absence of LORF5 has no effect on the adsorption of virus particles on the cell surface. Similarly, we tested the efficiency of the deletion strain invading the cell, and the results showed that there was no significant difference in the number of viral plaques in ΔLORF5, RΔLORF5, and CHv-BAC (Figure 4C), and the copies of the three viruses were roughly the same (Figure 4D), which was consistent with the results of Figures 4A,B, indicating that LORF5 is not related to DPV infection in cells.

Next, to explore the effect of LORF5 on DPV genome replication, ΔLORF5, RΔLORF5, and CHV-BAC [multiplicity of infection (MOI) = 1] were separately inoculated into DEF cells, and cell samples were collected to measure virus copies by q-PCR. The results demonstrated that the viral genome content increased over time; ΔLORF5, RΔLORF5, and CHv-BAC exhibited similar proliferation patterns during this process; and the genomic copies of the three viruses at each time node were almost equal (Figure 4E), explaining why LORF5 does not function in the replication stage of the viral genome.

In that way, will knocking out LORF5 affect the maturation of virus particles? Therefore, we observed the replication of ΔLORF5 in DEFs through transmission electron microscopy. In Figure 5A, a quantitative analysis of virions at different morphogenetic stages was performed using infected cells that were randomly observed under low-magnification electron microscopy. The data presented in Table 2 show the percentages of virions in different morphogenetic stages. The arrow shows the vesicles that reassembled the viruses, and such vesicles were abundant in the cytoplasm. Nucleocapsids and empty capsids were observed in the nucleus, of which nucleocapsids accounted for 66.7%, and there was no obvious perinuclear aggregation (Figures 5Ab). Moreover, we observed that the cells lysed after infection exhibited many soluble vacuoles, and the rupture of the cell membrane released a large amount of virus in the intercellular space (Figures 5Ac,d). Simultaneously, we infected DEF cells with 1-MOI virus, detected the titer of infectious virus particles released into the supernatant, and found that there was no significant difference in the mature virus particles in the supernatant (Figure 5B), while the viral titers in the cells of the deletion strain, the revertant strain, and the parent strain were at the same level (Figure 5C). The data show that LORF5 has no effect on the assembly and release of virus particles. Based on the above experiments, we can conclude that the presence or absence of LORF5 does not affect the adsorption and invasion of the virus or the replication, assembly, and release of the viral genome in vitro.

Then, we investigated if LORF5 plays a role in DPV replication through plaque size assays. As introduced in Materials and Methods, a 1.5% methylcellulose semisolid cell culture medium was used to ensure that adjacent cells were infected only by viral cell-to-cell spread; then we tested the transmission of deletion virus, WT, and reverted virus through plaque experiments.

Here, we used two methods for plaque statistics. In the first method, we detected the area of fluorescent plaques produced by ΔLORF5, RΔLORF5, and CHv-BAC in infected cells at 36 h (Figure 6A). The fluorescent area formed by the ΔLORF5 mutant virus was smaller than that of RΔLORF5 and CHv-BAC. On average, the diameter of the fluorescent plaque formed by the ΔLORF5 strain was 43.6 mm, while that of RΔLORF5 was 51.7 mm and that of CHv-BAC was 54.0 mm. And Figure 6B was the statistical analysis of Figure 6A; abrogation of LORF5 slightly impaired viral spread by 19.3% compared with WT (100%). In the second method, we counted the cytopathic plaques by crystal violet staining when the cells were infected for 5 days (Figure 6C). The average size of plaques produced by ΔLORF5 was 0.860 mm, while RΔLORF5 was 1.087 mm and CHv-BAC was 1.090 mm. And Figure 6D was the statistical analysis of Figure 6C; the size of the plaques formed by ΔLORF5 was reduced by 21.1% compared with WT, and there was no difference between WT and the reverting strain (data were considered significantly different if the P-value was ≤0.05). Then we come to the conclusion that the loss of the LORF5 gene truly reduces the transmission efficiency of DPV between DEF cells. In other words, we found that LORF5 has a function in cell-to-cell transmission, which was the first report.

All duck embryo experiments were approved by the Committee of Experimental Operational Guidelines and Animal Welfare of Sichuan Agricultural University, with approval number S20167031-1707. Experiments were conducted in accordance with approved guidelines.

DEF cells were prepared from 9- to 12-day-old unfertilized duck embryos containing 10% serum MEM (serum, MEM from Gibco) as the growth medium and 2% serum as a maintenance medium. The DPV CHv strain was separated and preserved in the laboratory (accession no. JQ647509). All primers used in this paper were designed by Primer 5 software. The primer sequences and products are shown in Table 3.

GS1783-pBAC-DPV was constructed by Ying Wu in our laboratory, in which the DPV genome was cloned into a BAC for the goal of DPV modification in bacteria. The targeted fragments were amplified using primers (Table 3) and then targeted into Escherichia coli GS1783-pBAC-DPV competent cells. In the first step of homologous recombination, the LORF5 gene was replaced by the Kan resistance gene (1,354 bp). Then, the I-SecI site was cleaved followed by a second homologous recombination, resulting in the deletion of the Kan fragment (245 bp). To obtain the mutant viruses, the ΔLORF5 or RΔLORF5 plasmid (1 μg) using the QIAGEN Plasmid Midi Kit (cat. no. 12143) from a positive colony was transfected into DEF cells by Lipofectamine^® 3000 Invitrogen (L3000001), the MEM medium with 2% calf serum was replaced, and the culture was continued at 37°C after incubating for 6 h. PCR analysis was performed on fourth-generation virus DNA with primers of US8 and LORF5 (Table 3). Positive BAC clones extracted with the Qiagen Plasmid Midi Kit were confirmed by RFLP analysis, with the system (25 μL) including 1 μg plasmid, 2 μL restriction enzyme EcoRI, 4 μL 10× Q.Cut G.Buffer, and ddH₂O to replenish, analyzed by 1% agarose gel electrophoresis at 50 V electrophoresis for 2–4 hpi after cutting at 37°C for 2 hpi.

DEF cells in 12-well plates were infected with 0.01 MOI of CHv-BAC-ΔLORF5, CHv-BAC-RΔLORF5, or CHv-BAC; the plates were shaken every 15 min during the 37°C incubation for 2 h; and then the MEM medium with 2% calf serum was changed. Samples of the infected cells and their supernatants were collected separately at 12, 24, 48, 72, and 96 hpi, and the volume of each sample was increased to 1,000 μL with MEM, and then the samples were frozen and thawed three times. Intracellular viral titers and supernatant viral titers were detected by determining the 50% tissue culture infectious dose (TCID₅₀) to assess the virus. Statistical analyses were performed using GraphPad Prism version 8 (San Diego, CA, United States), and data were considered significantly different if the P-value was ≤0.05. Growth kinetics data were repeated three independent times. Asterisks indicate significant differences compared to WT virus (^∗∗∗P < 0.001; ^∗∗P < 0.01, ^∗P < 0.1).

Culture DEF cells in a monolayer were placed in a six-well plate. After the cells were full for 12–24 h, they were precooled at 4°C for 1 h, washed with 4°C precooled PBS buffer to remove dead cell debris, then infected with 0.001 MOI of virus, incubated at 4°C for 2 h (agitated every 5 min), and then washed five times with precooled PBS. The viral genome was extracted from the cell samples (TaKaRa MiniBEST Viral RNA/DNA Extraction Kit Ver. 5.0), and the copies were detected by q-PCR. Alternatively, the medium was replaced with 1.5% solid medium after washing, and the number of plaques was counted after 36 h of incubation at 37°C. To detect invasion, cells were washed after incubation at 4°C, incubated at 37°C for 1–2 h, and then subjected to the same experiments. The number of plaques was calculated and analyzed by GraphPad Prism 8.0.2 software.

Total RNA was isolated from virus-infected cells using the RNeasy Plus Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. DNase I (Promega, Fitchburg, WI, United States) was used to remove any genomic DNA, and then cDNA was generated using the high-capacity cDNA reverse-transcription kit (Applied Biosystems/Thermo Fisher Scientific, Waltham, MA, United States). Premix Ex Taq^TM (Probe q-PCR) (Takara, Dalian, China) was used to determine viral cDNA copies. The primers and probe used to detect the BAC-CHv UL30 gene by q-PCR were designed previously in our laboratory. q-PCR amplifications were performed under the following conditions: 95°C for 30 s, followed by 40 cycles at 95°C for 5 s and 60°C for 30 s. Then, the q-PCR products were quantified by comparison with the established standard curve of the laboratory. For mRNA transcription level, the LORF5, UL55, and LORF4 genes were detected using the previously described primers. The conditions were set as follows: initial denaturation at 95°C for 1 min, followed by 45 cycles of denaturation at 95°C for 5 s, annealing at 59°C for 20 s, and extension at 72°C for 25 s. All reactions were performed in triplicate and in at least three independent experiments. The cycle number at threshold (Ct value) was determined to analyze these gene transcriptions, and the results were calculated using the 2^–ΔΔCt method.

DEF cells in 60-mm dishes were infected with CHv-BAC-ΔLORF5 at an MOI of 5, collected in a centrifuge tube by scraping at 14 hpi, and centrifuged at 1,000 rpm for 5 min to collect the cells; and the supernatant was discarded. The cells were added with 0.5% glutaraldehyde fixative solution, resuspended, and fixed for 10 min at 4°C; then centrifuged at 12,000 rpm for 10 min; and then readded with 3% glutaraldehyde fixative solution for fixation. All samples were then sent to Chengdu Lilai Biological Technology Co., Ltd., for analysis under a transmission electron microscope (Hitachi H-7650, Tokyo, Japan).

DEF cells in six-well plates were infected with 0.001 MOI of CHv-BAC-ΔLORF5, CHv-BAC-RΔLORF5, or CHv-BAC. After incubation for 2 h at 37°C, 1.5% methylcellulose (Solarbio, Beijing, China) was added to cover the cells. Here, we used two methods for plaque statistics. In the first method, the green fluorescent plaques produced by ΔLORF5, RΔLORF5, or CHv-BAC-infected cells at 36 h were calculated, and the cells were observed under a fluorescence microscope (Nikon TI-SR, Japan). Fifty randomly selected green fluorescent plaques were photographed per experiment. The second method counted the cytopathic plaques by crystal violet staining when the cells were infected for 5 days (cultured in a 37°C, 5% CO₂ incubator). In particular, the medium was discarded, and cells were fixed with 500 μL of precooled 4% paraformaldehyde at room temperature for 20 min, washed twice with sterile PBS, and added with 500 μL 0.5% crystal violet for staining for 30 min; they were then rinsed with tap water to remove the staining solution and to observe and count the plaques. All the average plaque size was measured using Image-Pro Plus software (Bio-Rad, CA, United States). The plaque size of the deletion virus and the reverted virus was calculated and compared with the plaque size of the parental virus set at 100%. Data were considered significantly different if the P-value was ≤0.05 (^∗∗P < 0.01, ^∗∗∗P < 0.001). All reactions were performed in triplicate and in at least three independent experiments.