Background
Duck Viral Enteritis (DVE), also known as Duck Plague (DP), is an acute, febrile, septic and contagious disease of ducks, geese, swans and many other species of birds within the order Anseriformes caused by Duck Enteritis Virus (DEV) [
1]. The morbidity and mortality of infected young ducklings or unprotected ducks reaches up to 100%, resulting in huge economic losses of domestic and wild waterfowls worldwide [
2‐
4]. In addition, after primary infection, the viruses persist in their host for life. They hide from the immune system in a latent state in which their genome is almost dormant. Eventual episodes of reactivation allow them to infect naive individuals, causing a long-term of prevalence in the high density duck raising farms after DVE outbreak [
5]. Therefore, even live attenuated DEV vaccines have been used in prevention and control of this lethal disease since 1960s, the DEV infection is not completely prevented [
6‐
10]. Thus, investigation of DEV gene functions and pathogenesis will be an intelligent policy to novel vaccine development and disease controlling.
According to the reports, studying the resulting variations in phenotype of virus mutants can be very useful for understanding the fundamental information of virus gene function and pathogenesis, which provide theoretical basis for development of novel vaccines and chemotherapeutics [
11]. A bacteria artificial chromosome (BAC) can take up the complete genome of a herpes virus as an infectious clone for randomly mutagenesis without requirement of restriction sites or cloning steps in E. coli have been developed for herpes virus mutagenesis and reconstitution which extensively facilitated functional studies of viral genes [
11]. After the first case of infectious herpes virus BAC clone has been successfully applied in murine cytomegalovirus (MCMV) in 1997 [
12], the strategy of cloning full length of herpesvirus genome as a BAC has been adopted for herpes virus. In most recent years, infectious BAC clone has been gradually applied in DEV study. Wang et al. firstly generated an infectious BAC clone of the European virulent DEV strain 2085 and expressed hemagglutinin (H5) of high pathogenicity H5N1 avian influenza virus (AIV) based on that. The insertion of BAC components into DEV 2085 genome UL44 (gC) gene caused disruption of gC function [
13]. In China, full length of DEV vaccine strains VAC and C-KCE were extensively cloned into BAC to generate bivalent vaccines for protection of ducks against DEV and other poultry pathogens. DEV glycoprotein gC, junctions of UL26 and gB, regions between UL15B and UL18 gene, SORF3 and US2 junctions were selected for insertion of BAC Mini-F sequence [
6,
14‐
18]. From these literatures, we found that DEV infectious BAC clones were predominantly considered as an ideal vector for constructing live bivalent vaccines. However, rarely reports focused on the functional characterization and pathogenesis of lethal DEV that will lead to high morbidity and mortality of infected ducks. Although bivalent live vaccine can protect ducks against DEV and other causative pathogens infection at one time, but lacking of molecular background of the DEV carrier and integration of two pathogens may lead to potential security risks in a relatively long period which we cannot predicted. Therefore, constructing a platform that can be used for functional characterization and pathogenesis investigation of virulent clinical isolate DEV will benefit the long-term development of novel vaccines and disease control.
To our knowledge, previously studies about DEV almost focused on the epidemiology and prevention of this disease. Limited molecular biology data are available regarding DEV genome and its encoding proteins except some significant genes related to virus infection, replication and antiviral effect. As one of the least known ORFs among total 78 genes of DEV genome, homologues of the DEV UL55 protein (pUL55) are encoded only among alpha herpes viruses [
19]. Reports about herpes simplex virus 2 (HSV-2) UL55 protein revealed that the product of HSV-2 UL55 gene may play an accessory role in virion assembly or maturation and has some relationship with UL26.5 distribution [
20], but the corresponding homologue gene of EHV-1 was supposed to mediate persistent infection [
21]. Pre-existing data suggested that the HSV-1 UL55 gene was not critical for intraperitoneal virulence or establishment of latent infection [
22], but subsequent documents suggested this dispensable gene was thought to be important for virus growth and spread in the natural host. However, the characteristics of DEV UL55 gene in DEF cells remains unclear due to limited data. Thus, carrying out some functional studies of UL55 gene will provide some data for further studies of DEV pathogenesis.
In this study, we firstly constructed an infectious BAC clone of the Chinese virulent DEV strain (DEV CHv) to generate a platform that can be used for functional characterization of DEV genes in DEF cells. The UL23 gene of DEV, which has been demonstrated nonessential for DEV replication, was used for inserting BAC mini-F sequence and screening marker EGFP [
23‐
26]. The functional studies of DEV CHv UL55 gene in DEF cells was carried out based on this newly established platform by using two-step RED recombination for generation of UL55 deletion and revertant mutants in
E. coli. The data yielded by this study demonstrated that UL55 gene is nonessential for virus replication and has no effect on distribution of UL26.5 protein in infected cells. As a result, we can conclude that this DEV BAC system will facilitate the studying of the biology and gene functions of DEV field strain. Moreover, the nonessential UL55 gene could be a candidate locus for developing bivalent vaccine after attenuation.
Methods
Cells, virues, plasmids and antibodies
The DEV CHv strain (Accession NO. JQ647509) was used for the construction of infectious BAC clone. Duck embryo fibroblast (DEF) cultures prepared from 10-day-old cherry valley duck embryos were used for the propagation of DEV CHv and its derived mutants, which was maintained in minimal essential medium (MEM, Gibico) supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin (1% P/S) and 10% new calf serum (NCS,Gibico) at 37 °C under a 5% CO2 atmosphere. The help plasmids pKD46, pKD4 and Pcp20 were kindly donated by Prof. Kelly T. Hughes (University of Utah), while the plasmid pBeloBAC11 and E. coli strain DH10B was supplied by Yunfeng Wang (Harbin veterinary research institute, Chinese Academy of Agricultural Science). Our lab prepared the antibodies against UL55 protein and UL26.5 protein by immune healthy rabbits.
Constrution of transfer vector pUC18/EGFP-TKAB-BAC11
The plasmid pBeloBAC11 containing the mini-F sequence was used for cloning of the DEV CHv complete genome. It was carried out broadly on the same principle used for the construction of the BAC clones of MDV-1, HVT, VZV, BoHV-1 and EHV-1 etc. by inserting the bacterial mini-F sequence and the enhanced green fluorescent protein (EGFP) into UL23 (TK) gene of DEV CHv through homologous recombination. The strategy for constructing the transfer vector containing the essential functional components (mini-F) of BAC, EGFP and UL23 homologous sequence to facilitate recombination was shown in Additional file
1: Figure S1. The generated transfer vector pUC18/EGFP-TKAB-BAC11 harboring the homologous regions of 1357 bp on the left and 1039 bp on the right flanking regions of TK insertion site, mini-F sequence and a cellular screening marker EGFP.
Constrution of BAC clone of DEV CHv
The cloning of DEV CHv complete genome into BAC plasmid was performed by homologous recombination in DEF cells (Additional file
2: Figure S2 a, b). To the details, approximately 2.5 μg pUC18/EGFP-TKAB-BAC11 was transfected into freshly seeded primary DEF cells infected with DEV CHv strain by using Lipofectamine 3000 (Invitrogen). Meanwhile, the DEV CHv infected cells and mock-infected cells transfected with pEGFP-∆MCS were did in parallel as controls. When the green fluorescence plaques appeared, several rounds of EGFP positive plaques selection were performed for recombinant virus purification and enrichment. Genomic DNA of the cultures was obtained at the next step by sodium dodecyl sulfate (SDS)-proteinase K extraction as described earlier [
27] for identification of the purified BAC-recombinant DEV CHv by PCR using the primers listed in Table
1. 100 ng identified BAC-recombinant DEV CHv DNA was electroporated (1.8 KV, 200 Ω, 25 μF) into E. coli DH10B cells using a Bio-Rad E. coli Pulser with 0.1 cm cuvettes. The colonies grown on chloramphenicol plate were detected by PCR to confirm the presence of EGFP, essential functional components of BAC and some important genes related to replication, virulence and structure of DEV CHv virions. Furthermore, plasmid extracted from the PCR identified chloramphenicol-resistant BAC clones was digested with
EcoR I for RFLP analysis, the correct plasmid was named pBAC-DEV and further identified by sequencing (Invitrogen).
Table 1
Primers used in this paper
1 | TKA-HOMO-for | gaattcatgcttgccatcataaccgtattctc | TK left homology arm |
TKA-HOMO-rev | tctagaataacttcgtataatgtatgctatacgaagttatcacctcgagcttttctttcctgtg |
2 | TKB-HOMO-for | gcatgcacatagcaacaactgacgcaaaagc | TK right homology arm |
TKB-HOMO-rev | aagctttcccagaaagctcgcctaggtcctc |
3 | EGFP-for | tctagatagttattaatagtaatcaattacg | EGFP |
EGFP-rev | gtcgacatgcagtgaaaaaaatgct |
4 | sopB-for | attcgttaattgcgcgcgtagg | sopB |
sopB-rev | gaatattcaggccagttatgct |
5 | repA-for | catggcggaaacagcggttatc | repA |
repA-rev | atgtatgagaggcgcattggag |
6 | TK-for | cgcggatcccactgaatgtcactgc | TK |
TK-rev | cccaagctttcaattaattgtcatctcggt |
7 | ΔUL55-KanR-for | gaaaggcggttggaataagaggaacgaggcggtagacgtgaccgacaacagtgtaggctggagctgcttc | KanR gene flanked by homology arms of UL55 |
ΔUL55-KanR-rev | tttcttatggttttaataaaacgctttattacattgtagtgtaacaagaccatatgaatatcctccttag |
8 | ΔUL55/ΔUL55R-for | tgcaaattagtgggaggtacg | ΔUL55/ΔUL55R identification product |
ΔUL55/ΔUL55R-rev | cccaaataccctgttagtagctt |
9 | ΔUL55R-UL55-for | atggccgacgcgaaggcggt | UL55 fragment with left homology arm of UL55 |
ΔUL55R-UL55-rev | gaagcagctccagcctacactcatacattagctttgtg
|
10 | ΔUL55R-KanR-for |
cacaaagctaatgtatgagtgtaggctggagctgcttc | UL55 fragment with right homology arm of UL55 |
ΔUL55R-KanR-rev | tttcttatggttttaataaaacgctttattacattgtagtgtaacaagaccatatgaatatcctccttag |
Reconstitution of infectious virus from pBAC-DEV clone
For the rescue of infectious recombinant virus in host cells, 2.5 μg freshly prepared pBAC-DEV DNA by QIAGEN Midi Kit was transfected into DEF cells using Lipofectamine 3000 (Invitrogen). Cells transfected with or without pEGFP-ΔMCS were taken as controls. Cells with green fluorescence CPE were harvested and passaged for enrichment of infectious virus production. The infectious recombinant virus named DEV CHv-BAC-G will be obtained until complete cytopathic effect had been developed.
For determination of the reconstitute DEV CHv-BAC-G, PCR was performed firstly for the identification of DEV CHv gene and the functional components of BAC. A rabbit polyclonal antibody against DEV UL23 gene was used to detect the expression of TK protein in DEV CHv-BAC-G infected DEF cells by indirect immunofluorescence analysis (IFA). Mock and wild type DEV CHv infected DEF cells were used as controls to demonstrate the specificity of anti-TK for DEV CHv. In IFA, DEF cells were grown on glass coverslips and infected with either wild type DEV CHv or reconstitute DEV CHv-BAC-G at a multiplicity of infection (MOI) of 0.02, 200 μl anti-TK IgG at a 1:50 dilution was used as first antibody. The corresponding pDEV-BAC that can produce green fluorescence plaques was isolated for sequencing.
Chracterization of the rescued virus in vitro
Growth curve analyses were performed to compare the growth kinetics of the wild type DEV CHv with that of the reconstitute virus DEV CHv-BAC-G. Briefly, DEF monolayers grown on 24-well plates were infected with 0.02 MOI of wild type and reconstitute virus. Supernatants containing virus were harvested directly at 7, 12, 24, 36, 48, 72 h post infection (h.p.i), while the cells and mixture of cultures were harvested at the indicated time points by treating with trypsin. The amount of infectious wild type and reconstitute virus in the harvested supernatants and cells were determined by Reed-Muench assay as previously studies described [
28]. Growth curves assay was performed in triplicate in three independent experiments.
To determine plaque areas of DEV CHv and DEV CHv-BAC-G, DEF cells were infected at an MOI of 0.02. The inoculated viruses were discarded at 1 h post infection before adding 500 μl MEM containing 0.5% methylcellulose and 10% NCS. After 3–5 days, the methylcellulose in each well was replaced with 500 μl pre-cold 4% paraformaldehyde for cells immobilization. Cells were then washed and stained with 500 μl crystal violet for 5 min. Thorough washing were performed subsequently to remove the staining solution. Plaques were photographed, and the average plaque areas were determined using the Image J software (
http://rsb.info.nih.gov/ij/). Values of DEV CHv-BAC-G were calculated and compared to wild type DEV CHv plaque areas, which were set to 100%. T-tests were used to assess the significance of the results (
P < 0.05). Average percentages of plaque diameters and standard deviations were determined from at least three independent experiments.
Transmission electron microscopy of the 0.1 MOI DEV CHv-BAC-G infected DEF cells were performed according to previously reports [
29]. Briefly, cells were washed with PBS at 36 h.p.i, and fixed with 2.5% glutaraldehyde at 4 °C for 2 h. After that, the fixed adherent cells were collected by scraping from the flasks and centrifuged at 40,000 rpm/min for 2 h. Then the pellets were mixed with 2% low melting-temperature agarose at 37 °C, and centrifuged at 6000 rpm/min for 10 min. Samples were post-fixed in 1.0% osmium tetroxide. After a stepwise dehydration in acetone, samples were embedded in epoxy resin 618 and polymerized at 80 °C for 72 h. Then, 50 nm ultra-thin sections were prepared with an LKB ultra tome (LKB Instruments, Inc., Rockville, MD), collected on grids, and stained with uranyl acetate and lead citrate for subsequent examination with the Hitachi H-600-A2 transmission electron microscope at an accelerating voltage of 75 kV. Images were recorded on Kodak electron microscope film and compared to that of DEV CHv virions.
Generation and characterization of UL55 deletion and its revertant mutant based on established recombinant DEV CHv-BAC-G
Cloning of wild type DEV CHv as infectious BAC clones facilitates easy manipulation of their genomes. In order to test the amenability of pBAC-DEV clone to genomic modification for DEV CHv gene study, we constructed a UL55 gene deletion and its revertant mutants by two-step RED recombination [
30]. In brief, 100 ng PCR product of kanR cassette flanked by the FRT sites and homology arms of UL55 gene was electroporated into DH10B cells containing pBAC-DEV clone and pKD46 to complete the first round of recombination induced by 100 mM L-arabinose for expression of RED recombination related proteins exo, beta, gam in pKD46. Kanamycin resistant colonies were isolated on LB/Cm/Kana plates. In the second round of RED recombination, the temperature sensitive plasmid Pcp20 was transformed into DH10B cells containing kanR colonies of the 1st recombination to trigger the excising of kanR FRT cassette. The induced Pcp20 can be removed at 42 °C and the positive colonies identified by PCR and sequencing was named as DEV CHv-BAC-GΔUL55 (Additional file
2: Figure S2 b,c).
The construction of DEV CHv-BAC-GΔUL55 revertant following the same mutagenesis protocol. In the 1st round of RED recombination, overlapping PCR product of complete UL55 gene and kanamycin resistance gene flanked by the homology arms of UL55 deletion region was inserted into the UL55 mutation locus (Additional file
2: Figure S2 c,d). The revertant mutant DEV CHv-BAC-GΔUL55R was obtained after excising of KanR in the 2nd round of RED recombination (Additional file
2: Figure S2 d). Infectivity of the DEV CHv-BAC-GΔUL55 and DEV CHv-BAC-GΔUL55R mutants were examined by transfection of them into DEF cells as their parental strain DEV CHv-BAC-G did. PCR, IFA and RFLP analysis were carried out as we described above to confirm the presence of DEV CHv genome and BAC functional components, and the mutation of UL55 region. Growth characteristics of DEV CHv-BAC-GΔUL55 and DEV CHv-BAC-GΔUL55R were compared to their parental virus DEV CHv-BAC-G as we previously described to determine the function of UL55 in DEV replication.
The effect of UL55 protein on the distribution of UL26.5 protein in infected DEF cells
To investigate the intracellular distribution correlation of UL55 and UL26.5 proteins, DEF cells infected with 0.02 MOI wild type DEV CHv and recombinant viruses were collected for IFA in two independent experiments. In the first experiment, DEV CHv infected DEF cells were collected at 60 h.p.i. and fixed with 4% paraformaldehyde. After washing step, the fixed cells were incubated with 1:50 diluted mouse anti-UL55 or 1:100 rabbit anti-UL26.5 IgG prior to addition of 1:100 diluted TRITC-conjugated Goat anti-mouse IgG or FITC-conjugated Goat anti-rabbit IgG for the single distribution analysis of UL55 and UL26.5 encoded proteins in host cells. For co-localization analysis, 1:50 diluted mouse anti-UL55 and 1:100 rabbit anti-UL26.5 IgG were added together into harvested DEF cells as primary antibody while the mixture of 1:100 diluted TRITC-conjugated Goat anti-mouse IgG and FITC-conjugated Goat anti-rabbit IgG were added as secondary antibody. DEF cells infected with 0.02 MOI parental DEV CHv-BAC-G and its derived mutants DEV CHv-BAC-GΔUL55 and DEV CHv-BAC-GΔUL55R were collected and fixed as mentioned above to demonstrate the contribution of UL55 gene to UL26.5 protein distribution. 1:100 diluted rabbit anti-UL26.5 IgG and TRITC-conjugated Goat anti-mouse IgG were used as the first and secondary antibodies for IFA Fluorescence of samples above were inspected under invert fluorescence microscope.
Discussion
As full genome sequence and organization of DEV have been published to date, specific gene functions could be characterized. To our knowledge, however, the specific characteristic and function of the UL55 protein in DEV CHv are still unknown due to the limited technology of genome manipulation in host eukaryotic cells [
15]. Recent advance in biologic technologies bring in many genome editing approach including the most popular CRISPR/Cas9 system. Based on this, scientists have been successfully knocking out protein-coding genes in several model organisms [
34]. It is now become an important approach for understanding gene functions and for engineering genetic regulatory systems. However, the successful rate of gene knock out/down is not 100% because of its limitations. The non-specificity of guide RNA target sequence and the uncertainty of DNA damage repair both may result unknown mutations in the other locus of genome (off-target effect), especially in organisms with large genomes [
35]. Compared to this, a full length genomic cloning technology based on BAC allowing the recombinants to be reconstituted as an infectious virus in host cells after mutagenesis in
E. coli become an optimal alternative approach for herpes virus genome manipulation. In this strategy, large herpes virus genomes can be manipulated accurately and rapidly in
E. coli without unwanted recombination events or rearrangements due to the only homologous double-strand ends will be used as a substrate [
36,
37]. The stability of the recombinant BAC clone in
E. coli can be maintained because of the strict replicons in BAC mini-F sequence. On basis of that, infectious BAC clone is considered as a preferred large-insert cloning system for genomic analysis and gene discovery in herpes virus, and undoubtedly contributed a lot to investigations of virus life cycles, vaccines, gene functions and pathogenesis [
37‐
39]. Thus, we firstly constructed an infectious BAC clone of the Duck Enteritis Virus Chinese virulent strain (DEV CHv) for investigation of UL55 gene function in infected DEF cells in this paper. We believe outcomes will provide some clues for understanding mechanisms of DEV life cycle and that involved in DEV pathogenicity.
Based on the established infectious DEV CHv BAC clone, mutagenesis of UL55 gene was generated by two-step RED recombination for testing the amenability of this platform for DEV CHv gene study. The major advantage of the Red recombination system is that only short homologous sequences of 30 to 50 bp are required for the recombination to proceed. For the recombination in bacteria, components of the Red or RecE/T recombination system can be delivered
in trans by plasmids such as pKD46 that allow inducible expression of Alpha, Beta and Gam. Once the mutagenesis procedure is completed, pKD46 can be cured from bacteria by its temperature-sensitive replication mechanism. Furthermore, the unwanted selection markers introduced by 1st round RED recombination can be excised from the recognition sites by Flp recombinase [
30,
37]. Besides, compared to the popular en passant protocol based on BAC maintained in GS1783, the traditional two-step recombination methods needs several help plasmids was much more accurate in the first round of mutagenesis. Thus, the combination of DEV infectious BAC clone system and RED recombination or other mutagenesis technologies allows easily generation of a variety of different modifications of DEV CHv. To exclude the random mutation of UL55 gene in genetics procedures, the deletion of UL55 gene was repaired by revertant mutation using RED recombination. The subsequent transfection into DEF cells demonstrated that UL55 gene is dispensable for DEV CHv replication and the UL55 deletion mutants exhibit identical growth characteristics to its parental and revertant virus in vitro
.
Study of HSV-2 UL55 gene suggested that UL55 protein abutted on and partially overlapped with the UL26.5 gene encoded capsid protein ICP35 in intracellular [
20]. To our knowledge, virus relies on host synthetic machine to complete its replication, biological functions of mature proteins must occur at the specific locations in cells [
40]. The intracellular compartmentation distribution can affect the folding, aggregation and posttranscriptional modification process of protein and further effects on function of cells [
41]. The abutting and overlapping distribution features of UL55 and UL26.5 protein in DEV CHv infected cells revealed functional correlations of UL55 and UL26.5 protein in virus replication. The co-localization features could be explained as the UL26.5 encoded protein is involved in viral capsid assembly as a scaffold protein. The assembled capsid protein gets transferred to the periphery of nuclei close to nuclear membrane for releasing into cytoplasm and subsequently the nearby UL55 included assemblons will accomplish the viral assembly in cytoplasm subsequently. The abutting feature in intracellular was aimed to execute the viral package function of UL55 protein. The explanation for the overlapping distribution of UL55 and UL26.5 protein could be attributed to the evacuation and degradation of UL26.5 protein after the capsid synthesis. The resulted aggregation of falling off UL26.5 protein from inner capsid structure in cytoplasm and nucleus may be detected by anti UL26.5 IgG and overlapped with fluorescence produced by the widely distributed UL55 protein in cytoplasm. Further distribution analysis of UL26.5 protein in the UL55 negative mutant infected DEF cells revealed that the absence of UL55 protein didn’t affect the distribution and function of UL26.5. Herein, we proposed a hypothesis that the assembly process of capsid constituted by UL26.5 and other component part may involves in multiple proteins. DEV UL55 protein plays a role in viral assembly, but it can be substituted. Further work is being carried out to testify the possibility of UL55 gene locus as an exogenous gene insertion site for developing vectored vaccine and virus attenuation. We believe the obtained data will supply some important information for virus pathogenesis research and vaccine development.
Acknowledgements
We would like to thank Yunfeng Wang (Harbin veterinary research institute, Chinese Academy of Agricultural Science) and Prof. Kelly T. Hughes (University of Utah) for kindly donating plasmids and E. coli strains.