Reconstitution and Mutagenesis of Avian Infectious Laryngotracheitis Virus from Cosmid and Yeast Centromeric Plasmid Clones

An ILTV cosmid library was constructed using pSuper Cos with partially Sau3AI-digested genomic DNA. Random colonies were selected, and the sizes of the inserted ILTV DNA fragments were evaluated following digestion with NotI. Boundaries of the ILTV inserts in clones containing large inserts (39-45 Kb) were partially sequenced by dideoxy sequencing with vector-specific primers. Subsequently, the clones were digested with the restriction endonucleases XhoI, PvuI, and EcoRI to confirm the appropriate digestion patterns for the ILTV genomic insertions (data not shown). Four cosmids (pCIZ34, pCIS28, pCISB27, and pCIZ52) containing inserts with overlapping genomic coordinates collectively representing 89.9% of the genome were selected and further characterized by Illumina MiSeq sequencing. The genomic coordinates based on the USDA strain (JN542564) are provided in Table 1. Based on the USDA genome, 15,271 bp (1,073 and 14,198 bp at the 5′ and 3′ termini, respectively) map outside the coverage of the 4 cosmids (See Fig. 1). Because of this, a “Bridge” construct that traverses the TR_S/UL junction was assembled in the YCp vector, pYES1L (Thermofisher). This construct (YCp-BC) contains 5′ sequences with genomic coordinates that overlap those at the 3′ end of pCIZ52 by 2,358 bp at the U_S/TR_S junction; it spans the entire TR_S region and overlaps UL sequences at the 5′ end of pCIZ34 by 1,026 bp. In addition, 2 other bridge constructs (YCp-KLO and YCp-ModKLO) were created in which the inverted pac2 (ipac2) sequences (USDA map coordinates 830 to 932) were replaced individually with a BamHI RE site or a GFP/kanamycin cassette, respectively.

There were no discernible growth differences among the bacteria containing cosmids pCIZ34, pCIS28, pCISB27, and pCIZ52 when propagated in Miller LB medium at 37°C and 42°C (Fig. 2). However, a slower growth rate was observed for the ModKLO YCp host relative to the cosmid constructs when propagated at 37°C. Notably, there were differences in the purity of cosmid DNA isolated from E. coli harboring pCIZ34 relative to the purity of other cosmids (pCIS28, pCISB27, and pCIZ52) isolated from the same E. coli strain when grown at 37°C (Fig. 3). When isolated from bacteria harvested during the exponential phase of growth at 37°C, a high level of chromosomal DNA was evidenced in restriction enzyme digests of pCIZ34 DNA. To improve upon this, different growth media (e.g., terrific broth, super broth, and 2X YT) were tested for E. coli propagation at 37°C. However, these growth media failed to improve the quality of the isolated cosmid DNA. Only when varying growth temperatures were tested did the purity of pCIZ34 dramatically improve. In addition to growth at 37°C, cultures were propagated in LB broth at the lower temperature of 30°C and the higher temperature of 42°C. The purity of the cosmid DNA was best when E. coli XL-1 MR containing pCIZ34 was propagated at 42°C. This purification anomaly may result from the induction of heat shock proteins, which stabilized the cosmid, or a toxic gene product from a cryptic prokaryotic promoter on pCIZ34, inactivated at the higher temperature.

The stability of the exceptionally long palindromes comprising replication origins was a concern during these experiments. The 2 cosmids pCIZ34 and pCIZ52 containing OriL (268 bp) and 1 copy of OriS (280 bp), respectively, are stable within the E. coli XL-1 MR host, and amenable to RedET mediated recombination. We serially passaged these clones in LB broth at 37°C (pCIZ52) and 42°C (pCIZ34) for 10 days and found no deletions in the palindromes. However, once these cosmids are reintroduced into different strains of E. coli via electroporation, it becomes exceedingly difficult to find recombinants with intact origins of replication. To investigate this, both types of palindromes were cloned into the yeast/E. coli shuttle vector pYES1L via transformation-associated recombination (TAR) in yeast and screened for intactness using yeast colony PCR. Nearly 98% of the yeast recombinants contained intact OriL and OriS. Unsurprisingly, no recombinants that had intact origins could be identified after shuttling these to the common electrocompetent E. coli strain TOP10 (and other strains). Even when using E. coli strains 6262 and 6787 (ΔrecA) and 6786 (recA+) (Scarab Genomics), which were engineered to stably harbor plasmids with pronounced secondary “stem-loop” structures, no recombinants were identified that contained intact origins. These strains were engineered to contain deletions in destabilizing Insertion Sequences (IS) and can harbor plasmids carrying direct repeats, inverted repeats, palindromes, or other sequences commonly unstable in standard E. coli strains (40). The E. coli recombinants had 243 nucleotides deleted in OriL and between 203 and 213 nucleotides deleted in OriS. We successfully created a YCp recombinant containing the UL insert of pCIZ34 in yeast and identified stable E.coli recombinants with intact origins (OriL). Similarly, YCp recombinants with inserts from pCIS28 and pCISB27 have been generated by TAR cloning in yeast and successfully transferred to E. coli, but we have not been able to accomplish this feat with the OriS-containing insert from pCIZ52.

Mirus transfection of LMH cells with the core set of cosmids pCIZ34, pCIS28, pCISB27, and pCIZ52 along with one of the TR_S/UL “bridge” clones YCp-BC, YCp-KLO, or YCp-ModKLO yielded infectious virus within 5 to 6 days (Fig. 4). Addition of recombinant vectors expressing the immediate early protein ICP4 and VP-16 (encoded by the U_L48 gene) to the transfection mixture enhanced infectivity (with plaques appearing earlier, sometimes by as much as 2 days earlier). While the cosmids and YCps could produce infectious virus without these ancillary plasmids, their addition increased the reliability and enhanced the efficiency of virus production. Plaque morphologies and the size of the plaques of vBC, vKLO, and vModKLO were similar to those observed with the USDA reference strain (Fig. 5). The accession numbers of the complete genome sequences of the reconstituted viruses vBC, vModKLO, and vKLO are MN784692, MN784693, and MN792995, respectively; and the USDA ILTV strain sequenced using Illumina MiSeq (USDA reference strain) is MN518177.

The pCIZ34, pCIS28, pCISB27, and pCIZ52 cosmid sequences were deposited in GenBank under the accession numbers MN784688, MN784689, MN784690, and MN784691, respectively. Sequence analysis of 77 complete ILTV genomes indicated near-perfect sequence identities of the cosmids with corresponding sequences from the genome of the USDA strain of ILTV. GenBank contains the complete nucleotide sequences of two USDA reference sequences (JN542534 and MN518177). For the cosmids pCIZ34, pCIS28, and pCISB27 with UL sequences, there was perfect sequence identity with that of the USDA (JN542534), and only 4 intragenic SNPs, which differed with the MN518177 sequence. A list of sequence differences can be found in Table 2. There were 2 different T/C substitutions in the U_L27 gene (gB) within pCIZ34 resulting in E189G and N678D substitutions. An A/G substitution in the U_L21 gene within the overlapping regions of pCIS28 and pCISB27 resulted in a synonymous substitution. A C/A substitution in the U_S7 gene (gI) also resulted in a synonymous substitution. Cosmid pCIZ52 contained the most genetic variations with INDELs in the 855/857 repeat region (~ 500 nuc upstream of the start codon of ICP4) and single nucleotide insertions in cytosine homopolymer stretches that flank the U_S10 gene.

In multiple alignments of the nucleotide sequences of the cosmids and YCps with genomic sequences of the reconstituted viruses, few base pair changes were identified. Only 2 base pair changes were identified immediately after the ipac2 site within the genomes of vModKLO and vKLO. This region is within the first 3,725 nuc of UL subgenomic region and is transcriptionally silent (unpublished results). No substitutions were identified in the vBC genome within this region.

The growth of recombinants vModKLO, vBC, and vKLO were measured in LMH cells (Fig. 6A). Between 0 and 24 h postinoculation, titers for USDA, vKLO, and vModKLO increased exponentially at similar rates. This contrasted with the lag growth noticed for vBC. However, by 48 h postinoculation (pi), vBC titers increased by 2 logs but never rose to the levels of USDA, vKLO, or vModKLO until 120 h pi. Statistically, there were no significant differences between the growth rates of the 4 viruses. Inspection of the entry data (Fig. 6B) might suggest a sight enhancement of entry for recombinants lacking the ipac2 sequence (vKLO with its BamHI linker and the large GFP/kanamycin cassette within vModKLO), however, this too, was not statistically significant with P values ranging from 0.07 to 1.0.

The virulence of the vBC, vKLO, and vModKLO recombinants compared to the USDA parental strain in specific pathogen-free (SPF) chickens inoculated via the intra-tracheal/ocular route was evaluated. At 3 days pi, the groups of chickens inoculated with the recombinant viruses and the USDA strain showed the highest clinical sign scores (CSS). Mortalities were recorded within the groups inoculated with vModKLO and the USDA parental strain (Fig. 7A). There were no significant differences (P < 0.05) in the median clinical sign scores of vKLO and vModKLO groups at any time point postinoculation. The median clinical sign scores observed for the vBC groups at day 4 pi were significantly lower (P < 0.05) than the median CSS for the USDA group, but clinical sign scores were not different at 3-, 5-, and 6-days pi (Fig. 7B). No clinical signs were observed in sham inoculated chickens. Overall, the recombinants induced significant clinical symptoms over time postinoculation. With the higher variance in the USDA group, there is no statistical support for a mean difference between the groups inoculated with the reconstituted viruses and those receiving the USDA parental control. The tracheal virus load in chickens inoculated with the recombinant viruses was not statistically different from that in chickens inoculated with the USDA strain (Fig. 8).

Leghorn male hepatoma (LMH) cells were grown in tissue culture dishes coated with 0.2% gelatin and propagated in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10 units/mL of penicillin, 10 μg/mL streptomycin, and 10% fetal bovine serum (FBS). Reconstituted ILTV was propagated on confluent monolayers of chicken kidney (CK) cells (48).

CK cells were infected with the USDA strain of ILTV using a multiplicity of infection (MOI) of 0.01. Four days postinfection, the cells were harvested and processed using a modified Hirt supernatant procedure (49, 50). Briefly, cells were trypsinized, pelleted, and lysed at 37°C for 16 h in 1.0 mL (per 75cm2 of cells) in a buffer containing 50 mM Tris pH 8.0, 10 mM EDTA, 0.5% SDS, and 1 mg/mL of proteinase K (New England Biolabs). High molecular weight DNA was precipitated with 350 μL of a solution containing 3M CsCl, 1 M potassium acetate, and 0.67 M acetic acid. After centrifugation, the herpesvirus DNA was extracted thrice with phenol-chloroform and precipitated with ammonium acetate and ethanol.

To obtain enriched virion DNA, chicken kidney cell cultures were infected as above and harvested 7 days p.i., lysed by freeze-thawing, and centrifuged for 10 min at 4000 rpm. Next, the supernatant was transferred to SW32 ultracentrifuge tubes, underlaid with 5 mL 30% sucrose in PBS, and centrifuged for 2 h at 20,000 rpm and 4°C. Then, the supernatant was completely aspirated, the pellet was washed once with TEN buffer, sedimented again for 1 h at 20,000 rpm, and resuspended in TEN buffer for DNA preparation, as described previously (51).

A pathogenic isolate of ILTV (obtained from D. Lűtticken, Boxmeer, NL) was used to construct the overlapping cosmid clones pCIZ34, pCIS28, pCISB27, and pCIZ52. ILT virion DNA was partially digested with Sau3AI, and fragments ranging from approximately 30 to 50 kbp were isolated from 0.6% agarose gels using diethylaminioethyl (DEAE) cellulose membranes (52) and cloned into the BamHI site of pSuperCos according to the instructions supplied by Stratagene (now Agilent). Ligated inserts were packaged using Gigapack gold packaging extracts, diluted in SM buffer (100 mM NaCl, 8 mM MgSO4 ·7H2O, 50 mM Tris-HCl, pH 7.5, and 0.01% gelatin), and used to infect XL1-Blue MR E. coli. Antibiotic-resistant clones were selected on LB agar plates supplemented with 0.2% (vol/vol) maltose, 10M MgSO4, and 100 μg/mL ampicillin. Mini prep DNA was subjected to restriction endonuclease profiling. Additional clones (pBC-114, pKLO-26, and pModKLO-10) spanning the TR_S/UL junction were generated with PCR products ligated into pUC19 using Gibson Assembly fusion cloning (New England Biolabs). The primers used in amplification reactions to generate PCR products are listed in Table 3.

To construct the pBC-114 clone containing one pac1 site and two pac2 sites (wild type configuration), primers “For3’TRS ILTV” and “REV5’UL ILTV” were used along with ILTV Hirt supernatant DNA (62 ng) to generate a 3,221 bp PCR product. Plasmid pUC19 was used as a template to generate a 2,712 bp PCR product using the primer pairs “For pUC19 04202014” and “REV pUC19 04202014” and 50 ng of pUC19. The fragments were amplified using Platinum Pfx polymerase (Invitrogen) in a buffer containing 2 mM MgSO4 with the following amplification parameters. After denaturing the templates at 94°C for 2 min, the parameters for 40 cycles were: 94°C for 30 s, annealing at 62°C for 30 s, and extension at 68°C for 2 min, respectively, followed by a final extension at 68°C for 15 min. These conditions were used for both amplification reactions.

These amplification reagents and conditions were also used to generate pKLO-26 and pModKLO. The pKLO-26 clone containing the deletion in the downstream ipac2 site was created using Gibson Assembly fusion cloning with 2 inserts that were generated in pfx polymerase amplification reactions with the 2 primer pairs (“For3’ TRS ILTV” and “REV PAC2 ILTV”) (“For PAC2 ILTV” and “REV5’UL ILTV”) and 62 ng of ILTV Hirt supernatant DNA. These 1,896 bp L and 1,282 bp O fragments were ligated with the 2,712 bp pUC19 PCR product. Fusion cloning was also used to generate the pModKLO clone in which the downstream ipac2 site was replaced with a GFP/kanamycin cassette. Two inserts were generated with the 2 primer pairs (“MluI FOR KLO” and “BsrGI REV KLO”) (“AsiSI FOR KLO” and “EcoRI REV KLO”) and 62 ng of ILTV Hirt supernatant DNA. The sizes of the PCR products were 1,989 and 1,293 bp, respectively. The third insert was generated with the primer pair (“BsrGI For KLO” and “AsiSI REV KLO”) with 100 ng of DNA isolated from the GΔORFC virus (53) and yielded a 3,979 bp product containing the GFP/kanamycin cassette. The pUC19 vector component was amplified using the primers pair (“EcoRI FOR KLO” and “MluI REV KLO”) with 50 ng of pUC19 DNA to generate a 2,709 bp PCR product. All fusion cloning procedures were performed with a 2-fold molar excess of inserts relative to the vector in a 20 μL reaction mixture for 30 min at 50°C. E. coli DH5α cells were transformed with the ligation products from all of the Gibson ligation experiments and propagated on Luria-Bertani (LB) plates containing 100 μg/mL of carbenicillin.

Constructs (YCp-BC, YCp-KLO, and YCp-ModKLO) containing 2.0 Kb segment from the 3′ end of the U_S, the complete 14.5 Kb TR_S sequences and variants of the 5′ UL sequences with or without alteration in the ipac2 site were assembled using the GeneArt High-Order Genetic Assembly System (Invitrogen) with PCR products and restriction endonuclease fragments from cosmid pCIZ52 and the pYes1L vector. A 5.1 Kb PCR product spanning the U_S/TR_S junction was amplified using the primer pairs (“ILTV BsrGI For” and “ILTV SpeI REV”) and 100 ng of GΔORFC virus (53), digested with BsrGI and SpeI and cloned into the synthetic minigene (MiniRepair -pIDT-SMART-AMP) to generate the recombinant pFX5.1. Cosmid pCIZ52 was digested with EcoRV, SpeI, AseI, and KpnI, and a 13.9 Kb KpnI–SpeI fragment was excised and used in assembly reactions with the oligonucleotide “stitches” (“Linker 2 FW rep2” and “Linker 2 RV rep2”). Likewise, the BsrGI-SpeI insert was gel purified from digested pFX5.1. Three assembly reactions were performed that contained the common fragments 5.1 Kb BsrGI- SpeI fragment and the 13.9 Kb KpnI-SpeI RE fragment, with either the 6.9 Kb MluI-EcoRV ModKLO fragment, the 3.0 Kb PstI-EcoRV KLO fragment or the 3.1 Kb PstI-EcoRV BC fragment.

PCR products containing the origins of replication were also cloned using the GeneArt High-Order Genetic Assembly System. The OriS and OriL fragments of ILTV were generated using primers “R2_14Kb” and “14Kb_1922F” and primers “cos 34 seq 87 F” and “cos34 Rev” with 50 ng of cosmid template pCIZ52 or pCIZ34, respectively. The amplification reaction conditions were as described above with similar cycling parameters, except the extension time was reduced to 1 min. The OriS PCR product (876 bp) was recombineered into the vector pYES1L using the stitching oligonucleotides OriS_FW1/OriS_RW1 and OriS_FW2/OriS_RVs to generate YCp-OriS. Similarly, the OriL PCR product (717 bp) was cloned using stitching oligonucleotides OriL_FW1/OriL_RW1 and OriL_FW2/OriL_RV2 to generate YCp-OriL.

All the assemblies of PCR products and restriction endonuclease fragments were performed in yeast (strain MaV203) using the PEG/lithium acetate transformation procedure with 100 ng of pYES1L vector, 200 ng of each insert, and 20 pmol of stitching oligonucleotides. Yeast transformants were selected on minimal yeast plates without tryptophan and screened using yeast colony PCR. Colonies were lysed in 50 μL of water with 2.5 Units of Lyticase (Millipore-Sigma) for 30 min at 37°C. DNA was released from spheroplasts by heating the solution to 95°C for 10 min, and 1.25 μL was used in a standard 12.5 μL PCR with HotStart Taq polymerase (New England Biolabs). Positive transformants from the PCR screens were propagated overnight at 30°C in yeast media (YPD) without tryptophan. DNA was isolated using a Zymolyase mini prep procedure and used to transform electrocompetent E. coli (TOP10) to generate recombinants YCp-ModKLO, YCp-KLO, and YCp-BC. E. coli recombinants were propagated in LB broth containing 50 μg/mL of spectinomycin. Miniprep DNA was isolated using alkaline lysis without phenol-chloroform extractions and characterized with restriction digestions.

Common electrocompetent strains from New England Biolabs (DH10β, NEB 5α), ThermoFisher (TOP10), and Agilent (XL-1 Blue) were initially used to test whether they would be suitable hosts for the stable propagation of YCp-OriS and YCp-OriL. Briefly, transformed yeast cells were screened for intact OriL and OriS palindrome using the lyticase PCR procedure. The mini prep DNA (100 ng) from positive yeast colonies was first treated with Exonuclease V (Plasmid Safe, Lucigen), then used to transform the various electrocompetent cells. Next, recombinants were selected on LB plates containing 50 μg/mL spectinomycin. Finally, E. coli recombinants were again screened for the intact origin of replication using the E. coli colony PCR procedure.

The electrocompetent E. coli strains 6262, 6786, and 6787 from Scarab Genomics were also tested for their usefulness in the stable propagation of the origin palindromes. The transformation protocol was identical to the above fore-mentioned, except the recombinants were selected on agar plates containing Modified Korz Medium with Glucose supplemented with 100 μg/mL of spectinomycin (54).

A mini gene was synthesized (IDT DNA technologies) that contained short DNA sequences necessary for homologous recombination with pSuperCos vector sequences that flanked the ILTV genomic inserts. Specifically, DNA block 1 contained 151 bp of sequence homologous to those 5′ of the T7 primer binding site, and block M had 157 bp of sequence homologous to those found immediately downstream of the Amp resistance ORF in pSuperCos. These DNA blocks, which contain a unique BamHI restriction endonuclease site between them, were cloned into pYES1L using standard cloning procedures with NotI and PacI. The final construct, linearized with BamHI, was used in assembly reactions with SfiI-linearized pCIZ34, pCISB27, pCIS28, and pCIZ52 (partially digested). All assemblies were performed in yeast (strain MaV203) using the PEG/lithium acetate transformation procedure as previously stated. E. coli TOP10 (ThermoFisher) was transformed with mini prep DNA isolated from yeast colonies and selected on LB plates containing 100 μg/mL ampicillin. E .coli recombinants were screened using colony PCR with the origin of replication-flanking primers, and miniprep DNA was characterized using restriction digestion.

The nucleotide sequences of the cosmids were determined using the Illumina MiSeq platform. Paired-end library preparation was performed by standard Illumina methods. Briefly, each sample was sheared, size-fractionated to 500 bp in length, and ligated to Illumina adapters with a unique barcode per sample. Each library was generated using the Illumina TruSeq Nano LT Sample Preparation Kit, according to the manufacturer’s specifications. The quantity and quality of the DNA were assessed using the Qubit BR dsDNA assay (Invitrogen) and an Agilent DNA High Sensitivity series chip assay. Libraries were standardized to 2 nM, and paired-end sequencing was performed on an Illumina MiSeq using version 2 kits to generate ca. 1 to 3 million 150 nuc paired-end reads per sample. Residual adapter-containing reads were removed, and reads were trimmed from the 3′ to a median Phred score of 30 and a minimum length of 50 nucleotides. The sequences of the U_S-TR_S-UL clones (BC-114, KLO-26, and ModKLO-10), and the origin of replication clones (YCp-OriS and YCp-OriL) were determined using Sanger sequencing with custom oligonucleotides. The genomic coordinates of the cosmid and YCp inserts are presented in Table 1.

Read preprocessing and assembly were performed using workflows run with Nextflow v18.10.1.5003 (55). Briefly, raw reads were quality-checked using FastQC v0.11.7 (56) and trimmed using Trim Galore v0.4.5 (57) (minimum quality = 8, minimum trimmed length = 40, stringency = 4). Host sequence removal was performed using the BBDuk tool of the BBTools package v.38.06 (58) (k = 35, hdist = 0). Digital normalization was performed using the BBNorm tool from BBTools (tgtcov = 75, minabund = 4, fixspikes=t). De novo assembly was performed using MIRA v4.9.6 (59) (parameter file available upon request). Circularization was performed using a modified Minimus2 script (60) from the AMOS package. Manual finishing was performed using Staden Gap5 v2.0.0b11 (61). Annotation was performed using an in-house avian herpesvirus-specific pipeline.

DNA was purified from 500 mL cultures (LB medium plus 100 μg/mL of ampicillin [cosmids] or 50 μg/mL spectinomycin [YCp]) using Nucleobond BAC100 (TaKaRa Bio). To facilitate homologous recombination, the cosmid, and YCp recombinants were digested with restriction endonucleases. The restriction endonuclease PacI was used to linearize the TR_S/UL clone YCp-ModKLO. NotI was used to linearize cosmid clones pCIZ34 and pCISB27. Cosmids pCIS28 and pCIZ52 were linearized with restriction endonucleases AscI and PacI. Clones YCp-KLO and YCp-BC were digested with PacI and SwaI, respectively. The DNA was subjected to microdialysis and concentrated via vacuum centrifugation after heat inactivation of the restriction endonucleases. The DNA concentrations were measured using a Qubit mini fluorometer (Invitrogen). Equimolar amounts (50 femtomoles) of the 4 digested cosmids (pCIZ34, pCIS28, pCISB27, and pCIZ52) and YCp recombinants (either YCp-BC, YCp-KLO, or YCp-ModKLO) with 2.0 μg of ancillary expression plasmids encoding ILTV U_L48 and ICP4 were mixed with 12 μL of Mirus TransIT-LT1 reagent in 250 μL of serum-free medium. After a 30-minute incubation, the DNA- lipid complexes were added to semi-confluent monolayers of LMH cells. Transfected cells were incubated at 37°C under 5% CO2 and monitored daily for signs of cytopathic effect (CPE). CPE was observed on day 5, and whole-cell freezer preps were made on day 6 posttransfection. Briefly, the medium containing cell-free ILTV was removed and stored at −80°C. The infected cells (with 500 μL of medium) were also stored at −80°C and subjected to 3 rounds of freeze-thaw cycles to release the cell-associated virus. The stability of the reconstituted ILTV recombinants was investigated by sequential passage (3 times) in monolayers of CK cells. The rescued viruses were allowed to infect subconfluent monolayers of CK cells for 16 h, after which the medium was replaced with fresh medium, and the infection was allowed to proceed for an additional 5 days. The supernatants were removed, and both the supernatants and infected cells (with 1.0 mL of medium) were frozen at −80°C. Following thawing, cell scrapers were used to disrupt the infected monolayer. After an additional freeze/thaw cycle, the cells were pelleted at 350 × g for 10 min. Freshly prepared CK cells were then infected with dilutions of these cell-free virus preparations. Virus stocks of the USDA reference strain of ILTV and viruses generated from cosmids/YCp recombinants were obtained by infecting CK cells with a MOI of 0.001. After 6 dpi, the infected cells were harvested, followed by 3 cycles of freeze/thawing and titered (see below).

The replication properties and spread of the reconstituted vBC, vKLO, vModKLO, and the USDA strain were analyzed in LMH cells. Briefly, LMH cells were seeded into 0.2% gelatin-treated 6-well plates, 5 × 105 cells in 3.0 mL of DMEM containing 10% FBS and antibiotics, and incubated at 39°C for 24 h. The medium was then removed, and the cells were inoculated with the recombinants or the USDA reference strain (MOI of 0.001). Viruses were absorbed for 60 min at 39°C. For the plaquing essay, the inoculum was removed after absorption, and the cells were rinsed once with media and overlaid with 1% wt/vol methylcellulose in DMEM, with 10% FBS, 50 μg/mL ampicillin, and 50 μg/mL gentamicin. Cells were incubated at 39°C for 5 days, fixed with 4% paraformaldehyde, and stained with 1% crystal violet (CV) in 20% ethanol and dH2O. All plaques were photographed from 6-well dishes, the plaque areas were outlined manually, and the areas were measured using ImageJ Fiji (https://imagej.net/software/fiji/downloads). Plaque diameters were determined from area calculations and compared to those obtained with USDA ILTV reference control. For the growth kinetics experiments, cells were infected with the same MOI of 0.001, and supernatants from 3 independent wells of each virus were collected at 0, 6, 24, 48, 72, 96, and 120 h postinoculation for qPCR measurements. Virus titers were determined in LMH and CK cells and expressed as the mean log₁₀ TCID50 mL-1. The vBC, vKLO, and vModKLO recombinant viruses and the USDA strain utilized in the animal experiments were propagated and titered in CK cells. For the viral entry kinetics, after 60 min at 39°C virus adsorption, infected LMH cell monolayers were covered with methyl-cellulose overlay media (1% wt/vol methylcellulose in DMEM, with 10% FBS, 50 μg/mL ampicillin, and 50 μg/mL gentamicin), and the number of plaques formed after inoculation (2.5, 7.5, 15, 30, and and 50 min) was compared to the number of plaques formed after an inoculation period of 60 min (endpoint). Copy numbers per 5 μL template were determined using Real-time qPCR as previously described (62). Further details on the in vitro multi-step virus growth and entry kinetics can be found in Devlin et al. (63) and Lee et al. (64).

The objective of the first experiment was to determine the virulence of the 3 reconstituted viruses in specific pathogen-free (SPF) chickens compared to the USDA ILTV reference strain. Briefly, 75 1-day-old SPF chickens (Merial Select) were distributed into 5 groups of 15 birds and placed in colony houses (Poultry Diagnostic Research Center, Athens, GA). At 3 weeks of age, 4 of the 5 groups were inoculated intratracheally (100 μL) and in the eye conjunctiva (50 μL each eye) with a dose of 3.5 (log₁₀) TCID50 per chicken of the vBC, vKLO, and vModKLO recombinants or the USDA strain. The remaining 15 chickens were mock-inoculated with tissue culture media. On days 3, 5, and 7 postinoculation, tracheal swabs were collected from each group of chickens and analyzed by real-time PCR to determine the trachea virus load.

Clinical signs of conjunctivitis, respiratory distress, lethargy, and mortality were scored from days 3 to 6 postinoculation as described by Vagnozzi et al. (65). Clinical sign categories of conjunctivitis, respiratory distress, and lethargy were evaluated in individual chickens and given a score. No clinical signs were given a score of 0; mild, a score of 1; moderate, a score of 2; and severe, a score of 3. In the case of mortality, a score of 6 was given. Each chicken received a total clinical sign score, and a mean clinical sign score was calculated at each time point. At the time point where the peak of the clinical signs was observed, the median clinical sign score for each group was compared.

Tracheal swabs were placed in 2 mL tubes containing 1X PBS solution with 2% antibiotic-antimycotic (100X, Invitrogen) and 2% newborn calf serum. Swabs were stored at −80°C until processed. Viral DNA extraction from tracheal samples was performed using the MagaZorbH DNA mini prep 96-well kit (Promega), as described by Johnson et al. (66).

The ILTV viral load in each of the tracheal swab samples was quantified by real-time PCR (qPCR) in a duplex assay normalized to a host gene as described by Vagnozzi et al. (67). The amount of ILTV DNA was measured by the qPCR assay on an Applied Biosystems 7500 Fast real-time PCR system (Life Technologies) with specific primers and probes of the U_L44 gene of ILTV (encoding glycoprotein C) and the endogenous control gene (avian α2-collagen gene) as described previously (68). Briefly, the duplex reactions for ILTV were set up to a final volume of 25 μL as follows: 12.5 μL of 2x master mix (TaqMan universal master mix II with UNG, Applied Biosystems), 1.25 μL of collagen primers to a final concentration of 0.5 μM, 1.25 μL of ILTV primers to a final concentration of 0.5 μM, 1.25 μL probes to a final concentration of 0.1 μM, and 5 μL of DNA template. In both PCR methods, the thermal cycling profile used was 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 60 sec. The relative amount of ILTV genomes detected per sample was expressed as the log₁₀ 2-ΔΔCt (67, 69).

Plaque size data of ILTV recombinant viruses were analyzed as plaque diameters using one-way analysis of variance (ANOVA), with Bonferroni correction for multiple comparison in the case of a P value of <0.01. The two-way ANOVA was utilized to highlight differences in mean viral titers of vBC, vKLO, vModKLO, and its parental USDA strain when grown in LMH cells. Genome copy numbers were compared at 0, 24, 48, 72, 96, and 120 h postinoculation in vitro. Error bars indicate SD from 3 replicates. P values < 0.05 were considered significant. The one-way ANOVA Kruskal–Wallis test (P < 0.05) was used to highlight differences in: (i) median clinical signs scores among groups of chickens inoculated via the intratracheal route with vBC, vKLO, vModKLO, and the USDA strain; (ii) mean viral genome load in trachea from chickens that received the recombinants and the USDA strain via the intratracheal-ocular route. Statistical analysis was performed using GraphPad PRISM 6.0 software (GraphPad Software).

All animal experiments described in this study were performed under the Animal Use Protocol AUP A2016-10-010-Y2-A2 approved by the Animal Care and Use Committee (IACUC) in accordance with regulations of the Office of the Vice President for Research at the University of Georgia.