BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, researchlibraries, and research funders in the common goal of maximizing access to critical research.

In Ovo Vaccination of Commercial Broilers with a Glycoprotein J Gene-DeletedStrain of Infectious Laryngotracheitis VirusAuthor(s): Anna Mashchenko, Sylva M. Riblet, Guillermo Zavala, and Maricarmen GarcíaSource: Avian Diseases, 57(2s1):523-531. 2013.Published By: American Association of Avian PathologistsDOI: http://dx.doi.org/10.1637/10413-100512-Reg.1URL: http://www.bioone.org/doi/full/10.1637/10413-100512-Reg.1

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, andenvironmental sciences. BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies, associations, museums, institutions, and presses.

Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance ofBioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.

Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercialinquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

In Ovo Vaccination of Commercial Broilers with a Glycoprotein J Gene-Deleted Strainof Infectious Laryngotracheitis Virus

Anna Mashchenko, Sylva M. Riblet, Guillermo Zavala, and Maricarmen GarcıaA

Poultry Diagnostic and Research Center, Department of Population Health, College of Veterinary Medicine, University of Georgia,Athens, GA 30602

Received 5 October 2012; Accepted 6 February 2013; Published ahead of print 9 February 2013

SUMMARY. Conventional live attenuated vaccines have been used as the main tool worldwide for the control of infectiouslaryngotracheitis. However, their suboptimal attenuation combined with poor mass administration practices allowed chickenembryo origin vaccine-derived isolates to circulate in the field, regain virulence, and be the cause of continuous outbreaks of thedisease. Previous studies indicated that stable attenuation of infectious laryngotracheitis virus (ILTV) can be achieved by thedeletion of individual viral genes that are not essential for viral replication in vitro. One of these genes is the glycoprotein J (gJ)gene. Its deletion provided significant attenuation to virulent ILTV strains from Europe and the United States. The objective of thisstudy was to construct an attenuated gJ-deleted ILTV strain and evaluate its safety and efficacy for in ovo (IO) administration ofcommercial broilers. A novel gJ-deleted virus (NDgJ) was constructed, and a 103 median tissue culture infective dose administeredat 18 days of embryo age was considered safe because it did not affect hatchability or survivability of chickens during the first weekposthatch. Broilers vaccinated IO and IO + eye drop at 14 days of age presented a significant reduction in clinical signs andreduction of virus loads after challenge, as compared with the nonvaccinated challenged group of chickens. Therefore, this studypresents initial proof that the NDgJ strain is a potential ILTV live-attenuated vaccine candidate suitable for IO vaccination ofcommercial broilers.

RESUMEN. Vacunacion in ovo de pollos de engorde comerciales con una cepa del virus de la laringotraqueıtis infecciosa aviarque presenta delecion del gene de la glicoproteına J.

Las vacunas vivas atenuadas convencionales han sido utilizadas como la herramienta principal en todo el mundo para el controlde la laringotraqueıtis infecciosa. Sin embargo, su atenuacion suboptima combinada con malas practicas de administracionmasiva permitio que derivados de la vacuna con origen en embrion de pollo circularan en el campo, que recuperaran suvirulencia, y que fueran la causa de los brotes continuos de la enfermedad. Estudios anteriores indicaron que la atenuacion establedel virus de la laringotraqueıtis infecciosa (ILTV) se puede lograr mediante la eliminacion de cada uno de los genes virales que noson esenciales para la replicacion viral in vitro. Uno de estos genes, es el de la glicoproteına J (gJ). Su supresion ha ofrecido unaatenuacion significativa de cepas virulentas del virus de laringotraqueıtis de Europa y de los Estados Unidos. El objetivo de esteestudio fue construir una vacuna atenuada con supresion del gene gJ y evaluar su seguridad y eficacia para una administracion inovo (IO) en pollos de engorde comerciales. Se construyo un virus nuevo con el gene gJ eliminado (NDgJ) y se considero que unadosis media de 103 dosis infectantes para cultivos de tejidos administrada a los 18 dıas de desarrollo embrionario fue seguradebido a que no afecto la incubabilidad o la supervivencia de los pollos durante la primera semana despues de la eclosion. Lospollos vacunados in ovo o vacunados in ovo y con una vacunacion ocular a los 14 dıas de edad, mostraron una reduccionsignificativa de los signos clınicos y una reduccion de las cargas virales en la traquea y en la conjuntiva despues del desafıo encomparacion con el grupo de pollos no vacunados y que fueron desafiados. Por lo tanto, este estudio presenta una prueba inicialde que la cepa NDgJ es un candidato potencial de una vacuna viva atenuada contra la laringotraqueıtis que es adecuada para lavacunacion in ovo de pollos de engorde.

Key words: infectious laryngotracheitis virus, glycoprotein J, in ovo vaccination

Abbreviations: CAM 5 chorioallantoic membrane; CEO 5 chicken embryo origin; CMV 5 cytomegalovirus; CSS 5 clinicalsign score; DMEM 5 Dulbecco’s minimal essential media; DPC 5 days postchallenge; DPH 5 days posthatch; ED 5 eye drop;GCN 5 genome copy number; gD 5 glycoprotein D; GFP 5 green fluorescent protein; gJ 5 glycoprotein J; GDgJ 5 green Dglycoprotein J; IBDV 5 infectious bursal disease virus; IBV 5 infectious bronchitis virus; ILT 5 infectious laryngotracheitis;ILTV 5 infectious laryngotracheitis virus; IO 5 in ovo; LMH 5 leghorn male hepatoma; NDV 5 Newcastle disease virus;nt 5 nucleotide; NVx 5 nonvaccinated; NVx-Ch 5 nonvaccinated-challenged; NVx-NCh 5 nonvaccinated-nonchallenged;NDgJ 5 novel gJ-deleted virus; ORF 5 open reading frame; PBS 5 phosphate-buffered saline; SPF 5 specific-pathogen-free;TCID50 5 50% tissue culture infectious dose; TCO 5 tissue culture origin; USDA 5 U.S. Department of Agriculture;Vx 5 vaccinated; Vx-Ch 5 vaccinated-challenged

Infectious laryngotracheitis virus (ILTV) is a member of the genusIltovirus, family Herpesviridae, subfamily Alphaherpesvirinae. ILTVis taxonomically classified as Gallid herpesvirus 1 (5), commonlyknown as infectious laryngotracheitis virus (ILTV), the causativeagent of the respiratory disease infectious laryngotracheitis (ILT) ofchickens. ILT is spread worldwide, and epidemics of the disease have

a devastating impact, particularly in areas of dense poultryproduction. The virus causes severe production loss due to increasedmortality, decreased egg production, delayed body weight gain, andpredisposition to other respiratory pathogens (12). The control ofthe disease is based on vaccination and biosecurity. During the past50 yr, live vaccines, attenuated by multiple passages in embryonatedchicken eggs (chicken embryo origin [CEO]) and in tissue culture(tissue culture origin [TCO]), have been used for the prevention ofILT in chickens (12). Both the CEO and TCO live attenuatedACorresponding author. E-mail: mcgarcia@uga.edu

AVIAN DISEASES 57:523–531, 2013

523

vaccines induce protection (1) and prevent clinical signs andmortality (13,14), and, in particular, the CEO vaccines significantlyreduce shedding of the challenge virus (26,31). However, both vaccineviruses persist in apparently healthy birds (1,15) and are able to spreadfrom bird to bird (13,26). Uncontrolled circulation of vaccines in thefield results in increased virulence, and in the case of the CEO vaccines,causes severe respiratory disease and mortality (11). The CEO vaccine isadministered via the drinking water or by coarse spray to commerciallayers and to broiler flocks only during outbreaks of the disease.However, if application of the vaccine is not performed properly, a largeproportion of the flock will not be exposed to the virus; consequently,the flock will not develop protective immunity, and the vaccine virus willbe allowed to circulate (27). The optimal delivery method forconventional live attenuated vaccines is via eye drop (ED) because, incontrast to water and spray vaccinations, it ensures uniformimmunization of the flock; however, its elevated manpower cost limitsthis practice to breeder flocks (8). Suboptimal attenuation and reducedsafety of the CEO vaccines encouraged the broiler industry to use newlyviral vector ILT vaccines. Two viral vector vaccines, one using thefowlpox virus, carrying ILTV glycoprotein B and UL34 genes, and theother using herpesvirus of turkeys, carrying ILTV glycoproteins I and Dgenes, have been used by the broiler industry via the in ovo route.Experimentally, viral vector vaccines do improve bird performance afterchallenge but do not reduce shedding of the challenge virus (16,31). InILT endemic areas, viral vector vaccines have not been efficient to detainthe spread of the disease (25). Therefore, a significant need exists forimproving the protective immunity, safety, and mass applicationmethods of ILTV vaccines. The functional redundancy of herpesvirusgenes permits the deletion of nonessential genes to produce stableattenuated viral phenotypes without affecting their replication in vitro.Previous studies have indicated that attenuation of ILTV can be achievedby the deletion of the surface glycoprotein J gene (gJ) (7). Although notessential for viral replication in vitro, the deletion of the gJ gene from theU.S. Department of Agriculture (USDA) virulent strain (BDgJ)significantly reduced its replication in the chorioallantoic membrane(CAM) of 10-day-old specific-pathogen-free (SPF) embryos (21).Production of ILTV live attenuated vaccines is contingent on theeffective propagation of these viruses in the CAM of SPF embryos.Therefore, the BDgJ strain was not a suitable ILTV vaccine candidate.

IO vaccination is becoming the preferred method of massvaccination, because it offers uniform vaccination coverage, fastdelivery, and reduces the stress on birds by multiple fieldvaccinations, and it is less costly than field vaccinations (24). Anincreasing number of studies have demonstrated the safety andefficiency of administering live attenuated vaccines by the in ovoroute. Some examples of live vaccines that have been evaluated forthe IO route inoculation are Newcastle disease virus (NDV)(18,22,23), infectious bursal disease virus (IBDV) (4,9), infectiousbronchitis virus (IBV) (2,29), or a combination of Marek’s diseasevirus, IBV, and IBDV (30). Recently, IO vaccination with aglycoprotein G gene-deleted strain of ILTV has been reported to besafe and efficacious for SPF chickens (20). The objectives of thisstudy were to 1) construct a novel attenuated gJ deletion strain of

ILTV that can reach titers comparable to those of the CEO vaccinewhen propagated in SPF embryos; 2) to evaluate the safety andefficacy of the novel gJ-deleted strain when administered IO tocommercial broilers.

MATERIALS AND METHODS

Generation of ILT novel gJ-deleted virus (NDgJ) virus. The parentstrain used to generate the NDgJ was previously described (21). This wasthe green D gJ (GDgJ) strain in which the gJ expression was inactivatedby insertion of the cytomegalovirus (CMV) promoter and the enhancedgreen fluorescent protein (EGFP) cassette (Fig. 1a(i)). For the design ofthe NDgJ virus, the full genome sequence of the ILTV GenBankaccession number NC_006623.1 was used. The open reading frame(ORF) for glycoprotein D (gD) US6 was predicted to be fromnucleotide (nt) 132675 to 133808, with a 12 nt overlap (129739 to132696) with the ORF of gJ US5. In the USDA wild-type virulentstrain, the transcription start site for gD has not been mapped; however,it was assumed that the US6 ORF overlapped the US5 starting at nt132504 as indicated by the wild type (Fig. 1a(ii)). A synthetic DNA(GenScript, Piscataway, NJ) was designed containing a markedly alteredUS5 ORF together with an unmodified 321-bp fragment located at the39 end of US5 and the putative start codon and promoter region of US6.The US5 coding sequence (CDS) was altered by randomly deleting10-bp sections and inserting them downstream and by introducingapproximately 50 CG and AT exchanges that resulted in a nonsensesequence but without changing the original GC content of the US5ORF. For homologous recombination with the GDgJ genome, authenticsequences were ligated to the 59 and 39 ends of the synthetic DNA. The59 end homologous recombination arm 479 bp was comprised of 270 ntof the US4 and 209 nt corresponding to the sequence located betweenthe stop codon of US4 and the disrupted US5 start codon. The 450 bp39 end homologous recombination arm comprised the nonoverlappingpart of US6 (Fig. 1a(iii)). The resulting 3876-bp DNA fragment waspurified and cloned in the bacterial plasmid pUC57 (GenScript). Therecombinant plasmid was digested with HindIII and EcoRI andseparated by agarose gel electrophoresis. The 3876-bp insert wascotransfected with GDgJ viral DNA (Fig. 1a(i)) into 80% confluentleghorn male hepatoma (LMH) cells (19) using the TransITH-2020transfection reagent (Roche, San Francisco, CA). Five days aftertransfection, cultures were inspected by light and fluorescencemicroscopy for cytopathic effect characteristic of ILTV, and viralplaques lacking green fluorescent expression (black plaques) wereaspirated under visual inspection. Selected plaques were subjected toat least three rounds of viral plaque purification on LMH cells to ensurethat no GDgJ (EGFP-expressing virus) virus was present. To confirm thecorrect construction of the NDgJ (Fig. 1a(iv)) by PCR, plaque-purifiedviruses were propagated in LMH cells. Five days postincubation, totalDNA was extracted from infected LMH cultures and analyzed by PCR.

Confirmation of NDgJ virus by PCR analysis. Three plaque-purifiedNDgJ isolates were analyzed by PCR assays. One PCR assay used the primerpair 25upUS4fw (59-TCA GCTCGA AGT CTG AAG AG-30) andClaIgIrev (59-CAG AAG ACG ATC GAT GAG TGC-39) that binds toregions within the viral genome that lie outside of the recombinationregion. The second PCR assay used the primer pair NgJ1390fw (59-CGGCACTGAAACTGAGCCGC-39) and NgJ2483rev (59-GGCAA-TGCTCTCTGCCTCCG-39) that binds to the markedly altered sequence

R

Fig. 1. Generation and confirmation of NDgJ virus. (a) Schematic diagram of steps necessary for the development of NDgJ virus. (i) Sph I 7958-bp fragment of the GDgJ virus US region containing the GFP-expression cassette instead of the first 2291 bp of the US5 CDS; (ii) Sph I 7958-bpfragment encompassing the ORF’s US4, US5, US6, and US7 from the US region of the wild-type strain (USDA); (iii) recombinant plasmidcontaining the altered US5 sequence flanked by the 59 end homologous recombination arm (479 bp) and the 39 end homologous recombination arm(771 bp); (iv) Sph I 7958-bp fragment of the NDgJ virus US region containing the altered 2357-bp sequence followed by the unaltered US5 321 bp

524 A. Mashchenko et al.

in the 39 end; (v) Primer pair designed to amplify from US4 (25upUS4FW) to upstream of the US7 gene (ClaIgIrev) that should amplify a 5591-bpfragment for both the USDA wild-type strain and the NDgJ. Primer pair (NgJ1390fw/NgJ2483rev) designed to amplify a 1112-bp fragment of thealtered US5 ORF sequence present only on the NDgJ virus. (b) PCR analysis of the USDA wild-type stain and the NDgJ virus. Lanes 1, 2, and 3show amplification performed with primer pair 25upUS4fw/ClaIgIrev on total DNA isolated from NDgJ virus-infected LMH cells, wild-typeUSDA virus-infected LMH cells, and uninfected LMH cells, respectively. As expected, a PCR product of 5591 bp was observed for the amplificationof the NDgJ and USDA-infected cells (lanes 1 and 2). Lane 4, molecular weight marker with bands of 10.0, 8.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.5, 1.0, and0.5 kb. Lanes 5, 6, and 7 show amplification performed with primer pair NgJ1390fw/NgJ2483rev on total DNA isolated from NDgJ virus-infectedLMH cells, wild-type USDA virus-infected LMH cells, and uninfected LMH cells, respectively. As expected, a PCR product 1112 bp was onlyobserved for the amplification of the NDgJ virus.

In ovo vaccination with gene-deleted ILTV 525

that replaced the US5 gene (Fig. 1e). All amplifications were performedwith GoTaqH DNA Polymerase (Promega, Madison, WI). Eachamplification reaction was performed in a 25-ml volume, containing250 mM of deoxynucleotide triphosphates, 3 mM of MgCl2, 50 mM of eachprimer, 1 U Taq polymerase, 10 ml of buffer, and 1 ml of template DNA.Amplification reactions for both sets of primers used an initial denaturingstep at 97 C for 2 min, followed by 30 amplification cycles of 94 C for30 sec, annealing temperature of 55 C for 30 sec and extension at 72 C for2 min, and a final extension at 72 C for 10 min. PCR product sizes wereevaluated after electrophoresis in 1% agarose gels stained with ethidiumbromide and compared with 1-Kb molecular weight DNA ladder (NewEngland Biolabs, Ipswich, MA) after exposure to ultraviolet light forvisualization.

Chicken embryo passages of NDgJ virus. To determine if the NDgJvirus can be propagated in the CAM of chicken embryos to sufficienttiters, the virus propagated in LMH cells was passaged 10 times inembryonated SPF chicken embryos and viral titers determined for eachpassage. Briefly, eight embryos were inoculated with 0.1 ml of the virussuspension diluted 1:10 in Dulbecco’s minimal essential media(DMEM) containing 1% antibiotic-antimycotic 1003 (Gibco, GrandIsland, NY) onto the CAM using a 25-gauge, 5/8-inch needle. Five to6 days postincubation the CAMs were removed from embryonated eggs.Areas of necrosis and thickness within each CAM were collected, washedin cold phosphate-buffered saline (PBS), and homogenized. Homoge-nates were frozen (220 C) and fast thawed (37 C) twice, thencentrifuged at 5000 3 g for 20 min. The cleared supernatant wastitrated in chicken kidney cells in 96-well plates as previously describedby Rodriguez-Avila et al. (26). Viral titers were expressed as the 50%tissue culture infectious dose per milliliter (TCID50/ml) and estimatedby the Reed and Muench method. Each passage was titrated in triplicateand presented as the average titer.

IO vaccination, hatchability, mortality, and weight gain. Threehundred and twenty-five broilers eggs from a local commercial sourcewere incubated in a small-scale hatchery (NatureForm Inc., Jacksonville,FL) from 0 to 21 days at 37.5 C, 55% humidity, and turned hourly.Embryos were vaccinated (Vx) manually IO, at 18 days of age, with0.1 ml of the NDgJ delivered into the allantoic cavity using a 22-gauge,1K-inch needle. Briefly, a total of 289 embryos were distributed intofour groups: 106 were Vx with 103 TCID50 (group IONDgJ103Vx), 60were Vx with 104 TCID50 (group IONDgJ104Vx), and 60 with 105

TCID50 (group IONDgJ105Vx) per embryo of the NDgJ virus. Onegroup of 63 embryos was mock inoculated with 0.1 ml of DMEM.Hatchability was calculated as the percentage of hatched chickens overthe total number of IO inoculated embryos within each group. Afterhatch, all Vx and nonvaccinated (NVx) chickens were identified by wingtags, distributed in separated floor pens at the Poultry Diagnostic andResearch Center (University of Georgia, Athens, GA), and fed astandard diet and water ad libitum. During 34 days, chickens weremonitored twice daily; those showing severe signs of respiratory distresswere euthanatized as approved in our institution animal use proposal(A2012 07-016-Y1-A0). The number of euthanatized chickens wasrecorded as mortalities, and the percentage of mortality per group wasestimated using Kaplan-Meire survival analysis, including chickenseuthanatized at 0, 2, 4, and 6 days posthatch (DPH) for lung and tracheacollection. To estimate the percentage of weight gain, 12 chickens pertreatment (IONDgJ103Vx, IONDgJ104Vx, IONDgJ105Vx, and NVx)were weighed at 13 and 34 days of age. At 18 days of age, 12 chickenswithin the IONDgJ103 group were vaccinated a second time with a doseof 103 TCID50 of the NDgJ virus delivered via eye drop (ED) (groupIONDgJ103 + EDVx).

Replication of NDgJ after IO vaccination. To evaluate thereplication pattern of the NDgJ virus at 0, 2, 4, and 6 DPH, 5 to 10chickens were randomly selected per group. Chickens were euthanatizedby CO2 inhalation, and trachea and lung tissues were collected. Tissueswere individually homogenized in 1 ml of PBS containing 2%antibiotic-antimycotic 1003 (Gibco) by centrifugation in the FastPreptubes (MP Biomedical, Solon, OH). Samples were frozen at 280 Cuntil processing.

ILTV real-time PCR. Total DNA was extracted from trachea andlung tissues collected at 0 to 6 DPH and from trachea and conjunctivaswabs collected 5 days postchallenge (DPC). DNA extraction ofindividual samples was done using MagaZorbH DNA mini-prep96-well kit (Promega) following the manufacturer’s recommendations.

Challenge experiment. Only chickens within the IONDgJ103Vxand IONDgJ103 + EDVx treatment groups were challenged. Briefly, at34 days of age, 46 chickens were transferred to positive-pressureHorsfall-Bauer-type isolation units located at the Poultry DiagnosticResearch Center. Ten IONDgJ103Vx chickens were distributed into twounits, with five chickens per unit. Twelve IONDgJ103 + EDVx chickenswere distributed into two isolation units, with six chickens per unit.Twenty-four mock-inoculated (NVx) chickens were distributed in fourunits, with six chickens per unit. Ten IONDgJ103Vx, 12 IONDgJ103 +EDVx, and 12 NVx chickens were challenged with the virulent ILTVfield isolate 63140 (28) at a dose of 104 TCID50 per chicken. Eachchicken received 0.1 ml intratracheally, and 0.05 ml was administered inthe conjunctiva in each eye via ED. Twelve NVx chickens remainedunchallenged (NVx-NCh) to serve as negative controls. Protection wasevaluated by scoring clinical signs from 3, 4, 5, and 6 DPC, as previouslydescribed (31). Briefly, clinical signs for individual chickens were scoredbased on the severity of breathing patterns, depression, and conjunc-tivitis on a scale of normal [0], mild (1), moderate (2), and severe (3).The mean clinical sign scores (CSSs) per group were calculated daily,and differences in CSSs among groups were analyzed statistically. Viralloads in the trachea and conjunctiva of Vx and NVx chickens weredetermined at 5 DPC by real-time PCR (3). The challenge virus loadsdetected in Vx chickens were compared with loads detected in NVxchickens. A significant reduction in challenge virus loads in Vx chickengroups was considered the result of protection induced by vaccination.At 7 DPC, all chickens were humanely euthanatized.

Statistical analysis. The percentage of mortality posthatch per groupof chickens was estimated using Kaplan-Meire survival analysis,including chickens euthanatized for lung and trachea collection at 0,2, 4, and 6 DPH. Mean CSSs obtained at 4, 5, and 6 DPC from Vxgroups of chickens were compared with the mean CSSs obtained fromthe NVx challenge group of chickens using the Kruskal-Wallis test.Multiple pairwise comparisons were performed for post hoc analysis.Weight gain percentages and postchallenge genome copy number(GCN) were compared independently among groups using one-wayanalysis of variance. When significant differences were found at the 5%level of significance, Bonferroni’s method for multiple pairwisecomparisons was used to detect differences between pairs. Differenceswere considered significant at P , 0.05. Statistical tests were performedusing GraphPad Prism 5 software (GraphPad, San Diego, CA).

RESULTS

Generation, confirmation, and propagation of NDgJ virus.PCR analysis for one of the three plaque-purified NDgJ viruses isshown in Fig. 1b, as expected the 25upUS4fw/ClaIgIrev primer pairproduced a 5591-bp PCR product for the NDgJ and the USDAwild-type strain-infected cells. A PCR fragment of 112 bp resultedfrom the amplification with primer pair NgJ1390fw/NgJ2483rev forDNA extracted from the NDgJ-infected LMH cells but not from thewild-type USDA-infected cells (Fig. 1b). Sequencing analysis ofPCR products generated by primer pair 25upUS4fw/ClaIgIrevrevealed no nt changes at the homologous recombination arms of therescued NDgJ viruses and the correct insertion of the nonsensesequence (data not shown). The nonsense sequence replaced theCMV-EGFP cassette in the GDgJ virus (21), generating a marklessvirus devoid of gJ expression, as no expression is expected from theinserted nonsense sequence.

Chicken embryo passages of NDgJ virus. Mean titers of theNDgJ virus in CAM of SPF-embryonated eggs ranged from 4.0 to6.2 log 10 per milliliter. Working stocks derived from the fifth CAM

526 A. Mashchenko et al.

passage were used for IO vaccination of commercial broiler eggs at103, 104, and 105 TCID50 per egg.

Hatchability, mortality, and weight gain of NDgJ IOVx chickens. The hatchability of commercial broiler eggs for theIONDgJ103Vx group was 91% and for the IONDgJ104- andIONDgJ105-Vx groups of chickens was 93%, while hatchability forthe mock-inoculated (NVx) group of chickens was 92%. Therefore,no differences in hatchability were observed between Vx and NVxgroups (Table 1). Most of the mortalities recorded postvaccinationwere from chickens showing severe respiratory distress and wereimmediately euthanatized. Severe respiratory distress occurredduring the first week of age in some but not all the chickens, ratheronly a percentage of chickens within each Vx group were affected.The higher the NDgJ dose applied IO, the higher the percentages of

chickens that were affected with respiratory problems during the firstweek of age. Mortalities of 1.7%, 2.0%, 21%, and 29% wereobserved for the NVx group, the IONDgJ103Vx group, theIONDgJ104Vx group, and the IONDgJ105Vx group of chickens,respectively (Table 1). Lungs and tracheal tissues were collected fromchickens showing severe respiratory distress during the first week ofage and analyzed by real-time PCR. All samples collected werepositive for ILTV, with GCNs ranging from 0.25 to 7.6 log 10 persample. Samples collected from the mortality within the mock-inoculated group were negative for ILTV DNA.

The mean weight gain (%) from 13 to 34 days of age forcommercial broilers within IONDgJ103Vx, IONDgJ104Vx,IONDgJ105Vx, and mock-inoculated (NVx) groups of chickens was79.98%, 80.90%, 81.57%, and 81.45%, respectively. No significantdifferences were found at the 5% level of significance for weightgained among groups of chickens between 13 to 34 days of age.

Replication of NDgJ virus. Viral replication of the NDgJ viruswas detected at 0 DPH in the lungs of chickens from all Vx groups(IONDgJ103, IONDgJ104, and IONDgJ105), with mean GCNranging from 3 to 4 log 10. The peak of NDgJ virus replication in thelung was detected at four DPH, with mean GCN ranging from 5.5to 6.9 log 10 (Fig. 2a). In contrast to the replication in the lung,limited replication was observed in the trachea at 0 DPH.Independently of the administered IO dose, the peak of viralreplication in the trachea was observed at four DPH, with mean

Table 1. Hatchability and mortality after IO vaccination withNDgJ virus.

GroupsA % HatchabilityB % MortalityC

IONDgJ103 91 (96/106) 2.0IONDgJ104 93 (56/60) 21.4IONDgJ105 93 (56/60) 28.6NVx (mock) 92 (58/63) 1.7

ATCID50/embryo.BNumber of hatch/total number set.CCalculated using survival analysis Kaplan-Meire.

Fig. 2. Replication of NDgJ virus. 2a. NDgJ viral loads expressed as the mean GCN log 10 (+/2SD) for lung samples: collected at 0 DPH(n 5 10) were 3.27(+/20.74), 4.92 (+/20.65), and 3.16 (+/21.07); at 2 DPH (n 5 10) were 1.10 (+/21.10), 6.18 (+/20.33), and 4.67 (+/20.83);at 4 DPH (n 5 5) were 5.48 (+/20.36), 6.89 (+/20.25), and 5.73 (+/20.51); at 6 DPH (n 5 5) were 1.02 (+/20.67), 0.99 (+/20.72), and 2.49(+/21.44) for IONDgJ103Vx, IONDgJ104Vx, and IONDgJ105Vx, respectively. Error bars indicate SD 2b. Mean NDgJ viral loads expressed asGCN log 10 from tracheas samples: collected at 0 DPH (n 5 10) were 0, 0.74 (+/20.50), and 0.48 (+/20.48); 2 DPH (n 5 10), 0.80 (+/20.80), 4.95(+/20.95), and 4.90 (+/20.45); 4 DPH (n 5 5), 5.27 (+/20.77), 6.74 (+/20.20), and 6.68 (+/20.22); 6 DPH (n 5 5), 2.08 (+/21.28), 1.37(+/21.37), and 0.47 (+/20.47) for IONDgJ103Vx, IONDgJ104Vx, and IONDgJ105Vx, respectively. Error bars indicate SD.

In ovo vaccination with gene-deleted ILTV 527

GNC ranging from 5.3 to 6.7 log 10 among groups. The peak ofviral replication in the lung coincided with the peak of viralreplication in the trachea. By 6 DPH, viral replication in the lungsand trachea decreased. Mean GCN levels ranging from 1.0 to 2.5log 10 were detected in the lungs (Fig. 2a), and 0.5 to 2.0 log 10

GCN were detected in the trachea (Fig. 2b).Clinical signs postchallenge. Because of the undesirable

mortality observed during the first week of age for theIONDgJ104Vx and IONDgJ105Vx groups of chickens, furtherassessment of protection was only performed for the IONDgJ103

and IONDgJ103 + EDVx groups of Vx chickens. Chickens from theIONDgJVx103, IONDgJ103 + EDVx, and NVx were challenged at35 days of age. The mean CSSs for the NVx-challenged (NVx-Ch)group of chickens were significantly higher than the mean CSSsrecorded for Vx groups of chickens (IONDgJ103 and IONDgJ103 +ED) at 3, 4, 5, and 6 DPC (Fig. 3a). The peak of clinical signs,defined as the day in which the NVx-Ch group of chickens showedthe highest CSSs postchallenge, was identified at 5 DPC. At thistime point, no significant differences in CSSs were identifiedbetween Vx groups and the NVx-NCh group of chickens, while theNVx-Ch group developed CSSs significantly higher than those ofthe Vx groups (Fig. 3b).

Viral loads postchallenge. Another parameter that was used toevaluate the protection elicited by IO vaccination with the NDgJ was

quantification of challenge virus loads in the trachea and conjunctivaat 5 DPC by real-time PCR. Viral loads were expressed as GCNlog 10 in trachea and conjunctiva swabs for individual chickenswithin each group (Fig. 4a). The mean viral loads detected intrachea and conjunctiva swabs are presented in Fig. 4b. Nosignificant differences were observed for tracheal viral loads betweenVx groups and the NVx-NCh group of chickens, and mean trachealviral loads of Vx groups were significantly lower than those detectedfor the NVx-Ch group. On the other hand, average viral loads fromconjunctiva swabs of Vx groups were significantly higher than thoseof the NVx-NCh group but significantly lower than those of theNVx-NCh group of chickens (Fig. 4b).

DISCUSSION

This study reports the safety and efficacy of IO vaccination ofcommercial broilers with an ILTV NDgJ. Independently of theNDgJ dose used, hatchability was not affected, as hatchability of theVx groups of chickens was not different in the hatchability of themock-inoculated group of chickens. Likewise, independently of theNDgJ dose used, the replication of the NDgJ virus during the firstweek of age posthatch was first detected in the lungs. The replicationof the NDgJ virus in the lung is likely due to the method of delivery

Fig. 3. CSSs. (a) Mean CSSs recorded at 3, 4, 5, and 6 DPC for chickens in IONDgJ103Vx-Ch (n 5 10), IONDgJ103+EDVx-Ch (n 5 12),NVx-Ch (n 5 12), and NVx-NCh (n 5 12) groups. Error bars indicate SEM. Mean CSS for the NVx-Ch group of chickens were significantlyhigher at 3, 4, 5, and 6 DPC (indicated with *) as compared with the mean CSS recorded for the IONDgJ103Vx-Ch and IONDgJ103 + EDVx-Chgroups of Vx chickens at the same time points postchallenge (P , 0.05). (b) At the peak of clinical signs (5 DPC), the mean score +/2the SEMrecorded for the NVx-Ch group of chickens was 6.92 (+/20.45), significantly higher than mean CSS for Vx groups of chickens IONDgJ103Vx-Chand IONDgJ103 + EDVx-Ch of 0.30 (+/20.15) and 0.33 (+/20.19), respectively. Statistically, differences among groups CSS (P , 0.05) areindicated by different letters and error bars indicate the SEM.

528 A. Mashchenko et al.

rather than the tropism of the NDgJ virus. When the virus isdelivered by IO vaccination, most of the virus is deposited in theamniotic sac. At 18 days of age, the embryo swallows and inhales theamniotic fluid; therefore, the virus will be first deposited in the lungsfollowed by replication in the trachea. In contrast to hatchability andreplication, marked differences in mortality were observed among103, 104, and 105 TCID50 NDgJ virus doses administered IO duringthe first week post-hatch. Two percent mortality was recorded forthe group of chickens Vx with a dose of 103 TCID50 per egg. Whilemortalities of 20% and 29% were recorded for chickens Vx with 104

and 105 TCID50 doses per egg. Although gJ deleted strains hadshown significant attenuation for adult chickens (7), mortalitiesobserved in this experiment during the first week of age suggests thatthe NDgJ virus retains some of its virulence for younger birds, andvirulence was dependent of the dose administered IO. Therefore, theuse of a safe dose is of essential importance for the success of the IO

vaccination with the NDgJ virus. Recent IO vaccination studiesusing a 102 TCID50 dose of the NDgJ virus showed very similarhatchability, mortality, and protection outcomes to those obtainedfor the 103 TCID50 dose (data not shown), indicating that a safedose for the IO vaccination with NDgJ virus ranges from 102 to 103

TCID50.The weight gain (%) by chickens between 13 to 34 days of age

was not affected by increased mortality observed during the firstweek of age, as mean weight gain within the IONDgJ103Vx,IONDgJ104Vx, and IONDgJ105Vx groups was not different fromthat of the NVx group of chickens. Despite suitable weight gainamong all Vx groups of chickens, only the IONDgJ103Vx groupwas considered safe, and consequently the protection elicited byvaccination was evaluated. Both Vx groups of chickens,IONDgJ103Vx-challenged (Vx-Ch) and IONDgJ103 + EDVx-Ch,showed a significant decrease in clinical signs at 3, 4, 5, and 6 DPC.

Fig. 4. Viral loads 5 DPC. Viral loads postchallenge expressed as GCN log 10. (a) Challenge virus loads in trachea and conjunctiva forindividual chickens. Challenge virus was detected in 1/10, 3/12, 12/12, and 0/12 trachea samples for IONDgJ103Vx-Ch, IONDgJ103 + EDVx-Ch, NVx-Ch, and NVx-NCh, groups of chickens, respectively. Challenge virus was detected in 9/10, 7/12, 12/12, and 0/12 conjunctiva samplesfor IONDgJ103Vx-Ch, IONDgJ103 + EDVx-Ch, NVx-Ch, and NVx-NCh, groups of chickens, respectively. (b) Mean challenge virus loads(GCN log 10 +/2SD) in trachea and conjunctiva for Vx and NVx groups of chickens. Mean GCN log 10 in trachea for IONDgJ103Vx-Ch,IONDgJ103 + EDVx-Ch, NVx-Ch, and NVx-NCh was 0.081 (+/20.280), 0.577 (+/21.01), 3.5 (+/20.692), and 0, respectively. Statistically,differences in trachea viral loads, (P , 0.05), among groups are indicated by different letters, and error bar indicate SD of the mean. Mean GCNlog 10 in conjunctiva for IONDgJ103Vx-Ch, IONDgJ103 + EDVx-Ch, NVx-Ch, and NVx-NCh was 2.063 (+/21.369), 1.43 (+/21.577), 4.53(+/20.750), and 0, respectively. Statistically, differences are indicated with the number of asterisks in the bar graph and error bars indicate the SDof the mean.

In ovo vaccination with gene-deleted ILTV 529

No differences were observed between protection levels elicited bysolely IO vaccination (IONDgJ103Vx) as compared with IOfollowed by ED vaccination (IONDgJ103 + EDVx). The protectioninduced by the NDgJ was further confirmed by a significant decreasein challenge virus loads in the trachea of Vx group of chickens ascompared with the NVx group. Similarly, a significant decrease inviral loads was detected in the conjunctiva of Vx groups; however,the decrease was not as marked as in the trachea. Although theprotection elicited by the NDgJ virus after IO vaccination wasrobust, around 8% to 20% of Vx chickens had viral loads similar tothose observed in NVx-Ch chickens. One possible reason for thehigh viral loads observed in 8% to 20% of the Vx chickens may bethat the presence of maternal antibodies interfered with thereplication of the NDgJ virus and consequently compromised thelevel of protective immunity in a portion of the Vx chickens. Broilersused in this study had ILTV maternal antibodies as measured byviral glycoprotein-specific ELISAs. Maternal antibodies peaked at 5DPH and declined to negligible levels 18 DPH (data not shown).Early studies demonstrated that maternal antibodies to ILTV aretransmitted to offspring (17), are detected in a low percentage of theprogeny (10), but do not interfere with vaccination in the field (6).Nonetheless, the interference of maternal antibodies with ILTV IOvaccination has not been addressed. A recent study has shown thatantigen-antibody complexed live B1-LaSota NDV vaccine admin-istered IO provided full immunity against phylogenetically distantvirulent viruses at an early age in maternal antibody-positivechickens (18). The use of antigen-antibody vaccine complexesdelays vaccine virus replication, consequently, evading the peak ofmaternal antibodies and improving flock protection and may be auseful strategy to further improve the protection elicited by IOvaccination with the NDgJ virus.

REFERENCES

1. Andreasen, J. R., Jr., J. R. Glisson, M. A. Goodwin, R. S. Resurreccion,P. Villegas, and J. Brown. Studies of infectious laryngotracheitis vaccines:immunity in layers. Avian Dis. 33:524–530. 1989.

2. Avakian, A. P., P. S. Wakenell, D. Grosse, C. E. Whitfill, and D.Link. Protective immunity to infectious bronchitis in broilers vaccinatedagainst Marek’s disease either in ovo or at hatch and against infectiousbronchitis at hatch. Avian Dis. 44:536–544. 2000.

3. Callison, S. A., S. M. Riblet, I. Oldoni, S. Sun, G. Zavala, S.Williams, R. Resurreccion, E. Spackman, and M. Garcıa. Development andvalidation of a real-time TaqmanH PCR assay for the detection andquantification of infectious laryngotracheitis virus in poultry. J. Virol.Methods 139:31–38. 2007.

4. Coletti, M., E. Del Rossi, M. P. Franciosini, F. Passamonti, G.Tacconi, and C. Marini. Efficacy and safety of an infectious bursal diseasevirus intermediate vaccine in ovo. Avian Dis. 45:1036–1043. 2001.

5. Davison, A. J. Herpesvirus systematics. Vet. Microbiol. 143:52–69. 2010.6. Dufour-Zavala, L. Epizootiology of infectious laryngotracheitis and

presentation of an industry control program. Avian Dis. 52:1–7. 2008.7. Fuchs, W., D. Wiesner, J. Veits, J. P. Teifke, and T. C. Mettenleiter. In

vitro and in vivo relevance of infectious laryngotracheitis virus gJ proteins that areexpressed from spliced and non-spliced mRNAs. J. Virol. 79:705–716. 2005.

8. Fulton, R. M., D. L. Schrader, and M. Will. Effect of route ofvaccination on the prevention of infectious laryngotracheitis in commercialegg-laying chickens. Avian Dis. 44:8–16. 2000.

9. Gagic, M., C. A. St. Hill, and J. M. Sharma. In ovo vaccination ofspecific-pathogen-free chickens with vaccines containing multiple agents.Avian Dis. 43:293–301. 1999.

10. Gharaibeh, S., K. Mahmoud, and M. Al-Natour. Field evaluation ofmaternal antibody transfer to a group of pathogens in meat-type chickens.Poult Sci. 87:1550–1555. 2008.

11. Guy, J. S., H. J. Barnes, and L. Smith. Increased virulence ofmodified-live infectious laryngotracheitis vaccine virus following bird-to-bird passage. Avian Dis. 35:348–355. 1991.

12. Guy, J., and M. Garcıa. Infectious laryngotracheitis virus. In: Diseasesof poultry, 12th ed. Y. M. Saif, J. R. Glisson, A. M. Fadly, L. R.McDougald, L. K. Nolan, and D. E. Swayne, eds. Blackwell Publishing,Ames, IA. pp. 137–152. 2008.

13. Han, M. G., and S. J. Kim. Efficacy of live virus vaccines againstinfectious laryngotracheitis assessed by polymerase chain reaction-restrictionfragment length polymorphism. Avian Dis. 47:261–271. 2003.

14. Hilbink, F., H. L. Oei, and D. J. van Roozelaar. Virulence of five livevaccines against avian infectious laryngotracheitis and their immunogenicityand spread after eye-drop or spray application. Vet. Q. 9:215–225. 1987.

15. Hughes, C. S., R. M. Gaskell, R. C. Jones, J. M. Bradbury, and F. T.Jordan. Effects of certain stress factors on the re-excretion of infectiouslaryngotracheitis virus from latently infected carrier birds. Res. Vet. Sci.46:274–276. 1989.

16. Johnson, D. I., A. Vagnozzi, F. Dorea, S. M. Riblet, A. Mundt, G.Zavala, and M. Garcıa. Protection against infectious laryngotracheitis virus(ILTV) by in ovo vaccination by commercially available viral vectorrecombinant vaccines. Avian Dis. 54:1251–1259. 2010.

17. Jordan, F. T. W. Immunity to infectious laryngotracheitis. In: AvianImmunology, M. E. Ross, L. N. Payne, and B. M. Freeman, eds.pp. 245–254. 1981.

18. Kapczynski, D. R., A. Martin, E. E. Hadad, and D. J. King.Protection from clinical disease against three highly virulent strains ofNewcastle disease virus after in ovo application of an antibody–antigencomplex vaccine in maternal antibody–positive chickens. Avian Dis.56:555–560. 2012.

19. Kawaguchi, T., K. Nomura, Y. Hirayama, and T. Kitagawa.Establishment and characterization of a chicken hepato-cellular carcinomacell line, LMH. Cancer Res. 47:4460–4464. 1987.

20. Legione, A. R., M. J. C. Coppo, S. W. Lee, A. H. Noormohammadi, C.A. Hartley, G. F. Browning, J. R. Gilkerson, D. O’Rourke, and J. M. Devlin.Safety and vaccine efficacy of a glycoprotein G deficient strain of infectiouslaryngotracheitis virus delivered in ovo. Vaccine 30:7193–7198. 2012.

21. Mundt, A., E. Mundt, R. J. Hogan, and M. Garcıa. Glycoprotein J ofinfectious laryngotracheitis virus is required for efficient egress of infectiousvirions from cells. J. Gen. Virol. 92:2586–2589. 2011.

22. Ohta, H., S. Ezoe, K. Yamazaki, T. Kawai, and T. Honda.Application of aluminum hydroxide for an in ovo live Newcastle diseasevaccine. Avian Dis. 53:392–395. 2009.

23. Ramp, K., E. Topfstedt, R. Wackerlin, D. Hoper, M. Ziller, T. C.Mettenleiter, C. Grund, and A. Romer-Oberdorfer. Pathogenicity andimmunogenicity of different recombinant Newcastle disease virus clone 30variants after in ovo vaccination. Avian Dis. 56:208–217. 2012.

24. Ricks, C. A., A. Avakian, T. Bryan, R. Gildersleeve, E. Haddad,R. Llich, S. King, L. Murray, P. Phelps, R. Poston, C. Whitfill, andC. Williams. In ovo vaccination technology. Adv. Vet. Med. 41:495–515.1999.

25. Ritter, G. D. Incidence and control of infectious laryngotracheitis(ILT) in North Carolina: 2007–2008 outbreak. In: Proc. 43rd NationalMeeting on Poultry Health and Processing, Ocean City, MD. pp. 62–67.2008.

26. Rodrıguez-Avila, A., I. Oldoni, S. M. Riblet, and M. Garcıa.Replication and transmission of live-attenuated infectious laryngotracheitisvirus (ILTV) vaccines. Avian Dis. 51:905–911. 2007.

27. Samberg, Y., E. Cuperstein, U. Bendheim, and I. Aronovici. Thedevelopment of a vaccine against avian infectious laryngotracheitis. IV.Immunization of chickens with a modified laryngotracheitis vaccine in thedrinking water. Avian Dis. 15:413–417. 1971.

28. Spatz, S., J. J. D. Volkening, C. L. Keeler, G. F. Kutish, S. M. Riblet,C. M. Boettger, K. F. Clark, L. Zsak, C. L. Afonso, E. S. Mundt, and M.Garcıa. Comparative full genome analysis of four infectious laryngotracheitisvirus (Gallid herpesvirus-1) virulent isolates from the United States. VirusGenes 44:273–285. 2012.

29. Tarpey, I., S. J. Orbell, P. Britton, R. Casais, T. Hodgson, F. Lin,E. Hogan, and D. Cavanagh. Safety and efficacy of an infectious bronchitisvirus used for chicken embryo vaccination. Vaccine 24:6830–6838. 2006.

530 A. Mashchenko et al.

30. Tarpey, I., A. A. van Loon, N. de Hass, P. J. Davis, S. Orbell,D. Cavanagh, P. Britton, R. Casais, P. Sondermeijer, and R. Sundick. Arecombinant turkey herpesvirus expressing chicken interleukin-2 increasesthe protection provided by in ovo vaccination with infectious bursal diseaseand infectious bronchitis virus. Vaccine 25:8529–8535. 2007.

31. Vagnozzi, A., G. Zavala, S. M. Riblet, A. Mundt, and M. Garcıa.Protection induced by infectious laryngotracheits virus (ILTV) live-attenuated

and recombinant viral vector vaccines in broilers. Avian Pathol. 41:21–31.2012.

ACKNOWLEDGMENT

This work was supported by Lohmann Animal Health GmbH & CoKG.

In ovo vaccination with gene-deleted ILTV 531