当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 病菌学杂志 > 2005年 > 第18期 > 正文
编号:11203114
Attenuation of Marek's Disease Virus by Deletion o
     Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, New York

    Viral Oncogenesis Group, Institute for Animal Health, Compton, Berkshire, United Kingdom

    ABSTRACT

    Marek's disease (MD) in chickens is caused by the alphaherpesvirus MD virus (MDV) and is characterized by the development of lymphoblastoid tumors in multiple organs. The recent identification and cloning of RLORF4 and the finding that four of six attenuated strains of MDV contained deletions within RLORF4 suggested that it is involved in the attenuation process of MDV. To assess the role of RLORF4 in MD pathogenesis, its coding sequence was deleted in the pRB-1B bacterial artificial chromosome clone. Additionally, RLORF5a was deleted separately to examine its importance for oncogenesis. The sizes of plaques produced by MDV reconstituted from pRB-1BRLORF5a (rRB-1BRLORF5a) were similar to those produced by the parental pRB-1B virus (rRB-1B). In contrast, virus reconstituted from pRB-1BRLORF4 (rRB-1BRLORF4) produced significantly larger plaques. Replication of the latter virus in cultured cells was higher than that of rRB-1B or rRB-1BRLORF5a using quantitative PCR (qPCR) assays. In vivo, both deletion mutants and rRB-1B replicated at comparable levels at 4, 7, and 10 days postinoculation (p.i.), as determined by virus isolation and qPCR assays. At 14 days p.i., the number of PFU of virus isolated from chickens infected with rRB-1BRLORF4 was comparable to that from chickens infected with highly attenuated RB-1B and significantly lower than that from rRB-1B-infected birds. The number of tumors and kinetics of tumor production in chickens infected with rRB-1BRLORF5a were similar to those of P2a chickens infected with rRB-1B. In stark contrast, none of the chickens inoculated with rRB-1BRLORF4 died up to 13 weeks p.i.; however, two chickens had tumors at the termination of the experiment. The data indicate that RLORF4 is involved in attenuation of MDV, although the function of RLORF4 is still unknown.

    INTRODUCTION

    Marek's disease (MD) is a lymphoproliferative disease in chickens caused by a member of the Alphaherpesvirinae subfamily, MD virus (MDV). The pathogenesis of infection with MDV can be divided into four phases (3). Bursa-derived (B) lymphocytes are infected between 3 and 6 days postinfection (p.i.) and are the primary target cells for the first phase of lytic infection, which is followed by infection of activated CD4+ thymus-derived (T) lymphocytes in the second phase. During the second phase, viral replication typically decreases and a latent infection is established between 5 and 10 days p.i. in activated CD4+ T lymphocytes. The third phase is characterized by reactivation of MDV replication between 14 and 21 days p.i. After the third phase, lymphomas may develop depending on the genetic susceptibility of the chickens and the virulence of the virus strain.

    The MDV genome consists of two unique regions, the unique short region (US) and the unique long region (UL), flanked by inverted repeats known as the inverted repeat long (IRL) and short (IRS), and the terminal repeat long (TRL) and short (TRS). The MDV Eco Q (Meq or RLORF7) and viral interleukin-8 (IL-8) homologue (vIL-8 or RLORF2) genes and the open reading frames (ORFs) RLORF5a (previously named ORF L1) and RLORF4 have been identified within the IRL and TRL regions of the MDV genome (see Fig. 1). Meq is a basic leucine zipper protein with an amino acid sequence similar to the proteins encoded by the jun and fos proto-oncogenes (12), has been shown to possess transformation potential when transfected into Rat-2 cells (18), and has been found in all MDV-transformed chicken cell (MDCC) lines studied thus far (12, 16). RLORF5a transcripts are expressed more abundantly in MDCC lines than in infected chick kidney cell (CKC) cultures (20). Deletion of this ORF in attenuated CVI988 indicated that it is not essential for virus replication and the establishment of latency (31). RLORF4 transcripts were recently shown to be expressed in MDCC lines and MDV-infected CKC cultures. Interestingly, in four of six cell culture attenuated MDV strains, deletions were identified within this ORF (10), although the authors erroneously referred to this ORF as LORF4 in their report. vIL-8 was originally identified through examination of splice variants of Meq (17, 23) and shares high similarity with cellular IL-8. Baculovirus-expressed MDV vIL-8 can function as a chemoattractant for peripheral blood mononuclear cells, although the classically described attraction of neutrophils was not observed (22).

    The use of direct mutagenesis of the MDV genome has become a powerful tool to identify specific genes important for MDV replication and oncogenesis. Previous studies examining the functional role of vIL-8 using deletion mutants showed that deletion of all three exons of vIL-8 decreases the ability to replicate in vivo; however, the virus was still able to become latent and cause tumors, albeit at a much lower incidence (5, 6, 22). Deletion of Meq in the very virulent MDV strain Md5 showed that the virus was still able to replicate in vivo but at lower levels than those of wild-type Md5, and tumors did not develop up to 8 weeks p.i. (19). Deletion of RLORF5a in the attenuated vaccine strain CVI988 indicated that it is not essential for virus replication and the establishment of latency; however, the effect on oncogenesis could not be determined using this strain of MDV (31).

    Attenuated MDV shows two major characteristic changes compared to nonattenuated strains. First, attenuated strains replicate more efficiently in vitro, causing larger plaques and increased viral titer and DNA replication, than nonattenuated strains do (41). Second, attenuated strains replicate much less efficiently in vivo than nonattenuated strains do (30, 41). In this report, we examined the phenotype and oncogenic potential of two mutant viruses with deletions in ORFs encoded in the IRL and TRL regions of the genome, rRB-1BRLORF5a and rRB-1BRLORF4. The recombinant viruses were generated by targeted deletion from the recently established bacterial artificial chromosome (BAC) clone of RB-1B, pRB-1B (24). Chickens infected with rRB-1BRLORF5a replicated and produced tumors similar to those of rRB-1B, while rRB-1BRLORF4 caused no MD-related deaths for up to 13 weeks p.i., although two chickens had tumors at this time. These data show that RLORF4 is involved in the attenuation of MDV as hypothesized; however, its function remains unknown.

    MATERIALS AND METHODS

    Cell cultures and virus strains. CKC cultures were prepared from 14-day-old specific-pathogen-free (SPF) chickens by standard methods (32) and used to propagate MDV strains RB-1B/p17 (RB-1B at passage 17) (29) and rRB-1B reconstituted from pRB-1B (clone 5) (24) and the mutant viruses rRB-1BRLORF5a (clone 1E1-1D1), rRB-1BRLORF4 (clones 2A1-2V1, 2A1-2V2, and 2A1-2V3), rRB-1BRLORF4+/– (2A1), and rRB-1BRLORF4+/– (2A8). Chicken embryo cell (CEC) cultures were prepared from 10-day-old SPF embryos by standard methods (32) and used to propagate the attenuated RB-1B/p71 strain (30) and for transfections.

    Construction of mutant BAC clones. Gene disruptions of RLORF5a and RLORF4 in pRB-1B were performed by the method of Datsenko and Wanner (7), with recombination between short homologous DNA regions catalyzed by the phage Red recombinase. Escherichia coli EL250 cells (15) harboring the red genes and the arabinose promoter-driven flp recombinase gene were kindly provided by Neal G. Copeland (National Cancer Institute, Frederick, MD). Briefly, the mutagenesis strategy was to replace the targeted gene with the kanamycin resistance gene (kanr) by homologous recombination. kanr, flanked by flp recognition target (FRT) sites from pKD13 (kindly provided by Barry L. Wanner, Purdue University), was amplified by PCR using primers with 50-base-pair extensions that were homologous to the start and end of the coding sequence of the gene to be disrupted. The sequences of the primers used for deletion of RLORF5a and RLORF4 were synthesized by Integrated DNA Technologies (Coralville, IA) and were as follows (MDV-specific sequences shown in capital letters): RLORF5a forward, 5'-AATATATACAGGGATGCACAGACATACTCCTATGCACCGATACACAGGCAcgtgtaggctggagctgcttc-3';RLORF5a reverse, 5'-TCTACAAAAGATCTAAATTTTACGTAATGGATCCCGTCCCGATCGTCCCCcattccggggatccgtcgac-3'; RLORF4 forward, 5'-TTTTTACCTTAGTGCGTTTGTTGACGGAAATGTCCGCCATTATTTGAACcgtgtaggctggagctgcttc-3'; and RLORF4 reverse, 5'-GGGCGATTTCCCTGTTATTGGGTCTCCACCAACACGTGATATTcattccggggatccgtcgac-3'. On the basis of the published sequence of MDV strain Md5 (36), nucleotides were deleted between 4,564 to 4,898 and 136,722 to 137,056 for RLORF5a and between 3,673 to 4,259 and 137,361 to 137,947 for RLORF4. The resulting PCR products were used to transform the recipient EL250 cells expressing Red recombinase using electroporation, and recombinant clones were isolated as kanamycin-resistant colonies as previously described (33). BAC DNA was isolated and examined for insertion of kanr into the right locus using restriction digestion mapping with NheI, BamHI, and HindIII enzymes. Since there are two copies of RLORF5a and RLORF4 in the MDV genome, one of each in the IRL and TRL regions, NheI digests were used to examine into which repeat long region the selectable marker was inserted. NheI cleaves the MDV genome such that the IRL-specific bands (25.2 kb) and TRL-specific bands (36.8 kb) can be readily distinguished. BamHI and HindIII restriction digests (data not shown) were used to evaluate the banding patterns of the mutated BAC clones to confirm the absence of spurious mutations elsewhere in the genome. Additionally, Southern blot analysis of each clone was performed using digoxigenin (DIG)-labeled probes for kanr, the BamHI-L fragment (vIL-8 gene), and the whole MDV genome.

    Once individual clones were examined and confirmed to lack spurious changes, kanr was excised by induction of FLP recombination by incubation in Luria-Bertani (LB) medium containing 0.02% arabinose (Sigma, St. Louis, MO) for 12 h. Bacteria were diluted 1:1,000,000 in LB medium and plated onto LB agar containing 30 μg/ml chloramphenicol. Individual colonies were restreaked onto LB agar with chloramphenicol and LB agar with chloramphenicol and kanamycin to confirm that individual colonies were no longer kanamycin resistant. By using this technique, 100% of colonies screened were chloramphenicol resistant and kanamycin susceptible (K. W. Jarosinski, unpublished observations). Small-scale BAC DNA preparations were performed by the E. coli alkaline lysis method previously described (27). Individual clones were again screened as described above for the presence of kanr, and restriction digest banding patterns were evaluated.

    This protocol was repeated for deletion of RLORF5a and RLORF4 in the remaining repeat to delete the second copy of each ORF. Once both copies were deleted and restriction enzyme banding patterns were proven to be correct, the regions were sequenced to confirm the absence of the ORF of interest. Unexpected sequences were not detected in this region, except for one copy of the 72-bp FRT sequence resulting from the recombination and FLP recombination process (7). Also, two clones with RLORF4 deleted in only one repeat long region (pRB-1BRLORF4+/– 2A1 and 2A8) were used in one animal experiment (experiment 5). Unless otherwise noted, rRB-1BRLORF5a and rRB-1BRLORF4 were reconstituted from clones 1E1-1D1 and 2A1-2V3, respectively.

    Recombinant viruses were reconstituted by transfecting CEC cultures with purified BAC DNA as previously described (33), and CEC cultures were passed onto CKC cultures at 7 days posttransfection. Each virus was then expanded in CKC cultures until stocks could be stored.

    Southern blot analysis. Probes used in Southern blot analysis were prepared using either the DIG-High Prime (MDV) or PCR DIG Probe Synthesis (vIL-8 and kanr) kits from Roche Applied Science according to the manufacturer's protocols. The complete MDV genomic probe was prepared using sonicated pRB-1B DNA that was randomly labeled with DIG (kindly provided by Sascha Trapp, Ithaca, NY). A DIG-labeled probe for the vIL-8 gene was used to identify the BamHI-L fragment and was prepared using previously published primers spanning the vIL-8 start and stop codons (10) and pRB-1B DNA as the template. The DIG-labeled kanr probe was prepared using kanr-specific primers (KanDIG forward, 5'-ACCCAGCCGGCCACAGTCG-3'; KanDIG reverse, 5'-GGGCGCCCGGTTCTTTTTG-3') with pKD13 as the template. For hybridization, DNA was separated by 0.8% agarose gel electrophoresis, gels were stained with ethidium bromide, and DNA was transferred to nitrocellulose membranes (Schleicher & Schuell, Keene, NH) according to the manufacturer's instructions.

    DNA extraction from tissues or cultured cells. Splenocyte preparations or CKC cultures infected with different strains of MDV were used for DNA extraction. Splenocytes were obtained by forcing spleens through a 60-μm-pore-size screen, resuspended in cold phosphate-buffered saline (PBS), and centrifuged on Ficoll-Paque Plus (Pharmacia, Piscataway, NJ). DNA was extracted from 1 x 107 splenocytes or infected CKC cultures using DNA STAT-60 (Tel-Test, Inc., Friendswood, TX) according to the manufacturer's instructions. Dried DNA pellets were dissolved in 100 μl DNase-free water, stored at 4°C, and used for quantitative PCR (qPCR) assays.

    qPCR assays. Quantification of MDV genomic copies using qPCR was performed as previously described (9, 45). Briefly, primers and probe specific for the MDV-infected cell protein 4 (ICP4) gene were used in qPCR assays. DNA loading for each sample was normalized using primers and probe specific for the chicken inducible nitric oxide synthase (iNOS) gene.

    For the generation of standard curves in qPCR assays, plasmids containing either the ICP4 or iNOS gene were used. Serial 10-fold dilutions of each respective plasmid were used for generating standard curves, starting with approximately 500 pg of plasmid. Total copy numbers were determined using the formula [(pg of input)(1 pmol/340 pg)(1/template size [in bp])(1 mol/1 x 1012 pmol)]/(6.02 x 1023 copies/mole) as previously described (10), and standard curves were generated by plotting the cycle threshold (CT) value at each dilution with the total copies. When creating standard curves, the y intercept was set at 40, since a CT value of 40 indicated no amplification of the specific target DNA. Therefore, any sample with a CT value of 40 had a value of 0 for the target sequence. The coefficient of regression was always >0.99 for standard curves.

    All qPCR assays were performed in an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Inc.), and the results were analyzed using Sequence Detection System v.1.9.1 software as previously described (9-11). Each qPCR reaction mixture contained TaqMan Universal Master Mix (Applied Biosystems, Inc.), 100 ng DNA, 25 pmol of each gene-specific primer, and 10 pmol of gene-specific probe in 25-μl volumes. Thermal cycling conditions were as follows: 50°C for 3 min, followed by 40 cycles at 94°C for 20 s and 60°C for 1 min. Real-time fluorescence measurements were performed as previously described (10, 11). Using the standard curve generated for each gene, the number of copies for ICP4 and iNOS DNA was determined by using the CT value for that sample. Because the MDV genome contains two copies of ICP4 and the chicken genome contains two copies of the iNOS gene, MDV genome copies were determined by dividing the number of ICP4 copies by the number of iNOS copies. This number was multiplied by 1,000,000 to create whole numbers, and the values were expressed as the number of MDV copies per 1 x 106 cells.

    Virus isolation assays. In order to determine the number of MDV-infected cells in chicken spleens, 5 x 105 splenocytes from individual birds were plated onto CKC cultures in triplicate. MDV-PFU were counted 5 days p.i., and the mean number of PFU per 1 x 106 cells was determined for each treatment group and compared using Student's t test.

    Measurement of plaque areas. After initial transfection of the mutant viruses into CKC cultures, it appeared that rRB-1BRLORF4 virus produced larger plaques than rRB-1B or rRB-1BRLORF5a do. Therefore, we examined plaque sizes more thoroughly by infecting cells seeded in six-well plates with 100 PFU per well. In the first experiment, rRB-1B/p5, rRB-1BRLORF5a/p5, and rRB-1BRLORF4/p5 (2A1-2V3) were compared, while in a second experiment, rRB-1B/p5, rRB-1BRLORF4/p3 (2A1-2V1), rRB-1BRLORF4/p3 (2A1-2V2), and rRB-1BRLORF4/p3 (2A1-2V3) were compared. Cells were fixed at 5 days p.i. with cold 90% acetone for 10 min and air dried. Cells were rehydrated in PBS for 10 min, PBS plus 10% fetal bovine serum was added as a blocking solution for 30 min, 0.5 ml of anti-MDV chicken sera obtained from MDV-infected SPF chickens was added to each well at a 1:1,000 dilution, and plates were incubated at room temperature for 30 min. Cells were then washed three times with PBS for 10 min, and goat anti-chicken immunoglobulin G (heavy and light chains) labeled with Alexa Fluor 568 (Molecular Probes, Inc., Eugene, OR) was added. After 30 min, cells were washed twice with PBS, and digital images of 65 individual plaques were obtained using an epifluorescence Axiovert 25 inverted microscope and an AxioCam HRc digital camera (Zeiss, San Marcos, CA). Plaque areas were measured using the National Institutes of Health ImageJ software (rsb.info.nih.gov/ij/), and means were determined for each virus.

    RNA extraction and reverse transcription-PCR (RT-PCR) assays. Total RNA was extracted from 60-mm dishes of rRB-1B-, rRB-1BRLORF5a-, or rRB-1BRLORF4-infected CKC cultures at 0, 3, and 6 days p.i. using RNA STAT-60 (Tel-Test, Inc., Friendswood, TX) according to the manufacturer's instructions. The dried RNA pellets were dissolved in 100 μl RNase-free water and DNase treated with the DNA-free system from Ambion, Inc. (Austin, TX) using the manufacturer's instructions.

    RT-PCR assays were performed using 1 μg total RNA using the GeneAmp Gold RNA PCR kit (Applied Biosystems, Inc., Foster City, CA) in 20-μl reaction mixtures as previously described (10). To amplify cDNA, 3 μl of the RT mixture was used with each pair of specific primers as previously described for RLORF5a (ORF L1) (44), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), RLORF4, and vIL-8 (10). The "touch-down" PCR procedure previously described was used to amplify specific cDNA products (43). PCRs were electrophoresed through 1.5% agarose gels, stained with ethidium bromide, and visualized using an Eagle EyeII still video system (Stratagene, La Jolla, CA).

    In vitro replication. In vitro replication of mutant viruses was measured over time by qPCR assays. In trial 1, DNA replication was assessed in CKC cultures that were infected with 100 PFU of rRB-1B/p5, rRB-1BRLORF5a/p5, or rRB-1BRLORF4/p5 (2A1-2V3). In trial 2, rRB-1B/p5, rRB-rRB-1BRLORF4/p3 (2A1-2V1), rRB-1BRLORF4/p3 (2A1-2V2), and rRB-1BRLORF4/p3 (2A1-2V3) were used. For each virus, DNA was obtained from three dishes at 24-h intervals from the time of plating (0 h) to 6 days p.i. and analyzed as described above.

    In vivo experiments. SPF P2a (major histocompatibility complex B19B19) chickens (38) were obtained from departmental flocks and housed in isolation units. Water and food were provided ad libitum. All experimental procedures were conducted in compliance with approved Institutional Animal Care and Use Committee (IACUC) protocols. In all experiments, chickens were inoculated intra-abdominally with 2,000 (experiment 1) or 1,000 (experiments 2 to 7) PFU of the various viruses, while control chickens were inoculated with uninfected CKC cultures. MDV does not spread horizontally between animals before 14 days p.i. (40); therefore, chickens infected for 14 days or less were housed together (experiments 2, 4, and 6). For experiments 1, 3, 5, and 7, all groups of chickens were housed in separate isolation units. Chickens were assigned to treatment groups using a randomized table.

    Experiment 1. In order to determine whether rRB-1BRLORF4 was able to replicate in vivo, 2-day-old P2a chickens were inoculated with uninfected CKC cultures (n = 20), rRB-1B/p5 (n = 20), or rRB-1BRLORF4/p5 (n = 20). In addition to inoculated chickens, an additional 10 sentinel chickens were housed with infected chickens to serve as contact controls. At 6 days p.i., splenocytes were collected from 10 chickens from each inoculated group, and virus isolation assays were performed on each sample. At 22 days p.i., splenocytes were collected from all remaining chickens, including contact control chickens, and virus isolation assays were performed.

    Experiment 2. Groups of 24 4-day-old chickens were inoculated with uninfected CKC cultures (control), RB-1B/p17, RB-1B/p71, rRB-1B/p6, rRB-1BRLORF5a/p6, or rRB-1BRLORF4/p6. At 4, 7, 10, and 14 days p.i., six chickens per group were used to evaluate thymic and bursal atrophy, as well as virus replication. Whole-bird body weights were measured prior to euthanasia. All thymus lobes and the bursa from each bird were weighed after collection, and the relative weights of the thymus and bursa to the whole body were determined. Spleens were collected, and splenocytes were used for virus isolation assays and DNA extraction.

    Experiment 3. To test the oncogenic potential of rRB-1BRLORF5a and rRB-1BRLORF4, 20 4-day-old chickens per group were inoculated with the same viruses as in experiment 2, but each group was housed in a separate isolation unit. Chickens were evaluated daily for symptoms of MD and were euthanized and examined for gross MD lesions when the birds showed clinical evidence of MD. Groups with significant early incidence of MD were terminated at 6 weeks p.i. The number of birds in groups without evidence of MD was reduced at 7 weeks p.i. to provide the required space/bird according to IACUC protocols. The remaining chickens were kept until 13 weeks p.i. when the experiment was terminated. Chickens euthanized at 7 and 13 weeks p.i. were examined for gross MD lesions.

    Experiment 4. To determine whether single deletions of RLORF4 would cause the same phenotype as double deletions, 24 4-day-old chickens per group were inoculated with uninfected CKC cultures (controls), rRB-1B/p5, rRB-1BRLORF4+/–/p4 (2A1), rRB-1BRLORF4+/–/p4 (2A8), or rRB-1BRLORF4/p5. At 4, 7, 10, and 14 days p.i., six chickens from each group were euthanized and splenocytes were used in virus isolation assays and DNA extraction.

    Experiment 5. Four-day-old chickens were inoculated with CKC cultures (n = 10), rRB-1B/p5 (n = 9), rRB-1BRLORF4+/–/p4 (2A1) (n = 9), rRB-1BRLORF4+/–/p4 (2A8) (n = 10), or rRB-1BRLORF4/p5 (n = 9) and housed in separate isolation units. Chickens were evaluated daily for symptoms of MD and sacrificed when clinical symptoms became apparent as in experiment 3. The experiment was terminated at 52 days p.i., since a majority of chickens in all groups, except uninfected controls and animals inoculated with rRB-1BRLORF4, showed clinical MD symptoms.

    Experiment 6. To determine whether the attenuated phenotype of rRB-1BRLORF4 clone 2A1-2V3 was clone specific, two additional clones were evaluated for in vivo replication. Chickens in each group (4 days old) were inoculated with uninfected CKC cultures (controls), rRB-1B/p5, rRB-1BRLORF4/p3 (2A1-2V1), rRB-1BRLORF4/p3 (2A1-2V2), or rRB-1BRLORF4/p3 (2A1-2V3). At 4, 7, 10, and 14 days p.i., four or five chickens from each group were euthanized and splenocytes were used in virus isolation assays and DNA extraction for use in qPCR assays.

    Experiment 7. To examine the oncogenic potential of three individual clones of RLORF4 deletion mutants, 10 chickens in each group of 4-day-old chickens were inoculated with the same viruses as in experiment 6. Chickens were evaluated as in experiments 3 and 5, and the experiment was terminated at 52 days p.i.

    Statistical analysis. Significant differences in the means for virus isolation, plaque size measurements, and qPCR assays were determined using either Student's t test or Tukey-Kramer comparison of means.

    RESULTS

    Construction of MDV deletion mutants. RLORF5a is located at the junction of the BamHI-Q2 and -L fragments, while RLORF4 is located solely in the BamHI-L fragment. There are two copies of each ORF, with one copy in both the IRL and TRL regions of the MDV genome (Fig. 1A and B). Using the virulent pRB-1B BAC clone (24), kanr was inserted into a single copy of RLORF5a and RLORF4 separately by Red recombination (Fig. 1C). The positive selection marker was subsequently removed by FLP recombination, which completely removed each respective ORF and left minimal residual sequences consisting of one FRT and a BamHI site. In order to delete both copies within the IRL and TRL, each ORF was deleted in one repeat and then the procedure was repeated to remove the second copy in the other repeat.

    Southern blot analysis of each clone during each step of the recombination/deletion process was performed following digestion of the BAC clones with NheI to distinguish each repeat long region (Fig. 1D). kanr was first inserted into RLORF5a (clone 1D, TRL; clone 1E, TRL) and RLORF4 (clone 2A, IRL; clone 2B, TRL) and then removed by FLP recombination (clone 1E1 for RLORF5a and clone 2A1 for RLORF4). The presence of bands in the respective clone that were reacting with a kanr-specific probe is shown in the bottom panel of Fig. 1D. After the first deletion of each respective ORF, kanr was inserted into the other repeat using the same method. Multiple clones were screened for insertion and excision, and only one clone for each deletion is shown for simplicity. kanr was inserted into RLORF5a within the IRL in the 1E1 clone (1E1-1D) and into RLORF4 within the TRL of the 2A1 clone (2A1-2V) and subsequently recombined out to produce the following double-deletion clones for RLORF5a (1E1-1D1) and RLORF4 (2A1-2V3), which were used for further experiments. In animal experiment 5, clones 2A1 and 2A8 (not shown in Fig. 1) that were produced by the exact same methodology were used for inoculation of chickens.

    Southern blot analysis after BamHI digestion of the engineered virus mutants was performed. Using a whole-genome, repeat-specific, or kanr probe, no apparent alterations were visible in either clone, apart from the expected deletions. Insertion of kanr and subsequent FLP-mediated removal of the kanr resulted in an additional BamHI site. Insertion of kanr into RLORF5a did not result in any discernible alterations in the restriction pattern, because the additional restriction enzyme site was inserted only 30 bp from the BamHI site, creating the Q2 and L fragments (Fig. 1E, top panel). Following deletion of RLORF4, the L fragment is divided into two fragments of 2,231 bp (L1) and 360 bp, although the latter fragment was not detected on the membrane since the probe for vIL-8 binds downstream of this fragment (Fig. 1E, top and middle panels). Insertion of kanr into RLORF5a led to an increase of the size of the Q2 fragment to 2,231 bp (shown as Q2a). Using the kanr probe, the Q2a and L2 fragments could be readily identified (Fig. 1E, bottom panel). HindIII restriction digests were also used to examine the integrity of the various mutant BAC clones, and no obvious spurious changes were seen (data not shown). The retention of the FRT sites after either ORF was replaced was further examined by nucleotide sequencing and proved to be correct (data not shown).

    Figure 2 shows mRNA expression of RLORF5a, RLORF4, and vIL-8 at 0, 3, and 6 days p.i. in infected CKC cultures. Deletion of RLORF5a and RLORF4 sequences completely abolished RLORF5a and RLORF4 mRNA expression, respectively, while vIL-8 mRNA expression was not affected in vitro.

    Deletion of RLORF4 results in increased plaque sizes and replication rates in CKC cultures. First, the replication of mutant viruses was evaluated by measuring the plaque areas and the level of DNA replication in vitro. Plaque sizes were determined for rRB-1B, rRB-1BRLORF5a, and rRB-1BRLORF4 (Fig. 3). Determination of the mean plaque areas demonstrated that there was no significant difference in the plaque areas produced by rRB-1B and rRB-1BRLORF5a, while the average size of rRB-1BRLORF4-induced plaques was significantly larger than those of rRB-1B and pRB-1BRLORF5a (Fig. 3A and B).

    Figure 3C shows the number of MDV genomic copies in CKC cultures infected with rRB-1B, rRB-1BRLORF5a, or rRB-1BRLORF4 using qPCR assays. At 72 h p.i. the number of genomic copies of rRB-1BRLORF4 was significantly higher and continued to increase at a faster rate than those of both rRB-1B and rRB-1BRLORF5a, consistent with the increased plaque size. From these experiments we concluded that deletion of RLORF4 in the pRB-1B clone resulted in increased in vitro replication of the reconstituted virus, at least in CKC cultures.

    In vivo replication of rRB-1BRLORF4 and rRB-1BRLORF5a. In a preliminary animal experiment, chickens infected with rRB-1B or rRB-1BRLORF4 were examined for virus replication at 6 and 22 days p.i. (experiment 1) using virus isolation assays. There was no significant difference between the two viruses with 0.60 ± 0.26 PFU/1 x 106 splenocytes isolated from chickens inoculated with rRB-1B and 0.26 ± 0.20 PFU/1 x 106 splenocytes demonstrated for rRB-1BRLORF4-infected chickens at 6 days p.i. However, at 22 days p.i., there was a significant difference between rRB-1B-infected chickens (216 ± 94 PFU/1 x 106 splenocytes) and rRB-rRB-1BRLORF4-inoculated chickens (14 ± 4 PFU/1 x 106 splenocytes). MDV was not isolated or detected using qPCR assays from contact chickens in both groups at 22 days p.i.

    In a second animal experiment, chickens were inoculated with the following viruses including uninfected controls: RB-1B/p17, RB-1B/p71 (attenuated), rRB-1B/p6, rRB-1BRLORF5a/p6, or rRB-1BRLORF4/p6. Table 1 summarizes the relative weights of prepared thymi and bursas for each group at 4, 7, 10, and 14 days p.i. Animals in all groups, except those infected with rRB-1BRLORF5a, had a significant increase in the relative thymus weight compared to uninfected chickens at 4 days p.i. Apart from RB-1B/p17, which caused thymic atrophy as evidenced by decreased thymus weights compared to uninfected controls at 7, 10, and 14 days p.i., only RB-1B/p71 and rRB-1B caused thymic atrophy at 10 days p.i. compared to uninfected chickens, while rRB-1BRLORF5a and rRB-1BRLORF4 did not cause thymic atrophy at any time point. Only chickens infected with RB-1B/p17 showed bursal atrophy at 14 days p.i.

    Also in experiment 2, splenocytes were collected and used for virus isolation assays (Fig. 4A) and for collection of DNA to determine viral DNA copy titers (Fig. 4B). Only RB-1B/p17 and RB-1B/p71 could be isolated from infected chickens at 4 days p.i. (Fig. 4A). At 4, 7, 10, and 14 days p.i., RB-1B/p17 was isolated at significantly higher levels than all other strains. In comparisons of the two pRB-1B-derived deletion mutants to the parental rRB-1B, only rRB-1BRLORF4 had significantly lower viremia levels at 14 days p.i. Similar results were obtained using qPCR assays detecting the number of viral genomic copies (Fig. 4B). These data indicate that the pRB-1B-derived virus exhibited a slightly delayed onset of viremia compared to parental RB-1B and that the RLORF4 deletion mutant appeared greatly debilitated with respect to lytic virus replication at 14 days p.i. (Fig. 4A and B).

    Deletion of RLORF4 reduces the oncogenic potential of rRB-1B. In order to examine the oncogenic potential of rRB-1BRLORF5a and rRB-1BLORF4, a third animal experiment was performed. All chickens infected with RB-1B/p17 succumbed to MD by 4 weeks p.i., while a majority of rRB-1B- and rRB-1BRLORF5a-infected chickens developed MD by 6 weeks (Fig. 4C). Therefore, all the remaining chickens in the rRB-1B- and rRB-1BRLORF5a-infected groups were euthanized at 6 weeks p.i. Because no clinical signs had occurred in animals infected with RB-1B/p71 and rRB-1BRLORF4, 10 animals in these two groups were evaluated for gross MD lesions at 7 weeks p.i. Postmortem examinations showed that none of the chickens infected with RB-1B/p71 or rRB-1BRLORF4 had developed MD lesions at that time point. From 7 to 13 weeks p.i., no chickens in the remaining groups developed MD symptoms and the experiment was terminated at 13 weeks p.i., when MD-specific lesions were found in two birds infected with rRB-1BRLORF4.

    rRB-1B with a single deletion of RLORF4 is phenotypically similar to rRB-1B in vivo. Since deleting both copies of RLORF4 in pRB-1B resulted in attenuation of the virus reconstituted from the recombinant BAC, viral replication of two independent clones still containing one of the two copies of RLORF4 was examined in a fourth animal experiment. Figure 5A and B show that viruses reconstituted from two independent clones harboring only one copy of RLORF4, 2A1 and 2A8, maintained virus replication rates similar to that of rRB-1B in vivo by virus isolation (Fig. 5A) and qPCR (Fig. 5B) assays. In contrast and as described in experiment 3 above, rRB-1BRLORF4 replicated at significantly lower levels at 14 days p.i.

    In the fifth animal experiment, the oncogenic potential of the single deletion mutants was evaluated in P2a chickens over a 52-day period. Animals infected with rRB-1B or single-deletion mutant 2A1 or 2A8 started to develop MD symptoms and gross lesions by 5 weeks p.i. (Fig. 5C), similar to the rRB-1B-infected chickens in experiment 3 (Fig. 4C). Again, chickens infected with rRB-1BRLORF4 did not develop MD symptoms or lesions for up to 52 days p.i.

    Phenotypic changes in rRB-1BRLORF are not clone specific. Since only one rRB-1BRLORF4 mutant clone was examined in experiments 1 to 5, two additional clones were evaluated both in vitro and in vivo. In vitro analysis showed that there were no significant differences between all three rRB-1BRLORF4 deletion mutants (2A1-2V1, 2A1-2V2, and 2A1-2V3) using plaque measurement and qPCR assays, while each was significantly different from rRB-1B (data not shown), similar to Fig. 3. Analysis of viral replication in vivo showed that using virus isolation and qPCR assays, all three independent rRB-1BRLORF4 clones replicated similarly (Fig. 6A and B). Replication was significantly lower at 14 days p.i. than with parental rRB-1B. Additionally, none of the rRB-1BRLORF4 clones induced MD lesions for up to 52 days p.i., while rRB-1B induced MD (Fig. 6C), consistent with experiments 3 and 5.

    DISCUSSION

    MDV is an oncogenic alphaherpesvirus that can be attenuated by serial passage in cultured cells (41). The previous finding that four of six attenuated strains of MDV examined had deletions within RLORF4 suggested that this ORF may be important for in vivo replication of MDV (10) because attenuated MDV does not replicate as well as nonattenuated MDV in vivo (30). However, attenuation of MDV by serial passage in CEC or CKC cultures can lead to multiple genomic changes, including expansion of the 132-bp tandem direct repeat region within the IRL and TRL regions (8, 25, 26), although it has been shown that this alteration does not directly cause attenuation (34). In addition, transcripts in the 1.8-kb transcript family that has been associated with the oncogenic potential of MDV (1, 2) are truncated in attenuated strains (1, 2, 14, 26, 28). Expression of glycoprotein C (gC) is greatly reduced after serial passage in vitro as well (4, 37, 39, 42). To test the hypothesis that deletion of RLORF4 may lead to attenuation of MDV, RLORF4 was directly deleted in the virulent pRB-1B BAC (24) clone by Red recombination.

    Following attenuation of MDV, phenotypic changes in the virus include decreased potential for in vivo replication and oncogenicity and increased in vitro replication rates and plaque sizes compared to those of nonattenuated strains. When both copies of RLORF4 were deleted in pRB-1B, the average plaque size of rRB-1BRLORF4 was almost double the sizes of plaques induced by the parental rRB-1B and a second mutant, rRB-1BRLORF5a (Fig. 3A and B). When qPCR assays were used to measure the number of MDV genomes per cell in infected CKC cultures, the number of MDV copies in rRB-1BRLORF4-infected cultures was significantly higher than those observed in rRB-1B- and rRB-1BRLORF5a-infected cultures after 72 h p.i. These data indicate that deleting RLORF4 leads to in vitro growth of MDV phenotypically associated with attenuated viruses. These data are very similar to recent observations for the absence or presence of gC in MDV. Deletion of the gC-encoding ORF, whose expression is down-regulated by serial passage of the virus in cultured cells, results in increased plaque sizes (35). As such, the increase in plaque size after deletion of RLORF4 is the second example of the deletion of an MDV ORF that resulted in an increase in virus-induced plaque size.

    When the rRB-1BRLORF4 mutant virus was examined in vivo by virus isolation and qPCR assays, there was little difference between rRB-1BRLORF4 and the parental rRB-1B virus early during infection (Fig. 4 to 6). Generally, the numbers of infectious virus and genomic copies detected in rRB-1BRLORF4-infected chickens were lower than those in rRB-1B-infected chickens, although these differences were not statistically significant until 14 days p.i. and later (experiment 1 and Fig. 4 to 6). The decreased replication in vivo satisfies the second major characteristic of attenuated MDV in that deleting RLORF4 led to decreased in vivo replication. Additionally, when the oncogenic potential of rRB-1BRLORF4 was examined, the level of MD incidence was reduced from 95% induction in P2a chickens infected with parental rRB-1B down to 10% in rRB-1BRLORF4-infected chickens 13 weeks p.i. (Fig. 4C). None of the rRB-1BRLORF4-infected chickens developed MD lesions for up to 8 weeks, which is generally the period over which MD incidence is studied. Examination of two independently derived single-deletion RLORF4 mutants showed that deletion of only one copy of RLORF4 is not sufficient to induce the phenotypic changes seen in double-deletion mutants, suggesting that the RLORF4 product is indeed important in MD pathogenesis and that one copy of the diploid ORF is sufficient for full pathogenicity. Analysis of the single-deletion mutants also indicated that it is unlikely that other significant genomic changes were introduced during the construction of mutant rRB-1BRLORF4, which could be responsible for attenuation. This was further substantiated by the fact that viral replication and oncogenic potential of three independent clones of rRB-1BRLORF4 were identical. Taken together, the in vivo and in vitro data strongly suggest that deleting both copies of RLORF4 directly leads to attenuation of MDV.

    Schat et al. (30) have speculated that attenuated strains may lose their ability to establish a primary infection in lymphocytes; however, this does not appear to be the case with rRB-1BRLORF4 because significant differences did not occur until 14 days p.i. This suggests that the involvement of RLORF4 in MD pathogenesis may be primarily later during the infection, possibly during the latency/reactivation or transformation phase, since RLORF4 transcripts could be detected in both lytically infected CKC cultures and transformed MDCC lines (10). Deletion of RLORF4 significantly reduced incidence of MD lesions, but it is unknown whether this was due to a potential involvement in transformation or an involvement in reactivation since there was less viral replication in vivo beyond 14 days p.i. in chickens infected with the RLORF4 deletion mutant.

    During the recent cloning of oncogenic RB-1B as a BAC, one clone termed pRB-1B-2 containing a 7.7-kb deletion in the IRL and TRL regions was obtained. Virus reconstituted from that clone was nononcogenic (24), presumably because deletion of these regions resulted in the complete loss of a number of genes or ORFs, including Meq, RLORF5a, RLORF4, and vIL-8, which strongly suggests that genes within this region are involved in oncogenesis. Deletion of vIL-8 in oncogenic RB-1B (22) and Md5 (6) showed that vIL-8 is dispensable for the establishment of latency and transformation but is involved in the early cytolytic phase, likely by attracting B or T lymphocytes to primarily infected cells. Lupiani et al. (19) recently showed that deletion of Meq in the Md5 strain did not affect replication of the virus but that Meq was involved in transformation since no chickens produced tumors in rMd5Meq-infected chickens for up to 8 weeks p.i. Since deletion of RLORF4 did not cause MD lesions at 7 weeks p.i. but did at 13 weeks p.i., it would be interesting to see whether other mutant viruses that have significantly reduced incidences of MD tumors up to 8 weeks continue to remain tumor-free for longer periods. These data combined indicate that Meq, RLORF4, and/or vIL-8 are involved in attenuation of MDV, while RLORF5a is not.

    It has previously been suggested that the region upstream of vIL-8 contains the putative vIL-8 promoter (10). RLORF4 is also located in this region; therefore, deletion of RLORF4 could also potentially affect vIL-8 mRNA expression. Although vIL-8 mRNA production is still evident in rRB-1BRLORF4-infected CKC cultures using qRT-PCR assays (Fig. 2), it is unknown at this time whether the level of vIL-8 mRNA expression is altered in vivo. We are currently studying the potential importance of levels of vIL-8 in the deletion mutants.

    It is interesting that when the levels of in vivo replication and oncogenesis for RB-1B/p17 and rRB-1B were compared, the latter replicated at significantly lower levels and MD occurred with kinetics that were delayed by approximately 1 week compared to RB-1B/p17. Additionally, no contact birds were detectably infected with MDV at 22 days p.i. in experiment 1. These data suggest that the pRB-1B virus harbors some mutations compared to the original RB-1B/p17 virus used as a comparative virus in this study, although the recombinant virus generated from pRB-1B is still virulent and regularly causes between 90 and 100% MD incidence in P2a chickens (13) (Fig. 4C). There are at least two possible reasons for this discrepancy between rRB-1B and RB-1B. The first is that insertion of the mini-F plasmid pHA1 into US2 alters pathogenesis. Previous studies in which the US1, US2, US10, and the SORFs-1, -2, and -3 were deleted by lacZ insertion in RB-1B showed that horizontal transmission was reduced (21). These data combined with those presented in this report suggest that US2 is important for MDV pathogenesis, possibly affecting the virus' ability to efficiently infect feather follicle epithelium cells. The second is that the change in pathogenesis of rRB-1B may be a clone-specific effect, since it was constructed using a clone-purified fourth passage stock of RB-1B and it was phenotypically identical to its parental strain (24).

    It has previously been shown that deletion of RLORF5a in the CVI988 strain did not affect in vitro and in vivo replication and the establishment of latency (31); however, its involvement in oncogenesis could not be studied since CVI988 is nononcogenic. In this report, we examined the effect of deleting RLORF5a in the virulent pRB-1B BAC clone. Figures 3 and 4 show that deleting both copies of RLORF5a did not affect in vitro or in vivo replication of rRB-1B, nor did it change its oncogenic potential, conclusively showing that RLORF5a is dispensable for MDV replication and oncogenicity. However, deleting both copies of RLORF4 in the virulent pRB-1B BAC clone led to changes in MDV replication, both in vitro and in vivo, two main characteristics of attenuated MDV strains. These data, in addition to the fact that a number of serially passaged MDV strains contain deletions within RLORF4, suggest that RLORF4 is important in MDV pathogenesis. Future studies will concentrate on the identification of the RLORF4 gene product and its mechanism of function.

    ACKNOWLEDGMENTS

    This work was supported in part by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant 2002-35204-11615 to Karel A. Schat, grant 2001-35204-10152 to Keith W. Jarosinski, and grant 2003-02234 to Nikolaus Osterrieder.

    We thank Priscilla H. O'Connell for her technical help in the laboratory.

    REFERENCES

    Bradley, G., M. Hayashi, G. Lancz, A. Tanaka, and M. Nonoyama. 1989. Structure of the Marek's disease virus BamHI-H gene family: genes of putative importance for tumor induction. J. Virol. 63:2534-2542.

    Bradley, G., G. Lancz, A. Tanaka, and M. Nonoyama. 1989. Loss of Marek's disease virus tumorigenicity is associated with truncation of RNAs transcribed within BamHI-H. J. Virol. 63:4129-4135.

    Calnek, B. W. 2001. Pathogenesis of Marek's disease virus infection. Curr. Top. Microbiol. Immunol. 255:25-55.

    Churchill, A. E., R. C. Chubb, and W. Baxendale. 1969. The attenuation, with loss of oncogenicity of the herpes-type virus of Marek's disease (strain HPRS-16) on passage in cell culture. J. Gen. Virol. 4:557-564.

    Cortes, P. L., and C. J. Cardona. 2004. Pathogenesis of a Marek's disease virus mutant lacking vIL-8 in resistant and susceptible chickens. Avian Dis. 48:50-60.

    Cui, X., L. F. Lee, W. M. Reed, H. J. Kung, and S. M. Reddy. 2004. Marek's disease virus-encoded vIL-8 gene is involved in early cytolytic infection but dispensable for establishment of latency. J. Virol. 78:4753-4760.

    Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.

    Fukuchi, K., A. Tanaka, L. W. Schierman, R. L. Witter, and M. Nonoyama. 1985. The structure of Marek disease virus DNA: the presence of unique expansion in nonpathogenic viral DNA. Proc. Natl. Acad. Sci. USA 82:751-754.

    Jarosinski, K. W., B. L. Njaa, P. H. O'Connell, and K. A. Schat. 2005. Pro-inflammatory responses in chicken spleen and brain tissues after infection with very virulent plus Marek's disease virus. Viral Immunol. 18:148-161.

    Jarosinski, K. W., P. H. O'Connell, and K. A. Schat. 2003. Impact of deletions within the BamHI-L fragment of attenuated Marek's disease virus on vIL-8 expression and the newly identified transcript of open reading frame LORF4. Virus Genes 26:255-269.

    Jarosinski, K. W., R. Yunis, P. H. O'Connell, C. J. Markowski-Grimsrud, and K. A. Schat. 2002. Influence of genetic resistance of the chicken and virulence of Marek's disease virus (MDV) on nitric oxide responses after MDV infection. Avian Dis. 46:636-649.

    Jones, D., L. Lee, J. L. Liu, H. J. Kung, and J. K. Tillotson. 1992. Marek disease virus encodes a basic-leucine zipper gene resembling the fos/jun oncogenes that is highly expressed in lymphoblastoid tumors. Proc. Natl. Acad. Sci. USA 89:4042-4046.

    Kamil, J. P., B. K. Tischer, S. Trapp, V. K. Nair, N. Osterrieder, and H. J. Kung. 2005. vLIP, a viral lipase homologue, is a virulence factor of Marek's disease virus. J. Virol. 79:6984-6996.

    Kopacek, J., L. J. Ross, V. Zelnik, and J. Pastorek. 1993. The 132 bp repeats are present in RNA transcripts from 1.8 kb gene family of Marek disease virus-transformed cells. Acta Virol. 37:191-195.

    Lee, E. C., D. Yu, J. Martinez de Velasco, L. Tessarollo, D. A. Swing, D. L. Court, N. A. Jenkins, and N. G. Copeland. 2001. A highly efficient Escherichia coli-based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics 73:56-65.

    Levy, A. M., Y. Izumiya, P. Brunovskis, L. Xia, M. S. Parcells, S. M. Reddy, L. Lee, H. W. Chen, and H. J. Kung. 2003. Characterization of the chromosomal binding sites and dimerization partners of the viral oncoprotein Meq in Marek's disease virus-transformed T cells. J. Virol. 77:12841-12851.

    Liu, J. L., S. F. Lin, L. Xia, P. Brunovskis, D. Li, I. Davidson, L. F. Lee, and H. J. Kung. 1999. MEQ and V-IL8: cellular genes in disguise? Acta Virol. 43:94-101.

    Liu, J. L., Y. Ye, L. F. Lee, and H. J. Kung. 1998. Transforming potential of the herpesvirus oncoprotein MEQ: morphological transformation, serum-independent growth, and inhibition of apoptosis. J. Virol. 72:388-395.

    Lupiani, B., L. F. Lee, X. Cui, I. Gimeno, A. Anderson, R. W. Morgan, R. F. Silva, R. L. Witter, H. J. Kung, and S. M. Reddy. 2004. Marek's disease virus-encoded Meq gene is involved in transformation of lymphocytes but is dispensable for replication. Proc. Natl. Acad. Sci. USA 101:11815-11820.

    Ohashi, K., W. Zhou, P. H. O'Connell, and K. A. Schat. 1994. Characterization of a Marek's disease virus BamHI-L-specific cDNA clone obtained from a Marek's disease lymphoblastoid cell line. J. Virol. 68:1191-1195.

    Parcells, M. S., A. S. Anderson, and T. W. Morgan. 1995. Retention of oncogenicity by a Marek's disease virus mutant lacking six unique short region genes. J. Virol. 69:7888-7898.

    Parcells, M. S., S. F. Lin, R. L. Dienglewicz, V. Majerciak, D. R. Robinson, H. C. Chen, Z. Wu, G. R. Dubyak, P. Brunovskis, H. D. Hunt, L. F. Lee, and H. J. Kung. 2001. Marek's disease virus (MDV) encodes an interleukin-8 homolog (vIL-8): characterization of the vIL-8 protein and a vIL-8 deletion mutant MDV. J. Virol. 75:5159-5173.

    Peng, Q., M. Zeng, Z. A. Bhuiyan, E. Ubukata, A. Tanaka, M. Nonoyama, and Y. Shirazi. 1995. Isolation and characterization of Marek's disease virus (MDV) cDNAs mapping to the BamHI-I2, BamHI-Q2, and BamHI-L fragments of the MDV genome from lymphoblastoid cells transformed and persistently infected with MDV. Virology 213:590-599.

    Petherbridge, L., A. C. Brown, S. J. Baigent, K. Howes, M. A. Sacco, N. Osterrieder, and V. K. Nair. 2004. Oncogenicity of virulent Marek's disease virus cloned as bacterial artificial chromosomes. J. Virol. 78:13376-13380.

    Ross, L. J., B. Milne, and P. M. Biggs. 1983. Restriction endonuclease analysis of Marek's disease virus DNA and homology between strains. J. Gen. Virol. 64:2785-2790.

    Ross, N., M. M. Binns, M. Sanderson, and K. A. Schat. 1993. Alterations in DNA sequence and RNA transcription of the BamHI-H fragment accompany attenuation of oncogenic Marek's disease herpesvirus. Virus Genes 7:33-51.

    Sambrook, J., and D. W. Russell (ed.). 2001. Plasmids and their usefulness in molecular cloning, p. 1.1-1.162. In Molecular cloning: a laboratory manual, 3rd ed., vol. 1. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

    Schat, K. A., A. Buckmaster, and L. J. Ross. 1989. Partial transcription map of Marek's disease herpesvirus in lytically infected cells and lymphoblastoid cell lines. Int. J. Cancer 44:101-109.

    Schat, K. A., B. W. Calnek, and J. Fabricant. 1982. Characterization of two highly oncogenic strains of Marek's disease virus. Avian Pathol. 11:593-605.

    Schat, K. A., B. W. Calnek, J. Fabricant, and D. L. Graham. 1985. Pathogenesis of infection with attenuated Marek's disease virus strains. Avian Pathol. 14:127-146.

    Schat, K. A., B. J. Hooft van Iddekinge, H. Boerrigter, P. H. O'Connell, and G. Koch. 1998. Open reading frame L1 of Marek's disease herpesvirus is not essential for in vitro and in vivo virus replication and establishment of latency. J. Gen. Virol. 79:841-849.

    Schat, K. A., and H. G. Purchase. 1998. Cell-culture methods, p. 223-234. In D. E. Swayne, J. R. Glisson, M. W. Jackwood, J. E. Pearson, and W. M. Reed (ed.), A laboratory manual for the isolation and identification of avian pathogens, 4th ed. American Association of Avian Pathologists, Kennett Square, Pa.

    Schumacher, D., B. K. Tischer, W. Fuchs, and N. Osterrieder. 2000. Reconstitution of Marek's disease virus serotype 1 (MDV-1) from DNA cloned as a bacterial artificial chromosome and characterization of a glycoprotein B-negative MDV-1 mutant. J. Virol. 74:11088-11098.

    Silva, R. F., S. M. Reddy, and B. Lupiani. 2004. Expansion of a unique region in the Marek's disease virus genome occurs concomitantly with attenuation but is not sufficient to cause attenuation. J. Virol. 78:733-740.

    Tischer, B. K., D. Schumacher, D. Chabanne-Vautherot, V. Zelnik, J. F. Vautherot, and N. Osterrieder. 2005. High-level expression of Marek's disease virus glycoprotein C is detrimental to virus growth in vitro. J. Virol. 79:5889-5899.

    Tulman, E. R., C. L. Afonso, Z. Lu, L. Zsak, D. L. Rock, and G. F. Kutish. 2000. The genome of a very virulent Marek's disease virus. J. Virol. 74:7980-7988.

    von Bülow, V., and P. M. Biggs. 1975. Differentiation between strains of Marek's disease virus and turkey herpesvirus by immunofluorescence assays. Avian Pathol. 4:133-146.

    Weinstock, D., and K. A. Schat. 1987. Virus specific syngeneic killing of reticuloendotheliosis virus transformed cell line target cells by spleen cells. Prog. Clin. Biol. Res. 238:253-263.

    Wilson, M. R., R. A. Southwick, J. T. Pulaski, V. L. Tieber, Y. Hong, and P. M. Coussens. 1994. Molecular analysis of the glycoprotein C-negative phenotype of attenuated Marek's disease virus. Virology 199:393-402.

    Witter, R. L. 1972. Epidemiology of Marek's disease: a review, p. 111-122. In P. M. Biggs, G. de The, and L. N. Payne (ed.), Oncogenesis and herpesviruses. IARC, Lyon, France.

    Witter, R. L. 2001. Protective efficacy of Marek's disease vaccines. Curr. Top. Microbiol. Immunol. 255:57-90.

    Witter, R. L., and L. Offenbecker. 1979. Nonprotective and temperature-sensitive variants of Marek's disease vaccine viruses. JNCI 62:143-151.

    Xing, Z., and K. A. Schat. 2000. Expression of cytokine genes in Marek's disease virus-infected chickens and chicken embryo fibroblast cultures. Immunology 100:70-76.

    Yamaguchi, T., S. L. Kaplan, P. Wakenell, and K. A. Schat. 2000. Transactivation of latent Marek's disease herpesvirus genes in QT35, a quail fibroblast cell line, by herpesvirus of turkeys. J. Virol. 74:10176-10186.

    Yunis, R., K. W. Jarosinski, and K. A. Schat. 2004. Association between rate of viral genome replication and virulence of Marek's disease herpesvirus strains. Virology 328:142-150.(Keith W. Jarosinski, Niko)