Horizontal transmission of Marek's disease virus requires US2, the UL13 protein kinase, and gC

Abstract

Marek's disease virus (MDV) causes a general malaise in chickens that is mostly characterized by the development of lymphoblastoid tumors in multiple organs. The use of bacterial artificial chromosomes (BACs) for cloning and manipulation of the MDV genome has facilitated characterization of specific genes and genomic regions. The development of most MDV BACs, including pRB-1B-5, derived from a very virulent MDV strain, involved replacement of the US2 gene with mini-F vector sequences. However, when reconstituted viruses based on pRB-1B were used in pathogenicity studies, it was discovered that contact chickens housed together with experimentally infected chickens did not contract Marek's disease (MD), indicating a lack of horizontal transmission. Staining of feather follicle epithelial cells in the skins of infected chickens showed that virus was present but was unable to be released and/or infect susceptible chickens. Restoration of US2 and removal of mini-F sequences within viral RB-1B did not alter this characteristic, although in vivo viremia levels were increased significantly. Sequence analyses of pRB-1B revealed that the UL13, UL44, and US6 genes encoding the UL13 serine/threonine protein kinase, glycoprotein C (gC), and gD, respectively, harbored frameshift mutations. These mutations were repaired individually, or in combination, using two-step Red mutagenesis. Reconstituted viruses were tested for replication, MD incidence, and their abilities to horizontally spread to contact chickens. The experiments clearly showed that US2, UL13, and gC in combination are essential for horizontal transmission of MDV and that none of the genes alone is able to restore this phenotype.

FIG. 1.

Analysis of U_S2 restoration by PCR. Shown is a schematic representation of the expected PCR product sizes using the SORF3vCre F5 (F5) and US2vCre B8 (B8) or SopC F and Us3-SeqUs2 AS (AS) primers and diagnostic PCR assays for insertion of U_S2 and removal of mini-F vector sequences from reconstituted viruses. (A to C) The expected PCR products using the two primer sets in reconstituted viruses lacking U_S2 and containing mini-F sequences (A), containing restored U_S2 and mini-F sequences (B), or containing restored U_S2 but lacking mini-F sequences after Cre excision (C) are shown. Using primers SopC F and AS, a 1,508- or 988-bp fragment is amplified in the presence or absence of U_S2, respectively, if mini-F sequences are present. No product will be amplified if mini-F sequences are removed, since the SopC F primer binds within the mini-F vector sequences. Using the F5 and B8 primers, a 714-bp fragment is amplified in U_S2-restored viruses lacking mini-F sequences. If vector sequences are present, the primers would be able to bind their respective complementary sequences but would not be amplified using our PCR parameters. In the absence of U_S2, the B8 primers would not bind. (D) The top panel shows PCR assays for the restoration of the U_S2 gene in reconstituted viruses (v). BAC DNA was transfected into CEC cultures with (+) or without (−) the pCAGGS-NLS/Cre plasmid, and reconstituted viruses were passaged in CKC cultures up to five times. DNA was extracted from infected CKC cultures and used in PCRs using previously described methods (33). Amplification of U_S2 was evident in each virus in which U_S2 was restored. The bottom panel shows PCR assays for the removal of mini-F sequences when viruses were reconstituted with the Cre enzyme present. The asterisks indicate BAC DNA used as controls.

FIG. 2.

In vivo replication and MD incidence in chickens infected with U_S2-restored vRB-1B with (+F) or without mini-F sequences. (A) Mean MDV genomic copies/1 × 10⁶ blood cells ± standard errors of the means using qPCR assays. Values for each group that were significantly higher (∧) or lower (*) than those of all other groups (P < 0.05) are indicated. (B) Percent MD incidence for each group during the 9-week evaluation period.

FIG. 3.

Frameshift mutations within U_L13, gC, and gD in pRB-1B and their predicted protein sequences. The nucleotide sequences of U_L13, gC, and gD in which frameshift mutations were identified are shown in panels A, B, and C, respectively, compared to the published Md5 sequence (48). Numbers indicate the respective nucleotide, and bold capitalized letters indicate the additional nucleotide causing the frameshift. Predicted protein sequences for each gene are compared to the Md5 sequence. Potential translation initiation sites are indicated with a bold M. Truncated proteins are shown for U_L13, gC, and gD. Underlined amino acid sequences indicate the serine/threonine protein kinase active site for U_L13 (A) and transmembrane domains for gC (B) and gD (C).

FIG. 4.

Western blot analysis of secreted gC in repaired viruses. Glycoproteins of uninfected or MDV-infected CKC cultures were precipitated with concanavalin A-Sepharose beads and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, followed by detection of gC using the gC-specific monoclonal antibody A6 as previously described (45). The gC protein was detected only in gC-repaired viruses with the expected sizes of 57 to 65 kDa, while uninfected and 1194- or 1216-infected CKC cultures were negative.

FIG. 5.

In vivo replication and MD incidence in chickens infected with repaired vRB-1B. (A) Mean MDV genomic copies/1 × 10⁶ blood cells ± standard errors of the means (error bars) using qPCR assays for trial 1 with v1193, v1194, v1216, v1231, or v1232 (Table 2). Values for each group that were significantly higher (∧) or lower (*) than those of all other groups without the same annotation (P < 0.05) are indicated. (B) Percent MD incidence for each group during the 52-day evaluation period in trial 1. (C) Same as in panel A for trial 1 except that v1193, v1216, v1218, v1265, v1232, v1272 (Table 2), or wild-type RB-1B was used. (D) Same as in panel B for trial 1 during an evaluation period of 37 days.

FIG. 6.

Incidence of MD in inoculated or contact chickens with repaired vRB-1B viruses. (A) The top panel shows the MD incidence of chickens inoculated with v1194, v1216, v1217, v1218, v1231, or v1232 over a 7-week period. Below is a table showing the number of chickens positive for gross lesions or amplification of MDV genomic copies using qPCR assays at the termination of the experiment. Chickens that died prior to the termination of the experiment due to MD were not used for blood DNA collection; thus, the total number of chickens sampled for qPCR assays may vary from the total number of chickens evaluated for gross lesions. The asterisk (*) denotes that chickens were excluded because the control qPCR (iNOS amplification) was undetectable, and thus they were excluded from the group for qPCR assays. (B) Same as in panel A, except that the experiment was extended to 13 weeks p.i. and used v1193, v1216, v1218, v1265, v1232, v1272, or wild-type RB-1B.