Cellular sequences in pestivirus genomes encoding gamma-aminobutyric acid (A) receptor-associated protein and Golgi-associated ATPase enhancer of 16 kilodaltons

Abstract

The presence of cellular protein coding sequences within viral RNA genomes is a unique and particularly interesting feature of cytopathogenic (cp) pestiviruses. Here we report the identification and characterization of two novel cellular sequences in the genomes of cp bovine viral diarrhea virus (BVDV) strains. In BVDV strain CP X604, we detected a duplication of the genomic region encoding NS3, NS4A, and part of NS4B, together with an insertion of sequences that code for cellular gamma-aminobutyric acid (A) receptor-associated protein [GABA(A)-RAP]. Transient-expression studies showed that the GABA(A)-RAP sequence leads to additional processing of the viral polyprotein and thereby to the expression of nonstructural protein NS3. Transfection of bovine cells with RNA transcribed from an infectious cDNA clone revealed that the GABA(A)-RAP-encoding insertion together with the duplicated viral sequences constitutes the genetic basis for the cytopathogenicity of strain CP X604. Surprisingly, molecular analysis of another cp BVDV strain (CP 721) resulted in the identification of a cellular Golgi-associated ATPase enhancer of 16 kDa (GATE-16)-encoding insertion together with duplicated viral sequences. To our knowledge, the genomes of CP X604 and CP 721 are the first viral RNAs found with cellular sequences encoding GABA(A)-RAP and GATE-16, respectively. Interestingly, the two cellular proteins belong to a family of eukaryotic proteins involved in various intracellular trafficking processes. Processing after the C-terminal glycine residue of GABA(A)-RAP and GATE-16 by cellular proteases is essential for covalent attachment to target molecules. Accordingly, it can be assumed that these cellular proteases also recognize the cleavage sites in the context of the respective viral polyproteins and thereby lead to the generation of NS3, the marker protein of cp BVDV.

FIG. 1.

Analysis of BVDV-1 strain CP X604. (A) Northern blot analysis of total RNA from MDBK cells infected with BVDV CP X604 or NCP7 and from noninfected cells (n.i.). RNA was separated by denaturing agarose gel electrophoresis, blotted onto a nylon membrane, and hybridized with a 2.5-kb NotI-NsiI fragment from the cDNA clone pCP7-5A (3, 7). Numbers refer to RNA ladder sizes (in kilobases). Migration positions of the viral genomic RNAs are marked with arrows. (B) Genome organizations of the BVDV-1 strains NCP7 and CP X604. For NCP7, the line below the polyprotein demarcates the viral genome, with the scale being in bases (26, 35). The genome of CP X604 contains a duplication of viral sequences encoding NS3, NS4A, and part of NS4B (NS4B∗) (grey box) together with a host cell-derived insertion (black box with X). The inserted cellular sequence encodes aa 14 to 116 of GABA(A)-RAP [GABA(A)-RAP∗]. The bars are not drawn to scale. The underlined part of the CP X604 genome has been sequenced. (C) Deduced amino acid sequence encoded by part of the CP X604 sequence. The positions of NS4A, NS4B∗, GABA(A)-RAP∗, and NS3 are indicated. The cellular insertion is underlined.

FIG. 2.

(A) Schematic representations of the fusion proteins encoded by the constructs p604-A and p604-B used for transient expression. The T7 expression plasmids were based on plasmid pCITE-CP7-NS34AB, which is a derivative of the vector pCITE (Invitrogen, De Schelp, The Netherlands) and encompasses the genomic region encoding NS3, NS4A, and part of NS4B of BVDV strain CP7. To establish a construct for the expression of GABA(A)-RAP∗, NS3, NS4A, and part of NS4B, the CP X604-specific cDNA fragment was obtained by RT-PCR with primer Ol X604-INS [5′-TAGTCATGAGGAAAAAGGAAGGAGAAAAG-3′, sense; the underlined nucleotides correspond to the 5′ part of the GABA(A)-RAP∗ coding sequence], which includes a BspHI site, and primer Ol X604-AgeIR (5′-CAACCGGTCTCCAGACCCCTCCT-3′, antisense; located in the NS3 coding region), which includes an AgeI site, and cloned into pCITE-CP7-NS34AB (precut with NcoI and AgeI). This plasmid was termed p604-B. For construction of p604-A, the genomic region encoding NS4B∗ (N-terminal 52 aa of NS4B), GABA(A)-RAP∗, and part of NS3 was generated by RT-PCR with primer Ol X604-NS4B (5′-GAGTCATGAGCGCGGGTGACGTGGAG-3′, sense; the underlined nucleotides encode aa 2 to 6 of NS4B of CP X604), which includes a BspHI site, and primer Ol X604-AgeIR and cloned into pCITE-CP7-NS34AB (precut with NcoI and AgeI). (B) Immunoblot. MDBK cells infected with CP X604 and BHK-21 cells transfected with p604-A or p604-B were lysed 48 h postinfection and 16 h posttransfection, respectively. The samples were separated by sodium dodecyl sulfate-8% polyacrylamide gel electrophoresis under reducing conditions, transferred to nitrocellulose, and incubated with the anti-NS3 monoclonal antibody 8.12.7, kindly provided by E. J. Dubovi (Cornell University, Ithaca, N.Y.). Mock-infected or -transfected cells (mock) served as negative controls. The sizes (in kilodaltons) of marker proteins are indicated on the left. The position of NS3 is indicated by an arrow. In cells infected with CP X604, NS2-3 was clearly visible after longer exposure times (not shown); for expression of NS2-3 and NS3 of CP X604, see also Fig. 3D.

FIG. 3.

Transfection experiments with RNA transcribed from infectious clones pNCP7-5A, pCP7-5A, and p7/X604. (A) Genome organizations of noncp BVDV NCP7-5A, cp BVDV CP7-5A, and the chimeric cDNA construct p7/X604, whose structure mirrors the genome structure of CP X604. The BVDV full-length cDNA clones pNCP7-5A and pNCP7-5A-(AgeI−) have been described previously (2, 9). These plasmids are noncp derivatives of the cp BVDV infectious full-length cDNA clone pCP7-5A (7). Construction of p7/X604 was based on pNCP7-5A-(AgeI−), which differs from pNCP7-5A by the absence of the single AgeI site (nt 5309 to 5314 of the NCP7-5A sequence). A CP X604-specific MluI-AgeI fragment was obtained by RT-PCR with primer Ol X604-MluI (5′-TTACGCGTCTGGAGGACTGGAATTCG-3′, sense), which encompasses an MluI site that corresponds to nt 7434 to 7439 of the NCP7-5A sequence, and primer Ol X604-AgeIR, which encompasses an AgeI site that corresponds to nt 5309 to 5314 of the NCP7-5A sequence, and cloned into pCR2.1 (Invitrogen). The resulting plasmid was termed pX604-MluI. Next, the SacI (nt 5842 to 5847)-MluI (nt 7434 to 7439) fragment of pNCP7-5A was cloned into p604-MluI that had been precut with SacI (located in the polylinker) and MluI. Finally, adding the pNCP7-5A-derived AgeI-SalI fragment (nt 5310 to 7720) completed the cloning of the SacI-SalI fragment, which was then introduced into pNCP7-5A-(AgeI−) that had been precut with SacI and SalI. For 7/X604, the CP X604-specific part of the genome and the positions of the MluI site (▪) and the AgeI site (▴) used for cDNA cloning are indicated by a line below the bar. (B) Crystal violet staining of MDBK cells 2 days after transfection with the indicated RNAs. In vitro synthesis of RNA and transfection of MDBK cells were performed as described previously (7). About 2 μg of RNA was used for each transfection. (C and D) Northern blot (C) and immunoblot (D) analyses of MDBK cells infected with the indicated viruses at a multiplicity of infection of 0.05. Northern blotting and immunoblotting were performed as described in the legends for Fig. 1A and 2B. n.i., noninfected cells. (C) RNAs were extracted 24 h after infection. The intensities of bands were determined with a phosphorimager. The relative amounts of viral genomic RNAs are indicated below the blot (percentages of the value for CP7-5A [100%]). Migration positions of the viral genomic RNAs are marked with arrows.

FIG. 4.

Analysis of BVDV-1 strain CP 721. (A) Northern blot analysis of total RNA from MDBK cells infected with 721 (mixture of cp and noncp virus), NCP 721, or CP 721. Migration positions of the viral genomic RNAs are indicated. (B) Genome organizations of NCP7 and CP 721. The genome of CP 721 contains a duplication of viral sequences encoding part of NS2 (NS2∗), NS3, NS4A, and part of NS4B (NS4B∗) (grey box), together with an insertion encoding part of GATE-16 (GATE-16∗; black box with G). The bars are not drawn to scale. The underlined parts of the CP 721 genome have been sequenced (this study and reference 5). (C) Deduced amino acid sequence of part of the CP 721 sequence. The positions of NS4A, NS4B∗, NS2∗, GATE-16∗, and NS3 are indicated. The cellular insertion is underlined.

FIG. 5.

Alignment of GABA(A)-RAP (GenBank accession no. AV605665), GATE-16 (accession no. AY117147), LC3 (accession no. U05784), and related proteins from C. elegans (accession no. Z69385 and U23511), A. thaliana (accession no. AC002387), S. cerevisiae (Apg8p; accession no. X79489), and Laccaria bicolor (accession no. U93506). Black and grey shading indicate identical and similar amino acids, respectively. Note that not all related sequences present in the data banks are included; e.g., several additional homologous sequences are found in A. thaliana. The proposed cleavage site after the highly conserved C-terminal glycine residue is marked (▴). Computer analysis of sequence data was performed with HUSAR (Deutsches Krebsforschungszentrum, Heidelberg, Germany), which provides the Genetics Computer Group software package (15).