An ARID family protein binds to the African swine fever virus encoded ubiquitin conjugating enzyme, UBCv1

Edited by Jesus Avila
  • Institute for Animal Health, Pirbright Laboratory, Ash Road, Pirbright, Surrey GU24 0NF, UK

Abstract

The NH2-terminal end of a protein, named SMCp, which contains an ARID (A/T rich interaction domain) DNA binding domain and is similar to the mammalian SMCY/SMCX proteins and retinoblastoma binding protein 2, was shown to bind the African swine fever virus encoded ubiquitin conjugating enzyme (UBCv1) using the yeast two hybrid system and in in vitro binding assays. Antisera raised against the SMCp protein were used to show that the protein is present in the cell nucleus. Immunofluorescence showed that although UBCv1 is present in the nucleus in most cells, in some cells it is in the cytoplasm, suggesting that it shuttles between the nucleus and cytoplasm. The interaction and co-localisation of UBCv1 with SMCp suggest that SMCp may be a substrate in vivo for the enzyme.

Keywords

  • Ubiquitin conjugating enzyme;
  • African swine fever virus;
  • A/T rich interaction domain;
  • Ubiquitin;
  • SMCX

1. Introduction

African swine fever virus (ASFV) is a large DNA virus which replicates in the cytoplasm of infected cells and has a tropism for macrophages. The 170 kb genome [8] and [32] encodes the only known virus encoded ubiquitin conjugating enzyme, UBCv1 [15] and [27].

These enzymes catalyse the attachment of ubiquitin to substrate proteins which tags them for proteolytic degradation via the proteasome. In addition to UBC enzymes, ubiquitin activating and E3 or ubiquitin protein ligase enzymes (for review see [6], [14], [28] and [29]) are involved in attaching ubiquitin to substrate proteins. Selective protein degradation mediated by protein ubiquitination regulates many processes including transcription and cell cycle control [6], [26] and [29].

Recombinant UBCv1 can, in vitro, ubiquitinate itself as well as histones and an ASFV virion protein, PIG1 [15] and [16]. However, it is not known whether these are in vivo substrate(s) for the enzyme. To investigate putative substrates for UBCv1, we identified interacting host proteins using the yeast two hybrid system. Among the UBCv1 interacting proteins was a truncated protein containing 215 amino acids including an ARID (A/T rich interaction domain) DNA binding domain. This protein, named SMCp, is very similar to the SMCX/Y [1] and [31] and retinoblastoma binding protein 2 (RBBP2) [10]. We demonstrated specific binding in vitro of UBCv1 to this 215 amino acid domain of SMCp and showed that both proteins are predominantly localised in the nucleus in cells. These results suggest either that SMCp is an in vivo substrate for UBCv1 or that the interaction targets UBCv1 to another substrate.

2. Materials and methods

2.1. Cells and viruses

Primary lamb testis (LT), bovine kidney (BK) cells, IBRS2 (ECACC 84100503), RK-13 (ECACC 88062427), HeLa (ECACC 84121901), CHO-K1 (ECACC 85051005), Vero (ECACC 84113001) and BSC 1 (ECACC 85011422) cells were grown under 5% CO2 at 37°C in HEPES buffered Dulbecco’s modified Eagle’s medium supplemented with 10% foetal calf serum. BA71V tissue culture adapted ASFV isolate has been described previously [4].

2.2. Plasmids

A SmaI, SalI fragment containing the UBCv1 gene was cloned into pAS2-CYH2 (Clontech) as an in frame fusion with the GAL4 DNA binding domain. The insert in the plasmid was sequenced to confirm that the UBCv1 ORF was in frame with the GAL4 DNA binding domain.

SMCp was amplified by PCR from the pACT2.6 AD plasmid, using primers SMCp-FOR; AAA View the MathML source ATG GAG CCG GGG TCA GAC G and SMCp-REV; AAA View the MathML source GGG TTT TTC AGT ATC TAC GAT TC. The amplified fragment of 650 bp was digested with the BamHI and EcoRI and ligated into pGEX2TK vector (Pharmacia-Amersham) to give the pGEX2TK-SMCp plasmid.

The UBCv1 gene was cloned as a BamHI fragment downstream and in frame with an eight amino acid epitope tag derived from human influenza virus haemagglutinin (HA) protein to give pCDNA-HA-UBCv1.

The HA tagged UBCv1 gene cassette was released from pCDNA-HA-UBCv1 using HindIII and XbaI, end repaired and ligated to SmaI digested p32-OP plasmid (L.K. Dixon and C.C. Abrams, unpublished results), downstream of the ASFV early/late p32 promoter [25] in the pT7 plasmid (Invitrogen), to give the plasmid p32-OP-HA-UBCv1.

Primers NLS Mut1 (5′-GAT CTView the MathML sourceC ATA TGG ATT GG-3′) and NLS Mut2 (5′-ATView the MathML sourceT CCT CTT TAT ATA GAT ATT TAC GGT AG-3′) were used to amplify 700 bp containing the p32 promoter, the HA epitope and the 5′ portion of UBCv1 upstream of the putative NLS. Primers NLS Mut3 (5′-GAG GAView the MathML sourceA TCA TAC CCC ATG GAA GAG TGT TCA GCG GAA GAC ATA G-3′) and NLS Mut4 (5′-GTG TGA View the MathML source CTA CTC ATC ATC CAT CTC TT-3′) amplified 200 bp of the 3′ terminus of UBCv1 after the putative NLS. The 700 bp and 200 bp PCR fragments were digested at XbaI sites within the primers, ligated together and then cloned into pT7Blue2 to give p32-OP-HA-UBCv1 (NLS Mut). The 700 bp 5′ fragment was also cloned separately into pCDNA3 giving p32-OP-HA-UBCv1 ΔC.