Replication of porcine circoviruses
© Faurez et al. 2009
Received: 05 March 2009
Accepted: 18 May 2009
Published: 18 May 2009
Skip to main content
© Faurez et al. 2009
Received: 05 March 2009
Accepted: 18 May 2009
Published: 18 May 2009
Porcine circoviruses are circular single-stranded DNA viruses that infect swine and wild boars. Two species of porcine circoviruses exist. Porcine circovirus type 1 is non pathogenic contrary to porcine circovirus type 2 which is associated with the disease known as Post-weaning Multisystemic Wasting Syndrome. Porcine circovirus DNA has been shown to replicate by a rolling circle mechanism. Other studies have revealed similar mechanisms of rolling-circle replication in plasmids and single-stranded viruses such asGeminivirus. Three elements are important in rolling-circle replication: i) a gene encoding initiator protein, ii) a double strand origin, and iii) a single strand origin. However, differences exist between viruses and plasmids and between viruses. Porcine circovirus replication probably involves a "melting pot" rather than "cruciform" rolling-circle mechanism.
This review provides a summary of current knowledge of replication in porcine circoviruses as models of theCircovirus genus. Based on various studies, the factors affecting replication are defined and the mechanisms involved in the different phases of replication are described or proposed.
It is apparent from the nucleic acid sequence that the PCV genome might be at the interface of Geminivirus and Nanovirus genomes . The Rep protein sequence suggests that the PCV genome results from a Nanovirus and Calicivirus recombination .
Based on common features which include the initiator Rep protein and stem-loop structure, it has been proposed that PCVs, likeGeminiviridae andNanoviridae, replicate using a rolling circle replication (RCR) strategy [3, 4]. This mechanism is also involved in replication of the pT181 plasmid of Staphylococcus aureus which encodes resistance to tetracycline . Studies of PCV replication have been based on the RCR model but a novel mechanism or "melting-pot RCR model" has been proposed which involves modifications at the origin of replication . This review provides a synthesis of current knowledge of replication in PCVs as models of theCircovirus genus.
PCV replication studies have generally been performed on the PCV-1 model. Nevertheless, PCV-1 and PCV-2 exhibit 79.5% and 82% sequence homologies for the replication origin and Rep gene, respectively , and one replication factor can be replaced by another . Therefore, hypotheses formulated for PCV-1 replication might probably be valid for PCV-2.
Replication of the PCV genome, unlike that of plasmids andGeminiviruses, requires the REP complex which consists of 2 viral proteins Rep and Rep' . Therep gene encoding for these 2 proteins is located on the positive strand downstream from the replication origin and its promoter includes the replication origin and to a certain extent ORF2. Rep consists of 314 amino acids and 37.5 kDa protein, whereas Rep', which is a spliced form of the Rep mRNA, consists of 178 amino acids and 20.2 kDa protein . The C-term parts of Rep' and Rep are also different. The N-term portions of Rep and Rep' contain three conserved amino acid sequences typical of proteins initiating RCR [14, 15].
The three-dimensional structures of the geminivirus Tomato Yellow Leaf Curl Sardinia Virus (TYLCSV) , in the nanovirus Faba Bean Necrotic Yellow Virus (FBNYV)  and in the circovirus PCV-2  have been described. According to structural data, the three RCR motives in TYLCSV, FBNYV and PCV-2 are similarly positioned on the various elements of the Rep proteins structure.
The first step in RCR is synthesis of the lagging strand which results in passage from single-strand DNA to the replicating double-strand DNA. As yet, neither the PCV genomic DNA sequence nor the proteins involved in this synthesis have been determined. However, the replication mechanisms described for RCR plasmids, Nanoviruses and Geminiviruses might provide some indication of the PCV mechanism.
In RCR plasmids, synthesis is initiated and terminated at a single-stranded origin. This origin is a non-coding region that can generate an imperfect stem-loop and is usually located away from the double-stranded origin involved in leading strand synthesis. The single-stranded origin contains a promoter recognized by a host RNA polymerase that synthesizes an RNA primer essential to DNA synthesis by cellular DNA polymerase I . Two mechanisms have been described for different groups of Geminiviruses. Mastreviruses (subgroup I) encapsidate DNA molecules about 80 nucleotides in length that are complementary to a single-stranded Small Intergenic Region (SIR) of the virus genome. This small intergenic region is different from the intergenic region which contains the origin of replication. The complementary DNA molecule consists of ribonucleotides at the 5' end which induce synthesis of the lagging strand . Such DNA molecules have not been detected for Curtoviruses (subgroup II) or Begomoviruses (subgroup III), except in the African cassava mosaic virus that belongs to subgroup III . The origin of the negative strand of Curtoviruses and Begomoviruses seems to be located in the intergenic region which contains the origin of replication and, contrary to Mastreviruses, a RNA primer is probably synthesized after infection . The presence of a DNA molecule with the same function as in Mastreviruses was also described for one Nanovirus but no ribonucleotide was identified .
The origin of replication of the lagging strand in PCV has not yet been identified. However the Geminivirus and Nanovirus models might help to explain PCV replication. It is possible that the origin of synthesis of the negative strand of the PCV genome is located in the intergenic region, which may or may not contain the origin of replication. A DNA molecule might thus either be encapsidated like Mastrevirus or produced after infection like Begomovirus.
Initiation of the leading strand first requires binding of the REP complex to the replicating form of the PCV genome, followed by nicking of the DNA by this complex.
The replication origin contains the elements essential to REP complex binding. The Rep protein binds to hexamers H1 and H2 and to the right strand of the stem-loop whereas the Rep' protein binds only to hexamers H1 and H2 . Rep protein binding requires cytosines at positions 3 and 10 of the right strand of the stem-loop motif (figure 3) and an appropriate spatial distance between these nucleotides . Binding of Rep and/or Rep' to the viral genome might be influenced by other elements in the PCV genome such as the octanucleotide in the loop . The binding interface of the Rep protein includes two structural elements. The first element is the C terminal of the α helix (α2 helix and α3 helix). The second element is the loop between the β4 sheet and the β5 sheet  (figure 4).
The replication origin is destabilized by the REP complex bound to the DNA. The REP complex nicks then the AxTAxTAC sequence of positive single stranded DNA (+) between nucleotides T and A (figure 3). The porcine circovirus cleavage mechanism presents similarities to that of Geminivirus. In both cases, cleavage is dependent on the presence of divalent cations such as Mg2+, but not of ATP or of the stem-loop structure [12, 28]. In PCV, the catalytic site responsible for this cleavage is tyrosine 96 (Tyr96) located on motif III of the Rep protein. Cleavage of the phosphodiester bond usually occurs by nucleophilic attack of the hydroxyl group of tyrosine 96. This attack generates a free 3'OH end and a phosphotyrosine diester binding at the 5' site . Replication of the positive viral DNA strand by the host DNA polymerase is initiated at the free 3'OH end. This 3'OH end is used as a primer. Several steps are needed to initiate replication: recruitment of the REP complex, cleavage and recruitment of the host DNA polymerase. Each step may interact with each of the others.
The termination mechanism of PCV leading strand synthesis has not yet been studied. However the involvement of tyrosine in PCV replication suggests that the mechanism might be similar to that of the pT181 plasmid.
This mini-review summarizes the various mechanisms and elements necessary for rolling circle replication of porcine circovirus. Although not all these mechanisms have been clarified, the explanatory indications are based on sequence homology between porcine circovirus and other models such as Geminivirus or pT181 plasmids. Supplementary studies on porcine circovirus should validate or not the hypotheses associated with the various models.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.