- Volume 94, Issue 6, 2013

Adenovirus (AdV) is thought to follow a sequential assembly pathway similar to that observed in dsDNA bacteriophages and herpesviruses. First, empty capsids are assembled, and then the genome is packaged through a ring-like structure, referred to as a portal, located at a unique vertex. In human AdV serotype 5 (HAdV5), the IVa2 protein initiates specific recognition of viral genome by associating with the viral packaging domain located between nucleotides 220 and 400 of the genome. IVa2 is located at a unique vertex on mature capsids and plays an essential role during genome packaging, most likely by acting as a DNA packaging ATPase. In this study, we demonstrated interactions among IVa2, 33K and DNA-binding protein (DBP) in virus-infected cells by in vivo cross-linking of HAdV5-infected cells followed by Western blot, and co-immunoprecipitation of IVa2, 33K and DBP from nuclear extracts of HAdV5-infected cells. Confocal microscopy demonstrated co-localization of IVa2, 33K and DBP in virus-infected cells and also in cells transfected with IVa2, 33K and DBP genes. Immunogold electron microscopy of purified HAdV5 showed the presence of IVa2, 33K or DBP at a single site on the virus particles. Our results provide indirect evidence that IVa2, 33K and DBP may form a complex at a unique vertex on viral capsids and cooperate in genome packaging.

Human bocavirus (HBoV), closely related to canine minute virus (MVC) and bovine parvovirus (BPV), is a new member of the Bocavirus genus within the Parvoviridae family. The non-structural protein NP1 of HBoV is a nuclear localized protein and plays an important role in DNA replication as well as in the evasion of host innate immunity. In the current study, we provide the first evidence that NP1 possesses a non-classical nuclear localization signal (ncNLS) (amino acids 7–50). Embedded within this ncNLS is a classical bipartite nuclear localization signal (cNLS) (amino acids 14–28), capable of transporting a heterologous cytoplasmic protein β-galactosidase fusion protein (β-gal-EGFP) to the nucleus via the classical importin α/β1-mediated pathway. Amino acids 7–50 containing the cNLS and the ncNLS of NP1 or full-length NP1 interact with importin α1, importin β1 and importin β1Δ, which lacks the importin α binding domain, indicating that the nuclear import of NP1 is through both conventional importin α/β1 heterodimer- and non-classical importinß1-mediated pathways. Given that the arrangement of a cNLS embedded within an ncNLS is unusual in viral proteins, our data together reveal a novel molecular mechanism underlying the nuclear import of HBoV NP1, providing a basis for further understanding its biological function.

A large number of studies have revealed that persistent infections with certain human papillomavirus (HPV) types are necessary for the development of invasive cancer of the cervix. Recent studies have shown that not only do the major carcinogenic HPV types 16 and 18 encode E6 and E7 oncoproteins with immortalizing activity but also the very weakly or non-carcinogenic types 53, 66, 70 and 82. Currently, it is unknown whether transcriptional differences exist between these viruses that account for carcinogenicity in vivo. Therefore, we compared for the first time the activities of the upstream regulatory regions (URRs) that drive E6 and E7 expression derived from HPV16, -18, -31, -53, -66, -70 and -82 in the absence and presence of the viral E2 transcriptional regulator. URR activities in the absence of E2 varied widely and were further modulated by the cellular background. The co-expression of homologous E2 proteins resulted in repression of the URRs of only some HPV types and this varied with cell type. Activation by E2 proteins was less cell-type dependent but differed in an HPV-type-dependent manner. However, basal URR activity, repression of the URR by E2 and transcriptional activation by E2 did not correlate with HPV carcinogenicity in vivo. In summary, our data do not support the model that the transcriptional activity of human alphapapillomavirus types correlates with epidemiological risk classification.

The aim of this work is to identify antigenic regions within the ORF1 protein of Torque teno sus virus 1 (TTSuV1) and Torque teno virus sus 2 (TTSuV2) that could be used as antigens to detect virus-specific antibodies following infection in pigs. Protein sequences of TTSuV ORF1 genes were analysed to predict linear antigenic epitopes. Synthesized peptides were analysed for serological reactivity with swine sera. Such an antigenic region was identified at the C terminus of the ORF1 protein of both viruses and showed serological reactivity with 78 % (TTSuV1) and 88 % (TTSuV2) of swine sera. An ELISA with an immunodominant peptide as antigen was used to examine the sera of piglets, aged 4–20 weeks, and adults. Results indicated that TTSuV1- and TTSuV2-specific antibodies were detectable at 4 weeks. Antibody titres increased from week 10 and peaked at week 20. A relatively high antibody titre persisted to adulthood.

To examine polyomavirus (PyV) infection in wildlife, we investigated the presence of PyVs in Zambia with permission from the Zambia Wildlife Authority. We analysed 200 DNA samples from the spleens and kidneys (n = 100 each) of yellow baboons and vervet monkeys (VMs) (n = 50 each). We detected seven PyV genome fragments in 200 DNA samples using a nested broad-spectrum PCR method, and identified five full-length viral genomes using an inverse PCR method. Phylogenetic analysis of virally encoded proteins revealed that four PyVs were closely related to either African green monkey PyV or simian agent 12. Only one virus detected from a VM spleen was found to be related, with relatively low nucleotide sequence identity (74 %), to the chimpanzee PyV, which shares 48 % nucleotide sequence identity with the human Merkel cell PyV identified from Merkel cell carcinoma. The obtained entire genome of this virus was 5157 bp and had large T- and small t-antigens, and VP1 and VP2 ORFs. This virus was tentatively named vervet monkey PyV 1 (VmPyV1) as a novel PyV. Comparison with other PyVs revealed that VmPyV1, like chimpanzee PyV, had a longer VP1 ORF. To examine whether the VmPyV1 genome could produce viral proteins in cultured cells, the whole genome was transfected into HEK293T cells. We detected VP1 protein expression in the transfected HEK293T cells by immunocytochemical and immunoblot analyses. Thus, we identified a novel PyV genome from VM spleen.

Papillomaviruses appear to be species-specific pathogens, and it was suggested that each animal species might harbour its own set of papillomaviruses. However, all approaches addressing the underlying evolutionary phenomena still suffer from very limited data about animal papillomaviruses. In case of the horse for example, only three equine papillomaviruses (EcPVs) have been identified. To further address the situation in this host, suspected papillomavirus-associated lesions were tested for EcPV DNA. Four novel EcPV types were detected and their genomes entirely cloned and sequenced. They display the characteristic organization, with early (E) and late (L) regions harbouring the seven classical open reading frames divided by non-coding regions. They were named EcPVs 4, 5, 6 and 7, according to their dissimilarity to other papillomaviruses. Most L1 nucleotide identities were shared with EcPV2 in case of EcPV4 (62 %) and EcPV5 (60 %) or with EcPV3 in case of EcPV6 (70 %) and EcPV7 (71 %). Thus, EcPVs 4 and 5 may establish novel species within the genus Dyoiota, while EcPVs 6 and 7 might fit into the genus Dyorho and belong to the same species as EcPV3. They were found in genital plaques (EcPV4), aural plaques (EcPV5, EcPV6) or penile masses (EcPV7). Interestingly, PCR analysis revealed the DNA of EcPV2 and EcPV4 as well as of EcPV3 and EcPV6 together in the same tissue samples, respectively. In conclusion, the DNA of four novel EcPV types was identified and cloned. They cluster with the known types and support broad genetic EcPV diversity in at least two of the known clades. Furthermore, PCR assays also provide evidence for EcPV co-infections in horses.

Recent studies have suggested that the small ubiquitin-related modifier (SUMO) conjugation pathway may play an important role in intrinsic antiviral resistance and thus for repression of herpesviral infections. In particular, it was shown that the herpes simplex virus type-1 regulatory protein ICP0 acts as a SUMO-targeted ubiquitin ligase (STUbL), inducing the widespread degradation of SUMO-conjugated proteins during infection. As the IE1 protein of human cytomegalovirus (HCMV) is known to mediate a de-SUMOylation of PML, we investigated whether HCMV uses a similar mechanism to counteract intrinsic antiviral resistance. We generated primary human fibroblasts stably expressing FLAG-SUMO-1 or FLAG-SUMO-3 and analysed the SUMOylation pattern after HCMV infection or isolated IE1 expression. However, Western blot experiments did not reveal a global loss of SUMO conjugates, either in HCMV-infected or in IE1-expressing cells, arguing against a function of IE1 as an STUbL. Interestingly, we observed that FLAG-SUMO-1 and FLAG-SUMO-3, subsequent to IE1-mediated promyelocytic leukemia protein (PML) de-SUMOylation and the consequent disruption of PML nuclear bodies, were recruited into viral replication compartments. This raised the question of whether FLAG-SUMO-1/3 might promote HCMV replication. Intriguingly, overexpression of FLAG-SUMO-1/3 enhanced accumulation of viral DNA, which correlated with an increase in viral replication and in virus particle release. Together, these data indicate that HCMV, in contrast to other herpesviruses, has evolved subtle mechanisms enabling it to utilize the SUMO conjugation pathway for its own benefit, resulting in an overall positive effect of SUMO conjugation for HCMV replication.

P1a and P1b are two serine proteases of Cucumber vein yellowing virus (an ipomovirus). They belong to the group of P1 factors present at the N terminus of the polyproteins of most members of the family Potyviridae. The present work compares the protease activities of P1a and P1b in different experimental systems. The findings made regarding how these two proteases work, such as the requirement for a host factor by P1a but not by P1b, underscore important differences in their catalytic activity that point towards their undergoing divergent evolution involving the acquisition of mechanistic variations. The expression of several truncated forms of P1b in bacteria and in planta helped define the protease domain of P1b, along with other important features such as its apparently in cis mode of action. Recent phylogenetic data, together with the present results, allow an appealing hypothesis to be proposed regarding P1 evolution and its involvement in potyvirid speciation.

Bean pod mottle virus (BPMV) is a bipartite, positive-sense (+) RNA plant virus of the family Secoviridae. Its RNA1 encodes all proteins needed for genome replication and is capable of autonomous replication. By contrast, BPMV RNA2 must utilize RNA1-encoded proteins for replication. Here, we sought to identify RNA elements in RNA2 required for its replication. The exchange of 5′ untranslated regions (UTRs) between genome segments revealed an RNA2-specific element in its 5′ UTR. Further mapping localized a 66 nucleotide region that was predicted to fold into an RNA stem–loop structure, designated SLC. Additional functional analysis indicated the importance of the middle portion of the stem and an adjacent two-base mismatch. This is the first report of a cis-acting RNA element in RNA2 of a bipartite secovirus.

A cell line from the small brown planthopper (SBPH; Laodelphax striatellus) was established to study replication of rice stripe virus (RSV), a tenuivirus. The SBPH cell line, which had been subcultured through 30 passages, formed monolayers of epithelial-like cells. Inoculation of cultured vector cells with RSV resulted in a persistent infection. During viral infection in the SBPH cell line, the viral non-structural protein NS3 co-localized with the filamentous ribonucleoprotein particles of RSV, as revealed by electron and confocal microscopy. The knockdown of NS3 expression due to RNA interference induced by synthesized double-stranded RNAs from the NS3 gene significantly inhibited viral infection in the SBPH cell line. These results demonstrated that NS3 of RSV might be involved in viral replication or assembly. The persistent infection of the SBPH cell line by RSV will enable a better understanding of the complex relationship between RSV and its insect vector.

A new disorder was observed on southern highbush blueberries in several south-eastern states in the USA. Symptoms included irregularly shaped circular spots or blotches with green centres on the upper and lower surfaces of leaves. Double-stranded RNA was extracted from symptomatic leaves suggesting the presence of virus(es) possibly involved in the disease. Sequencing revealed the presence of a novel RNA virus with a ~14 kb genome divided into four RNA segments. Sequence analyses showed that the virus, for which we propose the name Blueberry necrotic ring blotch virus (BNRBV), possesses protein domains conserved across RNA viruses in the alpha-virus-like supergroup. Phylogenetic inferences using different genes placed BNRBV in a clade that includes the Bromoviridae, the genus Cilevirus (CiLV) and the recently characterized Hibiscus green spot virus (HGSV). Despite the strong genetic relationships found among BNRBV, Cilevirus and HGSV, the genome of BNRBV contains three features that distinguish it significantly from its closest relatives: (i) the presence of two helicase domains with different evolutionary pathways, (ii) the existence of three conserved nucleotide stretches located at the 3′ non-coding regions of each RNA segment and (iii) the conservation of terminal nucleotide motifs across each segment. Furthermore, CiLV and HGSV possess poly(A)-tailed bipartite and tripartite genomes, respectively, whereas BNRBV has a quadra-partite genome lacking a poly(A) tail. Based on these genetic features a new genus is proposed for the classification of BNRBV.