- Volume 97, Issue 1, 2016

In order to explore the roles of microRNA(s) [miRNA(s)] in the influenza A virus life cycle, we compared the miRNA profiles of 293T and HeLa cell lines, as influenza A virus can replicate efficiently in 293T cells but only poorly in HeLa cells. We analysed differentially expressed miRNAs and identified five, including miR-33a, that could disturb influenza A virus replication significantly. Using TargetScan analysis, we found that ARCN1 could be a potential target of miR-33a. To confirm whether miR-33a could truly target ARCN1, we generated a luciferase reporter for the ARCN1 3′ untranslated region (UTR) and performed a luciferase assay. The data indicated that miR-33a could suppress the luciferase activity of the reporter for the ARCN1 3′ UTR but not a reporter in which the predicted miR-33a targeting sites on ARCN1 3′ UTR were mutated. We performed immunoblotting to confirm that miR-33a could downregulate the protein level of ARCN1. Consistently, the level of ARCN1 protein in HeLa cells was significantly lower than that in 293T cells. We also demonstrated that ectopic expression of ARCN1 could partially rescue the inhibitory effect of miR-33a on virus replication. Furthermore, we demonstrated that miR-33a could impede virus replication at the stage of virus internalization, which was similar to the pattern for knockdown of ARCN1, indicating that miR-33a inhibits influenza virus infection by suppressing ARCN1 expression. In addition, we found that miR-33a could also weaken the viral ribonucleoprotein activity in an ARCN1-independent manner. In conclusion, we found that miR-33a is a novel inhibitory factor for influenza A virus replication.

Avian influenza viruses of subtype H9N2 that are found worldwide are occasionally transmitted to humans and pigs. Furthermore, by co-circulating with other influenza subtypes, they can generate new viruses with the potential to also cause zoonotic infections, as observed in 1997 with H5N1 or more recently with H7N9 and H10N8 viruses. Comparative analysis of the adaptive mutations in polymerases of different viruses indicates that their impact on the phylogenetically related H9N2 and H7N9 polymerases is higher than on the non-related H7N7 and H1N1pdm09 polymerases. Analysis of polymerase reassortants composed of subunits of different viruses demonstrated that the efficient enhancement of polymerase activity by H9N2-PB2 does not depend on PA and PB1. These observations suggest that the PB2 subunit of the H9N2 polymerase has a high adaptive potential and may therefore be an important pandemic risk factor.

Swine vesicular disease virus (SVDV) emerged around 1960 from a human enterovirus ancestor, coxsackievirus B5 (CVB5), and caused a series of epizootics in Europe and Asia. We characterized a coxsackievirus B4 strain that caused an epizootic involving 24 488 pigs in the Soviet Union in 1975. Phylogenetic evidence suggested that the swine virus emerged from a human ancestor between 1945 and 1975, almost simultaneously with the transfer of CVB5.

The quasispecies diversity of RNA viruses is mainly determined by the fidelity of RNA-dependent RNA polymerase (RdRp) during viral RNA replication. Certain amino acid residues play an important role in determining the fidelity, and such residues can be substituted with other amino acids to produce virus strains with higher fidelity. In this study, two amino acid substitutions (A283T and H421Y) in the RdRp of porcine reproductive and respiratory syndrome virus (PRRSV) were identified under the selection of ribavirin. Preliminary data showed that two substitutions were involved in conferring PRRSV with the properties of increased ribavirin resistance and restricted quasispecies diversity. The results indicated that these two amino acid residues (Ala283 and His421) play a crucial role in PRRSV replication by affecting the fidelity of its RdRp. The results have important implications for understanding the molecular mechanism of PRRSV evolution and pathogenicity, and developing a safer modified live-attenuated vaccine (MLV) against PRRSV.

Coxsackievirus B3 strain 28 (CVB3/28) is less stable at 37 °C than eight other CVB3 strains with which it has been compared, including four in this study. In a variant CVB3/28 population selected for increased stability at 37 °C, the capsid proteins of the stable variant differed from the parental CVB3/28 by two mutations in Vp1 and one mutation in Vp3, each of which resulted in altered protein sequences. Each of the amino acid changes was individually associated with a more stable virus. Competition between CVB3/28 and a more stable derivative of the strain showed that propagation of the less stable virus was favoured in receptor-rich HeLa cells.

There is growing evidence that virus particles also contain host cell proteins, which provide viruses with certain properties required for entry and release. A proteomic analysis performed on double-gradient-purified hepatitis C virus (HCV) from two highly viraemic patients identified the phosphatidylinositol 3,5-bisphosphate 5-phosphatase FIG4 (KIAA0274) as part of the viral particles. We validated the association using immunoelectron microscopy, immunoprecipitation and neutralization assays in vitro as well as patient-derived virus particles. RNA interference-mediated reduction of FIG4 expression decreased cholesteryl ester (CE) levels along with intra- and extracellular viral infectivity without affecting HCV RNA levels. Likewise, overexpressing FIG4 increased intracellular CE levels as well as intra- and extracellular viral infectivity without affecting viral RNA levels. Triglyceride levels and lipid droplet (LD) parameters remained unaffected. The 3,5-bisphosphate 5-phosphatase active site of FIG4 was found to strongly condition these results. Whilst FIG4 was found to localize to areas corresponding to viral assembly sites, at the immediate vicinity of LDs in calnexin-positive and HCV core-positive regions, no implication of FIG4 in the secretory pathway of the hepatocytes could be found using either FIG4-null mice, in vitro morphometry or functional assays of the ERGIC/Golgi compartments. This indicates that FIG4-dependent modulation of HCV infectivity is unrelated to alterations in the functionality of the secretory pathway. As a result of the documented implication of CE in the composition and infectivity of HCV particles, these results suggest that FIG4 binds to HCV and modulates particle formation in a CE-related manner.

Hepatitis C virus (HCV) is a major cause of chronic hepatitis and liver carcinoma and new therapies based on novel targets are needed. The tight junction protein claudin 1 (CLDN-1) is essential for HCV cell entry and spread, and anti-CLDN-1 rat and mouse mAbs are safe and effective in preventing and treating HCV infection in a human liver chimeric mouse model. To accelerate translation of these observations into a novel approach to treat HCV infection and disease in humans, we screened a phage display library of human single-chain antibody fragments by using a panel of CLDN-1-positive and -negative cell lines and identified phage specifically binding to CLDN-1. The 12 clones showing the highest levels of binding were converted into human IgG4. Some of these mAbs displayed low-nanomolar affinity, and inhibited infection of human hepatoma Huh7.5 cells by different HCV isolates in a dose-dependent manner. Cross-competition experiments identified six inhibitory mAbs that recognized distinct epitopes. Combination of the human anti-SRB1 mAb C-1671 with these anti-CLDN-1 mAbs could either increase or reduce inhibition of cell culture-derived HCV infection in vitro. These novel human anti-CLDN-1 mAbs are potentially useful to develop a new strategy for anti-HCV therapy and lend support to the combined use of antibodies targeting the HCV receptors CLDN-1 and SRB1, but indicate that care must be taken in selecting the proper combination.

Sphingosine kinase (SK) 1 is a host kinase that enhances some viral infections. Here we investigated the ability of SK1 to modulate dengue virus (DENV) infection in vitro. Overexpression of SK1 did not alter DENV infection; however, targeting SK1 through chemical inhibition resulted in reduced DENV RNA and infectious virus release. DENV infection of SK1− / −  murine embryonic fibroblasts (MEFs) resulted in inhibition of infection in an immortalized line (iMEF) but enhanced infection in primary MEFs (1°MEFs). Global cellular gene expression profiles showed expected innate immune mRNA changes in DENV-infected WT but no induction of these responses in SK1− / −  iMEFs. Reverse transciption PCR demonstrated a low-level induction of IFN-β and poor induction of mRNA for the interferon-stimulated genes (ISGs) viperin, IFIT1 and CXCL10 in DENV-infected SK1− / −  compared with WT iMEFs. Similarly, reduced induction of ISGs was observed in SK1− / −  1°MEFs, even in the face of high-level DENV replication. In both iMEFs and 1°MEFs, DENV infection induced production of IFN-β protein. Additionally, higher basal levels of antiviral factors (IRF7, CXCL10 and OAS1) were observed in uninfected SK1− / −  iMEFs but not 1°MEFs. This suggests that, in this single iMEF line, lack of SK1 upregulates the basal levels of factors that may protect cells against DENV infection. More importantly, regardless of the levels of DENV replication, all cells that lacked SK1 produced IFN-β but were refractory to induction of ISGs such as viperin, IFIT1 and CXCL10. Based on these findings, we propose new roles for SK1 in affecting innate responses that regulate susceptibility to DENV infection.

A full-length genome sequence of 27 739 nt was determined for the only known European turkey coronavirus (TCoV) isolate. In general, the order, number and size of ORFs were consistent with other gammacoronaviruses. Three points of recombination were predicted, one towards the end of 1a, a second in 1b just upstream of S and a third in 3b. Phylogenetic analysis of the four regions defined by these three points supported the previous notion that European and American viruses do indeed have different evolutionary pathways. Very close relationships were revealed between the European TCoV and the European guinea fowl coronavirus in all regions except one, and both were shown to be closely related to the European infectious bronchitis virus (IBV) Italy 2005. None of these regions of sequence grouped European and American TCoVs. The region of sequence containing the S gene was unique in grouping all turkey and guinea fowl coronaviruses together, separating them from IBVs. Interestingly the French guinea fowl virus was more closely related to the North American viruses. These data demonstrate that European turkey and guinea fowl coronaviruses share a common genetic backbone (most likely an ancestor of IBV Italy 2005) and suggest that this recombined in two separate events with different, yet related, unknown avian coronaviruses, acquiring their S-3a genes. The data also showed that the North American viruses do not share a common backbone with European turkey and guinea fowl viruses; however, they do share similar S-3a genes with guinea fowl virus.

Hepatitis B virus (HBV) is one of the world's major unconquered infections, resulting in progressive liver disease, and current treatments rarely cure infection. A limitation to discovering new therapies is our limited knowledge of HBV entry and dissemination pathways that hinders the development of in vitro culture systems. To address this gap in our understanding we optimized the genesis of infectious lentiviral pseudoparticles (HBVpps). The recent discovery that the bile salt transporter sodium taurocholate co-transporting polypeptide (NTCP) acts as a receptor for HBV enabled us to assess the receptor dependency of HBVpp infection. HBVpps preferentially infect hepatoma cells expressing NTCP, whereas other non-liver cells engineered to express NTCP do not support infection, suggesting that additional hepatocyte-specific factors are required for HBVpp internalization. These results highlight the value of the HBVpp system to dissect the pathways of HBV entry and dissemination.

We describe a novel papillomavirus – Rusa alfredi papillomavirus 1 (RalPV1) – which causes endemic fibropapillomatosis in the European conservation breeding population of the highly endangered Visayan spotted deer (Rusa alfredi). Degenerated papillomavirus-specific primers were used to amplify and sequence parts of the viral DNA. Subsequently, the complete genomic DNA was cloned and the sequence was determined. The RalPV1 genome has a length of 8029 bp, encodes the early proteins E6, E7, E1, E2 and E5, the two late proteins L1 and L2 and contains an upstream regulatory region. Highest sequence identities were observed with two deltapapillomaviruses, the Capreolus capreolus PV1 and Cervus elaphus PV1. Pairwise comparisons and phylogenetic analysis based on the ORF L1 suggested that RalPV1 is a putative new type of the papillomavirus species Deltapapillomavirus 5.

Previous reports have demonstrated that hepatitis B virus (HBV) X protein (HBx) represses E-cadherin expression to induce epithelial–mesenchymal transition (EMT), an essential component of cancer progression to more aggressive phenotypes characterized by tumour invasion, migration and metastasis; however, the underlying mechanism for this phenomenon is still unclear. In this study, we found that ectopic expression of HBx in human hepatocytes using overexpression and 1.2-mer WT HBV replicon systems upregulated levels of the transcriptional repressors E12 and E47, resulting in inactivation of the E-cadherin promoter, containing three E-box motifs, and subsequent repression of its expression. E12/E47 knockdown using a specific small interfering RNA almost completely abolished the potential of HBx to repress E-cadherin expression. HBx inhibited the ubiquitin-dependent proteasomal degradation of E12/E47 without affecting their expression at the transcriptional level. Upregulation of E12/E47 by HBx ultimately led to EMT in human hepatocytes, as demonstrated by morphological changes, altered protein levels of EMT markers, including E-cadherin, plakoglobin, fibronectin, vimentin and N-cadherin, and increased capacity for cell detachment and migration.

Cyclin-dependent kinases (CDKs) are multifaceted regulators involved in the replication of human cytomegalovirus. Recently, we demonstrated an interaction of CDK9–cyclin T1 as well as viral CDK orthologue pUL97 with the viral regulator pUL69, thereby leading to pUL69-activating phosphorylation. Here, we demonstrate that colocalization and direct pUL69–cyclin T1 interaction is independent of viral strains and host cell types. In vitro phosphorylation of pUL69 by CDK9 or pUL97 did not occur in a single site-specific manner, but at multiple sites. The previously described fine-speckled nuclear aggregation of pUL69 was assigned to the late phase of viral replication. CDK inhibitors, including a novel inhibitor of the CDK-activating kinase CDK7, massively intensified this fine-speckled accumulation. Interestingly, we also observed spontaneous pUL69 accumulation in the absence of inhibitors at a lower frequency. These findings provide new insight into pUL69 kinase interregulation and emphasize the importance of pUL69 phosphorylation for correct intranuclear localization.

Colorectal cancer (CRC) is amongst the leading causes of cancer-related mortality worldwide. Emerging evidence suggests that human cytomegalovirus (HCMV) exists in the tumour tissue of CRC and is associated with disease outcome. To study whether tumoral HCMV is related to viral reactivation in blood, tumour specimens and pre- and post-operative blood samples from CRC patients were collected prospectively. PCR and quantitative PCR were performed to detect HCMV DNA. HCMV IgG and IgM antibodies were measured using a microparticle enzyme immunoassay. Transcription of a spliced HCMV UL73 gene transcript was analysed by quantitative reverse transcription PCR. HCMV was detected in 42.2 % (35/83) of the tumour samples, with a low median viral load (30.08, range 2.33–5704 copies per 500 ng genomic DNA). The vast majority (80/81, 98.8 %) of the CRC patients were seropositive for HCMV IgG. HCMV DNA was positive in 11.3 % (22/194) of the pre-operative and 8.9 % (15/168) of the post-operative blood samples. However, presence of HCMV and its viral load in tumours were not associated with the detection or viral loads in blood samples. About 26.67 % (8/30) of the HCMV-positive tumours with available RNA had detectable viral UL73 transcripts, whilst none of the blood samples were positive for viral RNA (P < 0.0001). Therefore, presence of HCMV in tumours does not correlate with the serological or viraemic status of CRC patients. Active viral gene transcription occurred in the tumour but not in the blood of CRC patients. HCMV reactivation in CRC patients is possibly due to virus–cancer interactions in the CRC tumour microenvironment.

Commonly used human adenovirus (HAdV)-5-based vectors are restricted by their tropism and pre-existing immunity. Here, we characterized novel HAdV-5 vectors pseudotyped with hypervariable regions (HVRs) and surface domains (SDs) of other HAdV types. Hexon-modified HAdV-5 vectors (HV-HVR5, HV-HVR12, HV-SD12 and HV-SD4) could be reconstituted and amplified in human embryonic kidney cells. After infection of various cell lines, we measured transgene expression levels by performing luciferase reporter assays or coagulation factor IX (FIX) ELISA. Dose-dependent studies revealed that luciferase expression levels were comparable for HV-HVR5, HV-SD12 and HV-SD4, whereas HV-HVR12 expression levels were significantly lower. Vector genome copy numbers (VCNs) from genomic DNA and nuclear extracts were then determined by quantitative real-time PCR. Surprisingly, determination of cell- and nuclear fraction-associated VCNs revealed increased VCNs for HV-HVR12 compared with HV-SD12 and HV-HVR5. Increased nuclear fraction-associated HV-HVR12 DNA molecules and decreased transgene expression levels were independent of the cell line used, and we observed the same effect for a hexon-modified high-capacity adenoviral vector encoding canine FIX. In conclusion, studying hexon-modified adenoviruses in vitro demonstrated that HVRs but also flanking hexon regions influence uptake and transgene expression of adenoviral vectors.

Equid herpesvirus 5 (EHV-5) is related to the human Epstein–Barr virus (human herpesvirus 4) and has frequently been observed in equine populations worldwide. EHV-5 was previously assumed to be low to non-pathogenic; however, studies have also related the virus to the severe lung disease equine multinodular pulmonary fibrosis (EMPF). Genetic information of EHV-5 is scanty: the whole genome was recently described and only limited nucleotide sequences are available. In this study, samples were taken twice 1 year apart from eight healthy horses at the same professional training yard and samples from a ninth horse that was diagnosed with EMPF with samples taken pre- and post-mortem to analyse partial glycoprotein B (gB) gene of EHV-5 by using next-generation sequencing. The analysis resulted in 27 partial gB gene sequences, 11 unique sequence types and five amino acid sequences. These sequences could be classified within four genotypes (I–IV) of the EHV-5 gB gene based on the degree of similarity of the nucleotide and amino acid sequences, and in this work horses were shown to be identified with up to three different genotypes simultaneously. The observations showed a range of interactions between EHV-5 and the host over time, where the same virus persists in some horses, whereas others have a more dynamic infection pattern including strains from different genotypes. This study provides insight into the genetic variation and dynamics of EHV-5, and highlights that further work is needed to understand the EHV-5 interaction with its host.

Several herpesviruses have acquired the gene for the CD200 membrane protein from their hosts and can downregulate myeloid activity through interaction of this viral CD200 orthologue with the host receptor for CD200, namely CD200R, which can give inhibitory signals. This receptor is a ‘paired receptor’, meaning proteins related to the inhibitory CD200R are present but differ in that they can give activating signals and also give a negligible interaction with CD200. We showed that the viral orthologues e127 from rat cytomegalovirus and K14 from human herpesvirus 8 do not bind the activating CD200R-like proteins from their respective species, although they do bind the inhibitory receptors. It is thought that the activating receptors have evolved in response to pathogens targeting the inhibitory receptor. In this case, the CD200 orthologue is not trapped by the activating receptor but has maintained the specificity of the host from which it was acquired, suggesting that the activating members of the CD200R family have evolved to protect against a different pathogen.

Cytomegalovirus (CMV) is a ubiquitous virus, causing the most common congenital infection in humans, yet a vaccine against this virus is not available. Experimental studies of immunity against CMV in animal models of infection, such as the infection of mice with mouse CMV (MCMV), have relied mainly on parenteral infection protocols, although the virus naturally transmits by mucosal routes via body fluids. To characterize the biology of infections by mucosal routes, we compared the kinetics of virus replication, latent viral load and CD8 T-cell responses in lymphoid organs upon experimental intranasal (targeting the respiratory tract) and intragastric (targeting the digestive tract) infection with systemic intraperitoneal infection of two unrelated mouse strains. We observed that intranasal infection induced robust and long-term virus replication in the lungs and salivary glands but limited replication in the spleen. CD8 T-cell responses were somewhat weaker than upon intraperitoneal infection but showed similar kinetic profiles and phenotypes of antigen-specific cells. In contrast, intragastric infection resulted in abortive or poor virus replication in all tested organs and poor T-cell responses to the virus, especially at late times after infection. Consistent with the T-cell kinetics, the MCMV latent load was high in the lungs but low in the spleen of intranasally infected mice and lowest in all tested organs upon intragastric infection. In conclusion, we showed that intranasal but not intragastric infection of mice with MCMV represents a robust model to study the short- and long-term biology of CMV infection by a mucosal route.