Characterization Of Heterotypic Interaction Between E1B55K And E4orf6Proteins From Adenovirus Type5and41

Adenovirus (Ad) vector is widely used in gene therapy and recombinant vaccine because of its distinctive characteristics, such as the ease of manipulation, effective gene delivery and expression, the capability to grow to high titer and its biosafety as an episomal vector. The commonly used Ad vector is constructed based on Ad2and Ad5genome (adenovirus species C), and HEK293cells, which constitutively express Ad5E1genes, are used as the packaging cells. Human adenoviruses are classified into seven species (A-G) including57types. Generally, different species has different tropism (targeting different organs or tissues). The development of adenoviral vectors from different types has many advantages. For example, tissue-specific characteristic can guide recombinant virus to targeting organ, and infect cells of interest more efficiently with less amount of vector administration. On the other hand, change in use of adenoviral vectors from various types helps escape host immune response against vectors. However,293cells can not be used to package all the El-deleted adenoviruses. The reason may be that the E1genes of Ad5can not perfectly complement the function of those of other adenovirus. The molecular mechanism under this is unknown. The development of adenoviral vectors based on various types is still labor-intensive and time-consuming since establishment of packaging cells is often indispensable.Ad41belongs to adenovirus species F. It can cause diarrhea in infants and young children, so it is also called enteric adenovirus. Because it is a natural gastrointestinal pathogen, it has great potential to be developed into a gene delivery vector targeting gastrointestinal tract. Ad41is also well known for its fastidious culture characteristic. In2008, an Ad41E1B55K-transduced293cell line (293E12) was established in our lab.293E12can be used for packaging of wild-type or recombinant Ad41. To further increase the propagation of Ad41vector, a cell line with improved expression of Ad41E1B55K (293TE7) was established in our lab.In this study, we evaluated the packaging capability of293TE7firstly. E1B55K is the early protein of adenovirus, and the multiple functions of E1B55K require its interaction with another adenovirus early protein E4orf6. Several reports have demonstrated that Ad5E1B55K connects with Ad5E4orf6to form an E3ubiquitin ligase complex, which also includes cellular protein Cullin5, Rbxl and Elongin B/C. The E3ubiquitin ligase complex sequesters specific cellular substrates (p53, Mre11, DNA ligase Ⅳ, Integrin α3) into ubiquitin/proteasomal degradation pathway and is essential for adenovirus replication. Ad41can’t be grown well in293cells, but it can be cultivated in293E12and293TE7cells which stably expressed Ad41E1B55K protein. Based on these facts, we assumed that Ad5E1B55K expressed in293cells could not substitute the function of Ad41E1B55K. To reveal the molecular mechanism why293TE7cells instead of293cells can be used in recombinant Ad41culture, interaction between heterotypic E1B55K and E4orf6and the consequence were studied in this study.1. Evaluation of the packaging capability of293TE7cellsHuman adenovirus type41(HAdV-41) is notoriously difficult to cultivate under laboratory conditions. Tripartite leader sequence (TPL) of HAdV-41was cloned, inserted into eukaryotic expression plasmid, and used to establish a HAdV-41E1B55K-transduced cell line (293TE7). HAdV-41E1B55K was expressed more abundantly in293TE7than in293E12, a previously developed HAdV-41ElB55K-expressing cell line. After being infected with E1-deleted HAdV-41vector (HAdV-41-GFP),293TE7synthesized more viral genomic DNA and structural proteins, which led ultimately to a significant increase of the yield of progeny viruses. Typically,293TE7produced progeny viruses3-15times more than293E12did, depending on the amount of seed viruses and culture time. These data demonstrated that293TE7was an effective packaging cell line, and implied its application in wild-type HAdV-41isolation, HAdV-41virological study and recombinant HAdV-41construction.2. The interaction between heterotypic E1B55K and E4orf6proteinIn this section, we studied the interaction of adenovirus E1B55K and E4orf6from Ad5and Ad41, the formation of E3ubiquitin ligase complexes and degradation of the substrates (p53and Mre11) of the complex. To facilitate the detection of E1B55K and E4orf6proteins, HA tag or Flag tag was added to the N-terminal of them, respectively.(1) Plasmid construction. First, Ad5E1B55K and Ad41E1B55K carrying HA tag were amplified by PCR and then cloned to pcDNA3to construct p5HAE1B and p41HAElB expression vector; meanwhile, Ad5E4orf6and Ad41E4orf6carrying Flag tag were amplified and cloned to pcDNA3to construct p5FlagE4and pcDNA3-SP41FE4N2expression vector. Many researchers employed cDNA to generate plasmid or viral vectors to express E4orf6in the absence of other viral products. Besides the normal34kDa product, such vectors consistently produce a polypeptide of about8kDa. During the PCR amplification of E4orf6, PCR-directed mutagenesis was employed to eliminate the aberrant splice product. The expression of E1B55K and E4orf6were detected by indirect immunofluorescence and western blot after transfecting to H1299cells.(2) The interaction between E1B55K and E4orf6and the degradation of the substrates.H1299cells were transfected with plasmids encoding Ad5E4orf6and Ad5E1B55K or Ad41E1B55K. The whole cell extracts were subjected to co-immunoprecipitation, and the results demonstrated that Ad5E4orf6could connect with Ad5E1B55K or Ad41E1B55K to form E3ubiquitin ligase complex. However, the E3ubiquitin ligase complex formed by Ad5E4orf6and Ad41E1B55K could not degrade the p53and Mrell. H1299cells were transfected with plasmids encoding Ad41E4orf6and Ad5E1B55K or Ad41E1B55K. The whole cell extracts were subjected to co-immunoprecipitation, and the results demonstrated that Ad41E4orf6could connect with not only Ad41E1B55K but also Ad5E1B55K to form E3ubiquitin ligase complex. Unexpectly, any E3ubiquitin ligase complexes formed could degrade the p53and Mre11.(3) The effect of E3ubiquitin ligase complex on viral late mRNA nuclear export.Cytoplasmic and nuclear mRNAs were extracted from293cells or293TE7cells after infected with Ad41GFP at an MOI of150vp/cell at indicated times, respectively. Real-time quantitative PCR was employed to analyze the viral late mRNA nuclear export. For protein penton, the ratio of cytoplasmic to nuclear mRNA in293TE7cells was2.3times to293cells at24hours postinfection with Ad41GFP, and for protein long fiber, this value was4and it was2.5times for protein hexon. However, this ratio was decreased to0.5for beta-actin which belongs to cellular endogenous protein. The results demonstrated that expression of the Ad41E1B55K protein in293TE7cells caused an increase in viral late protein (long fiber, penton, hexon) synthesis and promoted viral nuclear mRNA export, meanwhile, it also inhibited host cell mRNA export. That is to say, the Ad5E1B55K expressed in293cells could not perfectly substitute the function of Ad41E1B55K which was stably expressed in293TE7cells.In conclusion, the interaction between E1B55K and E4orf6and the study of the substrates degraded by E3ubiquitin ligase complex suggested that293cells might be used for effective packaging of recombinant Ad41(Ad5E1B55K could interact with Ad41E4orf6and further degraded substrates). However, nuclear export efficiency of viral late mRNA in293cells was lower than that in293TE7cells after Ad41infection. It partly explained the reason why293TE7cells were more efficient for Ad41propagation.293TE7is an ideal packaging cell line for the rescue and amplification of recombinant Ad41. After characterization of the heterotypic interaction between E1B55K and E4orf6protein, we found that the Ad5E1B55K could not replace the function of Ad41E1B55K, which might result from the relatively lower nuclear export efficiency of the viral late mRNA. This study preliminarily clarified the molecular mechanism of fastidious growth of Ad41in293cells. Moreover, our resutls suggested that the E3ubiquitin ligase complex could not explain all the functions of the E1B55K and E4orf6protein, or there might be other unknown substrates for this complex.