Two related Epstein-Barr virus membrane proteins are encoded by separate genes.

The structures of the 2.3- and 2.0-kilobase Epstein-Barr virus (EBV) mRNAs, partially encoded within the EcoRI J fragment DNA of the viral genome, were determined by analysis of their cDNAs. Both mRNAs are transcribed across the fused terminal repeats of the EBV episome and consist of nine exons. The mRNAs are transcribed from different promoters and have a unique 5' exon from the U5 region of the genome but eight common exons from the U1 region. One principal open reading frame is present in each mRNA and is predicted to encode 54,000- and 40,000-dalton integral membrane proteins. This result was confirmed by in vitro translation of RNAs in the presence of canine pancreatic microsomes. The 2.3-kilobase mRNA is not expressed in Raji cells, owing to the deletion of the 5' regulatory and coding region of this gene, whereas neither mRNA is expressed in Namalwa cells, owing to inactivation as a result of integration of the EBV genome via the terminal repeats. Since these mRNAs are readily detected in largely latently infected cells and do not increase in abundance with EBV replication, these putative latent-infection membrane proteins are tentatively designated LMP-2A and LMP-2B, respectively.     

Cytoplasmic expression of mRNAs containing the internal ribosome entry site and 3' noncoding region of hepatitis C virus: effects of the 3' leader on mRNA translation and mRNA stability

Translation initiation in many eukaryotic mRNAs is modulated by an interaction between the cap binding protein complex, bound to the 5' end of the mRNA, and the polyadenosine binding protein, bound to the 3'-terminal polyadenosine sequences. A few cellular and viral mRNAs, such as the hepatitis C virus (HCV) mRNA genome, lack 3'-terminal polyadenosine sequences. For such mRNAs, the question of whether their 3'-end sequences also regulate the initiation phase of protein synthesis via an interaction with their 5' ends has received intense scrutiny. For HCV mRNA, various experimental designs have led to conflicting interpretations, that the 3' end of the RNA can modulate translation initiation either in a positive or in a negative fashion. To examine the possibility of end-to-end communication in HCV in detail, mRNAs containing the HCV internal ribosome entry site linked to a luciferase coding region, followed by different 3' noncoding regions, were expressed in the cytoplasm of cultured cells by T7 RNA polymerase. The intracellular translation efficiencies, steady-state levels, stabilities, and 3'-end sequences of these chimeric RNAs were examined. It was found that the HCV 3' noncoding region modulates neither the translation nor the stability of the mRNAs. Thus, there is no detectable end-to-end communication in cytoplasmically expressed chimeric mRNAs containing the HCV noncoding regions. However, it remains an open question whether end-to-end communication occurs in full-length HCV mRNAs in the infected liver.     

Integration and transcription of group C human adenovirus sequences in the DNA of five lines of transformed rat cells

We have used the Southern blotting technique to analyze the integration patterns of human adenovirus sequences in the DNA of four rat cell lines, F17, 8617, T2C4 and F4, which were transformed by Ad-2 virus, and 5RK clone 6, which was transformed by Ad-5 HindIII-G fragment. We have also analyzed the Ad-specific messenger RNAs synthesized in these cell lines, in 293 cells (an Ad-5 transformed human cell line), and in Ad-2 early infected human KB cells, using the RNA geltransfer hybridization technique. We were interested in whether the Ad sequences are integrated, what the integration patterns are, whether the transforming region is present in an intact form, and whether the transforming region and other early regions are expressed at the mRNA level.Our results show that the integration patterns of Ad sequences range from simple to quite complex. Cells from line 8617 contain a single copy of right-end sequences flanked by left-end sequences. T2C4 cells have four different left-end sequences and two different right-end sequences. 5RK cells contain multiple different pieces of left-end sequences. In agreement with the results of Sambrook et al. (1979a,b), F17 cells contain a single copy of the left 17% of the genome, and F4 cells contain multiple copies of the right 5% of the genome fused to the left ~ 68% of the genome. The complete Ad genome is not present in any of the cell lines, and different regions may not be equimolar. There are no specific sites on the cellular or viral genome at which integration occurs. In 8617, F17 and F4 cells the Ad-2 sequences appear to be located close together on a single chromosome, suggesting that the Ad sequences in these cells arose from a single integration event. F17, 8617, T2C4, F4, and probably 5RK, cells all have an intact early region E1a (map position 1·3–4·6); F17, 8617, T2C4 and F4 cells also have E1b (m.p. 4·6–11·2) intact. E1a and E1b are the regions responsible for transformation. 8617 cells also have an intact early region E4 (m.p. 99-91·5) and T2C4 cells have an intact early region E3 (m.p. 76–86).Ad-2 early infected KB cells were shown to synthesize major E1a-specific mRNAs of 13 S, 12 S and 9 S, and major E1b-specific mRNAs of 22 S and 13 S. All the transformed cells synthesize the E1a 13 S and E1a 12 S mRNAs, and all cells except 5RK synthesize the E1b 22 S and E1b 13 S mRNAs. Early infected KB cells synthesize E3-specific mRNAs of 26 S, 24 S, 22 S, 19 S, 12 S and 9 S: T2C4 cells synthesize the major 22 S and 19 S RNA species, and possibly the less pronounced E3 mRNAs. Early infected cells and 8617 cells synthesize E4-specific mRNAs of 19 S, 17 S, 14 S, 12 S, 11 S, 9 S and 8 S. 8617 cells also synthesize E4 mRNAs of about 23 to 24 S and 21 S. F4 cells synthesize 24 S and 19 S hybrid mRNAs that contain both E4 and E1a sequences: these RNAs arise because F4 cells contain a portion of the E4 region fused to the left end (m.p. 0) of the genome.Our results, as well as those from other laboratories, are consistent with the idea that the transformed phenotype of Ad transformed cells is maintained by expression of Ad genes in E1a and E1b.     

Genomic deletions created upon LINE-1 retrotransposition

LINE-1 (L1) retrotransposition continues to impact the human genome, yet little is known about how L1 integrates into DNA. Here, we developed a plasmid-based rescue system and have used it to recover 37 new L1 retrotransposition events from cultured human cells. Sequencing of the insertions revealed the usual L1 structural hallmarks; however, in four instances, retrotransposition generated large target site deletions. Remarkably, three of those resulted in the formation of chimeric L1s, containing the 5' end of an endogenous L1 fused precisely to our engineered L1. Thus, our data demonstrate multiple pathways for L1 integration in cultured cells, and show that L1 is not simply an insertional mutagen, but that its retrotransposition can result in significant deletions of genomic sequence.