We previously showed that the right third of HindIII fragment L (0.59 to 0.65) of herpes simplex virus type 1 (HSV-1) encodes a family of mRNAs some members of which appear to be related by splicing. In the experiments described in this

We described the detailed characterization and high-resolution mapping of nine herpes simplex virus type 1 mRNAs encoded in EcoRI fragment I. Four of these mRNAs are partially colinear and encode the same sized polypeptide in vitro. Nucleotide sequence analysis of the DNA around the 5' ends of these mRNAs suggested that the larger may encode a small (ca. 100-dalton) polypeptide not resolvable by in vitro translation.

We have isolated as recombinant DNA clones, in the plasmid pBR322, regions of the herpesvirus type 1 genome spanning the region between 0.53 and 0.6 on the prototypical arrangement. This 11,000-base-pair region corresponds to 10% of the large unique region and encodes five major and several minor mRNA species abundant at different times after infection, which range in length from 7 to 1 kilobase. In this report, we have used RNA transfer blots and S1 nuclease digestion of hybrids between viral DNA and polyribosomal RNA to precisely localize (±0.1 kilobase) these mRNA's. Comparison of neutral and alkaline gels of S1 nuclease-digested hybrids indicates no intemal introns in the coding sequences of these mRNA's, although noncontiguous leader sequences near (ca. 0.1 kilobase) the 5' ends of any or all mRNA's could not be excluded. The 5' ends of several late rnRNA's that are encoded opposite DNA strands map very close to one another, and the 3' ends of a major late and a major early mRNA, which are partially colinear, terminate in the same region. In vitro translation of the viral mRNA's isolated by hybridization with DNA bound to cellulose and fractionation of mRNA species on denaturing agarose gels allowed us to assign specific polypeptide products to each of the mRNA's characterized. Among other results, it was demonstrated unequivocally that two major late mRNA's, which partially overlap, encode the same polypeptide.

We have used DNA bound to cellulose to isolate and translate in vitro herpes simplex virus type 1 (HSV-1) mRNA's encoded by HindIII fragment L (mapping between 0.592 and 0.647), an 8,450-base-pair (8.45-kb) portion of the long unique region of the viral genome. Readily detectable, late mRNA's 2.7 and 1.9 kb in size encoding 69,000-and 58,000-dalton polypeptides, respectively, were isolated. A very minor late mRNA family composed of two colinear forms, one 2.6 kb and one 2.8 kb, was isolated and found to encode only an 85,000-dalton polypeptide. A major early mRNA, 1.8 kb in size encoding a 64,000-dalton polypeptide, was also isolated. High-resolution mapping of these mRNA's by using Si nuclease and exonuclease VII digestion of hybrids between them and 5' and 3' end-labeled DNA fragments from the region indicated that the major early mRNA contained no detectable splices, and about half of its 3' end was complementary to the 3' region of the very minor 2.6-to 2.8-kb mRNA's encoded on the opposite strand. These mRNA's also contained no detectable splices. The major late 2.7-kb mRNA was found to be a family made up of members with no detectable splices and members with variable-length (100 to 300 bases) segments spliced out very near (ca. 50 to 100 bases) the 5' end. Like all herpesviruses, herpes simplex virus type 1 (HSV-1) is characterized by an unusual arrangement of its genome. The viral DNA has a molecular weight of 95 x 106 to 100 x 106 daltons (d; reviewed in reference 44), which corresponds to a length of 150,000 base pairs (150 kb; 50). The linear HSV-1 genome is segmented into a long unique region (UL, ca. 105 kb, 70%)

The sequences ofthe DNAs encoding the 5' ends of one early and one late herpes simplex virus type 1 mRNA were analyzed, and the 5' ends of these mRNA species were precisely located. Neither mRNA species is spliced and the noncoding strand ofthe DNA contains recognizable T-A-T-A and C-A-T boxes upstream from their respective 5 The question ofwhat factors control the temporal regulation of DNA virus gene expression is an active field ofresearch. The development of faithful uninfected cell transcription systems (3), as well as other techniques, has led to the following observations concerning this regulation in the smaller DNA viruses. In the papovaviruses (3, 4), the uninfected cell polymerase system favors the recognition of promoters for early genes. However, this preference is not absolute; at relatively high template DNA concentrations, the late promoters are efficiently recognized. In adenovirus (5), the situation is somewhat different in that the major late promoter appears to be readily recognized during the early phases of gene expression. Here, the appearance of abundant late mRNA involves changes in processing patterns for the late mRNA precursors.We have recently reported the detailed characterization of a number of early and late HSV-1 mRNAs (6-8). Interestingly, although HSV-1 mRNAs share general properties of cellular mRNA, splicing of HSV-1 mRNAs is rare. To date, only two well-characterized viral mRNAs, one a (9) and one y (8), have recognizable splices in them. Such a low incidence of splicing suggests that posttranscriptional processing of intervening sequences does not play a major role in the biogenesis of many HSV-1 mRNAs.In this study, we have examined the nucleotide sequences of the DNAs encoding the 5' ends of two well-characterized major unspliced HSV-1 mRNAs (6, 7). The DNA sequences upstream of these mRNAs share features with the HSV-1 thymidine kinase (tk) gene (10, 11) and with many other eukaryotic mRNAs (12, 13). The promoter for the 5.2-kilobase (kb) early mRNA was readily recognizable by the "Manley" HeLa cell polymerase system (3), while that for the 6-kb late mRNA was not. Preliminary data indicate that this difference between early and late HSV-1 mRNA recognition is general.MATERIALS AND METHODS Cells and Virus. Monolayer cultures of HeLa cells were grown at 370C in Eagle's minimal essential medium/10% calf serum without antibiotics. Plaque-purified virus of the KOS strain of HSV-1 was used for all infections.Enzymes. All enzymes were obtained from Bethesda Research Laboratories (Rockville, MD) and digestion was carried out in buffers recommended by that supplier.Isolation, Labeling, and Size Fractionation ofPolyribosomal RNA. Viral RNA synthesized in the absence of HSV-1 DNA synthesis (early RNA) was prepared as described (6). The drugs were a gift of C. Shipman of the University of Michigan.Polyribosomes were isolated from the cytoplasm of HSV-1-infected cells by the magnesium precipitation method of Palmiter (14) as described (6-8). Details of the isolation of poly(A)...