key: cord-0844638-ov92qh6h
authors: Nowak, Barbara; Gmeiner, Agnes; Sarnow, Peter; Levine, Arnold J.; Fleckenstein, Bernhard
title: Physical mapping of human cytomegalovirus genes: Identification of DNA sequences coding for a virion phosphoprotein of 71 kDa and a viral 65-kDa polypeptide
date: 1984-04-15
journal: Virology
DOI: 10.1016/0042-6822(84)90275-7
sha: f34b197c85846303884872e78abe532b8cdcb725
doc_id: 844638
cord_uid: ov92qh6h

Abstract Polyadenylated RNA was isolated from fibroblast cultures infected with human cytomegalovirus (HCMV) strain AD169 during the late phase of viral replication. The RNA was selected by hybridization to a series of cosmid clones containing the entire viral genome in partially overlapping segments. Translation of this RNA in a reticulocyte cell-free system allowed the mapping of virus specific polypeptides. Nine polypeptides synthesized in vitro comigrated with major virion structural proteins. An in vitro-translated protein of 71 kDa was precipitated by a monoclonal antibody directed against the phosphorylated internal envelope protein of 71 kDa. The map coordinates of viral DNA coding for this phosphoprotein were localized by hybrid selection with subcloned DNA fragments, and the direction of transcription was determined by hybrid selection with single-stranded DNA cloned in bacteriophage vector M13mp9. An in vitro translation with size-fractionated RNA, combined with immunoprecipitation and Northern blot analyses, indicated that an mRNA of 4 kb encodes the 71-kDa phosphoprotein. An mRNA of the same size, map coordinates, and orientation was translated into an abundant 65-kDa polypeptide which had the same size as the major structural phosphoprotein of HCMV.

The genome of human cytomegalovirus (HCMV) AD169 is a linear, doublestranded DNA molecule of about 235 kb (Geelen et uZ., 1978; Fleckenstein et a& 1982) . Like some other herpesvirus DNA molecules, the HCMV genome consists of a large (L) and a small (S) unique copy segment, which can be oriented in either of two directions. This results in four isomerit conformations of virion DNA. These unique regions are bracketed by a pair of inverted repeats (LaFemina and Hayward, 1980; Ebeling et al, 1983) . The HCMV genome has been cloned in plasmid (Thomsen and Stinski, 1981; Oram et al, 1982; Tai aL, 1982) and cosmid vectors (Fleckenstein et al, 1982) , providing a basis for the physical and functional mapping of viral DNA. A small, contiguous region of the viral genome is abundantly transcribed into three or four species of RNA during the immediate early phase of viral replication (DeMarchi, 1981; Wathen and Stinski, 1982; McDonough and Spector, 1983; Jahn et aL, submitted) , and the coding region for one immediate early gene product of 72 kDa has been identified (Stinski et aL, 1983) . On the other hand, transcripts from most regions of the viral genome are found in infected cells during subsequent phases of viral replication (DeMarchi, 1981; Wathen and Stinski, 1982; McDonough and Spector, 1983) , but no correlation has been reported between a defined virion protein and its respective coding sequence. This study describes a procedure which may be generally useful for the identification of late viral genes. Sequences of DNA coding for viral proteins were mapped by mRNA hybrid selection, followed by in vitro translation of late mRNA. Immunoprecipiation of these in vitro-synthesized polypeptides with a monoclonal antibody identified the coding region of a phosphorylated structural protein of 71 K molecular weight (Nowak et aZ., 1984) . Analysis of the transcripts and translated products indicated that the same region of the genome also encoded an abundant viral protein of 65 kDa.

Virus and cell culture. HCMV strain AD169 was propagated on human foreskin fibroblasts (HFF) using standard procedures. To obtain pure virus stocks, virion DNA was transfected by the calcium phosphate precipitation method (Graham and van der Eb, 1973; Copeland and Cooper, 1979) , and resulting virus was plaque purified. Labeled virus was purified from cultures incubated with 50 &i/ml [""S]methionine (1200 Ci/mmol. Amersham-Buchler, Braunschweig) at 4 days postinfection (Ebeling et al, 1983) . Proteins from cell extracts were prepared from [35S]methionine-labeled, AD169-or mockinfected cells as described . For isolation of immediate early (IE) RNA, cells were infected and maintained in the presence of 0.1 mg/ml cycloheximide, and total cellular RNA was harvested 14 to 16 hr postinfection. Late RNA was isolated after 4 to 6 days of infection. Mockinfected cell RNA was extracted from confluent, uninfected HFF cultures.

Cloning of DNA in plplasmids and bacteriophage Ml3 Recombinant clones containing subfragments of cosmid-cloned AD169 DNA (Fleckenstein et al. 1982) was constructed by using the plasmid vector pAcyc148 (Chang and Cohen, 1978) . Procedures for isolation of cosmid and plasmid DNA and construction of recombinant clones were carried out as described (Fleckenstein et aL, 1982; Knust et c& 1983) . Each cloned viral DNA fragment was identified by separation on agarose gels after restriction endonuclease digestion and Southern blot hybridization (Southern, 1975) . Cloned, single-stranded viral DNA was obtained by using the bacteriophage derivative M13mp9 and the procedure described by Messing et al. (1981 Messing et al. ( , 1982 . Isolation of RNA, size fracticmaticm, and Northern blot anal&s. Cellular RNA was isolated from cultures by a modification of the guanidine hydrochloride method (Liu et aL, 1979) . HCMV-or mock-infected cells were rinsed twice with ice-cold phosphatebuffered saline (PBS), scraped into PBS, centrifuged at 1500 rpm for 5 min, and washed once more. The cells (10' to 5 X 10' per pellet) were resuspended in 7 ml of lysis buffer (6 M guanidine hydrocholride, 0.1 M P-merceptoethanol, 0.5% N-lauroylsarkosine, 50 mM lithium citrate, pH 6.5). The lysate was layered on a cushion of 4 ml of 5.7 M CsCl in 0.1 M EDTA, pH 7.5. After centrifugation in a Spinco SW41 rotor at 28,000 rpm at 20' for 20-24 hr, the gelantinous pellet was dissolved in a small amount of HaO. Following ethanol precipitation and digestion with proteinase K, the RNA was extracted with phenol and chloroform-isoamyl alcohol and precipitated with ethanol. Polyadenylated [poly (A)+] RNA was prepared by passing total RNA over a column of oligodeoxythymidylic acid-cellulose (Bethesda Research Laboratories) (Aviv and Leder, 1972) . Poly (A)+ RNA was size fractionated on sucrose-formamide gradients (Siddell et al., 1980) . A 50-pg sample of poly (A)+ RNA was denatured in 60% deionized formamide, 10 mM Tris-HCI, pH 7.5,1 mM EDTA at 37" for 5 min before centrifugation. The RNA was layered on top of a to 5 to 20% sucrose formamide gradient containing 10 mM Tris-HCI, pH 7.5, 1 mM EDTA, 100 mM LiCl, 0.5% sodium dodecyl sulfate (SDS) and 50% deionized formamide. The gradients were centrifuged for 20 hr at 28,000 rpm and 20" in an SW41 rotor. Fractions were collected from the bottom of the gradient, and 30 pg of calf liver tRNA per ml was added as a carrier. The RNA of each fraction was precipitated with ethanol and divided into two parts. One aliquot was analyzed by Northern blot hybridization, the other part was used for in vitro translation assays, Labeled ribosomal RNA from owl monkey (28 S, 5.2 kb; 18.5 S, 2.0 kb) and Escherichia coli ( Eight cosmid clones representing the entire viral genome are indicated by horizontal brackets (Fleckenstein et al, 1982 to cloned viral probes. The respective cosmid clones are marked by their numbers above the lanes. Selected mRNA was eluted and translated in a reticulocyte cell-free system containing [85S]methionine. The translation products were ana-S, 1.5 kb) were used for size estimation. For Northern blot analysis, RNA was denatured with glyoxal (McMaster and Charmichael, 1977) . The RNA was fractionated by electrophoresis on 1% agarose gels, transferred to nitrocellulose filters (Thomas, 1980) , and hybridized with nickrepair, 3zP-labeled DNA probes (Knust et al, 1983 ).

H&ridization select&n and cd-free translation. The procedure for hybrid selection (Ricciardi et d, 1979) was modified, following suggestions of H. Esche (personal communication). Twenty micrograms of cosmid-plasmidor phage-cloned viral DNA was bound to a OA-cm-diameter nitrocellulose filter (Schleicher-Schull, BA 85). Cytoplasmic polyadenylated RNA (0.15-0.2 mg/ml) was incubated for 4 hr at 56" with several filter pieces in 20 mM PIPES (1,4-piperazinediethanesulfonic acid) (pH 6.4), 0.2% SDS, 50% formamide, lyzed on a 15% SDS-PAGE, followed by fluorography. The sizes of polypeptides, indicated in kilodaltons, were determined by coelectrophoresis of marker proteins. Lane a, p"S] methionine-labeled proteins of purified AD169 virions; lane b, translated products of nonselected late poly (A)+ RNA from AD169-infected cells; lane c, translated products of nonselected RNA from mock-infected cells; lane d, endogenous proteins synthesized in the cell-free system without addition of RNA. Note. Sizes of polypeptides (X10-*) obtained by in vitro translation of late RNA selected with seven cosmid clones covering the entire HCMV genome. These polypeptides were detected in repeated experiments. 

' Identified as structural, phosphorylated viral proteins by monoclonal antibodies (Nowak et al, 1984) .

b Comigrating with phosphorylated virion proteins.

cording to the manufacturer's direction (Amersham-Buchler, Braunschqeig). Immunoprecipiation and SDS-polyacrylamide gel electrophoresis. Precipitations were carried out as described by Ross et al. (1980a) . After preincubation with Staphylococcus aureus (Kessler, 1975) , samples of in vitro translation assays were incubated on ice for 30 min with human immune sera or monoclonal antibodies. Antigen-antibody complexes were adsorbed to protein A bearing S. aureua The washed immunoprecipitates were denatured and solubilized by the addition of SDS-sample buffer (Laemmli, 1970) , followed by boiling for 3 min. In vitro translation products and immunoprecipitates were separated on 15% linear polyacrylamide gels (PAGE) (Bodemer et aL, 1980) . Fixed gels were fluorographed by the method of Bonner and Laskey (1974) .

For hybridization selection of late viral RNA, eight cosmid clones were employed which cover the entire genome of HCMV strain AD169 (Fig. 1) . The clones were derived from partial cleavage of purified virion DNA by Hind111 (Fleckenstein et o2, 1982) , and contain overlapping segments of viral DNA. Purified, cosmid-cloned DNA, immobilized on nitrocellulose filters, was hybridized with polyadenylated RNA obtained at late times from infected cell cultures. Bound mRNA was eluted from the solid phase, and translated in vitro in a rabbit reticulocyte lysate system. Figure  2 gives an example showing the polypeptide patterns obtained with a set of cosmid clones. The in vitro translation products from hybrid-selected RNA gave a polypeptide pattern specific for each cosmid clone. Many identical bands were seen after selection with the cosmids pCM101'7 and 1015, which contain largely overlapping segments of HCMV DNA. In four independent experiments, the most prominent polypeptides were synthesized after selection with the clones pCM1007 and pCM1017/pCM1015, respectively (Fig. 2) . Table 1 summarizes the results of these experiments, listing the sizes of polypeptides that could reproducibly be hybrid selected and translated with individual cosmid clones. None of these polypeptides comigrated with the products of in vitrotranslated RNA from mock-infected cells. Some of the polypeptides listed in Table 1 appeared to comigrate with virion proteins or polypeptides from extracts of infected cells; other in vitro translation products found after hybrid selection and translation could correspond to virus-specific proteins that have not been identified as yet. It remains possible that some peptides may be due to premature termination or aberrant initiation during in vitro translation.

The apparent comigration of some in vitro-translated polypeptides with native viral proteins allowed a tentative correlation with the DNA sequence encoding these products. Hybrid selection with overlapping cosmids and subcloned Hind111 fragments allowed finer resolution of the putative genes for these proteins in individual restriction fragments. The results of these experiments are listed in Table 2.

Correlation of in vitro-translation prod; ucts with native virion proteins is possible FIG. 3 . Immunoprecipitation of cell-free translation products of RNA selected by hybridization to cloned viral DNA fragments. Late poly(A)+ RNA from AD169-infected cells was selected by hybridization to viral DNA fragments (indicated by numbers above the lanes). The eluted RNA was translated in the reticulocyte cell-free system and the translation products were immunoprecipitated with monoclonal antibodies. Immunoprecipitation with the monoclonal antibody 355, directed against the phosphorylated 'Il-kDa HCMV protein, is shown in the left half; precipitation with the anti-adenovirus monoclonal antibody 2A6 is shown in the right half. The precipitated translation products were analyzed on a 15% SDS-PAGE, followed by fluorography. Lane a, $SJmethionine-labeled proteins of purified AD169 virions. Translated products of nonselected late poly (A)+ RNA from AD169 infected cells were precipitated with a human anti-HCMV immune serum (lane b), with the monoclonal antibody 355 (lane c), or with monoclonal antibody 2A6 (lane d).

by precipitation with defined monoclonal antibodies. In a separate study a monoclonal antibody designated as PAb355 was obtained that was directed against a virion internal phosphorylated envelope protein of '71 kDa (Nowak et &, 1934) . Total RNA and poly (A)+ mRNA were prepared from lytically infected cells during the late phase of viral replication and employed for in vitro translation. Monoclonal antibody 355 precipitated a single 'Il-kDa polypeptide from the mixture of in v&o-synthesized products (Fig. 3, lane c) . Combining this method with hybrid selection, the sequence coding for the phosphorprotein of 71 kDa (pp 71) could be located within the cosmid clone pCM1007 (Fig. 3) . To determine more precisely the position of this structural gene, subcloned Hind111 fragments of cosmid pCM1007 were employed for subsequent hybrid-selected in vitro translation combined with immunoprecipitation with monoclonal antibody PAb355. As shown in Fig. 4 , the mRNA coding for pp71 hybridized with Hind111 fragments b, c, and L ( Fig. 1) . Similarly, the mRNA encoding the in vitro-translation product of the 65-kDa protein was selected by the same subcloned abcdefgh FIG. 4. Immunoprecipitation of cell-free translation products of RNA selected by hybridization to cloned viral Hind111 DNA fragments. Late poly(A)+ RNA from AD169-infected cells was selected by hybridization to the Hind111 fragments S, P, a, b + c, b, c, and L + D (lanes b through h, respectively). The eluted RNA was translated and the products were immunoprecipitated with the monoclonal antibody 355. The immunoprecipitated translation products were analyzed on a 15% SDS-PAGE, followed by fluorography. Lane a shows [?l]methionine-labeled proteins of purified AD169 virions.

Hind111 fragments (L, b, c) of cosmid clone pCM 1007. The 65kDa in vitro-translated polypeptide comigrated with the major structural protein of HCMV strain AD169 (Fig. 2, Table 2 ). In vitro translation of RNA selected by Hind111 fragments b and c resulted in about equivalent amounts of 65 and 71-kDa polypeptides; in contrast, Hind111 L fragments selected the RNA molecules encoding the 71-kDa,polypeptide more efficiently than the RNA for the 65-kDa protein (data not shown).

and 71 kDa The possibility that the coding sequences of the 65-and 71-kDa polypeptides over-lapped led us to analyze the direction of transcription of these two mRNAs. Hind111 fragment c was cloned in bacteriophage M13mp9 in both orientations. The orientation of the inserted Hind111 c fragment was determined by taking advantage of an EcoRI recognition site which cleaves the insert of about 800 bp asymmetrically (Fig. 5) . The message coding for pp71 was hybridized with single-stranded, cloned HCMV DNA in either of the two orientations (termed plus and minus, arbitrarily). The plus-strand DNA of extracellular bacteriophage contained the HCMV DNA which was transcribed as tested by hybrid selection and in vitro translation (Fig. 6A ). This demonstrates that the mRNA coding for pp71 is transcribed from left to right in the prototype arrangement of the HCMV AD169 genome (Fig. 5) . The message for the 65-kDa polypeptide was hybrid selected by the same Ml3 clone in the plus orientation as shown by in vitro translation of this mRNA and an analysis of the products directly (Fig. 6B) .

Next, experiments were carried out to determine the sizes of the mRNAs encoding 65-and 71-kDa virion polypeptides. RNA was isolated from lytically infected cells during late times, fractionated to obtain poly(A)+ RNA, and analyzed by Northern blot hybridizations. 32P-Labeled Hind111 fragments that were subcloned from cosmid pCM1007 were used as radioactive probes. A single type of an abundant poly(A)+ RNA of about 4 kb was detected when Hind111 fragments L and D, b, or c were used as probes (Fig. 7) . Hind111 fragment U, on the other hand, did not hybridize with this late 4-kb mRNA. This implies that the 4-kb mRNA hybridized with the same set of Hind111 DNA fragments which hybrid selected the RNA encoding the two polypeptides of 65 and 71 kDa. Finally, late poly(A)+ RNA was fractionated by sedimentation in sucrose gradients. RNA from each fraction was analyzed in parallel by Northern blot hybridization with the labeled Hind111 c fragment and by in vitro translation. The products of in vitro protein synthesis were immunoprecipitated by monoclonal antibody 355 and by human immune sera. As shown in Fig.  028 030 0.32 I 034 036 map ""115 Yir-

endonuclease maps of the DNA region of HCMV (strain Ad169) that codes for the viral structural polypeptides of 65 and 71 kDa. The Hind111 fragments U, b, c, and L of the cosmid clone pCMlSO7 are drawn on expanded scale (Fleckenstein et al, 1982) . Positions of cleavage sites for BamHI are as given by Greenaway et al. (1982) . The arrow indicates the map location of the coding sequences for the polypeptides of 65 and 71 kDa, and the direction of transcription for both polypeptides. (A+) and nonadenylated (A-) fractions, denatured, electrophoresed on agarose gels, and transferred to nitrocellulose filters. Cosmid clone pCM1075 (Hind111 L and D) and cloned Hind111 fragments b, c, or U were 9 labeled in vdtro and hybridized to 2.5 gg of blotted poly(A)+ or poly(A)-RNA, to 7.5 pg of Eschwichia coli RNA, or to 10 pg of RNA from uninfected HFF (Mock). 8B, the highest concentration of mRNA instructing pp71 was found in fraction no. 12; the same fraction contained the ma-jority of mRNA for 65kDa polypeptide (Fig. 8A) . Northern blot hybridization demonstrated that most of 4-kb mRNA was in fraction 12 (data not shown). This implies that the mRNA molecules coding the structural proteins of 65 and 71 kDa have about the same sizes and map coordinates, and are transcribed from the same DNA strand. As shown in Fig. 8A , a small level of in vitro translation of ?l-kDa polypeptide was also seen with smaller RNAs (>1.8 kb), suggesting that processed or degraded poly(A)+ mRNA molecules can be active in the in vitro translation assay.

This is the first report mapping the structural genes of virion proteins from HCMV. The method employed to map DNA sequences coding for virion polypeptides was based on cell-free translation of viral RNA selected by hybridization to cosmidcloned DNA fragments. This approach localized the map coordinates for nine structural genes encoding nine polypeptides which comigrate in SDS-polyacrylamide gels with authentic virion proteins. For a more unambiguous identification, it was useful to correlate in vitro-synthesized polypeptides with native viral proteins by immunoprecipitation using monoclonal antibodies. Employing this approach the map coordinates and direction of transcription were determined for a 4-kb mRNA encoding the phosphorylated 71-kDa internal envelope protein of HCMV. Surprisingly, viral polyadenylated RNA of the same size, genomic location, and direction of transcription was also translated in vitro into a 65-kDa polypeptide. This 65- FIG. 8. Immunoprecipitation of cell-free translation products of RNA size-fractionated in sucrose gradients. Fifty micrograms of late poly (A)+ RNA from AD169-infected cells was size fractionated on a 5 to 20% sucrose-formamide gradient. The RNA in each fraction (1 through 2'7) was ethanol precipitated and half of the RNA was translated in the reticulocyte cell-free system. The suspension with the translated products was divided into two aliquots. One aliquot of each fraction was precipitated with a human anti-HCMV immune serum (A), the other aliquot was precipitated with the monoclonal antibody 355 (B). The immunoprecipitated products were analyzed on a 15% SDS-PAGE, followed by fluorography. kDa protein appeared to be equivalent to the major structural protein of virus particles and dense bodies. The coding sequences for the 65kDa polypeptide might be localized more upstream (5') within the 4-kb mRNA than the sequence instructing the 71-kDa protein. This suggestion is based upon the efficiency of hybrid selection and in vitro translation in subcloned Hind111 fragments and from translation of size-fractionated mRNA (Figs. 6, 8) .

There are several possible interpretations of the experimental resuIt8 presented here. First, the two polypeptides, 71 and 65 kDa, could be encoded in two different 4-kb m-RNA8 that derive from the same region of the CMV genome. These mRNAs could arise from two distinct transcripts or from differential splicing patterns of a primary transcript, giving rise to two related but distinct 4-kb mRNAs. Alternatively, a single 4-kb mRNA species could give rise to two different proteins, 71 and 65 kDa, by using two different reading frames in the same mRNA. This is apparently the case with the Adenovirus Elb-22 S mRNA and Elb-19 kDa and Elb-58 kDa proteins read in two different reading frames (Bos et c& 1981) . Finally, it is formally possible that the 71-kDa protein is a precursor of the 65-kDa protein. The reason this appears to be an unlikely hypothesis is that two different monoclonal antibodies directed against the 71-kDa protein do not react with the 65-kDa protein (Nowak et uL, 1984) . Furthermore, the cleavage of the 'Il-kDa protein into a 65-kDa protein would have to occur very efficiently in the rabbit reticulocyte in vitro translation extract, and that possibility appears to be unlikely as well. Peptide maps of the 71-and 65-kDa proteins should help to distinguish between some of these alternatives. What is clear from these data, however, is that identical or overlapping coding regions of the CMV genome are employed to produce two structural proteins of this virus. ACKNOWLEDGMENTS This work was supported by Bundesministerium fur Forschung und Technologie, PTB 8353; and by grants from the New York State Science and Technology Foundation, the Sklarow Foundation, and Ac-ademic Research Associates. We thank W. Bodemer for helpful discussions. The excellent technical assistance of Agnes Gmeiner and Gail Urban is greatly appreciated.

Purification of biologically active globin messenger RNA by chromatography on oligothymidylic acid-cellulose

Detection of virus-speci& antigen in EB-(PSHR-1) Virus superinfected Raji cells by immunoprecipitation

A film detection method for tritium labeled proteins and nucleic acids in polyacrylamide gels

Construction and characterization at ampliable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid

Transfection by exogenous and endogenous murine retrovirus DNAs

Human cytomegalovirus DNA: Restriction enzyme cleavage maps and map locations for immediate-early, early, and late RNAs

Genome structure and virion polypeptides of the primate herpesvirus He-rpemiw a&us Types 1 and 3: Comparison with human cytomegalovirus

Cloning of the complete human ctyomegalovirus genome in cosmids

Human cytomegalovirus DNA, I. Molecular weight and infectivity

A new technique for the assay of infectivity of human adenovirus 5 DNA

Predominant immediate early transcripts of human cytomegalovirus strain AD169

Rapid isolation of antigens from cells with a staphylococcal protein A antibody adsorbent: Parameters of the interaction of antibody: antigen complexes with protein A

Cloning of Herpesvirus saimira DNA fragments representing the entire L-region of the genome

Cleavage of structural proteins during the assembly of the head of bacteriophage T4

Structural organization of the DNA molecules from human cytomegalovirus

Biological detection of specific mRNA molecules by microinjection

Transcription in human fibroblasts permissively infected by human cytomegalovirus strain AD169

Analysis of single and double-stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and acridine orange

A system for shotgun DNA sequencing

An integrative strategy of DNA sequencing and experiments beyond

Anatomy of HSV DNA. X. Mapping of viral genes by analysis of polypeptides and functions specified by HSV-1 X HSV-2 recombinants

Characterization of monoclonal antibodies and polyclonal immune sera directed against human CMV virion proteins

Use of recombinant plasmids to investigate the structure of the human cytomegalovirus genome

19'76). An efficient mRNA-dependent translation system from reticulocyte lysates

Purification and mapping of specific mRNAs by hybridization-selection and cell-free translation

Identification of the adenovirus early proteins and their genomic map position. virdogy 100

Adenovirus Elb-58kd tumor antigen and SV40 large tumor antigen are physically associated with the same 54 kd cellular protein in transformed cells

A monoclonal antibody detecting the adenovirus type 5 Elb-58 kDa tumor antigen: Characterization of the Elb-58 kDa tumor antigen in adenovirus infected and transformed cells

Coronavirus JHMCell-free synthesis of structural protein p60

Detection of specific sequences among DNA fragments separated by gel electrophoresis

Organization and expression of the immediate early genes of human cytomegalovirus

Construction of a cloned library of the EcoRI fragments from the cutomegalovirus strain AD169

Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose

Cloning of the human cytomegalovirus genome as endonuclease XbaI fragments

Temporal patterns of human cytomegalovirus transcription: Mapping the viral RNAs synthesized at immediate early, early, and late times after infection