A maturation protein in adenovirus morphogenesis

Adenovirus assembly is impaired by BMI1-related histone deacetylase activity

Polycomb ring finger oncogene BMI1 (B cell-specific Moloney murine leukemia virus integration site 1) plays a critical role in development of several types of cancers. Here, we report an inverse relationship between levels of BMI1 expression and adenovirus (Ad) progeny production. Enforced BMI1 expression in A549 cells impaired Ad progeny production. In contrast, knocking-down of endogenous BMI1 expression enhanced progeny production of a conditionally replicating Ad and wild-type Ad5 and Ad11p. Ad vectors overexpressing BMI1 were not impaired in the replication of progeny genomes and in the expression of E1A and Ad structural proteins. However, 293 cells infected by Ad vector overexpressing BMI1 contained a large proportion of morphologically irregular Ad particles. This effect was reversed in 293 cells pre-treated with the histone deacetylase (HDAC) inhibitor trichostatin A (TSA) in parallel with the production of infectious Ad particles. Our findings suggest an inhibitory role of BMI1 in Ad morphogenesis that can be implied in Ad tropism and Ad-mediated cancer therapy.

The structure and function of the adenovirus major late promoter

The adenovirus major late promoter (MLP) has played a pre-eminent role in the analysis of transcription initiation in mammalian cells, and is an outstanding example of the ways in which the study of adenovirus has led to fundamental insights into general cellular processes. The aim of this chapter is to give a comprehensive review of the structure and function of this model mammalian promoter. After a brief description of late transcription in the adenovirus replication cycle, the experimental evidence for the current consensus on the genetic structure of the MLP, including a consideration of non-primate adenovirus MLPs, will be reviewed. Next, the functions of the MLP in the viral life cycle will be examined, and some of the problems that remain to be resolved will be addressed. The review ends with some ideas on how the knowledge of the structure and function of the MLP can be used in designing virus vectors for specific experimental purposes.

Truncation of the human adenovirus type 5 L4 33-kDa protein: evidence for an essential role of the carboxy-terminus in the viral infectious cycle

The subgroup C human adenovirus L4 33-kDa protein is a nuclear phosphoprotein that plays a direct, but dispensable, role in virion assembly. The r-strand open reading frame (ORF) for this protein lies opposite to the 5' end of the l-strand E2 early (E2E) transcription units. To facilitate studies of regulation of E2E transcription, we wished to construct a mutant virus in which the 33-kDa ORF was truncated to serve as a background into which specific E2E mutations could be introduced without also altering the 33-kDa protein. We constructed viral DNA (vDNA) containing within the 33-kDa ORF two tandem, premature stop codons that should prevent translation of the C-terminal 47 amino acids of the protein (Delta47). We report here the unanticipated lethality of such truncation of the L4 33-kDa protein. Viral DNA harboring the Delta47 mutations did not produce infectious virus when transfected into cultured cells. In contrast, infectious virus was recovered upon transfection of revertant vDNA, indicating that the Delta47 mutations were responsible for the observed phenotype. The Delta47 mutations did not affect E2E transcription or production of the E2 DNA-binding protein. Transfected Delta47 vDNA was replicated and directed the production of early and late viral proteins, including hexon protein in the trimer conformation. However, no virus particles of any kind were produced. We propose that truncation of the adenovirus 33-kDa protein results in a lethal, late block in the infectious cycle during the assembly of progeny virions and discuss the implications of this phenotype for the mechanism of virion assembly.

Identification of a cellular repressor of transcription of the adenoviral late IVa(2) gene that is unaltered in activity in infected cells

The gene encoding the adenovirus type 2 IVa(2) protein, a sequence-specific activator of transcription from the viral major late promoter, is itself transcribed only during the late phase of infection. We previously identified a cellular protein (IVa(2)-RF) that binds specifically to an intragenic sequence of the IVa(2) transcription unit. We now report that precise substitutions within the IVa(2)-RF-binding site that decreased binding affinity increased the efficiency of IVa(2) transcription in in vitro reactions containing IVa(2)-RF. Consistent with the conclusion that this cellular protein represses IVa(2) transcription, mutations that led to more efficient transcription in the presence of IVa(2)-RF were without effect in reactions lacking this cellular protein. No change in the concentration or activity of IVa(2)-RF could be detected in adenovirus-infected cells during the period in which the IVa(2) gene is transcribed. We therefore propose that restriction of IVa(2) transcription to the late phase is the result of titration of this cellular repressor as the number of copies of the IVa(2) promoter increases upon replication of the viral genome.

Encapsidation of viral DNA requires the adenovirus L1 52/55-kilodalton protein

Previous work demonstrated that the adenovirus L1 52/55-kDa protein is required for assembly of viral particles, although its exact role in the assembly process is unclear. The 52/55-kDa protein's early expression, however, suggests that it might have other roles at earlier times during infection. To uncover any role the 52/55-kDa protein might have at early times and to better characterize its role in assembly, a mutant adenovirus incapable of expressing the 52/55-kDa protein was constructed (H5pm8001). Analysis of the onset and extent of DNA replication and late protein synthesis revealed that H5pm8001-infected 293 cells entered the late stage of infection at the same time as did adenovirus type 5 (Ad5)-infected cells. Interestingly, H5pm8001-infected cells displayed slightly lower levels of replicated viral DNA and late proteins, suggesting that although not required, the 52/55-kDa protein does augment these activities during infection. Analysis of transcripts produced from the major late and IVa2 promoters indicated a slight reduction in H5pm8001-infected compared to Ad5-infected cells at 18 h postinfection that was not apparent at later times. Analysis of particles formed in H5pm8001 cells revealed that empty capsids could form, suggesting that the 52/55-kDa protein does not function as a scaffolding protein. Subsequent characterization of these particles demonstrated that they lacked any associated viral DNA. These findings indicate that the 52/55 kDa-protein is required to mediate stable association between the viral DNA and empty capsid and suggest that it functions in the DNA encapsidation process.

DNA packaging mutant: repression of the vaccinia virus A32 gene results in noninfectious, DNA-deficient, spherical, enveloped particles

The vaccinia virus A32 open reading frame was predicted to encode a protein with a nucleoside triphosphate-binding motif and a mass of 34 kDa. To investigate the role of this protein, we constructed a mutant in which the original A32 gene was replaced by an inducible copy. The recombinant virus, vA32i, has a conditional lethal phenotype: infectious virus formation was dependent on isopropyl-beta-D-thiogalactopyranoside (IPTG). Under nonpermissive conditions, the mutant synthesized early- and late-stage viral proteins, as well as viral DNA that was processed into unit-length genomes. Electron microscopy of cells infected in the absence of IPTG revealed normal-appearing crescents and immature virus particles but very few with nucleoids. Instead of brick-shaped mature particles with defined core structures, there were numerous electron-dense, spherical particles. Some of these spherical particles were wrapped with cisternal membranes, analogous to intracellular and extracellular enveloped virions. Mutant viral particles, purified by sucrose density gradient centrifugation, had low infectivity and transcriptional activity, and the majority were spherical and lacked DNA. Nevertheless, the particle preparation contained representative membrane proteins, cleaved and uncleaved core proteins, the viral RNA polymerase, the early transcription factor and several enzymes, suggesting that incorporation of these components is not strictly coupled to DNA packaging.

The product of the adenovirus intermediate gene IVa2 is a transcriptional activator of the major late promoter

During the course of lytic infection, the adenovirus major late promoter (MLP) is induced to high levels after replication of viral DNA has started. We had previously shown that sequence elements located downstream of the MLP start site were implicated in this late-specific transcriptional activation (DE1, between +85 and +98; DE2, between +100 and +120). Two positive transcription factors involved in this activation have been detected. DEF-A, which specifically binds to DE1 and also to the 3' portion of DE2 (DE2a), and DEF-B, which interacts with the 5' part of DE2 (DE2b). When present together, these two proteins cooperatively assemble onto the DE2 element. We now report the purification of DEF-B and show that it is identical to the product of the adenovirus IVa2 gene product. This conclusion is based on microsequence analysis of DEF-B as well as on the inhibitory effect of antibodies against IVa2 on the DNA-binding activity of DEF-B and also on DE-dependent in vitro transcription. In addition, we show that bacterially synthesized IVa2 protein binds to the DE sequences with the same specificity as DEF-B. Finally, in transfected cells, a recombinant IVa2 protein stimulates MLP activity in a DE-dependent fashion. The physiological implications of these findings are discussed.

Gene A32 product of vaccinia virus may be an ATPase involved in viral DNA packaging as indicated by sequence comparisons with other putative viral ATPases

Statistically significant sequence similarity was revealed between the gene A32 product of vaccinia virus (VV), gene I products (gpI) of filamentous single-stranded DNA bacteriophages, and IVa2 gene products of adenoviruses. Four conserved sequence motifs were delineated, the two N-proximal of which correspond to the A and B motifs of the purine NTP-binding pattern. Based on the role of gpI and IVa2 proteins in virion morphogenesis, and on the conservation of the NTP-binding pattern in these proteins, we hypothesize that the A32 gene product might be involved in an ATP-consuming function in VV virion formation, e.g., packaging of the DNA in the virus particle.

Conserved sequences of the adenovirus genome for detection of all human adenovirus types by hybridization

The application of DNA hybridization directly to clinical specimens has the potential of improving the diagnosis of fastidious types of adenovirus. In this study, the genome of one adenovirus type from each human subgenus (A to F) was systematically evaluated by hybridization for homologous sequences to find the optimal common probe for detection of all human adenovirus types. The area of cross-hybridization, most closely defined with adenovirus type 2 (Ad2), mapped from map units 11.4 to 16.1 and 26.9 to 29.7 and, principally, to a central area of the genome between map units 47.5 and 65.2. The last area, enclosing the hexon gene, was highly conserved. Cloned probes generated from this area demonstrated the greatest homology to heterologous types by hybridization analysis. A HindIII-BglII clone containing the hexon gene of Ad2 within narrow confines reacted most evenly with all adenoviral types and detected the DNA of all other subgenera with a sensitivity 2 logs greater than that of a complete genomic Ad2 probe. The most homologous adenoviral gene sequences were observed in genes involved with DNA replication or intimately connected to the hexon in the early capsid formation. These results show that the hexon gene constitutes the best single region of the adenovirus genome for use as a genus-specific probe for the diagnosis of all human adenoviral subgenera by DNA hybridization.

Viral proteins containing the purine NTP-binding sequence pattern

A compilation is presented of viral proteins containing the NTP-binding sequence pattern, and criteria are suggested for assessment of the functional significance of the occurrence of this pattern in protein sequences. It is shown that the distribution of NTP-binding pattern-containing proteins through the viral kingdom is strongly non-random. Sequence comparisons led to delineation of several families of these proteins, some of which could be brought together into superfamilies including also cellular proteins. The available biochemical evidence is compatible with the proposal that viral proteins in which the NTP-binding pattern is evolutionarily conserved might all be NTPases involved in: i) duplex unwinding during DNA and RNA replication, transcription, recombination and repair, and possibly mRNA translation; ii) DNA packaging, and iii) dNTP generation.

Restriction analysis and homology studies of the bovine adenovirus 7 genome

We located the cleavage sites for restriction endonucleases EcoRI, HindIII, and BamHI on the genome of bovine adenovirus 7. Cross-hybridization at reduced stringency revealed two regions of homology shared by the DNA of human adenovirus 2 and bovine adenoviruses 7 and 3. These regions correspond to the hexon and the IVa2 protein genes of the human adenovirus. Another region of homology shared only by the human adenovirus and bovine adenovirus 7 corresponded to the penton or the polypeptide IIIa genes. These results allowed us to align the restriction map of bovine adenovirus 7 with respect to the other adenoviruses.

Morphogenesis of human adenovirus type 2: sequence of entry of proteins into previral and viral particles

The initial steps of adenovirus capsid morphogenesis and the sequence of entry of structural and nonstructural proteins into assembly-intermediate (IM) particles were investigated by pulse-chase labeling, temperature shifts, and cycloheximide inhibition of particle formation. The experiments were performed on wild-type and two assembly-defective, temperature-sensitive mutants, H2 ts 112 and H2 ts 107. The sequence of events in the adenovirus assembly can be schematized as follows. (i) Hexons, pentons, and protein IX assembled with scaffolding proteins 100K, PVIII, and PVII, precursor to the major core protein, to form a previral particle banding at a density of 1.285 in CsCl; (ii) additional incorporation of maturation and/or stabilization proteins IIIa, 50K, 39K, 28K, and PVI led to 1.295 IM; (iii) exit of 100K, 39K, and 28K, and entry of viral DNA gave rise to 1.370 IM; (iv) dephosphorylation and/or exit of 50K and exchange with core protein V and processing of precursors to VII, VI, VIII, and DNA-terminal protein resulted in formation of infectious 1.345 virion. The polypeptide composition of the new class of assembly-intermediate particles elicited by H2 ts 107 (1.285 IM), suggested that 100K, PVIII, and also PVII might serve as scaffold components for adenovirus capsid building.

Specific complex of the late nonstructural 100,000-dalton protein with newly synthesized hexon in adenovirus type 2-infected cells

Analysis of cellular extracts of HeLa cells infected with adenovirus type 2 (Ad2) by immunoprecipitation with antiserum against the late nonstructural 100,000-dalton (100K) protein revealed the presence of a specific complex between the 100K protein and newly synthesized hexon molecules. Serological analysis of the hexon molecule in the 100K/hexon complex with antibodies specific for hexon monomers or trimers showed that only monomeric hexon molecules were associated with the 100K protein. By immunofluorescence microscopy this monomeric hexon was primarily found in the cytoplasm, whereas the trimeric form was mainly confined to the nucleus of infected cells. We conclude that in the cytoplasm of Ad2-infected cells newly synthesized, monomeric hexon molecules can interact with the 100K protein. This suggests that the 100K protein may play some role either in trimerization of newly synthesized, monomeric hexon molecules and/or in its transport from the cytoplasm into the nucleus.

The nucleotide sequence of the gene encoding protein IVa2 in human adenovirus type 7

The nucleotide sequences of cloned DNA segments encoding the IVa2 gene from Ad7 and a portion of Ad12 (group B and group A human adenoviruses, respectively) have been determined. When compared to Ad5, a group C adenovirus, these sequences have been found to be 80% homologous. Most changes are transitions or transversions. This high degree of nucleotide homology results in a high degree of amino acid conservation in the predicted polypeptides encoded from these genes; most nucleotide changes occur at the third position in the codon. The predicted polypeptide contains 448 amino acids and has a calculated Mr-value of 50700. The positions of the 5' end of the mRNA and of the donor and acceptor splice sites of Ad7 and Ad12 can be inferred by analogy to those of Ad5. A long open reading frame starting upstream from the IVa2 gene overlaps the N-terminal portion of the polypeptide but is encoded in a different reading frame. Within this overlapping region, the long open reading frame is more conserved in amino acid sequence than is the presumed IVa2 polypeptide, suggesting that evolutionary pressure was exerted on the longer protein, a product of viral early region 2B. The high degree of conservation of this E2B region within the overlapping segment suggests that its activities must be more important for adenovirus infection than are the functions encoded in the amino-terminus of the IVa2 gene.

Characterization of a temperature-sensitive fiber mutant of type 5 adenovirus and effect of the mutation on virion assembly

A temperature-sensitive, fiber-minus mutant of type 5 adenovirus, H5ts142, was biochemically and genetically characterized. Genetic studies revealed that H5ts142 was a member of one of the three apparent fiber complementation groups which were detected owing to intracistronic complementation. Recombination analyses showed that it occupied a unique locus at the right end of the adenovirus genetic map. At the nonpermissive temperature, the mutant made stable polypeptides, but they were not glycosylated like wild-type fiber polypeptides. Sedimentation studies of extracts of H5ts142-infected cells cultured and labeled at 39.5 degrees C indicated that a limited number of the fiber polypeptides made at the nonpermissive temperature could assemble into a form having a sedimentation value of 6S (i.e., similar to the trimeric wild-type fiber), but that this 6S structure was not immunologically reactive. When H5ts142-infected cells were shifted to the permissive temperature, 32 degrees C, fiber polypeptides synthesized at 39.5 degrees C were as capable of being assembled into virions as fibers synthesized in wild type-infected cells; de novo protein synthesis was not required to allow this virion assembly. In H5ts142-infected cells incubated at 39.5 degrees C, viral proteins accumulated and aggregated into particles having physical characteristics of empty capsids. These particles did not contain DNA or its associated core proteins. However, when the infected culture was shifted to 32 degrees C, DNA appeared to enter the empty particles and complete virions developed. The intermediate particles obtained had the morphology of adenoviruses, but they contained less than unit-length viral genomes as measured by their buoyant density in a CsCl density gradient and the size of their DNA as determined in both neutral and alkaline sucrose gradients. The reduced size of the intermediate particle DNA was demonstrated to be the result of incompletely packaged DNA molecules being fragmented during the preparative procedures. Hybridization of labeled DNA extracted from the intermediate particles to filters containing restriction fragments of the adenovirus genome indicated that the molecular left end of the viral genome preferentially entered these particles.

The nucleotide sequence of the gene for protein IVa2 and of the 5' leader segment of the major late mRNAs of adenovirus type 5

We present here the primary structure of the region of human adenovirus 5 (Ad5) DNA from nucleotide 4001 through the HindIII site at nucleotide 6246 (map positions 0.11 to 0.17). The corresponding region in the closely related adenovirus type 2 (Ad2) encodes the spliced mRNA for viral protein IVa2 (Chow et al., 1979; Persson et al., 1979). Reverse transcription of the Ad5 pIVa2 mRNA localized the 5' terminus of the mRNA to approximately position 5840, and its splice coordinates to positions 5706 and 5427. From the data of Aleström et al. (1980) for Ad2, the 3' end of this mRNA was inferred to be specified by Ad5 nucleotide 4060. The nucleic acid data allow us to predict an Mr of 50873 for the IVa2 protein of Ad5, which is close to the experimentally determined value of 50000 (Persson et al., 1979). The DNA sequence described here also includes the information for the 5'-terminal leader segment of the major late mRNAs of Ad5.

Late nonstructural 100,000- and 33,000-dalton proteins of adenovirus type 2. II. Immunological and protein chemical analysis

For an immunological analysis of the late adenovirus type 2 nonstructural 100,000-dalton (100K) and 33K proteins, we prepared antisera against sodium dodecyl sulfate-denatured, gel-purified 100K and 33K proteins. These antisera were tested for potential cross-reactivity, since according to a previous report (Axelrod, Virology 87:366--383, 1978) these two proteins exhibit extensive amino acid homologies. However, immunoprecipitations of 100K and 33K proteins, as well as a sensitive immune replica technique, did not reveal any immunological relationship between these proteins. Therefore, using fingerprint peptide analysis, we investigated the structural relationship between 100K and 33K proteins labeled with a 14C-amino acid mixture or with [14C]proline after digestion with trypsin. We detected only minor, if any, amino acid homologies, indicating that the 100K and 33K proteins are not structurally related.

The gene for polypeptide IX of human adenovirus type 7

This paper describes the nucleotide sequence of subgroup B human adenovirus type 7 (Ad7) between positions 3351-4010, and of cloned cDNA derived from mRNAs encoded by this segment. One of these (mRNA VII) is shown to be unspliced, and has its 5'- and 3'-ends encoded by positions 3460 (determined by the nuclease S1 technique) and 3939, respectively. The mRNA sequence contains a single open reading frame for protein biosynthesis between the first available initiation triplet AUG at position 3481 to the stop codon UAA at position 3895. It can specify a polypeptide of 138 amino acids (14 098 daltons) which must be polypeptide IX of Ad7, as is evident from its mapping position, and from a comparison with the sequence of the protein IX gene of subgroup C adenovirus type 5.

In vitro synthesis of adenovirus type 5 T antigens. I. Translation of early region 1-specific rna from lytically infected cells

Translation of early RNA specific for the leftmost early region 1 of adenovirus type 5 DNA in a rabbit reticulocyte cell-free system resulted in the synthesis of proteins with the following molecular weights: 65,000 (65K), 54K, 42 to 34K, 29K, 25K, 19K, and 18K. All of these proteins could be immunoprecipitated with hamster antitumor serum. The 42 to 34K proteins mapped in early region 1a, and those with molecular weights of 65K, 19K, and 18K mapped in early region 1b. The gene coding for the 54K protein may be localized outside of early region 1. We could not map unambiguously the 29K and 25K proteins. The identification of a 65K protein among products synthesized in vitro suggests that this protein may be identical to the 65K major T antigen present in adenovirus type 5-infected and -transformed cells, and this indicates that it is indeed encoded by the viral genome. This protein is encoded by a 23S mRNA. The other early region 1-specific proteins appear to be encoded by mRNA of approximately 13S, except for the 19K protein, which is synthesized with RNA sedimenting at both 13S and 23S.

Comparison of the nuclear ribonucleoproteins containing the transcripts of adenovirus-2 and HeLa cell dna

A method as devised allowing the preparation of hnRNA-containing ribonucleproteins (hnRNP) frm HeLa cells infected with adenovirus type 2 under conditions where the extraction of viral replication complexes as minimal. Approximately 60% of the RNA from such hnRNP hybridized with adenovirus DNA. The hnRNP from infected cells had the same general characteristics as those from uninfected cells. Their size was heterogeneous (30-260 S) and depended upon that of their RNA. Their CsCl densities were identical (1.39-1.40 g/ml), indicating the same protein:RNA ratio. Their proteins were found in the same molecular weight range, between 25 000 and 200 000. The major proteins of hnRNP from HeLa cells were present in hnRNP from adenovirus-infected cells. As 60% of the cellular RNA was replaced by adenovirus RNA in hnRNP, this in dicated that there was not stringent specificity in the RNA-protein interactions. The relative proportions of the proteins were identifical in both cases, suggesting that the hnRNP assembly was independent of nucleotide sequences at least for the major proteins. The hnRNP from infected and uninfected cells differed by the size of their RNA, which was larger after infection, and by the presence of six additional minor polypeptides after infection. However, it cannot be excluded that the presence of these polypeptides in hnRNP resulted from non-specific adsorption.