Characterization of the major mRNAs from adenovirus 2 early region 4 by cDNA cloning and sequencing

Accurate detection of tumor-specific gene fusions reveals strongly immunogenic personal neo-antigens

Cancer-associated gene fusions are a potential source for highly immunogenic neoantigens, but the lack of computational tools for accurate, sensitive identification of personal gene fusions has limited their targeting in personalized cancer immunotherapy. Here we present EasyFuse, a machine learning computational pipeline for detecting cancer-specific gene fusions in transcriptome data obtained from human cancer samples. EasyFuse predicts personal gene fusions with high precision and sensitivity, outperforming previously described tools. By testing immunogenicity with autologous blood lymphocytes from patients with cancer, we detected pre-established CD4⁺ and CD8⁺ T cell responses for 10 of 21 (48%) and for 1 of 30 (3%) identified gene fusions, respectively. The high frequency of T cell responses detected in patients with cancer supports the relevance of individual gene fusions as neoantigens that might be targeted in personalized immunotherapies, especially for tumors with low mutation burden.

Selective multi-kinase inhibition sensitizes mesenchymal pancreatic cancer to immune checkpoint blockade by remodeling the tumor microenvironment

KRAS-mutant pancreatic ductal adenocarcinoma (PDAC) is highly immunosuppressive and resistant to targeted and immunotherapies. Among the different PDAC subtypes, basal-like mesenchymal PDAC, which is driven by allelic imbalance, increased gene dosage and subsequent high expression levels of oncogenic KRAS, shows the most aggressive phenotype and strongest therapy resistance. In the present study, we performed a systematic high-throughput combination drug screen and identified a synergistic interaction between the MEK inhibitor trametinib and the multi-kinase inhibitor nintedanib, which targets KRAS-directed oncogenic signaling in mesenchymal PDAC. This combination treatment induces cell-cycle arrest and cell death, and initiates a context-dependent remodeling of the immunosuppressive cancer cell secretome. Using a combination of single-cell RNA-sequencing, CRISPR screens and immunophenotyping, we show that this combination therapy promotes intratumor infiltration of cytotoxic and effector T cells, which sensitizes mesenchymal PDAC to PD-L1 immune checkpoint inhibition. Overall, our results open new avenues to target this aggressive and therapy-refractory mesenchymal PDAC subtype.

Intratumoral heterogeneity in cancer progression and response to immunotherapy

Most (if not all) tumors emerge and progress under a strong evolutionary pressure imposed by trophic, metabolic, immunological, and therapeutic factors. The relative impact of these factors on tumor evolution changes over space and time, ultimately favoring the establishment of a neoplastic microenvironment that exhibits considerable genetic, phenotypic, and behavioral heterogeneity in all its components. Here, we discuss the main sources of intratumoral heterogeneity and its impact on the natural history of the disease, including sensitivity to treatment, as we delineate potential strategies to target such a detrimental feature of aggressive malignancies.

Targeting metastatic cancer

Despite recent therapeutic advances in cancer treatment, metastasis remains the principal cause of cancer death. Recent work has uncovered the unique biology of metastasis-initiating cells that results in tumor growth in distant organs, evasion of immune surveillance and co-option of metastatic microenvironments. Here we review recent progress that is enabling therapeutic advances in treating both micro- and macrometastases. Such insights were gained from cancer sequencing, mechanistic studies and clinical trials, including of immunotherapy. These studies reveal both the origins and nature of metastases and identify new opportunities for developing more effective strategies to target metastatic relapse and improve patient outcomes.

Evolutionary dynamics of neoantigens in growing tumors

Cancers accumulate mutations that lead to neoantigens, novel peptides that elicit an immune response, and consequently undergo evolutionary selection. Here we establish how negative selection shapes the clonality of neoantigens in a growing cancer by constructing a mathematical model of neoantigen evolution. The model predicts that, without immune escape, tumor neoantigens are either clonal or at low frequency; hypermutated tumors can only establish after the evolution of immune escape. Moreover, the site frequency spectrum of somatic variants under negative selection appears more neutral as the strength of negative selection increases, which is consistent with classical neutral theory. These predictions are corroborated by the analysis of neoantigen frequencies and immune escape in exome and RNA sequencing data from 879 colon, stomach and endometrial cancers.

Large database for the analysis and prediction of spliced and non-spliced peptide generation by proteasomes

Proteasomes are the main producers of antigenic peptides presented to CD8+ T cells. They can cut proteins and release their fragments or recombine non-contiguous fragments thereby generating novel sequences, i.e. spliced peptides. Understanding which are the driving forces and the sequence preferences of both reactions can streamline target discovery in immunotherapies against cancer, infection and autoimmunity. Here, we present a large database of spliced and non-spliced peptides generated by proteasomes in vitro, which is available as simple CSV file and as a MySQL database. To generate the database, we performed in vitro digestions of 55 unique synthetic polypeptide substrates with different proteasome isoforms and experimental conditions. We measured the samples using three mass spectrometers, filtered and validated putative peptides, identified 22,333 peptide product sequences (15,028 spliced and 7,305 non-spliced product sequences). Our database and datasets have been deposited to the Mendeley (doi:10.17632/nr7cs764rc.1) and PRIDE (PXD016782) repositories. We anticipate that this unique database can be a valuable source for predictors of proteasome-catalyzed peptide hydrolysis and splicing, with various future translational applications.

Cell transformation by the adenovirus oncogenes E1 and E4

Extensive studies on viral-mediated oncogenic transformation by human adenoviruses have revealed much of our current understanding on the molecular mechanisms that are involved in the process. To date, these studies have shown that cell transformation is a multistep process regulated by the cooperation of several adenoviral gene products encoded in the early regions 1 (E1) and 4 (E4). Early region 1A immortalizes primary rodent cells, whereas co-expression of early region protein 1B induces full manifestation of the transformed phenotype. Beside E1 proteins, also some E4 proteins have partial transforming activities through regulating many cellular pathways. Here, we summarize recent data of how adenoviral oncoproteins may contribute to viral transformation and discuss the challenge of pinpointing the underlying mechanisms.

Epigenetics and the dynamics of chromatin during adenovirus infections

The DNA genome of eukaryotic cells is compacted by histone proteins within the nucleus to form chromatin. Nuclear-replicating viruses such as adenovirus have evolved mechanisms of chromatin manipulation to promote infection and subvert host defenses. Epigenetic factors may also regulate persistent adenovirus infection and reactivation in lymphoid tissues. In this review, we discuss the viral proteins E1A and protein VII that interact with and alter host chromatin, as well as E4orf3, which separates host chromatin from sites of viral replication. We also highlight recent advances in chromatin technologies that offer new insights into virus-directed chromatin manipulation. Beyond the role of chromatin in the viral replication cycle, we discuss the nature of persistent viral genomes in lymphoid tissue and cell lines, and the potential contribution of epigenetic signals in maintaining adenovirus in a quiescent state. By understanding the mechanisms through which adenovirus manipulates host chromatin, we will understand new aspects of this ubiquitous virus and shed light on previously unknown aspects of chromatin biology.

Gene Therapy for Prostate Cancer by Controlling Adenovirus E1a and E4 Gene Expression with PSES Enhancer

PSES is a chimeric enhancer containing enhancer elements from prostate-specific antigen (PSA) and prostate-specific membrane antigen (PSMA) genes that are prevalently expressed in androgen-independent prostate cancers. PSES shows strong activity equivalent to cytomegalovirus (CMV) promoter, specifically in PSA/PSMA-positive prostate cancer cells, the major cell types in prostate cancer in the absence of androgen. We developed a recombinant adenovirus (AdE4PSESE1a) by placing adenoviral E1a and E4 genes under the control of the bidirectional enhancer PSES and enhanced green fluorescent protein gene for the purpose of intratumoral virus tracking under the control of CMV promoter. Because of PSES being very weak in nonprostatic cells, including HEK293 and HER911 that are frequently used to produce recombinant adenovirus, AdE4PSESE1a can only be produced in the HER911E4 cell line which expresses both E1 and E4 genes. AdE4PSESE1a showed similar viral replication and tumor cell killing activities to wild-type adenovirus in PSA/PSMA-positive prostate cancer cells. The viral replication and tumor cell killing activities were dramatically attenuated in PSA/PSMA-negative cells. To test whether AdE4PSESE1a could be used to target prostate tumors in vivo, CWR22rv s.c. tumors were induced in nude mice and treated with AdE4PSESE1a via intratumoral and tail vein injection. Compared to tumors treated with control virus, the growth of CWR22rv tumors was dramatically inhibited by AdE4PSESE1a via tail vein injection or intratumoral injection. These data show that adenoviral replication can be tightly controlled in a novel fashion by controlling adenoviral E1a and E4 genes simultaneously with a single enhancer.

Analysis of early region 4 of porcine adenovirus type 3

The early region 4 (E4) of porcine adenovirus (PAdV)-3, located at the right-hand end of the genome is transcribed in a leftward direction and has the potential to encode seven (p1-p7) open reading frames (ORFs). To determine the role of each protein in viral replication, we constructed full-length PAdV-3 genomic clones containing deletions of individual E4 ORF or combined deletions of the neighboring ORFs. Transfection of swine testicular (ST) cells with individual E4 mutant plasmid DNAs generated PAdV-3 E4 mutant viruses except with plasmids containing a deletion of ORF p3, ORF p2+ p3 or ORF p3+ p4. Each of the mutants was further analyzed for growth kinetics, and early/late protein synthesis. Mutant viruses carrying deletions in ORF p1, ORF p2 or ORF p4 showed growth characteristics similar to that of wild-type PAdV-3. Early/late protein synthesis was also indistinguishable from that of wild-type PAdV-3. However, mutant viruses carrying deletions in ORF p5, ORF p6 or ORF p7 showed a modest effect in their ability to grow in porcine cells and express early proteins. These results suggest that the E4 ORF p3 (showing low homology with non-essential human adenovirus (HAdV)-9-E4 ORF1 encoded proteins) is essential for the replication of PAdV-3 in vitro. In contrast, the E4 ORF p7 (showing homology to essential HAdV-2 34 kDa protein) is not essential for replication of PAdV-3 in vitro. Moreover, successful deletion of 1.957 kb fragment in E4 region increased the available capacity of replication-competent PAdV-3 (E3 + E4 deleted) to approximately 4.3 kb and that of replication-defective PAdV-3 (E1 + E3 + E4 deleted) to approximately 7 kb. This is extremely useful for the construction of PAdV-3 vectors that express multiple genes and/or regulatory elements for gene therapy and vaccination.

Cell transformation by human adenoviruses

The last 40 years of molecular biological investigations into human adenoviruses have contributed enormously to our understanding of the basic principles of normal and malignant cell growth. Much of this knowledge stems from analyses of their productive infection cycle in permissive host cells. Also, initial observations concerning the carcinogenic potential of human adenoviruses subsequently revealed decisive insights into the molecular mechanisms of the origins of cancer, and established adenoviruses as a model system for explaining virus-mediated transformation processes. Today it is well established that cell transformation by human adenoviruses is a multistep process involving several gene products encoded in early transcription units 1A (E1A) and 1B (E1B). Moreover, a large body of evidence now indicates that alternative or additional mechanisms are engaged in adenovirus-mediated oncogenic transformation involving gene products encoded in early region 4 (E4) as well as epigenetic changes resulting from viral DNA integration. In particular, detailed studies on the tumorigenic potential of subgroup D adenovirus type 9 (Ad9) E4 have now revealed a new pathway that points to a novel, general mechanism of virus-mediated oncogenesis. In this chapter, we summarize the current state of knowledge about the oncogenes and oncogene products of human adenoviruses, focusing particularly on recent findings concerning the transforming and oncogenic properties of viral proteins encoded in the E1B and E4 transcription units.

Adenovirus early E4 genes in viral oncogenesis

Previous investigations into potential transforming activities of adenovirus (Ad) early genes were largely overshadowed by the more obvious roles of E1A and E1B products. One exception was an Ad9 E4 protein (ORF1) shown to enhance transformation of cultured cells and promote mammary tumors in female rats. Recently, significant advances in understanding Ad E4 gene products at the molecular level have revealed that these proteins possess an unexpectedly diverse collection of functions, which not only orchestrate many viral processes, but overlap with oncogenic transformation of primary mammalian cells. Operating through a complex network of protein interactions with key viral and cellular regulatory components, Ad E4 products are apparently involved in transcription, apoptosis, cell cycle control, DNA repair, cell signaling, posttranslational modifications and the integrity of nuclear multiprotein complexes known as PML oncogenic domains (PODs). Some of these functions directly relate to known transforming and oncogenic processes, or implicate mechanisms such as modulating the function and subcellular localization of cellular PDZ domain-containing proteins, POD reorganization, targeted proteolytic degradation, inhibition of DNA double-strand break repair and 'hit-and-run' mutagenesis. Here, we summarize the recent data and discuss how E4 gene product interactions may contribute to viral oncogenesis.

The role of adenovirus E4orf4 protein in viral replication and cell killing

It has only been within the last few years that insights have been gained into the remarkable diversity of functions of the adenovirus early transcription region 4 (E4) products. The polypeptide encoded by E4 open reading frame 4 (E4orf4) has emerged as an enigmatic product. Although it accomplishes certain functions that propel viral replication, it has also been shown to be highly toxic, an effect that could dampen the infectious cycle, but that also might serve to facilitate release of viral progeny. When expressed alone, E4orf4 induces a novel form of p53-independent apoptosis in cancer cells but not in normal human cells, thus making it of potential use in cancer gene therapy. In addition, knowledge of its mechanism of action, especially with regard to its interaction with protein phosphatase 2A (PP2A), could provide insights to develop new small molecule anti-cancer drugs. Thus future studies on E4orf4 should be both informative and potentially valuable therapeutically. In this study we review the current status of knowledge on E4orf4.

Mutational analysis of early region 4 of bovine adenovirus type 3

The primary objective of characterizing bovine adenovirus type 3 (BAV3) in greater detail is to develop it as a vector for gene therapy and vaccination of humans and animals. A series of BAV3 early region 4 (E4) deletion-mutant viruses, containing deletions in individual E4 open reading frames (Orf) or combinations of Orfs, were generated by transfecting primary fetal bovine retinal cells with E4-modified genomic DNA. Each of these mutants was further analyzed for growth kinetics, viral DNA accumulation, and early-late protein synthesis. Mutant viruses carrying deletions in Orf1, Orf2, Orf3, or Orf4 showed growth characteristics similar to those of the E3-deleted BAV3 (BAV302). DNA accumulation and early/late protein synthesis were also indistinguishable from those of BAV302. However, mutant viruses carrying a deletion in Orf5, Orfs 1-3 (BAV429), or Orfs 3-5 (BAV430) were modestly compromised in their ability to grow in bovine cells and express early/late proteins. E4 mutants containing larger deletions, Orfs 1-3 (BAV429) and Orfs 3-5 (BAV430), were further tested in a cotton rat model. Both mutants replicated as efficiently as BAV3 or BAV302 in the lungs of cotton rats. BAV3-specific IgA and IgG responses were detected in serum and at the mucosal surfaces in cotton rats inoculated with mutant viruses. In vitro and in vivo characterization of these E4 mutants suggests that none of the individual E4 Orfs are essential for viral replication. Moreover, successful deletion of a 1.5-kb fragment in the BAV3 E4 region increased the available insertion capacity of replication-competent BAV3 vector (E3-E4 deleted) to approximately 4.5 kb and that of replication-defective BAV3 vector (E1a-E3-E4 deleted) to approximately 5.0 kb. This is extremely useful for the construction of BAV3 vectors that express multiple genes and/or regulatory elements for gene therapy and vaccination.

Molecular regulation and biological function of adenovirus early genes: the E4 ORFs

Over the past few years there have been a number of interesting advances in our understanding of the functions encoded by the adenovirus early transcription unit 4 (Ad E4). A large body of recent data demonstrates that E4 proteins encompass an unexpectedly diverse collection of functions required for efficient viral replication. E4 gene products operate through a complex network of protein interactions with key viral and cellular regulatory components involved in transcription, apoptosis, cell cycle control and DNA repair, as well as host cell factors that regulate cell signaling, posttranslational modifications and the integrity of nuclear multiprotein complexes known as nuclear bodies (NBs) or PML oncogenic domains (PODs). As understood at present, some of the lytic functions overlap with roles in oncogenic transformation of primary mammalian cells. These observations, together with findings that E4 proteins substantially affect cell toxicity and the immune response of the host have profound implications for the development of Ad vectors for gene therapy. In this article we will summarize recent findings regarding the diverse functions of E4 gene products in the context of earlier work. We will emphasize the interaction of E4 proteins with cellular and viral interaction partners, the role of these interactions for lytic virus growth and how these interactions may contribute to viral oncogenesis. Finally, we will discuss their role in Ad vector and adeno-associated virus infections.

Transcription map and expression of bovine herpesvirus-1 glycoprotein D in early region 4 of bovine adenovirus-3

Early region 4 (E4) of bovine adenovirus type 3 (BAV-3) was analyzed by Northern blotting, RT-PCR analysis, cDNA sequencing, and S1 nuclease protection assays. The transcriptional map of the E4 region of BAV-3 has marked dissimilarities from those of mouse adenovirus-1, ovine adenovirus-287, and human adenovirus-2, for which the transcriptional maps have been constructed. The E4 region of BAV-3, located between 98.6 and 89.8 MU transcribes seven distinct classes of bovine adenovirus type 3 mRNA. The seven mRNA species formed by the removal of one to three introns share both the 3' end and a short 5' leader (25 nucleotides). The E4 mRNAs can encode at least five unique polypeptides, namely, 143R1, 69R, 143R2, 268R, and 219R. Isolation of a replication-competent recombinant "BAV404" containing 1.9-kb insertion [glycoprotein (gD) of bovine herpesvirus 1, under the control of a SV40 early promoter and poly(A)] in the region between E4 and the right ITR suggested that this region is nonessential for BAV-3 replication. Expression of gD by BAV404 recombinant virus was confirmed by immunoprecipitation with gD-specific monoclonal antibodies. Analysis of the kinetics of protein expression indicated that gD is expressed at both early and late times postinfection. These results suggest that: (a) E4 produces seven 5'-3' coterminal mRNAs and (b) the right terminal region of BAV-3 can be used for the expression of vaccine antigens.

An arginine-faced amphipathic alpha helix is required for adenovirus type 5 e4orf6 protein function

A region in the carboxy terminus of the protein encoded by open reading frame 6 in early region 4 (E4orf6) of adenovirus type 5 was determined to be required for directing nuclear localization of the E1B 55-kDa protein and for efficient virus replication. A peptide encompassing this region, corresponding to amino acids 239 through 255 of the E4orf6 protein, was analyzed by circular dichroism spectroscopy. The peptide showed evidence of self-interaction and displayed the characteristic spectra of an amphipathic alpha helix in the helix-stabilizing solvent trifluoroethanol. Disrupting the integrity of this alpha helix in the E4orf6 protein by proline substitutions or by removing amino acids 241 through 250 abolished its ability to direct the E1B 55-kDa protein to the nucleus when both proteins were transiently expressed in HeLa cells. Expression of E4orf6 variants that failed to direct nuclear localization of the E1B 55-kDa protein failed to enhance replication of the E4 mutant virus, dl1014, whereas expression of the wild-type E4orf6 protein restored growth of dl1014 to near-wild-type levels. These results suggest that the E4orf6 protein contains an arginine-faced, amphipathic alpha helix that is critical for a functional interaction with the E1B 55-kDa protein in the cell and for the function of the E4orf6 protein during a lytic infection.

Integrated viral genomes can be lost from adenovirus type 12-induced hamster tumor cells in a clone-specific, multistep process with retention of the oncogenic phenotype

In adenovirus type 12 (Ad12)-induced tumor cells, in Ad12-transformed cells and in continuously passaged cell lines from these sources, the viral DNA is integrated in multiple copies, usually at a single chromosomal location. In different tumors or cell lines, the sites of integration of Ad12 DNA are all different. Rare exceptions exist. In most instances, the integrated viral DNA resides very stably in the host cell genomes. However, upon continuous serial passage of such cell lines, the integrated viral DNA can be destabilized and lost. In two instances, i.e. in the Ad12-induced hamster tumor cell lines H1111(1) and CLAC1, we have investigated the loss of integrated viral DNA in detail. After extended serial passage, these two cell lines seemed to be devoid of Ad12 DNA sequences, as detectable by Southern blot hybridization, but continued to induce tumors after reinjection into hamsters. Cells from these two cell lines were now recloned three times, and DNAs from cultures derived from several individual clones were reinvestigated for the presence of several parts of the viral genome by the polymerase chain reaction (PCR). Some of the clones still carried parts of the Ad12 genome. However, several clones were isolated that proved free of all parts of the viral genome, except for minute segments from the right terminus of the Ad12 genome. Apparently, the loss of integrated viral DNA from these cell lines proceeded as a continuous, gradual, multistep process whose pattern could differ from cell clone to cell clone, once destabilization had been initiated. The mechanism of destabilization is not understood. Cell populations of 2 x 10(6) to 3 x 10(7), and as low as 10(2), cells from the clones, that contained only minimal remnants from the right viral DNA terminus, were reinjected into newborn or 13-20 day-old weanling Syrian hamsters (Mesocricetus auratus). Tumors developed within 5-17 days after injection. Tumor cell clones also grew in soft agar. The injection of primary hamster skin fibroblasts never elicited tumor formation. The tumor cells induced by this reinjection proved repeatedly free of Ad12 DNA both by Southern blot hybridization and by PCR, except for those cell and tumor clones that contained small segments of the right terminal E4 region of the Ad12 genome. The tumor cells, however, retained their oncogenic phenotype. The results raise questions about the cell clone-specific excision patterns of integrated foreign DNA from the recipient genome and the possibility of a hit-and-run mechanism of adenoviral oncogenesis.

The early region 4 orf4 protein of human adenovirus type 5 induces p53-independent cell death by apoptosis

Previous studies by our group showed that infection of human and rodent cells by human adenovirus type 5 (Ad5) results in the induction of p53-independent apoptosis and cell death that are dependent upon transactivation of early region 4 (E4). To identify which E4 products are involved, studies were conducted with p53-deficient human SAOS-2 cells infected with various Ad5 E4 mutants. An E4orf6-deficient mutant was defective in cell killing, whereas another that expressed only E4orf6 and E4orf4 killed like wild-type virus, suggesting that E4orf6 may be responsible for cytotoxicity; however, a mutant expressing only E4orf4 induced high levels of cell death, indicating that this E4 product may also be able to induce cytotoxicity. To define the E4 cell death-inducing functions more precisely, cDNAs encoding individual E4 products were introduced into cells by DNA transfection in the absence of other Ad5 proteins. In cotransfections with a cDNA encoding firefly luciferase, enzymatic activity was high in all cases except with E4orf4, where luciferase levels were less than 20% of those in controls. In addition, drug selection of several cell types following transfection with retroviral vector DNA encoding individual E4 products as well as puromycin resistance yielded a large number of cell colonies except when E4orf4 was expressed. These data demonstrated that E4orf4 is the only E4 product capable of independent cell killing. Cell death induced by E4orf4 was due to apoptosis, as evidenced by 4',6-diamidino-2-phenylindole (DAPI) staining of cell nuclei in E4orf4-expressing cells. Thus, although E4orf6 may play some role, these results suggested that E4orf4 may be the major E4 product responsible for induction of p53-independent apoptosis.

Adenovirus type 5 E4 open reading frame 4 protein induces apoptosis in transformed cells

Adenovirus type 5 E4 open reading frame 4 (E4orf4) protein has been previously shown to counteract transactivation of the junB and c-fos genes by cyclic AMP plus E1A protein and to interact with protein phosphatase 2A (PP2A). Here, we show that the wild-type E4orf4 protein induces apoptosis in the E1A-expressing 293 cells, in NIH 3T3 cells transformed with v-Ras, and in the lung carcinoma cell line H1299. The induction of apoptosis is not accompanied by enhanced levels of p53 in 293 cells and occurs in the absence of p53 in H1299 cells, indicating involvement of a p53-independent pathway. A mutant E4orf4 protein that had lost the ability to induce apoptosis also lost its ability to bind PP2A. We suggest that E4orf4 antagonizes continuous signals to proliferate, like those given by E1A or v-Ras, and that the conflicting signals lead to the induction of cell death.

Characterization of the early region 4 of porcine adenovirus type 3

The nucleotide sequence of a 3028 bp DNA segment, located between map co-ordinates 100 and 92 in the genome of porcine adenovirus type 3 (PAV-3), was determined. The segment includes the entire early region 4 (E-4) and the right inverted terminal repeat sequences. There were two TATA boxes and one canonical polyadenylation signal on the 1 strand. Homology searches of the GenBank data base for the predicted amino acid sequences revealed that, of the eight open reading frames (ORFs) on the 1 strand, and four ORFs on the r strand, only ORF 8 on the 1 strand showed homology with the 34 kDa E-4 protein of human adenovirus types 2, 12 and 34. Northern blot analysis showed that transcription from the E-4 region of PAV-3 began 4 h after infection, peaked at 8 h and declined after 10 h, before DNA replication began 16 h after infection. The E-4 region of PAV-3 was further characterized by 5' and 3' end mapping of the transcription unit.

Early region 4 modulates adenovirus DNA replication by two genetically separable mechanisms

Three viral proteins, all products of early region 2 (E2), participate directly in adenovirus DNA replication. Three products of early region 4 (E4) also affect viral DNA synthesis: the product of E4 ORF4 inhibits viral DNA accumulation, while the products of E4 ORFs 3 and 6 antagonize that effect of ORF4 expression. Because no E4 products are required for DNA synthesis, these proteins probably act indirectly. The E4 ORF3, 4, and 6 proteins all participate in aspects of the regulation of viral gene expression. To determine whether they modulate DNA replication by effects on expression of viral replication proteins, we examined E2 expression in E4 mutant-infected cells. In cells infected by ORF3-, 6- mutants, expression of ORF4 substantially depressed the steady-state levels of replication proteins and E2 mRNAs, reduced E2 transcription rates, and profoundly inhibited viral DNA replication. Thus, in the absence of E4 ORFs 3 and 6, ORF4 acts as a transcriptional regulator of E2 expression, and reduced replication protein levels largely account for the inhibition of DNA replication by ORF4. Cells infected by viruses that express ORFs 3 and 6 in addition to ORF4 accumulated much larger quantities of viral DNA than did cells infected by the ORF3-, 6-, 4+ mutant. Increased DNA accumulation was not accompanied by a comparable increase in E2 expression. Therefore, the ORF3 and 6 products counteract the ORF4-induced reduction of DNA replication by a mechanism other than reversing the inhibitory effect of ORF4 on E2 expression. The effect of ORF4 on E2 expression is consistent with its ability to regulate levels of the transcription factor AP-1 (Müller et al., 1992, J. Virol. 66, 5867-5878); the mechanism by which ORFs 3 and 6 enhance replication is unknown.

A gene transfer vector-cell line system for complete functional complementation of adenovirus early regions E1 and E4

The improvements to adenovirus necessary for an optimal gene transfer vector include the removal of virus gene expression in transduced cells, increased transgene capacity, complete replication incompetence, and elimination of replication-competent virus that can be produced during the growth of first-generation adenovirus vectors. To achieve these aims, we have developed a vector-cell line system for complete functional complementation of both adenovirus early region 1 (E1) and E4. A library of cell lines that efficiently complement both E1 and E4 was constructed by transforming 293 cells with an inducible E4-ORF6 expression cassette. These 293-ORF6 cell lines were used to construct and propagate viruses with E1 and E4 deleted. While the construction and propagation of AdRSV beta gal.11 (an E1-/E4- vector engineered to contain a deletion of the entire E4 coding region) were possible in 293-ORF6 cells, the yield of purified virus was depressed approximately 30-fold compared with that of E1- vectors. The debilitation in AdRSV beta gal.11 vector growth was found to correlate with reduced fiber protein and mRNA accumulation. AdCFTR.11A, a modified E1-/E4- vector with a spacer sequence placed between late region 5 and the right inverted terminal repeat, efficiently expressed fiber and grew with the same kinetic profile and virus yield as did E1- vectors. Moreover, purified AdCFTR.11A yields were equivalent to E1- vector levels. Since no overlapping sequences exist in the E4 regions of E1-/E4- vectors and 293-ORF6 cell lines, replication-competent virus cannot be generated by homologous recombination. In addition, these second-generation E1-/E4- vectors have increased transgene capacity and have been rendered virus replication incompetent outside of the new complementing cell lines.

Adenovirus type 5 early region 4 is responsible for E1A-induced p53-independent apoptosis

In the absence of E1B, the 289- and 243-residue E1A products of human adenovirus type 5 induce p53-dependent apoptosis. However, our group has shown recently that the 289-residue E1A protein is also able to induce apoptosis by a p53-independent mechanism (J. G. Teodoro, G. C. Shore, and P. E. Branton, Oncogene 11:467-474, 1995). Preliminary results suggested that p53-independent cell death required expression of one or more additional adenovirus early gene products. Here we show that both the E1B 19-kDa protein and cellular Bcl-2 inhibit or significantly delay p53-independent apoptosis. Neither early region E2 or E3 appeared to be necessary for such cell death. Analysis of a series of E1A mutants indicated that mutations in the transactivation domain and other regions of E1A correlated with E1A-mediated transactivation of E4 gene expression. Furthermore, p53-deficient human SAOS-2 cells infected with a mutant which expresses E1B but none of the E4 gene products remained viable for considerably longer times than those infected with wild-type adenovirus type 5. In addition, an adenovirus vector lacking both E1 and E4 was unable to induce DNA degradation and cell killing in E1A-expressing cell lines. These data showed that an E4 product is essential for E1A-induced p53-independent apoptosis.

Adenovirus E4 open reading frame 4 protein autoregulates E4 transcription by inhibiting E1A transactivation of the E4 promoter

Here we show that the adenovirus early region 4 (E4) open reading frame 4 (ORF4) protein autoregulates its own transcription by inhibiting adenovirus E1A-induced activation of E4 transcription both in transient transfection experiments and during lytic virus growth. The inhibitory activity of E4-ORF4 was selective for E1A-CR3-dependent transactivation and had no effect on CR1 transactivation. The inhibitory activity of E4-ORF4 was relieved by okadaic acid treatment, which inhibits the cellular protein phosphatase 2A (PP2A), suggesting that E4-ORF4 controls the phosphorylated status of transcription factors important for E4 promoter activity. This conclusion agrees with previous demonstrations that E4-ORF4 associates with PP2A and causes a partial dephosphorylation of certain transcription factors, including E1A (U. Müller, T. Kleinberger, and T. Shenk, J. Virol. 66:5869-5878, 1992; T. Kleinberger and T. Shenk, J. Virol. 67:7556-7560, 1993). However, our results indicate that dephosphorylation of E1A itself might not be the primary target for E4-ORF4. Instead, the E4-ORF4-PP2A complex appears to work by dephosphorylation of multiple cellular transcription factors that are involved in E1A transactivation of the E4 promoter.

Adenovirus replication is coupled with the dynamic properties of the PML nuclear structure

Wild-type PML and at least four other novel proteins are localized within discrete nuclear structures known as PODs. We demonstrate here that during adenovirus infection, immediate early viral proteins from the E1 and E4 transcription units associate with the POD, which in turn undergoes a dramatic morphological change. During this process, the auto-antigen Sp-100 and NDP55 but not PML, relocate from the POD to the viral inclusion bodies, the sites of adenovirus DNA replication and late RNA transcription. The E4-ORF3 11-kD protein alone will induce this reorganization and reciprocally, viruses carrying mutations in the E4-domain fail to do so. These same viral mutants are defective in viral replication as well as the accumulation of late viral mRNAs and host cell transcription shutoff. We show that interferon (INF) treatment enhances the expression of PML, reduces or blocks PODs reorganization, and inhibits BrdU incorporation into viral inclusion bodies. In addition, cell lines engineered to overexpress PML prevent PODs from viral-induced reorganization and block or severely delay adenovirus replication. These results suggest that viral replication relies on components of the POD and that the structure is a target of early viral proteins.

Efficient dual transcomplementation of adenovirus E1 and E4 regions from a 293-derived cell line expressing a minimal E4 functional unit

Transgene expression after the administration of recombinant adenovirus with E1 deleted is constantly transient. It is admitted that E1A-substituting activities of cellular or viral origin allow viral antigen synthesis and trigger cytotoxic lymphocyte-mediated clearance of the recipient cells. Our approach to solving this problem relies on the additional deletion of the E4 region from the vector backbone as this region upregulates viral gene expression at both transcriptional and posttranscriptional levels. As a prerequisite to the construction of E1 E4 doubly defective adenoviruses, we investigated the possibility of transcomplementing both functions within a single cell. In particular, the distal ORF6+ORF7 segment from the E4 locus of adenovirus type 5 was cloned under the control of the dexamethasone-inducible mouse mammary tumor virus long terminal repeat. Following transfection into 293 cells, clone IGRP2 was retained and characterized as it can rescue the growth defect of all E1+ E4- adenoviral deletants tested. DNA and RNA analysis experiments verified that the mouse mammary tumor virus promoter drives the expression of the ORF6+ORF7 unit and permits its bona fide alternative splicing, generating ORF6/7 mRNA in addition to the ORF6-expressing primary transcript. Importantly, IGRP2 cells sustain cell confluence for a period longer than that of 293 parental cells and allow the plaque purification of E1- or E4- defective viruses. The dual expression of E1 and E4 regulatory genes within IGRP2 cells is demonstrated by the construction, plaque purification, and helper-free propagation of recombinant lacZ-encoding doubly defective adenoviruses harboring different E4 deletions. In addition, the emergence, if any, of replicative particles during viral propagation in this novel packaging cell line will be drastically impaired as only a limited segment of E4 has been integrated.

Development of cell lines capable of complementing E1, E4, and protein IX defective adenovirus type 5 mutants

The cloning capacity of currently available E1- and E3-deleted adenovirus (Ad) vectors does not exceed 8 kb. To increase capacity and improve vector safety further, we have explored the possibility that Early Region 4 (E4) and the gene encoding protein IX (pIX) might also be deleted. To generate cell lines expressing sufficient levels of E4 and pIX proteins in trans in addition to E1-encoded proteins to complement mutations in these genes, we transformed 293 cells with constructs containing the E4 transcription unit and pIX coding sequences under the control of inducible mouse mammary tumor virus (MMTV) and metallothionein promoters, respectively. We obtained two lines, VK2-20 and VK10-9, that express both E4 and pIX proteins as well as E1. The lines could be efficiently transfected with DNA, and allowed the rescue and propagation of an adenovirus; recombinant, Ad5dlE3,4, containing a 2.7-kb E3 deletion and a 2.8-kb E4 deletion in addition to an insertion of plasmid DNA sequences in E1A. Because the E4 sequences within VK2-20 and VK10-9 cells do not overlap with the DNA sequence of Ad5dlE3,E4, the probability of regeneration of the wild-type E4 during virus propagation should be very low. Using the cell lines described in this study, it should be possible to generate Ad vectors lacking E1, pIX, E3, and E4. This would not only increase capacity over that of currently available vectors (to approximately 11 kb) but would also result in more severely attenuated vectors than those with deletions only of E1 or of E1 and E3 and, hence, safer for use in gene therapy protocols.

Targeting of adenovirus E1A and E4-ORF3 proteins to nuclear matrix-associated PML bodies

The PML protein was first identified as part of a fusion product with the retinoic acid receptor alpha (RAR alpha), resulting from the t(15;17) chromosomal translocation associated with acute promyelocytic leukemia (APL). It has been previously demonstrated that PML, which is tightly bound to the nuclear matrix, concentrates in discrete subnuclear compartments that are disorganized in APL cells due to the expression of the PML-RAR alpha hybrid. Here we report that adenovirus infection causes a drastic redistribution of PML from spherical nuclear bodies into fibrous structures. The product encoded by adenovirus E4-ORF3 is shown to be responsible for this reorganization and to colocalize with PML into these fibers. In addition, we demonstrate that E1A oncoproteins concentrate in the PML domains, both in infected and transiently transfected cells, and that this association requires the conserved amino acid motif (D)LXCXE, common to all viral oncoproteins that bind pRB or the related p107 and p130 proteins. The SV-40 large T antigen, another member of this oncoprotein family is also found in close association with the PML nuclear bodies. Taken together, the present data indicate that the subnuclear domains containing PML represent a preferential target for DNA tumor viruses, and therefore suggest a more general involvement of the PML nuclear bodies in oncogenic processes.

Human adenovirus encodes two proteins which have opposite effects on accumulation of alternatively spliced mRNAs

All mRNAs expressed from the adenovirus major late transcription unit have a common, 201-nucleotide-long 5' leader sequence, which consists of three short exons (the tripartite leader). This leader has two variants, either with or without the i-leader exon, which, when present, is spliced between the second and the third exons of the tripartite leader. Previous studies have shown that adenovirus early region 4 (E4) encodes two proteins, E4 open reading frame 3 (E4-ORF3) and E4-ORF6, which are required for efficient expression of mRNAs from the major late transcription unit. These two E4 proteins appear to have redundant activities, and expression of one has been shown to be sufficient for efficient major late mRNA accumulation during a lytic virus infection. In this report, we provide evidence that E4-ORF3 and E4-ORF6 both regulate major late mRNA accumulation by stimulating constitutive splicing. Moreover, we show that the two proteins have different effects on accumulation of alternatively spliced tripartite leader exons. In a DNA transfection assay, E4-ORF3 was shown to facilitate i-leader exon inclusion, while E4-ORF6 preferentially favored i-leader exon skipping. In addition, E4-ORF3 and E4-ORF6 had the same effects on accumulation of alternatively spliced chimeric beta-globin transcripts. This finding suggests that the activities of the two proteins may be of more general relevance and not restricted to splicing of major late tripartite leader-containing pre-mRNAs. Interestingly, E4-ORF6 expression was also shown to stimulate i-leader exon skipping during a lytic virus infection.

Human adenovirus encodes two proteins which have opposite effects on accumulation of alternatively spliced mRNAs

All mRNAs expressed from the adenovirus major late transcription unit have a common, 201-nucleotide-long 5' leader sequence, which consists of three short exons (the tripartite leader). This leader has two variants, either with or without the i-leader exon, which, when present, is spliced between the second and the third exons of the tripartite leader. Previous studies have shown that adenovirus early region 4 (E4) encodes two proteins, E4 open reading frame 3 (E4-ORF3) and E4-ORF6, which are required for efficient expression of mRNAs from the major late transcription unit. These two E4 proteins appear to have redundant activities, and expression of one has been shown to be sufficient for efficient major late mRNA accumulation during a lytic virus infection. In this report, we provide evidence that E4-ORF3 and E4-ORF6 both regulate major late mRNA accumulation by stimulating constitutive splicing. Moreover, we show that the two proteins have different effects on accumulation of alternatively spliced tripartite leader exons. In a DNA transfection assay, E4-ORF3 was shown to facilitate i-leader exon inclusion, while E4-ORF6 preferentially favored i-leader exon skipping. In addition, E4-ORF3 and E4-ORF6 had the same effects on accumulation of alternatively spliced chimeric beta-globin transcripts. This finding suggests that the activities of the two proteins may be of more general relevance and not restricted to splicing of major late tripartite leader-containing pre-mRNAs. Interestingly, E4-ORF6 expression was also shown to stimulate i-leader exon skipping during a lytic virus infection.

Activation of RNA polymerase III transcription of human Alu repetitive elements by adenovirus type 5: requirement for the E1b 58-kilodalton protein and the products of E4 open reading frames 3 and 6

We found that transcription of endogenous human Alu elements by RNA polymerase III was strongly stimulated following infection of HeLa cells with adenovirus type 5, leading to the accumulation of high levels of Alu transcripts initiated from Alu polymerase III promoters. In contrast to previously reported cases of adenovirus-induced activation of polymerase III transcription, induction required the E1b 58-kDa protein and the products of E4 open reading frames 3 and 6 in addition to the 289-residue E1a protein. In addition, E1a function was not required at high multiplicities of infection, suggesting that E1a plays an indirect role in Alu activation. These results suggest previously unsuspected regulatory properties of the adenovirus E1b and E4 gene products and provide a novel approach to the study of the biology of the most abundant class of dispersed repetitive DNA in the human genome.

Regulated splicing of adenovirus type 5 E4 transcripts and regulated cytoplasmic accumulation of E4 mRNA

The E4 gene of human type C adenoviruses has been shown previously to give rise to an array of mRNAs via differential splicing. In this study, the pattern of expression of these mRNAs during lytic infection was examined, and two distinct temporal classes were defined. mRNAs of the early class were distinguished from those of the late class by the presence, in the early class, of a sequence in the 3' half of the mRNA that was removed as an intron in the late class. A single mRNA of the late class was found to show a strong dependence on the presence of the 55-kDa protein from region E1b and the open reading frame 6 protein from region E4 for its normal cytoplasmic accumulation. One feature of this mRNA that distinguishes it from other E4 mRNAs expressed at late times is the retention within it of an intron from the 5' half of E4; it may therefore be recognized as incompletely spliced by the host cell and retained in the nucleus. It is proposed that the E1b 55-kDa/E4 open reading frame 6 protein complex facilitates accumulation of this mRNA by overcoming this retention mechanism.

Activation of RNA polymerase III transcription of human Alu repetitive elements by adenovirus type 5: requirement for the E1b 58-kilodalton protein and the products of E4 open reading frames 3 and 6

We found that transcription of endogenous human Alu elements by RNA polymerase III was strongly stimulated following infection of HeLa cells with adenovirus type 5, leading to the accumulation of high levels of Alu transcripts initiated from Alu polymerase III promoters. In contrast to previously reported cases of adenovirus-induced activation of polymerase III transcription, induction required the E1b 58-kDa protein and the products of E4 open reading frames 3 and 6 in addition to the 289-residue E1a protein. In addition, E1a function was not required at high multiplicities of infection, suggesting that E1a plays an indirect role in Alu activation. These results suggest previously unsuspected regulatory properties of the adenovirus E1b and E4 gene products and provide a novel approach to the study of the biology of the most abundant class of dispersed repetitive DNA in the human genome.

Adenovirus E4orf4 protein reduces phosphorylation of c-Fos and E1A proteins while simultaneously reducing the level of AP-1

Adenovirus E1A protein and cyclic AMP cooperate to induce transcription factor AP-1 and viral gene expression in mouse S49 cells. We report that a protein encoded within the viral E4 gene region acts to counterbalance the induction of AP-1 DNA-binding activity by E1A and cyclic AMP. Studies with mutant adenoviruses demonstrated that in the absence of E4orf4 protein, AP-1 DNA-binding activity is induced to substantially higher levels than in wild-type virus-infected cells. The induction is the result of increased production of JunB and c-Fos proteins. Hyperphosphorylated forms of c-Fos and E1A proteins accumulate in the absence of functional E4orf4 protein. We propose that the E4orf4 protein acts to inhibit the activity of a cellular kinase that phosphorylates both the E1A and c-Fos proteins. Phosphorylation-dependent alterations in the activity of c-Fos, E1A, or some unidentified protein might, then, lead to decreased synthesis of AP-1 components. This E4 function likely plays an important role in natural infections, since a mutant virus unable to express the E4orf4 protein is considerably more cytotoxic than the wild-type virus.

Transcription mapping of mouse adenovirus type 1 early region 4

Early region 4 (E4) of mouse adenovirus type 1 was analyzed by Northern blotting, cDNA sequencing, and S1 nuclease protection and primer extension assays. The transcription map of this region was dissimilar to the consensus human adenovirus E4 transcription map in which all transcripts have identical 5' and 3'-terminal sequences. Seven classes of mouse adenovirus type 1 mRNAs were identified; all shared the same 3' end. Three classes of unspliced mRNAs differed at their 5' start sites, two classes of spliced transcripts differed in the locations of their splice acceptors, and two classes of spliced messages differed in their splice donors and acceptors. From the structure of the various transcripts, translational products were predicted. In addition to a predicted polypeptide with similarity to the human adenovirus 2 E4 34K protein previously identified (A. O. Ball, C. W. Beard, P. Villegas, and K. R. Spindler, 1991, Virology 180, 257-265), two open reading frames with similarity to human adenovirus 2 E4 open reading frames 2 and 3 were found.

Enteric adenovirus type 40: complementation of the E4 defect in Ad2 dl808

The enteric adenovirus type 40 cannot be passaged in HeLa cells, but will grow productively in cells that express the E1B region of adenovirus types 2 or 5. Even in such permissive cells, the lytic cycle is prolonged, there is an abnormal pattern of E1B early gene expression and a failure to switch off host cell functions, suggesting that other gene functions might be impaired in Ad40. For Ad2, E4 ORF 6 and ORF 3 proteins are known to have an essential role in progressing from the early to the late phase of lytic infection and the shutoff of host functions requires an interaction between the E4 ORF 6 34K protein and the E1B 55K protein. To test whether E4 functions of Ad40 are impaired, complementation tests have been made between Ad40 and the E4 deletion mutant Ad2 dl808, which lacks all but ORF 1 of the E4 region. In HeLa and Vero cells, Ad40 complements dl808 to levels equivalent to an Ad2 wild-type infection, as demonstrated by measuring virion packaged DNA, virus titration, and viral protein synthesis. Surprisingly, Ad2 dl808 fails to reciprocally complement Ad40. The results show that Ad40 produces functional E4 ORF 6 and/or ORF 3 activity, and that their expression precedes DNA replication.

The metabolism of host RNAs in cells infected by an adenovirus E4 mutant

Mutants of adenovirus type 5 (Ad5) that lack early region 4 (E4) are defective in the expression of viral late genes. E4 mutants exhibit dramatically reduced levels of both cytoplasmic and nuclear viral late RNAs compared to wild-type virus, due principally to reduced stability of unprocessed viral late RNA in the nucleus of mutant-infected cells. To determine whether E4 products also affect the metabolism of host RNAs in infected cells, steady-state levels of beta-actin RNA and triose phosphate isomerase (TPI) RNA were measured in the cytoplasms and nuclei of HeLa cells infected by either wild-type Ad5 or the E4 deletion mutant H5dl1004, and were compared to levels in uninfected HeLa cells. S1 nuclease analyses revealed only slight reductions in beta-actin mRNA levels in the cytoplasm and in levels of spliced and unspliced beta-actin RNA in the nucleus of cells infected by either Ad5 or H5dl1004. RNase protection analyses showed that cytoplasmic TPI RNA levels were not affected by infection of HeLa cells with either Ad5 or H5dl1004. Steady-state levels of nuclear TPI RNA, both spliced and unspliced, were slightly reduced in cells infected by wild-type virus but not in HeLa cells infected by H5dl1004. These results indicate that the reduced stability of RNA in HeLa cells infected by E4 mutants is a virus-specific phenotype which does not extend to host cell RNAs.

Adenovirus E4-dependent activation of the early E2 promoter is insufficient to promote the early-to-late-phase transition

The adenovirus E4 ORF6/7 protein has been shown to activate the cellular transcription factor E2F. E2F activation leads to activation of the adenovirus early E2 promoter which controls the production of viral DNA replication proteins. In the present study an adenovirus type 5 cDNA mutant, H5ilE4L, was constructed. This mutant is capable of making the ORF6/7 polypeptide but lacks the coding sequences for all other E4 products. H5ilE4L trans activates the early E2 promoter to wild-type levels, but still it is defective for viral DNA replication. A mutant expressing ORF6 in addition to ORF6/7, H5ilE4I, is normal for viral DNA replication. This indicates that activation of the early E2 promoter is insufficient to promote efficient viral DNA replication and that another E4-encoded function is necessary. The ORF6 protein seems to provide this function. We suggest that ORF6/7-induced activation of E2F is not necessary for adenovirus growth in HeLa cells. Rather, this activation might be of importance in the normal, growth-arrested host cell, since E2F has been shown to bind to the promoter regions of a number of immediate-early genes involved in regulation of cell proliferation (M. Mudryj, S. W. Hiebert, and J. R. Nevins, EMBO J. 9:2179-2184, 1990).

Early region 4 sequence and biological comparison of two isolates of mouse adenovirus type 1

The DNA sequence of 88-100 map units of mouse adenovirus type 1 (MAV-1) was determined. One translational open reading frame showed 48% sequence similarity to a human adenovirus type 2 early region 4 protein. Based on the protein similarity, genome location, and transcriptional polarity, we concluded that this region of MAV-1 corresponds to early region 4. A 241-bp sequence consisting of 10 imperfect direct repeats with sequence similarity to minisatellite DNA was found in this region. Two virus isolates with different passage histories were examined and were found to have a sequence polymorphism within this region. The two viruses were compared for growth in cell culture and mice and small quantitative differences were observed only in vivo.

The adenovirus E4 17-kilodalton protein complexes with the cellular transcription factor E2F, altering its DNA-binding properties and stimulating E1A-independent accumulation of E2 mRNA

E2F is a cellular DNA-binding factor. Its binding activity is changed within adenovirus-infected cells so that it binds cooperatively to pairs of properly spaced and oriented E2F recognition sites. In the work described in this report, the conversion to cooperative binding was shown to require the adenovirus E4 17-kilodalton (kDa) polypeptide. Mutant viruses carrying alterations within the E4 17-kDa coding region failed to generate the infection-specific, cooperatively binding form of E2F. It was possible to alter E2F from uninfected cells so that it bound cooperatively by incubation with a partially purified fraction obtained from infected cells. The E4 17-kDa protein copurified with this activity and was also found to be present in a complex containing E2F. Consistent with its ability to alter the binding of E2F to its recognition sites within the E2 promoter, the E4 17-kDa polypeptide contributed to maximal expression of E2 mRNAs in some cell types. Its ability to enhance E2 transcription did not require expression of the E1A transactivator protein. These results are consistent with a model which proposes that the E4 17-kDa polypeptide binds to the cellular E2F factor, altering its binding behavior and thereby enhancing its ability to stimulate transcription.

An adenovirus early region 4 gene product is required for induction of the infection-specific form of cellular E2F activity

E2F is a cellular, sequence-specific DNA-binding factor that binds to pairs of sites that occur upstream of the E1A and E2 early mRNA cap sites. During adenovirus infection, there is induction of a form of E2F that binds cooperatively to the pair of sites in the E2 control region. Production of the infection-specific E2F activity is dependent on early region 4 (E4), as extracts of cells infected with a mutant that lacks E4 did not contain this activity. Instead, two new forms of E2F were seen with the E4 mutant. Infection with mutant viruses unable to make E1A gene products produced the wild-type infection-specific E2F activity after a delay. Mutations in the E1B-55 kD-, E1B-21 kD-, E2-72 kD-, and E3-coding regions had no effect on production of infection-specific E2F. Analysis of cell lines confirmed the results obtained with mutant viruses. Cells that expressed E1A but not E4 genes (e.g., 293 cells) did not contain infection-specific E2F. Cell lines that expressed the E4 gene contained the activity. These observations demonstrate that E4 participates in the infection-induced change in E2F-binding activity. The data are consistent with E1A playing an indirect role in the process by mediating the efficient expression of E4 gene products which, in turn, induce the alteration in E2F activity.

Adenovirus early region 4 encodes two gene products with redundant effects in lytic infection

In order to assign specific functions to individual gene products encoded by adenovirus type 5 early region 4 (E4), we have constructed and analyzed a set of mutant viruses that express individual E4 open reading frames or combinations of open reading frames. The results of these analyses demonstrate that the gene products of E4 open reading frames 3 and 6 have redundant effects in viral lytic infection. These E4 products independently augment viral DNA replication, viral late protein synthesis, the shutoff of host cell protein synthesis, and the production of infectious virus. The product of open reading frame 6 is more efficient in the regulation of these processes than is the product of open reading frame 3. The regulation of viral DNA replication and the control of viral and cellular protein synthesis appear to be separable functions associated with both E4 gene products. The role of early region 4 in adeno-associated virus helper function, however, is mediated only by the product of open reading frame 6. Finally, we demonstrate that E4 mutant viruses display a multiplicity-leakiness phenotype which is consistent with the regulatory role that this region plays in viral infection.

Complementation of adenovirus E4 mutants by transient expression of E4 cDNA and deletion plasmids

Human adenovirus mutants that carry a large deletion in early region 4 (E4) are severely defective in the synthesis of viral late proteins. Plasmids that carry intact E4 sequences can complement the late protein synthetic defect of such mutants when introduced into infected cells by transfection, presumably due to the transient expression of E4 products. Cells transfected with cDNA clones capable of expressing E4 open reading frame (ORF) 6, or deletion mutant clones expected to express either E4 ORF 6 or E4 ORF 3, also complement the mutants' defects. Thus, these E4 ORFs can individually satisfy the requirement for E4 products in viral late gene expression, and function effectively in the absence of other E4 products. Some E4 deletion mutants also exhibit a defect in the production of viral DNA. All of the clones that stimulate late gene expression also enhance one such mutant's ability to accumulate viral DNA. Thus, the ORF 3 and ORF 6 products are also individually sufficient to provide an E4 function necessary for normal viral DNA replication in the absence of other E4 products.

Adenovirus early region 4 is essential for normal stability of late nuclear RNAs

H2dl808 is a deletion mutant of adenovirus type 2 lacking most of transcriptional early region E4. In most normal adenovirus host cells this virus displayed a complex mutant phenotype that included a dramatic reduction in the level of cytoplasmic late RNA, a corresponding defect in late protein synthesis, and a 5- to 10-fold defect in viral DNA accumulation. H5dl1004 is a deletion mutant of adenovirus type 5 that also lacks a portion of E4. It exhibited a reduction in levels of cytoplasmic late RNAs that was somewhat less severe than that of H2dl808 and a corresponding late protein synthetic defect but no defect in the production of viral DNA. In addition to the defect in the accumulation of late cytoplasmic mRNAs, HeLa cells infected by either H2dl808 or H5dl1004 showed substantially reduced levels of viral RNAs in their nuclei at late times after infection. Both mature mRNAs and apparent mRNA precursors were affected. The late transcription rates of the deletion mutant viruses were similar to that of wild-type virus. These results suggest that the underaccumulation of RNA in H2dl808- and H5dl1004-infected cells is caused by a reduction in the stability of viral RNA in the nucleus, and they implicate E4 products in a novel aspect of the regulation of viral gene expression.

Redundant control of adenovirus late gene expression by early region 4

A series of human adenovirus type 5 derivatives carrying deletion mutations in early region 4 (E4) were constructed and characterized with respect to viral late protein synthesis, viral cytoplasmic late message accumulation, viral DNA accumulation, and plaquing ability. Viral late protein synthesis was essentially normal in cells infected by mutants expected to produce either the E4 open reading frame (ORF) 3 product or the E4 ORF 6 product. In cells infected by mutants lacking both ORF 3 and ORF 6, late protein synthesis was dramatically reduced. The basis for this reduction appears to be a concomitant reduction in cytoplasmic late message levels. Our results suggest that the products of ORFs 3 and 6 are redundant, since they are individually able to satisfy the requirement for E4 in late gene expression. Two of the mutants examined were defective for viral late protein synthesis but showed no measurable defect in viral DNA accumulation. The defect in late gene expression is not, therefore, a reflection of a primary defect in viral DNA synthesis. Finally, mutants expected to express ORF 3 or ORF 6 formed plaques with normal or only modestly reduced efficiency, whereas mutants expected to express neither ORF formed plaques with an efficiency less than 10(-6) that of wild-type virus. Thus, plaque-forming ability reflected late protein synthetic ability, suggesting that among these mutants late protein synthetic proficiency is the principle determinant of plaquing efficiency.

Gene product of region E4 of adenovirus type 5 modulates accumulation of certain viral polypeptides

An adenovirus type 5 mutant, designated H5ilE4I, was constructed in which region E4 was replaced by a cloned cDNA. The cDNA was a copy of an mRNA which exclusively contains open translational reading frames 6 and 7. The phenotype of the mutant was compared with that of the previously characterized E4 mutant H2dl808 and wild-type adenovirus 5. Although the H5ilE4I mutant lacked at least five E4 genes, it was nondefective for growth in HeLa cells. The defects in viral DNA replication, late protein synthesis, and shutoff of host cell protein synthesis associated with the phenotype of the H2dl808 mutant were not observed in HeLa cells infected with the H5ilE4I mutant. However, differences were observed regarding the time of onset of viral DNA replication and the accumulation of the hexon polypeptide as well as the 72-kilodalton adenovirus-specific DNA-binding protein. The results thus indicate that open reading frame 6 or 7 or both contain all genetic information required for viral replication in tissue culture cells, whereas another E4 gene modulates the accumulation of certain viral polypeptides. The early onset of viral DNA replication in H5ilE4I-infected cells may be an indirect effect of the enhanced expression of the 72-kilodalton DNA-binding protein.

Adenovirus early region 4 is required for efficient virus particle assembly

H2dl807, a defective deletion mutant of human adenovirus type 2 lacking parts of early regions 3 and 4 and all of late region 5, was severely defective for virus particle assembly on HeLa cells, producing about 1% of the normal yield of particles. On Vero cells, H2dl807 produced only 5% as many particles as wild type, while on W162 cells, a Vero cell derivative which supports the growth of early region 4 mutants, H2dl807 produced nearly 40% of the wild-type level of particles. Two other defective deletion mutants, H2dl802 and H5dl1021, which lack parts of early region 3 and which are incapable of making fiber, the product of late region 5, were wild type for virus assembly. These data suggest that the cause of the assembly defect of H2dl807 is the lack of a diffusible early region 4 product. H2dl807-infected Vero cells accumulated nearly wild-type amounts of viral late proteins in the nucleus and cytoplasm. Thus, the defect of the mutant in assembly on Vero cells is not due to a general lack of late proteins. Finally, the fact that H2dl802 and H5dl1021 make wild-type amounts of virus particles suggests that fiber is not essential for adenovirus assembly.