Phillip H. Gallimore

Adenovirus E1A: remodelling the host cell, a life or death experience

It came as a considerable surprise to virologists in 1962 when John Trentin and his colleagues (Trentin et al., 1962) reported that a human adenovirus was oncogenic. At Baylor University (Houston, Texas) they had been inoculating newborn hamsters with human adenoviruses and found that adenovirus serotype 12 (Ad12) induced tumours in a high proportion of the animals injected. Other Ad serotypes (Ads) were found to be non-tumorigenic (this included Ad2, which with its closely related non-oncogenic serotype Ad5, have been the most extensively studied Ads; referred to in this review as Ad2/5). This important observation was initially greeted with some scepticism (see Gross, 1966). However, Ad12 oncogenesis was rapidly con®rmed by others (Huebner et al., 1962) and lead to more extensive in vivo and in vitro studies on the capacity of human Ads to disrupt normal cell growth control. It is not possible to review all of the literature on this subject here but there are a number of reviews that cover this subject (Gross, 1966; Gallimore et al., 1984; Williams et al., 1995). From the in vivo studies in the 1960s it soon became clear that there were indeed tumorigenic (Huebner et al., 1962; Trentin et al., 1962, Girardi et al., 1964; Pereira et al., 1965) and nontumorigenic adenoviruses (Trentin et al., 1962, 1968). As well as Ad serotype, adenovirus induced tumorigenicity was found to be dependent on virus dose (Yabe et al., 1962), host genetic constitution (Yabe et al., 1964, Yohn et al., 1965; Allison et al., 1967), age at inoculation (Yabe et al., 1962) and the hosts immune status (Yohn et al., 1965; Allison et al., 1967). Around this time tissue culture studies revealed that rodent cells were susceptible to adenovirus transformation and that both tumorigenic (McBride and Wiener, 1964) and non-tumorigenic Ads (Freeman et al., 1967) induced morphological transformation from which immortal cell lines could be easily derived. Studies on immortal rodent cell lines showed that the cells transformed by non-oncogenic viruses could be tumorigenic, but only in immunocompromised hosts (Gallimore, 1972, Gallimore et al., 1977). The next logical step in adenovirus research was to identify the virus genes responsible for transformation and/or tumorigenicity. Unlike today, no restriction enzymes were readily available and there was no Southern blotting technique (Southern, 1975). Fujinaga and Green used membrane hybridization to show that adenovirus sequences were indeed retained and expressed in Ad transformed cells (Fujinaga and Green, 1970). This was followed by the ®rst attempt to utilize restriction endonucleases and Cot analysis (Pettersson and Sambrook, 1973) to map the viral sequences in an Ad2 transformed line, Ad2/8617. In the following year, a more extensive study found that a minimal region of the virus genome representing the left hand 14% was common to nine independently isolated Ad2 transformed rat embryo cell lines (Gallimore et al., 1974). This strongly suggested, but did not prove, that the Ad transforming gene resided in this region of the Ad genome. With the development of the calcium phosphate DNA transfection technique (Graham and van der Eb, 1973) and the use of de®ned Ad DNA fragments, it was convincingly shown that this region was responsible for the induction of transformation (Graham et al., 1974a,b). RNA mapping was carried out (Sharp et al., 1974; Flint et al., 1975) which de®ned the early region transcripts (reviewed by Shenk and Flint, 1991; Shenk, 1996) and later designated E1 (E1A+E1B) E2, E3 and E4 (Kitchingman et al., 1977). Viral RNA selected on speci®c Ad DNA fragments was then used in transcription/translation assays to identify the proteins expressed from these regions and their molecular weights (Lewis et al., 1976; Harter and Lewis, 1978; Halbert et al., 1979). The E1A proteins were found to have a molecular weight range from approximately 28 to 58 kDa. The next major step in our understanding of E1A was facilitated by the isolation of the now famous 293 cells, human embryo kidney cells transformed by Ad5 DNA (Graham et al., 1977). These cells expressed the Ad5 E1 transforming proteins that provided a permissive environment for replicationand transformation-defective E1 region mutants (Frost and Williams, 1978; Graham et al., 1978). Similar cell lines were developed for the isolation of Ad12 mutants (Byrd et al., 1982). The E1 regions of Ad2/5 and Ad12 Oncogene (2001) 20, 7824 ± 7835 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01

Detection of the large external transformation-sensitive protein on some epithelial cells.

15.00

Identification of the human papilloma virus-1a E4 gene products.

Abstract The large, external, transformation-sensitive (LETS) protein is detected on the surface of epithelial cells. Whereas fibroblasts build a massive network of fibrillar LETS protein in one-week-old confluent culture, epithelial cells do not. This observation may be useful for distinguishing epithelial cells from fibroblasts. When rat liver epithelial cells are continuously passaged in culture, surface LETS protein undergoes a significant alteration first from fibrillar organization to patch-like, then to complete absence.

In vitro traits of adenovirus-transformed cell lines and their relevance to tumorigenicity in nude mice

Antibodies prepared against a human papilloma virus‐1 (HPV‐1) E4/beta‐galactosidase fusion protein identified several polypeptides in HPV‐1, but not HPV‐2 or 4, induced papillomas. The major E4 protein, that represented up to 30% of total cellular protein, was a 16/17‐K doublet which was purified by column chromatography and analysed for amino acid content. A peptide derived by chymotryptic digestion was purified by h.p.l.c. and subjected to amino acid sequencing. The unique sequence obtained, Gly‐His‐Pro‐Asp‐Leu‐Ser‐Leu, identified the 16/17‐K doublet as a product of the HPV‐1 E4 gene region. Antibodies to both the E4/beta‐galactosidase fusion protein and the 16/17‐K doublet identified two smaller polypeptides (10/11‐K) which may represent spliced products of E4. We propose that the products of the HPV‐1 E4 gene region are not classical DNA tumor virus early proteins and suggest that they play a role in virus maturation.

Comparison of viral RNA sequences in adenovirus 2-transformed and lytically infected cells☆

Six independently isolated adenovirus 2-transformed rat cell lines and one adenovirus 5-transformed human cell line have been examined in vitro for serum growth requirements, saturation density, anchorage-independent growth, proteolytic enzyme activity and the presence of LETS glycoprotein and T antigen. This series of adenovirus-transformed cell lines exhibits an oncogeni spectrum ranging from being tumorigenic in immunocompetent rats through to nontumorigenic in adult nude mice. The relevance of the in vitro findings to growth potential in vivo is discussed.

Viral DNA sequences in cells transformed by simian virus 40, adenovirus type 2 and adenovirus type 5.

The complementary strands of fragments of 32P-labelled adenovirus 2 DNA generated by cleavage with restriction endonucleases EcoRI or Hpa1 were separated by electrophoresis. Saturation hybridization reactions were performed between these fragment strands and unlabelled RNA extracted from the cytoplasm of adenovirus 2-transformed rat embryo cells or from human cells early after adenovirus 2 infection. The fraction of each fragment strand complementary to RNA from these sources was measured by chromatography on hydroxylapatite. Maps of the viral DNA sequences complementary to messenger RNA in different lines of transformed cells and early during lytic infection of human cells were constructed. Five lines of adenovirus 2-transformed cells were examined. All contained the same RNA sequences, complementary to about 10% of the light strand of EcoRI fragment A. DNA sequences coding for this RNA were more precisely located using Hpa1 fragments E and C and mapped at the left-hand end of the genome. Thus any viral function expressed in all adenovirus 2-transformed cells, tumour antigen, for example, must be coded by this region of the viral genome. Two lines, F17 and F18, express only these sequences; two others, 8617 and REM, also contain mRNA complementary to about 7% of the heavy strand of the right-hand end of adenovirus 2 DNA; a fifth line, T2C4, contains these and many additional viral RNA sequences in its cytoplasm. The viral RNA sequences found in all lines of transformed cells are also present in the cytoplasm of human cells during the early phase of a lytic adenovirus infection. The additional cytoplasmic sequences in the 8617 and REM cell lines also correspond to “early” RNA sequences.

The APC/C and CBP/p300 cooperate to regulate transcription and cell-cycle progression

Until recently, the study of DNA tumor viruses has been an essentially syntactic subject. Those who have worked in the field commonly believe, for example, that there are interconnections between the expression of integrated viral genomes, the structure of cell surfaces and the growth properties of cells. On the whole, however, the anastomosis of various features has worked out well: it has resulted in a groundplan on which the apparently diverse elevations of a fast-growing field can quickly be sited with respect to one another.

Regulation of the 26S proteasome by adenovirus E1A

The anaphase-promoting complex/cyclosome (APC/C) is a multicomponent E3 ubiquitin ligase that, by targeting protein substrates for 26S proteasome-mediated degradation through ubiquitination, coordinates the temporal progression of eukaryotic cells through mitosis and the subsequent G1 phase of the cell cycle. Other functions of the APC/C are, however, less well defined. Here we show that two APC/C components, APC5 and APC7, interact directly with the coactivators CBP and p300 through protein–protein interaction domains that are evolutionarily conserved in adenovirus E1A. This interaction stimulates intrinsic CBP/p300 acetyltransferase activity and potentiates CBP/p300-dependent transcription. We also show that APC5 and APC7 suppress E1A-mediated transformation in a CBP/p300-dependent manner, indicating that these components of the APC/C may be targeted during cellular transformation. Furthermore, we establish that CBP is required in APC/C function; specifically, gene ablation of CBP by RNA-mediated interference markedly reduces the E3 ubiquitin ligase activity of the APC/C and the progression of cells through mitosis. Taken together, our results define discrete roles for the APC/C–CBP/p300 complexes in growth regulation.

Modulation of morphological differentiation of human neuroepithelial cells by serine proteases: independence from blood coagulation.

We have identified the N‐terminus of adenovirus early region 1A (AdE1A) as a region that can regulate the 26S proteasome. Specifically, in vitro and in vivo co‐precipitation studies have revealed that the 19S regulatory components of the proteasome, Sug1 (S8) and S4, bind through amino acids (aa) 4–25 of Ad5 E1A. In vivo expression of wild‐type (wt) AdE1A, in contrast to the N‐terminal AdE1A mutant that does not bind the proteasome, reduces ATPase activity associated with anti‐S4 immunoprecipitates relative to mock‐infected cells. This reduction in ATPase activity correlates positively with the ability of wt AdE1A, but not the N‐terminal deletion mutant, to significantly reduce the ability of HPV16 E6 to target p53 for ubiquitin‐mediated proteasomal degradation. AdE1A/proteasomal complexes are present in both the cytoplasm and the nucleus, suggesting that AdE1A interferes with both nuclear and cytoplasmic proteasomal degradation. We have also demonstrated that wt AdE1A and the N‐terminal AdE1A deletion mutant are substrates for proteasomal‐mediated degradation. AdE1A degradation is not, however, mediated through ubiquitylation, but is regulated through phosphorylation of residues within a C‐terminal PEST region (aa 224–238).

T-cell responses to human papillomavirus type 16 among women with different grades of cervical neoplasia

We have previously shown that a serum protein, termed differentiation reversal factor (DRF), is responsible for neurite retraction in differentiated cultures of an adenovirus 12 (Ad12) transformed human retinoblast cell line. Data is presented here to show that DRF is identical to the serine protease prothrombin. Both proteins have been immunoprecipitated using an antibody raised against purified prothrombin and have been shown to hydrolyse a specific thrombin substrate only after activation by the snake venom ecarin. Following addition to Ad12 HER 10 cells, which had previously been differentiated by culture in the presence of 2 mM dibutyryl cAMP in serum‐free medium, thrombin and prothrombin caused half‐maximal retraction of neurites at concentrations of 0.5 ng/ml and 20 ng/ml respectively. Interestingly, activation of prothrombin was shown to be unnecessary for biological activity. Using the inhibitor di‐isopropylfluorophosphate (DIP), we have shown that abrogation of the proteolytic activity of thrombin also results in a loss (greater than 2000 fold) of differentiation reversal activity. Thrombin and its zymogen both stimulated the mitosis of differentiated Ad12 HER 10 cells to a similar extent. In addition, differentiation reversal was highly specific since, at physiologically significant concentrations, closely related serine proteases did not cause neurite retraction. Prothrombin and thrombin also reversed morphological differentiation in the SK‐N‐SH neuroblastoma cell line and in heterogeneous cultures of cells from various regions in the human foetal brain.