Adenovirus type-12 tumor antigen. III. Tumorigenicity and immune response to syngeneic rat cells transformed with virions and isolated transforming fragment of adenovirus 12 DNA

Identification of genes differentially expressed as result of adenovirus type 5- and adenovirus type 12-transformation

BACKGROUND

Cells transformed by human adenoviruses (Ad) exhibit differential capacities to induce tumours in immunocompetent rodents; for example, Ad12-transformed rodent cells are oncogenic whereas Ad5-transformed cells are not. The E1A gene determines oncogenic phenotype, is a transcriptional regulator and dysregulates host cell gene expression, a key factor in both cellular transformation and oncogenesis. To reveal differences in gene expression between cells transformed with oncogenic and non-oncogenic adenoviruses we have performed comparative analysis of transcript profiles with the aim of identifying candidate genes involved in the process of neoplastic transformation.

RESULTS

Analysis of microarray data revealed that a total of 232 genes were differentially expressed in Ad12 E1- or Ad5 E1-transformed BRK cells compared to untransformed baby rat kidney (BRK) cells. Gene information was available for 193 transcripts and using gene ontology (GO) classifications and literature searches it was possible to assign known or suggested functions to 166 of these identified genes. A subset of differentially-expressed genes from the microarray was further examined by real-time PCR and Western blotting using BRK cells immortalised by Ad12 E1A or Ad5 E1A in addition to Ad12 E1- or Ad5 E1-transformed BRK cells. Up-regulation of RelA and significant dysregulation of collagen type I mRNA transcripts and proteins were found in Ad-transformed cells.

CONCLUSION

These results suggest that a complex web of cellular pathways become altered in Ad-transformed cells and that Ad E1A is sufficient for the observed dysregulation. Further work will focus on investigating which splice variant of Ad E1A is responsible for the observed dysregulation at the pathway level, and the mechanisms of E1A-mediated transcriptional regulation.

Adenovirus 12-mediated down-regulation of the major histocompatibility complex (MHC) class I promoter: identification of a negative regulatory element responsive to Ad12 E1A

In highly oncogenic adenovirus (Ad) 12-transformed cells, major histocompatibility complex (MHC) class I gene expression is down-regulated by the products of the viral E1A oncogene at the level of initiation of transcription. However, class I gene expression is unaltered or elevated in non-oncogenic Ad2- or Ad5-transformed cells. These changes in class I expression may permit Ad12-transformed cells to escape host immune surveillance and elicit tumour formation. Here we show that the 2kb of 5' flanking region of the mouse H-2Kb class I gene is sufficient to mediate down-regulation of transcription driven from homologous or heterologous (HSV thymidine kinase) basal promoter elements in cells expressing Ad12 E1A, but not in Ad2 E1A-expressing cells. Deletion analysis of the 2kb region showed that sequences from -1.18 to -1.44kb (relative to the cap site) were a target for Ad12 E1A-mediated transcriptional down-regulation. Deletion of this entire region from the 2kb flanking sequence of the H-2Kb gene abolished Ad12 E1A-mediated down-regulation of transcription. Computer analysis of the -1.18 to -1.44kb sequence identified two 6/7bp matches with the AP-1 transcription factor consensus sequence and two matches with the pig MHC class I PD1 repressor element. Gel retardation analysis using overlapping DNA fragments derived from the -1.18 to -1.44kb sequence revealed several DNA:protein complexes formed using nuclear extract derived from Ad12-, but not from Ad2- or Ad5-transformed cells. Some of these DNA:protein complexes were also present, but at lower levels, in nuclear extracts from untransformed rat cells suggesting the possible involvement of cellular factors in the mechanism of down-regulation mediated by Ad12 E1A. A binding site for the AP-1 factor failed to compete for protein binding to fragments within the -1.18 to -1.44 sequence, while the PD1 site competed for binding only in the -1.15 to -1.23 region. These results indicate that novel factors (as well as a previously identified class I repressor, PD1) may be involved in Ad12 E1A-mediated down-regulation of MHC class I transcription.

Tumorigenicity of adenovirus-transformed rodent cells is influenced by at least two regions of adenovirus type 12 early region 1A

Chimeric adenovirus type 5 (Ad5)/Ad12 early region 1A (E1A) genes were used to transform primary baby rat kidney cells in cooperation with Ad12 E1B, and the resulting cell lines were assayed for tumorigenicity in syngeneic rats. It was found that lines were nontumorigenic when transformed by hybrid E1A genes consisting of the amino-terminal 80 amino acids from Ad12 including conserved region 1 (CR1), with the remaining portion from Ad5. In contrast, cell lines transformed by hybrids containing Ad12 E1A sequences from the amino terminus to the leftmost border of CR3 or beyond were tumorigenic. To extend these results, sequences spanning CR2 and CR3 of Ad5 E1A were replaced with the homologous regions of Ad12 E1A and additional transformed cell lines were established. These lines were weakly-to-moderately tumorigenic, suggesting that Ad12 E1A sequences between CR2 and CR3 may be involved in tumorigenicity but are not the sole factors influencing it. Interestingly, examination of an E1A sequence alignment indicated that the region between CR2 and CR3 of Ad12 E1A is also conserved in the corresponding sequence of simian adenovirus type 7, which, like Ad12, is highly oncogenic. This region is characterized by the presence of a stretch of several alanine residues and is similar to a motif present in a number of proteins with transcriptional repression activity. The possibility that this region may influence tumorigenicity by means of a transcriptional regulatory mechanism is discussed.

Human adenovirus type 9-induced rat mammary tumors

Following subcutaneous inoculation of newborn Wistar-Furth rats with human adenovirus type 9 (Ad9), 16 of 16 female and 0 of 11 male rats developed mammary tumors. Tumor-positive animals usually developed tumors in multiple glands. Histopathological analyses indicated that three general categories of tumor could be identified. Mammary fibroadenomas were the most common tumor type encountered, but phyllodeslike tumors and solid sarcomas were also frequently found. In situ hybridization and immunohistochemical techniques established that benign fibroadenomas were derived from mammary fibroblasts (collagen type I- and vimentin-positive cells) and that malignant tumors were derived from myoepithelial cells (collagen type IV-, vimentin-, and muscle-specific actin-positive cells). The fact that mammary tumors were limited to female rats suggested that female hormones are essential for tumor growth and development. In this regard, ovariectomy of Ad9-infected female rats prevented tumor development, while subsequent diethylstilbestrol (DES) treatment elicited tumor formation. In addition, Ad9-infected and castrated male rats which received DES also developed mammary tumors. Established male mammary tumors regressed when DES treatment was stopped and reappeared after DES treatment was resumed. Together, these results indicate that estrogen is required for both initiation and maintenance of Ad9-induced mammary tumors. Southern blot analysis of high-molecular-weight tumor DNA showed that mammary tumor cells contained single or multiple integrated copies of the entire Ad9 genome. RNase protection experiments established that estrogen receptor as well as Ad9 E1a and E4 mRNAs were expressed in mammary tumors, but Ad9 E3 and, surprisingly, E1b mRNAs were not expressed at detectable levels.

Eradication of adenovirus E1-induced tumors by E1A-specific cytotoxic T lymphocytes

Cytotoxic T lymphocyte (CTL) clones against adenovirus type 5 (Ad5) early region 1 (E1)-transformed cells were generated in C57BL/6 (B6) mice. A defined peptide encoded by Ad5 E1A is the target structure for H-2Db-restricted CTLs. Upon intravenous injection into B6 nude mice bearing Ad5 E1-induced tumors, these CTLs, if combined with recombinant IL-2, destroy subcutaneous tumor masses up to 10 cm3. The in vivo action of CTLs is highly specific, and long-term "memory" persists in treated nude mice months after tumor regression. Our data show an important role for CTLs directed against a viral nuclear oncogene product in tumor eradication.

Nuclear proteins binding to an enhancer element of the major histocompatibility class I promoter: differences between highly oncogenic and nononcogenic adenovirus-transformed rat cells

Two major DNA-binding activities specific for the major histocompatibility (MHC) class I regulatory element (CRE) were detected in adenovirus (Ad)-transformed cells. One activity, term CRE1, had similar binding properties to a previously described positive-acting transcription factor specific for MHC class I genes termed H2TF1. The other activity, termed CRE2, bound to a region on the CRE separate from CRE1, and was present in Ad12, but not in Ad5-transformed cells. A CRE2-like activity was also present in non-adenovirus-transformed mouse L929 cells, indicating that CRE2 may be a cellular, rather than a viral, factor. The CRE2 activity did not correspond to any previously described transcription factor with a potential binding site in the CRE.

Expression of hamster MHC class I antigens in transformed cells and tumours induced by human adenoviruses

The steady-state levels of hamster MHC class I mRNA and cell surface protein were analysed in cells transformed by either adenovirus type 12 (Ad12) or types 2 or 5 (Ad2 or Ad5). All cell lines were oncogenic in new born and adult hamsters. A great reduction in both class I mRNA and protein was observed in Ad12 transformed cells compared to cells transformed by Ad2 or Ad5. Analysis of class I mRNA in solid tumours induced in hamsters by Ad transformed cell lines also showed greatly reduced mRNA levels in tumours induced by Ad12 compared to those induced by Ad2 or Ad5. This suggests that, in the Ad transformed hamster cell system, reduction in the level of MHC class I gene expression is not necessarily associated with tumour formation.

Region E1a of highly oncogenic adenovirus 12 in transformed cells protects against NK but not LAK cytolysis

The sensitivity of a library of adenovirus-transformed rat cell lines to lysis with highly enriched populations of rat NK cells and LAK cells activated in vitro by culture with recombinant human IL-2 was studied and correlated with the tumorigenic potential of these cell lines. The cell lines studied express the transforming E1 region of highly oncogenic Ad12 or nononcogenic Ad2. Two cell lines express recombinant E1A regions. In one the E1A genes were of Ad12 origin and the E1B region was derived from nononcogenic Ad5. In the other, the E1A region was from Ad5 and the E1B genes from Ad12. All cell lines tested which express the early region E1 of Ad12 are tumorigenic in syngeneic rats. The two cell lines which express only the E1A or the E1B genes of Ad12, and the Ad2-transformed cells did not induce tumors. Transformed cell lines which express the E1A region of nononcogenic Ad2 or Ad5 are efficiently killed by rat NK cells, but cells which express the Ad12 E1A genes are resistant to lysis by NK-enriched cell fractions even at high effector:target ratio; cells containing the Ad12 E1 region are also resistant to IFN-activated NK cells. Although such NK-resistant cells have a uniformly low level of class I MHC antigen, their resistance is not affected by MHC antigen level modulation by rat IFN. Ad12-transformed cells resistant to endogeneous NK cells, however, are efficiently lysed by LAK cells stimulated in vitro by recombinant IL-2. Sensitivity to LAK killing is unaffected by IFN treatment of target cells. These results show that expression of the E1A region of highly oncogenic Ad12 in the transformed cells, which confers resistance to endogeneous NK cells, fails to protect against lysis by LAK cells.

Adenovirus tumor-specific transplantation antigen is a function of the E1A early region

Viable recombinant adenoviruses that carry a portion of the type 12 E1A and E1B transcription units in a type 5 background were used to identify genes controlling expression of the adenovirus tumor-specific transplantation antigen (TSTA). The TSTA immunity is not crossreacting between the group A and group C adenovirus serotypes. Viruses carrying the E1A region (sub370-12E1A), or both E1A and E1B (sub370-12E1AB) regions of Ad12, induce a strong transplantation immunity against tumors induced by syngeneic cells transformed with adenovirus 12, but fail to induce any protection against syngeneic cells transformed with adenovirus 2. Immunization with the virus carrying only the E1B region (sub370-12E1B) of adenovirus 12 induces no immunity to adenovirus 12 transformed cell line, but confers a strong protection against cells transformed with adenovirus 2. These results provide strong evidence that the adenovirus tumor-specific transplantation antigen is a function of the E1A early region.

Tumorigenicity of adenovirus-transformed cells: region E1A of adenovirus 12 confers resistance to natural killer cells

Sensitivity of a library of rat cells transformed in vitro with viable recombinant adenoviruses to natural killer (NK) cells and allogeneic cytotoxic T cells has been studied and correlated with their oncogenic potential in syngeneic rats. All cell lines transformed with the sub370-12E1AB virus (containing E1A and E1B regions of Ad12) and with the sub370-12E1A virions (containing the E1A region of Ad12 and the E1B region of Ad5) showed a high degree of resistance to NK cells. The cell lines transformed with the sub370-12E1B virus (containing the E1A region of Ad5 and the E1B region of Ad12) were highly sensitive to NK cytotoxicity. While all cell lines transformed with virions containing the E1A genes of Ad5 expressed high levels of class I MHC antigen, only three of eight cell lines containing the E1A region of Ad12 showed detectable levels by flow cytometric analysis after staining with specific antibodies. All cell lines containing E1A genes of Ad5 were killed by in vitro generated allogeneic cytolytic T cells. Only three of eight cell lines containing the E1A region of Ad12 were killed by such CTLs; the level of cytotoxicity, however, did not reach that seen with the cells containing the E1A genes of Ad5. All cell lines containing the E1A and E1B genes of Ad12 were highly tumorigenic. Only two of four cell lines transformed with virus containing the E1A genes of Ad12 and E1B region of Ad5 were tumorigenic. The efficiency of tumor induction was low and the latent period was long confirming the importance of the E1B region. None of the cell lines transformed with virus containing the E1A region of Ad5 and the E1B genes of Ad12 were tumorigenic, reflecting their high degree of sensitivity to both natural and induced cellular immunity. Expression of the E1A region of Ad12 in transformed cells modulates not only the level of class I MHC antigens, but also confers resistance to NK cell cytotoxicity.

Deletion of the gene encoding the adenovirus 5 early region 1b 21,000-molecular-weight polypeptide leads to degradation of viral and host cell DNA

The adenovirus 5 mutant H5dl337 lacks 146 base pairs within early region 1B. The deletion removes a portion of the region encoding the E1B 21,000-molecular-weight (21K) polypeptide, but does not disturb the E1B-55K/17K coding region. The virus is slightly defective for growth in cultured HeLa cells, in which its final yield is reduced ca. 10-fold compared with wild-type virus. The mutant displays a striking phenotype in HeLa cells. The onset of cytopathic effect is dramatically accelerated, and both host cell and viral DNAs are extensively degraded late after infection. This defect has been described previously for a variety of adenovirus mutants and has been termed a cytocidal (cyt) phenotype. H5dl337 serves to map this defect to the loss of E1B-21K polypeptide function. In addition to its defect in the productive growth cycle, H5dl337 is unable to transform rat cells at normal efficiency.

Transformation by human adenoviruses

When, approximately 10 years ago, it was shown that the functions essential for cell transformation were localized in a small region of the adenovirus genome, a DNA segment which at that time was thought to be capable of encoding two or three average-sized proteins at most, it seemed reasonable to hope that an understanding of the mechanisms by which adenoviruses transform cells might be quickly achieved. While such optimism might be forgiven, it was quite clearly naive in the extreme. As a consequence of mRNA splicing and the use of overlapping reading frames the number of proteins encoded within E1 is 2-3-times greater than would have been predicted a decade ago, and post-translational modifications may add another dimension of complexity. In fact it has taken nearly all of the past decade just to identify the proteins encoded in E1 and to characterize them in the most rudimentary way. However, we have now entered a period in which new information is accumulating at an extremely rapid rate as a result of several major technical and fundamental advances. Chief among these are the use of recombinant DNA techniques, particularly site-directed mutagenesis, which combined with methods for introducing mutations made in cloned sequences back into infectious virus, clearly represents a powerful approach to studying the functions of transforming proteins. In addition, the ability to express transforming proteins in bacteria and to produce large amounts of highly purified proteins which previously were only just detectable in infected and transformed cells is a major breakthrough. Advances in immunological techniques, particularly the development of monoclonal antibodies and antisera against synthetic peptides, have enormously simplified the task of detecting and characterizing E1 proteins. Finally, recent results suggesting that adenovirus transforming proteins may be functionally and structurally similar to other oncogenes brings a new perspective to the study of oncogenic transformation. Have all the proteins involved in transformation by adenoviruses been identified? It seems probable that all those virally coded proteins which play a major role are now known but of course minor players in the cast could still be waiting in the wings. We have pointed out that viral functions encoded outside region E1 may have some importance at least in initiation of transformation by virions and have speculated on the possibility that one or more of these may be involved in the integration of viral DNA into the host cell chromosome.(ABSTRACT TRUNCATED AT 400 WORDS)

Adenovirus type 12 specific cell surface antigen in transformed cells is a product of the E1b early region

Six syngeneic rat cell lines transformed with isolated or cloned left end fragments of adenovirus type 12 (Ad12) DNA were used in a study of the Ad12-specific cell surface antigen. Cells transformed with EcoRI-C, SalI-C, and HindIII-G fragments of Ad12 DNA fragment-transformed express the E1a and a part of or a complete E1b early regions. Two AccI-H DNA fragment-transformed cell lines contain and express only the E1a region. All these cells contain nuclear Ad12 antigen. Cytotoxic antibodies raised against syngeneic EcoRI-C DNA fragment-transformed cells kill EcoRI-C, SalI-C and HindIII-G fragment-transformed cells, but fail to kill cells transformed with AccI-H DNA fragment. Immunofluorescence analysis shows that such antibodies, which stain the surface of cells expressing the E1b region, do not stain the surface of AccI-H DNA fragment-transformed cells. Cells transformed with AccI-H fragment are also not killed by secondary cytolytic T cells, raised and effective against cells transformed with EcoRI-C, SalI-C, and HindIII-G DNA fragments. Cells transformed with AccI-H fragment do not elicit cytolytic T cells against any of the studied cell lines. The only Ad12-specific product shared by all cell lines killed in cytolytic assays which is absent from AccI-H fragment-transformed cells is the E1b 18K protein. Since it has also been shown in other studies that this protein is associated with the cellular membrane, the simplest interpretation of these data is that the Ad12-specific cell surface antigen in transformed rat cells is a product of the left end of the E1b region.