E1A Transformation Research Articles

The Cellular Protein Complex Associated with a Transforming Region of E1A Contains c-MYC.

The cell-transforming activity of human adenovirus 5 (hAd5) E1A is mediated by the N-terminal half of E1A, which interacts with three different major cellular protein complexes, p300/CBP, TRRAP/p400, and pRb family members. Among these protein interactions, the interaction of pRb family proteins with conserved region 2 (CR2) of E1A is known to promote cell proliferation by deregulating the activities of E2F family transcription factors. The functional consequences of interaction with the other two protein complexes in regulating the transforming activity of E1A are not well defined. Here, we report that the E1A N-terminal region also interacted with the cellular proto-oncoprotein c-MYC and the homolog of enhancer of yellow 2 (ENY2). Our results suggested that these proteins interacted with an essential E1A transforming domain spanning amino acid residues 26 to 35 which also interacted with TRRAP and p400. Small interfering RNA (siRNA)-mediated depletion of TRRAP reduced c-MYC interaction with E1A, while p400 depletion did not. In contrast, depletion of TRRAP enhanced ENY2 interaction with E1A, suggesting that ENY2 and TRRAP may interact with E1A in a competitive manner. The same E1A region additionally interacted with the constituents of a deubiquitinase complex consisting of USP22, ATXN7, and ATXN7L3 via TRRAP. Acute short hairpin RNA (shRNA)-mediated depletion of c-MYC reduced the E1A transforming activity, while depletion of ENY2 and MAX did not. These results suggested that the association of c-MYC with E1A may, at least partially, play a role in the E1A transformation activity, independently of MAX. The transforming region of adenovirus E1A consists of three short modules which complex with different cellular protein complexes. The mechanism by which one of the transforming modules, CR2, promotes cell proliferation, through inactivating the activities of the pRb family proteins, is better understood than the activities of the other domains. Our analysis of the E1A proteome revealed the presence of the proto-oncoprotein c-MYC and of ENY2. We mapped these interactions to a critical transforming module of E1A that was previously known to interact with the scaffolding molecule TRRAP and the E1A-binding protein p400. We showed that c-MYC interacted with E1A through TRRAP, while ENY2 interacted with it independently. The data reported here indicated that depletion of c-MYC in normal human cells reduced the transforming activity of E1A. Our result raises a novel paradigm in oncogenic transformation by a DNA viral oncogene, the E1A gene, that may exploit the activity of a cellular oncogene, the c-MYC gene, in addition to inactivation of the tumor suppressors, such as pRb.

Abstract C297: PARP1 and BIN1 cooperate to act as a novel co-repressor for E2F1.

Abstract Background: Bridging integrator 1 (BIN1) is a transcriptional co-repressor and inhibits oncogenic activities of c-MYC and E1A oncoproteins1. BIN1 inhibits c-MYC through direct interaction, whereas the mechanism by which BIN1 attenuates the oncogenic transformation mediated by adenovirus E1A remains elusive. Since BIN1 does not physically interact with E1A, we hypothesized that BIN1 diminishes a cellular effector, which can be activated by E1A and is essential for E1A transformation. One of the well-recognized cellular effectors for E1A transformation is the E2F1 transcription factor. Although retinoblastoma (RB) protein is an authentic E2F1 inhibitor, loss of RB does not immediately result in cancer development, suggesting that an anti-E2F1 function, which is independent of RB, compensates for RB deficiency. BIN1 was recently shown to interact with poly (ADP-ribose) polymerase1 (PARP1), a component of transcriptional complex2. Given that PARP1 also interacts with E2F1, we hypothesized functional cross talk between BIN1 and PARP1 to curb E2F1 activity. Experimental Procedures: E2F1 activity was determined using an adenovirus-driven E2F1-sensitive luciferase reporter vector, E2A-Luc. Protein-protein interaction was studied by co-immunoprecipitation followed by Western blot analysis. Chromatin immunoprecipitation assay was used to demonstrate protein-DNA interaction. Colony formation and foci formation assays were performed to investigate BIN1-mediated tumor suppression in the presence and absence of PARP1. Results: We discovered that BIN1 physically interacts with E2F1 and inhibits its transactivation. Interestingly, depletion of PARP1 released endogenous E2F1 activity, regardless of the status of RB expression. Furthermore, BIN1 failed to inhibit E2F1 activity in the absence of PARP1, but RB does not need PARP1 to inhibit E2F1 activity. Chromatin immunoprecipitation assays suggested that BIN1 and PARP1 co-existed on an E2F-responsive promoter. Moreover, in the absence of PARP1, the binding affinities of both BIN1 and E2F1 to the promoter were significantly reduced. In addition, only in the presence of PARP1, ectopically expressed BIN1 inhibited tumor colony formation and HPV16 E7 oncoprotein-dependent cellular transformation. Our results suggest that BIN1 and PARP1 cooperate to inhibit E2F1 activity and suppress oncogenic transformation. Conclusion: We conclude that BIN1 is a novel E2F1 corepressor in cooperation with PARP1. BIN1-PARP1 interaction may serve as a safety device to control deregulated E2F1 activity due to RB loss. This study was supported by NIH R01CA140379 (to D.S.). Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C297. Citation Format: Alpana Kumari, Slovénie Pyndiah, Erica K. Cassimere, Daitoku Sakamuro. PARP1 and BIN1 cooperate to act as a novel co-repressor for E2F1. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C297.

Increased OXPHOS activity precedes rise in glycolytic rate in H-RasV12/E1A transformed fibroblasts that develop a Warburg phenotype

BackgroundThe Warburg phenotype in cancer cells has been long recognized, but there is still limited insight in the consecutive metabolic alterations that characterize its establishment. We obtained better understanding of the coupling between metabolism and malignant transformation by studying mouse embryonic fibroblast-derived cells with loss-of-senescence or H-RasV12/E1A-transformed phenotypes at different stages of oncogenic progression.ResultsSpontaneous immortalization or induction of senescence-bypass had only marginal effects on metabolic profiles and viability. In contrast, H-RasV12/E1A transformation initially caused a steep increase in oxygen consumption and superoxide production, accompanied by massive cell death. During prolonged culture in vitro, cell growth rate increased gradually, along with tumor forming potential in in vitro anchorage-independent growth assays and in vivo tumor formation assays in immuno-deficient mice. Notably, glucose-to-lactic acid flux increased with passage number, while cellular oxygen consumption decreased. This conversion in metabolic properties was associated with a change in mitochondrial NAD+/NADH redox, indicative of decreased mitochondrial tricarboxic acid cycle and OXPHOS activity.ConclusionThe high rate of oxidative metabolism in newly transformed cells is in marked contrast with the high glycolytic rate in cells in the later tumor stage. In our experimental system, with cells growing under ambient oxygen conditions in nutrient-rich media, the shift towards this Warburg phenotype occurred as a step-wise adaptation process associated with augmented tumorigenic capacity and improved survival characteristics of the transformed cells. We hypothesize that early-transformed cells, which potentially serve as founders for new tumor masses may escape therapies aimed at metabolic inhibition of tumors with a fully developed Warburg phenotype.

Adenovirus E1A oncoprotein liberates c‐Myc activity to promote cell proliferation through abating Bin1 expression via an Rb/E2F1‐dependent mechanism

Adenovirus E1A oncogene transforms primary rodent fibroblasts in cooperation with activated Ras. Conversely, the c-Myc oncoprotein-binding tumor suppressor, Bin1, inhibits Ras/E1A-mediated cell transformation. Since E1A does not directly bind to and inhibit Bin1, the primary mechanism by which E1A counteracts Bin1 to liberate oncogenic c-Myc activity is poorly understood. Here we show that wild-type E1A, but not an Rb binding-defective E1A mutant, suppresses endogenous Bin1 expression in cultured rodent fibroblasts. Similarly, other anti-Rb agents, such as human papillomavirus E7, mitogenic stimuli, and small interfering RNA (siRNA) for Rb, consistently decrease Bin1 promoter activity. In contrast, serum starvation, which activates Rb, enhances endogenous Bin1 levels. These findings suggest that Bin1 may be a novel component of Rb-mediated G1 arrest. Consistent with this premise, chromatin immunoprecipitation assays demonstrate that Rb protein directly interacts with the Bin1 promoter only upon removal of serum. Furthermore, ectopically expressed E2F1, which is primarily inhibited by Rb under serum-starved condition, represses Bin1 promoter activity in a manner that is dependent on the DNA-binding and transactivation domains of E2F1. Lastly, depletion of endogenous Bin1 per se is biologically meaningful since antisense or siRNA of Bin1 transfection releases endogenous c-Myc transcriptional activity and, concomitantly, accelerates cell proliferation. Our results suggest that Bin1 gene suppression caused by oncogenic E1A via Rb inactivation is an essential step in cell cycle progression promoted by c-Myc, and subsequently, E1A transformation. We propose a novel G1 arrest signaling mechanism by which Rb indirectly curbs oncogenic c-Myc activity via sustaining Bin1 expression.

Nutlin-3 Protects Kidney Cells during Cisplatin Therapy by Suppressing Bax/Bak Activation

Nutlins, the newly developed small molecule antagonists of MDM2, activate p53 and induce apoptosis in cancer cells, offering a novel strategy of chemotherapy. Recent studies have further suggested synergistic effects of nutlins with other chemotherapeutic drugs. However, it is unclear whether nutlins increase or decrease the side effects of these drugs in normal non-malignant cells or tissues. Cisplatin is a widely used chemotherapy drug, which has a major side effect of kidney injury. Here we show that Nutlin-3 protected kidney cells against cisplatin-induced apoptosis. The cytoprotective effects of Nutlin-3 were not related to its regulation of p53 or consequent gene expression during cisplatin treatment. Moreover, the protective effects were shown in MDM2-, MDM4-, or p53-deficient cells. On the other hand, Nutlin-3 suppressed mitochondrial events of apoptosis during cisplatin incubation, including Bax activation and cytochrome c release. Nutlin-3 attenuated cisplatin-induced oligomerization of Bax and Bak but not their interactions with Bcl-XL. In isolated mitochondria, Nutlin-3 inhibited cytochrome c release induced by Ca2+, Bim peptide, and recombinant tBid. Importantly, it blocked both Bax and Bak oligomerization under these conditions. Together, the results have uncovered a new pharmacological function of nutlins, i.e. suppression of Bax and Bak, two critical mediators of apoptosis.

The p400 Complex Is an Essential E1A Transformation Target

Here, we report the identification of a new E1A binding protein complex that is essential for E1A-mediated transformation. Its core component is a SWI2/SNF2-related, 400 kDa protein (p400). Other components include the myc- and p/CAF-associated cofactor, TRRAP/PAF400, the DNA helicases TAP54α/β, actin-like proteins, and the human homolog of the Drosophila Enhancer of Polycomb protein. An E1A mutant, defective in p400 binding, is also defective in transformation. Certain p400 fragments partially rescued this phenotype, underscoring the role of E1A-p400 complex formation in the E1A transforming process. Furthermore, E1A and c- myc each alter the subunit composition of p400 complexes, implying that physiological p400 complex formation contributes to transformation suppression.

The CXXC Zn binding motifs of the human papillomavirus type 16 E7 oncoprotein are not required for its in vitro transforming activity in rodent cells.

The conserved region 3 (CR3) of the E7 protein of human papillomaviruses contains two CXXC motifs involved in zinc binding and in the homodimerization of the molecule. Studies have suggested that the intact CXXC motifs in the CR3 of HPV16 and HPV18 E7 are required for the in vitro transforming activity of these proteins. CR3 also contains a low affinity pRb binding site and is involved in the disruption of the E2F/Rb1 complex. E7 is structurally and functionally related to Adenovirus E1A protein, which also has two CXXC motifs in CR3. However, the Ad E1A transforming activity appears to be independent of the presence of such domains. In fact, this viral protein exists in vivo as two different forms of 289 and 243 amino acids. The shorter Ad E1A form (Ad E1A243), where both CXXC motifs are deleted by internal splicing, retains its in vitro transforming activity. We have investigated if the HPV16 E7 CR3 can be functionally replaced by the Ad E1A243 CR3, which lacks both CXXC motifs. A chimeric protein (E7/E1A243) containing the CR1 and CR2 of HPV16 E7 fused to the CR3 of Ad E1A 243 was constructed. The E7/E1A243 while not able to homodimerize in the S. cerevisiae two-hybrid system retains several of the properties of the parental proteins, HPV16 E7 and Ad E1A. It associates with the 'pocket' proteins, induces growth in soft agar of NIH3T3 cells and immortalizes rat embryo fibroblasts. These data suggest that the CXXC motifs in CR3 of E7 do not play a direct role in the transforming properties of this viral protein but probably are important for maintaining the correct protein configuration.

An essential domain of the c-myc protein interacts with a nuclear factor that is also required for E1A-mediated transformation.

Cell transformation by nuclear oncogenes such as c-myc presumably involves the transcriptional activation of a set of target genes that participate in the control of cell division. The function of a small evolutionarily conserved domain of the c-myc gene encompassing amino acids 129 to 145 was analyzed to explore the relationship between cell transformation and transcriptional activation. Deletion of this domain inactivated the c-myc oncogene for cell transformation while retaining the ability to activate transcription of either myc consensus binding sites or a GAL4-dependent promoter when the c-myc N-terminus was fused to the GAL4 DNA-binding domain. Point mutations that altered a conserved tryptophan (amino acid 136) within this domain had similar effects. Expression of the wt c-Myc N terminus (amino acids 1 to 262) as a GAL4 fusion was a dominant inhibitor of cell transformation by the c-myc oncogene, and this same domain also inhibited transformation by the adenovirus E1A gene. Surprisingly, deletion of amino acids 129 to 145 eliminated the dominant negative activity of GAL4-Myc on both c-myc and E1A transformation. Expression of the GAL4-Myc protein in Cos cells led to the formation of a specific complex between the Myc N terminus and a nuclear factor, and this complex was absent with the dl129-145 mutant. These results suggest that an essential domain of the c-Myc protein interacts with a specific nuclear factor that is also required for E1A transformation.

Counteracting Effects of E1a Transformation on cAMP Growth Inhibition

Several signaling molecules have been identified which act as inhibitors of epithelial cell growth. The mechanisms for this negative growth regulation are still poorly understood. In the case of TGF-β, inhibition of keratinocyte cell growth can be totally prevented by transformation with an intact early region 1a (E1a) oncogene. We show here that E1a-transformed keratinocytes become also partially resistant to growth inhibition by elevated 3′,5′-cyclic adenosine monophosphate (cAMP) levels, as induced by treatment with forskolin, dibutyryl-cAMP, 8Br-cAMP, or 8C1-cAMP. Resistance to cAMP is due to interference of E1a with signaling pathways downstream of protein kinase A (PKA) activation, as intracellular cAMP levels and PKA activity were found to be similar in control and E1a-transformed cells. Induction of c-fos expression by 8Br-cAMP occurs at the same time in both cell lines. Interestingly however, this effect is maintained longer in the case of E1a-transformed cells compared to the control. A truncated E1a mutant which is still able to bind to the p105-Rb gene product, p107, and p60/cyclin A, induces cAMP resistance at levels which are only slightly lower than those induced by an intact E1a oncogene. In contrast, an E1a mutant which binds only to a p300 cellular protein and induces a substantial level of TGF-β resistance fails to induce cAMP resistance. Thus, E1a transformation counteracts the growth-inhibitory effects of cAMP as well as TGF-β, but to a different degree and through an only partially overlapping mechanism.

Genetic analysis of carcinogen enhancement of type 5 adenovirus transformation of cloned Fischer rat embryo fibroblast cells.

Pretreatment of CREF cells with methyl methanesulfonate (MMS) before infection with the host-range cold-sensitive type 5 adenovirus (Ad5) mutant H5hr1 results in a dose-dependent carcinogen enhancement of viral transformation (CET). The properties of CET observed with H5hr1, which include both an MMS dose-dependent enhancement in the number of transformed foci and an increase in transformation frequency after correction for cell toxicity, are not observed in carcinogen-pretreated wild-type (wt) Ad5 (H5wt)-infected CREF cells. This study was conducted to determine the role of the viral E1A and E1B transforming genes of H5hr1 in mediating the unique CET phenotype of H5hr1. Coinfection of MMS-pretreated CREF cells with H5wt or H5sub309 (which displays a wt Ad5 phenotype) and H5hr1 resulted in a suppression of the unique CET phenotype that was directly related to the multiplicity of infection with wt Ad5. Suppression of the unique H5hr1 CET phenotype was also apparent in MMS-pretreated CREF cells coinfected with H5hr1 and an Ad5 mutant expressing either a wt 13S E1A-encoded 289 amino-acid (aa) protein and an intact wt E1B gene or a wt 13S E1A-encoded 289-aa protein and a 22S E1B-encoded 495-aa protein. In contrast, the unique H5hr1 CET phenotype was not suppressed in MMS-pretreated CREF cells coinfected with H5hr1 and Ad5 or Ad2 mutants expressing either a wt 12S E1A-encoded 243-aa protein and both wt E1B gene products or an intact wt E1A gene and a wt E1B 13S-encoded 175-aa protein. That genetic changes in both the E1A and E1B viral regions of H5hr1 were required to induce the unique CET phenotype was also indicated by the inability of a recombinant Ad5 containing the 0-4.5 map-unit region of H5hr1 and the 4.5-100 map-unit region of H5sub309 to display the H5hr1 unique CET phenotype. Direct confirmation of the requirement for both gene regions of H5hr1 to mediate its unique CET was obtained by generating CREF cells stably expressing a wt Ad5 E1A 13S-encoded 289-aa protein and a wt E1B 22S-encoded 495-aa protein. In these CREF transformants (which displayed a CREF-like morphology), transformation by H5hr1 was not reduced, but the unique CET phenotype after MMS pretreatment was eliminated. These results suggest that alterations in both the 13S-encoded E1A and 22S-encoded E1B gene products of H5hr1 contribute to its unique CET.

Escape from transforming growth factor beta control and oncogene cooperation in skin tumor development.

Control of tumor development by surrounding normal cells has been suggested by a number of in vitro studies. In vivo, tumorigenicity of ras-transformed primary keratinocytes can be suppressed by addition of normal dermal fibroblasts. Here, we report that dermal fibroblasts produce a diffusible inhibitory factor belonging to the transforming growth factor beta (TGF-beta) family and possibly corresponding to TGF-beta 3. This factor can suppress growth of ras-transformed primary keratinocytes in culture and after injection into mice. As with primary cells, tumorigenicity of a ras-transformed, TGF-beta-sensitive keratinocyte line is substantially inhibited by adding dermal fibroblasts, leading to the formation of much smaller and differentiated tumors. Introduction of an intact E1a oncogene into these cells induces concomitant resistance to TGF-beta, to the effect of dermal-fibroblast inhibitory factor, and to dermal-fibroblast tumor suppression. Similar results are obtained with a transformation-deficient truncated E1a mutant, which binds to a reduced subset of cellular proteins (including the retinoblastoma gene product). Thus, genetic events such as those elicited by E1a transformation enable keratinocytes to escape from the inhibitory influences of a normal cellular environment and lead, together with ras transformation, to skin tumor development.

Requirement of the adenovirus E1A transformation domain 1 for inhibition of PC12 cell neuronal differentiation.

Expression of the adenovirus early gene E1A inhibits the nerve growth factor (NGF)-induced differentiation of PC12 pheochromocytoma cells. Expression of the 12S form of E1A, which lacks the transcription activation region, also inhibited PC12 cell differentiation in a manner similar to the wild-type gene. Three cellular proteins--the retinoblastoma susceptibility gene product referred to as 105(Rb)-, 107-, and 300-kDa proteins--stably interacted with the different E1A polypeptides. Analysis of the association of these cellular proteins with mutant E1A polypeptides demonstrated that a functional domain 1, which is minimally involved in the association of the 300-kDa protein with E1A, was sufficient to inhibit neuronal differentiation. Deletion of transformation domain 2, which encodes sequences necessary for the binding of the 105(Rb)- and 107-kDa proteins, did not influence the ability of the mutant E1A polypeptide to inhibit PC12 cell differentiation. E1A was also shown to alter the expression of mRNAs for the early response genes c-fos, c-myc, egr-1, and c-jun and their regulation in response to NGF. In clones expressing either 12S or 13S E1A, NGF stimulation of c-fos and c-myc was repressed. In contrast, basal mRNA levels for c-jun and egr-1 were constitutively elevated and not significantly affected further by challenge with NGF. Simply expressing c-jun by gene transfer, however, did not mimic the action of E1A because constitutively expressing c-jun clones differentiated in response to NGF. Thus, expression of the E1A polypeptide disrupts NGF control of early transcription events that have been shown to be critical for PC12 cell neuronal differentiation.

Isolation of Swiss 3T3 cell variants with altered heparan sulfate

The replica filter technique has been used to isolate variants of Swiss mouse 3T3 cells which produce heparan sulfates with altered levels of sulfation. These changes in the extent of sulfation correlate with alterations in cell morphology, in the organization of cytoskeletal elements, focal contacts, and the extracellular matrix, and in the growth regulation of cells, as expressed by saturation density. An increase in the extent of heparan sulfate sulfation occurs concomitantly with a decreased saturation density and enhanced focal contact formation. In contrast, graded decreases in sulfation correlate with graded increases in saturation density and losses of cytoskeletal and extracellular matrix organization. These graded responses appear very similar to those which have been reported for the transformation of cells with fusiform mutants of Rous sarcoma virus or the adenovirus type 2 E1a transforming gene and suggest that the morphological changes observed in the transformed cells can be controlled by cellular systems.

Adenovirus E1a ras cooperation activity is separate from its positive and negative transcription regulatory functions.

The E1a gene of adenovirus encodes two proteins, 289 and 243 amino acids long, which have positive (transactivator) and negative (enhancer repressor) RNA polymerase II transcriptional regulatory properties and cell transformation activities including cooperation with an activated ras gene. The E1a transforming functions more closely correlate with the repressor property than with transactivation in that both E1a proteins express the repressor and transformation functions while only the 289-amino-acid protein is an efficient transactivator. To understand whether the transcriptional regulatory activities of E1a are related to its ras cooperation activity, we generated a series of mutant E1a expression vectors by linker insertion mutagenesis of the 289-amino-acid protein. Here we describe a new class of mutants which although defective for enhancer repression still can cooperate with the ras oncogene in cell transformation. The mutants are also defective in transcription transactivation. Our data suggest that enhancer repression and transformation via ras cooperation are separate E1a functions and that cooperation with ras does not rely on either of the RNA polymerase II transcription regulatory functions of E1a. We also show that mutations which inactivate enhancer repression are not confirmed to a single critical domain necessary for repression. We therefore propose that the integrity of the overall configuration of the E1a proteins is important for the repression activity.

Adenovirus E1a ras cooperation activity is separate from its positive and negative transcription regulatory functions.

The E1a gene of adenovirus encodes two proteins, 289 and 243 amino acids long, which have positive (transactivator) and negative (enhancer repressor) RNA polymerase II transcriptional regulatory properties and cell transformation activities including cooperation with an activated ras gene. The E1a transforming functions more closely correlate with the repressor property than with transactivation in that both E1a proteins express the repressor and transformation functions while only the 289-amino-acid protein is an efficient transactivator. To understand whether the transcriptional regulatory activities of E1a are related to its ras cooperation activity, we generated a series of mutant E1a expression vectors by linker insertion mutagenesis of the 289-amino-acid protein. Here we describe a new class of mutants which although defective for enhancer repression still can cooperate with the ras oncogene in cell transformation. The mutants are also defective in transcription transactivation. Our data suggest that enhancer repression and transformation via ras cooperation are separate E1a functions and that cooperation with ras does not rely on either of the RNA polymerase II transcription regulatory functions of E1a. We also show that mutations which inactivate enhancer repression are not confirmed to a single critical domain necessary for repression. We therefore propose that the integrity of the overall configuration of the E1a proteins is important for the repression activity.

Mutations in the E1a gene of type 5 adenovirus result in oncogenic transformation of Fischer rat embryo cells.

Transformation of a specific clone of Fischer rat embryo (CREF) cells with wild-type 5 adenovirus (Ad5) or the E1a plus E1b transforming gene regions of Ad5 results in epithelioid transformants that grow efficiently in agar but that do not induce tumors when inoculated into nude mice or syngeneic Fischer rats. In contrast, CREF cells transformed by a host-range Ad5 mutant, H5hrl, which contains a single base-pair deletion of nucleotide 1055 in E1a resulting in a 28-kd protein (calculated) in place of the wild-type 51-kd acidic protein, display a cold-sensitive transformation phenotype and an incomplete fibroblastic morphology but surprisingly do induce tumors in nude mice and syngeneic rats. Tumors develop in both types of animals following injection of CREF cells transformed by other cold-sensitive Ad5 E1a mutants (H5dl101 and H5in106), which contain alterations in their 13S mRNA and consequently truncated 289AA proteins. CREF cells transformed with only the E1a gene (0-4.5 m.u.) from H5hrl or H5dl101 also produce tumors in these animals. To directly determine the role of the 13S E1a encoded 289AA protein and the 12S E1a encoded 243AA protein in initiating an oncogenic phenotype in adenovirus-transformed CREF cells, we generated transformed cell lines following infection with the Ad2 mutant pm975, which synthesizes the 289AA E1a protein but not the 243AA protein, and the Ad5 mutant H5dl520 and the Ad2 mutant H2dl1500, which do not produce the 289AA E1a protein but synthesize the normal 243AA E1a protein. All three types of mutant adenovirus-transformed CREF cells induced tumors in nude mice and syngeneic rats. Tumor formation by these mutant adenovirus-transformed CREF cells was not associated with changes in the arrangement of integrated adenovirus DNA or in the expression of adenovirus early genes. These results indicate, therefore, that oncogenic transformation of CREF cells can occur in the presence of a wild-type 13S E1a protein or a wild-type 12S E1a protein when either protein is present alone, but does not occur when both wild-type E1a proteins are present.