Insulin gene enhancer activity is inhibited by adenovirus 5 E1a gene products

Disease Tolerance and Pathogen Resistance Genes May Underlie Trypanosoma cruzi Persistence and Differential Progression to Chagas Disease Cardiomyopathy

Chagas disease is caused by infection with the protozoan Trypanosoma cruzi and affects over 8 million people worldwide. In spite of a powerful innate and adaptive immune response in acute infection, the parasite evades eradication, leading to a chronic persistent infection with low parasitism. Chronically infected subjects display differential patterns of disease progression. While 30% develop chronic Chagas disease cardiomyopathy (CCC)—a severe inflammatory dilated cardiomyopathy—decades after infection, 60% of the patients remain disease-free, in the asymptomatic/indeterminate (ASY) form, and 10% develop gastrointestinal disease. Infection of genetically deficient mice provided a map of genes relevant for resistance to T. cruzi infection, leading to the identification of multiple genes linked to survival to infection. These include pathogen resistance genes (PRG) needed for intracellular parasite destruction, and genes involved in disease tolerance (protection against tissue damage and acute phase death—DTG). All identified DTGs were found to directly or indirectly inhibit IFN-γ production or Th1 differentiation. We hypothesize that the absolute need for DTG to control potentially lethal IFN-γ PRG activity leads to T. cruzi persistence and establishment of chronic infection. IFN-γ production is higher in CCC than ASY patients, and is the most highly expressed cytokine in CCC hearts. Key DTGs that downmodulate IFN-γ, like IL-10, and Ebi3/IL27p28, are higher in ASY patients. Polymorphisms in PRG and DTG are associated with differential disease progression. We thus hypothesize that ASY patients are disease tolerant, while an imbalance of DTG and IFN-γ PRG activity leads to the inflammatory heart damage of CCC.

Pancreas duodenum homeobox-1 transcriptional activation requires interactions with p300

The homeodomain transcription factor, pancreas duodenum homeobox (PDX)-1, is essential for pancreas development, insulin production, and glucose homeostasis. Mutations in pdx-1(ipf-1) are associated both with maturity-onset diabetes of the young and type 2 diabetes. PDX-1 interacts with multiple transcription factors and coregulators, including the coactivator p300, to activate the transcription of the insulin gene and other target genes within pancreatic beta-cells. In characterizing the protein-protein interactions of PDX-1 and p300, we identified mutations in PDX-1 that disrupt its function and are associated with increased or decreased interactions with p300. Several mutant PDX-1 proteins that are associated with heritable forms of diabetes in humans, in particular the mutant P63fsdelC, exhibited increased binding to a carboxy-terminal segment of p300 in the setting of decreased DNA-binding activities, suggesting that sequestration of p300 by mutant PDX-1 proteins may be an additional mechanism by which insulin gene expression is reduced in heterozygous carriers of pdx-1(ipf-1) mutations. The introduction of the point mutations S66A/Y68A in the highly conserved amino-terminal PDX-1 transactivation domain reduced the ability of PDX-1 to interact with p300, substantially diminished the transcriptional activation of PDX-1, and reduced the synergistic activation of glucose-responsive insulin promoter enhancer sequences by PDX-1, E12, and E47. We propose that interactions of PDX-1 with p300 are required for the transcriptional activation of PDX-1 target genes. Impairment of interactions between PDX-1 and p300 in pancreatic beta-cells may limit insulin production and lead to the development of diabetes.

Modulation of PI3K/Akt pathway by E1a mediates sensitivity to cisplatin

In order to investigate the molecular mechanisms implicated in the induction of chemo sensitivity by adenovirus E1a gene expression, we decided to investigate which signal transduction pathways could be affected by the E1a gene in Human Normal Fibroblast (IMR90). No effect was observed in SAPK pathways (p38MAPK and JNK), but E1a was able to affect the Akt activation mediated by insulin. This result was confirmed by transient transfection experiments performed in Cos-7 cells and also observed in other transformed cell lines such as A431. Furthermore, E1a expression induces a decrease in the basal status of Akt activity. Finally we demonstrated that E1a is able to block the Akt activation mediated by cisplatin and correlates with a sensitive phenotype. In summary, our data demonstrate that specific inhibition of the PI3K/Akt pathway mediates some of the biological properties of E1a such as induction of chemosensitivity.

E1A: tumor suppressor or oncogene? Preclinical and clinical investigations of E1A gene therapy

In the late 1980s, we have shown that the E1A gene can downregulate HER-2/neu overexpression, thus reversing the tumorigenic and metastatic phenotype. Further, E1A can function as a tumor suppressor gene by inducing apoptosis and inhibiting metastasis. At The University of Texas M. D. Anderson Cancer Center, we have been investigating the adenovirus type 5 E1A gene as a potential therapeutic gene in breast and ovarian cancer since 1995 by using cationic liposome as gene delivery system. In this chapter, we recount our development of E1A as a therapeutic gene.

Induction of susceptibility to tumor necrosis factor by E1A is dependent on binding to either p300 or p105-Rb and induction of DNA synthesis

The introduction of the adenovirus early region 1A (E1A) gene products into normal cells sensitizes these cells to the cytotoxic effects of tumor necrosis factor (TNF). Previous studies have shown that the region of E1A responsible for susceptibility is CR1, a conserved region within E1A which binds the cellular proteins p300 and p105-Rb at nonoverlapping sites. Binding of these and other cellular proteins by E1A results in the induction of E1A-associated activities such as transformation, immortalization, DNA synthesis, and apoptosis. To investigate the mechanism by which E1A induces susceptibility to TNF, the NIH 3T3 mouse fibroblast cell line was infected with viruses containing mutations within E1A which abrogate binding of some or all of the cellular proteins to E1A. The results show that TNF susceptibility is induced by E1A binding to either p300 or p105-Rb. E1A mutants that bind neither p300 nor p105-Rb do not induce susceptibility to TNF. Experiments with stable cell lines created by transfection with either wild-type or mutant E1A lead to these same conclusions. In addition, a correlation between induction of DNA synthesis and induction of TNF sensitivity is seen. Only viruses which induce DNA synthesis can induce TNF sensitivity. Those viruses which do not induce DNA synthesis also do not induce TNF sensitivity. These data suggest that the mechanisms underlying induction of susceptibility to TNF by E1A are intimately connected to E1A's capacity to override cell cycle controls.

Transcriptional repression by human adenovirus E1A N terminus/conserved domain 1 polypeptides in vivo and in vitro in the absence of protein synthesis

The human adenovirus E1A 243R protein (243 residues) transcriptionally represses a set of cellular genes that regulate cellular growth and differentiation. We describe two lines of evidence that E1A repression does not require cellular protein synthesis but instead involves direct interaction with a cellular protein(s). First, E1A 243R protein represses an E1A-repressible promoter in the presence of inhibitors of protein synthesis, as shown by cell microinjection-in situ hybridization. Second, E1A 243R protein strongly represses transcription in vitro from promoters of the E1A-repressible genes, human collagenase, and rat insulin type II. Repression in vitro is promoter-specific, and an E1A polypeptide containing only the N-terminal 80 residues is sufficient for strong repression both in vivo and in vitro. By use of a series of E1A 1-80 deletion proteins, the E1A repression function was found to require two E1A sequence elements, one within the nonconserved E1A N terminus, and the second within a portion of conserved region 1 (40-80). These domains have been reported to possess binding sites for several cellular transcription regulators, including p300, Dr1, YY1, and the TBP subunit of TFIID. The in vitro transcription-repression system described here provides a powerful tool for the further analysis of molecular mechanism and the possible role of these cellular factors.

Negative regulation of transcription in eukaryotes

Images

Repression of liver-specific hepatitis B virus enhancer 2 activity by adenovirus E1A proteins

Two regions of the hepatitis B virus (HBV) genome have been shown to display properties of a transcriptional enhancer. Enhancer 1 is active in most hepatoma lines examined as well as in some non-hepatocyte-derived cell lines. In contrast, enhancer 2 activity is strictly liver specific. In this study, we show that adenovirus E1A expression in the highly differentiated human hepatoma line Huh6 strongly inhibits HBV enhancer 2-stimulated transcription while having no effect on HBV enhancer 1 activity. A sequence motif in HBV enhancer 2 which is essential for its enhancer function is the target for E1A-mediated repression. The repression of HBV enhancer 2 activity is mediated through the N-terminal region of the E1A proteins known to bind a 300-kDa cellular protein. Our results suggest that HBV enhancer function may be modulated by a cellular mechanism similar to E1A-mediated transcriptional repression.

The adenovirus E1A 243R protein purified from Escherichia coli under nondenaturing conditions is found in association with dnaK

The adenovirus E1A 243R protein immortalizes primary cells in culture and induces part of the phenotypes required for transformation. It has also been shown to interact with a number of cellular polypeptides, including the product of the retinoblastoma gene. To understand more fully the molecular activities of the E1A 243R protein in association with these proteins as well as its role in the processes of cellular growth, we have developed a method for rapidly purifying this protein from genetically engineered Escherichia coli under nondenaturing conditions. The plasmid-encoded E1A protein, when expressed in a protease-deficient mutant, is found to have the same length and amino acid sequence as that which is produced in a mammalian cell. The procedure for purifying the E1A 243R protein from bacteria relies primarily upon immunoaffinity chromatography and the use of a peptide comprising the epitope recognized by an E1A-specific antibody. Elution of the E1A protein under this condition allows for gentle isolation and a purity that ranges from 90 to 96%. However, without the addition of micromolar amounts of ATP prior to its elution from the antibody column, the E1A protein is found in association with an E. coli protein of 70 kDa. Immunoblot analysis with a specific antibody showed that this bacterial protein was the heat shock protein dnaK, which is known to have extensive homology with the hsp-hsc70 family of proteins in mammalian cells. Recognition of E1A by the dnaK protein may very well reflect a situation that also occurs between the mammalian heat shock proteins and the E1A 243R protein after adenovirus infection.

Extinction of the human insulin gene expression in insulinoma x fibroblast somatic cell hybrids involves cis-acting DNA elements

Insulin gene expression in rat insulinoma (RIN) cells is extinct in RIN x fibroblast hybrids and can reappear upon loss of DNA contributed by the fibroblast parent. (Besnard et al., Exp. Cell Res. 185:101-108, 1989). In the present study, we looked for the role of 5'-flanking sequences of the human insulin gene in the negative control observed in the hybrids. RIN cells were transformed with composite genes which consisted of the coding sequence of the gpt gene placed under the control of 5'-flanking regions of the human insulin gene (Ins.gpt gene). Upon hybridization of these cells with mouse fibroblasts, the expression of both Ins.gpt and endogenous rat insulin genes were suppressed together. The results obtained indicate that cis-acting DNA elements are involved in the negative control of the gene. These elements are located in a fragment spread from -258 to +241 of the transcription origin of the human insulin gene.