Extinction of insulin gene expression in hybrids between beta cells and fibroblasts is accompanied by loss of the putative beta-cell-specific transcription factor IEF1

Genome-wide analysis of hepatic gene silencing in mammalian cell hybrids

Silencing of tissue-specific gene expression in mammalian somatic cell hybrids is a well-documented epigenetic phenomenon which is both profound (involving a large number of genes) and enigmatic. Our aim was to utilize whole-genome microarray analyses to determine the true extent of gene silencing on a genomic level. By comparing gene expression profiles of hepatoma×fibroblast cell hybrids with those of parental cells, we have identified over 300 liver-enriched genes that are repressed at least 5-fold in the cell hybrids, the majority of which are repressed at least 10-fold. Also, we identify nearly 200 fibroblast-enriched genes that are repressed at least 5-fold. Silenced hepatic genes include several that encode transcription factors and proteins involved in signal transduction pathways. These data suggest that extensive reprogramming occurs in cell hybrids, leading to a nearly global (although not complete) loss of tissue-specific gene expression.

Dissociation of the Hepatic Phenotype from HNF4 and HNF1α Expression

Dedifferentiated cells have served as tools to understand the molecular consequences of the loss of tissue-specific pathways. Here we report the characterization of one of these cell lines, M29, which lacks the liver-enriched HNF4-HNF1α pathway, in order to determine if this class of variant cell lines could provide additional information regarding requirements for tissue-type expression. We report that although the liver-specific α1-antitrypsin (α1AT) gene remains silent despite reactivation of the HNF4/HNF1α pathway in the M29 cells, the frequency of activation of an integrated α1AT-APRT transgene is increased 1000-fold in response to these transcription factors. The human α1AT locus (introduced via chromosome transfer) also remained silent on these cells, despite HNF4 and HNF1α expression. Results from cell fusion experiments suggest that the defect in the M29 cells is recessive. Results suggest that the M29 cells contain a defect that represses liver gene expression despite the presence of the HNF4/HNF1α pathway.

Transcription factor BETA2 acts cooperatively with E2A and PDX1 to activate the insulin gene promoter

The insulin gene is efficiently expressed only in pancreatic beta cells. Using reverse transcriptase-polymerase chain reaction analysis, we show that insulin mRNA levels are at least 10(5)-fold higher in beta cells than non-beta cells. To examine the underlying mechanisms, we expressed beta cell transcription factors by transfection of non-beta cells. Separate expression of BETA2, E2A, or PDX1 led to modest (<10-fold) activation of the insulin promoter, whereas co-expression of the three proteins produced synergistic, high level activation (160-fold). This level of activity is approximately 25% that observed in transfected beta cell lines. Of the three factors studied, BETA2 appears to play a dominant role. Efficient transcription required a C-terminal activation domain of BETA2 and an N-terminal region, which does not function as an independent activation domain. The myogenic basic helix-loop-helix (bHLH) protein MyoD was unable to bind and activate the promoter, even when its DNA binding region was replaced with that of BETA2. Our results demonstrate the central importance of BETA2 in insulin gene transcription and the importance of sequences outside the canonical DNA binding domain in permitting efficient DNA binding and cell-specific activity of the insulin gene promoter.

Hepatocyte nuclear factor-4 prevents silencing of hepatocyte nuclear factor-1 expression in hepatoma x fibroblast cell hybrids

Hepatocyte nuclear factors-1alpha (HNF1alpha) and -4 (HNF4) are components of a liver-enriched transcription activation pathway which is thought to play a critical role in hepatocyte-specific gene expression, including activation of alpha1-antitrypsin gene expression. HNF1alpha, HNF4 and alpha1-antitrypsin (alpha1AT) genes are extinguished in hepatoma/fibroblast somatic cell hybrids, suggesting that fibroblasts contain a repressor-like activity. To determine the molecular basis for silencing of these genes in cell hybrids, ectopic expression of HNF1alpha and HNF4 was used. Results show that constitutive expression of HNF4 prevents extinction of HNF1alpha gene expression in hepatoma/fibroblast hybrids. In contrast, forced HNF1alpha expression failed to prevent extinction of the HNF4 locus in cell hybrids. Likewise, the alpha1AT gene remained silent in the presence of both HNF1alpha and HNF4. These results suggest that extinction of HNF1alpha is a simple lack-of-activation phenotype, whereas extinction of HNF4 andalpha1AT loci is more complex, perhaps involving negative regulation.

Insulin gene expression in immortalized rat hippocampal and pheochromocytoma-12 cell lines

Employing reverse transcription-polymerase chain reaction and clonal cell lines derived by retroviral transduction of the temperature sensitive simian virus 40 large T-antigen into dispersed rat embryonic hippocampal cells, we detected the ancestral gene-insulin II mRNA in three progenitor subcloned cell lines. These cell lines upon differentiation are known to express markers indicative of commitment to either neuronal (H19-7; NF + , GFAP -), glial (H19-5; GFAP +, NF -), or bipotential (H583-5, NF +, GFAP + ) lineages. No duplicated, i.e., insulin I gene expression, was observed in any of the three cell lines. Induction of differentiation was associated with the persistence of insulin II mRNA and in the cells expressing a neuronal phenotype (H19-7; NF +, GFAP -) a relative doubling in insulin II mRNA level was present (P < 0.05). Minimal cellular insulin immunoreactivity was detected only in a subpopulation of cells with a differentiated neuronal phenotype. Radioimmunoassayable insulin peptide in the H19-7 cellular conditioned medium revealed a 5-fold increase in the differentiated state. In contrast, peripheral sympathetic PC-12 neuronal cells both in the undifferentiated and nerve growth factor-driven differentiated states, failed to express both insulin I and insulin II genes. We conclude that insulin II is expressed by cultured rat hippocampal clonal cell lines, and not by the peripheral sympathetic PC-12 neuronal cell line.

Direct and indirect mechanisms of repression participate in suppression of T-cell-specific gene expression in T x L-cell hybrids

Expression of tissue-specific genes can be altered upon fusion of mammalian cells of different types. To resolve the genetic basis of this phenomenon and to identify components of the regulatory circuits that are involved, we have established a series of somatic cell hybrids between mouse T cells and L cells. These hybrids have an unusual and interesting phenotype. Unlike many hybrid cells studied, in which the expression of an entire set of tissue-specific genes was coordinately extinguished, in our T x L-cell hybrids only two out of seven T-cell-restricted genes were completely extinguished, whereas the other genes were repressed to various degrees. These hybrids extinguish the production of TCR beta and Thy-1 mRNA, repress the expression of TCR alpha, GATA-3, TCF-1, and LEF-1 genes to different extents, exhibit small changes in the level of CD3-epsilon mRNA, and continue to express the fibroblast-specific fibronectin gene, and the ets-1 gene. In this study we have evaluated for the first time the molecular mechanisms that underlie the repression of TCR alpha and TCR beta chain genes in T x L-cell hybrids. We have shown that multiple repression mechanisms, both direct and indirect, contribute to TCR alpha and TCR beta suppression. Repression of the expression of these genes correlated not only with the downregulation of GATA-3, TCF-1, and LEF-1 transcription factor expression, and with a change in the chromatin structure, but more importantly, with the activation of the silencer activity. Our study provides evidence for the existence of at least two negatively regulating elements, located at the TCR alpha enhancer-containing fragment and at the silencer region, which are active in our hybrid cells. We have shown that there was no correlation between the levels of GATA-3, TCF-1, and LEF-1 expression versus the level of TCR alpha mRNA in the independent hybrids. In contrast, both the silencer activity and the ability of the TCR alpha enhancer to downregulate thymidine kinase (TK) promoter activity were found to be in an inverse correlation with the ability of the different hybrid cells to express TCR alpha mRNA.

Negative regulation of transcription in eukaryotes

Images

Extinction of liver/bone/kidney alkaline phosphatase in osteosarcoma hybrid cells

We have constructed a series of interspecific somatic cell hybrids between the human osteoblast-like osteosarcoma, TE85, and a mouse fibrosarcoma, La-t-. In these whole-cell hybrids, we observed a 10-fold reduction of human liver/bone/kidney (L/B/K) alkaline phosphatase steady-state mRNA and alkaline phosphatase protein activity. The phenomenon of loss of tissue-specific gene expression has been termed extinction. Subclones of these hybrids were isolated, which reexpressed the alkaline phosphatase gene product. These late-passage hybrids had a reduced number of mouse fibroblast chromosomes when compared to earlier passages. This suggests that a trans-acting negative regulatory element, encoded in the fibroblast genome, regulates expression of L/B/K alkaline phosphatase. This is the first evidence that extinction plays a role in the regulation of osteoblast gene expression.

Insulin-producing cells contain a cell-specific repressor activity that functions through multiple E-box sequences

The cis-acting DNA element known as the E box (consensus sequence CAxxTG) plays an important role in the transcription of a number of cell-specifically expressed genes. The rat insulin I gene, for example, contains two such sequences (IEB1 and IEB2) that are recognized specifically by a characteristic beta cell nuclear factor insulin enhancer factor 1 (IEF1). To define the role of these elements better, we tested for cooperative interactions between the IEB sequences. Transfection experiments were performed with a series of plasmids containing the elements separated by different distances. Transcriptional activity in vivo is only modestly affected (less than two-fold) when the distances between the IEB elements are changed by a half-integral number of double-helical turns. Surprisingly, plasmids bearing four and six copies of the IEB motif showed sharply reduced activity as compared to those with two copies. In vitro DNA-binding studies revealed that this effect was not due to inability of IEF1 to bind to multiple copies of IEB. Moreover, multiple copies of the IEB sequence were able to inhibit activity of a cis-linked Moloney sarcoma virus (MSV) or insulin enhancer upon transfection to beta cells but not to other cell types. The above data are consistent with the view that beta cells contain a cell-specific repressor molecule capable of binding to multiple copies of IEB and thereby inhibiting transcription. This interpretation was further strengthened by in vivo competition and trans-activation experiments. The beta-cell-specific repressor activity identified by these studies may play an important role in mediating gene expression in insulin-producing cells, perhaps by regulating the access of helix-loop-helix transcription factors to E-box sequence elements.