Independent transcriptional regulation of a single VL30 element by epidermal growth factor and activators of protein kinase C

Complex interactions with direct repeats of a mitogen-responsive VL30 enhancer

The RVL-3 VL30 enhancer is an LTR-derived triple direct repeat of 35 base pairs that mediates gene induction in response to several different intracellular signaling pathways. Using mobility shift assays, methylation interference and DNase I footprinting, we have investigated the physical interactions between the RVL-3 enhancer and components of nuclear extracts from Rat-1 cells. Each enhancer repeat unit contains a single binding site. Our studies suggest that binding to the double or triple repeat enhancer is cooperative, involving simultaneous occupation of two sites, with a preference for adjacent sites. Binding cooperativity would have implications for the mechanism of gene activation directed from the native VL30 enhancer.

Regulation of transin/stromelysin and VL30 gene expression by intracellular calcium

Changes in intracellular Ca++ levels are observed as a second messenger in response to a number of cellular agonists, including epidermal growth factor, transforming growth factor beta 1, and endothelin-1. The role of elevated intracellular Ca++ in transducing the effects of these three agonists on gene expression has been studied using two target genes: transin/stromelysin-1 and the endogenous murine retrovirus VL30. Although the effects of EGF and TGF beta 1 on transin/stromelysin-1 mRNA expression appear to be independent of these agonists' effects on intracellular Ca++ levels, elevated Ca++ interacted synergistically with activators of pkC to induce transin expression, even though neither agent alone could induce transin/stromelysin-1 expression. In contrast, the integrated VL30 retrovirus could be induced by Ca++ ionophores alone, and induction of VL30 mRNA by other agonists was blocked if intracellular Ca++ levels were held below a threshold value of 165 nM with Ca++ chelators. Genetic analysis of the VL30 upstream regulatory region indicated that a triple-repeat element present in the VL30 long-terminal repeat could function as an inducible enhancer, but responsiveness to either EGF or pkC activation required the concomitant elevation of intracellular Ca++. Because EGF was capable of inducing expression even in pkC-depleted cells, providing Ca++ levels were elevated, these results indicate that elevated intracellular Ca++ is capable of interacting synergistically with multiple signaling pathways to stimulate increased gene expression.

Protein kinase C-independent activation of c-jun and c-fos transcription by epidermal growth factor

Phorbol esters, epidermal growth factor (EGF) and serum induce the transient expression of the c-jun and c-fos proto-oncogenes in quiescent fibroblasts. While phorbol esters such as phorbol 12-myristate 13-acetate (PMA) are thought to induce the transcription of these genes by activating protein kinase C (PKC), the signal transduction pathway(s) mediating the effects of EGF and serum are still unclear. We have investigated whether PKC and/or calcium play a role in mediating EGF-stimulated c-jun and c-fos RNA and protein expression in quiescent NIH 3T3 fibroblasts. PMA, EGF or serum stimulated a rapid, transient increase in c-jun and c-fos expression and cJun protein synthesis in quiescent NIH 3T3 cells. Depletion of whole cell PKC activity by pretreatment with PMA abolished any subsequent response to PMA, but had no effect on the ability of EGF or serum to induce c-jun and c-fos RNA and cJun protein expression. Nuclear run-on analysis indicated that EGF-induced gene expression was due to an increase in the rate of transcription of c-jun and c-fos in both naive and PKC-depleted cells. The role of calcium in the EGF-induced expression of c-jun and c-fos was also investigated using an NIH 3T3 cell line (HER-14) overexpressing the wild type human EGF receptor. Removal of extracellular calcium by chelation with excess EGTA or use of the non-specific calcium channel blocker lanthanide, both of which abolish the EGF-induced calcium transient in HER-14 cells, had no effect on the PMA or EGF induced c-jun or c-fos response. These findings suggest that EGF induces c-jun and c-fos transcription and cJun protein synthesis in a manner independent of an increase in intracellular calcium or activation of PKC in quiescent NIH 3T3 cells.

Altered levels of endogenous retrovirus-like sequence (VL30) RNA during mouse epidermal cell carcinogenesis

Gene expression during mouse keratinocyte carcinogenesis was examined in a clonal cell model. Tumor cells from three separate initiated cell lineages were compared with their nontumorigenic precursors and with the progenitor cell strain prior to treatment with 7,12-dimethylbenz[a]anthracene (DMBA). The steady-state levels of VL30 RNA in normal and papilloma cells were regulated by extracellular Ca2+ (which controls proliferation and differentiation in normal epidermal keratinocytes) and culture density. In contrast, steady-state levels of VL30 RNA were not regulated by these factors in the squamous cell carcinoma or the anaplastic carcinoma cells. VL30 expression was Ca2+ dependent in the initiated cell precursors within each tumor cell lineage, suggesting that the loss of response to extracellular Ca2+ was associated with the malignant conversion stage of carcinogenesis. No differences between normal and tumor cells were found in the cellular RNA levels of five additional proto-oncogenes. The mouse epidermal cell model should provide a means for direct assessment of a potential functional role of VL30 sequences in cancer development.

Retrotransposon-like VL30 elements are efficiently induced in anoxic rat fibroblasts

VL30 elements are a multigene family within the class of retroviruses and retrotransposons. We have characterized the response of normal rat fibroblasts to anoxia, in which endogenous VL30 element expression is strongly induced. Optimal induction up to 500-fold occurs under complete anoxia, although a lesser response is seen under atmospheres up to 2% oxygen. Phorbol esters and diacylglycerol, which induce mouse VL30 RNA approximately eightfold, show no effect on the rat VL30 system. The hypoxic conditions optimal for rat VL30 induction represent a mild cellular stress, with no reduction in cell viability during the induction period. Although the precise physiological role of this fibroblast response to temporary anoxia is unknown, it may occur during wound healing. The induction of VL30 by anoxia provides a unique model system wherein a member of the mammalian retrovirus/retrotransposon family is highly responsive to a common physiological signal.