Molecular cloning and characterization of the yeast gene for squalene synthetase.

Squalene synthetase (farnesyl-diphosphate: farnesyl-diphosphate farnesyltransferase, EC 2.5.1.21) is a critical branch point enzyme of isoprenoid biosynthesis that is thought to regulate the flux of isoprene intermediates through the sterol pathway. The structural gene for this enzyme was cloned from the yeast Saccharomyces cerevisiae by functional complementation of a squalene synthetase-deficient erg9 mutant. Identification of this ERG9 clone was confirmed by genetic linkage analysis in yeast and expression of enzyme activity in Escherichia coli. The predicted squalene synthetase polypeptide of 444 amino acids (Mr, 51,753) lacks significant homology to known protein sequences, except within a region that may represent a prenyl diphosphate (substrate) binding site. The ERG9-encoded protein contains a PEST consensus motif (rich in proline, glutamic acid, serine, and threonine) present in many proteins with short cellular half-lives. Modeling of the protein suggests that it contains at least one, and possibly two, membrane-spanning domains. Disruption of the chromosomal squalene synthetase coding region by insertional mutagenesis indicates that ERG9 is a single copy gene that is essential for cell growth in yeast.

E1A-dependent trans-activation of the c-fos promoter requires the TATAA sequence.

Previous experiments have demonstrated that transcription of the human c-fos oncogene is activated through the action of the 289-amino acid adenovirus E1A gene product. In this study we have utilized a series of c-fos promoter deletion and substitution mutants to define regulatory sequences that allow the induction by E1A. Although the deletion of upstream promoter sequences has varying degrees of effect on overall promoter activity, these deletions retain inducibility by E1A. This includes the deletion of the serum response element and two elements that bind the ATF transcription factor. In fact, a c-fos promoter deleted to position -53, which leaves the TATA element but no other known functional element, retains inducibility, indicating a role for the TATA element in E1A control. Indeed, substitution of the c-fos TATA element (TATAA) with a TATA sequence from the simian virus 40 early promoter (TATTTAT) abolishes E1A inducibility; this promoter does retain responsiveness to cAMP induction, however, demonstrating that this TATTTAT substitution is functional. We conclude that the E1A-dependent activation of c-fos transcription is mediated through an effect on a TATA-binding protein that has specificity for the TATAA sequence.

An AP1-binding site in the c-fos gene can mediate induction by epidermal growth factor and 12-O-tetradecanoyl phorbol-13-acetate.

We have demonstrated that two sequence elements in the c-fos promoter can mediate the response of the gene to epidermal growth factor and the tumor promoter 12-O-tetradecanoyl phorbol-13-acetate (TPA). The first is the previously described serum response element. The second is a sequence element highly homologous to the consensus binding site for the HeLa cell transcription factor AP1. Although recent reports have shown that fos protein binds to AP1-binding sites through an interaction with AP1 protein and have raised the speculation that fos protein may negatively regulate expression of the c-fos gene via this interaction, we found no role for the AP1 consensus homology in the downregulation of c-fos expression following induction by epidermal growth factor and TPA.

Multiple sequence elements in the c-fos promoter mediate induction by cAMP.

Induction of c-fos by cyclic AMP in NIH-3T3 cells is distinct from induction by serum. Whereas induction by serum is mediated by the serum response element (SRE), induction by cAMP does not require this element. In fact, no single sequence element in the c-fos promoter/enhancer is stringently required for the cAMP response. Rather, multiple sequence elements in the c-fos promoter/enhancer can mediate induction by cAMP independently. These elements are: (1) the region from -72 to -54, which contains a binding site for a cellular activating transcription factor (ATF); (2) the region from -225 to -99; (3) the region from -303 to -281, which is homologous to the consensus binding site for the transcription factor AP1; and (4) the region from -317 to -298, which contains the SRE. These sequence elements convey cAMP inducibility when fused to the cAMP-unresponsive 'minimal' fos promoter (-53 to +42). In addition, the c-fos regions from -700 to -63 and from -71 to -48 can confer cAMP inducibility to a heterologous promoter.

Definition of multiple, functionally distinct TATA elements, one of which is a target in the hsp70 promoter for E1A regulation.

We have dissected the human hsp70 promoter to define sequence elements allowing response to E1A. Alterations of sequence upstream of the TATA element, either with Bal 31 nuclease or by site-directed mutagenesis, had little or no effect on the response of the promoter to E1A. In general, the basal level was reduced, indicating that these sites interact with factors important for transcription, but regulation persisted. Although a CAT gene driven by just the hsp70 TATA (void of upstream sequences) could be stimulated by E1A, a similar construct containing the early SV40 TATA element was not. Analysis of several additional such constructions indicated that the specific sequence TATAA was crucial. Substitution of the TATAA sequence with the SV40 TATTTAT element in the context of the wild-type hsp70 promoter resulted in loss of E1A inducibility, but maintenance of heat inducibility. Replacement of this element with sequences not related to any TATA element resulted in loss of activity and inducibility. Thus, the SV40 TATA equivalent is functional in the context of the hsp70 promoter but cannot be induced by E1A. We conclude that the target for E1A induction of the hsp70 promoter is TATAA, and that multiple functionally distinct TATA elements, and presumably cognate transcription factors, can be distinguished in eukaryotic cells.

c-fos sequence necessary for basal expression and induction by epidermal growth factor, 12-O-tetradecanoyl phorbol-13-acetate and the calcium ionophore.

We have investigated the sequence requirements for induction of the human c-fos gene by epidermal growth factor (EGF), 12-O-tetradecanoyl-13-acetate (TPA), and the calcium ionophore A23187 by transfecting c-fos promoter mutants into HeLa and A431 cells. Induction by both EGF and TPA in HeLa cells required the presence of the c-fos enhancer located at -317 to -298 relative to the mRNA cap site. A23187, however, did not induce expression of the transfected gene, even though it strongly induced expression of the endogenous gene, suggesting that it has different requirements for induction than do EGF and TPA. We have also investigated the role of promoter sequences downstream of the enhancer in general expression and induction of c-fos. A sequence between -97 and -76, which includes an 8-base-pair perfect direct repeat, was needed for efficient general expression but not for induction of the gene. A factor in nuclear extracts that bound specifically to this sequence was detected by a gel mobility shift assay. A 7-base-pair sequence, located between -63 and -57 relative to the mRNA cap site and previously shown to be important for general expression of mouse c-fos, was also important for general expression of the human gene. In addition, this element was important for inducibility by EGF and TPA, since induction was significantly reduced when internal deletion mutants that retained the enhancer but lacked the -63 to -57 sequence element were analyzed in transfecting assays.