ERF (ETS2 Repressor Factor) is a novel member of the ets family of genes, which was isolated by virtue of its interaction with the ets binding site (EBS) within the ETS2 promoter. The 2.7 kb ubiquitously expressed ERF mRNA encodes a 548 amino acid phosphoprotein that exhibits strong transcriptional repressor activity on promoters that contain an EBS. The localization of the DNA‐binding domain of the protein at the N‐terminus and th repression domain at the C‐terminus is reminiscent of the organization of ELK1‐like members of the ets family; however, there is no significant homology between ERF and ELK1 or any other ets member outside the DNA‐binding domain. The repressor activity of ERF can antagonize the activity of other ets genes that are known transcriptional activators. Furthermore, ERF can suppress the ets‐dependent transforming activity of the gag‐myb‐ets fusion oncogene of ME26 virus. Although ERF protein levels remain constant throughout the cell cycle, the phosphorylation level of the protein is altered as a function of the cell cycle and after mitogenic stimulation. The ERF protein is also hyperphosphorylated in cells transformed by the activated Ha‐ras and v‐src genes and the transcription repressor activity of ERF is decreased after co‐transfection with activated Ha‐ras or the kinase domain of the c‐Raf‐1 gene, indicating that ERF activity is probably regulated by the ras/MAPK pathway. Consistent with the in vivo phosphorylation and inactivation by ras, ERF is efficiently phosphorylated in vitro by Erk2 and cdc2/cyclin B kinases, at sites similar to those detected in vivo. Furthermore, a single mutation at position 526 results in the loss of a specific phosphopeptide both in in vivo and in vitro (by Erk2) labeling. Substitution of Thr526 for glutamic acid also decreases the repression ability of ERF. Our data suggest a model in which modulation of ERF activity is involved in the transcriptional regulation of genes activated during entry into G1 phase. Obstruction of the ERF repressor function by the transactivating members of the ets family of genes (i.e.gag‐myb‐ets) may be essential for the control of genes involved in cell proliferation and may also underlie their tumorigenic effects.

A limited number of transcription factors have been suggested to be regulated directly by Erks within the Ras/mitogen-activated protein kinase signaling pathway. In this paper we demonstrate that ERF, a ubiquitously expressed transcriptional repressor that belongs to the Ets family, is physically associated with and phosphorylated in vitro and in vivo by Erks. This phosphorylation determines the ERF subcellular localization. Upon mitogenic stimulation, ERF is immediately phosphorylated and exported to the cytoplasm. The export is blocked by specific Erk inhibitors and is abolished when residues undergoing phosphorylation are mutated to alanine. Upon growth factor deprivation, ERF is rapidly dephosphorylated and transported back into the nucleus. Phosphorylation-defective ERF mutations suppress Ras-induced tumorigenicity and arrest the cells at the G 0 /G 1 phase of the cell cycle. Our findings strongly suggest that ERF may be important in the control of cellular proliferation during the G 0 /G 1 transition and that it may be one of the effectors in the mammalian Ras signaling pathway.Mitogen-activated protein kinase (MAPK) pathways are a central relay of many extracellular signals leading to change in gene expression. At least three MAPK pathways, which have high structural homology and identity in biochemical mechanisms of activation, have been identified in mammalian cells. The JNK (c-Jun amino-terminal kinase) and p38 pathways are involved primarily in the transduction of stress and cytokine stimuli. The Erk (extracellular signal-regulated kinase) pathway plays a major role in transduction of mitogenic and differentiation stimuli (for reviews, see references 41 and 47). Ras small GTPases have a pivotal role in regulation of proliferation from both receptor tyrosine kinases (RTK) and G proteinmediated receptors, (for reviews, see references 6 and 30). Notably, Ras plays an essential role in the activation of the Raf kinase, which directly phosphorylates and activates the Mek kinase, leading to the activation of Erk1 and Erk2 by phosphorylation on threonine and tyrosine residues. Phosphorylated Erks form homodimers (22) and translocate to the nucleus, where they phosphorylate proteins involved in gene regulation. Besides the Raf/Mek/Erk kinase cascade, other downstream Ras effectors are known to participate in the proliferative response (for a review, see reference 31). For example, phosphoinositide 3-OH kinase (PI3-K) (42) and members of the Rho family (for reviews, see references 17 and 23) have been shown to be responsible for morphological changes induced by Ras and are required for Ras-dependent transformation. Nevertheless, although the implication of Ras pathways, and in particular the Raf/Erk pathway, in the control of proliferation is well established, links with the control of the cell cycle machinery are not clear. Ras-dependent exit from G 0 (45) and progression through G 1 via the control of the retinoblastoma tumor suppressor protein (Rb) (34, 37) have been demonstrated. However, the transcription...

We developed a simple and accurate method to define the sequence recognition properties of DNA-binding proteins. The method employs polymerase chain reaction (PCR) amplification of sequences selected from a mixture of random oligonucleotides by the gel mobility-shift assay. We used this method to define the sequence requirement of the binding domain of the yeast transcriptional activator GCN4. Using a total of 200 ng of purified protein and four cycles of binding and subsequent amplification, we identified the TGA-(C/G)TCA sequence as the binding consensus of GCN4, which is consistent with the previously reported recognition sequence. In addition, our data indicate that GCN4 can bind with lower affinity to sequences that differ from the optimal sequence in one or even two positions. The most common variation was the C to A at position +2. The majority of the substitutions that still allowed binding were 3' to the central C residue indicating that the two sides of the palindromic recognition sequence are not equivalent.