The yeast CHA1 promoter is activated in the presence of serine or threonine. Activation requires the Cha4p activator, and it results in perturbation of a nucleosome that incorporates the TATA element under noninducing conditions. We show that in yeast lacking the amino terminus of histone H3, the promoter is constitutively active and the chromatin is concomitantly perturbed. This derepression occurs in the absence of elevated intracellular levels of serine or threonine and is not observed in cells lacking Rpd3p, Tup1p, or the amino terminus of histone H4. Furthermore, derepression in the absence of the H3 amino terminus requires the primary activator of this promoter, Cha4p, which we show by chromatin immunoprecipitation to be constitutively bound to the CHA1 promoter in WT yeast. Thus, the H3 amino terminus is required to prevent Cha4p from activating CHA1 in the absence of inducer. We also present results of a microarray experiment showing that the H3 amino terminus has a substantial repressive effect on a genome-wide scale. The principal protein components of chromatin are the histones. The four core histones, H2A, H2B, H3, and H4, can be divided into central structured domains and relatively unstructured amino and, in some cases, carboxyl-terminal regions, often referred to as the histone tails (1, 2). These latter regions do not contribute to core nucleosome structure (3), but are subject to numerous posttranslational modifications that contribute to multiple aspects of chromatin function (4-6). Specific modifications have been correlated with transcriptional activation and others with repression, including the formation of stably repressed domains of heterochromatin (7).One approach to studying the role of the histone tails in transcriptional regulation has been to focus on specific histonemodifying enzymes, or specific residues in the histone amino termini. An alternative, complementary approach has been to examine the effects of deleting individual histone amino termini. Such work has shown that in budding yeast (Saccharomyces cerevisiae), the individual histone amino termini are dispensable, but the H3 and H4 tails cannot be simultaneously removed without loss of viability, although they can be interchanged (8)(9)(10)(11). Other experiments have shown that the H3 and H4 tails are involved in silencing at telomeres and the silent mating-type loci (12, 13), and have implicated the H3 and H4 amino termini in both positive and negative aspects of gene regulation (14)(15)(16)(17).In this work we demonstrate a role for the histone H3 amino terminus in the regulation of the CHA1 promoter in yeast. Our initial interest in the CHA1 promoter stems from its tightly regulated chromatin structure. In its uninduced state, the CHA1 TATA element is incorporated near the center of a positioned nucleosome, and this nucleosome is perturbed on activation of CHA1 by increased intracellular levels of serine or threonine (18). Nucleosome perturbation requires the principal CHA1 activator, Cha4p, and can be elicited under cond...

In Saccharomyces cerevisiae Fat1p and fatty acyl-CoA synthetase (FACS) are hypothesized to couple import and activation of exogenous fatty acids by a process called vectorial acylation. Molecular genetic and biochemical studies were used to define further the functional and physical interactions between these proteins. Multicopy extragenic suppressors were selected in strains carrying deletions in FAA1 and FAA4 or FAA1 and FAT1. Each strain is unable to grow under synthetic lethal conditions when exogenous long-chain fatty acids are required, and neither strain accumulates the fluorescent long-chain fatty acid C 1 -BODIPY-C 12 indicating a fatty acid transport defect. By using these phenotypes as selective screens, plasmids were identified encoding FAA1, FAT1, and FAA4 in the faa1⌬ faa4⌬ strain and encoding FAA1 and FAT1 in the faa1⌬ fat1⌬ strain. Multicopy FAA4 could not suppress the growth defect in the faa1⌬ fat1⌬ strain indicating some essential functions of Fat1p cannot be performed by Faa4p. Chromosomally encoded FAA1 and FAT1 are not able to suppress the growth deficiencies of the fat1⌬ faa1⌬ and faa1⌬ faa4⌬ strains, respectively, indicating Faa1p and Fat1p play distinct roles in the fatty acid import process. When expressed from a 2 plasmid, Fat1p contributes significant oleoyl-CoA synthetase activity, which indicates vectorial esterification and metabolic trapping are the driving forces behind import. Evidence of a physical interaction between Fat1p and FACS was provided using three independent biochemical approaches. First, a C-terminal peptide of Fat1p deficient in fatty acid transport exerted a dominant negative effect against long-chain acyl-CoA synthetase activity. Second, protein fusions employing Faa1p as bait and portions of Fat1p as trap were active when tested using the yeast two-hybrid system. Third, co-expressed, differentially tagged Fat1p and Faa1p or Faa4p were co-immunoprecipitated. Collectively, these data support the hypothesis that fatty acid import by vectorial acylation in yeast requires a multiprotein complex, which consists of Fat1p and Faa1p or Faa4p.