Regulation of FoxO activity by CBP/p300-mediated acetylation Lars P. van der Heide and Marten P. Smidt Rudolf Magnus Institute of Neuroscience, UMC Utrecht, Universiteitsweg 100, 3584 CG, Utrecht, The Netherlands Forkhead box, class O (FoxO) transcription factors are inhibited by insulin-induced FoxO phosphorylation. Recently, acetylation of FoxO factors by calcium response element-binding (CREB)-binding protein (CBP) and/or p300 has been identified as a novel regulatory pathway, although the exact consequences of acetylation remain unclear. We propose that binding of CBP/p300 to FoxO factors is essential for FoxO- mediated transcription. CBP and p300 act as FoxO cofactors by weakening histone–DNA interactions. Acetylation of FoxO factors, however, attenuates FoxO-mediated transcriptional activity by disrupting the interaction between FoxO factors and target DNA. Therefore, acetylation shifts the function of FoxO from cell-cycle arrest and protection against oxidative stress towards cell death. Introduction Initial mammalian studies on forkhead box, class O (FoxO) transcription factors were performed using fork- head [a transcription factor with a forkhead (winged helix) domain as the DNA-binding interface] in rhabdomyosar- coma (FKHR), which is a FoxO factor implicated in the pathogenesis of alveolar rhabdomyosarcomas [1,2]. FKHR has now been renamed as FOXO1, according to the new nomenclature, and is the first mammalian member of the FoxO family of transcription factors. Recently, FoxO proteins have been identified in several different organ- isms, including Caenorhabditis elegans, zebrafish, Droso- phila, mouse, rat and humans. In the mouse, four different FoxO members have been identified to date: FoxO1, FoxO3, FoxO4 and FoxO6 [3,4]. Over recent years, it has become evident that FoxO factors are sensitive to insulin– phosphatidylinositol 3-kinase (PI3K)–protein kinase B (PKB) signalling and have an array of downstream targets and interacting partners (Box 1). Central to insulin- mediated inhibition of FoxO factors is a shuttling mechanism that shifts FoxO localization to the cytosol, thereby terminating its transcriptional function. Phos- phorylation-dependent regulation of FoxO factors was long considered the sole FoxO regulatory pathway, however, several recent papers discuss the acetylation of FoxO factors by calcium response element-binding (CREB)-binding protein (CBP) and/or p300 and the deacetylation by Sir2 as a novel regulatory pathway [5–9]. Whereas the majority describes that acetylation of FoxO factors represses target gene transcription, some suggest that FoxO acetylation increases target gene transcription. This discrepancy has been attributed to cell type and the target-gene-specific effects of FoxO. For understanding this dual role of CBP/p300 in FoxO-mediated transcription, it is important to differen- tiate between acetylation of histone–DNA and acetylation of FoxO factors (Box 2). Therefore, we propose that CBP/ p300 functions as a cofactor in FoxO-mediated transcrip- tional activity, whereas FoxO acetylation attenuates FoxO-mediated transcription of target genes. The mech- anism of FoxO acetylation by CBP/p300 shares simi- larities with that of p53 regulation by CBP/p300, and might regulate FoxO DNA-binding capabilities. Box 1. In vivo FoxO functions DAF-16 is the FoxO homologue expressed in Caenorhabditis elegans, and is regulated by a signalling pathway similar to the mouse insulin–PI3K–PKB pathway. DAF-16 is remarkably similar to mouse FoxO proteins, and can be partially substituted by FOXO3 [11], which exemplifies the evolutionary conservation between FoxO factors. Direct activation of DAF-16, or mutation of the insulin–PI3K– PKB pathway in C. elegans results in life-span extension, stress resistance and arrest at the dauer diapause stage [12]. Besides cell- autonomous inputs, DAF-16 also responds to environmental inputs: starvation, heat and oxidative stress all activate DAF-16, whereas nutrient-rich conditions deactivate it. A unique FoxO homologue in Drosophila, dFOXO [13–15], seems to have similar roles in metabolism to DAF-16 in C. elegans. The insulin–PI3K–PKB signal- ling cascade [13–15] and nutrients [15] negatively regulate dFOXO. However, although dFOXO-knockout flies are viable and of normal size, they are more vulnerable to oxidative stress, which suggests that dFOXO provides protection against oxidative stress. The possible in vivo functions of FoxO proteins have recently been extended because of the increasing complexity of the specific phenotypes of FOXO1, FOXO3, and FOXO4 in studies of mammalian (mouse) models. FOXO1 homozygous null-mutants die before birth due to several embryonic defects [16], including incomplete vascular development [17]. Analysis of heterozygote null-mutants indicates that FOXO1 is involved in pancreatic b-cell function, hepatic glucose metabolism and adipocyte differentiation [16,18,19]. Inspection of the FOXO3 knockout reveals haematological abnormalities, a decreased glucose uptake in glucose-tolerance tests [20] and a distinct ovarian phenotype due to premature follicular activation. Thus, it has been suggested that FOXO3 functions during the early stages of follicular growth as a suppressor of follicular activation [17,20]. The FOXO4 knockout does not have any obvious abnor- mality [17]. Taken together, in vivo studies implicate FoxO proteins in the homeostasis of metabolism; FoxO factors respond to nutrients, growth factors and stress to ‘fine-tune’ cellular metab- olism and optimally adapt the cell to an ever-changing environment. Corresponding author: Smidt, M.P. (m.p.smidt@med.uu.nl). Available online 24 December 2004 Opinion TRENDS in Biochemical Sciences Vol.30 No.2 February 2005 www.sciencedirect.com 0968-0004/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tibs.2004.12.002