Sterol Regulatory Element-binding Proteins (SREBPs): Key Regulators of Nutritional Homeostasis and Insulin Action

Abstract

sterol regulatory element-binding protein sterol regulatory element sterol regulatory element-binding protein-1c or adipocyte differentiation and determination factor-1 basic helix-loop-helix bHLH leucine zipper low density lipoprotein cAMP-response element-binding protein vitamin D receptor-interacting protein activator-recruited cofactor fatty acid synthase polyunsaturated fatty acid(s) The sterol regulatory element-binding proteins (SREBPs)1 were first identified by two groups working independently on cholesterol metabolism (1Yokoyama C. Wang X. Briggs M.R. Admon A. Wu J. Hua X. Goldstein J.L. Brown M.S. Cell. 1993; 75: 185-197Crossref Google Scholar, 2Hua X. Yokoyama C. Wu J. Briggs M.R. Brown M.S. Goldstein J.L. Wang X. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11603-11607Crossref PubMed Scopus (501) Google Scholar) and fat cell differentiation (3Tontonoz P. Kim J.B. Graves R.A. Spiegelman B.M. Mol. Cell. Biol. 1993; 13: 4753-4759Crossref PubMed Scopus (534) Google Scholar). Subsequent studies have demonstrated there are three major SREBP isoforms encoded by two different genes (4Brown M.S. Goldstein J.L. Cell. 1997; 89: 331-340Abstract Full Text Full Text PDF PubMed Scopus (3004) Google Scholar). These unique members of the basic helix-loop-helix leucine zipper (bHLHLZ) family of transcriptional regulatory proteins can be distinguished from other family members by two characteristics. The first is they are synthesized as precursors that are threaded into membranes of the endoplasmic reticulum and nuclear envelope in a hairpin orientation such that the amino and carboxyl tails both face the cytoplasm. The amino-terminal half of the precursor is clipped out of the membrane in two steps responding to regulatory cues that signal the need for increased cellular cholesterol (5Sakai J. Duncan E.A. Rawson R.B. Hua X. Brown M.S. Goldstein J.L. Cell. 1996; 85: 1037-1046Abstract Full Text Full Text PDF PubMed Scopus (434) Google Scholar). The released amino-terminal fragment, which contains the transcriptional activation and DNA binding domains, is targeted to the nucleus where it activates expression of SREBP target genes. The second distinguishing feature of the SREBPs is that they have a unique dual DNA binding specificity, which is discussed below. Two of the three major isoforms are produced from the SREBP-1 gene, which contains two promoters (6Miserez A.R. Cao G. Probst L. Hobbs H.H. Genomics. 1997; 40: 31-40Crossref PubMed Scopus (90) Google Scholar). Transcription from each promoter produces an mRNA with a unique first exon that encodes one of the alternative amino termini referred to as 1a and 1c, respectively (Fig.1). These alternate exons are attached during mRNA splicing to a common second exon in the same reading frame, and therefore, the remaining protein coding information of both isoforms is identical. There is alternative mRNA splicing at the 3′-end as well (7Hua X. Wu J. Goldstein J.L. Brown M.S. Hobbs H.H. Genomics. 1995; 25: 667-673Crossref PubMed Scopus (247) Google Scholar), but this does not appear to be conserved in all mammalian species and its functional significance remains unclear (8Shimomura L. Shimano H. Horton J.D. Goldstein J.L. Brown M.S. J. Clin. Invest. 1997; 99: 838-845Crossref PubMed Scopus (641) Google Scholar). In contrast, mRNAs produced through alternative promoter usage at the 5′-end yield proteins with significant differences in their capacity to activate gene expression (9Shimano H. Horton J.D. Shimomura L. Hammer R.E. Brown M.S. Goldstein J.L. J. Clin. Invest. 1997; 99: 846-854Crossref PubMed Scopus (684) Google Scholar). The longer amino-terminal region in SREBP-1a contains a high percentage of acidic amino acids that make it a potent transcriptional activation domain. The SREBP-1c isoform is a much weaker activator of gene expression because it lacks 29 acidic amino acids present in 1a. Using a nuclease mapping technique to evaluate the relative levels of SREBP-1a and SREBP-1c mRNA, the ratio was shown to vary over an ∼50–100-fold range in different tissues of the body. In liver and adipocytes, 1c mRNA is 9- and 3-fold, respectively, more abundant than SREBP-1a, whereas in spleen SREBP-1a is 10 times more abundant than 1c (8Shimomura L. Shimano H. Horton J.D. Goldstein J.L. Brown M.S. J. Clin. Invest. 1997; 99: 838-845Crossref PubMed Scopus (641) Google Scholar). Whether these ratios reflect similar differences in the levels of each protein remains to be firmly established. In all cultured cell lines examined, SREBP-1a was expressed at higher levels (8Shimomura L. Shimano H. Horton J.D. Goldstein J.L. Brown M.S. J. Clin. Invest. 1997; 99: 838-845Crossref PubMed Scopus (641) Google Scholar). 1c mRNA is the predominant isoform in adult liver and adipocytes, so it is likely to be the key protein involved in SREBP-1-dependent processes in these tissues. Why different tissues express different ratios of SREBP-1a and -1c is not clear. It is possible that the more active 1a isoform is preferentially expressed when there is a high demand for cholesterol and fatty acids such as when new membrane is required during periods of rapid cell division. SREBP-1a stimulates gene expression in vitro and in cultured cells by interacting with the transcriptional coactivators CBP and P300 (10Oliner J.D. Andresen J.M. Hansen S.K. Zhou S. Tjian R. Genes Dev. 1996; 10: 2903-2911Crossref PubMed Scopus (140) Google Scholar, 11Ericsson J. Edwards P.A. J. Biol. Chem. 1998; 273: 17865-17870Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) (Fig. 1). These are large, ubiquitous transcriptional coactivator proteins that are recruited to specific promoters through binding to activation domains of several DNA binding transcription factors in addition to SREBPs (12Kwok R.P. Lundblad J.R. Chrivia J.C. Richards J.P. Bachinger H.P. Brennan R.G. Roberts S.G. Green M.R. Goodman R.H. Nature. 1994; 370: 223-226Crossref PubMed Scopus (1282) Google Scholar). The shorter activation domain of SREBP-1c does not interact efficiently with CBP or P300, and how this isoform activates transcription is not clearly understood. 2K. A. Dooley and T. F. Osborne, unpublished data. The single SREBP-2 isoform similarly interacts with CBP and P300 to activate transcription (Fig. 1). The amino-terminal domain of SREBP-1a interacts also with a separate multisubunit complex alternately called vitamin D receptor-interacting protein (DRIP) or activator-recruited cofactor (ARC) (13Nåår A.M. Beauring P.A. Zhou S. Abraham S. Solomon W. Tjian R. Nature. 1999; 398: 828-832Crossref PubMed Scopus (373) Google Scholar). This heterogeneous complex increases transcription through interacting with activation domains of several other DNA-binding transcriptional regulatory proteins in addition to SREBP-1a. The DRIP/ARC interaction is independent of CBP/P300. As members of the bHLHLZ family of DNA-binding proteins, SREBPs form dimers that recognize the inverted repeat E-box 5′-CANNTG-3′ (where N represents any base). bHLH subfamilies can be classified according to their preference for specific bases at the middle positions of the E-box (14Murre C. Bain G. van Dijk M.A. Engel I. Furnari B.A. Massari M.E. Matthews J.R. Quong M.W. Rivera R.R. Stuiver M.H. Biochim. Biophys. Acta. 1994; 1218: 129-135Crossref PubMed Scopus (411) Google Scholar). SREBPs belong to the same subfamily as Myc/Max and USF, which all prefer the 5′-CACGTG-3′ E-box. However, SREBPs are further distinguished because they not only bind to this inverted repeat but also to the direct repeat sterol regulatory element (SRE) 5′-TCACNCCAC-3′ or to related sites (15Millinder-Vallett S. Sanchez H.B. Rosenfeld J.M. Osborne T.F. J. Biol. Chem. 1996; 271: 12247-12253Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). This flexibility is because of a unique tyrosine residue in the SREBP basic domain that corresponds to an arginine in all other E-box-binding bHLH proteins. Kim et al. (16Kim J.B. Spotts G.D. Halvorsen Y.-D. Shih H.-M. Ellenberger T. Towle H.C. Spiegelman B.M. Mol. Cell. Biol. 1995; 15: 2582-2588Crossref PubMed Scopus (296) Google Scholar) demonstrated the importance of the tyrosine residue by changing it to an arginine to mimic other bHLH proteins. The resulting protein bound only to E-boxes. The reciprocal mutation changing the arginine of the related USF1 protein to a tyrosine converted it into an E-box- and SRE-recognizing protein. When the x-ray structure of the DNA binding domain of SREBP-1 bound to the SRE element was compared with DNA-bound structures for other bHLH proteins it was revealed that the tyrosine permits the basic domain to adopt a slightly different conformation allowing it to recognize specifically the direct repeat site (17Parraga A. Bellsolell L. Ferre-D'Amare A.R. Burley S.K. Structure. 1998; 6: 661-672Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). This duality in DNA recognition has significant functional implications because all cholesterol-regulated SREBP-dependent promoters that have been carefully evaluated contain direct repeat SRE type sites and not E-boxes (18Athanikar J.N. Osborne T.F. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 4935-4940Crossref PubMed Scopus (52) Google Scholar). SREBPs are the only mammalian bHLH proteins that have been identified with the unique tyrosine residue in their DNA binding domain, and they are not present in the nucleus until a low sterol level activates their proteolytic release from their membrane tether. There are several other E-box-binding proteins in the nucleus independent of the cholesterol level of the cell. If cholesterol-regulated genes had E-box sites they could be activated by these other proteins before SREBP entered the nucleus. Thus, the net difference in target gene expression before and after cholesterol depletion would be small. However, direct repeat SRE sites would ensure that no other bHLH protein could activate target genes in the absence of nuclear SREBP. This would effectively maximize the regulatory response and amplify the difference between the uninduced and induced state. Several distinct genes of both cholesterol and fatty acid metabolism were directly activated by SREBPs in studies performed in cultured cells (Ref. 19Magaña M.M. Koo S.-H. Towle H.C. Osborne T.F. J. Biol. Chem. 2000; 275: 4726-4733Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, and references therein). Even genes of fatty acid metabolism appear to be activated through SRE recognition and not through E-boxes even though SREBPs are capable of binding and activating promoters containing E-boxes in transient transfection studies (3Tontonoz P. Kim J.B. Graves R.A. Spiegelman B.M. Mol. Cell. Biol. 1993; 13: 4753-4759Crossref PubMed Scopus (534) Google Scholar, 18Athanikar J.N. Osborne T.F. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 4935-4940Crossref PubMed Scopus (52) Google Scholar). As more promoters are analyzed in sufficient detail, it is possible some genes will be regulated through SREBP binding to E-boxes. This could link gene activation with different regulatory signals that are relayed through multiple E-box-binding proteins including SREBPs. Studies evaluating SREBP expression in response to dietary and genetic manipulation in animals have provided additional strong evidence that SREBPs are fundamentally involved in both lipogenesis and cholesterol homeostasis (20Shimano H. Horton J.D. Hammer R.E. Shimomura I. Brown M.S. Goldstein J.L. J. Clin. Invest. 1996; 98: 1575-1584Crossref PubMed Scopus (698) Google Scholar, 21Shimano H. Shimomura L. Hammer R.E. Herz J. Brown M.S. Goldstein J.L. Horton J.D. J. Clin. Invest. 1997; 100: 2115-2124Crossref PubMed Scopus (353) Google Scholar, 22Horton J.D. Shimomura I. Brown M.S. Hammer R.E. Goldstein J.L. Shimano H. J. Clin. Invest. 1998; 101: 2331-2339Crossref PubMed Google Scholar, 23Horton J.D. Bashmakov Y. Shimomura I. Shimano H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5987-5992Crossref PubMed Scopus (537) Google Scholar, 24Pai J. Guryev O. Brown M.S. Goldstein J.L. J. Biol. Chem. 1998; 273: 26138-26148Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar, 25Kim J.B. Sarraf P. Wright M. Yao K.M. Mueller E. Solanes G. Lowell B.B. Spiegelman B.M. J. Clin. Invest. 1998; 101: 1-9Crossref PubMed Scopus (612) Google Scholar, 26Shimomura I. Shimano H. Korn B.S. Bashmakov Y. Horton J.D. J. Biol. Chem. 1998; 273: 35299-35306Abstract Full Text Full Text PDF PubMed Scopus (318) Google Scholar, 27Shimano H. Yahagi N. Amemiya-Kudo M. Hasty A.H. Osuga J. Tamura Y. Shionoiri F. Iizuka Y. Ohashi K. Harada K. Gotoda T. Ishibashi S. Yamada N. J. Biol. Chem. 1999; 274: 35832-35839Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar) (Fig.2). Specific analyses of individual isoforms suggest SREBP-1 may be selectively involved in activation of genes involved in fatty acid metabolism and de novolipogenesis whereas SREBP-2 may be more selective for genes involved directly in cholesterol homeostasis (22Horton J.D. Shimomura I. Brown M.S. Hammer R.E. Goldstein J.L. Shimano H. J. Clin. Invest. 1998; 101: 2331-2339Crossref PubMed Google Scholar, 24Pai J. Guryev O. Brown M.S. Goldstein J.L. J. Biol. Chem. 1998; 273: 26138-26148Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar). Moreover, aberrant expression of SREBPs in mice resulted in metabolic syndromes with physiologic effects similar to specific disorders of lipid metabolism in humans (28Shimomura I. Hammer R.E. Richardson J.A. Ikemoto S. Bashmakov Y. Goldstein J.L. Brown M.S. Genes Dev. 1998; 12: 3182-3194Crossref PubMed Scopus (679) Google Scholar). Also, overexpression of SREBP-1 and -2 has been documented in livers and adipose tissue of the leptin-deficient ob/ob mouse or obese Zucker rat, respectively (29Shimomura I. Bashmakov Y. Horton J.D. J. Biol. Chem. 1999; 274: 30028-30032Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar, 30Boizard M. LeLiepvre X. Lemarchand P. Foufelle F. Ferre P. Dugail I. J. Biol. Chem. 1998; 273: 29164-29171Abstract Full Text Full Text PDF PubMed Scopus (109) Google Scholar). Addition of excess cholesterol resulted in the inhibition of processing for membrane-bound precursor forms of both SREBP-1 and -2 in experiments performed in both animals (31Shimomura I. Bashmakov Y. Shimano H. Horton J.D. Goldstein J.L. Brown M.S. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12354-12359Crossref PubMed Scopus (126) Google Scholar) and cultured cells (32Hua X. Sakai J. Brown M.S. Goldstein J.L. J. Biol. Chem. 1996; 271: 10379-10384Abstract Full Text Full Text PDF PubMed Scopus (170) Google Scholar). However, when hamsters were fed a diet supplemented with a bile acid-binding resin and a 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitor to short-circuit liver cholesterol regulation and set up a pseudo-low cholesterol environment, both expression and proteolytic activation of SREBP-2 were increased (33Sheng Z. Otani H. Brown M.S. Goldstein J.L. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 935-938Crossref PubMed Scopus (279) Google Scholar). In contrast, expression of SREBP-1 was not altered nor was processing of membrane-bound SREBP-1 accelerated in these animals. In fact, SREBP-1 processing was inhibited (33Sheng Z. Otani H. Brown M.S. Goldstein J.L. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 935-938Crossref PubMed Scopus (279) Google Scholar). Overexpression of SREBP-1a in cultured cells or animal livers resulted in activation of genes of cholesterol and fatty acid metabolism (20Shimano H. Horton J.D. Hammer R.E. Shimomura I. Brown M.S. Goldstein J.L. J. Clin. Invest. 1996; 98: 1575-1584Crossref PubMed Scopus (698) Google Scholar). Interestingly, in the animal studies this was associated with an increase in hepatic levels of cholesterol and triglycerides, but serum levels were largely unaffected. Crossing these animals with LDL receptor knockout animals resulted in a dramatic increase in circulating levels of cholesterol and triglycerides (34Horton J.D. Shimano H. Hamilton R.L. Brown M.S. Goldstein J.L. J. Clin. Invest. 1999; 103: 1067-1076Crossref PubMed Scopus (158) Google Scholar). Thus, lack of accumulation of serum lipids in the SREBP-1a overexpressing strain was likely due to unregulated expression of hepatic LDL receptors mediated by the overexpression of SREBP-1a. Similar overexpression of SREBP-2 also resulted in accumulation of both lipid classes, although there was more cholesterol relative to fatty acids in both cultured cells and livers than was found in SREBP-1a overexpressors (22Horton J.D. Shimomura I. Brown M.S. Hammer R.E. Goldstein J.L. Shimano H. J. Clin. Invest. 1998; 101: 2331-2339Crossref PubMed Google Scholar, 24Pai J. Guryev O. Brown M.S. Goldstein J.L. J. Biol. Chem. 1998; 273: 26138-26148Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar). Because the activation potential of SREBP-1c is significantly lower than that of either SREBP-1a or SREBP-2 (9Shimano H. Horton J.D. Shimomura L. Hammer R.E. Brown M.S. Goldstein J.L. J. Clin. Invest. 1997; 99: 846-854Crossref PubMed Scopus (684) Google Scholar), its overexpression in liver resulted in a much lower level of activation for genes of both fatty acid and cholesterol metabolism and a correspondingly lower level of accumulation of fatty acids and cholesterol (22Horton J.D. Shimomura I. Brown M.S. Hammer R.E. Goldstein J.L. Shimano H. J. Clin. Invest. 1998; 101: 2331-2339Crossref PubMed Google Scholar, 24Pai J. Guryev O. Brown M.S. Goldstein J.L. J. Biol. Chem. 1998; 273: 26138-26148Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar, 34Horton J.D. Shimano H. Hamilton R.L. Brown M.S. Goldstein J.L. J. Clin. Invest. 1999; 103: 1067-1076Crossref PubMed Scopus (158) Google Scholar). Selective overexpression of SREBP-1c in cultured preadipocytes activated genes involved in fat cell differentiation and lipid accumulation (35Kim J.B. Spiegelman B.M. Genes Dev. 1996; 10: 1096-1107Crossref PubMed Scopus (846) Google Scholar). A more specific role for SREBPs in glucose homeostasis and fat metabolism was first provided by studies in which SREBP-1c was overexpressed in adipose cells (28Shimomura I. Hammer R.E. Richardson J.A. Ikemoto S. Bashmakov Y. Goldstein J.L. Brown M.S. Genes Dev. 1998; 12: 3182-3194Crossref PubMed Scopus (679) Google Scholar). These animals developed insulin-resistant hyperglycemia and a fatty liver and accumulated high levels of serum triglycerides, signs reminiscent of the human disorder congenital generalized lipodystrophy. Two follow-up studies further support a key role for SREBP-1c in insulin action. In one report it was noted that these same lipodystrophic animals also had very low levels of serum leptin and leptin administration resulted in a reversal of the insulin resistance (36Shimomura I. Hammer R.E. Ikemoto S. Goldstein J.L. Brown M.S. Nature. 1999; 401: 73-76Crossref PubMed Scopus (859) Google Scholar). In a related study, it was demonstrated that SREBP-1c mRNA levels decreased in rats treated with streptozotocin to induce diabetes and insulin administration reversed this effect (37Shimomura I. Bashmakov Y. Ikemoto S. Horton J.D. Brown M.S. Goldstein J.L. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13656-13661Crossref PubMed Scopus (627) Google Scholar). Feeding a high carbohydrate diet to rodents after a period of fasting resulted in a significant activation of the entire lipogenic program, which is a signature insulin response (38Towle H.C. Kaytor E.N. Shih H.M. Annu. Rev. Nutr. 1997; 17: 405-433Crossref PubMed Scopus (248) Google Scholar). SREBP-1c mRNA expression was activated during this fasting/refeeding regimen in normal animals (23Horton J.D. Bashmakov Y. Shimomura I. Shimano H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5987-5992Crossref PubMed Scopus (537) Google Scholar, 25Kim J.B. Sarraf P. Wright M. Yao K.M. Mueller E. Solanes G. Lowell B.B. Spiegelman B.M. J. Clin. Invest. 1998; 101: 1-9Crossref PubMed Scopus (612) Google Scholar). Also, overexpression of SREBP-1c in liver prevented the down-regulation of lipogenic genes during fasting (23Horton J.D. Bashmakov Y. Shimomura I. Shimano H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5987-5992Crossref PubMed Scopus (537) Google Scholar). Additionally, re-activation of the hepatic lipogenic program during the feeding phase was not observed in animals in which the SREBP-1 gene was disrupted even though the SREBP-2 gene was expressed at normal levels (27Shimano H. Yahagi N. Amemiya-Kudo M. Hasty A.H. Osuga J. Tamura Y. Shionoiri F. Iizuka Y. Ohashi K. Harada K. Gotoda T. Ishibashi S. Yamada N. J. Biol. Chem. 1999; 274: 35832-35839Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar). These studies provide compelling evidence that SREBP-1c is a key transcriptional activator for early events in the initiation of lipogenesis. Interestingly, the SREBP-1 gene in these knockout animals produced a truncated mutant mRNA, which was still induced normally during the refeeding stage (27Shimano H. Yahagi N. Amemiya-Kudo M. Hasty A.H. Osuga J. Tamura Y. Shionoiri F. Iizuka Y. Ohashi K. Harada K. Gotoda T. Ishibashi S. Yamada N. J. Biol. Chem. 1999; 274: 35832-35839Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar). The ubiquitous bHLHLZ proteins USF1 and USF2 were proposed to be involved in the fasting/refeeding response because FAS mRNA was expressed at reduced levels during the refeeding phase in animals where either USF1 or USF2 genes were inactivated by homologous recombination (39Casado M. Vallett V.S. Kahn A. Vaulont S. J. Biol. Chem. 1999; 274: 2009-2013Abstract Full Text Full Text PDF PubMed Scopus (120) Google Scholar). Exactly how SREBP-1c and USF proteins might both be involved in the fasting and refeeding response is unknown at present, but it is possible that USF functions as a SREBP-1 co-regulator similar to Sp1 in the LDL receptor promoter (40Bennett M.K. Osborne T.F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 6340-6344Crossref PubMed Scopus (101) Google Scholar). SREBP-1 is a more likely primary target of insulin because it is subject to multiple levels of regulation by hormones and nutrients that affect fuel homeostasis, whereas USF levels are largely unchanged by these same manipulations (27Shimano H. Yahagi N. Amemiya-Kudo M. Hasty A.H. Osuga J. Tamura Y. Shionoiri F. Iizuka Y. Ohashi K. Harada K. Gotoda T. Ishibashi S. Yamada N. J. Biol. Chem. 1999; 274: 35832-35839Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar). In freshly isolated hepatocytes, SREBP-1c mRNA was activated by insulin (41Foretz M. Pacot C. Dugail I. Lemarchand P. Guichard C. LeLiepvre X. Berthelier-Lubrano B. Spiegelman B. Kim J.B. Ferre P. Foufelle F. Mol. Cell. Biol. 1999; 19: 3760-3768Crossref PubMed Scopus (452) Google Scholar). However, induction of mRNA for a lipogenic gene such as FAS required simultaneous addition of insulin and a high level of glucose. A recombinant adenovirus expressing a dominant negative version of SREBP-1c prevented induction of FAS mRNA by insulin and glucose. These investigators also demonstrated that glucagon and cAMP decrease SREBP-1c mRNA levels. The antagonistic effects of insulin/glucose and glucagon/cAMP on SREBP-1c mRNA levels are consistent with the known effects of the two opposing hormone systems on gene expression and metabolism. Using similar methods, Foretz et al. (42Foretz M. Guichard C. Ferre P. Foufelle F. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 12737-12742Crossref PubMed Scopus (594) Google Scholar) also showed that the insulin-induced expression of the glucokinase gene is mediated through a mechanism requiring SREBP-1c as well. Thus, when these studies are taken together with the results from fasting/refeeding and gene-targeting studies mentioned above, all of the data strongly indicate SREBP-1c is an important early mediator in the pathway of insulin action in liver. Also, because the defective SREBP-1c mRNA was still increased by fasting and refeeding in the SREBP-1 knockout animals but the activation of the classic lipogenic program was abolished, a model can be envisaged where insulin activates SREBP-1c, which subsequently activates downstream metabolic events. It is also tempting to speculate that another transcriptional regulatory protein that activates the SREBP-1c promoter is a direct target of insulin. Because insulin treatment of primary hepatocytes increased SREBP-1c mRNA but activation of FAS gene expression required both insulin and high glucose, SREBP-1c is probably activated at both the transcriptional and post-transcriptional levels by insulin. Evidence for activation of SREBP-1c protein activity by insulin was noted (25Kim J.B. Sarraf P. Wright M. Yao K.M. Mueller E. Solanes G. Lowell B.B. Spiegelman B.M. J. Clin. Invest. 1998; 101: 1-9Crossref PubMed Scopus (612) Google Scholar), and SREBP activity was stimulated in cultured cells through a mitogen-activated protein kinase signaling pathway (43Kotzka J. Müller-Wieland D. Koponen A. Njamen D. Kremer L. Roth G. Munck M. Knobel B. Krone W. Biochem. Biophys. Res. Commun. 1998; 249: 375-379Crossref PubMed Scopus (72) Google Scholar). A better understanding of exactly how SREBP expression and activity are modulated by insulin signaling and dietary changes is clearly an important area that requires further study. Because SREBPs are regulated directly by cholesterol and they are involved in both cholesterol and fatty acid metabolism, it was important to determine whether they are directly regulated by fatty acids as well. The addition of oleic acid and other longer chain unsaturated fatty acids inhibited sterol regulatory element-mediated transcription and decreased processing of membrane-bound SREBP-1 and -2 in cultured cells (44Worgall T.S. Sturley S.L. Seo T. Osborne T.F. Deckelbaum R.J. J. Biol. Chem. 1998; 273: 25537-25540Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar, 45Thewke D.P. Panini S.R. Sinensky M. J. Biol. Chem. 1998; 273: 21402-21407Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar). A series of reports followed that evaluated SREBP regulation by different types of fatty acids in animal feeding studies. Three studies reported SREBP-1 mRNA expression was significantly suppressed when diets were supplemented with specific polyunsaturated fatty acids (PUFA) or fish oil (46Xu J. Nakamura M.T. Cho H.P. Clarke S.D. J. Biol. Chem. 1991; 274: 23577-23583Abstract Full Text Full Text PDF Scopus (403) Google Scholar, 47Kim H.-J. Takahashi M. Ezaki O. J. Biol. Chem. 1999; 274: 25892-25898Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar, 48Mater M. Thelen A.P. Pan D.A. Jump D.B. J. Biol. Chem. 1999; 274: 32725-32732Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar). One of these studies provided further evidence that SREBP-1 mRNA stability was decreased in livers of PUFA-fed animals (46Xu J. Nakamura M.T. Cho H.P. Clarke S.D. J. Biol. Chem. 1991; 274: 23577-23583Abstract Full Text Full Text PDF Scopus (403) Google Scholar). In a separate animal study SREBP-1 mRNA was only slightly affected, but the level of processed SREBP-1 protein was significantly reduced when PUFA were added to the diet (49Yahagi N. Shimano H. Hasty A.H. Amemiya-Kudo M. Okazaki H. Tamura Y. Iizuka Y. Shionoiri F. Ohashi K. Osuga J. Harada K. Gotoda T. Nagai R. Ishibashi S. Yamada N. J. Biol. Chem. 1999; 274: 35840-35844Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar). SREBP-2 processing in these studies was not altered. There were significant differences in feeding regimens, dietary composition (other than the fatty acid), and choice of animal model that are likely to account for differences reported in these separate studies. Therefore, it is likely that SREBP-1 mRNA levels and protein processing are both affected by fatty acids. Since their identification as bHLHLZ transcriptional regulatory proteins in 1993, the SREBPs have been shown to possess highly unique functional characteristics that define them as key regulators of nutritional homeostasis. There is compelling evidence of a key role for coordinate regulation of fatty acid and cholesterol metabolism through SREBP proteins (Fig. 2). SREBP-1c gene expression and protein activity are both directly subject to significant regulation by dietary and hormonal factors, and SREBP-1c is likely an important mediator of insulin action in the liver. The physiological role of SREBP-1a is less clear, but because it is a much more potent activator of gene expression than 1c, it is probably required to ensure that cells with a relatively high need for cholesterol and fatty acids can activate synthesis to maximal levels to keep up with the demand. The available evidence suggests SREBP-2 may be selectively involved in cholesterol metabolism. An important question is exactly how the different SREBPs activate genes of cholesterol or fatty acid metabolism in preferential ways at the level of the individual promoters. The answer is likely to come from studies that carefully analyze how individual SREBP isoforms function to activate key promoters in each pathway. SREBPs are weak activators of gene expression by themselves and function in a synergistic manner with more generic transcriptional co-regulatory proteins such as Sp1, nuclear factor Y, and CREB/ATF (40Bennett M.K. Osborne T.F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 6340-6344Crossref PubMed Scopus (101) Google Scholar). The identity of generic SREBP co-regulator(s) and the position of their binding site(s) relative to the position and number of SREBP binding sites are quite variable from promoter to promoter (Ref. 19Magaña M.M. Koo S.-H. Towle H.C. Osborne T.F. J. Biol. Chem. 2000; 275: 4726-4733Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, and references therein). Additionally, the requirement for specific co-regulators in the same promoter can be distinct depending on the individual SREBP isoform that is activating gene expression (19Magaña M.M. Koo S.-H. Towle H.C. Osborne T.F. J. Biol. Chem. 2000; 275: 4726-4733Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar). Thus, future studies that unravel the complexities in the SREBP system will provide significant insights into mechanisms of both gene regulation and nutritional homeostasis.