Metabolic Regulation of Gene Expression
Howard C. Towle1
Nutritional factors can influence virtually every aspect of the functioning of the human organism. This influence extends to the realm of gene expression. By influencing gene expression in specific tissues of the organism, nutritional factors help to adapt the organism to changes in the environment. This review will describe the technologies that have emerged for analyzing the effects of nutrition on gene expression, with particular emphasis on the process of gene transcription and its control. Several model systems then will be described in which changes in specific gene expression in response to nutritional factors have been elucidated and efforts to understand the molecular basis of these changes have been made. While this field is still in its infancy, the pace of change suggests that great strides will soon be forthcoming in understanding these important mechanisms and how they may relate to human health and disease.
Gene expression refers to those processes by which the genetic information stored in the DNA is converted into proteins (including enzymes) within the cell. This is a multistep process that involves gene transcription, mRNA processing (capping, splicing, and polyadenylation), and mRNA transport and translation. Each of these processes in turn involves a complex series of biochemical events. Consequently, control of gene expression can be exerted at many different sites in the cell, and in fact, examples of regulation occurring at each step of this pathway have been elucidated. Such complexity of control is undoubtedly critical to the fine tuning of cellular function within the context of the overall organism. Despite this richness in terms of potential sites of control, it is clear that transcriptional regulation and, more specifically, control of transcriptional initiation provide the most commonly employed site for regulation. Given the importance of transcriptional regulation, much attention has been focused on this process in the past decade and much has been learned. Hence, this chapter will focus primarily on regulation occurring at the level of gene transcription.
TRANSCRIPTION AND ITS REGULATION
Transcription of protein-coding genes in all eucaryotes is performed by the enzyme RNA polymerase II. However, this enzyme lacks the inherent ability to recognize the proper site in DNA for initiation of transcription. Rather, RNA polymerase II functions together with a battery of general ''transcription factors" to perform these processes (for review, see Zawel and Reinberg, 1993). To date, six factors essential to the process of promoter selection and initiation have been identified. These factors, together with RNA polymerase II, recognize specific sequences in the DNA helix at the initiation site that are frequently termed the basal promoter site. The most commonly recognized of these signals is the TATA box, a 7 base pair (bp)-conserved sequence occurring approximately 30 bp upstream from (to the 5' end of) the site of initiation in many genes. Other less-conserved signals at the site of initiation also play a role in the site selection. These sequences are recognized by general transcription factors to initiate assembly of the RNA transcription complex.
Although RNA polymerase II in combination with the general transcription factors is competent to recognize and initiate transcription from the basal promoter, the rate of this process is very low. To achieve effective production of mRNA, other transcription factors need to be brought into play. Because these transcription factors only function on a limited set of genes, they are termed specific transcription factors. There are in excess of 100 such factors present in any particular cell. These specific transcription factors function by binding to specific DNA sequences in the vicinity of the basal promoter site. Many times these sites are located immediately upstream of the basal promoter, within 100 to 200 bp of the initiation site. Other times these factors can function by binding at sites that can be several thousand base pairs removed from the initiation site in regions known as enhancers. These sites serve to localize the specific tran-
scription factors in proximity to the basal transcriptional machinery. The factors in turn influence the rate of initiation, presumably by making protein-protein contacts with RNA polymerase II or its associated factors (Choy and Green, 1993). These contacts are thought to alter the stability or kinetics of initiation-complex formation to stimulate transcription. Thus, the rate of initiation from any specific promoter in any particular cell is determined in large part by the qualitative and quantitative nature of the binding sites for specific transcription factors present on the gene and the concentration and activity of the corresponding transcription factors present in the cell. Regulation can occur by controlling the activity of these specific transcription factors. Thus, understanding the control of transcriptional initiation requires the elucidation of the regulatory sequences present for binding specific transcription factors and the elucidation of the nature of the factors that bind to these sites. Technologies for unraveling these components have developed in the past decade and are essential tools in the arsenal of the molecular biologist.
TECHNOLOGIES FOR STUDYING GENE TRANSCRIPTION
The first issue that generally needs to be tackled when analyzing the control of specific gene transcription is the localization of DNA regulatory sequences. These sequences represent the binding sites for specific transcription factors and can be used as tools to help identify these factors. The fundamental assay that has been developed for addressing this question is the transfection assay. To perform a transfection assay, there are several prerequisites. First, one needs a cultured cell that is capable of responding to the nutrient or metabolite of interest. Second, one needs cloned DNA sequences from a gene that is transcriptionally regulated in response to the effector. Third, one needs a means of introducing the cloned DNA into the cultured cell. Finally, one needs an assay to assess promoter activity. This latter need is most often fulfilled by linking the gene of interest to a "reporter gene," which contains the coding sequences for an easily assayed enzyme activity. The most commonly employed reporter genes are chloramphenicol acetyl transferase, ß-galactosidase, and luciferase. In the transfection assay, DNA sequences from the gene of interest containing the potential regulatory regions and basal promoter, as well as a reporter gene, are introduced into a cultured cell. These sequences are then transcribed by the endogenous machinery of the cell. By culturing the transfected cell in the presence of varying concentrations of a specific nutrient or metabolite, the presence of regulatory sequences can be detected by assessing changes in reporter gene activity. In combination with techniques that allow specific mutagenesis of the cloned DNA, the location of the regulatory sequences can be pinpointed. This is generally accomplished first by using deletion mutations to define the boundaries of the region of interest and then by making finer point mutations to locate the critical bases for control.
Once the regulatory sequences of a particular gene have been identified, they can be used to search for a particular transcription factor that binds to this site. A variety of assays are available for examining specific protein-DNA interactions, but of these the electrophoretic mobility shift ("band shift") assay has proven the most useful (Fried and Crothers, 1981). In this assay, a short DNA oligonucleotide is radiolabeled with 32P. This radiolabeled oligonucleotide is mixed with nuclear extracts from the cell or tissue of interest in the presence of excess, unlabeled DNA. The latter serves to react with all nonsequence-specific binding proteins, so that only proteins with high affinity will bind to the radiolabeled probe. The presence of a bound protein is then detected by electrophoresis in a nondenaturing polyacrylamide gel. Under these conditions, the migration of the radiolabeled oligonucleotide will be retarded if bound to a specific protein. In this manner, the presence of a specific transcription factor can often be detected even when it is present in low abundance and purity.
Purification of specific transcription factors can be a technical challenge, given the small quantities of material often present and the very low relative abundance of these proteins. Great strides have been made in this process due to the development of DNA-affinity purification technology (Kadonaga and Tjian, 1986). In this case, a specific oligonucleotide containing the binding site for the transcription factor of interest is coupled to an inert support such as cellulose. Nuclear extracts are incubated with the DNA affinity columns. Again, excess nonspecific DNA is generally added to compete for nonspecific binding of other nuclear proteins with lower affinity. After binding, the transcription factor of interest is eluted by increasing ionic strength. DNA affinity can lead to purifications on the orders of several thousand-fold in a single pass and may make it feasible to recover sufficient quantities of specific transcription factors for biochemical analysis, such as protein sequencing. Alternatively, the genes encoding specific transcription factors can be directly cloned by screening cDNA expression libraries with radiolabeled probes containing the relevant binding site (Singh et al., 1988).
REGULATION OF GENE EXPRESSION BY CHOLESTEROL
All cells require cholesterol for membrane biosynthesis. Cholesterol can be derived from the diet or synthesized by cells. In order to achieve a balance in their cholesterol needs, cells have developed mechanisms to control these two sources (for review see Goldstein and Brown, 1990). Cholesterol uptake is mediated by the low-density lipoprotein receptor (LDL receptor). When cholesterol levels are low, the production of this receptor is induced to provide for more uptake, and when cholesterol levels are high, receptor production is repressed. The rate-limiting enzymes for cholesterol biosynthesis in the cells are HMG-CoA synthase and HMG-CoA reductase. These enzymes are regulated in terms of both their enzymatic activity and production. Cholesterol limitation leads to induction of the synthesis of these two enzymes, while excess choles-
terol leads to repression. In this manner, the cell strives to ensure that adequate supplies of cholesterol are available for membrane biosynthesis and that excess levels that can lead to deleterious cell effects do not accumulate.
Efforts to understand the transcriptional regulation of the genes encoding LDL receptor, HMG-CoA synthase, and HMG-CoA reductase were conducted in the laboratories of Goldstein and Brown (1990). First, using transfection assays, the critical DNA regulatory sequences necessary for control by cholesterol were mapped. The 5'-flanking regions of each of these genes were linked to reporter genes and introduced into Chinese hamster ovary fibroblast cells. Cells were maintained in the absence of cholesterol or in the presence of exogenous cholesterol. After 48 hours, cells maintained in the absence of cholesterol were found to have markedly higher levels of reporter gene activity than cells maintained in the presence of cholesterol. This result implied that the regulatory sequences were present in the cloned 5'-flanking regions of each gene. By mutational analysis, the regulatory sequences of each gene were mapped and compared. A specific DNA motif with the sequence (5')CACCCCAC was found to be present in the regulatory sequences of each of the cholesterol metabolizing genes. Goldstein and Brown (1990) proposed that this motif, which they designated the SRE-1 for sterol response element, served as the binding site for a specific transcription factor involved in the coordinate control of these three genes. This factor presumably would be activated under conditions of low cholesterol to stimulate transcription of the corresponding genes.
To further analyze this system, the specific transcription factor recognizing the SRE-1 from the LDL receptor gene was purified to homogeneity (Wang et al., 1993). This purification took advantage of DNA affinity chromatography using the SRE-1 containing oligonucleotide. The purified factor was designated SREBP-1 for SRE-binding protein. A highly homologous gene, designated SREBP-2, was subsequently identified (Hua et al., 1993). Several properties suggest that these genes are critical for mediating the cholesterol regulatory pathway. First, binding of SREBP to SRE-1 oligonucleotides containing mutations of the binding site correlated with functional activity of the mutant binding sites (Briggs et al., 1993). Mutations that interfered with activity in the transfection assay blocked SREBP-1 binding in vitro , whereas mutations that did not interfere with activity did not block binding. Second, introducing a plasmid into a cultured cell, which led to overexpression of SREBP, led to increased promoter activity from cotransfected DNA containing the SRE-1 element (Sato et al., 1994). Third, a sterol-resistant mutant cell line was found to have a defect in the gene encoding SREBP-2 (Yang et al., 1994).
Identification of the specific transcription factor and its DNA binding site led to efforts to understand how the activity of this factor is regulated by cellular cholesterol levels (Wang et al., 1994). Cloning of the gene for SREBP-1 revealed that it encoded a protein of 125 kilodaltons (kDa), significantly larger than the 68 kDa SREBP-1 isolated by DNA affinity chromatography. Using specific antibodies to the SREBP-1, the 125-kDa form was found to be localized
in the endoplasmic reticulum of the cell as an integral membrane-bound protein. The 68-kDa nuclear form represented the amino (N)-terminal segment of this larger precursor form. Control by cholesterol involved the cleavage of the 68-kDa N-terminal fragment from its endoplasmic reticulum precursor and subsequent nuclear localization of the active fragment. While it is unclear how cholesterol regulates this process, it is tempting to speculate that cholesterol as a normal membrane substituent may influence the properties of the protease involved in cleavage of the SREBP-precursor. In this manner, the levels of intracellular cholesterol can be directly linked to formation of the transcription factor involved in controlling intracellular cholesterol levels.
Many questions remain to be answered but are now experimentally tractable given the progress in this area. Is cholesterol itself a direct mediator of the control pathway, or does it need to be metabolized to an active metabolite? In cells, the oxysterol 25-hydroxycholesterol is more potent than cholesterol, suggesting that a metabolite of cholesterol may be the mediator. What is the nature of the protease involved in cleavage of the SREBP-1 precursor, and how is its activity regulated? Are other genes that are regulated by cholesterol controlled through the same pathway? Preliminary evidence indicates that the answer to this last question may be "no" and that other regulatory pathways may be important (Osborne, 1991; Spear et al., 1994). It also is known that expression of the genes involved in cholesterol metabolism may be regulated at steps other than transcription, and these pathways are largely unexplored. Clearly, there is much to learn, but given the progress of the past few years, the prognosis for answering these questions is excellent.
REGULATION OF HEPATIC GENE EXPRESSION BY CARBOHYDRATE
A second example of gene transcription that is regulated in response to nutritional factors involves the liver and enzymes involved in triglyceride formation. When mammals are fed a diet high in simple carbohydrates and low in fats, a significant portion of the excess carbohydrate is taken up by the liver and converted to triglycerides. Feeding of such a diet induces a response in the liver that involves both rapid changes in the enzymatic activity of the key rate-limiting enzymes in this pathway and a longer-term induction of the cellular concentration of these enzymes (for review, see Hillgartner et al., 1995). The latter is presumably an adaptive response of the organism and is the focus of this discussion. Enzymes that have been shown to be induced by carbohydrate feeding include enzymes of glycolysis, such as pyruvate kinase; enzymes of fatty acid synthesis, such as acetyl CoA carboxylase and malic enzyme; and enzymes of triglyceride formation, such as glycerol-3-phosphate acyltransferase. In all cases, the induction in enzyme production is due to increased mRNA levels. In several cases, but not all, transcription represents the key step in this regulation (Hillgartner et al., 1995).
The actual intracellular pathway leading to increased transcription in response to carbohydrate feeding is poorly understood. Feeding of a high-carbohydrate diet causes increased glucose metabolism in the liver, as well as increased insulin secretion and decreased glucagon secretion. All of these factors play a role in the induction. Using cultured primary hepatocytes as a model system, an important role of carbohydrate metabolism has been implicated. Comparing hepatocytes cultured in low (5.5 mM) or high (27.5 mM) concentrations of glucose, most of the enzymes that are induced in whole animals fed a high-carbohydrate diet also are induced in the hepatocytes cultured in the presence of high concentrations of glucose. This occurs in the presence of a constant concentration of insulin. Other carbohydrates that can feed into the glycolytic pathway at or above the level of pyruvate also are able to induce enzyme production (Mariash and Oppenheimer, 1983). This has led to the hypothesis that increased carbohydrate metabolism is responsible for initiating an intracellular signaling pathway that coordinately regulates this set of genes. The role of insulin appears to be to facilitate effective carbohydrate metabolism in the cell. In particular, the glucokinase step of glucose metabolism is highly insulin-sensitive in the hepatocyte (Lefrancois-Martinez et al., 1994).
To explore this signaling pathway, Towle and coworkers have attempted first to elucidate the DNA regulatory sequences and specific transcription factors responsible for transcriptional regulation. Two genes have been chosen to compare for this purpose: the liver-type pyruvate kinase (PK) and S14 genes. The latter encodes a polypeptide of unknown physiological function that is expressed in the liver, adipose tissue, and lactating mammary gland, all sites of active fatty acid metabolism (Oppenheimer et al., 1987). S14 mRNA is induced rapidly in the rat (=30 minutes) after feeding a high-carbohydrate meal, and this response is due to changes in gene transcription (Jump et al., 1990). By comparing the induction of the PK and S14 genes, the scientists in this laboratory hoped to identify common components that might be involved in coordinate regulation of this family.
Using transfection assays in primary hepatocytes, it was shown that the 5'-flanking region of either the PK or S14 genes were capable of supporting increased promoter activity in cells maintained in high glucose compared to cells in low glucose (Jacoby et al., 1989; Thompson and Towle, 1991). The regulatory sequences responsible for this effect were mapped by mutagenesis: for the S14 gene, critical sequences were found between -1457 and -1422 upstream from the promoter site (Shih and Towle, 1992, 1994). For the PK gene, the critical regulatory sequences mapped to a region between -172 and -144 (Bergot et al., 1992; Liu et al., 1993). Comparing the two regulatory sequences to each other revealed some significant similarities. In both cases, the binding motif (5')CACGTG is found within the regulatory site in two copies. In both cases, the spacing between the two motifs is 5 bp, and this spacing is critical to control by carbohydrate (Shih et al., 1995). The CACGTG motif is recognized as the core binding site for a family of transcription factors known as the c-myc family. All
members of this family possess a similar DNA binding motif composed of a basic region that contacts the DNA and adjacent helix-loop-helix and leucine zippers motifs involved in dimerization (Kadesch, 1993). Based on this information, it was hypothesized that a member of the c-myc family expressed in liver binds to each of two similar sites oriented on the same side of the DNA helix. These proteins may interact with each other directly or form a contact site for a third component. These factors serve as the end of the signaling cascade that is activated by increased carbohydrate metabolism in the rat. The identity of the factor binding to the regulatory sequences of these two genes is currently unknown and is the target of future investigation.
METABOLITES AS DIRECT EFFECTORS OF TRANSCRIPTION FACTORS
Recent work has established a direct pathway by which nutrients and metabolites can influence gene transcription. This work comes from studies of a large family of genes known as the steroid receptor family. This family includes receptors for a wide variety of hormones that directly enter the cell to elicit their biological activity: steroid hormones, thyroid hormones, retinoic acids, and vitamin D3. In this family, the receptor itself serves as a transcription factor for which the activity is regulated by binding of its ligand (for review see Evans, 1988).
During the cloning of the steroid receptors, a large group of related gene products of unknown physiological function were discovered. These gene products contained sequence homology with the steroid receptors, particularly in the DNA-binding domain, and hence were postulated to function as transcription factors as well. However, no ligand was known for activating these factors. This led to their designation as "orphan receptors" (O'Malley and Conneely, 1992). The hypothesis was proposed that these family members would be activated by yet unidentified ligands. Recently, activators have been found for several of the orphan receptors, and the nature of these activators suggests that intracellular nutrients or metabolites may be the natural ligands.
The peroxisome proliferator-activated receptors (PPARs) represent the best characterization of the orphan receptors. These receptors were first shown to be activated in response to a diverse group of xenobiotic substances known to induce a massive accumulation of peroxisomes in rodent hepatocytes (Issemann and Green, 1990). In addition to peroxisome proliferation, these agents also induce enzymes of the peroxisomal and microsomal fatty acid oxidation systems. This induction has been shown to be due to direct interaction of the PPARs with regulatory sequences in the genes encoding these enzymes, such as acyl-CoA oxidase (Dreyer et al., 1992; Tugwood et al., 1992). Although peroxisome proliferators were first shown to be activators of PPARs, the question arose as to what the natural ligands for these receptors might be. Recent work has shown that natural fatty acids activate PPARs (Keller et al., 1993). This finding is con-
sistent with observations that high dietary fat intake induces the peroxisomal ß-oxidation system. A preference for polyunsaturated fatty acids over monoun-saturated or saturated fatty acids was found. With polyunsaturated fatty acids, activation was observed with concentrations of 50 µM, which is within the range found in blood. To date, the mechanism by which fatty acids activate PPARs is unknown. No direct binding of fatty acids to the PPARs has been demonstrated. It is reasonable to speculate that a product formed by metabolism of fatty acids may be the actual ligand for these receptors. It also is conceivable that fatty acids could act by release of an unknown second messenger. PPARs represent the first group of mammalian transcription factors that are activated by nutrients.
A second group of orphan receptors from the steroid receptor family that appears to be activated in response to a nutrient has been recently identified. In this case, the orphan receptor was found by pharmacologic screening to be activated by farnesol and certain of its metabolites (Forman et al., 1995). Again, activation occurred within the micromolar range for the most active agents, which is thought to be physiological for these compounds. Similar to the PPARs, direct binding of farnesol to the receptor has not been demonstrated, which suggests that a farnesoid-induced metabolite may serve as an authentic ligand for the receptor. The farnesol derivative, farnesyl pyrophosphate, is a key metabolic intermediate as the last common precursor in the mevalonate pathway. This pathway leads to formation of cholesterol, bile acids, dolichol, ubiquinone, steroid and retinoic hormones, and farnesylated proteins. Thus, farnesol stands at a critical step in regulating these biosynthetic pathways, and its potential role as a mediator of gene transcription is intriguing. To date, no target genes for the farnesol receptor have been identified, likely due to its recent discovery.
AUTHOR'S CONCLUSIONS AND RECOMMENDATIONS
The discovery of nutrients as regulators of gene expression in mammals is clearly a rapidly emerging field. Although such regulation was predicted over 20 years ago (Tomkins, 1975), the technology to explore this important role has developed only in the more recent past. Using these techniques, the principal components in the regulatory systems—the DNA regulatory sites and specific transcription factors—are beginning to be identified. Work to explore the mechanisms by which these components function to control gene expression will largely await identification of these key components. Clearly, this work is at a basic stage at present, and its applicability to more physiological questions of health and disease must await further advances in knowledge.
The impact of nutrients on gene transcription is likely to be imparted not in terms of short-term (seconds to minutes) control, but rather in longer-term (hours to days) adaptive responses. By changing the concentration of key enzymes and proteins that are involved in cellular processes such as metabolism,
these effects will presumably allow the organism to operate more efficiently in the face of changing nutritional status. The impact of such changes is difficult to measure, as they occur within the context of changes occurring at many levels in the body. However, the conservation of these mechanisms across many species lines argues for a significant role. There is much to learn in this area in the future. Continued research efforts along these lines will help to address many of these unresolved questions.
Bergot, M-O., M-J.M. Diaz-Guerra, N. Puzenat, M. Raymondjean, and A. Kahn 1992 Cis-regulation of the L-type pyruvate kinase gene promoter by glucose, insulin and cyclic AMP. Nucleic Acids Res. 20:1871–1877.
Briggs, M.R., C. Yokoyama, X. Wang, M.S. Brown, and J.L. Goldstein 1993 Nuclear protein that binds sterol regulatory element of low-density lipoprotein receptor promoter. I. Identification of the protein and delineation of its target nucleotide sequence. J. Biol. Chem. 268:14490–14496.
Choy, B., and M.R. Green 1993 Eukaryotic activators function during multiple steps of preinitiation complex assembly. Nature 366:531–536.
Dreyer, C., G. Krey, H. Keller, F. Givel, G. Helftenbein, and W. Wahli 1992 Control of the peroxisomal ß-oxidation pathway by a novel family of nuclear hormone receptors. Cell 68:879–887.
Evans, R.M. 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895.
Forman, B.M., E. Goode, J. Chen, A.E. Oro, D.J. Bradley, T. Perlmann, D.J. Noonan, and L.T. Burka 1995 Identification of a nuclear receptor that is activated by farnesol metabolites. Cell 81:687–693.
Fried, M., and D.M. Crothers 1981 Equilibria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9:6505–6525.
Goldstein, J.L., and M.S. Brown 1990 Regulation of the mevalonate pathway. Nature 343:425–430.
Hillgartner, F.B., L.M. Salati, and A.G. Goodridge 1995 Physiological and molecular mechanisms involved in nutritional regulation of fatty acid synthesis. Physiol. Rev. 75:47–76.
Hua, X., C. Yokoyama, J. Wu, M.R. Briggs, M.S. Brown, J.L. Goldstein, and X. Wang 1993 SREBP-2, a second basic-helix-loop-helix-leucine zipper protein that stimulates transcription by binding to a sterol regulatory element. Proc. Natl. Acad. Sci. USA 90:11603–11607.
Issemann, I., and S. Green 1990 Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 347:645–650.
Jacoby, D.B., N.D. Zilz, and H.C. Towle 1989 Sequences within the 5'-flanking region of the S14 gene confer responsiveness to glucose in primary hepatocytes. J. Biol. Chem. 264:17623–17626.
Jump, D.B., A. Bell, and V. Santiago 1990 Thyroid hormone and dietary carbohydrate interact to regulate rat liver S14 gene transcription and chromatin structure. J. Biol. Chem. 265:3474–3478.
Kadesch, T. 1993 Consequences of heteromeric interactions among helix-loop-helix proteins. Cell Growth Differ. 4:49–55.
Kadonaga, J.T., and R. Tjian 1986 Affinity purification of sequence-specific DNA binding proteins. Proc. Natl. Acad. Sci. U.S.A. 83:5889–5893.
Keller, H., C. Dreyer, J. Medin, A. Mahfoudi, K. Ozato, and W. Wahli 1993 Fatty acids and retinoids control lipid metabolism through activation of peroxisome proliferator-activated receptor-retinoid X receptor heterodimers. Proc. Natl. Acad. Sci. USA 90:2160–2164.
Lefrancois-Martinez, A-M., M-J.M. Diaz-Guerra, V. Vallet, A. Kahn, and B. Antoine 1994 Glucose-dependent regulation of the L-type kinase gene in a hepatoma cell line is independent of insulin and cyclic AMP. FASEB J. 8:89–96.
Liu, Z., K.S. Thompson, and H.C. Towle 1993 Carbohydrate regulation of the rat L-type pyruvate kinase gene requires two nuclear factors: LF-A1 and a member of the c-myc family. J. Biol. Chem. 268:12787–12795.
Mariash, C.N., and J.H. Oppenheimer 1983 Stimulation of malic enzyme formation in hepatocyte culture by metabolites: Evidence favoring a nonglycolytic metabolite as the proximate induction signal. Metabolism 33:545–552.
O'Malley, B.W., and O.M. Conneely 1992 Orphan receptors: In search of a unifying hypothesis for activation. Mol. Endocrinol. 6:1359–1361.
Oppenheimer, J.H., H.L. Schwartz, C.N. Mariash, W.B. Kinlaw, N.C. Wong, and H.C. Freake 1987 Advances in our understanding of thyroid hormone action at the cellular level. Endocr. Rev. 8:288–308.
Osborne, T.F. 1991 Single nucleotide resolution of sterol regulatory region in promoter for 3-hydroxy-3-methylglutaryl Coenzyme A reductase. J. Biol. Chem. 266:13947–13951.
Sato, R., J. Yang, X. Wang, M.J. Evans, Y.K. Ho, J.L. Goldstein, and M.S. Brown 1994 Assignment of the membrane attachment, DNA binding, and transcriptional activation domains of sterol regulatory element-binding protein-1 (SREBP-1). J. Biol. Chem. 269:17267–17273.
Shih, H-M., and H.C. Towle 1992 Definition of the carbohydrate response element of the rat S14 gene. J. Biol. Chem. 267:13222–13228.
1994 Definition of the carbohydrate response element of the rat S14 gene. J. Biol. Chem. 269(12):9380–9387.
Shih, H-M., Z. Liu, and H.C. Towle 1995 Two CACGTG motifs with proper spacing dictate the carbohydrate regulation of hepatic gene transcription. J. Biol. Chem. 270:21991–21997.
Singh, H., J.H. LeBowitz, A.S. Baldwin, Jr., and P.A. Sharp 1988 Molecular cloning of an enhancer binding protein: Isolation by screening of an expression library with a recognition site DNA. Cell 52:415–423.
Spear, D.H., J. Ericsson, S.M. Jackson, and P.A. Edwards 1994 Identification of a 6-base pair element involved in the sterol-mediated transcriptional regulation of farnesyl diphosphate synthase. J. Biol. Chem. 269:25212–25218.
Thompson, K.S., and H.C. Towle 1991 Localization of the carbohydrate response element of the rat L-type pyruvate kinase gene. J. Biol. Chem. 266:8679–8682.
Tomkins, G.M. 1975 The metabolic code. Science 189:760–763.
Tugwood, J., I. Issemann, R.G. Anderson, K.R. Bundell, W.L. McPheat, and S. Green 1992 The mouse peroxisome proliferater activated receptor recognizes a response element in the 5' flanking sequence of the rat acyl CoA oxidase gene. EMBO J 11:433–439.
Wang, X., M.R. Briggs, X. Hua, C. Yokoyama, J.L. Goldstein, and M.S. Brown 1993 Nuclear protein that binds sterol regulatory element of low density liporprotein receptor promoter. II. Purification and characterization. J. Biol. Chem. 268:14497–14504.
Wang, X., R. Sato, M.S. Brown, X. Hua, and J.L. Goldstein 1994 SREBP-1, a membrane-bound transcription factor released by sterol-regulated proteolysis. Cell 77:53–62.
Yang, J., R. Sato, J.L. Goldstein, and M.S. Brown 1994 Sterol-resistant transcription in CHO cells caused by gene rearrangement that truncates SREBP-2. Genes Dev. 8:1910–1919.
Zawel, L., and D. Reinberg 1993 Initiation of transcription by RNA polymerase II: A multistep process. Prog. Nucleic Acids Res. Mol. Biol. 44:67–108.
GUY MILLER: When the primary cells of the liver and all the zone one [periportal] hepatocytes essentially convert to zone three [perivenous] metabolism in the face of a relatively hypoxic environment, they revert from a profoundly oxidative state to a highly glycolytic state, and they are dynamically transformed as this proceeds. Clearly, that influences carbohydrate utilization as a model for what is happening in vivo. Do you have any suggestions for how we would start approaching that from the gene therapy or gene manipulation side of substrate utilization?
HOWARD TOWLE: That is a tough one. You are absolutely right that what we are looking at are hepatocytes, which are fairly homogeneous in their response, and, of course, in the liver there is significant zonation in terms of function, so that perivenous hepatocytes are more active in carbohydrate metabolism and lipogenesis. I really do not know how we would handle that in terms of gene therapy.
DENNIS BIER: Is this inverse effect that you mention between the induction of lipogenic enzymes by carbohydrate and the repression of these enzymes by polyunsaturated fatty acid limited to polyunsaturated fatty acids because acetyl-CoA is an allosteric regulator of pyruvate kinase and that is what happens when you take away glucose?
HOWARD TOWLE: Yes, in fact, it is quite specific for polyunsaturated fatty acids, so that saturated or monosaturated fatty acids do not show that effect at all.
DENNIS BIER: What about acetate or acetyl-CoA?
HOWARD TOWLE: Acetate does not have an effect on the hepatocytes in terms of this response, and I presume that acetate would be converted to acetyl-CoA.
ALLISON YATES: Do you see a difference between omega-3 and omega-6 fatty acids?
HOWARD TOWLE: In terms of the repression of the lipogenic enzyme expression by fatty acids? If I remember correctly, both are capable of repressing, and I am not aware of any studies showing differential actions of one group compared to another.
ROBERT NESHEIM: Thank you very much, Howard. Very interesting techniques here and it is going to be interesting to see what happens over the years. Probably, if I am around 40 years from now, I will learn a new language.