Search PubMed⌕ Search

Biomedical subjects

G I Bell

Publications and source records attributed to G I Bell.

At least 181 records · Page 10Linked to original sources
PubMed1993

A molecular inventory of human pancreatic islets: sequence analysis of 1000 cDNA clones.

The islets of Langerhans play a central role in glucose homeostasis by secreting the polypeptide hormones insulin and glucagon. They are comprised primarily of four endocrine cell types: insulin-secreting beta-cells which represent about 70% of the cells in the islet along with smaller number of cells secreting glucagon, somatostatin and pancreatic polypeptide. Diabetes mellitus results from the specific loss or dysfunction of the beta-cells. Because of the central role of the islets of Langerhans in the regulation of glucose homeostasis, we are preparing a database of genes expressed in this tissue. One thousand cDNA clones randomly isolated from a human pancreatic islet library were partially sequenced yielding 280 kilobases of sequence. Database searches indicated that 397 of the cDNAs represented known human genes or human homologs of genes identified in other species and a further 58 sequences corresponded to expressed sequence tags identified in other tissues or cells (contamination by exocrine pancreatic tissue was estimated to be less than 10%). 545 of the cDNAs were not related to any other sequences in the databases. The islet cDNA collection provides a unique source of genes for genetic studies of diabetes as well as for molecular studies of islet function in normal and diabetic states.

Animals↗
PubMed1993

Sequence, tissue distribution, and functional characterization of the rat fructose transporter GLUT5.

cDNA clones encoding rat GLUT5-small intestinal facilitative hexose transporter were isolated from a jejunum library by cross-hybridization with a human GLUT5 cDNA probe. The cDNA sequence indicates that rat GLUT5 is composed of 502 amino acids and has 81.5% amino acid identity and 87.3% similarity with the sequence of human GLUT5. Expression of synthetic rat GLUT5 mRNA in Xenopus oocytes showed that rat GLUT5 was able to mediate the uptake of fructose and, to a lesser extent, of glucose. RNA blotting studies showed that GLUT5 mRNA was present in rat small intestine, kidney, and brain. Although GLUT5 mRNA is expressed in human testis, adipose tissue, and skeletal muscle, it could not be detected by RNA blotting in these rat tissues. Developmental studies showed low levels of GLUT5 mRNA in rat small intestine and kidney during the prenatal period with a rapid induction of GLUT5 expression occurring postnatally. In situ hybridization studies of GLUT5 mRNA expression in the small intestine revealed differential expression along the crypt-villus axis with the highest levels of mRNA being in the midvillus region. In addition, there was quantitatively more GLUT5 mRNA detected in the proximal as opposed to the distal small intestine.

Amino Acid Sequence↗
PubMed1993

Human thyroxine-binding globulin gene: complete sequence and transcriptional regulation.

T4-binding globulin (TBG) is a glycoprotein of hepatic origin which transports thyroid hormone in serum. To characterize the human TBG (hTBG) gene, we studied its genomic organization, promoter activity, and regulation. To this purpose, we isolated from liver a complete hTBG cDNA clone containing the 5'-untranslated region and localized the transcription start site (TSS). The analysis of genomic clones revealed that the hTBG gene consists of five exons and that its exon-intron organization is similar to that of other members of the serine protease inhibitor family. The first exon (exon 0) is a short noncoding sequence located 1.62 kilobase pairs (kbp) upstream from exon 1. Potential cis-acting transcriptional regulatory elements including a TATA box, a CAAT box, and a hepatocyte nuclear factor-1 binding motif were identified in the upstream region. A reporter gene in which 3.2 kbp of the 5'-flanking region, including exon 0, was inserted upstream of the bacterial chloramphenicol acetyltransferase gene showed significant activity when transfected into a hepatblastoma-derived (HepG2) cell line. The phorbol ester, 12-O-tetradecanoylphorbol-13-acetate, down-regulated the promoter activity by more than 80% and completely inhibited hTBG synthesis, whereas thyroid hormone, glucocorticoid, estrogen, and nicotinic acid had little, if any, effect. A series of 5'-deletions revealed that the fragment -218 to +4 from the TSS had the highest promoter activity, nearly 1000-fold greater than the promoterless chloramphenicol acetyltransferase construct. When nonhepatocyte-derived cell lines (CV-1 and CHO) were tested, promoter activity was reduced by a factor of 100, showing that the promoter works in liver-specific manner. The region -218 to -102 contains liver-specific enhancer elements, since deletion to nucleotide -101 resulted in a profound reduction of the promoter activity in HepG2 cells but not in CV-1 or CHO cells. On the other hand, mutational disruption of the putative hepatocyte nuclear factor-1 site (located 65 bp upstream of the TSS) completely abolished the promoter activity in all cell lines, indicating that this site is absolutely required for the transcription of the hTBG gene.

Amino Acid Sequence↗
PubMed1993

Organization of the human GLUT2 (pancreatic beta-cell and hepatocyte) glucose transporter gene.

The gene encoding the predominant facilitative glucose transporter expressed in pancreatic beta-cells and hepatocytes, termed GLUT2, has been cloned and characterized. The human GLUT2 gene is composed of 11 exons spanning approximately 30 kilobases. The sequence of the promoter region and all exons and adjacent intron regions has been determined and deposited in the GenBank database. Two highly polymorphic simple tandem repeat DNA polymorphisms useful for linkage studies were localized in introns 1 and 4a. In addition, a 168-base pair insertion/deletion polymorphism was identified in intron 3. The characterization of the human GLUT2 gene will facilitate studies of its role in the development of diabetes mellitus.

Amino Acid Sequence↗
PubMed1993

Identification of a simple tandem repeat DNA polymorphism in the human glycogen synthase gene and linkage to five markers on chromosome 19q.

We have identified a simple tandem repeat DNA polymorphism in the human glycogen synthase gene of the form (TG)n. This DNA polymorphism has 10 alleles and a heterozygosity of 0.82 and can be easily typed using the polymerase chain reaction. It has been localized within the framework genetic map of chromosome 19 and is located in the region of the apolipoprotein C-II and histidine-rich calcium-binding protein genes. This DNA polymorphism will facilitate genetic studies of the role of the glycogen synthase gene in the development of insulin resistance and NIDDM.

Base Sequence↗
PubMed1993

Expression of calcium channel mRNAs in rat pancreatic islets and downregulation after glucose infusion.

Recent studies have shown that two different voltage-dependent Ca2+ channels are expressed in pancreatic islets, the beta-cell/neuroendocrine-brain and the cardiac subtypes. The effects of chronic hyperglycemia on the levels in pancreatic islets of the mRNAs encoding the alpha 1-subunits of the beta-cell and cardiac subtype Ca2+ channels were studied in rats made hyperglycemic by infusion of glucose for 48 h. A competitive reverse transcriptase-polymerase chain reaction procedure was used to obtain quantitative data on the levels of these two transcripts in islets obtained from individual rats. The quantitative polymerase chain reaction data indicate that the levels of mRNA encoding the alpha 1-subunit of the beta-cell Ca2+ channel are 2.5-fold greater than those for the cardiac subtype. The levels of beta-cell Ca2+ channel mRNA were 72.9% lower in the glucose-infused animals when compared with the saline-infused animals (P < 0.005) and those of the cardiac channel were 72.1% lower in the animals infused with glucose (P < 0.02). In contrast, glucose infusion resulted in a twofold increase in insulin mRNA levels and did not significantly alter levels of beta-actin mRNA. In situ hybridization studies revealed that the mRNAs for these two Ca2+ channels are expressed at higher levels in normal rat islets than in the surrounding acinar tissue, which suggests that the observed changes in mRNA levels occur within cells of the pancreatic islet. To assess the possible functional consequences of this reduction in expression of mRNA for the Ca2+ channels, the insulin secretory responses of perfused pancreases to the Ca2+ channel agonist Bay K8644 were studied.(ABSTRACT TRUNCATED AT 250 WORDS)

3-Pyridinecarboxylic acid, 1,4-dihydro-2,6-dimethy↗
PubMed1993

Human glucagon-like peptide-1 receptor gene. Localization to chromosome band 6p21 by fluorescence in situ hybridization and linkage of a highly polymorphic simple tandem repeat DNA polymorphism to other markers on chromosome 6.

Glucagon-like peptide-1 is a fragment of proglucagon secreted by intestinal L-cells. It has potent glucose-dependent insulin secretory effects and also suppresses gastric acid secretion in the stomach. The biological actions of GLP-1 are mediated by the GLP-1 receptor, the structure of which has recently been determined. Defects in insulin secretion are a common feature of NIDDM and as such the GLP-1 receptor is a candidate for contributing to the development of this clinically and genetically heterogeneous disorder. As a first step in determining the role of the GLP-1 receptor in the development of NIDDM, we have isolated the human GLP-1 receptor gene and mapped it to chromosome 6, band p21.1, using the technique of fluorescence in situ hybridization. We also identified a simple tandem repeat DNA polymorphism in the human GLP-1 receptor gene of the form (TG)n. This DNA polymorphism has 14 alleles and a heterozygosity of > 0.8. We have used this DNA polymorphism to localize the GLP-1 receptor gene within the genetic map of the short arm of chromosome 6. This DNA polymorphism will facilitate genetic studies of the contribution of the GLP-1 receptor gene to impaired beta-cell function and NIDDM.

Base Sequence↗
PubMed1993

Splice variant of the somatostatin receptor 2 subtype, somatostatin receptor 2B, couples to adenylyl cyclase.

The diverse biological actions of somatostatin (SRIF) are mediated by a family of receptors, of which five have been cloned and characterized. One of the SRIF receptor subtypes, SSTR2, has been shown to exist in two forms. SSTR2A and SSTR2B are 369 and 346 amino acids in size, respectively, and differ in length and amino acid sequence in their intracellularly located carboxyl termini. SSTR2A and SSTR2B are generated by alternative splicing of SSTR2 mRNA. We previously characterized mouse SSTR2A and showed that it could be distinguished from other cloned SRIF receptor subtypes by its high affinity for MK-678 and its lack of coupling to adenylyl cyclase. To determine whether the properties of mouse SSTR2A and SSTR2B differ, we have expressed both in COS-7 cells and characterized their ligand-binding properties and ability to couple to adenylyl cyclase. The two receptors exhibited similar affinities for a number of SSTR2-selective agonists such as MK-678. Pretreatment with SRIF of COS-7 cells expressing each receptor reduced high affinity agonist binding to both SSTR2A and SSTR2B, indicating that both receptors can be regulated. Furthermore, agonist binding to both receptors was reduced by GTP analogs and Na+, indicating that they both associate with G proteins. As shown previously, SSTR2A could not mediate SRIF inhibition of forskolin-stimulated cAMP formation. In contrast, SSTR2B was coupled to adenylyl cyclase and was able to mediate SRIF inhibition of forskolin-stimulated cAMP formation. Thus, SSTR2A and SSTR2B differ in their ability to couple to adenylyl cyclase. Because SSTR2A and SSTR2B differ only in the length and amino acid sequence of their carboxyl termini, these findings imply that the carboxyl-terminal 15 residues of SSTR2B may be involved in coupling this receptor to adenylyl cyclase.

Adenylyl Cyclases↗
PubMed1993

Cloned somatostatin receptors: identification of subtype-selective peptides and demonstration of high affinity binding of linear peptides.

The recent molecular cloning of the genes encoding three somatostatin (SRIF) receptor subtypes has allowed for the individual expression of these receptors in mammalian cells and characterization of their respective pharmacological profiles. In the present study, we have investigated the affinities of a battery of SRIF analogues to bind to SRIF receptor subtypes SSTR1 (cloned somatostatin complex), SSTR2, and SSTR3, as well as their abilities to inhibit the release of growth hormone from anterior pituitary cells in vitro. We labeled SSTR1 and SSTR3 receptors expressed in Chinese hamster ovary and COS-1 cells, respectively, with the metabolically stable SRIF analogue 125I-CGP 23996. SSTR2 receptors expressed in Chinese hamster ovary cells were labeled with the SSTR2-specific radioligand 125I-MK-678. Inhibition studies were performed using SRIF analogues of differing structures, including hexapeptide analogues similar to MK-678, octapeptide analogues similar to SMS 201-995, pentapeptide analogues similar to c[Ahep-Phe-D-Trp-Lys-Thr(Bzl)] (SA), and linear SRIF analogues. SSTR1 bound SRIF and SRIF-28 with high affinity and the peptide SA and its structural analogues with low affinity. The hexapeptides did not interact with SSTR1 at concentrations as high as 1 microM, and only a few of the octapeptides or linear peptides bound, with very low affinities. In contrast, 125I-MK-678 binding to SSTR2 was potently inhibited by the hexapeptides, octapeptides, and some of the linear compounds, whereas SA and its analogues did not bind to SSTR2. The potencies of the various SRIF agonists to inhibit growth hormone release in vitro was highly correlated with their potencies to inhibit radioligand binding to SSTR2, but not to SSTR1 or SSTR3. SSTR3 bound analogues of each class but with moderate to low affinities, with the exception of several linear peptides and one of the octapeptides. We report for the first time the binding affinities of linear analogues of SRIF, some of which display subnanomolar affinities and are highly selective for SRIF receptor subtypes. Most importantly, these studies identify several peptide analogues that are highly potent, specific, and selective for individual subtypes of SRIF receptors. Such information, coupled with the knowledge of the distribution of these receptor subtypes in normal and pathological tissues, will be critical for more specific experimental and therapeutic interventions.

Amino Acid Sequence↗
PubMed1993

Mutation of an aspartate at residue 89 in somatostatin receptor subtype 2 prevents Na+ regulation of agonist binding but does not alter receptor-G protein association.

Sodium ions have been shown to reduce the binding of agonists to a number of G protein-linked receptors. They are believed to do so by interacting with aspartate residues in the second membrane-spanning region of these receptors to cause G protein uncoupling, resulting in a diminished affinity of the receptors for agonists. To investigate Na+ regulation of agonist binding to somatostatin receptors, Na+ was tested for its effect on the binding of agonists to cloned somatostatin receptor type 1 (SSTR1) and somatostatin receptor type 2 (SSTR2) stably expressed in Chinese hamster ovary cells. Na+ reduced agonist binding to SSTR2 but not to SSTR1. Because high affinity agonist binding to SSTR1 does not depend on G protein coupling but agonist binding to SSTR2 is reduced by guanosine-5'-(beta, gamma-imido)triphosphate and pertussis toxin treatment, the selective Na+ effect on SSTR2 is consistent with previous findings with other receptors showing that Na+ uncouples receptors from G proteins, thereby reducing the affinity of the receptors for agonists. Conversion of Asp89 to Asn89 in SSTR2 resulted in a mutant receptor whose affinity for agonists was not altered by Na+, indicating that Asp89 is involved in mediating the effects of Na+ on agonist binding to SSTR2. However, the affinities of the mutant and wild-type receptors for somatostatin were the same, and both guanosine-5'-O-(gamma-thio)triphosphate and pertussis toxin treatment reduced agonist binding to the mutant and wild-type receptors. These findings differ from the results of similar mutagenesis studies on other G protein-linked receptors, in that the mutant and wild-type SSTR2 forms associate with G proteins in similar ways. These results indicate that Asp89 acts in a novel manner to regulate agonist binding and G protein interaction with SSTR2.

Animals↗
PubMed1993

Characterization of cloned somatostatin receptors SSTR4 and SSTR5.

The recent molecular cloning of the genes and cDNAs encoding multiple somatostatin (SRIF) receptor subtypes has allowed for the individual expression of these receptors in mammalian cells and characterization of their respective pharmacological profiles. Previously, we fully described and compared the pharmacological properties of the first three SRIF receptor subtypes, SRIF receptor type (SSTR)1, SSTR2, and SSTR3. In the present study, we have investigated the properties of the newly cloned SRIF receptor subtypes SSTR4 and SSTR5 with regard to pharmacological profiles, the regulation of high affinity agonist binding to these receptors by stable GTP analogues, Na+, or prior exposure to agonists, and the inhibition of forskolin-stimulated cAMP accumulation mediated by these receptors. We labeled SSTR4 and SSTR5 expressed in Chinese hamster ovary (CHO-K1) and COS-1 cells, respectively, with the metabolically stable SRIF analogue 125I-CGP 23996. Radioligand binding competition studies were performed using SRIF analogues of differing structures, including hexapeptide analogues similar to MK-678, octapeptide analogues similar to SMS 201-995, pentapeptide analogues similar to c[Ahep-Phe-D-Trp-Lys-Thr(Bzl)], and linear SRIF analogues. SSTR4 bound compounds in all structural classes with high to moderate affinities, and several compounds were identified that are > 100-fold selective for SSTR4, compared with the other cloned SRIF receptors, including the linear SRIF analogue BIM-23052 and the CGP 23996-like SRIF analogue L-362,855. In contrast, SSTR5 bound very few SRIF analogues with high affinity. Both receptors could be regulated by prior exposure to agonist. In addition, agonist binding to SSTR4 was reduced by stable GTP analogues, Na+, and pertussis toxin, but agonist binding to SSTR5 was not affected by these treatments. SSTR4 is efficiently coupled to the inhibition of adenylyl cyclase activity, whereas SSTR5 appears not to couple to this cellular effector system. Such differences between the cloned SRIF receptors provide useful strategies for identifying regions of these receptor subtypes that may be involved in ligand-binding specificities and G protein and cellular effector system coupling. The identification of subtype-selective SRIF analogues may lead to more specific therapeutic interventions.

Amino Acid Sequence↗
PubMed1992

Mammalian facilitative glucose transporters: evidence for similar substrate recognition sites in functionally monomeric proteins.

Four facilitative glucose transporters isoforms, GLUT1/erythrocyte, GLUT2/liver, GLUT3/brain, and GLUT4/muscle-fat, as well as chimeric transporter proteins were expressed in Xenopus oocytes, and their properties were studied. The relative Km's of the transporters for 2-deoxyglucose were GLUT3 (Km = 1.8 mM) > GLUT4 (Km = 4.6 mM) > GLUT1 (Km = 6.9 mM) > GLUT2 (Km = 17.1 mM). In a similar fashion, the uptake of 2-deoxyglucose by GLUT1-, GLUT2-, and GLUT3-expressing oocytes was inhibited by a series of unlabeled hexoses and pentoses and by cytochalasin B in a similar hierarchical order. To determine if the functional unit of the glucose transporter was a monomer or higher-order multimer, the high-affinity transporter GLUT3 was coexpressed with either the low-affinity GLUT2 or a GLUT3 mutant which contained a transport inactivating Trp410-->Leu substitution. In oocytes expressing both GLUT2 and GLUT3, the transport activity associated with each transporter isoform could be distinguished kinetically. Similarly, there was no alteration in the kinetic parameters of GLUT3, or the ability of glucose or cytochalasin B to inhibit 2-deoxyglucose uptake, when coexpressed with up to a 3-fold greater amount of functionally inactive mutant of GLUT3. These studies suggest that the family of glucose transporters have similar binding sites which may be in the form of a functional monomeric unit when expressed in Xenopus oocytes.

Animals↗
PubMed1992

Cloning of a novel somatostatin receptor, SSTR3, coupled to adenylylcyclase.

The gene encoding a novel mouse somatostatin receptor termed mSSTR3 was isolated and characterized. The sequence of mSSTR3 shows 46 and 47% identity with mSSTR1 and mSSTR2, respectively. mSSTR3 binds somatostatin-14 and somatostatin-28 with high affinity, but shows very low affinity for the somatostatin analogs MK-678 and SMS-201-995. In addition, mSSTR3 is coupled to pertussis toxin-sensitive G proteins and mediates somatostatin inhibition of forskolin-stimulated and dopamine D1 receptor-stimulated cAMP formation, indicating that it is coupled to adenylylcyclase. The pharmacological properties of mSSTR3 and its ability to couple with adenylylcyclase distinguish SSTR3 from the other cloned somatostatin receptors and indicates that it mediates biological functions different from SSTR1 or SSTR2. In situ hybridization indicates that SSTR3 mRNA is widely distributed in the mouse brain, and its expression in the nucleus of the lateral olfactory tract and in the piriform cortex, the primary olfactory cortex in the rodent brain, suggests that SSTR3 may participate in the processing and modulation of primary sensory information.

Adenylyl Cyclases↗
PubMed1992

Human Oct3 gene family: cDNA sequences, alternative splicing, gene organization, chromosomal location, and expression at low levels in adult tissues.

Transcription factors containing the POU-domain have been shown to be important regulators of tissue-specific gene expression in the pituitary and lymphoid cells. Using a polymerase chain reaction (PCR)-based strategy, we have searched for similar factors that may be expressed in adult human pancreatic islets. This approach resulted in the amplification of sequences encoding the octamer binding proteins Oct1 and Oct3 (also called Oct4). The isolation of cDNAs encoding Oct3 revealed the expression of two isoforms of this transcription factor termed Oct3A and Oct3B that are generated by alternative splicing. Human Oct3A and Oct3B are composed of 360 and 265 amino acids, respectively, of which the 225 amino acids at the COOH-termini are identical. The sequence of human Oct3A shows 87% amino acid identity with mouse Oct3. Reverse-transcriptase PCR showed low levels of expression of both Oct3A and Oct3B mRNA in all adult human tissues examined. We also isolated and characterized the human Oct3 gene (OTF3) and a related gene, OTF3C. The human Oct3 gene, localized to human chromosome 6 in the region of the MHC complex, spans about 7 kb and consists of five exons. The Oct3-related gene, OTF3C, is a retroposon and has been localized to human chromosome 8. Southern blotting and PCR amplification of human DNA indicated the presence of other OTF3-related genes as has been previously noted in the mouse. Two polymorphisms which can be typed using PCR were identified in OTF3 which will facilitate genetic studies of this gene.

Amino Acid Sequence↗
PubMed1992

Repression of the insulin-like growth factor II gene by the Wilms tumor suppressor WT1.

The Wilms tumor suppressor gene wt1 encodes a zinc finger DNA binding protein, WT1, that functions as a transcriptional repressor. The fetal mitogen insulin-like growth factor II (IGF-II) is overexpressed in Wilms tumors and may have autocrine effects in tumor progression. The major fetal IGF-II promoter was defined in transient transfection assays as a region spanning from nucleotides -295 to +135, relative to the transcription start site. WT1 bound to multiple sites in this region and functioned as a potent repressor of IGF-II transcription in vivo. Maximal repression was dependent on the presence of WT1 binding sites on each side of the transcriptional initiation site. These findings provide a molecular basis for overexpression of IGF-II in Wilms tumors and suggest that WT1 negatively regulates blastemal cell proliferation by limiting the production of a fetal growth factor in the developing vertebrate kidney.

Animals↗
PubMed1992

Fructose transporter in human spermatozoa and small intestine is GLUT5.

We recently reported that the glucose transporter isoform, GLUT5, is expressed on the brush border membrane of human small intestinal enterocytes (Davidson, N. O., Hausman, A. M. L., Ifkovits, C. A., Buse, J. B., Gould, G. W., Burant, C. F., and Bell, G. I. (1992) Am. J. Physiol. 262, C795-C800). To define its role in sugar transport, human GLUT5 was expressed in Xenopus oocytes and its substrate specificity and kinetic properties determined. GLUT5 exhibits selectivity for fructose transport, as determined by inhibition studies, with a Km of 6 mM. In addition, fructose transport by GLUT5 is not inhibited by cytochalasin B, a competitive inhibitor of facilitative glucose transporters. RNA and protein blotting studies showed the presence of high levels of GLUT5 mRNA and protein in human testis and spermatozoa, and immunocytochemical studies localize GLUT5 to the plasma membrane of mature spermatids and spermatozoa. The biochemical properties and tissue distribution of GLUT5 are consistent with a physiological role for this protein as a fructose transporter.

Animals↗
PubMed1992

Expression and regulation of the human GLUT4/muscle-fat facilitative glucose transporter gene in transgenic mice.

To study the molecular basis of tissue-specific expression of the GLUT4/muscle-fat facilitative glucose transporter gene, we generated lines of transgenic mice carrying 2.4 kilobases of the 5'-flanking region of the human GLUT4 gene fused to a chloramphenicol acetyltransferase (CAT) reporter gene (hGLUT4[2.4]-CAT). This reporter gene construct was specifically expressed in tissues that normally express GLUT4 mRNA, which include both brown and white adipose tissues as well as cardiac, skeletal, and smooth muscle. In contrast, CAT reporter activity was not detected in brain or liver, two tissues that do not express the GLUT4 gene. In addition, the relative levels of CAT mRNA driven by the human GLUT4 promoter in various tissues of these transgenic animals mirrored those of the endogenous mouse GLUT4 mRNA. Since previous studies have observed alterations in GLUT4 mRNA levels induced by fasting and refeeding (Sivitz, W. I., DeSautel, S. L., Kayano, T., Bell, G. I., and Pessin, J. E. (1989) Nature 340, 72-74), the regulated expression the hGLUT4[2.4]-CAT transgene was also assessed in these animals. Fasting was observed to decrease CAT activity in white adipose tissue which was super-induced upon refeeding. These alterations in CAT expression occurred in parallel to the changes in endogenous mouse GLUT4 mRNA levels. Although CAT expression in skeletal muscle and brown adipose tissue was unaffected, the endogenous mouse GLUT4 mRNA was also refractory to the effects of fasting/refeeding in these tissues. These data demonstrate that 2.4 kilobases of the 5'-flanking region of the human GLUT4 gene contain all the necessary sequence elements to confer tissue-specific expression and at least some of the sequence elements controlling the hormonal/metabolic regulation of this gene.

Adipose Tissue↗