Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Dual Roles for Glucokinase in Glucose Homeostasis as Determined by Liver and Pancreatic β Cell-specific Gene Knock-outs Using Cre Recombinase

1999; Elsevier BV; Volume: 274; Issue: 1 Linguagem: Inglês

10.1074/jbc.274.1.305

ISSN

1083-351X

Autores

Catherine Postic, Masakazu Shiota, Kevin D. Niswender, Thomas L. Jetton, Yeujin Chen, J. Michael Moates, Kathy D. Shelton, J Lindner, Alan D. Cherrington, Mark A. Magnuson,

Tópico(s)

Diabetes and associated disorders

Resumo

Glucokinase (GK) gene mutations cause diabetes mellitus in both humans and mouse models, but the pathophysiological basis is only partially defined. We have used cre-loxPtechnology in combination with gene targeting to perform global, β cell-, and hepatocyte-specific gene knock-outs of this enzyme in mice. Gene targeting was used to create a triple-loxed gk allele, which was converted by partial or total Cre-mediated recombination to a conditional allele lacking neomycin resistance, or to a null allele, respectively. β cell- and hepatocyte-specific expression of Cre was achieved using transgenes that contain either insulin or albumin promoter/enhancer sequences. By intercrossing the transgenic mice that express Cre in a cell-specific manner with mice containing a conditional gk allele, we obtained animals with either a β cell or hepatocyte-specific knock-out of GK. Animals either globally deficient in GK, or lacking GK just in β cells, die within a few days of birth from severe diabetes. Mice that are heterozygous null for GK, either globally or just in the β cell, survive but are moderately hyperglycemic. Mice that lack GK only in the liver are only mildly hyperglycemic but display pronounced defects in both glycogen synthesis and glucose turnover rates during a hyperglycemic clamp. Interestingly, hepatic GK knock-out mice also have impaired insulin secretion in response to glucose. These studies indicate that deficiencies in both β cell and hepatic GK contribute to the hyperglycemia of MODY-2. Glucokinase (GK) gene mutations cause diabetes mellitus in both humans and mouse models, but the pathophysiological basis is only partially defined. We have used cre-loxPtechnology in combination with gene targeting to perform global, β cell-, and hepatocyte-specific gene knock-outs of this enzyme in mice. Gene targeting was used to create a triple-loxed gk allele, which was converted by partial or total Cre-mediated recombination to a conditional allele lacking neomycin resistance, or to a null allele, respectively. β cell- and hepatocyte-specific expression of Cre was achieved using transgenes that contain either insulin or albumin promoter/enhancer sequences. By intercrossing the transgenic mice that express Cre in a cell-specific manner with mice containing a conditional gk allele, we obtained animals with either a β cell or hepatocyte-specific knock-out of GK. Animals either globally deficient in GK, or lacking GK just in β cells, die within a few days of birth from severe diabetes. Mice that are heterozygous null for GK, either globally or just in the β cell, survive but are moderately hyperglycemic. Mice that lack GK only in the liver are only mildly hyperglycemic but display pronounced defects in both glycogen synthesis and glucose turnover rates during a hyperglycemic clamp. Interestingly, hepatic GK knock-out mice also have impaired insulin secretion in response to glucose. These studies indicate that deficiencies in both β cell and hepatic GK contribute to the hyperglycemia of MODY-2. glucokinase maturity onset diabetes of the young, type 2 polymerase chain reaction base pair(s) phosphate-buffered saline human growth hormone postnatal day 2. Glucokinase (GK)1 plays an essential role in blood glucose homeostasis. In humans, GK gene mutations cause maturity onset diabetes of the young, type 2 (MODY-2) (1Froguel P. Vaxillaire M. Sun F. Velho G. Zouali H. Butel M.O. Lesage S. Vionnet N. Clément K. Fougerousse F. Tanizawa Y. Weissenbach J. Beckman S.J. Lathrop J.M. Passa P. Permutt M.A. Cohen D. Nature. 1992; 356: 162-164Crossref PubMed Scopus (566) Google Scholar, 2Velho G. Froguel P. Clement K. Pueyo M.E. Rakotoambinina B. Zouali H. Passa P. Cohen D. Robert J.J. Lancet. 1992; 340: 1162-1163Abstract PubMed Scopus (212) Google Scholar, 3Velho G. Blanche H. Vaxillaire M. Bellanne-Chantelot C. Pardini V.C. Timsit J. Passa P. Deschamps I. Robert J.-J. Weber I.T. Marotta D. Pilkis S.J. Lipkind G.M. Bell G.I. Froguel P. Diabetologia. 1997; 40: 217-224Crossref PubMed Scopus (225) Google Scholar, 4Vionnet N. Stoffel M. Takeda J. Yasuda K. Bell G.I. Zouali H. Sesage S. Lesage S. Velho G. Iris F. Passa P. Froguel P. Cohen D. Nature. 1992; 356: 721-722Crossref PubMed Scopus (561) Google Scholar), a disease that is characterized by early onset and persistent hyperglycemia (5Velho G. Froguel P. Diabetes Metab. Rev. 1997; 23: 34-37Google Scholar). The pathophysiological basis of GK-deficient diabetes is generally viewed as involving both pancreatic β cells and hepatocytes (4Vionnet N. Stoffel M. Takeda J. Yasuda K. Bell G.I. Zouali H. Sesage S. Lesage S. Velho G. Iris F. Passa P. Froguel P. Cohen D. Nature. 1992; 356: 721-722Crossref PubMed Scopus (561) Google Scholar, 6Froguel P. Zouali H. Vionnet N. Velho G. Vaxillaire M. Sun F. Lesage S. Stoffel M. Takeda J. Passa P. Permutt M.A. Beckman J.S. Bell G.I. Cohen D. N. Engl. J. Med. 1993; 328: 697-702Crossref PubMed Scopus (716) Google Scholar, 7Hager J. Blanche H. Sun F. Vionnet N. Vaxillaire M. Poller W. Cohen D. Czernichow P. Velho G. Robert J.-J. Cohen N. Froguel P. Diabetes. 1994; 43: 730-733Crossref PubMed Google Scholar, 8Stoffel M. Froguel P. Takeda J. Zouali H. Vionnet N. Nishi S. Weber I.T. Harrison R.W. Pilkis S.J. Lesage S. Vaxillaire M. Velho G. Sun F. Iris F. Passa P. Cohen D. Bell G.I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7698-7702Crossref PubMed Scopus (222) Google Scholar, 9Sun F. Knebelmann B. Pueyo M.E. Zouali H. Lesage S. Vaxillaire M. Passa P. Cohen D. Velho G. Antignac C. Froguel P. J. Clin. Invest. 1993; 92: 1174-1180Crossref PubMed Scopus (26) Google Scholar, 10Zouali H. Vaxillaire M. Lesage S. Sun F. Velho G. Vionnet N. Chiu K. Passa P. Permutt A. Demenais F. Cohen D. Beckman J.S. Froguel P. Diabetes. 1993; 42: 1238-1245Crossref PubMed Scopus (76) Google Scholar), although the relative contributions of each cell type have not been determined. Both β cells and hepatocytes are key sites of GK expression, and both play central roles in glucose homeostasis. In pancreatic β cells GK determines glucose utilization and thus is necessary for glucose-stimulated insulin secretion. In liver, GK is thought to determine rates of both glucose uptake and glycogen synthesis and is also viewed as being essential for the regulation of various glucose-responsive genes (11Girard J. Annu. Rev. Nutr. 1997; 17: 325-352Crossref PubMed Scopus (302) Google Scholar). Although GK is also expressed in certain hypothalamic nuclei of the brain, where it might contribute to feeding behavior and counter regulatory responses, and in the gut, where it might contribute to the secretion of enteroincretins such as GLP-1 (12Jetton T.L. Liang Y. Petterpher C.C. Zimmerman E.C. Cox F.G. Horvath K. Matchinsky F.M. Magnuson M.A. J. Biol. Chem. 1994; 269: 3641-3654Abstract Full Text PDF PubMed Google Scholar), the functional importance of the enzyme in both of these sites is undefined. Both GK gene knock-outs (13Bali D. Svetlanov A. Lee H.-W. Fusco-DeMane D. Leiser M. Li B. Barzilai N. Surana M. Hou H. Fleischer N. DePinho R. Rossetti L. Efrat S. J. Biol. Chem. 1995; 270: 21464-21467Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 14Grupe A. Hultgren B. Ryan S. Ma Y.H. Bauer M. Stewart T.A. Cell. 1995; 83: 69-78Abstract Full Text PDF PubMed Scopus (229) Google Scholar, 15Terauchi Y. Sakura H. Yasuda K. Iwamoto K. Takahashi N. Ito K. Kasai H. Suzuki H. Ueda O. Kamada N. Jishage K. Komeda K. Noda M. Kanazawa Y. Taniguchi S. Miwa I. Akanuma Y. Kodama T. Yazaki Y. Kadowaki T. J. Biol. Chem. 1995; 270: 30253-30256Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar) and GK transgenic mice (16Ferre T. Riu E. Bosch F. Valera A. FASEB J. 1996; 10: 1213-1218Crossref PubMed Scopus (164) Google Scholar, 17Hariharan N. Farrelly D. Hagan D. Hillyer D. Arbeeny C. Sabrah T. Treloar A. Brown K. Kalinowski S. Mookhtiar K. Diabetes. 1997; 46: 11-16Crossref PubMed Google Scholar, 18Niswender K.D. Shiota M. Postic C. Cherrington A.D. Magnuson M.A. J. Biol. Chem. 1997; 272: 22570-22575Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar) have been generated and used as model systems to determine the effects of either increased or diminished GK gene expression on blood glucose homeostasis. These studies have revealed a reciprocal relationship between GK gene copy number and the blood glucose concentration (19Niswender K.D. Postic C. Shiota M. Jetton T.L. Magnuson M.A. Biochem. Soc. Trans. 1997; 25: 113-117Crossref PubMed Scopus (14) Google Scholar). GK gene knock-out mice have provided useful animal models for MODY-2, but analysis of these mice has been limited by several issues. First, mice that totally lack both islet and hepatic isoforms of GK, or even just the islet isoform, do not live for more than a few days, thereby preventing detailed physiological studies in adult animals. Second, the phenotype of mice that globally lack GK remains unsettled. Two different lines of global GK null mice differ markedly in their age of death; one is lethal at mid-gestation (13Bali D. Svetlanov A. Lee H.-W. Fusco-DeMane D. Leiser M. Li B. Barzilai N. Surana M. Hou H. Fleischer N. DePinho R. Rossetti L. Efrat S. J. Biol. Chem. 1995; 270: 21464-21467Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar), whereas the other dies shortly after birth (14Grupe A. Hultgren B. Ryan S. Ma Y.H. Bauer M. Stewart T.A. Cell. 1995; 83: 69-78Abstract Full Text PDF PubMed Scopus (229) Google Scholar). Third, the role of the liver in the pathogenesis of hyperglycemia that characterizes MODY-2 has not been clearly resolved. Characterization of the cell-specific roles of GK in glucose homeostasis requires an animal model in which the enzyme can be selectively eliminated in selected sites without affecting expression in other sites. To achieve this, we made use of the cre-loxPgene targeting strategy since it enables mice containing cell-specific gene deletions to be created (20Tsien J.Z. Chen D.F. Gerber D. Tom C. Mercer E.H. Anderson D.J. Mayford M. Kandel E.R. Tonegawa S. Cell. 1996; 87: 1317-1326Abstract Full Text Full Text PDF PubMed Scopus (944) Google Scholar, 21Kuhn R. Schenk F. Aguet M. Rajewsky K. Nature. 1995; 269: 1427-1429Google Scholar, 22Metzger D. Clifford J. Chiba H. Chambon P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 6991-6995Crossref PubMed Scopus (426) Google Scholar, 23Feil R. Brocard J. Mascrez B. LeMeur M. Metzger D. Chambon P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 10887-10890Crossref PubMed Scopus (675) Google Scholar, 24Kellendonk C. Tronche F. Monaghan A.-P. Angrand P.-O. Stewart F. Schutz G. Nucleic Acids Res. 1996; 24: 1404-1411Crossref PubMed Scopus (213) Google Scholar). We describe here the generation of mice bearing two variant conditional gk alleles. We converted one of these conditional, loxed gk alleles to a null allele, and we also generated mice that express either insulin-cre (Rip-cre) or albumin-cre (Alb-cre) transgenes. By intercrossing two cell-specific Cre transgenes with the loxedgk allele, we generated mice with either hepatic or β cell-specific defects in GK expression. Analysis of these different global and tissue-specific GK gene knock-outs has allowed us to more clearly define how deficiencies in both hepatic and β cell GK cause MODY-2. The key features of the targeting vector used, shown in Fig. 1 a, are a phosphoglycerol kinase-neomycin resistance gene cassette (neoR), a phosphoglycerol kinase-herpes simplex virus type I thymidine kinase gene cassette, and three 34-bp loxP sequences (25Sauer B. Methods Enzymol. 1993; 225: 890-900Crossref PubMed Scopus (261) Google Scholar). Two of the loxP sites flank neoR, and the third is located between exons 8 and 9 of the GK gene. The vector was assembled in pNTK(A) (a gift from Dr. Richard Mortensen, Harvard Medical School) using loxP sites isolated from pBS246 (25Sauer B. Methods Enzymol. 1993; 225: 890-900Crossref PubMed Scopus (261) Google Scholar) and mouse GK gene fragments isolated from clone λ21 which was obtained from a 129/Ola P1 clone (26Postic C. Niswender K.D. Decaux J.-F. Parsa R. Shelton K.D. Gouhot B. Pettepher C.C. Granner D.K. Girard J. Magnuson M.A. Genomics. 1995; 29: 740-750Crossref PubMed Scopus (31) Google Scholar). A detailed description of the 11 DNA manipulations required to assemble the targeting vector are available on request. Correct assembly of this vector was confirmed by DNA sequencing. The sequences altered in the gk lox allele were deposited in GenBankTM (accession numbers AF047362 and AF47830). 50 μg of the targeting vector was linearized with NotI and then electroporated into 5 × 107 TL-1 ES cells, a line derived from 129/SvEvTacBR mice (27Labosky P.A. Barlow D.P. Hogan B.L.M. Development. 1994; 120: 3197-3204Crossref PubMed Google Scholar). Analysis of several clones resistant to both G418 and ganciclovir revealed one clone, BG7, that had undergone the desired recombination event. NeoR was removed from this clone by partial Cre-mediated recombination. Briefly, 5 × 107 cells were electroporated with 50 μg of uncut pBS185, a CMV-cre expression vector (25Sauer B. Methods Enzymol. 1993; 225: 890-900Crossref PubMed Scopus (261) Google Scholar). The transfected cells were plated at low dilution without G418 or ganciclovir selection, and ∼200 clones were picked and characterized. From these, a single clone (BE8) was identified that had undergone partial recombination, thereby removing neoR and one loxP site but leaving both exons 9 and 10 and its flanking loxP sites intact. Both the BG7 and BE8 ES cell clones were microinjected into C57Bl/6 blastocysts and implanted into pseudopregnant ICR female recipients (28Robertson E.J. Robertson E.D. Teratocarcinomas and Embryonic Stem Cells: A Practical Approach. IRL Press at Oxford University Press, Oxford1987: 71-112Google Scholar). Male chimeras were mated with C57Bl/6 females, and germline transmission was identified by both Southern blot and PCR analysis of tail DNA (29Hogan B. Beddington R. Costantini F. Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1994: 293-295Google Scholar). Except for the gk lox/lox mice, which were maintained as both inbred (129/SvEvTac) and hybrid lines, all other combinations of the conditional gk alleles and cell-specific Cre transgenes were studied as hybrid lines derived from 129/SvEvTac, C57Bl/6, and DBA-2 strains. All mice were housed in specific pathogen-free barrier facilities, maintained on a 12-h light/dark cycle, and fed a standard rodent chow (Purina Mills, Inc., St. Louis, MO). Pups from both sexes were used for analyses at day 2. For the rest of the studies only males were used. Animals were killed during the post-absorptive state (between 8:00 and 10:00 a.m.) except as specified. Cre-mediated recombination of thegk lox+neo allele into agk del allele was performed by pronuclear microinjection of supercoiled pBS185 DNA (25Sauer B. Methods Enzymol. 1993; 225: 890-900Crossref PubMed Scopus (261) Google Scholar) into single cell mouse embryos derived from mating gk lox+neo males with superovulated B6D2 F1 hybrid females. Conversion of thegk lox+neo allele into agk del allele was detected by both Southern blot and PCR analysis of tail DNA from all resulting pups. The removal of exons 9 and 10 eliminates 125 of the 465 amino acids in both isoforms of GK. An insulin-cre transgene (Rip-cre) was assembled by removing nls-cre coding sequences from pmL78 (kindly provided by Mark Lewandowsky, University of California, San Francisco) by digestion with SalI andEcoRI and cloning them into the corresponding sites of pBSIIKS. A 668-bp fragment of the rat insulin 2 promoter (Rip) was ligated into the XhoI andKpnI sites of this vector, and a 2.1-kilobase pair fragment of the human growth hormone (hGH) gene was inserted into theSpeI site. Finally, the plasmid was digested withSalI and religated to remove a small intervening DNA fragment that had been created. The albumin-cre transgene (Alb-cre) was made by removing the nls-cre and hGH fusion fragment from the Rip-cre-hGH vector by digestion with XbaI. This fragment was blunt-ended by Klenow fill-in and then ligated into an EcoRV site of a plasmid containing the 2335-bp rat albumin enhancer/promoter fragment (kindly provided by Richard Palmiter, University of Washington) (30Pinkert C.A. Ornitz D.M. Brinster R.L. Palmiter R.D. Genes Dev. 1987; 3: 268-276Crossref Scopus (295) Google Scholar). The Rip-cre-hGH and Alb-cre-hGH plasmids were digested with KpnI andNotI, and the transgenes were isolated by gel electrophoresis and agarase digestion. These DNA fragments were microinjected into the pronuclei of B6D2 F2 hybrid mice by the Vanderbilt Transgenic/ES Cell Shared Resource. Tail DNA from potential founder mice was screened by Southern blot analysis for DNA integration. Rip-cre founder 25 and Alb-crefounder 21, the two Cre transgenic mice selected for use in this study, were estimated to contain 9 and 7 transgene copies, respectively. Three different gk alleles (gk w, gk lox, andgk del) were routinely distinguished by PCR analysis. The gk w allele was detected using primers 1 (5′-TGTCTCAATTTGCTGTGTCCTCCA) and 2 (5′-TCTGTTAATGCAAATGCTCGTGTT), which amplify a 605-bp fragment. Thegk lox allele was detected as a 710-bp fragment after amplification with primers 1 and 2. Thegk del allele was detected as a 435-bp DNA fragment using primers 2 and 3 (5′-TTGAGACCCGTTTTGTGTCG). Both theAlb-cre and Rip-cre transgenes were detected using the primers 5′-ACCTGAAGATGTTCGCGATTATCT and 5′-ACCGTCAGTACGTGAGATATCTT, which amplify a 370-bp fragment. ThelacZ transgenes were detected using the primers 5′-TGCTGATGAAGCAGAACAACTT and 5′-TATTTAATCAGCGACTGTCC, which amplify a 602-bp fragment. RNA was isolated from mouse tissues according to Chomczynski and Sacchi (31Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-159Crossref PubMed Scopus (63169) Google Scholar). First strand cDNA synthesis was carried out by first diluting 1 μg of RNA in 20 μl of diethylpyrocarbonate-treated water, denaturing at 85 °C for 3 min, and chilling on ice. In a 20-μl reaction containing 750 ng of the denatured RNA, 2 μl of 10× reaction buffer, 3 μm MgCl2 (Perkin-Elmer), 1 mm dNTPs (U. S. Biochemical Corp.), 10 units of reverse transcriptase, 20 units of RNasin, and 5 μg of oligo(dT) primer, first strand cDNA synthesis was carried out by incubating the reaction at 25 °C for 10 min, 37 °C for 30 min, 42 °C for 30 min, 95 °C for 5 min, and 5 °C for 5 min. After cDNA synthesis, PCR was carried out in 50-μl reactions containing 5 μl of 10× PCR mix, 1.5 mm MgCl2, 40 μm dATP, dTTP, dGTP, 20 μm dCTP, 2.5 μCi of [α-32P]dCTP, 2.5 units of Taq DNA polymerase (Promega), and 10 pmol of the primers 5′-GAAGCCTATATCCCAAAGGAA and 5′-GACCTTCAACGGTGAGGTCA, which amplify a fragment of 548 bp. Thermal cycling consisted of a 5-min denaturation at 95 °C followed by 26 cycles of 1 min denaturation at 95 °C, 1 min annealing at 59 °C, and 1 min extension at 72 °C. After a final extension at 72 °C for 7 min, 5 μl of the final reaction was mixed with 5 μl of sequencing gel loading buffer and denatured, and 4 μl was separated in a 6% denaturing polyacrylamide gel prior to autoradiography. A Cre-inducible nls-lacZreporter gene (CMV-RAGE-β-Gal) was provided by Dr. Alan Naftalin (Division of Cardiology, Vanderbilt University), which contains, in sequential order, a cytomegalovirus promoter, loxP, yeast His3 (stop) sequences, loxP, and nls-lacZ. The transgene was separated from vector DNA by digestion withSphI and NotI and gel electrophoresis. CATZ transgenic mice (32Agah R. Frenkel P.A. French B.A. Michael L.H. Overbeek P.A. Schneider M.D. J. Clin. Invest. 1997; 100: 169-179Crossref PubMed Scopus (437) Google Scholar), which contain a β actin promoter-driven Cre-activable β-galactosidase fusion gene, were kindly provided by M. Schneider (Baylor College of Medicine). Both lacZ reporter mice were estimated to have ∼35 copies of the transgenes in their genome by quantitative Southern analysis. Tissues were rapidly frozen in liquid nitrogen after removal from the animal, and genomic DNA was prepared according to Hogan et al. (29Hogan B. Beddington R. Costantini F. Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1994: 293-295Google Scholar). The efficiency of Cre recombination was usually assessed by Southern blot analysis. DNA was digested with either BglII or XmnI, and the blots were probed with either an ∼900-bp DNA fragment containing GK exons 9 and 10 or a ∼600-bp BamHI-EcoRI fragment located at the 3′ end of the GK gene locus (Fig.1 a). Blood glucose concentrations were determined with a Hemocue blood glucose analyzer (Hemocue, Mission Viejo, CA). Plasma glucose concentrations were determined using a micro-assay procedure (33Bergmeyer H.U. Bergmeyer J. Grabl M. Methods in Enzymatic Analysis. Verlag Chemie, Weinheim, Germany1984: 163-172Google Scholar). The values obtained from the Hemocue were ∼20% lower than the values determined by plasma glucose assay. Plasma insulin concentrations were determined by radioimmunoassay (RIA) using a rat insulin RIA kit from Linco Research (St. Louis, MO). Plasma non-esterified fatty acids were measured with a NEFA C kit from Wako Pure Chemical Industries. Hepatic glycogen content was analyzed as described by Keppler et al. (34Keppler D. Decker K. Bergmeyer H.U. Methods of Enzymatic Analysis. Verlag Chemie, Weinheim, Germany1984: 11-18Google Scholar). Plasma triglycerides and β-hydroxybutyrate concentrations were measured using a colorimetric kit from Sigma. Western blot analysis and GK activities in crude liver extracts were determined as described previously (35Niswender K.D. Postic C. Jetton T.L. Bennet B.D. Piston D.W. Efrat S. Magnuson M.A. J. Biol. Chem. 1997; 272: 22564-22569Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar). Organs were quickly removed after cervical dislocation and fixed in 4.0% paraformaldehyde in 0.1m phosphate-buffered saline (PBS), pH 7.6, for 2–4 h, and washed in PBS overnight. Tissues were either infiltrated in 30% sucrose/PBS for cryostat sectioning or dehydrated in ethanol for paraffin embedding. lacZ expression was detected by incubating tissues in stain solution (8.4 mm KCl; 84 mm phosphate buffer, pH 7.5; 2 mmMgCl2; 5 mm K4Fe(CN)6; 5 mm K4FeCN6; 0.01% sodium deoxycholate; 0.02% Nonidet P-40; 1% 5-bromo-4-chloro-3-indolyl β-d-galactosidase) for 12 h at room temperature. Samples processed for histological examination were embedded in paraffin, sectioned, and stained with hematoxylin and eosin. For detection of glycogen, cryostat sections of liver were subjected to the periodic acid-Schiff technique. Sections were oxidized 10 min in 0.5% periodic acid, washed, and stained with Schiff's reagent for 5 min. For the detection of neutral lipid, liver cryosections were stained with the Oil Red O technique using 0.23% dye dissolved in 65% isopropyl alcohol for 10 min. Following washing, sections were counterstained with hematoxylin. Tissues were processed for immunocytochemical analysis as described previously (36Jetton T.L. Magnuson M.A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2619-2623Crossref PubMed Scopus (84) Google Scholar). For Cre immunostaining, whole pancreas was fixed and equilibrated in 30% sucrose in PBS overnight at 4 °C. Tissues were then frozen in OCT in a cryostat chamber at −20 to −35 °C, sectioned, mounted onto electrostatically charged slides, and allowed to dry at room temperature for at least 30 min. Cre was then detected using a rabbit polyclonal antisera to Cre (kindly provided by Barbara Morris, Novagen). Hyperglycemic clamp studies were performed using chronically cannulated, conscious mice as described previously (18Niswender K.D. Shiota M. Postic C. Cherrington A.D. Magnuson M.A. J. Biol. Chem. 1997; 272: 22570-22575Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar). All groups of mice were fasted 6–8 h prior to the experimentation. Body weight, hematocrit, and general appearance were used as indices of health. A variable infusion of 50% glucose was used to raise blood glucose levels to ∼300 mg/dl. A 4 μCi bolus of tracer ([3-3H]glucose, NEN Life Science Products) was given at −100 min, followed by a constant 0.04 μCi/min for the duration of the study. The glucose turnover rate (mg/kg/min) was calculated as the rate of tracer infusion (dpm/min) divided by the blood glucose specific activity (dpm/mg) corrected to the body weight of the mouse. Glycogen synthesis via the direct pathway was determined by measuring incorporation of [3H] into glycogen (dpm/g liver)/[3H] specific activity in plasma glucose (dpm/mg). All results are presented as the mean ± S.E. of the mean. Statistical significance was determined by Student's t test. A conditional GK gene allele (gk lox+neo) that contains three loxPsites and a neomycin resistance (neoR) cassette was created by gene targeting in ES cells (Fig.1 a). This allele contains twoloxP sites flanking exons 9 and 10 and a thirdloxP site downstream of neoR. By partial Cre-mediated recombination in ES cells, thegk lox+neo allele was converted to a second conditional allele (gk lox) that lacks neoR and contains only two loxP sites (Fig.1 a). Correct gene targeting and gene conversion in ES cells was confirmed by Southern blot (Fig. 1, b and c) and PCR analysis (not shown). ES cells containing either agk lox+neo or gk lox allele were then used to generate mice with these two conditional alleles, and each were bred to homozygosity. Both lines of mice obtained in this manner were viable, but each exhibited subtle perturbations in their blood glucose concentration. Eight-week-oldgk lox+neo/lox+neo mice had a blood glucose concentration of 242 ± 9 mg/dl compared with 194 ± 3 mg/dl for the gk lox/lox mice and 175 ± 8 mg/dl for mice with two high activity (gk a) wild type alleles (hereafter designated as gk w) (37Moates J.M. Postic C. Decaux J.-F. Girard J. Magnuson M.A. Genomics. 1997; 45: 185-193Crossref PubMed Scopus (4) Google Scholar). The mice were otherwise normal after at least six generations of inbreeding. Since the mixing of different inbred strains of mice may have contributed to the minor differences observed, we also compared the basal blood glucose concentrations of mice with two wild type alleles (gk w/w) with thegk lox/lox mice in an inbred 129SvEvTac background (Table I). A persistent difference was also observed, suggesting that the insertion of aloxP site and some flanking sequences between exons 8 and 9 caused a slight attenuation in GK gene expression. Indeed, measurements of hepatic GK activity indicated it was reduced by 28% ingk lox/lox mice compared withgk w/w mice (15 ± 1 milliunits/mg proteinversus 21 ± 3 milliunits/mg protein, p< 0.05).Table IComparison of post-absorptive (basal) plasma glucose concentrations (mg/dl) in different lines of mice2 days6–10 weeksHybrid micegk w/w73 ± 10175 ± 8n = 8n = 14gk del/w117 ± 10ap < 0.05, compared togkw/w mice.304 ± 21bp < 0.001 compared togk w/w mice.n = 24n = 6gk del/del566 ± 64bp < 0.001 compared togk w/w mice.dAll gk del/del mice and mostgk lox/lox Rip-cre mice die within the 1st week of birth of severe diabetes.n = 9gk lox/w75 ± 7171 ± 7n = 7n = 5gk lox/lox118 ± 8ap < 0.05, compared togkw/w mice.194 ± 3ap < 0.05, compared togkw/w mice.n = 23n = 28gk lox/w + Rip-cre115 ± 4ap < 0.05, compared togkw/w mice.,cp < 0.05 compared togk lox/w mice.229 ± 16ap < 0.05, compared togkw/w mice.,cp < 0.05 compared togk lox/w mice.n = 6n = 7gk lox/lox +Rip-cre335 ± 35bp < 0.001 compared togk w/w mice.,ep < 0.05 compared togk lox/lox mice.dAll gk del/del mice and mostgk lox/lox Rip-cre mice die within the 1st week of birth of severe diabetes.n = 12gk lox/w +Alb-creND193 ± 5ap < 0.05, compared togkw/w mice.,cp < 0.05 compared togk lox/w mice.n = 7gk lox/lox +Alb-cre113 ± 14ap < 0.05, compared togkw/w mice.215 ± 6ep < 0.05 compared togk lox/lox mice.n = 5n = 9129/SvEvTac inbred micegk w/wND132 ± 2n = 6gk lox/loxND165 ± 4ap < 0.05, compared togkw/w mice.n = 6Blood samples were collected, quickly centrifuged, and plasma removed and frozen immediately at −80 °C. Statistical analysis was performed with a two-tailed unpaired Student's t test. Values are represented as means ± S.E. Comparisons were made among mice of similar age and genetic background. Significance was accepted at p < 0.05. ND, not determined.a p < 0.05, compared togkw/w mice.b p < 0.001 compared togk w/w mice.c p < 0.05 compared togk lox/w mice.d All gk del/del mice and mostgk lox/lox Rip-cre mice die within the 1st week of birth of severe diabetes.e p < 0.05 compared togk lox/lox mice. Open table in a new tab Blood samples were collected, quickly centrifuged, and plasma removed and frozen immediately at −80 °C. Statistical analysis was performed with a two-tailed unpaired Student's t test. Values are represented as means ± S.E. Comparisons were made among mice of similar age and genetic background. Significance was accepted at p < 0.05. ND, not determined. To determine the effect of a total deletion of GK, we converted thegk lox+neo allele into a deleted allele (gk del) by microinjecting different concentrations of a CMV-cre expression plasmid (pBS185) into single cell gk lox+neo/w mouse embryos (38Araki K. Araki M. Miyazaki J. Vassalli P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 160-164Crossref PubMed Scopus (258) Google Scholar). A total of 56 pups were born from 398 microinjected embryos. Tail DNAs from these animals were analyzed by both Southern blot (Fig.2 a) and PCR analysis (Fig.2 b), and each recombination event was categorized as either complete or partial (results not shown). At concentrations of 0.1 ng/μl or greater, Cre

Referência(s)