Role of transglutaminase 2 in glucose tolerance ... - Semantic Scholar

Role of transglutaminase 2 in glucose tolerance: knockout mice studies and a putative mutation in a MODY patient FRANCESCA BERNASSOLA, MASSIMO FEDERICI,* MARCO CORAZZARI, ALESSANDRO TERRINONI, MARTA L. HRIBAL,* VINCENZO DE LAURENZI, MARCO RANALLI, ORNELLA MASSA,† GIORGIO SESTI,‡ W.H. IRWIN MCLEAN,§ GENNARO CITRO,储 FABRIZIO BARBETTI,†,†† AND GERRY MELINO1 Biochemistry Laboratory, IDI-IRCCS, 噦 Department of Experimental Medicine and Biochemical Sciences and *Department of Internal Medicine, University of Rome ‘Tor Vergata’, 00133 Rome, Italy; †IBCIT-Biomedical Science Park Rome S. Raffaele, 00144 Rome, Italy; ‡Department of Clinical and Experimental Medicine, University of Catanzaro ‘Magna Grecia’, 88100 Catanzaro, Italy; § Epithelial Genetics Group, Human Genetics Unit, Department of Molecular and Cellular Pathology, Ninewells Medical School, Dundee, DD1 9SY, Scotland, UK; 储SSd-SAFU, Ist. Regina Elena Institute, IFO, Rome, Italy; and ††IRCCS ‘Bambino Gesu`’ Children’s Hospital, 00165 Rome, Italy Transglutaminase 2 (TGase 2) is a Caⴙ2dependent enzyme that catalyzes both intracellular and extracellular cross-linking reactions by transamidation of specific glutamine residues. TGase 2 is known to be involved in the membrane-mediated events required for glucose-stimulated insulin release from the pancreatic ␤ cells. Here we show that targeted disruption of TGase 2 impairs glucose-stimulated insulin secretion. TGase 2ⴚ/ⴚ mice show glucose intolerance after intraperitoneal glucose loading. TGase 2ⴚ/ⴚ mice manifest a tendency to develop hypoglycemia after administration of exogenous insulin as a consequence of enhanced insulin receptor substrate 2 (IRS-2) phosphorylation. We suggest that the increased peripheral sensitivity to insulin partially compensates for the defective secretion in this animal model. TGase 2ⴚ/ⴚ mouse phenotype resembles that of the maturity-onset diabetes of young (MODY) patients. In the course of screening for human TGase 2 gene in Italian subjects with the clinical features of MODY, we detected a missense mutation (N333S) in the active site of the enzyme. Collectively, these results identify TGase 2 as a potential candidate gene in type 2 diabetes.—Bernassola, F., Federici, M., Corazzari, M., Terrinoni, A., Hribal, M. L., De Laurenzi, V., Ranalli, M., Massa, O., Sesti, G., Mclean, W. H. I., Citro, G., Barbetti, F., Melino, G. Role of transglutaminase 2 in glucose tolerance: knockout mice studies and a putative mutation in a MODY patient. FASEB J. 16, 1371–1378 (2002) ABSTRACT

Key Words: insulin 䡠 diabetes 䡠 mature-onset diabetes of the young Transglutaminases (TGases) are Ca⫹2-dependent enzymes, catalyzing intermolecular isopeptide bonds between the ␥-carboxamide groups of peptide-bound glutamine residues and the primary amino groups of 0892-6638/02/0016-1371 © FASEB

several compounds (1). The establishment of these covalent cross-links leads to the post-translational modification of proteins by forming either ε(␥-glutamyl)lysine isodipeptides or N,N-bis(␥-glutamyl)polyamine linkages and, in many instances, oligomerization of substrate proteins. Nine distinct types of TGases have been characterized in mammals: TGase 1, TGase 2 (or tissueTGase, tTG), TGase 3, TGase 4 (prostate), TGase 5, TGase 6, TGase 7, coagulation factor XIII, and band 4.2. TGase 2 (EC 2.3.2.13) is a monomeric cytosolic protein finely regulated at the post-translational level by Ca⫹2, GTP (1) and nitric oxide (2). In addition to its cross-linking activity, TGase 2 exhibits a signaling function as the G␣h subunit of a GTP binding protein, modulating ␣1-adrenergic receptor-stimulated phospholipase C activation (3). The TGase 2 gene is constitutively expressed both during development and in adult tissues, where a tight correlation between TGase 2 expression and occurrence of apoptosis has been found (4 – 6). We have generated TGase 2-deficient mice through homologous recombination techniques (7). The mice are viable and fertile and show no developmental abnormalities. Apoptosis induced in vitro or ex vivo with different agents in fibroblasts and thymocytes, respectively, appears to be normal (7). The endocrine release of insulin occurs by exocytotic fusion with the plasma membrane and is triggered by a rise in intracellular calcium (8). The TGase 2 enzyme has been implicated in a wide variety of intracellular and extracellular biological processes, including stimulus secretion coupling (9 –12) and receptor-mediated endocytosis of several proteins and hormones (13–15). A functional role for TGase 2 in Ca⫹2-dependent 1 Correspondence: IDI-IRCCS, Biochemistry Lab, c/o Dep. Experimental Medicine, D26/F153, University of Rome ‘Tor Vergata’, Via Tor Vergata 135, 00133 Rome, Italy. E-mail: [email protected]

1371

glucose-stimulated insulin release from pancreatic ␤ cells has been reported (9 –12). TGase 2 also mediates retinoid-induced insulin secretion in rat islets (12). In addition, TGase 2 has been implicated in insulin receptor (IR) aggregation, internalization, and intracellular processing, by cross-linking receptors in the area of clathrin-coated pits, thus participating in the regulation of insulin action in target tissues (13–15). These observations prompted us to analyze glucose homeostasis in TGase 2-deficient mice.

sodium orthovanadate, and 1 mM phenylmethylsulfonyl fluoride). The lysates were immunoprecipitated with anti-IR (Transduction Lab, Lexington, KY), anti-IRS-1 (UBI) and IRS-2 (UBI) antibodies coupled to protein A-agarose beads. For immunoblotting, the beads were boiled in SDS sample buffer, separated on 8% SDS-polyacrylamide gel, and transferred to nitrocellulose membranes. The filter was immunoblotted with anti-phosphotyrosine antibody 4G10 (Upstate Biotechnology, Lake Placid, NY). Bound antibodies were detected with horseradish peroxidase-coupled antibodies to rabbit immunoglobulin G using the ECL detection system (Amersham Pharmacia Biotech, Buckinghamshire, UK). Bands were quantitated with NIH Image software (version 1.6; National Institutes of Health, Bethesda, MD).

MATERIALS AND METHODS In vitro islet studies Animal husbandry TGase 2⫺/⫺ mice were generated as described (7). Animals were maintained on the original 129SvJ/C57Bl hybrid background at the Charles River Italia facility (Calco, Italy). All mice were kept in a 12 h dark-light cycle, fed standard chow ad libitum, and handled in accordance with the competent Institutional Review Boards. All animals undergoing experimental procedures were individually genotyped for the TGase 2 gene by PCR (7). Metabolic studies Wild-type and TGase 2⫺/⫺ mice had equivalent body weights ranging from 25 to 35 mg; 6-month-old male animals were used throughout to eliminate potential differences in glucose homeostasis introduced by the estrous. Glucose tolerance test (GTT) was performed as described (16, 17). After an overnight fast of 15 h, mice were injected with 2 g/kg body weight of d-glucose intraperitoneal (i.p.). Plasma glucose levels were determined from mouse tail vein at 0, 30, 60, and 120 min after the injection and measured using an automatic glucometer (Bayer). To determine the acute insulin release, overnight fasted animals were loaded with glucose (3 g/kg body weight) by i.p. infusion. Blood was obtained by retro-orbital bleeds on anesthetized mice and collected at different time points. Plasma insulin levels were measured by using an ELISA kit (CrystalChem, Chicago, IL), with mouse insulin as a standard. Insulin tolerance test (ITT) was performed as described (17–18) using an i.p. injection of human insulin (0.75 U/kg body weight) in anesthetized freely fed mice. Thereafter, blood was drawn at 0, 15, 30, and 60 min from the vein tail and plasma glucose levels were measured. Results were expressed as percentages of initial blood glucose concentrations.

Pancreatic islets were isolated as described using the intraductal collagenase technique (19). In vitro TGase activity was determined by measuring the incorporation of [3H] putrescine into N,N⬘-dimethylcasein as already described (7, 20). The reaction mixture contained 150 mM Tris-HCl buffer pH 8.3, 90 mM NaCl, 10 mM DTT, 15 mM CaCl2, 12.5 mg N,N⬘-dimethylcasein/mL, 0.2 mM putrescine containing 1 ␮Ci [3H] putrescine. Proteins from different tissue and cellular extracts (0.1– 0.3 mg) were incubated with the reaction mixture in a final volume of 150 ␮L at 37°C. After 20 min of incubation, the reaction was stopped by spotting 100 ␮L quadruplicate aliquots onto Whatman 3 MM filter paper. Unbound [3H] putrescine was removed by washing with large volumes of 15%, 10%, and 5% trichloroacetic acid and absolute ethanol. Filters were then air-dried and radioactivity was measured by liquid scintillation counting. For secretion experiments, islets of similar sizes were hand picked and cultured for 48 h at 37°C. All experiments were carried out in Krebs-Ringer bicarbonate buffer with islets from a single harvest pool that included islets extracted from four animals from each genotype. After incubation with different glucose concentrations (ranging from 5.5 to 22.2 mM) for 30 min, both the collected media and the islets extracted in acid ethanol at 4°C, were used for measurement of insulin content by radioimmunoassay as described (21–22). For histological, immunohistochemical and morphometric analysis of islets, pancreata were isolated from 6-month-old male mice and fixed overnight in a 4% paraformaldehyde solution in 0.1 M phosphate-buffered saline, pH 7.4 (Merck, Rahway, NJ). For morphometric analysis, we examined three animals for each genotype. Measurements of islet size were performed as described (16 –18). We performed immunohistochemistry with the avidin-biotin peroxidase system, using rabbit antiinsulin and rabbit anti-glucagon antibodies (Sigma, St. Louis, MO).

Immunoprecipitation and Western blot analyses of insulin signaling proteins

Families with maturity onset diabetes of the young

Insulin-induced tyrosine phosphorylation of IR and IRS proteins was analyzed in skeletal muscle (soleus) and epididymal fat tissue homogenates from 6-month-old male mice (n⫽3 for each genotype). Animals were injected i.p. with human insulin (1.5 U/kg body weight) and tissues were extracted at 0, 15, 30, and 60 min (18). Muscle was homogenized in homogenization buffer-A (20 mM Tris [pH 7.6], 10% glycerol, 1% Nonidet P-40, 140 mM sodium chloride, 2.5 mM calcium chloride, 1 mM magnesium chloride, 1 mM sodium orthovanadate, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride). Epididymal fat pads were homogenized in homogenization buffer-B (20 mM Tris [pH 7.6], 10% glycerol, 1% Nonidet P-40, 137 mM NaCl, 2 mM EDTA, 1 mM

We studied the entire coding region of the TGase 2 gene in 21 probands with maturity onset diabetes of the young (MODY) recruited in a pediatric setting using stringent clinical criteria (23). All subjects were negative for serological markers of autoimmune diabetes (ICA, anti-GAD, anti-IA2), had an affected parent and a third diabetic generation, and were diagnosed as having diabetes before 18 years of age. Those 21 subjects were also negative at screening for mutations by denaturing gradient gel electrophoresis of the coding region (all exons including intron/exon boundaries) of the hepatocyte nuclear factor 1␣ (HNF 1␣/MODY 3) and glucokinase (GK/MODY 2); ⬎ 100 GK mutants detected in the laboratory of F. Bernassola by this technique (23–26 and F. Bernassola,

1372

Vol. 16

September 2002

The FASEB Journal

BERNASSOLA ET AL.

unpublished observations). These two genetic defects combined represent the most frequent cause of MODY in Europe (23–28), accounting for ⬃65– 80% of all MODY cases. In 6 and 14 of those 21 individuals, mutations in the insulin promoter factor 1 gene (IPF-1/MODY 4), and NEUROD1/ MODY 6 were also excluded by single-strand conformation polymorphism (SSCP) (29 and L. Hansen, F. Bernassola, unpublished observations). In addition, we screened nine children bearing a glucokinase mutation as an internal negative control.

Exon 12, TGM2F10N2, 5⬘TGGGTACAGCTTTGTTCAG3⬘, TGM2R10N, 5⬘TGTGACGGTGGGAACTCTA3⬘; Exon 13, TGM2 F11N, 5⬘ATCCTCGGAGGTCACGTGAC3⬘, TGM2 R11N, 5⬘GATAAGGATTGGGATCAAGG3⬘. All clinical investigations, with fully explained patient’s consent, were examined and approved by the competent Institutional Review Boards, according to the Helsinki Declaration.

Human TGase 2 gene analysis

Statistical analyses were performed by GraphPad Prism Software using the one-way ANOVA with Bonferroni comparison or Student t test, as indicated. Values of P ⬍ 0.05 were considered statistically different.

We amplified the 12 exons of the TGase 2 gene by the polymerase chain reaction (PCR). DNA amplicons were then analyzed by denaturing high-pressure liquid chromatography (dHPLC, WAVE Transgenomics) and samples with abnormal peaks were subjected to DNA direct sequencing. The presence/absence of a heterozygous missense mutation (N333S) identified in a patient with MODY and his diabetic father was further analyzed in 100 normal controls. We took advantage of the fact that the mutation introduces an Alu-I restriction enzyme site suitable for a RFLP/PCR assay. For PCR and digestion analysis conditions, a 378 bp DNA fragment was amplified by PCR using primers TGM2F6N2: 5⬘-AAGGAGGGGAGAGGAGTTGA-3⬘ and TGM2R6N2: 5⬘-TTAACTCTCCAAGCCTCGTC-3⬘. PCR program parameters used were the following: a first step of 3 min at 95°C, 40 cycles of 40 s at 94°C; 40 s at 57°C; 40 s at 72°C, followed by a final step of 5 min at 72°C. The Alu-I (Promega, Madison, WI) digestion analysis was performed overnight at 37°C using 20 ␮L of a 50 ␮L PCR standard reaction, adding 3 ␮L of the appropriate 10⫻ buffer and 2 U of enzyme for a final volume of 30 ␮L. The band visible at 1.6% agarose gel after digestion of the wild-type DNA sequence is of 378 bp; the heterozygous mutant, used as positive control, showed the expected additional bands of 240 and 138 bp, respectively. To be sure that no mutations were missed by the dHPLC, WAVE, we then sequenced all 13 exons of 12 MODY patients, confirming the results obtained by dHPLC, WAVE. To sequence all the exons, we amplified the 13 exons of the TGase 2 gene by PCR. The following amplification program was used for all amplifications: 3 min at 95°C; 40 cycles of the following steps: 40 s at 94°C; 40 s at 57.5°C; 40 s at 72°C; a final step of 5 min at 72°C. For each exon we designed specific amplification primers based on the published sequence (GeneBank AL031651). Exon 1, TGM2-F0N, 5⬘AAGCGGGCTATAAGTTAGCG3⬘, TGM2-F0N, 5⬘TCAGGTCTTAGGGATTCAGC3⬘; Exon 2, TGM2 F1N, 5⬘CCGAGGCTCATGCGTCTCC3⬘, TGM2 R1N, 5⬘AGGGTTCTGGGGCATGTCG3⬘; Exon 3, TGM2-F2N, 5⬘CTGCCAGCCCTGCTACCCCT3⬘, TGM2-R2N, 5⬘CCCTGCCCACCCCAACGCTG3⬘; Exon 4, TGM2-F3N, 5⬘CCCAGAGCTGAGGTCCCTTT3⬘, TGM2-R3N, 5⬘CAGAAGGGCCTGAGATGGTT3⬘; Exon 5, TGM2-F4N, 5⬘ATTGTTACCCTCTTCGTCAC3⬘, TGM2-R4N, 5⬘ATGGTAGTTTGGTTAGTGTC3⬘; Exon 6, TGM2-F5N, 5⬘CCCAGGCTCAGTCCCCGTGC3⬘, TGM2-R5N, 5⬘CAGTCTCTCCAATGATAGCC3⬘; Exon 7, TGM2-F6N, 5⬘CCAGAGCAGGGCGGATGT3⬘, TGM2-R6N, 5⬘CATTCCAGACTCCCGACAA3⬘; Exon 8, TGM2F6N2, 5⬘AAGGAGGGGAGAGGAGTTGA3⬘, TGM2R6N2, 5⬘TTAACTCTCCAAGCCTCGTC3⬘; Exon 9, TGM2-F7N, 5⬘CTTGAGGTTTTTGGCTACAT3⬘, TGM2-R7N, 5⬘GGGCTGGGAAAACTGGATGC3⬘; Exon 10, TGM2-F8N, 5⬘GCTCTCTTCTCAACCCCTACA3⬘, TGM2R8N2, 5⬘TAGCATGTTGTCAGTTGGCGG3⬘; Exon 11, TGM2F9N2, 5⬘GGGTGATTCTGCATCATGAG3⬘, TGM2R9N2, 5⬘GTGTGTGGGGTGGGGGTAGG3⬘; DEFECTIVE GLUCOSE TOLERANCE IN MICE LACKING TGase 2

Statistical analysis

RESULTS In vivo glucose tolerance and insulin sensitivity in TGase 2ⴚ/ⴚ mice Glucose levels were measured in the fasted and randomly fed animals. In the randomly fed state, blood glucose concentrations were significantly higher in TGase 2⫺/⫺ mice compared with wild-type animals (167⫾16 and 107⫾35, respectively; n⫽20 for each genotype; P⬍0.0001). Fasting glucose and insulin levels were similar in both groups (data not shown). We examined the effect of the TGase 2 disruption on in vivo glucose metabolism. TGase 2⫺/⫺ mice showed significantly higher glucose levels than those of the wild-type controls during an i.p. GTT at all time points after glucose load (Fig. 1A; P⬍0.0001, ANOVA). Insulin levels were measured during the GTT, showing decreased levels upon glucose stimulation in TGase 2⫺/⫺ mice with respect to wild-type animals (Fig. 1B; P⬍0.05 at 60 and 120 min; ANOVA). To further evaluate the in vivo effect of TGase 2 deletion on pancreatic ␤ cell function, we performed an acute insulin release test on glucose stimulation. TGase 2⫺/⫺ mice showed significantly higher glucose levels than wild-type controls during this test (Fig. 1C; P⬍0.01 at 2 min and P⬍0.001 at 5 min; ANOVA). In TGase 2⫺/⫺ mice, acute insulin release at 2 min in response to glucose was reduced by 42% compared with wild-type mice (P⬍0.0047, ANOVA; Fig. 1D). These results suggest that TGase 2 deficiency may impair the glucose-stimulated insulin secretory response. Next, we determined in vivo insulin sensitivity by an i.p. ITT. TGase 2⫺/⫺ mice showed a greater hypoglycemic response to exogenous insulin compared with wild-type controls at 60 min (P⬍0.02, ANOVA; Fig. 1E). IR and IRS proteins tyrosine phosphorylation To investigate mechanisms responsible for increased in vivo insulin sensitivity during the ITT, we assessed tyrosine phosphorylation of IR and IRS proteins in skeletal soleus muscle and epididymal fat tissue during the ITT. Tyrosine phosphorylation of IR and IRS-2 at 60 1373

Figure 1. In vivo analysis of glucose tolerance and insulin sensitivity in wild-type and TGase 2⫺/⫺ mice. All tests have been performed in 6-month-old mice. The results are expressed as mean ⫾ sd and represent 20 animals per group. Glucose and insulin levels during an i.p. glucose tolerance tests are shown in panels A and B, respectively (continuous line, TGase 2 ⫹/⫹ mice; dotted line, TGase 2⫺/⫺mice). The acute-phase insulin secretory response to glucose is depicted in panels C (glucose levels) and D (insulin levels) (n⫽6 animals per each group). E) The behavior of blood glucose (expressed as percentage of initial blood glucose concentration) during the ITT test (0.75 U/kg body weight on fed animals). Data represent the mean ⫾ sd of 8 animals for each genotype; asterisks denote variable values of statistically significant differences (see Results).

min was increased by 23% and 35%, respectively, in skeletal muscle of insulin-stimulated TGase 2⫺/⫺ animals vs. wild-type mice (Fig. 2; P⬍0.05 for IR phosphorylation and P⬍0.001 for IRS-2 phosphorylation, respectively). In the adipose tissue, no differences in tyrosine phosphorylation of IR and IRS-2 could be detected between TGase 2⫺/⫺ and wild-type mice after insulin injection (not shown). IRS-1 tyrosine phosphorylation was similar between TGase 2⫺/⫺ and wild-type animals in both skeletal muscle and adipose tissue (Fig. 2, center: muscle; adipose tissue not shown). TGase 2 in pancreatic islet We have found that TGase 2 enzymatic activity was induced by 50% on glucose stimulation in pancreatic islet homogenates from wild-type mice. In contrast, no significant incorporation of [3H] putrescine into N,N⬘dimethylcasein could be detected in islet homogenates from TGase 2⫺/⫺ animals in either the absence or presence of glucose stimulation (not shown). To test directly the role of TGase 2 in islet function, glucosestimulated insulin secretion was assessed in isolated pancreatic islets in vitro. Reduced insulin secretory response was observed in islets from TGase 2⫺/⫺ mice vs. islets from controls at the two maximal glucose

concentrations used above basal value (11.1, 22.2 mM). (P⬍0.05, ANOVA, Fig. 3A). Total pancreatic insulin and glucagon content (Fig. 3B, C) was similar in TGase 2⫺/⫺ and wild-type mice, indicating that the reduced levels observed in vivo were not due to an impairment in hormone synthesis, but more likely to a secretory defect. Morphometric analysis of pancreata showed no differences in either islet size (Fig. 3D, E) or morphology (Fig. 3F--I). These data further support the notion that the impairment in glucose-stimulated insulin secretion observed in the TGase 2-deficient pancreatic islets derives from a discrete defect occurring at some level in the insulin secretory machinery rather than from gross structural modifications of the islets. TGase 2 mutation in a MODY patient Since the TGase 2⫺/⫺ mouse phenotype is reminiscent of subjects with a mild impairment in glucose metabolism, we analyzed the TGase 2 gene in 21 MODY patients by dHPLC. We also directly sequenced the entire coding region of TGase 2 gene, including intro/ exon boundaries, in 12 MODY-2/MODY-3 negative patients randomly chosen among the 21 screened in the present study: this analysis indicated that dHPLC

Figure 2. Insulin-stimulated tyrosine phosphorylation of IR and IRS proteins in skeletal soleus muscle from 6-month-old wild-type and TGase 2⫺/⫺ mice during an ITT at 0 and 60 min after insulin injection (see Fig. 1E). Data were normalized to the level of immunoprecipitated proteins, as indicated. Data are representative of results obtained from three untreated and three insulin-injected animals per genotype for each time point. *P ⬍ 0.05 for insulin-stimulated IR phosphorylation and **P ⬍ 0.001 for insulin-stimulated IRS-2 phosphorylation of wild-type mice vs. TGase 2 ⫺/⫺ mice. 1374

Vol. 16

September 2002

The FASEB Journal

BERNASSOLA ET AL.

Figure 3. Analysis of insulin secretion from isolated islets and pancreatic islets morphology. Islets of Langerhans were isolated by in situ perfusion of the pancreas followed by collagenase digestion. Islets from TGase 2⫹/⫹ (black symbols) and TGase 2⫺/⫺ (white symbols) mice were stimulated to release insulin by culturing for 30 min with the indicated glucose concentrations. Results are expressed as the amount of secreted insulin divided by insulin content. Results represent the mean ⫾ sd for 4 animals in each group (A); *P ⬍ 0.05. Total pancreatic (B) insulin and (C) glucagon content was measured in 6-month-old mice. Pancreata were homogenized and extracted in acid ethanol. Hormones were measured by radioimmunoassay and expressed per mg of pancreas weight. Values are means ⫾ sd of 6 animals for each group. D, E) Hematoxylin and eosin staining in pancreas sections from 6-month-old TGase 2⫹/⫹ (D) and TGase 2⫺/⫺ (E) mice. G, H) Immunohistochemical analysis for ␤ and ␣ cells. Pancreatic sections from 6-month-old TGase 2⫹/⫹(F, H) and TGase 2⫺/⫺ (G, I) animals were immunologically stained with rabbit anti mouse insulin (F, G) and rabbit anti mouse glucagon (H, I) antibodies. Representative sections are shown. Scale bar, 80 ␮m.

did not miss (at least in the 12 subjects examined) any DNA variation. dHPLC identified several sequence variations scattered throughout the gene that by direct sequencing disclosed silent mutations or intronic variations (data not shown). We identified a heterozygous transition (A998-⬎G) in the TGase 2, resulting in serine substituting for asparagine in codon 333 (N333S), in a single subject with MODY (Fig. 4A). To exclude that TGase 2/N333S was not a rare sequence variant we ran RFLP-PCR assay (see methods) on 100 control subjects with no family history of type 1 and type 2 diabetes mellitus. One band of 378 bp is visible on a 1.6% agarose gel after digestion with AluI of wild-type DNA samples whereas two additional bands of 240 and 138 bp, respectively, are detectable in the heterozygous MODY patient and in his father of the proband (Fig. 4B). In normal controls, this experiment has shown 200 wild-type alleles. The clinical records of the TGase 2/N333S mutant patient showed that the proband was lean (BMI 22.2 at 18 years of age) and had a mild fasting hyperglycemia (6.5– 6.9 mmol) since he was 13 years of age. At age of 14, an acute episode of severe hyperglycemia (27 mmol) without ketosis led to insulin treatment (0.5 U/kg/day). The patient always resulted negative to the search of serological markers of autoimmune diabetes and celiac disease (EMA). A glucagon test performed when the patient was 18 years old (insulin dose: 0.25 U/kg/day) showed detectable basal C peptide levels that moderately increased after stimulus (C peptide: 0⬘: 1.05, 6⬘ after glucagon: 1.83 ng/mL; normal reference basal values: 1.1–3.2 ng/mL). No trial with oral hypoglycemic agents has been attempted so far. The father, who also bears the mutation, is diabetic (fasting plasma glucose 7.2 mmol while taking oral hypoglycemic agents) and moderately overweight (BMI: 30). Five DEFECTIVE GLUCOSE TOLERANCE IN MICE LACKING TGase 2

other family members with diabetes were not available for DNA analysis. The N333 residue is conserved among TGase 1, 3 and factor XIII (Fig. 4C) and is located in the active site of the TGase 2 polypeptide (Fig. 4D), emphasizing its functional importance. The cross-linking enzymatic activity of transglutaminase was evaluated in the erythrocytes of the proband since these cells contain only TGase, as shown in the knockout mice (data not shown). The proband showed 1430 pmol/h/mg of protein and the father showed 1650 pmol/h/mg. The same analysis in 14 normal controls and the 2 unaffected relatives shown in Fig. 4B revealed TGase activity values of 2270 ⫾ 610 pmol/ h/mg of protein (range 1420 –3170). These results are in keeping with an heterozygote status of the proband and his father. The GTPase activity in erythrocytes was not measurable because of the high basal levels of other GTPases, as shown in the knockout mice (data not shown).

DISCUSSION Type 2 diabetes is a complex multifactorial metabolic disorder characterized by peripheral insulin resistance and ␤ cell insulin secretory dysfunction, playing a cooperative role to cause and sustain hyperglycemia (30). Insulin resistance and ␤ cell secretory capacity are genetically determined (31, 32). Genetic studies in diabetic subjects have allowed identification of several gene variations associated with the pathophysiological defects responsible for the onset and the progression of the disease, but so far no major genes have been identified (32). In this report, we analyzed the effect of TGase 2 1375

Figure 4. TGase 2 heterozygous transition mutation 998A ⬎ G (N333S) in a MODY patient. A) Direct sequencing of the TGase 2 exon 8 PCR products showing heterozygous mutation 998A ⬎ G, which predicts the amino acid change N333S within the active site of TGase 2. The reverse sequence is shown, corresponding to base numbers 996-1011 of the TGase 2 coding sequence and the last 11 bp of intron 7. B) Pedigree of the MODY family; arrow indicates the proband. AluI restriction enzyme analysis of TGase 2 exon 8 PCR products shows an additional band of 100 bp in the affected individuals due to the presence of mutation 998A ⬎ G. Both members of the family carrying the mutation displayed impaired glucose tolerance. C) Sequence alignment of the conserved region (active site) flanking the mutated residue of TGase 2, TGase 1, TGase 3, and clotting factor XIIIA. The position of mutated amino acid is indicated by an arrow; the triangle indicates H335, a component of the catalytic triad. D) Computerized homology model of TGase 2 showing the mutated amino acid in yellow and the catalytic triad (C277, D358, H335) in red. The site of entry of the substrate is shown, guided by H305 and D306 (orange). E) Close-up view of the active site with mutated residue (same color coding as panel D). The residues in blue-green have been found mutated in either TGase 1 (Lamellar ichthyosis) or clotting factor XIIIa.

disruption on glucose metabolism. In fact, the TGase 2 enzyme was previously implicated in regulation of Ca⫹2-dependent glucose-stimulated insulin release from pancreatic ␤ cells and in insulin receptor aggregation, internalization, and intracellular processing by cross-linking receptors in the area of clathrin-coated pits (9 –14, 15). Moreover, impaired TGase 2 function has been observed in lymphocytes from type 2 diabetic patients (33). These observations led us to hypothesize a role for the TGase 2 enzyme in regulating insulin secretion and insulin action in target tissues. Our results indicate that disruption of TGase 2 results in a complex phenotype characterized by glucose intolerance arising from impaired insulin secretion after a glucose load and increased insulin sensitivity of peripheral tissues when exogenous insulin is administered. Thus, it is conceivable that the TGase 2 disruption does not result in overt diabetes due to increased peripheral insulin sensitivity, partially attenuating the phenotype. Assessment of in vivo fast insulin release and glucosestimulated insulin secretion from isolated islets revealed that TGase 2⫺/⫺ disruption mildly impairs ␤ cell insulin secretory ability. Accordingly, no differences in total insulin content and islet mass were observed with respect to control mice. TGase 2 associates with cytoskeletal components, inducing their covalent modifications by its GTP and Ca⫹2-dependent cross-linking activity (34 –36). During stimulus secretion coupling in the pancreatic ␤ cell, the TGase 2-dependent cross1376

Vol. 16

September 2002

linking of specific membrane or membrane-associated proteins is an even that participates in the secretory mechanism (9 –12, 37). Heterotrimeric and low molecular mass GTP binding proteins have both been involved in regulating glucose-triggered exocytosis of insulin from pancreatic ␤ cells (38). Since peripheral insulin resistance has been reported to adversely affect insulin secretion (17–18, 30), ultimately resulting in pancreatic exhaustion, we determined in vivo insulin sensitivity by an i.p. ITT. Surprisingly, TGase 2⫺/⫺ mice showed a greater hypoglycemic response to exogenous insulin than did wild-type controls. These results indicate that TGase 2⫺/⫺ animals were indeed more sensitive than wild-type mice. At the molecular level, the higher sensitivity to insulin displayed by TGase 2⫺/⫺ mice during the ITT might be consequent to an increased activation of the insulin receptor and/or its endogenous substrates IRS-1 and IRS-2 in target tissues of insulin action such as skeletal muscle or fat. We observed an increased insulin-induced tyrosine phosphorylation of IR␤ and IRS-2 in skeletal muscle. TGase 2 has been involved in regulation of IR internalization, a process implicated in the transmission and compartmentalization of IR signals to IRS-1 and IRS-2 (39). Thus, our findings might be dependent on a redistribution of IR/IRS complexes in plasma membrane and endosomes of skeletal muscle cells. The phenotype observed in TGase 2⫺/⫺ mice can be

The FASEB Journal

BERNASSOLA ET AL.

recapitulated in a mild impairment in glucose metabolism reminiscent of observations in some individuals with type 2 diabetes and in MODY patients, an autosomal dominant form of type 2 diabetes characterized by defective insulin secretion (28). Current data indicate that ⬃15–25% of European patients with MODY do not bear a mutation in any of the six known MODY genes (28). GK/MODY 2 and HNF 1␣/MODY 3 mutations are found in 65– 80% of all MODY cases, irrespective of mode of selection (pediatric vs. adult diabetes clinic) (23, 27, 28). Furthermore, defects in the other MODY genes account collectively for no ⬎ 2–3% of all cases (27–29). Therefore, we chose to run analysis of variations of the TGase 2 in Italian subjects with the clinical diagnosis of MODY and negative for the search of mutations in MODY 2/MODY 3 genes. This investigation revealed a heterozygous transition (A998-⬎G) predicting amino acid change N333S in the active site of the TGase 2 polypeptide in one individual with MODY and his diabetic father. The molecular mechanism by which this mutation causes hyperglycemia is unknown. In conclusion, we have demonstrated that mice with disruption of the TGase 2 gene show a phenotype characterized by mild impairment in glucose-stimulated insulin secretion determining an hyperglycemic state in the postabsorptive phase. Furthermore, we have identified a mutation in the TGase 2 gene in a MODY family. Though this substitution is highly indicative of a role for TGase 2 in the disease and in keeping with the TGase 2 knockout mouse phenotype, the definitive involvement of the enzyme in the pathophysiology of MODY requires further substantiation. Nevertheless, the present data suggest that TGase 2 might be considered a modulator of ␤ cell function and further investigated as a candidate gene for human MODY and type 2 diabetes. We thank G. Bertini, G. Cortese, and P. Piccoli (S.S.D SAFU, Regina Elena Institute, Rome, Italy) for technical assistance and mouse husbandry, Dr. G. Chiumello (Hospital San Raffaele, Milano, Italy) for identification of MODY patients, and A. M. Davalli (Hospital San Raffaele, Milano, Italy) for advice on islet isolation. This work was supported by Telethon (grants E872, E417/bi), AIRC, MURST 40%, Min. Sanita` and EU Grant QLG1–1999-00739 to G.M., and EU Grant QLG1-CT-1999 – 00674 to G.S. The financial support of Telethon-Italy (grants E1257 to F. Bernassola, E946 to F. Barbetti, and E1224 to V.D.L.) is gratefully acknowledged. F. Bernassola is supported by a FIRC fellowship, and M.L.H. by a EU fellowship. W.H.I.M. is supported by a Wellcome Trust Senior Research Fellowship in Basic Biomedical Sciences.

3.

4.

5. 6.

7. 8. 9. 10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

20.

REFERENCES Achyuthan, K. E., and Greenberg, C. S. (1987) Identification of a guanosine triphosphate-binding site on guinea pig liver transglutaminase. Role of GTP and calcium ions in modulating activity. J. Biol. Chem. 262, 1901–1906 2. Melino, G., Bernassola, F., Knight, R. A., Corasaniti, M. T., Nistico`, G., and Finazzi-Agro, A. (1997) S-nitrosylation regulates apoptosis. Nature (London) 388, 432– 433

21.

1.

DEFECTIVE GLUCOSE TOLERANCE IN MICE LACKING TGase 2

22.

Nakaoka, H., Perez, D. M., Baek, K. J., Das, T., Husain, A., Misono, K., Im, M. J., and Graham, R. M. (1994) Gh: a GTP-binding protein with transglutaminase activity and receptor signaling function. Science 264, 1593–1596 Nagy, L., Thomazy, V. A., Saydak, M. M., Stein, J. P., and Davies, P. J. A. (1997) The promoter of mouse tissue transglutaminase gene directs tissue-specific, retinoid-regulated and apoptosislinked expression. Cell Death Differ. 4, 534 –547 Melino, G., and Piacentini, M. (1998) Tissue transglutaminase in apoptosis: a downstream or a multifunctional upstream effector? FEBS Lett. 430, 59 – 63 Melino, G., Annicchiarico-Petruzzelli, M., Piredda, L., Candi, E., Gentile, V., Davies, P. J. A., and Piacentini, M. (1994) TissueTransglutaminase and apoptosis: sense and antisense transfection studies in human neuroblastoma cells. Mol. Cell. Biol. 14, 6584 – 6596 De Laurenzi, V., and Melino, G. (2001) Gene disruption of tissue transglutaminase. Mol. Cell. Biol. 21, 148 –155 Matschinsky, F. M. (1996) Banting Lecture 1995. A lesson in metabolic regulation inspired by the glucokinase glucose sensor paradigm. Diabetes 45, 223–241 Gomis, R., Sener, A., Malaisse-Lagae, F., and Malaisse, W. J. (1983) Transglutaminase activity in pancreatic islets. Biochim. Biophys. Acta 760, 384 –388 Sener, A., Dunlop, M. E., Gomis, R., Mathias, P. C., MalaisseLagae, F., and Malaisse, W. J. (1985) Role of transglutaminase in insulin release. Study with glycine and sarcosine methyl esters. Endocrinology 117, 237–242 Bungay, P. J., Owen, R. A., Coutts, I. C., and Griffin, M. (1986) A role for transglutaminase in glucose-stimulated insulin release from the pancreatic beta-cell. Biochem. J. 235, 269 –278 Driscoll, H. K., Adkins, C. D., Chertow, T. E., Cordle, M. B., Matthews, K. A., and Chertow, B. S. (1997) Vitamin A stimulation of insulin secretion: effects on transglutaminase mRNA and activity using rat islets and insulin-secreting cells. Pancreas 15, 69 –77 Baldwin, D. Jr., Prince, M., Marshall, S., Davies, P., and Olefsky, J. M. (1980) Regulation of insulin receptors: evidence for involvement of an endocytotic internalization pathway. Proc. Natl. Acad. Sci. USA 77, 5975–5978 Davies, P. J., Davies, D. R., Levitzki, A., Maxfield, F. R., Milhaud, P., Willingham, M. C., and Pastan, I. H. (1980) Transglutaminase is essential in receptor-mediated endocytosis of alpha 2-macroglobulin and polypeptide hormones. Nature (London) 283, 162–167 Haigler, H. T., Maxfield, F. R., Willingham, M. C., and Pastan, I. (1980) Dansylcadaverine inhibits internalization of 125I-epidermal growth factor in BALB 3T3 cells. J. Biol. Chem. 255, 1239 –1241 Withers, D. J., Gutierrez, J. S., Towery, H., Burks, D. J., Ren, J. M., Previs, S., Zhang, Y., Bernal, D., Pons, S., Shulman, G. I., Bonner-Weir, S., and White, M. F. (1998) Disruption of IRS-2 causes type 2 diabetes in mice. Nature (London) 391, 900 –904 Lauro, D., Kido, Y., Castle, A. L., Zarnowski, M. J., Hayashi, H., Ebina, Y., and Accili, D. (1998) Impaired glucose tolerance in mice with a targeted impairment of insulin action in muscle and adipose tissue. Nat. Genet. 20, 294 –298 Bruning, J. C., Winnay, J., Bonner-Weir, S., Taylor, S. I., Accili, D., and Kahn, C. R. (1997) Development of a novel polygenic model of NIDDM in mice heterozygous for IR and IRS-1 null alleles. Cell 88, 561–572 Kulkarni, R. N., Wang, Z. L., Wang, R. M., Hurley, J. D., Smith, D. M., Ghatei, M. A., Withers, D. J., Gardiner, J. V., Bailey, C. J., and Bloom, S. R. (1997) Leptin rapidly suppresses insulin release from insulinoma cells, rat and human islets and, in vivo, in mice. J. Clin. Invest. 100, 2729 –2736 Melino, G., Candi, E., and Steinert, P. M. (2000) Assays for transglutaminases in cell death. Methods Enzymol. 322, 433– 472 Federici, M., Hribal, M. L., Ranalli, M., Marselli, L., Porzio, O., Lauro, D., Borboni, P., Lauro, R., Marchetti, P., Melino, G., and Sesti, G. (2001) The common Arg972 polymorphism in insulin receptor substrate-1 causes apoptosis of human pancreatic islets. FASEB J. 15, 22–24 Porzio, O., Federici, M., Hribal, M. L., Lauro, D., Accili, D., Lauro, R., Borboni, P., and Sesti, G. (1999) The Gly972– ⬎ Arg amino acid polymorphism in IRS-1 impairs insulin secretion in pancreatic beta cells. J. Clin. Invest. 104, 357–364

1377

23.

24.

25.

26.

27.

28.

29.

30.

1378

Massa, O. Meschi, F., Cuesta-Munoz, A., Caumo, A., Cerutti, F., Toni, S., Cherubini, V., Guazzarotti, L., Sulli, N., Matschinsky, F. M., Lorini, R., Iafusco, D., and Barbetti, F. (2001) High prevalence of glucokinase mutations in Italian children with MODY. Influence on glucose tolerance, first-phase insulin response, insulin sensitivity and BMI. Diabetologia 44, 898 –905 Prisco, F., Iafusco, D., Franzese, A., Sulli, N., and Barbetti, F. (2000) MODY 2 presenting as neonatal hyperglycemia: a need to re-shape the definition of ‘neonatal diabetes’? Diabetologia 43, 1331–1332 Njolstad, P. R., Sovik, O., Cuesta-Munoz, A., Bjorkhaug, L., Massa, O., Barbetti, F., Undlien, D., Shiota, C.,Magnuson, M. A., Molven, A., Matschinsky, F. M., and Bell, G. I. (2001) Neonatal diabetes mellitus due to complete glucokinase deficiency. N. Engl. J. Med. 344, 1588 –1592 Christesen, H. B. T., Brock Jacobsen, B., Odili, S., Buettger, C., Cuesta-Munoz, A., Hansen, T., Brusgaard, K., Massa, O., Magnuson, M. A., Shiota, C., Matschinsky, F. M., and Barbetti, F. (2002) The second activating glucokinase mutation (A456V): implications for glucose homeostasis and diabetes therapy. Diabetes In press Hattersley, A. T. (1998) Maturity onset diabetes of the young: clinical heterogeneity explained by genetic heterogeneity. Diabetes Med 15, 15–24 Fajans, S. S., Bell, G. I., and Polonsky, K. S. (2001) Molecular mechanisms and clinical pathophysiology of maturity onset diabetes of the young. N. Engl. J. Med. 345, 971–980 Hansen, L. Urioste, S., Petersen, H. V., Jensen, J. N., Eiberg, H., Barbetti, F., Serup, P., Hansen, T., and Pedersen, O. (2000) Missense mutations in the human insulin promoter factor-1 gene and their relation to maturity-onset diabetes of the young and late-onset type 2 diabetes mellitus in Caucasians. J. Clin. Endocrinol. Metab. 85, 1323–1326 Gerich, J. E. (1998) The genetic basis of type 2 diabetes mellitus: impaired insulin secretion versus impaired insulin sensitivity. Endocr. Rev. 19, 491–503

Vol. 16

September 2002

31. 32. 33. 34.

35.

36.

37. 38.

39.

Kahn, C. R., Vicent, D., and Doria, A. (1996) Genetics of non-insulin-dependent (type-II) diabetes mellitus. Annu. Rev. Med. 47, 509 –531 Almind, K., Doria, A., and Kahn, C. R. (2001) Putting the genes for type II diabetes on the map. Nat Med. 7, 277–279 Berntorp, E., Berntorp, K., Seiving, B., and Stenberg, P. (1987) Lymphocyte transglutaminase function may be impaired in type 2 diabetes mellitus. Diabetes Res. 5, 29 –33 Clement, S., Velasco, P. T., Murthy, S. N., Wilson, J. H., Lukas, T. J., Goldman, R. D., and Lorand, L. (1998) The intermediate filament protein, vimentin, in the lens is a target for crosslinking by transglutaminase. J. Biol. Chem. 273, 7604 –7609 Piredda, L., Farrace, M. G., Lo Bello, M., Malorni, W., Melino, G., Petruzzelli, R., and Piacentini, M. (1999) Identification of ‘tissue’ transglutaminase binding proteins in neural cells committed to apoptosis. FASEB J. 13, 355–364 Trejo-Skalli, A. V., Velasco, P. T., Murthy, S. N., Lorand, L., and Goldman, R. D. (1995) Association of a transglutaminaserelated antigen with intermediate filaments. Proc. Natl. Acad. Sci. USA 92, 8940 – 8944 Gomis, R., Alarcon, C., Valverde, I., and Malaisse, W. J. (1986) Role of transglutaminase in proinsulin conversion and insulin release. Adv. Exp. Med. Biol. 211, 443– 446 Kowluru, A., Seavey, S. E., Li, G., Sorenson, R. L., Weinhaus, A. J., Nesher, R., Rabaglia, M. E., Vadakekalam, J., and Metz, S. A. (1996) Glucose- and GTP-dependent stimulation of the carboxyl methylation of CDC42 in rodent and human pancreatic islets and pure beta cells. Evidence for an essential role of GTP binding proteins in nutrient-induced insulin secretion. J. Clin. Invest. 98, 540 –555 Inoue, G., Cheatham, B., Emkey, R., and Kahn, C. R. (1998) Dynamics of insulin signaling in 3T3–L1 adipocytes. Differential compartmentalization and trafficking of insulin receptor substrate (IRS)-1 and IRS-2. J. Biol. Chem. 273, 11548 –11555

The FASEB Journal

Received for publication January 18, 2002. Revised for publication April 1, 2002.

BERNASSOLA ET AL.