Visualization of purified fibronectin-transglutaminase complexes.

It has been reported previously (Turner, P.M., and Lorand, L. (1989) Biochemistry 28, 628-635) that human erythrocyte transglutaminase forms a noncovalent complex with human plasma fibronectin near its collagen-binding domain. In the present study, we show by nondenaturing electrophoresis that guinea pig liver transglutaminase, similarly to the erythrocyte enzyme, forms a complex with human fibronectin. Studies of anisotropic shifts of fluorescein-labeled liver and erythrocyte transglutaminases, upon addition of fibronectin, indicated that both transglutaminases bind to fibronectin with a stoichiometry of about 2:1. Polymerization of fibrinogen by human erythrocyte transglutaminase was inhibited after complex formation with fibronectin. Complexes of fibronectin with either erythrocyte or liver transglutaminase were isolated by glycerol gradient zone sedimentation and examined by rotary shadowing electron microscopy. The globular transglutaminase could be readily identified binding to the thin fibronectin strand. The binding site for transglutaminase was within 5-10 nm of the N terminus of fibronectin, consistent with its proximity to the collagen-binding domain. Under some experimental conditions, the complex of fibronectin with erythrocyte transglutaminase appeared as a ring-shaped structure in which two transglutaminase molecules had probably dimerized. The molecular weight of the erythrocyte transglutaminase was determined by sedimentation equilibrium to be 71,440 +/- 830.     

Increase of a calcium independent transglutaminase activity in the erythrocyte during the infection with Plasmodium falciparum

We have studied the activity of a calcium dependent transglutaminase (EC 2.3.2.13) during the growth of the parasite Plasmodium falciparum inside the infected human erythrocyte. There is only one detectable transglutaminase in the two-cell-system, and its origin is erythrocytic. No activity was detected in preparations of the parasite devoid of erythrocyte cytoplasm. The Michaelis Menten constants (Km) of the enzyme for the substrates N'N' dimethylcaseine and putrescine were undistinguishable whether the cell extracts used in their determination were obtained from normal or from infected red cells. The total activity of transglutaminase in stringently synchronized cultures, measured at 0.5 mM Ca2+, decreased with the maturation of the parasite. However, a fraction which became irreversibly activated and independent of calcium concentration was detected. The proportion of this fraction grew with maturation; it represented only 20% of the activity in 20 hr-old-trophozoites while in 48-hr-schizonts it was more than 85% of the total activity. The activation of this fraction of transglutaminase did not depend on an increase in the erythrocyte cytoplasmic calcium, since most of the calcium was shown to be located in the parasite.     

Inhibition of activity and quenching of intrinsic fluorescence of transglutaminase by acrylamide are independent events

Addition of low concentrations of acrylamide to the assay mixture of erythrocyte transglutaminase leads to a strong inhibition of the enzyme with a mechanism consistent with a non-competitive inhibition against both substrates. For the quenching of the intrinsic fluorescence of the protein, much higher concentrations of acrylamide are required so that both phenomena appear independent of each other. In this particular case, therefore, acrylamide can be employed to obtain information on ligand-triggered conformational changes.     

High-performance liquid chromatographic assay of transglutaminase and its application to the purification of human erythrocyte transglutaminase and platelet factor XIII

A high-performance liquid chromatographic method was developed for the assay of transglutaminase [EC 2.3.2.13] activity. Casein and dansylcadaverine were used as substrates and the reaction was stopped by adding an excess amount of EGTA. Casein-bound dansylcadaverine was separated from free dansylcadaverine by high-performance liquid chromatography on a TSK SW gel column on the basis of the differences in the molecular weight and hydrophobicity. The sensitivity was approximately 0.04 nmol of casein-bound dansylcadaverine in the assay mixture. With this assay method, human erythrocyte transglutaminase and platelet factor XIII were purified by successive chromatographies on DEAE-cellulose and Sephacryl S-300, which were common for both enzymes, followed by Blue Sepharose CL-6B and DEAE Bio-Gel A for erythrocyte transglutaminase or Phenyl-Sepharose CL-4B for platelet factor XIII. The purification factors and activity yields were 15,300-fold and 22% for erythrocyte transglutaminase and 43.8-fold and 33% for platelet factor XIII.     

Modification of human erythrocyte ghosts with transglutaminase

Guinea pig liver transglutaminase was shown to catalyze the incorporation of dansylcadaverine and putrescine into two major protein fractions of human erythrocyte ghosts. As judged by sodium dodecylsulfate gel electrophoresis under reducing conditions, one of these is a high molecular weight polymer which may contain spectrin. The other corresponds to band 3, an 88, 000 dalton polypeptide. Amine substrates of transglutaminase were synthesized with specific properties to further explore this useful enzymatic technique of covalently labelling proteins in erythrocyte ghosts and in other biological membranes.     

Activation of transglutaminase in mu-calpain null erythrocytes

Intracellular transglutaminases (protein-glutamine: amine gamma-glutamyltransferase, EC 2.3.2.13) are calcium-dependent thiol enzymes that catalyze the covalent cross-linking of proteins, including those in the erythrocyte membrane. Several studies suggest that the activation of some transglutaminases is positively regulated by the calcium-dependent cysteine protease, mu-calpain. Using mu-calpain null (Capn1(-/-)) mouse erythrocytes, we demonstrate that the activation of soluble as well as membrane-bound forms of transglutaminase (TG2) in mouse erythrocytes was independent of mu-calpain. Also, the absence of mu-calpain or any detectable cysteine protease did not affect the transglutaminase activity in the erythrocyte lysate. Our studies also identify physiological substrates of mu-calpain in the erythrocyte membrane and show that their cleavage has no discernible effect on the transglutaminase mediated cross-linking of membrane proteins. Taken together, these data suggest the existence of a calpain-independent mechanism for the activation of transglutaminase 2 by calcium ions in the mouse erythrocytes and presumably also in non-erythroid cells.     

Human erythrocyte transglutaminase: purification and preliminary characterisation

Erythrocyte transglutaminase was purified by anion-exchange chromatography, size exclusion and affinity chromatography. Homogeneity was achieved by an additional step of HPLC size-exclusion chromatography. The molecular mass of the purified enzyme was calculated to be 65,000 Da by size-exclusion chromatography and sucrose-gradient centrifugation, and 92,000 Da by SDS-PAGE, thus suggesting a high degree of asymmetry. The amino-acid composition of erythrocyte transglutaminase differed substantially from that of the guinea-pig liver enzyme, notably with respect to the number of histidine, cysteine and acidic amino-acid residues. The enzyme has an absolute requirement for divalent cations for activity: calcium, manganese, and the lanthanides terbium and gadolinium activate the enzyme in decreasing order of efficacy, while no activity is displayed in the presence of magnesium. In the presence but not in the absence of calcium ions, the enzyme is rapidly inactivated by N-ethylmaleimide and by diethylpyrocarbonate suggesting that the cation influences the reactivity of amino acids essential for catalysis. When erythrocyte proteins are employed as amine acceptors in the presence of calcium, the erythrocyte transglutaminase appears to preferentially modify membrane-associated proteins, although, in the absence of calcium ions and exogenous amines, it displays a pH-dependent interaction with soluble proteins.     

Retinoic acid-induced transglutaminase in mouse epidermal cells is distinct from epidermal transglutaminase

Elevated transglutaminase activity and formation of cornified envelopes are markers of terminal differentiation in mouse epidermal cells. Epidermal transglutaminase catalyzes cornified envelope formation and in cultured cells is inducible by calcium ion or phorbol ester tumor promoters. Retinoic acid also induces transglutaminase activity but inhibits cross-linked envelope formation. This apparent paradox might be resolved by the observation that the retinoic acid-induced transglutaminase appears to be either a different enzyme or a markedly altered form of the epidermal enzyme. The retinoic acid-induced transglutaminase is soluble in aqueous buffers, is thermolabile at pH 9.0, 37 degrees C, and elutes from an anion exchange column at 0.4 M NaCl. In contrast, the epidermal enzyme is particulate and requires detergent for solubilization, is relatively thermostable, and elutes from the anion exchanger at 0.25 M NaCl. The retinoic acid-induced enzyme is probably identical with the "tissue" transglutaminase present in liver and in other cells. It is proposed that the transglutaminase induced by retinoic acid may play a role in the inhibition by retinoids of calcium and tumor promoter-induced differentiation.     

Complexation of fibronectin with tissue transglutaminase

Previous work [Lorand, L., Dailey, J. E., & Turner, P. M. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 1057-1059] showed that fibronectin might serve as a specific carrier for transglutaminases accidentally discharged from erythrocytes or other cells into plasma. In the present study we examined the association of these proteins in purified systems. Complexation was readily demonstrable by nondenaturing electrophoresis, using dansylcadaverine-dependent activity staining as well as immunoblotting procedures, and also by HPLC gel filtration. The results indicate a stoichiometry of 2:1 for the binding of the human erythrocyte transglutaminase (80K) to human plasma fibronectin (440K). The attachment is noncovalent in nature and does not involve cross-linking of the proteins either to themselves or to each other. Binding occurs in the absence of Ca2+, suggesting that a domain on the transglutaminase molecule other than the catalytic site is needed for complexation with fibronectin. Limited proteolysis with chymotrypsin for delineating the relevant region in fibronectin yielded two gelatin- (collagen) binding fragments (56K and 46K), each displaying affinity for transglutaminase. Moreover, these fragments--like intact fibronectin--bound erythrocyte transglutaminase and gelatin simultaneously in ternary complexes.     

Intracellular distribution of transglutaminase activity during rat liver regeneration

Transglutaminase and ornithine decarboxylase activities have been assayed at intervals after partial hepatectomy in regenerating liver cells fractionated to obtain nuclear, cytoplasmic-particulate, and cytoplasmic-soluble fractions. Ornithine decarboxylase activity, localized entirely in the cytoplasmic fractions, undergoes a dramatic induction during the first 4 h after partial hepatectomy and remains elevated. This induction is very sensitive to inhibition by cycloheximide and actinomycin D, as previously reported. Transglutaminase activity is localized in both the cytoplasm and the nucleus with the highest specific activity in the nucleus. Nuclear transglutaminase activity approximately doubles in the first 2 h of liver regeneration, apparently as a result of a translocation of enzyme from the cytoplasm to the nucleus. Inhibitor studies indicate that the translocation is not dependent upon protein or RNA synthesis. In the first 2 h, actinomycin D slightly activates transglutaminase activity in the cytoplasmic-particulate and nuclear fractions. Only at 4 h after the onset of regeneration do actinomycin D and cycloheximide show some inhibition of transglutaminase activity indicating de novo synthesis at this time. A broad increase of transglutaminase activity occurs from hours 12–16 to hour 32 after partial hepatectomy in the nuclear and cytoplasmic-particulate fraction. These data suggest the existence of a function for transglutaminase in the nucleus of rat liver cells.     

Isolation and characterization of cDNA clones to mouse macrophage and human endothelial cell tissue transglutaminases.

The deduced amino acid sequences for tissue transglutaminases from human endothelial cells and mouse macrophages have been derived from cloned cDNAs. Northern blot analysis of both tissue transglutaminases shows a message size of approximately 3.6-3.7 kilobases. The molecular weights calculated from the deduced amino acid sequences were 77,253 for human endothelial tissue transglutaminase and 76,699 for mouse macrophage tissue transglutaminase. The deduced amino acid sequence for the human endothelial transglutaminase was confirmed by comparison with the amino acid sequence obtained by cyanogen bromide digestion of the human erythrocyte transglutaminase. The amino acid sequences of both human endothelial and mouse macrophage tissue transglutaminases were compared to other transglutaminases. A very high degree of homology was found between human endothelial, mouse macrophage, and guinea pig liver tissue transglutaminase (greater than 80%). Moreover, human endothelial tissue transglutaminase was compared with human Factor XIIIa and a very high degree of homology (75% identity) was found in the active site region.     

cDNA Sequence, Gene Sequence, and Properties of Murine Pallidin (Band 4.2), the Protein Implicated in the Murine pallid Mutation

Band 4.2, which plays an important but poorly understood role in erythrocyte function and survival, is a major component of erythrocyte membranes. Recently, it has been shown that the gene for murine protein band 4.2 colocalizes on chromosome 2 with the murine pallid mutation, which affects the formation or function of intracellular storage granules in melanocytes and platelets and lysosomes in kidney. As a first step in identifying the mutation responsible for the pallid phenotype, we have sequenced the entire normal murine band 4.2 gene. Our results show that the gene for murine band 4.2 is approximately 22 kb in size, with 13 exons and 12 intervening introns. The organization of the mouse band 4.2 gene is identical to that of the human band 4.2 gene and similar to that of the genes for the transglutaminase enzymes, reiterating the membership of protein band 4.2 in the transglutaminase gene superfamily. We also present 3.5 kb of normal murine erythroid band 4.2 cDNA sequence containing an open reading frame of 2073 bp and coding for 691 amino acids. This is the same size as the human erythrocyte protein, with which the murine protein shares a 72% amino acid identity.     

A role for transglutaminase in glucose-stimulated insulin release from the pancreatic beta-cell.

Preincubation of rat islets of Langerhans with the potent inhibitors of islet transglutaminase activity, monodansylcadaverine (30-100 microM) and N-(5-aminopentyl)-2-naphthalenesulphonamide (100-200 microM), led to significant inhibition of glucose-stimulated insulin release from islets. In contrast, the respective N'-dimethylated derivatives of these two compounds, which did not inhibit islet transglutaminase activity, were much less effective as inhibitors of glucose-stimulated insulin release. None of the compounds inhibited rat spleen protein kinase C activity at concentrations which gave rise to inhibition of glucose-stimulated insulin release. When tested for their effects on calmodulin-stimulated bovine heart phosphodiesterase activity, of the compounds that inhibited insulin release, only monodansylcadaverine did not act as an effective antagonist of calmodulin at concentrations (up to 50 microM) that gave rise to significant inhibition of glucose-stimulated insulin release. Furthermore, at 50 microM, monodansylcadaverine did not inhibit methylation of islet lipids. The inhibition of glucose-stimulated insulin release by monodansylcadaverine is therefore likely to be attributable to its interference with islet transglutaminase activity. The sensitivity of islet transglutaminase to activation by Ca2+ was investigated by using a modified assay incorporating dephosphorylated NN'-dimethylcasein as a substrate protein. The Km for Ca2+ obtained (approx. 3 microM) was an order of magnitude lower than previously reported for the islet enzyme [Bungay, Potter & Griffin (1984) Biochem. J. 219, 819-827]. Mg2+ (2 mM) was found to have little effect on the sensitivity of the enzyme to Ca2+. Investigation of the endogenous substrate proteins of islet transglutaminase by using the Ca2+-dependent incorporation of [14C]methylamine into proteins of islet homogenates demonstrated that most of the incorporated radiolabel was present in cross-linked polymeric aggregates which did not traverse 3% (w/v) acrylamide gels. The radiolabelled polymeric aggregates were present in 71 000 g-sedimented material of homogenates, and their formation was transglutaminase-mediated. These findings provide new evidence for the involvement of islet transglutaminase in the membrane-mediated events necessary for glucose-stimulated insulin release.     

Interaction of erythrocyte transglutaminase with calcium ions

We have investigated the interaction between calcium ions and erythrocyte transglutaminase and the enzyme activation. The binding involves both high and low affinity sites, but only the former ones are relevant for activation. The binding of calcium and the activation are modified by treatment with NBD-Cl and with PLP suggesting the presence of cysteine and lysine residues at the high affinity binding sites. The interaction of the enzyme with calcium is not calmodulin dependent and is easily detected as a shift in electrophoretic mobility in the presence of SDS.     

Retinoids induce tissue transglutaminase in NIH-3T3 cells.

We report that all-trans and 13-cis-retinoic acid as well as the synthetic compound CH-55 enhance tissue transglutaminase activity as they increase NIH-3T3 cell adhesiveness. The 4-hydroxyphenylretinamide (4-HPR) with low activity in inducing attachment, lectin binding and growth inhibition also fails to induce transglutaminase. Thyroxine (Thy), a compound with a response element common to RA, is inactive. The tumor promoter 12-tetradecanoyl-phorbol-13-acetate (TPA), which increases adhesiveness with different kinetics than RA, failed to enhance tranglutaminase. We conclude that retinoids with biological activity in inducing adhesion, inhibition of growth and increase of lectin binding, are also active in inducing transglutaminase activity.     

Induction of cellular transglutaminase biosynthesis by sodium butyrate

Cellular transglutaminase activity was induced in simian virus-transformed human embryonic lung fibroblasts (WI-38 VA13A) by sodium butyrate. The level of enzyme activity approached a maximum by 6 days; 9–11-fold higher in the presence of sodium butyrate (1 mM) than in its absence. The observed increases in cellular transglutaminase activity could be entirely accounted for by equivalent increases in the levels of enzyme protein measured by inhibition enzyme-linked immunosorbent assay. Sodium butyrate also increased the rate of enzyme synthesis, but had no effect on the rate of cellular transglutaminase degradation. The increase in the rate of enzyme synthesis was matched by an increased level of translatable transglutaminase mRNA as measured in a cell-free translation system. Our results suggest that sodium butyrate regulates cellular transglutaminase at the pretranslational level.     

Transglutaminase 2 interaction with small heat shock proteins mediate cell survival upon excitotoxic stress

Transglutaminase 2 has been postulated to be involved in the pathogenesis of central nervous system neurodegenerative disorders. However, its role in neuronal cell death remains to be elucidated. Excitotoxicity is a common event underlying neurodegeneration. We aimed to evaluate the protein targets for transglutaminase 2 in cell response to NMDA-induced excitotoxic stress, using SH-SY5Y neuroblastoma cells which express high tranglutaminase 2 levels upon retinoic acid-driven differentiation toward neurons. NMDA-evoked calcium increase led to transglutaminase 2 activation that mediated cell survival, as at first suggested by the exacerbation of NMDA toxicity in the presence of R283, a synthetic competitive inhibitor of transglutaminase active site. Assays of R283-mediated transglutaminase inhibition showed the involvement of enzyme activity in NMDA-induced reduction in protein basal levels of pro-apoptotic caspase-3 and the stress protein Hsp20. However, this occurred in a way different from protein cross-linking, given that macromolecular assemblies were not observed in our experimental conditions for both proteins. Co-immunoprecipitation experiments provided evidence for the interaction, in basal conditions, between transglutaminase 2 and Hsp20, as well as between Hsp20 and Hsp27, a major anti-apoptotic protein promoting caspase-3 inactivation and degradation. NMDA treatment disrupted both these interactions that were restored upon transglutaminase 2 inhibition with R283. These results suggest that transglutaminase 2 might be protective against NMDA-evoked excitotoxic insult in neuronal-like SH-SY5Y cells in a way, independent from transamidation that likely involves its interaction with the complex Hsp20/Hsp27 playing a pro-survival role.     

Cystamine inhibits caspase activity. Implications for the treatment of polyglutamine disorders

Huntington's disease (HD) is an autosomal dominant neurodegenerative disorder caused by an abnormally expended polyglutamine domain. There is no effective treatment for HD; however, inhibition of caspase activity or prevention of mitochondria dysfunction delays disease progression in HD mouse models. Similarly administration of cystamine, which can inhibit transglutaminase, prolonged survival of HD mice, suggesting that inhibition of transglutaminase might provide a new treatment strategy. However, it has been suggested that cystamine may inhibit other thiol-dependent enzymes in addition to transglutaminase. In this study we show that cystamine inhibits recombinant active caspase-3 in a concentration-dependent manner. At low concentrations cystamine is an uncompetitive inhibitor of caspase-3 activity, becoming a non-competitive inhibitor at higher concentrations. The IC(50) for cystamine-mediated inhibition of caspase-3 activity in vitro was 23.6 microm. In situ cystamine inhibited in a concentration-dependent manner the activation of caspase-3 by different pro-apoptotic agents. Additionally, cystamine inhibited caspase-3 activity to the same extent in cell lines stably overexpressing wild type tissue transglutaminase (tTG), a mutant inactive tTG, or an antisense for tTG, demonstrating that cystamine inhibits caspase activity independently of any effects it may have on the transamidating activity of tTG. Finally, treatment with cystamine resulted in a robust increase in the levels of glutathione. These findings demonstrate that cystamine may prolong neuronal survival and delay the onset of HD by inhibiting caspases and increasing the level of antioxidants such as glutathione.     

Tissue (type II) transglutaminase covalently incorporates itself, fibrinogen, or fibronectin into high molecular weight complexes on the extracellular surface of isolated hepatocytes. Use of 2-[(2-oxopropyl)thio] imidazolium derivatives as cellular transglutaminase inactivators.

Rabbit hepatocyte surface-expressed tissue (type II) transglutaminase is shown to act as a binding site for fibrinogen or fibronectin and to covalently incorporate these glycoproteins, in addition to itself, into extracellular high molecular weight complexes. This concept is supported by the observation that a nonpeptidyl, active site-directed transglutaminase inactivator (L683685) elicited concentration-dependent (0.1-10 microM) decreases in the calcium-dependent binding and covalent cross-linking of 125I-fibrinogen, 125I-fibronectin, or [14C]putrescine by hepatocyte suspensions. In corroboration with these findings, an antiserum against rabbit liver transglutaminase, which did not cross-react with rabbit factor XIII, elicited concentration-dependent decreases in the calcium-dependent binding and covalent cross-linking of 125I-fibrinogen or [14C]putrescine by hepatocyte suspensions. Western blots of sodium dodecyl sulfate/Triton-insoluble hepatocyte fractions conducted with this antiserum, with a polyclonal antiserum against human erythrocyte transglutaminase, or with a monoclonal antibody (CUB-7401) against guinea pig liver transglutaminase detected the 80-kDa tissue transglutaminase, as well as tissue transglutaminase-immunoreactive bands of higher molecular mass (range of 90 to greater than 200 kDa). The higher molecular weight species were preferentially incorporated, in a time- and calcium-dependent manner, into very high molecular weight complexes which did not enter the stacking gel. Incorporation of these tissue transglutaminase-containing bands into the high molecular weight complexes was inhibited by L683685, indicating that cross-linking by the enzyme was responsible for the assembly of the complexes of which tissue transglutaminase was itself a component. Cellular integrins did not mediate ligand binding under the experimental conditions, as evidenced by the failure of the Arg-Gly-Asp-Ser tetrapeptide or anti-integrin antibodies to inhibit binding or cross-linking of 125I-fibrinogen or 125I-fibronectin, in the presence or absence of transglutaminase inactivators.     

Bovine aortic endothelial cell transglutaminase. Enzyme characterization and regulation of activity.

Bovine aortic endothelial cells contain Ca2+-dependent tissue-type transglutaminase. Its activity in these cells was high, with apparent Km and Vmax. values with respect to putrescine of 0.203 mM and 18.5 nmol/min per mg of protein, and its activity was inhibited by the three competitive inhibitors dansylcadaverine, spermine and methylamine. The molecular mass of endothelial cell transglutaminase estimated by gel filtration chromatography was 88 kDa and it was immunoprecipitated by rabbit monospecific antiserum raised against rat liver transglutaminase. Its enzymic activity rose when the cell cultures reached confluence, and was further increased when their proliferation was arrested (synchronized at G0/G1 phase). Most of the enzymic activity was found in the 15,000 g soluble fraction, with only 4-22% of the activity found in the particulate fraction, depending on the state of cell proliferation. Examination of these cellular fractions by SDS/polyacrylamide-gel electrophoresis and immunoblotting revealed that at confluence endothelial cells have accumulated transglutaminase antigen in their 15,000 g particulate fraction. A series of experiments demonstrated the existence of a latent transglutaminase form in non-proliferating cells, and suggested that this might involve the formation of an inhibitory complex. Treatment of cell lysates and the 15,000 g particulate fraction with high salt concentration showed a significant increase in transglutaminase activity. Mixing experiments using the 100,000 g particulate fraction or purified rat liver transglutaminase on one hand and the cytosolic fraction on the other showed dose-dependent inhibition of the transglutaminase activity of the latter. It is concluded that endothelial cells contain a particulate fraction-residing inhibitor of transglutaminase which interacts via ionic interaction with the enzyme.