The amino-terminal region of the Escherichia coli T-protein of the glycine cleavage system is essential for proper association with H-protein.

T-protein is a component of the glycine cleavage system and catalyzes the tetrahydrofolate-dependent reaction. Our previous work on Escherichia coli T-protein (ET) showed that the lack of the N-terminal 16 residues caused a loss of catalytic activity [Okamura-Ikeda, K., Ohmura, Y., Fujiwara, K. and Motokawa, Y. (1993) Eur. J. Biochem. 216, 539-548]. To define the role of the N-terminal region of ET, a series of deletion mutants were constructed by site-directed mutagenesis and expressed in E. coli. Deletions of the N-terminal 4, 7 and 11 residues led to reduction in the activity to 42, 9 and 4%, respectively, relative to the wild-type enzyme (wtET). The mutant with 7-residue deletion (ETDelta7) was purified and analyzed. ETDelta7 exhibited a marked increase in Km (25-fold) for E. coli H-protein (EH) accompanied by a 10-fold decrease in kcat compared with wtET, indicating the importance of the N-terminal region in the interaction with EH. The role of this region in the ET-EH interaction was investigated by cross-linking of wtET-EH or ETDelta7-EH complex with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, a zero-length cross-linker, in the presence of folate substrates. The resulting tripartite cross-linked products were cleaved with lysylendopeptidase and V8 protease. After purification by reversed-phase HPLC, the cross-linked peptides were subjected to Edman sequencing. An intramolecular cross-linking between Asp34 and Lys216 of wtET which was not observed in wtET alone and an intermolecular cross-linking between Lys288 of wtET and Asp-43 of EH were identified. In contrast, no such cross-linking was detected from the cross-linked product of ETDelta7. These results suggest that EH, when it interacts with ET, causes a change in conformation of ET and that the N-terminal region of ET is essential for the conformational change leading to the proper interaction with EH.     

Molecular mechanism of the electron transfer reaction in cytochrome P450(cam)--putidaredoxin: roles of glutamine 360 at the heme proximal site

We characterized electron transfer (ET) from putidaredoxin (Pdx) to the mutants of cytochrome P450(cam) (P450(cam)), in which one of the residues located on the putative binding site to Pdx, Gln360, was replaced with Glu, Lys, and Leu. The kinetic analysis of the ET reactions from reduced Pdx to ferric P450(cam) (the first ET) and to ferrous oxygenated P450(cam) (the second ET) showed the dissociation constants (K(m)) that were moderately perturbed for the Lys and Leu mutants and the distinctly increased for the Glu mutant. Although the alterations in K(m) indicate that Gln360 is located at the Pdx binding site, the effects of the Gln360 mutations (0.66-20-fold of that of wild type) are smaller than those of the Arg112 mutants (25-2500-fold of that of wild type) [Unno, M., et al. (1996) J. Biol. Chem. 271, 17869-17874], allowing us to conclude that Gln360 much less contributes to the complexation with Pdx than Arg112. The first ET rate (35 s(-1) for wild-type P450(cam)) was substantially reduced in the Glu mutant (5.4 s(-1)), while less perturbation was observed for the Lys (53 s(-1)) and Leu (23 s(-1)) mutants. In the second ET reaction, the retarded ET rate was detected only in the Glu mutant but not in the Lys and Leu mutants. These results showed the smaller mutational effects of Gln360 on the ET reactions than those of the Arg112 mutants. In contrast to the moderate perturbations in the kinetic parameters, the mutations at Gln360 significantly affected both the standard enthalpy and entropy of the redox reaction of P450(cam), which cause the negative shift of the redox potentials for the Fe(3+)/Fe(2+) couple by 20-70 mV. Since the amide group of Gln360 is located near the carbonyl oxygen of the amide group of the axial cysteine, it is plausible that the mutation at Gln360 perturbs the electronic interaction of the axial ligand with heme iron, resulting in the reduction of the redox potentials. We, therefore, conclude that Gln360 primarily regulates the ET reaction of P450(cam) by modulating the redox potential of the heme iron and not by the specific interaction with Pdx or the formation of the ET pathway that are proposed as the regulation mechanism of Arg112.     

Crystal structure of human T-protein of glycine cleavage system at 2.0 A resolution and its implication for understanding non-ketotic hyperglycinemia

T-protein, a component of the glycine cleavage system, catalyzes the formation of ammonia and 5,10-methylenetetrahydrofolate from the aminomethyl moiety of glycine attached to the lipoate cofactor of H-protein. Several mutations in the human T-protein gene cause non-ketotic hyperglycinemia. To gain insights into the effect of disease-causing mutations and the catalytic mechanism at the molecular level, crystal structures of human T-protein in free form and that bound to 5-methyltetrahydrofolate (5-CH3-H4folate) have been determined at 2.0 A and 2.6 A resolution, respectively. The overall structure consists of three domains arranged in a cloverleaf-like structure with the central cavity, where 5-CH3-H4folate is bound in a kinked shape with the pteridine group deeply buried into the hydrophobic pocket and the glutamyl group pointed to the C-terminal side surface. Most of the disease-related residues cluster around the cavity, forming extensive hydrogen bonding networks. These hydrogen bonding networks are employed in holding not only the folate-binding space but also the positions and the orientations of alpha-helix G and the following loop in the middle region, which seems to play a pivotal role in the T-protein catalysis. Structural and mutational analyses demonstrated that Arg292 interacts through water molecules with the folate polyglutamate tail, and that the invariant Asp101, located close to the N10 group of 5-CH3-H4folate, might play a key role in the initiation of the catalysis by increasing the nucleophilic character of the N10 atom of the folate substrate for the nucleophilic attack on the aminomethyl lipoate intermediate. A clever mechanism of recruiting the aminomethyl lipoate arm to the reaction site seems to function as a way of avoiding the release of toxic formaldehyde.     

Mechanisms of factor Xa-catalyzed cleavage of the factor VIIIa A1 subunit resulting in cofactor inactivation

Activation of factor VIII by factor Xa is followed by proteolytic inactivation resulting from cleavage within the A1 subunit (residues 1-372) of factor VIIIa. Factor Xa attacks two sites in A1, Arg(336), which precedes the highly acidic C-terminal region, and a recently identified site at Lys(36). By using isolated A1 subunit as substrate for proteolysis, production of the terminal fragment, A1(37-336), was shown to proceed via two pathways identified by the intermediates A1(1-336) and A1(37-372) and generated by initial cleavage at Arg(336) and Lys(36), respectively. Appearance of the terminal product by the former pathway was 7-8-fold slower than the product obtained by the latter pathway. The isolated A1 subunit was cleaved slowly, independent of the presence of phospholipid. The A1/A3-C1-C2 dimer demonstrated an approximately 3-fold increased cleavage rate constant, and inclusion of phospholipid further enhanced this value by approximately 2-fold. Although association of A1 or A1(37-372) with A3-C1-C2 enhanced the rate of cleavage at Arg(336), inclusion of A3-C1-C2 did not affect the cleavage at Lys(36) in A1(1-336). A synthetic peptide 337-372 blocked the cleavage at Lys(36) (IC(50) = 230 microm) while showing little if any effect on cleavage at Arg(336). Proteolysis at Lys(36), and to a lesser extent Arg(336), was inhibited in a dose-dependent manner by heparin. These results suggest that inactivating cleavages catalyzed by factor Xa at Lys(36) and Arg(336) are regulated in part by the A3-C1-C2 subunit. Furthermore, cleavage at Lys(36) appears to be selectively modulated by the C-terminal acidic region of A1, a region that may interact with factor Xa via its heparin-binding exosite.     

Mapping of trypsin cleavage and antibody-binding sites and delineation of a dispensable domain in the beta subunit of Escherichia coli RNA polymerase.

We have mapped principal sites in the Escherichia coli RNA polymerase molecule that are exposed to attack by trypsin under limited proteolysis conditions. The 1342-amino acid-long beta subunit is alternatively cleaved at Arg903 or Lys909. The cleavage occurs adjacent to a dispensable domain (residues 940-1040) that is absent in the homologous RNA polymerase subunits from chloroplasts, eukaryotes, and archaebacteria. In E. coli, this region can be disrupted with genetic deletions and insertions without the loss of RNA polymerase function. Insertion of 127 amino acids into this region introduces a new highly labile site for trypsin proteolysis. The dispensable domain carries the epitope for monoclonal antibody PYN-6 (near residue 1000), which can be used for anchoring the catalytically active enzyme on a solid support. We also report the identification of a secondary trypsin cleavage at Arg81 of the beta' subunit within a putative zinc-binding domain that is conserved in prokaryotes and chloroplasts.     

The positive charge at Lys-288 of the glucagon-like peptide-1 (GLP-1) receptor is important for binding the N-terminus of peptide agonists

Lysine-288 in the glucagon-like peptide-1 receptor was predicted to be ideally positioned to play a role in hormone binding. Subsequent mutation of Lys-288 to Ala or Leu greatly reduced hormone affinity, while substitution with Arg had minimal effect. Compared to wild type, the Lys288-Ala receptor had a reduced affinity for three peptide ligands with complete N-terminal sequences but not for their N-truncated analogues. Hence, the role of this positively charged residue, which is conserved at the equivalent position in all other Family B receptors, was determined to be important for receptor interaction with the N-terminal eight residues of peptide agonists.     

Specificity of the wound-induced leucine aminopeptidase (LAP-A) of tomato activity on dipeptide and tripeptide substrates.

Wounding of tomato leaves results in the accumulation of an exoprotease called leucine aminopeptidase (LAP-A) that preferentially hydrolyzes amino acid-p-nitroanilide and -beta-naphthylamide substrates with N-terminal Leu, Met and Arg residues. To determine the substrate specificity of LAP-A on more natural substrates, the rates of hydrolysis of 60 dipeptide and seven tripeptide substrates were determined. For comparison, the specificities of the porcine and Escherichia coli LAPs were evaluated in parallel. Several marked differences in substrate specificities for the animal, plant and prokaryotic LAP enzymes were observed. Substrates with variable N-terminal (P1) residues (Xaa) were evaluated; these substrates had Leu or Gly in the penultimate (P1') position. The plant, animal, and prokaryotic LAPs hydrolyzed dipeptides with N-terminal nonpolar aliphatic (Leu, Val, Ile, and Ala), basic (Arg), and sulfur-containing (Met) residues rapidly, while P1 Asp or Gly were cleaved inefficiently from peptides. Significant differences in the cleavage of dipeptides with P1 aromatic residues (Phe, Tyr, and Trp) were noted. To systematically evaluate the impact of the P1' residue on cleavage of dipeptides, three series of dipeptides (Leu-Xaa, Gly-Xaa, and Arg-Xaa) were evaluated. The P1' residue strongly influenced hydrolysis of dipeptides and the magnitude of its effect was dependent on the P1 residue. P1' Pro, Asp, Lys and Gly slowed the hydrolysis rates of the tomato LAP-A, porcine LAP, and E. coli PepA markedly. Analysis six Arg-Gly-Xaa tripeptides showed that more diversity was tolerated in the P2' position. P2' Arg inhibited tripeptide cleavage by all three enzymes, while P2' Asp enhanced hydrolysis rates for the porcine and prokaryotic LAPs.     

Glycation and post-translational processing of human interferon-gamma expressed in Escherichia coli

Until recently, nonenzymatic glycosylation (glycation) was thought to affect the proteins of long living eukaryotes only. However, in a recent study (Mironova, R., Niwa, T., Hayashi, H., Dimitrova, R., and Ivanov, I. (2001) Mol. Microbiol. 39, 1061-1068), we have shown that glycation takes place in Escherichia coli as well. In the present study, we demonstrate that the post-translational processing (proteolysis and covalent dimerization) observed with cysteineless recombinant human interferon-gamma (rhIFN-gamma) is tightly associated with its in vivo glycation. Our results show that, at the time of isolation, rhIFN-gamma contained early (but not advanced) glycation products. Using reverse phase high performance liquid chromatography in conjunction with fluorescence measurements, enzyme-linked immunosorbent assay, and mass spectrometry, we found that advanced glycation end products arose in rhIFN-gamma during storage. The latter were identified mainly in the Arg/Lys-rich C terminus of the protein, which was also the main target of proteolysis. Mass spectral analysis and N-terminal sequencing revealed four major (Arg140/Arg141, Phe137/Arg138, Met135/Leu136, and Lys131/Arg132) and two minor (Lys109/Ala110 and Arg90/Asp91) cleavage sites in this region. Tryptic peptide mapping indicated that the covalent dimers of rhIFN-gamma originating during storage were formed mainly by lateral cross-linking of the monomer subunits. Antiviral assay showed that proteolysis lowered the antiviral activity of rhIFN-gamma, whereas covalent dimerization completely abolished it.     

Substrate specificity of cucumisin on synthetic peptides

The substrate specificity of cucumisin [EC 3.4.21.25] was identified by the use of the synthetic peptide substrates Leu(m)-Pro-Glu-Ala-Leu(n) (m = 0-4, n = 0-3). Neither Pro-Glu-Ala-Leu (m = 0) nor Leu-Pro-Glu-Ala (n = 0) was cleaved by cucumisin, however other analogus peptides were cleaved between Glu-Ala. The hydrolysis rates of Leu(m)-Pro-Glu-Ala-Leu increased with the increase of m = 1 to 2 and 3, but was however, essentially same with the increase of m = 3 to 4. Similarly, the hydrolysis rates of Leu-Leu-Pro-Glu-Ala-Leu(n) increased with the increase of n = 0 to 1 and 2, but was essentially same with the increase of n = 2 to 3. Then, it was concluded that cucumisin has a S5-S3' subsite length. In order to identify the substrate specificity at P1 position, Leu-Leu-Pro-X-Ala-Leu (X; Gly, Ala, Val, Leu, Ile, Pro, Asp, Glu, Lys, Arg, Asn, Gln, Phe, Tyr, Ser, Thr, Met, Trp, His) were synthesized and digested by cucumisin. Cucumisin showed broad specificity at the P1 position. However, cucumisin did not cleave the C-terminal side of Gly, Ile, Pro, and preferred Leu, Asn, Gln, Thr, and Met, especially Met. Moreover, the substrates, Leu-Leu-Pro-Glu-Y-Leu (Y; Gly, Ala, Ser, Leu, Val, Glu, Lys, Phe) were synthesized and digested by cucumisin. Cucumisin did not cleave the N-terminal side of Val but preferred Gly, Ser, Ala, and Lys especially Ser. The specificity of cucumisin for naturally occurring peptides does not agree strictly with the specificity obtained by synthetic peptides at the P1 or P1' position alone, but it becomes clear that the most of the cleavage sites on naturally occurring peptides by cucumisin contain suitable amino acid residues at P1 and (or) P1' positions. Moreover, cucumisin prefers Pro than Leu at P2 position, indicating that the specificity at P2 position differs from that of papain.     

Mechanisms of plasmin-catalyzed inactivation of factor VIII: a crucial role for proteolytic cleavage at Arg336 responsible for plasmin-catalyzed factor VIII inactivation

Plasmin not only functions as a key enzyme in the fibrinolytic system but also directly inactivates factor VIII and other clotting factors such as factor V. However, the mechanisms of plasmin-catalyzed factor VIII inactivation are poorly understood. In this study, levels of factor VIII activity increased approximately 2-fold within 3 min in the presence of plasmin, and subsequently decreased to undetectable levels within 45 min. This time-dependent reaction was not affected by von Willebrand factor and phospholipid. The rate constant of plasmin-catalyzed factor VIIIa inactivation was approximately 12- and approximately 3.7-fold greater than those mediated by factor Xa and activated protein C, respectively. SDS-PAGE analysis showed that plasmin cleaved the heavy chain of factor VIII into two terminal products, A1(37-336) and A2 subunits, by limited proteolysis at Lys(36), Arg(336), Arg(372), and Arg(740). The 80-kDa light chain was converted into a 67-kDa subunit by cleavage at Arg(1689) and Arg(1721), identical to the pattern induced by factor Xa. Plasmin-catalyzed cleavage at Arg(336) proceeded faster than that at Arg(372), in contrast to proteolysis by factor Xa. Furthermore, breakdown was faster than that in the presence of activated protein C, consistent with rapid inactivation of factor VIII. The cleavages at Arg(336) and Lys(36) occurred rapidly in the presence of A2 and A3-C1-C2 subunits, respectively. These results strongly indicated that cleavage at Arg(336) was a central mechanism of plasmin-catalyzed factor VIII inactivation. Furthermore, the cleavages at Arg(336) and Lys(36) appeared to be selectively regulated by the A2 and A3-C1-C2 domains, respectively, interacting with plasmin.     

The crystal structure of leucyl/phenylalanyl-tRNA-protein transferase from Escherichia coli

Leucyl/phenylalanyl-tRNA-protein transferase (L/F-transferase) is an N-end rule pathway enzyme, which catalyzes the transfer of Leu and Phe from aminoacyl-tRNAs to exposed N-terminal Arg or Lys residues of acceptor proteins. Here, we report the 1.6 A resolution crystal structure of L/F-transferase (JW0868) from Escherichia coli, the first three-dimensional structure of an L/F-transferase. The L/F-transferase adopts a monomeric structure consisting of two domains that form a bilobate molecule. The N-terminal domain forms a small lobe with a novel fold. The large C-terminal domain has a highly conserved fold, which is observed in the GCN5-related N-acetyltransferase (GNAT) family. Most of the conserved residues of L/F-transferase reside in the central cavity, which exists at the interface between the N-terminal and C-terminal domains. A comparison of the structures of L/F-transferase and the bacterial peptidoglycan synthase FemX, indicated a structural homology in the C-terminal domain, and a similar domain interface region. Although the peptidyltransferase function is shared between the two proteins, the enzymatic mechanism would differ. The conserved residues in the central cavity of L/F-transferase suggest that this region is important for the enzyme catalysis.     

Proteolytic Cleavage of the Linker Region of the Human P-glycoprotein Modulates Its ATPase Function

P-glycoprotein (Pgp), an anticancer drug-translocating ATPase, is responsible for multidrug resistance in cancer. We have previously shown (Nuti, S. L., Mehdi, A., and Rao, U. S. (2000) Biochemistry 39, 3424-3432) that tryptic cleavage of Pgp results in the activation of basal and drug-stimulated ATPase functions of Pgp. To understand this phenomenon, we determined the sites cleaved by trypsin and further examined whether the modulation of Pgp function is trypsin-specific or the result of proteolysis in general. The effects of chymotrypsin and proteinase K on Pgp ATPase function were studied. The results show that proteolysis of Pgp irrespective of the protease employed resulted in the activation of basal ATPase activity. However, drug-stimulated ATPase activities were differentially modulated. Immunoblot analysis of proteolytic digests indicated that, irrespective of the protease employed, Pgp was predominantly cleaved in the middle of the molecule. N-terminal amino acid sequencing of Pgp tryptic and chymotryptic peptides indicated Arg(680) and Leu(682) as the sites of cleavage, respectively. These two cleavage sites are part of the predicted linker region that joins the two halves of Pgp. Together, these results suggest that the linker region in Pgp is primarily accessible to protease action and that cleavage of this region modulates Pgp ATPase function.     

Bivalent inhibitor of the N-end rule pathway.

The N-end rule relates the in vivo half-life of a protein to the identity of its N-terminal residue. Ubr1p, the recognition (E3) component of the Saccharomyces cerevisiae N-end rule pathway, contains at least two substrate-binding sites. The type 1 site is specific for N-terminal basic residues Arg, Lys, and His. The type 2 site is specific for N-terminal bulky hydrophobic residues Phe, Leu, Trp, Tyr, and Ile. Previous work has shown that dipeptides bearing either type 1 or type 2 N-terminal residues act as weak but specific inhibitors of the N-end rule pathway. We took advantage of the two-site architecture of Ubr1p to explore the feasibility of bivalent N-end rule inhibitors, whose expected higher efficacy would result from higher affinity of the cooperative (bivalent) binding to Ubr1p. The inhibitor comprised mixed tetramers of beta-galactosidase that bore both N-terminal Arg (type 1 residue) and N-terminal Leu (type 2 residue) but that were resistant to proteolysis in vivo. Expression of these constructs in S. cerevisiae inhibited the N-end rule pathway much more strongly than the expression of otherwise identical beta-galactosidase tetramers whose N-terminal residues were exclusively Arg or exclusively Leu. In addition to demonstrating spatial proximity between the type 1 and type 2 substrate-binding sites of Ubr1p, these results provide a route to high affinity inhibitors of the N-end rule pathway.     

Limited digestion of calmodulin with trypsin in the presence or absence of various metal ions

The limited proteolysis of bovine brain calmodulin with trypsin in the presence or absence of various metal ions was reinvestigated in detail by HPLC. With metal ion-free calmodulin, limited proteolysis occurred at Arg 37 and Arg 106 with a cleavage ratio of 1 to 5, resulting in fragments consisting of residues 1-37, 38-148, 1-106 and 107-148. Fragments 1-37 and 107-148 accumulated under metal ion-free conditions. In the presence of calcium ions, the susceptibility of these sites to trypsin decreased and limited proteolysis occurred at Lys 77 as already reported by other workers. Fragment 78-148 accumulated, whereas fragment 1-77 was unstable under calcium-bound conditions, giving smaller peptides. Upon binding of manganese ions, calmodulin underwent a change of susceptibility to trypsin, resulting in cleavage at Lys 77, as observed for calcium-bound calmodulin. In the presence of zinc or magnesium ions, calmodulin was cleaved at the same sites as metal ion-free calmodulin under conditions where calmodulin would be expected to bind the respective ions.     

Limited proteolysis of Escherichia coli cytidine 5'-triphosphate synthase. Identification of residues required for CTP formation and GTP-dependent activation of glutamine hydrolysis.

Cytidine 5'-triphosphate synthase catalyses the ATP-dependent formation of CTP from UTP using either ammonia or l-glutamine as the source of nitrogen. When glutamine is the substrate, GTP is required as an allosteric effector to promote catalysis. Limited trypsin-catalysed proteolysis, Edman degradation, and site-directed mutagenesis were used to identify peptide bonds C-terminal to three basic residues (Lys187, Arg429, and Lys432) of Escherichia coli CTP synthase that were highly susceptible to proteolysis. Lys187 is located at the CTP/UTP-binding site within the synthase domain, and cleavage at this site destroyed all synthase activity. Nucleotides protected the enzyme against proteolysis at Lys187 (CTP > ATP > UTP > GTP). The K187A mutant was resistant to proteolysis at this site, could not catalyse CTP formation, and exhibited low glutaminase activity that was enhanced slightly by GTP. K187A was able to form tetramers in the presence of UTP and ATP. Arg429 and Lys432 appear to reside in an exposed loop in the glutamine amide transfer (GAT) domain. Trypsin-catalyzed proteolysis occurred at Arg429 and Lys432 with a ratio of 2.6 : 1, and nucleotides did not protect these sites from cleavage. The R429A and R429A/K432A mutants exhibited reduced rates of trypsin-catalyzed proteolysis in the GAT domain and wild-type ability to catalyse NH3-dependent CTP formation. For these mutants, the values of kcat/Km and kcat for glutamine-dependent CTP formation were reduced approximately 20-fold and approximately 10-fold, respectively, relative to wild-type enzyme; however, the value of Km for glutamine was not significantly altered. Activation of the glutaminase activity of R429A by GTP was reduced 6-fold at saturating concentrations of GTP and the GTP binding affinity was reduced 10-fold. This suggests that Arg429 plays a role in both GTP-dependent activation and GTP binding.     

The specificity of carboxypeptidase Y may be altered by changing the hydrophobicity of the S'1 binding pocket.

The S'1 binding pocket of carboxypeptidase Y is hydrophobic, spacious, and open to solvent, and the enzyme exhibits a preference for hydrophobic P'1 amino acid residues. Leu272 and Ser297, situated at the rim of the pocket, and Leu267, slightly further away, have been substituted by site-directed mutagenesis. The mutant enzymes have been characterized kinetically with respect to their P'1 substrate preferences using the substrate series FA-Ala-Xaa-OH (Xaa = Leu, Glu, Lys, or Arg) and FA-Phe-Xaa-OH (Xaa = Ala, Val, or Leu). The results reveal that hydrophobic P'1 residues bind in the vicinity of residue 272 while positively charged P'1 residues interact with Ser297. Introduction of Asp or Glu at position 267 greatly reduced the activity toward hydrophobic P'1 residues (Leu) and increased the activity two- to three-fold for the hydrolysis of substrates with Lys or Arg in P'1. Negatively charged substituents at position 272 reduced the activity toward hydrophobic P'1 residues even more, but without increasing the activity toward positively charged P'1 residues. The mutant enzyme L267D + L272D was found to have a preference for substrates with C-terminal basic amino acid residues. The opposite situation, where the positively charged Lys or Arg were introduced at one of the positions 267, 272, or 297, did not increase the rather low activity toward substrates with Glu in the P'1 position but greatly reduced the activity toward substrates with C-terminal Lys or Arg due to electrostatic repulsion. The characterized mutant enzymes exhibit various specificities, which may be useful in C-terminal amino acid sequence determinations.     

Proteolysis of non-phosphorylated and phosphorylated tau by thrombin

The microtubule-associated protein tau aggregates intracellularly by unknown mechanisms in Alzheimer's disease and other tauopathies. A contributing factor may be a failure to break down free cytosolic tau, thus creating a surplus for aggregation, although the proteases that degrade tau in brain remain unknown. To address this issue, we prepared cytosolic fractions from five normal human brains and from perfused rat brains and incubated them with or without protease inhibitors. D-Phenylalanyl-L-prolylarginyl chloromethyl ketone, a thrombin-specific inhibitor, prevented tau breakdown in these fractions, suggesting that thrombin is a brain protease that processes tau. We next exposed human recombinant tau to purified human thrombin and analyzed the fragments by N-terminal sequencing. We found that thrombin proteolyzed tau at multiple arginine and lysine sites. These include Arg(155)-Gly(156), Arg(209)-Ser(210), Arg(230)-Thr(231), Lys(257)-Ser(258), and Lys(340)-Ser(341) (numbering according to the longest human tau isoform). Temporally, the initial cleavage occurred at the Arg(155)-Gly(156) bond. Proteolysis of the resultant C-terminal tau fragment then proceeded bidirectionally. When tau was phosphorylated by glycogen synthase kinase-3beta, most of these proteolytic processes were inhibited, except for the first cleavage at the Arg(155)-Gly(156) bond. Furthermore, paired helical filament tau prepared from Alzheimer's disease brain was more resistant to thrombin proteolysis than following dephosphorylation by alkaline phosphatase. The results suggest a possible role for thrombin in proteolysis of tau under physiological and/or pathological conditions in human brains. They are consistent with the hypothesis that phosphorylation of tau inhibits proteolysis by thrombin or other endogenous proteases, leading to aggregation of tau into insoluble fibrils.     

Channelling and formation of 'active' formaldehyde in dimethylglycine oxidase

Here we report crystal structures of dimethylglycine oxidase (DMGO) from the bacterium Arthrobacter globiformis, a bifunctional enzyme that catalyzes the oxidation of N,N-dimethyl glycine and the formation of 5,10-methylene tetrahydrofolate. The N-terminal region binds FAD covalently and oxidizes dimethylglycine to a labile iminium intermediate. The C-terminal region binds tetrahydrofolate, comprises three domains arranged in a ring-like structure and is related to the T-protein of the glycine cleavage system. The complex with folinic acid indicates that this enzyme selectively activates the N10 amino group for initial attack on the substrate. Dead-end reactions with oxidized folate are avoided by the strict stereochemical constraints imposed by the folate-binding funnel. The active sites in DMGO are approximately 40 A apart, connected by a large irregular internal cavity. The tetrahydrofolate-binding funnel serves as a transient entry-exit port, and access to the internal cavity is controlled kinetically by tetrahydrofolate binding. The internal cavity enables sequestration of the reactive iminium intermediate prior to reaction with tetrahydrofolate and avoids formation of toxic formaldehyde. This mode of channelling in DMGO is distinct from other channelling mechanisms.     

Mutation in transforming growth factor beta induced protein associated with granular corneal dystrophy type 1 reduces the proteolytic susceptibility through local structural stabilization

Hereditary mutations in the transforming growth factor beta induced (TGFBI) gene cause phenotypically distinct corneal dystrophies characterized by protein deposition in cornea. We show here that the Arg555Trp mutant of the fourth fasciclin 1 (FAS1-4) domain of the protein (TGFBIp/keratoepithelin/βig-h3), associated with granular corneal dystrophy type 1, is significantly less susceptible to proteolysis by thermolysin and trypsin than the WT domain. High-resolution liquid-state NMR of the WT and Arg555Trp mutant FAS1-4 domains revealed very similar structures except for the region around position 555. The Arg555Trp substitution causes Trp555 to be buried in an otherwise empty hydrophobic cavity of the FAS1-4 domain. The first thermolysin cleavage in the core of the FAS1-4 domain occurs on the N-terminal side of Leu558 adjacent to the Arg555 mutation. MD simulations indicated that the C-terminal end of helix α3′ containing this cleavage site is less flexible in the mutant domain, explaining the observed proteolytic resistance. This structural change also alters the electrostatic properties, which may explain increased propensity of the mutant to aggregate in vitro with 2,2,2-trifluoroethanol. Based on our results we propose that the Arg555Trp mutation disrupts the normal degradation/turnover of corneal TGFBIp, leading to accumulation and increased propensity to aggregate through electrostatic interactions.     

Domain-domain interactions of HtpG, an Escherichia coli homologue of eukaryotic HSP90 molecular chaperone.

In the present study, we investigated the domain structure and domain-domain interactions of HtpG, an Escherichia coli homologue of eukaryotic HSP90. Limited proteolysis of recombinant HtpG, revealed three major tryptic sites, i.e. Arg7-Gly8, Arg336-Glu337 and Lys552-Leu553, of which the latter two were located at the positions equivalent to the major cleavage sites of human HSP90alpha. A similar pattern was obtained by papain treatment under nondenaturing conditions but not under denaturing conditions. Thus, HtpG consists of three domains, i.e. Domain A, Met1-Arg336; domain B, Glu337-Lys552; and domain C, Leu553-Ser624, as does HSP90. The domains of HtpG were expressed and their interactions were estimated on polyacrylamide gel electrophoresis under nondenaturing conditions. As a result, two kinds of domain-domain interactions were revealed: domain B interaction with domain A of the same polypeptide and domain C of one partner with domain B of the other in the dimer. Domain B could be structurally and functionally divided into two subdomains, the N-terminal two-thirds (subdomain BI) that interacted with domain A and the C-terminal one-third (subdomain BII) that interacted with domain C. The C-terminal two-thirds of domain A, i.e. Asp116-Arg336, were sufficient for the binding to domain B. We finally propose the domain organization of an HtpG dimer.