Semisynthesis of cytotoxic proteins using a modified protein splicing element

Two cytotoxic proteins, bovine pancreatic ribonuclease A (RNase A), and a restriction endonuclease from Haemophilus parainfluenzae (HpaI), were produced using a novel semisynthetic approach that utilizes a protein splicing element, an intein, to generate a reactive thioester at the C-terminus of a recombinant protein. Nucleophilic attack on this thioester by the N-terminal cysteine of a synthetic peptide ultimately leads to the ligation of the two reactants through a native peptide bond. This strategy was used to produce RNase A and HpaI by isolating inactive truncated forms of these proteins, the first 109 and 223 amino acids of RNase A and HpaI, respectively, as fusion proteins consisting of the target protein, an intein, and a chitin binding domain. Thiol-induced cleavage of the precursor led to the liberation of the target protein with a C-terminal thioester-tag. Addition of synthetic peptides representing the amino acids missing from the truncated forms led to the generation of full-length products that displayed catalytic activity indicative of the wild-type enzymes. The turnover numbers and Km for ligated and renatured RNase A were 8.2 s(-1) and 1.5 mM, in good agreement with reported values of 8.3 s(-1) and 1.2 mM (Hodges & Merrifield, 1975). Ligated HpaI had a specific activity of 0.5-1.5 x 10(6) U/mg, which compared favorably with the expected value of 1-2 x 10(6) U/mg (J. Benner, unpubl. obs.). Besides assisting in the production of cytotoxic proteins, this technique could allow the easy insertion of unnatural amino acids into a protein sequence.     

Autoproteolysis of herpes simplex virus type 1 protease releases an active catalytic domain found in intermediate capsid particles

The UL26 gene of herpes simplex virus type 1 (HSV-1) encodes a 635-amino-acid protease that cleaves itself and the HSV-1 assembly protein ICP35cd (F. Liu and B. Roizman, J. Virol. 65:5149-5156, 1991). We previously examined the HSV protease by using an Escherichia coli expression system (I. C. Deckman, M. Hagen, and P. J. McCann III, J. Virol. 66:7362-7367, 1992) and identified two autoproteolytic cleavage sites between residues 247 and 248 and residues 610 and 611 of UL26 (C. L. DiIanni, D. A. Drier, I. C. Deckman, P. J. McCann III, F. Liu, B. Roizman, R. J. Colonno, and M. G. Cordingley, J. Biol. Chem. 268:2048-2051, 1993). In this study, a series of C-terminal truncations of the UL26 open reading frame was tested for cleavage activity in E. coli. Our results delimit the catalytic domain of the protease to the N-terminal 247 amino acids of UL26 corresponding to No, the amino-terminal product of protease autoprocessing. Autoprocessing of the full-length protease was found to be unnecessary for catalysis, since elimination of either or both cleavage sites by site-directed mutagenesis fails to prevent cleavage of ICP35cd or an unaltered protease autoprocessing site. Catalytic activity of the 247-amino-acid protease domain was confirmed in vitro by using a glutathione-S-transferase fusion protein. The fusion protease was induced to high levels of expression, affinity purified, and used to cleave purified ICP35cd in vitro, indicating that no other proteins are required. By using a set of domain-specific antisera, all of the HSV-1 protease cleavage products predicted from studies in E. coli were identified in HSV-1-infected cells. At least two protease autoprocessing products, in addition to fully processed ICP35cd (ICP35ef), were associated with intermediate B capsids in the nucleus of infected cells, suggesting a key role for proteolytic maturation of the protease and ICP35cd in HSV-1 capsid assembly.     

Identification of the SH3 domain of GAP as an essential sequence for Ras-GAP-mediated signaling

Guanosine triphosphatase activating protein (GAP) is an essential component of Ras signaling pathways. GAP functions in different cell types as a deactivator and a transmitter of cellular Ras signals. A domain (amino acids 275 to 351) encompassing the Src homology region 3 (SH3) of GAP was found to be essential for GAP signaling. A monoclonal antibody was used to block germinal vesicle breakdown (GVBD) induced by the oncogenic protein Ha-ras Lys12 in Xenopus oocytes. The monoclonal antibody, which was found to recognize the peptide containing amino acids 275 to 351 within the amino-terminal domain of GAP, did not modify the stimulation of the Ha-Ras-GTPase by GAP. Injection of peptides corresponding to amino acids 275 to 351 and 317 to 326 blocked GVBD induced by insulin or by Ha-Ras Lys12 but not that induced by progesterone. These findings confirm that GAP is an effector for Ras in Xenopus oocytes and that the SH3 domain is essential for signal transduction.     

A novel DNA binding and nuclease activity in domain III of Mu transposase: evidence for a catalytic region involved in donor cleavage

The Mu A protein is a 75 kDa transposase organized into three structural domains. By severing the C-terminal region (domain III) from the remainder of the protein, we unmasked a novel non-specific DNA binding and nuclease activity in this region. Deletion analysis localized both activities to a 26 amino acid stretch (aa 575-600) which remarkably remained active in DNA binding and cleavage. The two activities were shown to be tightly linked by site-directed mutagenesis. To study the importance of these activities in the transposition process, an intact mutant transposase lacking the DNA binding and nuclease activity of domain III was constructed and purified. The mutant transposase was indistinguishable from wild-type Mu A in binding affinity for both the Mu ends and the enhancer, and in strand transfer activity when the cleavage step was bypassed. In contrast, the mutant transposase displayed defects in both synapsis and donor cleavage. Our results strongly suggest that the 26 amino acid region in domain III carries catalytic residues required for donor DNA cleavage by Mu A protein. Furthermore, our data suggest that an active site for donor cleavage activity in the Mu tetramer is assembled from domain II (metal ion binding) in one A monomer and domain III (DNA cleavage) in a separate A monomer. This proposal for active site assembly is in agreement with the recently proposed domain sharing model by Yang et al. (Yang, J.Y., Kim, K., Jayaram, M. and Harshey, R.M. [1995] EMBO J., 14, 2374-2384).     

Green fluorescent protein as a reporter in rapid screening of antituberculosis compounds in vitro and in macrophages

Trypanosoma cruzi trans-sialidase consists of a C-terminal domain composed essentially of immunodominant amino acid repeat units (SAPA-repeats) and an amino region responsible for the enzymatic activity (catalytic domain). To investigate the possible function(s) of SAPA-repeats, recombinant trans-sialidases either containing or lacking the C-terminal region were tested in mice. The presence of SAPA-repeats in the intravenously injected protein has two consequences. First, they enhance the persistence of the trans-sialidase activity in blood. Second, SAPA-repeats promoted the production of antibodies directed to the catalytic domain that inhibit trans-sialidase activity. These results suggest that SAPA-repeats modulate the trans-sialidase activity in blood.     

Modulation of protein splicing of the Saccharomyces cerevisiae vacuolar membrane ATPase intein

Protein splicing of the Saccharomyces cerevisiae vacuolar membrane ATPase intein involves four highly coordinated reactions that result in precise cleavage and formation of peptide bonds. In this study, we investigated the roles of the last N-extein residue (-1 residue) and the intein penultimate residue in modulating splicing reactions. Most of the 20 amino acid substitutions at the -1 position had no effect on overall protein splicing but could lead to significant accumulation of thioester intermediates when splicing was blocked by mutation. A subset of -1 substitutions attenuated the initiation of protein splicing and enabled us to demonstrate in vitro splicing of a mesophilic intein containing all wild-type catalytic residues. Substitutions involving the intein penultimate residue allowed modulation of the branch resolution and C-terminal cleavage reaction. Our data suggest that the N-S acyl rearrangement, which initiates splicing, may also serve as the rate-limiting step. Through appropriate amino acid substitutions, we were able to modulate splicing reactions in vitro by change in pH or temperature or addition of thiol reagents. Both insertion and deletion were tolerated in the central region of the intein although splicing or structure of the intein may have been affected.     

Multidomain structure and cellulosomal localization of the Clostridium thermocellum cellobiohydrolase CbhA

The nucleotide sequence of the Clostridium thermocellum F7 cbhA gene, coding for the cellobiohydrolase CbhA, has been determined. An open reading frame encoding a protein of 1,230 amino acids was identified. Removal of a putative signal peptide yields a mature protein of 1,203 amino acids with a molecular weight of 135,139. Sequence analysis of CbhA reveals a multidomain structure of unusual complexity consisting of an N-terminal cellulose binding domain (CBD) homologous to CBD family IV, an immunoglobulin-like beta-barrel domain, a catalytic domain homologous to cellulase family E1, a duplicated domain similar to fibronectin type III (Fn3) modules, a CBD homologous to family III, a highly acidic linker region, and a C-terminal dockerin domain. The cellulosomal localization of CbhA was confirmed by Western blot analysis employing polyclonal antibodies raised against a truncated enzymatically active version of CbhA. CbhA was identified as cellulosomal subunit S3 by partial amino acid sequence analysis. Comparison of the multidomain structures indicates striking similarities between CbhA and a group of cellulases from actinomycetes. Average linkage cluster analysis suggests a coevolution of the N-terminal CBD and the catalytic domain and its spread by horizontal gene transfer among gram-positive cellulolytic bacteria.     

Characterization of the lys2 gene of Penicillium chrysogenum encoding alpha-aminoadipic acid reductase

A DNA fragment containing a gene homologous to LYS2 gene of Saccharomyces cerevisiae was cloned from a genomic DNA library of Penicillium chrysogenum AS-P-78. It encodes a protein of 1409 amino acids (Mr 154859) with strong similarity to the S. cerevisiae (49.9% identity) Schizosaccharomyces pombe (51.3% identity) and Candida albicans (48.12% identity) alpha-aminoadipate reductases and a lesser degree of identity to the amino acid-activating domains of the non-ribosomal peptide synthetases, including the alpha-aminoadipate-activating domain of the alpha-aminoadipyl-cysteinyl-valine synthetase of P. chrysogenum (12.4% identical amino acids). The lys2 gene contained one intron in the 5'-region and other in the 3'-region, as shown by comparing the nucleotide sequences of the cDNA and genomic DNA, and was transcribed as a 4.7-kb monocistronic mRNA. The lys2 gene was localized on chromosome III (7.5 Mb) in P. chrysogenum AS-P-78 and on chromosome IV (5.6 Mb) in strain P2, whereas the penicillin gene cluster is known to be located in chromosome I in both strains. The lys2-encoded protein is a member of the aminoacyladenylate-forming enzyme family with a reductase domain in its C-terminal region.     

Circulatory arrest

UDP-GlcNAc: alpha-3-D-mannoside beta-1,2-N-acetylglucosaminyltransferase I (GnTI, EC 2.4.1.101) plays an essential role in the conversion of oligomannose to complex and hybrid N-glycans. Rabbit GnTI is 447 residues long and has a short four-residue N-terminal cytoplasmic tail, a 25-residue putative signal-anchor hydrophobic domain, a stem region of undetermined length and a large C-terminal catalytic domain, a structure typical of all glycosyltransferases cloned to date. Comparison of the amino acid sequences for human, rabbit, mouse, rat, chicken, frog and Caenorhabditis elegans GnTI was used to obtain a secondary structure prediction for the enzyme which suggested that the location of the junction between the stem and the catalytic domain was at about residue 106. To test this hypothesis, several hybrid constructs containing GnT I with N- and C-terminal truncations fused to a mellitin signal sequence were inserted into the genome of Autographa californica nuclear polyhedrosis virus (AcMNPV), Sf9 insect cells were infected with the recombinant baculovirus and supernatants were assayed for GnTI activity. Removal of 29, 84 and 106 N-terminal amino acids had no effect on GnTI activity; however, removal of a further 14 amino acids resulted in complete loss of activity. Western blot analysis showed strong protein bands for all truncated enzymes except for the construct lacking 120 N-terminal residues indicating proteolysis or defective expression or secretion of this protein. The data indicate that the stem is at least 77 residues long.     

A common 40 amino acid motif in eukaryotic RNases H1 and caulimovirus ORF VI proteins binds to duplex RNAs

Eukaryotic RNases H from Saccharomyces cerevisiae , Schizosaccharomyces pombe and Crithidia fasciculata , unlike the related Escherichia coli RNase HI, contain a non-RNase H domain with a common motif. Previously we showed that S.cerevisiae RNase H1 binds to duplex RNAs (either RNA-DNA hybrids or double-stranded RNA) through a region related to the double-stranded RNA binding motif. A very similar amino acid sequence is present in caulimovirus ORF VI proteins. The hallmark of the RNase H/caulimovirus nucleic acid binding motif is a stretch of 40 amino acids with 11 highly conserved residues, seven of which are aromatic. Point mutations, insertions and deletions indicated that integrity of the motif is important for binding. However, additional amino acids are required because a minimal peptide containing the motif was disordered in solution and failed to bind to duplex RNAs, whereas a longer protein bound well. Schizosaccharomyces pombe RNase H1 also bound to duplex RNAs, as did proteins in which the S.cerevisiae RNase H1 binding motif was replaced by either the C.fasciculata or by the cauliflower mosaic virus ORF VI sequence. The similarity between the RNase H and the caulimovirus domain suggest a common interaction with duplex RNAs of these two different groups of proteins.     

Primary structure of matrilin-3, a new member of a family of extracellular matrix proteins related to cartilage matrix protein (matrilin-1) and von Willebrand factor

A mouse cDNA encoding for matrilin-3, the third member of the novel matrilin family of extracellular matrix proteins, was cloned. The protein precursor of 481 amino acids consists of a putative signal peptide, a short positively charged sequence, a single vWFA-like domain followed by four epidermal growth factor-like modules and a potential coiled-coil alpha-helical oligomerization domain at the C-terminus. It is the smallest member of the matrilin family with a predicted Mr of the mature protein of 48 902. The primary structure of a C-terminal portion of 310 amino acids of the human matrilin-3 was determined and showed a sequence identity to the mouse matrilin-3 of 84.8%. Northern blot hybridization of mouse matrilin-3 mRNA showed a 2.9 kb mRNA expressed in sternum, femur and trachea and indicates a cartilage-specific expression.     

The molecular interaction of Fas and FAP-1. A tripeptide blocker of human Fas interaction with FAP-1 promotes Fas-induced apoptosis

Fas (APO-1/CD95), which is a member of the tumor necrosis factor receptor superfamily, is a cell surface receptor that induces apoptosis. A protein tyrosine phosphatase, Fas-associated phosphatase-1 (FAP-1), that was previously identified as a Fas binding protein interacts with the C-terminal 15 amino acids of the regulatory domain of the Fas receptor. To identify the minimal region of the Fas C-terminal necessary for binding to FAP-1, we employed an in vitro inhibition assay of Fas/FAP-1 binding using a series of synthetic peptides as well as a screen of random peptide libraries by the yeast two-hybrid system. The results showed that the C-terminal three amino acids (SLV) of human Fas were necessary and sufficient for its interaction with the third PDZ (GLGF) domain of FAP-1. Furthermore, the direct cytoplasmic microinjection of this tripeptide (Ac-SLV) resulted in the induction of Fas-mediated apoptosis in a colon cancer cell line that expresses both Fas and FAP-1. Since t(S/T)X(V/L/I) motifs in the C termini of several other receptors have been shown to interact with PDZ domain in signal transducing molecules, this may represent a general motif for protein-protein interactions with important biological functions.     

Chronic neurologic effects of pesticide overexposure

To investigate the molecular mechanisms involved in paramyxovirus-induced cell fusion, the function and structure of a peptide with a 20-amino-acid sequence from the leucine zipper region (heptad repeat region 2) of the Newcastle disease virus fusion protein (F) were characterized. A peptide with the sequence ALDKLEESNSKLDKVNVKLT (amino acids 478-497 of the F protein) was found to inhibit syncytia formation after virus infection and after transfection of Cos cells with the HN (hemagglutinin-neuraminidase) and F protein cDNAs. Using an F protein gene that requires addition of exogenous trypsin for cleavage, it was shown that the peptide exerted its inhibitory effect prior to cleavage. The three-dimensional conformation of the peptide in aqueous solution was determined through the use of NMR and molecular modeling. Results showed that the peptide formed a helix with properties between an alpha-helix and a 3(10)-helix and that leucine residues aligned along one face of the helix. Side chain salt bridges and hydrogen bonds likely contributed to the stability of the peptide secondary structure. Analysis of the aqueous solution conformation of the peptide suggested mechanisms for specificity of interaction with the intact F protein.     

The cleavage of pro-urokinase type plasminogen activator by stromelysin-1

Membrane binding of urokinase type plasminogen activator (u-PA) is thought to play a pivotal role in connective tissue remodeling and invasive processes. We compare the ability of different matrix-metalloproteinases involved in connective tissue turnover to cleave pro-urokinase type plasminogen activator between the catalytic domain and the receptor binding part to investigate a potential role for matrix-metalloproteinases in the regulation of membrane-associated proteolytic activity. We employed several forms of human stromelysin-1 (full length, C-truncated, and recombinant catalytic domain), rabbit C-truncated stromelysin-1, the human gelatinases A and B and the human catalytic domain of neutrophil collagenase. The gelatinases and the collagenase did not separate the receptor binding domain of pro-urokinase type plasminogen activator from the catalytic domain, whereas all stromelysin-1 forms cleaved the glutamic acid 143-leucine 144 bond of pro-urokinase type plasminogen activator. This reaction could be inhibited by specific inhibitors of matrix metalloproteinases and was not affected by inhibitors of serine proteinases. The M(r) 31000 cleavage product with leucine 144 as N-terminus displayed no proteolytic activity towards the pro-urokinase type plasminogen activator substrate pyroGlu-Gly-Arg-pNA-HCI (S2444), but it could be activated by an additional treatment with plasmin. Comparison between full length stromelysin-1 and its C-truncated forms, showed that both exhibited the same cleavage properties towards pro-urokinase type plasminogen activator. Thus, the cleavage of pro-urokinase type plasminogen activator by stromelysin-1 is not influenced by the presence or absence of the C-terminal domain. The recombinant catalytic domain of MMP-3 generated pro-urokinase type plasminogen activator, whereas incubation of pro-urokinase type plasminogen activator with the native forms of human or rabbit stromelysin-1 led to a moderate activation of pro-uPA due to an additional cleavage that is catalyzed by a serine proteinase.     

The microtubule-associated protein tau cross-links to two distinct sites on each alpha and beta tubulin monomer via separate domains

The interaction between tubulin subunits and microtubule-associated proteins (MAPs) such as tau is fundamental for microtubule structure and function. Previous work has suggested that the "microtubule binding domain" of tau (composed of three or four imperfect 18-amino acid repeats, separated by 13- or 14-amino acid inter-repeat regions) can bind to the C-terminal ends of both alpha and beta tubulin monomers. Here, using covalent cross-linking strategies, we demonstrate that there are two distinct tau cross-linking sites (designated as "C-terminal" and "internal") on each alpha and beta tubulin monomer. The C-terminal tau cross-linking site is located within the 12 C-terminal amino acids of both alpha and beta tubulin, while the internal tau cross-linking site is located within the C-terminal one-third of alpha and beta tubulin but not within the last 12 amino acids. In addition, we show that tau cross-links to the C-terminal site via its repeat 1 and/or the R1-R2 inter-repeat. The cross-linking of tau to the internal site is mediated by some subset of its other repeat units. Integrating these and earlier data with the 3.7 A resolution model of the alphabeta tubulin dimer recently presented by E. Nogales et al. [(1998), Nature 391, 199-203], we propose a new model for the tau-microtubule interaction.     

Drosophila Rrp1 domain structure as defined by limited proteolysis and biophysical analyses

Drosophila Rrp1 is a DNA repair nuclease whose C-terminal region shares extensive homology with Escherichia coli exonuclease III, has nuclease activity, and provides resistance to oxidative and alkylating agents in repair-deficient E. coli strains. The N-terminal 421 amino acid region of Rrp1, which binds and renatures homologous single-stranded DNA, does not share homology with any known protein. Proteolysis by endoproteinase Glu-C (protease V8) reduces the Rrp1 protein to a single, cleavage-resistant peptide. The peptide (referred to as Rrp1-C274) begins with the sequence TKTTV, corresponding to cleavage between Glu-405 and Thr-406 of Rrp1. We determined that nuclease activity is intrinsic to Rrp1-C274 although altered when compared with Rrp1; 3'-exonuclease activity is reduced 210-fold, 3'-phosphodiesterase activity is reduced 6.8-fold, and no difference in apurinic/apyrimidinic endonuclease activity is observed. Rrp1 and Rrp1-C274 are both monomers with frictional coefficients of 2.2 and 1.4, respectively. Circular dichroism results indicate that Rrp1-C274 is predominantly alpha-helical, while the N-terminal 399 amino acids is predominantly random coil. These results suggest that Rrp1 may have a bipartite structural organization; a highly organized, globular C-terminal domain; and an asymmetric, protease-sensitive random coil-enriched N-terminal region. A shape model for this bipartite structure is proposed and discussed.     

The activation of prothrombin by the prothrombinase complex. The contribution of the substrate-membrane interaction to catalysis

The conversion of prothrombin to thrombin requires the cleavage of two peptide bonds and is catalyzed by the prothrombinase complex composed of factors Xa and Va assembled on a membrane surface. Presteady-state kinetic studies of the effects of membranes on the proteolytic reaction were undertaken using model membranes composed of phosphatidylcholine and phosphatidylserine (PCPS). The concentration of PCPS was varied to alter the concentration of free phospholipid available for substrate binding without influencing the concentration of membrane-assembled prothrombinase. In fluorescence stopped-flow measurements, increasing concentrations of PCPS resulted in an increase in the rate of product formation. Assessment of bond cleavage by sodium dodecyl sulfate-polyacrylamide gel electrophoresis following rapid chemical quench using 125I-prothrombin revealed that the activation reaction proceeded through the ordered cleavage at Arg323-Ile324 followed by cleavage at Art274-Thr275 at all concentrations of PCPS. Increasing the PCPS concentration resulted in a large increase in the Arg323-Ile324 cleavage reaction with a much smaller effect on the subsequent cleavage at Arg274-Thr275, thereby leading to an increase in the extent of accumulation of the intermediate, meizothrombin. Fluorescence stopped-flow and rapid chemical quench measurements were also conducted using prethrombin 2 plus fragment 1.2 or meizothrombin as substrates to assess the influence of PCPS on the individual cleavage reactions. The rate of cleavage at Arg323-Ile324 by prothrombinase was increased approximately 60-fold with increasing PCPS, whereas the cleavage at Arg274-Thr275 was increased by a factor of approximately 5. These differential effects of PCPS on the two cleavage reactions adequately explain changes in the extent of accumulation of meizothrombin during prothrombin activation. Proteolytic removal of the membrane binding fragment 1 domain of the substrates, meizothrombin and prethrombin 2-fragment 1.2, had no effect on the cleavage at Arg274-Thr275 at saturating PCPS but completely eliminated the membrane-dependent rate enhancement for cleavage at Arg323-Ile324. Thus, membrane binding by the substrate is essential for the first cleavage reaction at Arg323-Ile324, which leads to the conversion of prothrombin to meizothrombin. In contrast, the substrate-membrane interaction mediated by the fragment 1 domain has no detectable effect on the second cleavage reaction at Arg274-Thr275 which is required for the conversion of meizothrombin to thrombin.     

Protein trans-splicing and functional mini-inteins of a cyanobacterial dnaB intein

A 429 aa theoretical intein is encoded in the dnaB gene (DNA helicase) of the cyanobacterium Synechocystis sp. strain PCC6803. This intein is shown to be capable of protein splicing with or without its native exteins when tested in E. coli cells. A centrally located 275 amino acid sequence (residues 107-381) of this intein can be deleted without loss of the protein splicing activity, resulting in a functional mini-intein of 154 aa in size. Efficient in vivo protein trans-splicing was observed when this mini-intein was split into a 106 aa N-terminal fragment containing intein motifs A and B, and a 48 aa C-terminal fragment containing intein motifs F and G. These results indicate that the N- and C-terminal regions of the Ssp DnaB intein, whether covalently linked with each other or not, can come together through non-covalent interaction to form a protein splicing domain that is functionally sufficient and structurally independent from the centrally located endonuclease domain of the intein.