Portal de Periódicos da CAPES

The ADP-ribosylating Mosquitocidal Toxin from Bacillus sphaericus

2002; Elsevier BV; Volume: 277; Issue: 14 Linguagem: Inglês

10.1074/jbc.m108463200

1083-351X

Jörg Schirmer, Ingo Just, Klaus Aktories,

Plant Virus Research Studies

The mosquitocidal toxin (MTX) from Bacillus sphaericus SSII-1 is a ∼97-kDa protein sharing sequence homology within the N terminus with the catalytic domains of various bacterial ADP-ribosyltransferases. Here we studied the proteolytic activation of the ADP-ribosyltransferase activity of MTX. Chymotrypsin treatment of the 97-kDa MTX holotoxin (MTX30–870) results in a 70-kDa putative binding component (MTX265–870) and a 27-kDa enzyme component (MTX30–264), possessing ADP-ribosyltransferase activity. Chymotryptic cleavage of an N-terminal 32-kDa fragment of MTX (MTX30–308) also yields MTX30–264, but the resulting ADP-ribosyltransferase activity is much greater than that of the processed MTX30–870. Kinetic studies revealed a Km NAD value of 45 μm for the processed 32-kDa MTX fragment, and a Km NAD value of 1300 μm for the processed holotoxin. Moreover, the kcat value for the activated MTX30–308 fragment was about 10-fold higher than that for the activated holotoxin (MTX30–870). Precipitation analysis showed that the 70-kDa proteolytic fragment of MTX remains noncovalently bound to the N-terminal 27-kDa fragment, thereby inhibiting ADP-ribosyltransferase and NAD glycohydrolase activities. Glu197 of MTX30–264 was identified as the "catalytic" glutamate that is conserved in all ADP-ribosyltransferases. Whereas mutated MTX30–264E197Q has neither ADP-ribosyltransferase nor NAD glycohydrolase activity, mutated MTX30–264E195Q possesses glycohydrolase activity but not transferase activity. Transfection of HeLa cells with a vector encoding a fusion protein of MTX30–264 with a green fluorescent protein led to cytotoxic effects characterized by cell rounding and formation of filopodia-like protrusions. These cytotoxic effects were not observed with the catalytically inactive MTX30–264E197Q mutant, indicating that the MTX enzyme activity is essential for the cytotoxicity in mammalian cells. The mosquitocidal toxin (MTX) from Bacillus sphaericus SSII-1 is a ∼97-kDa protein sharing sequence homology within the N terminus with the catalytic domains of various bacterial ADP-ribosyltransferases. Here we studied the proteolytic activation of the ADP-ribosyltransferase activity of MTX. Chymotrypsin treatment of the 97-kDa MTX holotoxin (MTX30–870) results in a 70-kDa putative binding component (MTX265–870) and a 27-kDa enzyme component (MTX30–264), possessing ADP-ribosyltransferase activity. Chymotryptic cleavage of an N-terminal 32-kDa fragment of MTX (MTX30–308) also yields MTX30–264, but the resulting ADP-ribosyltransferase activity is much greater than that of the processed MTX30–870. Kinetic studies revealed a Km NAD value of 45 μm for the processed 32-kDa MTX fragment, and a Km NAD value of 1300 μm for the processed holotoxin. Moreover, the kcat value for the activated MTX30–308 fragment was about 10-fold higher than that for the activated holotoxin (MTX30–870). Precipitation analysis showed that the 70-kDa proteolytic fragment of MTX remains noncovalently bound to the N-terminal 27-kDa fragment, thereby inhibiting ADP-ribosyltransferase and NAD glycohydrolase activities. Glu197 of MTX30–264 was identified as the "catalytic" glutamate that is conserved in all ADP-ribosyltransferases. Whereas mutated MTX30–264E197Q has neither ADP-ribosyltransferase nor NAD glycohydrolase activity, mutated MTX30–264E195Q possesses glycohydrolase activity but not transferase activity. Transfection of HeLa cells with a vector encoding a fusion protein of MTX30–264 with a green fluorescent protein led to cytotoxic effects characterized by cell rounding and formation of filopodia-like protrusions. These cytotoxic effects were not observed with the catalytically inactive MTX30–264E197Q mutant, indicating that the MTX enzyme activity is essential for the cytotoxicity in mammalian cells. ADP-ribosylation of eukaryotic target proteins is a major biochemical mechanism that is exploited by bacterial toxins to affect the eukaryotic host cells (1.Domenighini M. Pizza M. Rappuoli R. Moss J. Iglewski B. Vaughan M. Tu A.T. Bacterial Toxins and Virulence Factors in Disease. Marcel Dekker, Inc., New York, Basel, Hong Kong1995: 59-74Google Scholar). Among these toxins are diphtheria toxin and Pseudomonas exotoxin A, which ADP-ribosylate the eukaryotic elongation factor 2 at diphthamide, a modified histidine residue, to block protein synthesis (2.Middlebrook J.L. Dorland R.B. Microbiol. Rev. 1984; 48: 199-221Crossref PubMed Google Scholar). Cholera toxin and pertussis toxin are known to ADP-ribosylate G-proteins at specific arginine and cysteine residues, respectively, to modify signal transduction by G-protein-coupled receptors (3.Van den Akker F. Merritt E.A. Hol W.G.J. Aktories K. Just I. Bacterial Protein Toxins. Springer, Berlin2000: 109-125Google Scholar, 4.Locht C. Antoine R. Aktories K. Bacterial Toxins: Tools in Cell Biology and Pharmacology. Chapman & Hall, Weinheim, Germany1997: 33-41Crossref Scopus (4) Google Scholar). Members of the subfamily of binary ADP-ribosylating toxins such as Clostridium botulinum C2 toxin and Clostridium perfringens iota toxin specifically modify G-actin at Arg177 (5.Aktories K. Wegner A. Mol. Microbiol. 1992; 6: 2905-2908Crossref PubMed Scopus (79) Google Scholar). Moreover, several bacterial exoenzymes ADP-ribosylate small GTP-binding proteins of targets cells. Examples are C. botulinum C3 exoenzyme and related C3-like transferases, which ADP-ribosylate Rho GTPases at Asn41(6.Rubin E.J. Gill D.M. Boquet P. Popoff M.R. Mol. Cell. Biol. 1988; 8: 418-426Crossref PubMed Scopus (233) Google Scholar, 7.Braun U. Habermann B. Just I. Aktories K. Vandekerckhove J. FEBS Lett. 1989; 243: 70-76Crossref PubMed Scopus (96) Google Scholar, 8.Sekine A. Fujiwara M. Narumiya S. J. Biol. Chem. 1989; 264: 8602-8605Abstract Full Text PDF PubMed Google Scholar), and Pseudomonas aeruginosa exoenzyme S, which modifies Ras proteins at several arginine residues (9.Ganesan A.K. Frank D.W. Misra R.P. Schmidt G. Barbieri J.T. J. Biol. Chem. 1998; 273: 7332-7337Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). Another member of the family of ADP-ribosylating toxins is the mosquitocidal toxin (MTX), 1The abbreviations used are: MTXmosquitocidal toxinGSTglutathione S-transferaseSBTIsoybean trypsin inhibitorEGFPenhanced green fluorescent proteinMALDI-TOFmatrix-assisted laser desorption ionization time-of-flight 1The abbreviations used are: MTXmosquitocidal toxinGSTglutathione S-transferaseSBTIsoybean trypsin inhibitorEGFPenhanced green fluorescent proteinMALDI-TOFmatrix-assisted laser desorption ionization time-of-flightwhich is produced by the low-toxicity strain SSII-1 of Bacillus sphaericus. The toxin is lethal to Culex quinquefasciatus and Aedes aegypti mosquito larvae (10.Thanabalu T. Hindley J. Jackson-Yap J. Berry C. J. Bacteriol. 1991; 173: 2776-2785Crossref PubMed Google Scholar). The mature MTX (without the putative signal sequence of 29 amino acid residues) is a 97-kDa protein (MTX30–870). MTX30–870 is processed into a 27-kDa N-terminal fragment and a 70-kDa C-terminal fragment by crude mosquito larval gut extracts and trypsin (11.Thanabalu T. Hindley J. Berry C. J. Bacteriol. 1992; 174: 5051-5056Crossref PubMed Google Scholar). The cleavage site is reportedly amino acid 264 (phenylalanine) for the mosquito larval gut extracts and amino acid 262 (lysine) for trypsin as determined by N-terminal sequencing. This 97-kDa fragment, previously designated MTX21, was renamed MTX30–870 in accordance with the nomenclature of our MTX truncations. MTX30–870 retains its lethal effects upon C. quinquefasciatus larvae regardless of whether it is unprocessed or proteolytically cleaved. N-terminal or C-terminal truncations of the toxin alone, however, lacked any toxicity toward mosquito larvae (12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar). mosquitocidal toxin glutathione S-transferase soybean trypsin inhibitor enhanced green fluorescent protein matrix-assisted laser desorption ionization time-of-flight mosquitocidal toxin glutathione S-transferase soybean trypsin inhibitor enhanced green fluorescent protein matrix-assisted laser desorption ionization time-of-flight Sequence alignments revealed homologies in the N terminus of MTX with the catalytic domains of various bacterial ADP-ribosyltransferases, such as pertussis toxin, cholera toxin, or the Escherichia coli heat-labile enterotoxins (LT1 and LT2) (10.Thanabalu T. Hindley J. Jackson-Yap J. Berry C. J. Bacteriol. 1991; 173: 2776-2785Crossref PubMed Google Scholar). Recent crystal structure analysis of bacterial ADP-ribosylating toxins, including diphtheria toxin, E. coli heat-labile toxin, pertussis toxin, and the vegetative insecticidal proteins from Bacillus cereus, revealed a highly conserved structure of the NAD binding and catalytic domains of the toxins (13.Masignani V. Pizza M. Rappuoli R. Aktories K. Just I. Bacterial Protein Toxins. Springer, Berlin2000Google Scholar). A common feature of the catalytic domain is a "catalytic" glutamic acid residue essential for transferase activity (14.Carroll S.F. Collier R.J. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 3307-3311Crossref PubMed Scopus (109) Google Scholar, 15.Jung M. Just I. van Damme J. Vandekerckhove J. Aktories K. J. Biol. Chem. 1993; 268: 23215-23218Abstract Full Text PDF PubMed Google Scholar, 16.Domenighini M. Magagnoli C. Pizza M. Rappuoli R. Mol. Microbiol. 1994; 14: 41-50Crossref PubMed Scopus (97) Google Scholar) and often an additional glutamic acid residue located 2 residues upstream of the first one. This second glutamic acid residue is also essential for ADP-ribosylating activity (16.Domenighini M. Magagnoli C. Pizza M. Rappuoli R. Mol. Microbiol. 1994; 14: 41-50Crossref PubMed Scopus (97) Google Scholar, 17.Domenighini M. Rappuoli R. Mol. Microbiol. 1996; 21: 667-674Crossref PubMed Scopus (132) Google Scholar, 18.Takada T. Iida K. Moss J. J. Biol. Chem. 1995; 270: 541-544Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar) but has been reported not to be involved in NAD glycohydrolase activity (19.Barth H. Preiss J.C. Hofmann F. Aktories K. J. Biol. Chem. 1998; 273: 29506-29511Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 20.Radke J. Pederson K.J. Barbieri J.T. Infect. Immun. 1999; 67: 1508-1510Crossref PubMed Google Scholar). The putative catalytic domain of MTX possesses two glutamic acid residues at positions 195 and 197, as well as an arginine residue at position 97 and a serine residue and a threonine residue at positions 142 and 143, respectively, which appear to represent the consensus residues commonly found among bacterial ADP-ribosyltransferases (see Fig. 1). Therefore, MTX is considered to be a member of the family of ADP-ribosylating toxins (1.Domenighini M. Pizza M. Rappuoli R. Moss J. Iglewski B. Vaughan M. Tu A.T. Bacterial Toxins and Virulence Factors in Disease. Marcel Dekker, Inc., New York, Basel, Hong Kong1995: 59-74Google Scholar, 12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar, 16.Domenighini M. Magagnoli C. Pizza M. Rappuoli R. Mol. Microbiol. 1994; 14: 41-50Crossref PubMed Scopus (97) Google Scholar). However, a precise characterization of the MTX enzyme activity and its regulation by proteolytic activation is still pending. Here we report on the necessity of proteolytic activation of the enzyme, the regulation of enzyme activity by its putative binding component, and the identification of the catalytic glutamic acid residue by site-directed mutagenesis. In addition, we show for the first time a cytotoxic effect of the MTX enzyme component in mammalian cell culture. As shown by studies with catalytically inactive mutants, the ADP-ribosylation activity appears to be responsible for the observed biological effects. [Adenylate-32P]NAD (30 Ci/mmol) was purchased from PerkinElmer Life Sciences (Vilvoorde, Belgium). Chymotrypsin and soybean trypsin inhibitor were from Roche Diagnostics. All other reagents were from Sigma unless otherwise indicated. For cloning of MTX30–308 consisting of amino acid residues 30–308, DNA was amplified from plasmid pTH21 (encoding MTX30–870) by PCR with the following primers: MTX30–308 sense, 5′-AGATCTGCTTCACCTAATTCTCCAAAAG-3′; and MTX30–308antisense, 5′-GTCGACCTTTTATTTTTGATTTGATATTCTG-3′. For cloning of MTX265–870 consisting of amino acids 265–870, DNA was also amplified from plasmid pTH21 by polymerase chain reaction with the following primers: MTX265–870 sense, 5′-GGATCCATACTAGATTTAGATTATAATCAAG-3′; and MTX265–870antisense, 5′-GTCGACTCTAGGTTCTACACCTAATG-3′. After cloning into the pCR™II vector (Invitrogen), the MTX fragments were cut with BglII (MTX30–308) or BamHI (MTX265–870), respectively, and SalI, purified, and ligated into the digested pGEX4-TGL vector. This vector was previously designed in our laboratory and is a modification of the pGEX-4T vector from Amersham Biosciences, Inc. The vector contains an additional oligonucleotide in the multiple cloning site that codes for a glycine linker between the GST residue and the inserted gene, enabling a better removal of the GST protein by thrombin cleavage. The proper constructs were checked by DNA sequencing (ABI PRISM; PerkinElmer Life Sciences). Mutagenesis of MTX30–308 was performed by round circle PCR-based site-directed mutagenesis (QuikChange™; Stratagene) using the following sense primers and corresponding antisense primers: MTX_E195Q sense, 5′-CCCTTTCCTAACCAGGATGAAATAAC-3′; and MTX_E197Q sense, 5′-CCTAACGAGGATCAAATAACTTTTC-3′. All primers were from MWG (Ebersberg, Germany), and mutations were verified by DNA sequencing. For expression and purification of the GST fusion proteins, vectors were transformed into E. coli BL21 strains. Cells were grown in LB medium and induced with 0.2 mmisopropyl-1-thio-β-d-galactopyranoside at an OD of 0.6. After overnight incubation at 29 °C, cells were harvested, lysed by sonication in lysis buffer (20 mm Tris, pH 7.4, 10 mm NaCl, 5 mm MgCl2, 1% Triton X-100, 1 mm phenylmethylsulfonyl fluoride, and 5 mm dithiothreitol), and purified by affinity chromatography with glutathione-Sepharose beads (Amersham Biosciences, Inc.). Loaded beads were washed twice with lysis buffer and twice with thrombin cleavage buffer (50 mm Tris, pH 7.4, 150 mmNaCl, and 5 mm MgCl2). Induction of MTX30–870 protein synthesis was carried out as described by Thanabalu et al. (11.Thanabalu T. Hindley J. Berry C. J. Bacteriol. 1992; 174: 5051-5056Crossref PubMed Google Scholar). Briefly, the E. coli cells were grown to stationary phase at 37 °C, harvested, resuspended in fresh medium containing 1 mmisopropyl-1-thio-β-d-galactopyranoside, and incubated for 1 h at 30 °C. MTX constructs were cleaved with thrombin directly from the beads in thrombin cleavage buffer. Thrombin was removed with benzamidine-Sepharose beads (Amersham Biosciences, Inc.). 100 μg of MTX construct were incubated with 0.5 μg of chymotrypsin in a total volume of 250 μl for 60 min at room temperature. Chymotrypsin was inactivated with 2 μg of aprotinin. To check whether full cleavage was achieved, the proteins were subjected to SDS-PAGE. MTX30–264 and bovine serum albumin (serving as internal standard) were purified by a chloroform/methanol precipitation and dissolved in 0.1% trifluoroacetic acid to a final concentration of 20 pmol/μl each. A saturated matrix solution of 4-hydroxy-α-cyanocinnamic acid in a 1:1 solution of acetonitrile/aqueous 0.1% trifluoroacetic acid was prepared. The MTX30–264/bovine serum albumin solution was mixed with the matrix solution in equal parts, and using the dried droplet method of matrix crystallization, 1 μl of the sample/matrix mixture was placed on the mass spectrometer target and dried at room temperature, resulting in a fine granular matrix layer. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry was performed on a Bruker Biflex mass spectrometer equipped with a nitrogen laser (λ = 337 nm) to desorb and ionize the samples. Mass spectra were recorded in the linear positive mode. ADP-ribosylation was performed as follows: 10 μm soybean trypsin inhibitor or 12 μg of total HeLa cell lysate proteins were incubated with 100 μm [32P]NAD and 100 nm MTX fragment for 30 min at room temperature in the presence of 1 mm dithiothreitol, 2 mmMgCl2, and 20 mm Tris, pH 7.4, in a total volume of 20 μl. The reaction was stopped by the addition of Laemmli buffer and heating for 5 min at 95 °C, and the samples were subsequently subjected to SDS-PAGE according to the methods of Laemmli (21.Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (205531) Google Scholar). [32P]ADP-ribosylated proteins were detected with a PhosphorImager from Molecular Dynamics. To study the inhibition of MTX ADP-ribosyltransferase activity by its putative binding component, MTX30–264 (100 nm) was preincubated with varying concentrations of MTX265–870(25–200 nm) for 5 min at 4 °C. Thereafter, ADP-ribosyltransferase assays were performed as described above. For measurement of glycohydrolase activity, the MTX fragments (1 μm each) were incubated with 100 μm [32P]NAD in 10 μl of a buffer containing 2 mm MgCl2 and 50 mmTris, pH 7.4, for 1 h. 2 μl of each reaction mixture were separated by TLC on TLC aluminum sheets (Silica Gel 60 F254; Merck) with 66% 2-propanol/0.33% ammonium sulfate and analyzed by phosphorimaging. Initial rate data for the ADP-ribosyltransferase reaction was determined with regard to NAD binding by varying the NAD concentration from 30 to 500 μm for activated MTX30–308 and from 250 to 5000 μm for activated MTX30–870. Experiments were performed at fixed SBTI concentrations of 10 μm. The amount of SBTI utilized was <10%. In all experiments, toxin concentrations were 100 nm, and incubation time was 5 min for MTX30–308 and 30 min for MTX30–870 at room temperature. Kinetic values were obtained by quantifying the PhosphorImager data with the help of the ImageQuant software (Amersham Biosciences, Inc.) and transforming the data to the Lineweaver-Burk plot. GST and GST-MTX265–870were bound to glutathione-Sepharose beads and incubated with HeLa cytosol (0.2 μg/μl) for 30 min at 4 °C to block nonspecific binding sites. Beads were washed twice with buffer (10% glycerol, 50 mm Tris, pH 7.4, 100 mm NaCl, 1% Nonidet P-40, and 2 mm MgCl2) and once with thrombin cleavage buffer. The beads were then incubated with MTX30–264 for 30 min at room temperature. Beads were washed as described before and subjected to SDS-PAGE. Next, the proteins were either detected with Coomassie Blue or transferred to a nitrocellulose membrane for subsequent immunoblotting. For immunoblot analysis, the membrane was blocked for 60 min with 5% nonfat dry milk in phosphate-buffered saline containing 0.05% Tween 20 (PBS-T) followed by a 1-h incubation with an anti-MTX30–264 antibody (rabbit, 1:2000 in PBS-T). After washing with PBS-T, the blot was incubated for 1 h with goat anti-rabbit antibody coupled to horseradish peroxidase (1:5000 in PBS-T) and washed. Proteins were detected with the Amersham Biosciences, Inc. enhanced chemiluminescence system as instructed by the manufacturer. The eukaryotic expression vector pEGFP-C1 (CLONTECH), encoding EGFP under the cytomegalovirus promoter, was used for easy identification of transfected cells. MTX30–308 and MTX30–264were amplified from the pTH21 vector using the following primers adding BglII and SalI sites: MTX30–308/MTX30–264 sense, 5′-AGATCTGCTTCACCTAATTCTCCAAAAG-3′; MTX30–308 antisense, 5′-GTCGACCTTCTTATTTTTGATTTGATATTCTG-3′; and MTX30–264antisense, 5′-GTCGACAAAACCTTTAGAATCCATATTATTTC-3′. The amplified DNA fragments were digested and inserted into the digested pEGP-C1 vector. The E195Q mutation and the E197Q mutation, respectively, were introduced using the QuikChange™ kit (Stratagene) and the primers described for the pGEX-MTX30–308 mutants. Plasmid DNAs were propagated in E. coli and purified for the following transfection studies. For transfection studies, HeLa cells were cultured for 24 h in Dulbeccos's modified Eagle's medium supplemented with 10% fetal calf serum (PAN Systems, Aidenbach, Germany) and 4 mm glutamine/penicillin/streptomycin in 30-mm dishes at 37 °C and 5% CO2 for 24 h. HeLa cells were then transfected with a 5:1 ratio of the respective plasmids encoding the MTX constructs and the pEGFP-C1 vector or the pEGFP-C1 vector alone using the Polyfect transfection kit (Qiagen, Hilden, Germany) according to the manufacturer's manual. For actin cytoskeleton staining, cells were fixed and permeabilized with 4% formaldehyde plus 1% Triton X-100 in phosphate-buffered saline for 10 min at room temperature and then incubated with 1 μg/ml tetramethylrhodamine isothiocyanate-phalloidin for 60 min in the dark at room temperature. Cells were examined by fluorescence microscopy after 24 h of incubation, and pictures were obtained with a Zeiss microscope supplied with a digital camera. To study ADP-ribosyltransferase activity of the MTX and its possible regulation, we constructed the vector pGEX-MTX265–870, encoding the 70-kDa C-terminal part of the toxin (amino acids 265–870), and tried to construct the vector pGEX-MTX30–264, encoding the 27-kDa N-terminal part of the toxin (amino acids 30–264). However, no correct pGEX-MTX30–264 clone was obtained, despite several cloning strategies. Using the pET vector system (Novagen, Bad Soden, Germany), we obtained correct pET-MTX30–264 clones as long as nonexpression host cells were used. These cells lack the DE3-lysogen encoding the T7-polymerase necessary for transcription from the pET vector. Due to possible cytotoxicity of the MTX30–264 fragment toward E. coli, we constructed the vector pGEX-MTX30–308, encoding a 32-kDa N-terminal part of the toxin (amino acids 30–308). Fig. 2A gives an overview of the MTX constructs used in this study. We proteolytically cleaved MTX30–308 to generate the 27-kDa fragment. Cleavage of MTX30–870 resulted in the 27- and 70-kDa fragments of MTX. Because trypsin treatment resulted in incomplete cleavage, chymotrypsin was used. The expressed MTX proteins were analyzed by SDS-PAGE before and after chymotrypsin treatment (Fig. 2B). Note that in contrast to the MTX30–276 fragment reported by Thanabalu et al. (12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar), which runs like a 70-kDa protein, the MTX30–308 truncation runs as predicted from the amino acid sequence as a 32-kDa protein. The difference in migration on SDS-PAGE of the 27-kDa fragments after cleavage of MTX30–870 and MTX30–308 is due to the use of different expression vectors for the constructs. MTX30–308 carries an additional 12 amino acids at its N terminus (remnants of the glycine linker), whereas MTX30–870 carries only 3 additional amino acids of vector origin at its N terminus. These amino acids are not removed by chymotrypsin treatment, and therefore the 27-kDa cleavage products show a mass difference of about 600 Da (Fig. 2B). However, exact protein molecular mass determination by MALDI-TOF mass spectrometry data confirmed that chymotryptic cleavage of MTX30–308 results in the corresponding MTX fragment as reported for cleavage of MTX30–870, with phenylalanine 264 as the cleavage site (11.Thanabalu T. Hindley J. Berry C. J. Bacteriol. 1992; 174: 5051-5056Crossref PubMed Google Scholar) (Fig. 2C). Initial experiments showed that cleaved MTX ADP-ribosylates a large array of different proteins in lysates of insect cells (data not shown) and mammalian cultured cell lines (see also Fig. 9). To test the ADP-ribosyltransferase activity of the MTX constructs in more detail, SBTI was chosen as an in vitro model substrate. MTX and its fragments were incubated with SBTI and [32P]NAD for 30 min at room temperature. After SDS-PAGE, labeling was detected by a PhosphorImager. Only proteolytically cleaved MTX30–870 and proteolytically cleaved MTX30–308 led to labeling of SBTI. Interestingly, cleaved MTX30–308 catalyzed labeling to a much larger extent than processed MTX30–870 (Fig. 3A). Because several ADP-ribosyltransferases possess NAD glycohydrolase activity, i.e. hydrolysis of NAD in the absence of a protein substrate, we tested whether this was also true for MTX. Consistent with the ADP-ribosylation results, only cleaved MTX constructs were capable of hydrolyzing NAD. Again, the proteolytically cleaved MTX30–308 was much more active than cleaved MTX30–870 (Fig. 3B).Figure 3Activity of proteolytically cleaved MTX fragments. A, ADP-ribosylation of SBTI by different MTX constructs (100 nm each). SBTI (10 μm) was incubated with the indicated MTX constructs in the presence of [32P]NAD for 30 min at room temperature. Thereafter, labeled proteins were analyzed by SDS-PAGE and phosphorimaging (shown). B, NAD glycohydrolase activity of MTX. MTX constructs (1 μm each) were incubated in the presence of 100 μm [32P]NAD for 1 h at room temperature. The formation of ADP-ribose was analyzed by TLC and phosphorimaging (shown).View Large Image Figure ViewerDownload Hi-res image Download (PPT) For further characterization of the enzyme activity, we determined the Michaelis constant (Km) for NAD of the MTX transferase reaction. Constant amounts of SBTI were used in an ADP-ribosylation assay with either activated MTX30–308 or activated MTX30–870 at varying concentrations of NAD. The Km NAD value for activated MTX30–308 was found to be 45 μm, and kcat was calculated to be 2.5 min−1, resulting in a kcat:Km ratio of 0.06 min−1m−1. For MTX30–870, a Km NAD value of 1300 μm and a kcat value of 0.5 min−1 were obtained, resulting in a kcat:Km ratio of 0.0004 min−1m−1 (Table I).Table IKinetics of ADP-ribosyltransferase activityKm(NAD)kcatkcat:Kmμmmin−1min−1Activated MTX30–30845 ±72.5 ± 10.06Activated MTX30–8701300 ± 1960.5 ± 0.30.0004Enzyme kinetics of activated MTX30–308 and activated MTX30–870. ADP-ribosylation of SBTI was performed with activated MTX30–308 and activated MTX30–870 in the presence of varying concentrations of [32P]NAD as described. Labeled proteins were analyzed by SDS-PAGE and phosphorimaging. The data were quantified, and kinetic values were obtained from Lineweaver-Burk plot transformation of data. Data are given as means ± S.E. (n = 3). Open table in a new tab Enzyme kinetics of activated MTX30–308 and activated MTX30–870. ADP-ribosylation of SBTI was performed with activated MTX30–308 and activated MTX30–870 in the presence of varying concentrations of [32P]NAD as described. Labeled proteins were analyzed by SDS-PAGE and phosphorimaging. The data were quantified, and kinetic values were obtained from Lineweaver-Burk plot transformation of data. Data are given as means ± S.E. (n = 3). After chymotryptic cleavage of MTX30–870, the 27- and 70-kDa fragments could only be separated by denaturing methods. Therefore, we attempted to co-precipitate MTX30–264 (obtained by cleavage of MTX30–308) with GST-MTX265–870 bound to glutathione-Sepharose beads, and vice versa. After incubation of the loaded beads with the toxin fragments and washing, the beads were subjected to SDS-PAGE. As shown in Fig. 4, MTX30–264 was co-precipitated by GST-MTX265–870, whereas no MTX30–264 band was detectable with GST alone. The same binding and co-precipitation was found when MTX265–870 was incubated with GST-MTX30–264 immobilized to glutathione-Sepharose beads (data not shown). As described above, MTX265–870remained bound to MTX30–264 after proteolytic cleavage. Therefore, we wanted to investigate whether this interaction had any effect on the enzyme activity, possibly explaining why cleaved MTX30–870 showed markedly less enzyme activity than cleaved MTX30–308 (Fig. 3). Therefore, MTX30–264 (obtained by cleavage of MTX30–308) was preincubated with varying concentrations of MTX265–870, and, then the ADP-ribosylation was started. As shown in Fig. 5A, in the presence of equimolar concentrations of MTX30–264 and MTX265–870, ADP-ribosylation of SBTI was drastically reduced. A 1:1 ratio of both MTX fragments exhibited an activity similar to that of the proteolytically activated MTX30–870. The inhibition of the enzyme activity by MTX265–870 was concentration-dependent, and a 1-fold surplus of MTX265–870 almost completely inhibited the enzyme activity (Fig. 5B). Moreover, MTX265–870 also inhibited NAD glycohydrolase activity of MTX30–264 (data not shown). By sequence alignments with other ADP-ribosyltransferases, amino acids 195 and 197 (both glutamic acid residues) of MTX are proposed to be important for catalysis (1.Domenighini M. Pizza M. Rappuoli R. Moss J. Iglewski B. Vaughan M. Tu A.T. Bacterial Toxins and Virulence Factors in Disease. Marcel Dekker, Inc., New York, Basel, Hong Kong1995: 59-74Google Scholar, 16.Domenighini M. Magagnoli C. Pizza M. Rappuoli R. Mol. Microbiol. 1994; 14: 41-50Crossref PubMed Scopus (97) Google Scholar) (see Fig. 1). The respective MTX30–308E195Q and MTX30–308E197Q mutant proteins were expressed in E. coli and purified as described. The mutants were treated with chymotrypsin, and degradation was followed by SDS-PAGE (Fig. 6A). We did not observe any differences in the sensitivity of wild-type and mutant MTX proteins toward proteolytic cleavage, suggesting no major folding changes of the mutant proteins. Mutant MTX proteins were tested for ADP-ribosylation activity, but none was active, indicating that both glutamate residues, Glu195 and Glu197, were essential for transferase activity (Fig. 6B). It is reported that only the C-terminal glutamate of the glutamate-aspartate-glutamate (EXE) motif in the catalytic domain is essential for NAD glycohydrolase activity in "biglutamic" ADP-ribosyltransferases (19.Barth H. Preiss J.C. Hofmann F. Aktories K. J. Biol. Chem. 1998; 273: 29506-29511Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar, 20.Radke J. Pederson K.J. Barbieri J.T. Infect. Immun. 1999; 67: 1508-1510Crossref PubMed Google Scholar). To study whether this is also true for MTX, NAD glycohydrolase assays with the MTX30–308 mutants (cleaved/uncleaved) were performed. The E197Q mutant was not capable of hydrolyzing NAD regardless of proteolytic treatment. By contrast, the E195Q mutant hydrolyzed NAD, but only after proteolytic cleavage. The NAD glycohydrolase activity was similar to that of activated MTX30–308 (Fig. 6C). Next, we tested the MTX toxin for cytotoxic effects on mammalian cells. To this end, a plasmid encoding for a fusion protein of EGFP with MTX30–264 was constructed. Transfection of HeLa cells with this plasmid resulted in morphological changes in cells expressing the protein as observed by fluorescence microscopy (Fig. 7). The vector encoding the EGFP-MTX30–264 protein was co-transfected with the pEGFP-C1 vector in a 5:1 ratio to enhance fluorescence because fluorescence was very weak when the pEGFP-MTX30–264 construct was transfected alone. To obtain comparable experimental conditions, this co-transfection was done routinely with all other pEGFP-MTX plasmids used in this study. Cells expressing the 27-kDa MTX catalytically active fragment exhibited formation of filopodia-like protrusions and rounding up. Staining of the actin cytoskeleton with tetramethylrhodamine isothiocyanate-phalloidin proved that the protrusions contain actin (Fig. 8). In contrast, HeLa cells expressing the EGFP-MTX30–264E197Q fusion protein did not show morphological changes as seen in Fig. 7. Cells expressing the EGFP-MTX30–264E195Q mutant protein exhibited some changes in cell morphology, however, this effect was much less pronounced than that seen in cells expressing the wild-type enzyme component (Fig. 7). Additionally, a plasmid encoding a fusion protein of EGFP with MTX30–308 was cloned to check whether this truncation is inactive in vivo, as it is in vitro, giving further proof of the necessity of enzyme activation. Cells transfected with the vector encoding MTX30–308 and cells transfected with the vector alone exhibited no significant morphological changes (Fig. 7). Fig. 9 shows that MTX30–264 (derived from MTX30–308) ADP-ribosylated several proteins in lysates of HeLa cells. By contrast, MTX30–308, which was without biological activity after transfection, did not ADP-ribosylate any protein in the cell lysates, and the same was true for the processed MTX mutant proteins.Figure 8Actin cytoskeleton staining of pEGFP-MTX-transfected HeLa cells. Actin cytoskeleton staining with tetramethylrhodamine isothiocyanate-phalloidin of cells transfected with the pEGFP vector (control), pEGFP-MTX30–264, or pEGFP-MTX30–264E197Q. Single cells are shown at ×100 magnification.View Large Image Figure ViewerDownload Hi-res image Download (PPT) MTX30–870 is the native form of the ADP-ribosylating toxin from B. sphaericus SSII-1, which lacks the putative signal sequence of 29 amino acids. The toxin is reportedly proteolytically cleaved into a 27-kDa N-terminal fragment and a 70-kDa C-terminal fragment (11.Thanabalu T. Hindley J. Berry C. J. Bacteriol. 1992; 174: 5051-5056Crossref PubMed Google Scholar, 12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar). Preliminary studies suggested that the 27-kDa fragment harbors the ADP-ribosyltransferase activity of MTX. In the present study, we characterized the biochemical and biological activities of the 27-kDa fragment in more detail. Our approach for constructing an expression vector encoding the MTX30–264gene was only successful as long as the plasmid was transformed into nonexpression host cells. Transformation into host cells, which are capable of vector expression, led to either defective clones or no transformation at all. By contrast, the gene of an enzymatically defective MTX30–264 (MTX30–264E195/197Q) was easily cloned into a pGEX or pET vector and transformed into expression host cells. Therefore, we suggest that MTX30–264 is toxic to E. coli due to its enzyme activity. Notably, a putative ADP-ribosyltransferase (pierisin-1) from a cabbage butterfly that is homologous to MTX (32% identity at the amino acid level) was reported to be toxic to E. coli (22.Watanabe M. Kono T. Matsushima-Hibiya Y. Kanazawa T. Nishisaka N. Kishimoto T. Koyama K. Sugimura T. Wakabayashi K. Proc. Natl. Acad. Sci U. S. A. 1999; 96: 10608-10613Crossref PubMed Scopus (68) Google Scholar). Whereas the complete molecule of pierisin-1 exhibits toxicity to E. coli, MTX30–870 is nontoxic to the expressing bacteria. Therefore, MTX30–870 and a 32-kDa N-terminal MTX truncation (MTX30–308), which were well expressed in E. coli, were used to generate the 27-kDa fragment by chymotryptic cleavage and to study its enzyme activity. Study of the in vitro activity of our MTX constructs revealed potent activation by proteolytic cleavage of the toxin and ADP-ribosylation of several proteins in mammalian cell lysates. To analyze the enzyme activity in more detail, we used SBTI, which is a well-known artificial substrate for ADP-ribosylation by bacterial transferases (23.Coburn J. Wyatt R.T. Iglewski B.H. Gill D.M. J. Biol. Chem. 1989; 264: 9004-9008Abstract Full Text PDF PubMed Google Scholar), as a model substrate. By comparing the activities of the toxin fragments, we observed that cleaved MTX30–870was markedly less active than cleaved MTX30–308. This finding was surprising because both fragments yielded MTX30–264 as a cleavage product, which was shown in this study to harbor the catalytic activity. To clarify these discrepancies, we compared the enzyme kinetics of both fragments. For MTX30–264 derived from the 32-kDa MTX fragment (MTX30–308), the Km NAD value was 45 μm. This value fits well into the range of the Km values reported for other bacterial ADP-ribosyltransferases such as diphtheria toxin (24.Blanke S.R. Huang K. Wilson B.A. Papini E. Covacci A. Collier R.J. Biochemistry. 1994; 33: 5155-5161Crossref PubMed Scopus (50) Google Scholar) or exoenzyme S (25.Liu S. Kulich S.M. Barbieri J.T. Biochemistry. 1996; 35: 2754-2758Crossref PubMed Scopus (46) Google Scholar). The Km NAD value of the proteolytically activated MTX30–870, however, was about 30-fold higher, and the kcat value was decreased by a factor of about 10. With regard to the kcat:Km value of both constructs, the activated MTX holotoxin was 150-fold less active than the activated MTX30–308 construct. We suggest that the 70-kDa C-terminal fragment that is generated by cleavage of MTX30–870 is responsible for the low activity of activated MTX30–870. Therefore, we analyzed whether MTX30–264 and MTX265–870 do interact. Precipitation assays revealed that both fragments remain tightly bound to each other after cleavage. This interaction between MTX30–264 and MTX265–870 explains the inability to separate these two fragments by methods not involving denaturing techniques. For example, chromatographic or size exclusion methods failed, consistent with the reports of Thanabalu et al. (12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar). Thus, the two toxin fragments stick to each other after proteolytic cleavage due to strong noncovalent interactions. Disulfide bonds can be ruled out because MTX30–870 contains no cysteine residues. To study the consequences of the interaction, we determined the enzyme activity of MTX30–264 at increasing concentrations of MTX265–870. These studies revealed that the interaction of both fragments effectively blocked ADP-ribosylation as well as NAD glycohydrolase activity. With regard to the kinetic data of activated MTX30–308 and activated MTX30–870, the MTX binding component can be regarded as a noncompetitive autoinhibitor of the MTX enzyme component because both the Km value and the kcat value are changed. We further suggest that the 5-kDa C-terminal part of MTX30–308 is sufficient to shield the catalytic domain or to stabilize an inactive conformational structure of MTX30–308. During cleavage of MTX30–308, the 5-kDa fragment is further proteolytically degraded and does not inhibit MTX30–264 enzyme activity. Characterization of the catalytic domain of MTX revealed typical features of a biglutamic acid ADP-ribosyltransferase. Alignments with other bacterial ADP-ribosyltransferases suggested that the EXE motif in positions 195–197 plays an important role in ADP-ribosylation. Exchange of Glu195 or of Glu197 totally blocked the ADP-ribosyltransferase activity. As described for other bacterial ADP-ribosyltransferases such as P. aeruginosa exoenzyme S (20.Radke J. Pederson K.J. Barbieri J.T. Infect. Immun. 1999; 67: 1508-1510Crossref PubMed Google Scholar) or C. botulinum C2 toxin (19.Barth H. Preiss J.C. Hofmann F. Aktories K. J. Biol. Chem. 1998; 273: 29506-29511Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar), both glutamic acid residues in the catalytic domain are important for ADP-ribosyltransferase activity, but only the second glutamate in the EXE motif is necessary for NAD glycohydrolase activity. Whereas proteolytically cleaved MTX30–308E195Q possessed glycohydrolase activity, cleaved MTX30–308E197Q did not. These findings are in line with the notion that Glu197 of MTX is the catalytic glutamate residue, which is highly conserved among the family of ADP-ribosyltransferases, and that Glu195 is the second important glutamic residue in the catalytic domain, consistent with other biglutamic ADP-ribosyltransferases (19.Barth H. Preiss J.C. Hofmann F. Aktories K. J. Biol. Chem. 1998; 273: 29506-29511Abstract Full Text Full Text PDF PubMed Scopus (89) Google Scholar,20.Radke J. Pederson K.J. Barbieri J.T. Infect. Immun. 1999; 67: 1508-1510Crossref PubMed Google Scholar). The construction of enzymatically inactive MTX mutants was helpful in studying the role of ADP-ribosyltransferase activity in the biological action of MTX. Thus far, only a lethal effect of the MTX holotoxin on certain mosquito larvae and a cytotoxic effect on insect cell cultures had previously been reported (10.Thanabalu T. Hindley J. Jackson-Yap J. Berry C. J. Bacteriol. 1991; 173: 2776-2785Crossref PubMed Google Scholar, 11.Thanabalu T. Hindley J. Berry C. J. Bacteriol. 1992; 174: 5051-5056Crossref PubMed Google Scholar, 12.Thanabalu T. Berry C. Hindley J. J. Bacteriol. 1993; 175: 2314-2320Crossref PubMed Google Scholar). However, the cytotoxic effects (aggregation of cultured insect cells) reported were observed even with a C-terminal fragment of MTX lacking the enzyme domain. Here we report for the first time a cytotoxic effect related to the enzyme activity of MTX. Neither the MTX holotoxin nor the activated MTX30–308enzyme fragment alone showed any toxic effect when applied directly onto mammalian cell cultures. However, transfection of HeLa cells with a vector encoding the active 27-kDa enzyme component of MTX led to cytotoxic effects characterized by rounding up of cells and increased formation of filopodia-like structures. This in vivo effect of MTX on mammalian cells was not observed with the catalytically inactive MTX30–264E197Q mutant. MTX30–264E195Q, which possesses NAD glycohydrolase activity but not transferase activity, caused minor changes in HeLa cells. Whether this effect is caused by NAD glycohydrolase activity remains to be studied; however, this effect was much less pronounced than that induced by the wild-type toxin fragment. The results indicate that the MTX enzyme activity is responsible for the cytotoxicity observed, whereas the transferase activity seems to be decisive. Transfection of the vector encoding the N-terminal 32-kDa truncation of MTX (MTX30–308) was not toxic for HeLa cells. These findings support in vitro results, showing that proteolytic cleavage is necessary to activate the toxin. Thus far, the in vivo substrate of MTX and the molecular mechanism of cytotoxicity are unknown. The observed cell rounding and the formation of the actin-containing protrusions might be a hint that one or more proteins important in the regulation of the actin cytoskeleton are affected. This would be in line with numerous other bacterial ADP-ribosylating toxins that act on the actin cytoskeleton, such as C. botulinum C2 toxin that modifies actin (26.Aktories K. Bärmann M. Ohishi I. Tsuyama S. Jakobs K.H. Habermann E. Nature. 1986; 322: 390-392Crossref PubMed Scopus (380) Google Scholar) or C. botulinum C3 exoenzyme with Rho as a substrate (8.Sekine A. Fujiwara M. Narumiya S. J. Biol. Chem. 1989; 264: 8602-8605Abstract Full Text PDF PubMed Google Scholar, 27.Aktories K. Mohr C. Koch G. Curr. Top. Microbiol. Immunol. 1992; 175: 115-131PubMed Google Scholar). In vitro, MTX30–264 ADP-ribosylates numerous proteins in HeLa cell lysate as well as in other eukaryotic and prokaryotic cell lysates (data not shown). This phenomenon is also known for other ADP-ribosyltransferases, e.g. exoenzyme S, which appears to target Ras as a preferred substrate (23.Coburn J. Wyatt R.T. Iglewski B.H. Gill D.M. J. Biol. Chem. 1989; 264: 9004-9008Abstract Full Text PDF PubMed Google Scholar). Additional studies are under way to characterize the in vivo substrates of MTX in mammalian and insect cells. We thank Colin Berry (Cardiff School of Biosciences, University of Cardiff, Cardiff, United Kingdom) for providing the pTH21 plasmid encoding MTX30–870.