Structure of Shiga Toxin Type 2 (Stx2) from Escherichia coli O157:H7
2004; Elsevier BV; Volume: 279; Issue: 26 Linguagem: Inglês
10.1074/jbc.m401939200
ISSN1083-351X
AutoresM.E. Fraser, M. Fujinaga, M.M. Cherney, Angela R. Melton‐Celsa, E M Twiddy, Alison D. O’Brien, Michael N.G. James,
Tópico(s)Bacterial Genetics and Biotechnology
ResumoSeveral serotypes of Escherichia coli produce protein toxins closely related to Shiga toxin (Stx) from Shigella dysenteriae serotype 1. These Stx-producing E. coli cause outbreaks of hemorrhagic colitis and hemolytic uremic syndrome in humans, with the latter being more likely if the E. coli produce Stx2 than if they only produce Stx1. To investigate the differences among the Stxs, which are all AB5 toxins, the crystal structure of Stx2 from E. coli O157:H7 was determined at 1.8-Å resolution and compared with the known structure of Stx. Our major finding was that, in contrast to Stx, the active site of the A-subunit of Stx2 is accessible in the holotoxin, and a molecule of formic acid and a water molecule mimic the binding of the adenine base of the substrate. Further, the A-subunit adopts a different orientation with respect to the B-subunits in Stx2 than in Stx, due to interactions between the carboxyl termini of the B-subunits and neighboring regions of the A-subunit. Of the three types of receptor-binding sites in the B-pentamer, one has a different conformation in Stx2 than in Stx, and the carboxyl terminus of the A-subunit binds at another. Any of these structural differences might result in different mechanisms of action of the two toxins and the development of hemolytic uremic syndrome upon exposure to Stx2. Several serotypes of Escherichia coli produce protein toxins closely related to Shiga toxin (Stx) from Shigella dysenteriae serotype 1. These Stx-producing E. coli cause outbreaks of hemorrhagic colitis and hemolytic uremic syndrome in humans, with the latter being more likely if the E. coli produce Stx2 than if they only produce Stx1. To investigate the differences among the Stxs, which are all AB5 toxins, the crystal structure of Stx2 from E. coli O157:H7 was determined at 1.8-Å resolution and compared with the known structure of Stx. Our major finding was that, in contrast to Stx, the active site of the A-subunit of Stx2 is accessible in the holotoxin, and a molecule of formic acid and a water molecule mimic the binding of the adenine base of the substrate. Further, the A-subunit adopts a different orientation with respect to the B-subunits in Stx2 than in Stx, due to interactions between the carboxyl termini of the B-subunits and neighboring regions of the A-subunit. Of the three types of receptor-binding sites in the B-pentamer, one has a different conformation in Stx2 than in Stx, and the carboxyl terminus of the A-subunit binds at another. Any of these structural differences might result in different mechanisms of action of the two toxins and the development of hemolytic uremic syndrome upon exposure to Stx2. Several serotypes of Escherichia coli produce one or more protein toxins that are closely related to Stx from Shigella dysenteriae serotype 1 (1O'Loughlin E.V. Robins-Browne R.M. Microbes Infection. 2001; 3: 493-507Crossref PubMed Scopus (148) Google Scholar). Stx from S. dysenteriae was first identified by Kiyoshi Shiga for whom these toxins are named (2Shiga K. Zentralbl. Bakteriol. Orig. 1898; 24: 913-918Google Scholar). Collectively these E. coli are known as Stx-producing E. coli (STEC). 1The abbreviations used are: STEC, Stx-producing E. coli; Tris, Tris(hydroxymethyl)aminoethane; PPS, 3-(1-pyridino)-1-propanesulfonate; r.m.s., root mean square; Gb3, globotriaosylceramide; MES, 4-morpholineethanesulfonic acid.1The abbreviations used are: STEC, Stx-producing E. coli; Tris, Tris(hydroxymethyl)aminoethane; PPS, 3-(1-pyridino)-1-propanesulfonate; r.m.s., root mean square; Gb3, globotriaosylceramide; MES, 4-morpholineethanesulfonic acid.E. coli O157:H7 is the STEC responsible for many outbreaks of hemorrhagic colitis or bloody diarrhea in the U.S., Canada, and Japan. In some patients the prodrome of hemorrhagic colitis may progress to the hemolytic uremic syndrome, which culminates in kidney failure. Two types of Stx may be produced by STEC, Stx1 and/or Stx2. The Stx types all have an AB5 structure, in which a single A-subunit is associated with five B-subunits. The A-subunit embodies the N-glycosidase catalytic activity; it acts by removing a specific adenine base from the 28 S rRNA of the 60 S ribosomal subunit within infected cells. Because this adenine base is on a loop of rRNA that is important for elongation factor binding, the toxin is able to shut down protein synthesis in a targeted cell. The A-subunit of Stx1 is nearly identical to the A-subunit of Stx; it differs only by the change of a single serine residue at position 45 to a threonine (Table I). In contrast, the amino acid sequence identity between the A-subunits of Stx1 and Stx2 is only 55%. The B-pentamer of the Stxs contains the binding sites for glycolipids that are located on the surface of target cells, and thus mediates entry of the catalytic A-subunit into cells. Although the Stx1 group is homogeneous, the Stx2 group has a number of variants. Variants of Stx2 include Stx2c, Stx2d, Stx2d-activable, Stx2e, and Stx2f. Only the Stx2c variant of Stx2 has been found in the O157 serotypes. The Stx2 variants are distinguished by a difference in biological activity, immunological reactivity, or the receptor to which they bind. Stx1, Stx2, and the Stx2 variants bind preferentially to the glycosphingolipid globotriaosylceramide (Galα1–4Galβ1–4glucosyl ceramide, Gb3) with the exception of Stx2e, which binds preferentially to globotetraosylceramide (GalNAcβ1–3Galα1–4Galβ1–4glucosyl ceramide). These different binding properties lead to the ability of the toxin to target different cells.Table ISequence alignments of the A-subunits of Stx, Stx1, Stx2, ricin, and the B-subunits of Stx or Stx1 and Stx2 The sequence alignments were generated using ClustalW (39Thompson J.D. Higgins D.G. Gibson T.J. Nucleic Acids Res. 1994; 22: 4673-4680Crossref PubMed Scopus (55183) Google Scholar). For Stx, the sequence identifiers are CAA30741 and CAA30742 from Ref. 40Kozlov Y.V. Kabishev A.A. Fedchenko V.I. Baev A.A. Dokl. Biochem. 1987; 295: 740-744Google Scholar; for Stx1 A, the sequence identifier is AAA98151 from Ref. 41Jackson M.P. Newland J.W. Holmes R.K. O'Brien A.D. Microb. Pathog. 1987; 2: 147-153Crossref PubMed Scopus (133) Google Scholar; for Stx2, the sequence identifiers are CAA30714 and CAA30715 from Ref. 42Jackson M.P. Neill R.J. O'Brien A.D. Holmes R.K. Newland J.W. FEMS Microbiol. Lett. 1987; 44: 109-114Crossref Scopus (289) Google Scholar. Open table in a new tab The molecular structure of the Stx1 B-pentamer alone was first determined using x-ray crystallography. This structure showed five B-subunits in a pentameric arrangement (3Stein P.E. Boodhoo A. Tyrrell G.J. Brunton J.L. Read R.J. Nature. 1992; 355: 748-750Crossref PubMed Scopus (261) Google Scholar), similar to the arrangement of the B-subunits of the heat-labile enterotoxin (4Sixma T.K. Pronk S.E. Kalk K.H. Wartna E.S. van Zanten B.A.M. Witholt B. Hol W.G.J. Nature. 1991; 351: 371-377Crossref PubMed Scopus (464) Google Scholar). In heat-labile enterotoxin the five B-subunits surround a pore through which the carboxyl-terminal residues of the A-subunit traverse. When the structure of Stx was solved, the carboxyl-terminal tail of the A-subunit was found to be surrounded by its five B-subunits like heat-labile enterotoxin, whereas the remainder of the A-subunit lay on one side of the B-pentamer (5Fraser M.E. Chernaia M.M. Kozlov Y.V. James M.N. Nat. Struct. Biol. 1994; 1: 59-64Crossref PubMed Scopus (258) Google Scholar). The determinations of the structures of the Stx1 B-subunits and of their mutated forms, alone and in complex with sugars, have led to a better understanding of the binding properties of the B-pentamer (6Ling H. Boodhoo A. Hazes B. Cummings M.D. Armstrong G.D. Brunton J.L. Read R.J. Biochemistry. 1998; 37: 1777-1788Crossref PubMed Scopus (367) Google Scholar, 7Ling H. Structural Studies of the Interactions between Shiga-like Toxins and their Carbohydrate Receptor. Ph.D. thesis, Department of Biochemistry. University of Alberta, Edmonton, Alberta1999: 185Google Scholar, 8Ling H. Pannu N.S. Boodhoo A. Armstrong G.D. Clark C.G. Brunton J.L. Read R.J. Structure. 2000; 8: 253-264Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar) and have served as templates in the design of inhibitors to cell binding (9Kitov P.I. Sadowska J.M. Mulvey G. Armstrong G.D. Ling H. Pannu N.S. Read R.J. Bundle D.R. Nature. 2000; 403: 669-672Crossref PubMed Scopus (792) Google Scholar, 10Mulvey G.L. Marcato P. Kitov P.I. Sadowska J. Bundle D.R. Armstrong G.D. J. Infect. Dis. 2003; 187: 640-649Crossref PubMed Scopus (136) Google Scholar).The A-subunit of the Stxs can be proteolytically cleaved at a susceptible site (11Olsnes S. Reisbig R. Eiklid K. J. Biol. Chem. 1981; 256: 8732-8738Abstract Full Text PDF PubMed Google Scholar), but the two portions, A1 (27.5 kDa) and A2 (4.5 kDa), remain covalently associated through a disulfide bond until the two cysteine residues are reduced. The disulfide bond is between Cys-A241 and Cys-A260 in Stx2 (Table I). The amino acid residues forming the N-glycosidase catalytic site are on A1. The active site residues were initially predicted from the amino acid sequence similarity between A1 and the catalytic subunit of the toxin ricin found in castor beans (Table I) (12Ready M. Katzin B. Robertus J. Proteins. 1988; 3: 53-59Crossref PubMed Scopus (40) Google Scholar) and subsequently confirmed with genetic studies (summarized by Melton-Celsa and O'Brien (13Melton-Celsa A.R. O'Brien A.D. Aktories K. Just I. Handbook of Experimental Pharmacology. Springer, Freiburg2000: 385-406Google Scholar)). The catalytic mechanism of ricin has been investigated in detail: the adenine base binds between two tyrosine side chains, Tyr-80 and Tyr-123, whereas Arg-180 is thought to act as an acid and Glu-177 either acts as a base to activate a water molecule or stabilizes the transition state (summarized in Ref. 14Miller D.J. Ravikumar K. Shen H. Suh J.-K. Kerwin S.M. Robertus J.D. J. Med. Chem. 2002; 45: 90-98Crossref PubMed Scopus (83) Google Scholar). In the structure of Stx, the disulfide bond connecting A1 and A2 is located near the residues that form the N-glycosidase active site; a methionine residue, only one residue away from the cysteine residue of A2, blocked access to the active site (5Fraser M.E. Chernaia M.M. Kozlov Y.V. James M.N. Nat. Struct. Biol. 1994; 1: 59-64Crossref PubMed Scopus (258) Google Scholar). Thus, the A-subunit in the holotoxin was predicted to be enzymatically inactive. The A2 peptide includes the carboxyl-terminal tail, which in the threedimensional structure of Stx forms an α-helix surrounded by the five B-subunits. The carboxyl-terminal residues of A2 continue through the pore of the B-pentamer, but the last six residues appeared to be disordered in the Stx structure. This may result from the packing arrangement in the crystals, because two holotoxin molecules pack pentamer-to-pentamer, and these six residues would be located at this interface.Despite their similar sequences and expected similar structures, Stx1 and Stx2 cause different degrees and types of tissue damage. If the E. coli O157:H7 with which a patient is infected produces mainly Stx2, the patient is more likely to suffer from hemolytic uremic syndrome than if the E. coli O157:H7 only produces Stx1 (summarized by Siegler et al. (15Siegler R.L. Obrig T.G. Pysher T.J. Tesh V.L. Denkers N.D. Taylor F.B. Pediatr. Nephrol. 2003; 18: 92-96Crossref PubMed Scopus (132) Google Scholar)). The difference in pathogenicity between strains that produce only Stx2 and those that produce Stx1 alone or Stx1 and Stx2 was postulated to be due to the greater affinity of the Stx1 B-pentamer for glycolipids (16Head S.C. Karmali M.A. Lingwood C.A. J. Biol. Chem. 1991; 266: 3617-3621Abstract Full Text PDF PubMed Google Scholar). According to that theory, Stx1 would be more likely to be picked up by intestinal epithelial cells, where it could cause bloody diarrhea; Stx2 could then circulate through the bloodstream to the kidneys where it would cause the hemolytic uremic syndrome (17Tesh V.L. Burris J.A. Owens J.W. Gordon V.M. Wadolkowski E.A. O'Brien A.D. Samuel J.E. Inf. Immun. 1993; 61: 3392-3402Crossref PubMed Google Scholar). However, both protein toxins produce effects in animals that do not appear to be due solely to targeting particular cell types (reviewed by O'Loughlin and Robins-Browne (1O'Loughlin E.V. Robins-Browne R.M. Microbes Infection. 2001; 3: 493-507Crossref PubMed Scopus (148) Google Scholar)). The tissue damage may be due to these other effects of the toxins and could be mediated by residues of the holotoxin that are not involved in receptor binding.We have determined the structure of Stx2 and compared it to that of Stx to contribute to the understanding of what distinguishes Stx and Stx2 from each other. The experimental structures of the holotoxins can be used as templates upon which to model anti-infective compounds to treat the diseases caused by E. coli O157:H7 and other STECs. They can also be used to design chimera that could be used to decipher the different mechanisms of action of the two protein toxins in animal models.EXPERIMENTAL PROCEDURESExpression, Purification, and Crystallization of Stx2—Stx2 was produced in E. coli DH5α using the pLPSH3 plasmid and purified by immunoaffinity chromatography as previously described (13Melton-Celsa A.R. O'Brien A.D. Aktories K. Just I. Handbook of Experimental Pharmacology. Springer, Freiburg2000: 385-406Google Scholar). The toxin eluted from the column with 0.1 m glycine, pH 7.8, and was dialyzed in TEAN (50 mm Tris, pH 7.5, with 1 mm EDTA, 3 mm sodium azide, and 0.2 m NaCl). SDS-PAGE analysis of the purified protein in the presence of the reducing agent, 2-mercaptoethanol, showed that the A-subunit was intact and had not been proteolyzed (data not shown).Crystals of Stx2 were grown by vapor diffusion in hanging drops using sodium formate as the major precipitant. The drops were formed by mixing 1.5 μl of a 9.4 mg/ml solution of the protein with 1.5 μl of the reservoir solution. The reservoir solution contained 4 m sodium formate, 0.1 m MES (pH 6.0), 50 mm 3-(1-pyridino)-1-propanesulfonate (PPS), and 1% ethylene glycol. After a few days, crystals appeared as hexagonal bipyramids. These were seeded into fresh solutions where they grew to a maximum size of 0.35 mm by 0.30 mm within 3 weeks.Structure Determination and Analysis—The crystals of Stx2 diffract to better than 1.8-Å resolution. For the data collection, one crystal was transferred to a solution containing 30% glycerol, 4 m sodium formate, 0.1 m MES (pH 6.0), and 1% ethylene glycol then vitrified at 100 K in the nitrogen stream of a Molecular Structure Corp. low temperature device. Diffraction data were collected from this crystal using a Raxis IV++ image plate detector mounted on a Rigaku RU-H3R rotating anode x-ray generator producing CuKα radiation (λ = 1.5418 Å) that was monochromated and focused by Osmic mirrors. The data were processed using the programs DENZO and SCALEPACK (18Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref Scopus (38355) Google Scholar). From the molecular weight, the unit cell dimensions, and the space group, the crystals were predicted to contain one AB5-hexamer per asymmetric unit.The structure was solved by the molecular replacement method using the programs ALMN (19Collaborative Computational Project N. Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19703) Google Scholar) and BRUTE (20Fujinaga M. Read R.J. J. Appl. Crystallogr. 1987; 20: 517-521Crossref Scopus (176) Google Scholar). The structure solution was performed in two stages, finding first the orientation and position of the B-pentamer in the asymmetric unit and, subsequently, the orientation and location of the A-subunit. The search model for the B-pentamer was the structure of a mutant form of Stx2e (8Ling H. Pannu N.S. Boodhoo A. Armstrong G.D. Clark C.G. Brunton J.L. Read R.J. Structure. 2000; 8: 253-264Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar), identified as 2BOS in the Protein Data Bank (21Berman H.M. Westbrook J. Feng Z. Gilliland G. Bhat T.N. Weissig H. Shindyalov I.N. Bourne P.E. Nucleic Acids Res. 2000; 28: 235-242Crossref PubMed Scopus (26639) Google Scholar). The rotation function for the B-pentamer gave five peaks as expected, and the translation function clearly showed that the space group was P61. The search model for the A-subunit was the A-subunit of Stx (5Fraser M.E. Chernaia M.M. Kozlov Y.V. James M.N. Nat. Struct. Biol. 1994; 1: 59-64Crossref PubMed Scopus (258) Google Scholar), using chain A of the structure identified as 1DM0 in the Protein Data Bank (21Berman H.M. Westbrook J. Feng Z. Gilliland G. Bhat T.N. Weissig H. Shindyalov I.N. Bourne P.E. Nucleic Acids Res. 2000; 28: 235-242Crossref PubMed Scopus (26639) Google Scholar). The model was improved by cycles of maximum likelihood refinement using the Crystallography and NMR System (22Brunger A.T. Adams P.D. Clore G.M. Delano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.-S. Kuszewski J. Jilges N. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. Sect. D Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16929) Google Scholar) and model building using the molecular graphics programs XFIT (23McRee D.E. J. Struct. Biol. 1999; 125: 156-165Crossref PubMed Scopus (2019) Google Scholar) and TOM/FRODO (24Jones T.A. Methods Enzymol. 1985; 115: 157-171Crossref PubMed Scopus (934) Google Scholar). The quality of the model was judged with information from the programs PROCHECK (25Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-291Crossref Google Scholar) and WHATCHECK (26Hooft R.W.W. Vriend G. Sander C. Abola E.E. Nature. 1996; 381: 272Crossref PubMed Scopus (1791) Google Scholar), and the final unit cell dimensions were those suggested by WHATCHECK.The structure of Stx had been determined at lower resolution (2.5 Å) (27Fraser M.E. Chernaia M.M. Kozlov Y.V. James M.N.G. Parker M.W. Protein Toxin Structure. Springer, Heidelberg1996: 173-190Crossref Google Scholar) than that of Stx2 (1.77 Å), so we hypothesized that the model for Stx could be improved by using information from the structure of Stx2. As well, Stx had been refined prior to the use of the maximum likelihood target in refinement (28Pannu N.S. Read R.J. Acta Crystallogr. Sect. A. 1996; 52: 659-668Crossref Scopus (318) Google Scholar). For these reasons, the Stx model was refined for several additional cycles using the original data set (5Fraser M.E. Chernaia M.M. Kozlov Y.V. James M.N. Nat. Struct. Biol. 1994; 1: 59-64Crossref PubMed Scopus (258) Google Scholar). The final refined structures of Stx and Stx2 were analyzed and compared using programs from the CCP4 package (19Collaborative Computational Project N. Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19703) Google Scholar) as well as the molecular graphics programs O (29Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A. 1991; 47: 110-119Crossref PubMed Scopus (13003) Google Scholar) and Swiss PDB Viewer (30Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9466) Google Scholar). The atomic coordinates and the structure factors for both Stx2 and Stx have been deposited in the Protein Data Bank (21Berman H.M. Westbrook J. Feng Z. Gilliland G. Bhat T.N. Weissig H. Shindyalov I.N. Bourne P.E. Nucleic Acids Res. 2000; 28: 235-242Crossref PubMed Scopus (26639) Google Scholar), where they have been assigned the identifiers 1R4P for Stx2 and 1R4Q for Stx.RESULTSStx2—Table II presents the crystallographic details for Stx2, and Table III presents the statistics from the refinement. This is a well determined structure, as demonstrated by the low values of the R-factors and the excellent geometry of the refined model. The catalytic site of the A-subunit of Stx2 is accessible in the crystal structure, and Fig. 1 shows that it contains electron density. Three water molecules could be modeled into this electron density, and their positions and temperature factors were refined. The distances from these water molecules to atoms of the protein were reasonable hydrogen bonding distances, but the three were too close to each other for hydrogen-bonding (2.0, 2.1, and 2.2 Å). Instead, for the final model, a molecule of formic acid (used to crystallize the protein) has been modeled in two alternate conformations to account for the electron density. The crystal structure of Stx2 shows that most of the A-subunit is located on one side of the B-pentamer, but residues of A2 traverse the pore of the B-pentamer and form a short helix on the cell surface binding side (see Fig. 2). In the pore, residues of the A2 portion of Stx2 begin in an α-helical conformation (residues Ser-A278 to Leu-A285), changing to an extended conformation from Asn-A286 to Ser-A289. (The nomenclature indicates the amino acid residue followed by the chain letter and the position of the amino acid residue in that chain.) Ser-A289 initiates the final helix near the carboxyl terminus of the A-subunit. This terminal helix projects out of the B-pentamer pore at an angle of about 30° to the base of the B-pentamer and packs alongside two of the B-subunits. The final two residues of the A-subunit (Gly-A296 and Lys-A297) are not part of the helix, but they are well ordered in the structure.Table IIStatistics for the crystallographic data of Stx2Space groupP61Number of measurements187,198Number of unique reflections66,898Completeness (%) overall (1.86–1.80 Å)98.3% (99.1%)I/σI overall (1.86–1.80 Å)8.0 (2.8)RmergeaRmerge = ΣhklΣi|I – 〈I〉/ΣhklΣi(I) overall (1.86–1.80 Å)6.0% (25.2%)a Rmerge = ΣhklΣi|I – 〈I〉/ΣhklΣi(I) Open table in a new tab Table IIIStatistics for the refined model of Stx2 and for the refined model of StxToxin typeStx2StxResolution limit1.77 Å2.5 ÅCell dimensionsa = b = 143.96 Å,a = 133.00 Åc = 59.30 Åb = 147.18 Åα = β = 90°, γ = 120°c = 82.85 Åα = β = γ = 90°Number of data for refinement66,89053,527Completeness97.6%93.9%R-factor (number of data)15.1% (63,514)19.9% (50,780)R free (number of data)18.4% (3,376)25.7% (2,747)Number of protein atoms4,9819,687Number of water molecules51146Number of atoms in ions or ligands1250r.m.s. deviations from ideal geometry Bond lengths (Å)0.0170.016 Bond angles (°)2.01.9Ramachandran plot Number in most favored regions522 (92.6%)934 (85.8%) Number in additional allowed regions42 (7.4%)147 (13.5%) Number in generously allowed regions06 (0.6%) Number in disallowed regions01 (0.1%) Open table in a new tab Fig. 1Electron density for Stx2. The electron density from the 2 Fo - Fc, αc map is contoured at 1σ near residues forming the active site. The molecular model is drawn as sticks, with water molecules depicted as filled circles. This figure was drawn with the programs BOB-SCRIPT and MOLSCRIPT (35Kraulis P.J. J. Appl. Crystallogr. 1991; 24: 946-950Crossref Google Scholar, 36Esnouf R.M. J. Mol. Graphics. 1997; 15: 132-134Crossref Scopus (1794) Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig. 2Ribbon diagram of Stx2. The A-subunit is red, whereas the B-subunits are orange (chain B), cyan (chain C), green (chain D), yellow (chain E), and blue (chain F). The active site in the A-subunit is marked by the magenta letter A. The side chains of the cysteine residues that link A1 and A2 are depicted in yellow. The sites equivalent to the Gb3-binding sites on the B-pentamer of Stx1 are shown by magenta numbers that distinguish the type of binding site (6Ling H. Boodhoo A. Hazes B. Cummings M.D. Armstrong G.D. Brunton J.L. Read R.J. Biochemistry. 1998; 37: 1777-1788Crossref PubMed Scopus (367) Google Scholar). This figure and Figs. 6 and 7 were drawn with the programs MOLSCRIPT and RASTER3D (35Kraulis P.J. J. Appl. Crystallogr. 1991; 24: 946-950Crossref Google Scholar, 37Merritt E.A. Bacon D.J. Methods Enzymol. 1997; 277: 505-524Crossref PubMed Scopus (3869) Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Stx—We next took the information from the structure of Stx2 and refined further the structure of Stx. The new refinement improved the electron density in some regions, notably residues 43–46 and 184–188 of the A-subunit. Two and three additional residues could be fit at the carboxyl termini of A1 and A2, respectively, but the remaining carboxyl-terminal residues are disordered in the crystal structure. Data for the newly refined model for Stx are included in Table II.DISCUSSIONActive Site of Stx2—The active site of the A subunit is accessible in Stx2 in contrast to the active site of the A subunit in Stx. In the crystal structure of Stx2, a molecule of formic acid and a water molecule bound in the active site mimic the binding of adenine (Fig. 3). The water molecule interacting with Arg-A170 closely approximates the position of N3 of adenine when bound to ricin (31Weston S.A. Tucker A.D. Thatcher D.R. Derbyshire D.J. Pauptit R.A. J. Mol. Biol. 1994; 244: 410-422Crossref PubMed Scopus (106) Google Scholar). The oxygen atoms of formic acid occupy roughly the positions of N1 and N6 or N1 and N7 of the adenine in the two alternate orientations of the molecule of formic acid, and the carbon atom approximates the position of C6 of adenine. The tyrosine residue (residue A77) that forms part of the catalytic site is in a different conformation in Stx and Stx2, as well as in the structures of ricin alone and in ricin when complexed with adenine. The different conformations of the tyrosine residue in these active sites indicate that this tyrosine side chain could adopt different conformations during a catalytic cycle. The role of the tyrosine residue in catalysis is most likely to bind the substrate by stacking with the adenine ring. However, in the structure of Stx, this tyrosine residue forms a hydrogen bond through its hydroxyl group with the carbonyl oxygen atom of Pro-A258, one of the residues of A2. It is this stretch of A2 that lies across the active site of Stx, as shown in Fig. 4.Fig. 3Active site of Stx2 (A) and ricin (B).Dashed lines represent hydrogen bonding interactions, and the interatomic distances are noted. A, configuration of the two alternate conformations of formic acid and a water molecule in the active site of Stx2. B, adenine bound to the A chain of ricin (31Weston S.A. Tucker A.D. Thatcher D.R. Derbyshire D.J. Pauptit R.A. J. Mol. Biol. 1994; 244: 410-422Crossref PubMed Scopus (106) Google Scholar). This figure was drawn using the program LIG-PLOT (38Wallace A.C. Laskowski R.A. Thornton J.M. Prot. Eng. 1995; 8: 127-134Crossref PubMed Scopus (4250) Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig. 4Cα trace of Stx2 superposed on that of Stx. Stx2 is shown in solid lines, whereas Stx is shown in dashed lines in this stereoview. The superposition was based solely on structurally equivalent residues of the B-pentamers. This figure and Fig. 5 were drawn with the program MOLSCRIPT (35Kraulis P.J. J. Appl. Crystallogr. 1991; 24: 946-950Crossref Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Superpositions of Stx2 and Stx—The orientation of the A-subunit with respect to the B-pentamer differs in the two holotoxins Stx2 and Stx. Fig. 5 shows the superposition of the two holotoxins using only residues of the B-pentamers to align the structures, which superpose with an r.m.s. deviation of 1.3 Å for the 340 Cα atoms. The A-subunits superpose with an r.m.s. deviation of 0.84 Å for 246 Cα atoms. The sequences of the A-subunits are most different in the region of the cleavage loop and in the stretch of polypeptide following the second cysteine residue. The B-subunits of Stx2 are two residues longer at the carboxyl terminus than the B-subunits of Stx; in both toxins, the carboxyl-terminal residues of the B-subunits interact with the A-subunit. These different specific interactions may be responsible for the different orientations of the A-subunits with respect to their B-pentamers in the holotoxins.Fig. 5Active sites of Stx2 and Stx. In this stereoview, residues of Stx2 are drawn as gray stick models, whereas residues of Stx are drawn in black. Specific residues of Stx are labeled. The hydrogen bonding interaction between the active site tyrosine residue, Tyr-A77, and the carbonyl oxygen atom of Pro-A258 is represented by a dashed line. Residues Ala-257 to Ala-261 are the part of A2 that lies across the active site.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Comparison of the Receptor-binding Sites—Three distinct receptor-binding sites have been identified on the B-pentamer of Stxs (6Ling H. Boodhoo A. Hazes B. Cummings M.D.
Referência(s)