A simple method for purification of Shiga or Shiga-like toxin from Shigella dysenteriae and Escherichia coli O157: H7 by immunoaffinity column chromatography

Abstract A simple method was developed for purifying Shiga toxin and Shiga-like toxin from the culture supernatants and cell lysates of Shigella dysenteriae type 1 and Escherichia coli O157 : H7 grown in modified syncase media. Two steps, DEAE cellulose column chromatography and immunoaffinity column chromatography, were sufficient for obtaining purified toxin. By this procedure, about 0.32–0.75 mg of purified toxin was obtained from 5 1 of culture with high recovery rate (53–62%). The toxins purified by this method from the culture supernatants and cell lysates of S. dysenteriae and E. coli O15 : H7 were immunologically, biologically and structurally indistinguishable.     

Characterization of the purified molybdate-stabilized glucocorticoid receptor from rat liver. An in vitro transformable complex

Rat liver glucocorticoid receptor was purified in the presence of molybdate by a three-step procedure comprising protamine sulfate precipitation, affinity chromatography on a dexamethasone matrix and high-performance size-exclusion chromatography (HPSEC) on a TSK G 3000 SW column. The [3H]triamcinolone-acetonide-receptor complex was obtained in 20% yield with an overall 11 800-fold purification. The dissociation rate constant of this complex was 1.6 X 10(-4) min-1. The purified receptor sedimented at 8.3 S in high-salt and 9.4 S in low-salt sucrose gradients containing molybdate. A 7.0-nm Stokes radius was determined by HPSEC on a TSK G 4000 column in high-salt buffer. The calculated Mr was 278000. Dodecyl sulfate/polyacrylamide gel electrophoresis revealed an almost homogeneous 90 000-Mr band. Three minor bands with Mr of 78 000, 72 000 and 48 000 were also inconstantly seen. An apparent pI = 5.1 was observed for the [3H]steroid complex by isoelectric focusing in agarose gel. Furthermore high-performance ion-exchange chromatography of the purified complex on a DEAE 545 LKB column (DEAE HPLC) yielded a sharp peak eluted at a 315 mM potassium ion concentration. This peak was shown to contain almost all the 90 000-Mr protein. Moreover the purified receptor complex appeared to be transformable to a DNA-binding state after molybdate removal followed by warming 30 min at 25 degrees C in presence of 0.2% bovine serum albumin: 50-78% transformation yield could be demonstrated by DNA-cellulose chromatography. Partial transformation could also be obtained at 0 degrees C in the absence of any added protein and was followed by DEAE HPLC. The transformed complex was eluted by 180 mM potassium.     

Molecular Composition of the 16S Toxin Produced by a Clostridium botulinum Type D Strain, 1873

The 16S toxin was purified from a Clostridium botulinum type D strain 1873 (D-1873). Furthermore, the entire nucleotide sequences of the genes coding for the 16S toxin were determined. It became clear that the purified D-1873 16S toxin consists of neurotoxin, nontoxic nonhemagglutinin (NTNH), and hemagglutinin (HA), and that HA consists of four subcomponents, HA1, HA2, HA3a, and HA3b, the same as type D strain CB16 (D-CB16) 16S toxin. The nucleotide sequences of the nontoxic components of these two strains were also found to be identical except for several bases. However, the culture supernatant and the purified 16S toxin of D-1873 showed little HA activity, unlike D-CB16, though the fractions successively eluted after the D-1873 16S toxin peak from an SP-Toyopearl 650S column showed a low level of HA activity. The main difference between D-1873 and D-CB16 HA molecules was the mobility of the HA1 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Therefore it was presumed that the loss of HA activity of D-1873 16S toxin might be caused by the differences of processing HA after the translation.     

Purification and characterization of human plasma lecithin:cholesterol acyltransferase.

A highly purified (approximately 12 000-fold) homogeneous preparation of human plasma lecithin:cholesterol acyltransferase (LCAT) with 16% yield was obtained by a combination of density ultracentrifugation, high density lipoprotein affinity column chromatography, hydroxylapatite chromatography, and finally chromatography on anti-apolipoprotein D immunoglobulin-Sepharose columns to remove apolipoprotein D. This enzyme preparation was homogeneous by the following criteria: a single band by polyacrylamide gel electrophoresis in 8 M urea; a single band on sodium dodecyl sulfate gel electrophoresis with an apparent molecular weight of 68 000 +/- 1600; a single protein peak with a molecular weight of 70 000 on a calibrated Sephadex G-100 column. Its amino acid composition was different from human serum albumin and all other apoproteins isolated from lipoprotein fractions.     

Monoclonal antibody-based immunoassay for type A Clostridium botulinum toxin is comparable to the mouse bioassay.

A monoclonal antibody (BA11) has been produced against Clostridium botulinum type A neurotoxin by the fusion of myeloma cells (P3 NS1/1-Ag4-1) with spleen cells from BALB/c mice immunized with botulinum type A neurotoxoid. The antibody bound specifically to botulinum type A neurotoxin, showing no cross-reactivity with types B and E botulinum toxins or with any of several other bacterial toxins tested. The monoclonal antibody did not bind to botulinum type A neurotoxin which had been denatured with sodium dodecyl sulfate and bound only weakly to each of the separated heavy and light subunits of the neurotoxin, suggesting a conformational requirement for the antigenic determinant of the antibody. A sensitive immunoassay for C. botulinum type A toxin with monoclonal antibody BA11 in conjunction with an enzyme amplication system has been developed which allows detection of 5 to 10 mouse 50% lethal doses ml-1 of purified neurotoxin. The assay was equally sensitive when applied to the detection of crude toxin in food stuffs; the average value for the minimum level of detectable toxin in extracts of tinned salmon or corned beef was 9 +/- 3.1 mouse 50% lethal doses ml-1.     

Molecular composition of progenitor toxin produced by Clostridium botulinum type C strain 6813

The molecular composition of the purified progenitor toxin produced by a Clostridium botulinum type C strain 6813 (C-6813) was analyzed. The strain produced two types of progenitor toxins (M and L). Purified L toxin is formed by conjugation of the M toxin (composed of a neurotoxin and a non-toxic nonhemagglutinin) with additional hemagglutinin (HA) components. The dual cleavage sites at loop region of the dichain structure neurotoxin were identified between Arg444-Ser445 and Lys449-Thr450 by the analyses of C-terminal of the light chain and N-terminal of the heavy chain. Analysis of partial amino acid sequences of fragments generated by limited proteolysis of the neurotoxin has shown to that the neurotoxin protein produced by C-6813 was a hybrid molecule composed of type C and D neurotoxins as previously reported. HA components consist of a mixture of several subcomponents with molecular weights of 70-, 55-, 33-, 26~21- and 17-kDa. The N-terminal amino acid sequences of 70-, 55-, and 26~21-kDa proteins indicated that the 70-kDa protein was intact HA-70 gene product, and other 55- and 26~21-kDa proteins were derived from the 70-kDa protein by modification with proteolysis after translation of HA-70 gene. Furthermore, several amino acid differences were exhibited in the amino acid sequence as compared with the deduced sequence from the nucleotide sequence of the HA-70 gene which was common among type C (strains C-St and C-468) and D progenitor toxins (strains D-CB16 and D-1873).     

Monoclonal antibody-based immunoassay for type A Clostridium botulinum toxin is comparable to the mouse bioassay

A monoclonal antibody (BA11) has been produced against Clostridium botulinum type A neurotoxin by the fusion of myeloma cells (P3 NS1/1-Ag4-1) with spleen cells from BALB/c mice immunized with botulinum type A neurotoxoid. The antibody bound specifically to botulinum type A neurotoxin, showing no cross-reactivity with types B and E botulinum toxins or with any of several other bacterial toxins tested. The monoclonal antibody did not bind to botulinum type A neurotoxin which had been denatured with sodium dodecyl sulfate and bound only weakly to each of the separated heavy and light subunits of the neurotoxin, suggesting a conformational requirement for the antigenic determinant of the antibody. A sensitive immunoassay for C. botulinum type A toxin with monoclonal antibody BA11 in conjunction with an enzyme amplication system has been developed which allows detection of 5 to 10 mouse 50% lethal doses ml-1 of purified neurotoxin. The assay was equally sensitive when applied to the detection of crude toxin in food stuffs; the average value for the minimum level of detectable toxin in extracts of tinned salmon or corned beef was 9 +/- 3.1 mouse 50% lethal doses ml-1.     

Comparison of antigenicity of toxins produced by Clostridium botulinum type C and D strains.

C1 neurotoxin of Clostridium botulinum strains C-Stockholm (C-ST), C beta-Yoichi, C-468, CD6F, and C-CB19 and type D toxin of strains D-1873 and D-CB16 were purified by gel filtration, ion exchange, and affinity chromatographies. The purified toxins had di-chain structure made of heavy and light chains. The toxins of C beta-Yoichi, C-468, CD6F, and C-CB19 reacted with anti-C-ST heavy chain and anti-C-ST light chain in immunodiffusion tests and enzyme-linked immunosorbent assay, whereas D-CB16 toxin reacted with anti-D-1873 heavy chain and anti-D-1873 light chain. However, C-6813 toxin reacted with anti-D-1873 heavy chain and anti-C-ST light chain but not with anti-C-ST heavy chain or anti-D-1873 light chain immunoglobulin G. These results indicate common antigens in the heavy chains of C-6813 and D-1873 toxins and in the light chains of C-6813 and C-ST toxins. Further, they provide evidence for heterogeneity within type C1 toxin subunits.     

Comparison of antigenicity of toxins produced by Clostridium botulinum type C and D strains

C1 neurotoxin of Clostridium botulinum strains C-Stockholm (C-ST), C beta-Yoichi, C-468, CD6F, and C-CB19 and type D toxin of strains D-1873 and D-CB16 were purified by gel filtration, ion exchange, and affinity chromatographies. The purified toxins had di-chain structure made of heavy and light chains. The toxins of C beta-Yoichi, C-468, CD6F, and C-CB19 reacted with anti-C-ST heavy chain and anti-C-ST light chain in immunodiffusion tests and enzyme-linked immunosorbent assay, whereas D-CB16 toxin reacted with anti-D-1873 heavy chain and anti-D-1873 light chain. However, C-6813 toxin reacted with anti-D-1873 heavy chain and anti-C-ST light chain but not with anti-C-ST heavy chain or anti-D-1873 light chain immunoglobulin G. These results indicate common antigens in the heavy chains of C-6813 and D-1873 toxins and in the light chains of C-6813 and C-ST toxins. Further, they provide evidence for heterogeneity within type C1 toxin subunits.     

Human milk SIgA binds to botulinum type B 16S toxin and limits toxin adherence on T84 cells

Botulinum neurotoxin produced by Clostridium botulinum type B is in the form of a complex of 12S and 16S toxins. Food-borne botulism is caused by these complex toxins which are ingested orally and absorbed from the digestive tract. Here, we show that the human milk SIgA binds to the type B16S toxin. The binding of SIgA to 16S toxin and HA was inhibited by carbohydrates such as galactose, suggesting that the interaction of carbohydrate side chain of the SIgA with the HA of the 16S toxin is important for SIgA-16S complex formation. We also demonstrate that SIgA inhibits the attachment of 16S toxin to intestinal epithelial cells. These data suggest that the interaction of antigen nonspecific SIgA with 16S toxin has a large influence on the absorption of 16S toxin from the intestinal epithelium, and that SIgA may provide insight into developing a therapeutic agent for type B food-borne botulism.     

Modulation of the activity of purified phosphatidylinositol 4-phosphate kinase by phosphorylated and dephosphorylated B-50 protein

To investigate the modulation of phosphatidylinositol 4-phosphate kinase activity by the degree of phosphorylation of the B-50 protein, the enzyme was purified from rat brain cytosol by ammonium sulphate precipitation and DEAE-cellulose column chromatography. Purified rat brain B-50 was phosphorylated with protein kinase C and dephosphorylated with alkaline phosphatase. Incubation of the semi-purified phosphatidylinositol 4-phosphate kinase with 1 microgram of the B-50 preparation enriched in the dephospho-form, resulted in a small reduction of phosphatidylinositol 4-phosphate kinase activity (-16%), whereas incubation with the phospho B-50 preparation inhibited the enzyme activity by 40%. The effect of exogenous B-50 was studied in the presence of 10 micrograms albumin to minimize aspecific protein-protein interactions. The present data on the effect of exogenous B-50 protein on phosphatidylinositol 4-phosphate kinase activity, further support our hypothesis that the phosphorylation state of B-50 may be a regulatory factor in phosphoinositide metabolism in rat brain.     

Isolation of corticotropin-beta-lipotropin common precursor from ovine pituitary glands

Ovine corticotropin-beta-lipotropin common precursor was purified to homogeneity from commercial frozen ovine pituitary glands. A crude preparation was obtained following a procedure published elsewhere (Lee, T.H. and Lee, M.S. (1977) Biochemistry 16, 2824-2829) and was purified by gel filtration on Sephadex G-200 in the presence of 0.5% SDS and 0.1% 2-mercaptoethanol, and under an atmosphere of nitrogen. The gel filtration was repeated once. The partially purified preparation obtained from the second Sephadex G-200 gel filtration was further fractionated by preparative SDS-acrylamide gel electrophoresis, using immunoprecipitated and electrophoretically purified [125I]corticotropin-beta-lipotropin common precursor as a marker. The preparation was judged homogeneous by the appearance of a single protein band in analytical SDS-acrylamide gel electrophoresis, which exhibited both corticotropin and beta-lipotropin immunoreactivities, and a single symmetrical peak in high-pressure liquid chromatography on a reverse phase C18 column. The isolated ovine corticotropin-beta-lipotropin common precursor possessed specific activities of 116 micrograms of immunoreactive corticotropin and 210 micrograms of immunoreactive beta-lipotropin per mg of protein, equivalent to 89 and 62% of theoretical values, respectively. The amino acid composition of the homogeneous preparation was determined.     

Purification of heparinase and heparitinase by affinity chromatography on glycosaminoglycan-bound ah-sepha-rose 4b

Heparinase and heparitinase were separated from an extract of Flavobacterium heparinum, induced with heparin by using column chromatography on hydroxylapatite. As the heparinase preparation contained chondroitinases B and C, chondroitinase B was removed by rechromatography on a hydroxylapatite column. Chondroitinase C was then eliminated by column chromatography on O-phosphono(“phospho”)-cellulose. The heparinase preparation thus obtained was free from sulfoamidase for 2-deoxy-2-sulfoamino-D-glucose (GlcN-2S), sulfatase for 2-amino-2-deoxy-6-O-sulfo D-glucose (GlcN-6S), as well as (δ_4,5glycosiduronase for the unsaturated disaccharides obtained from heparin. The remaining sulfatase for 4-deoxy-α-L-thero-hex-4-enopyranosyluronic acid 2-sulfate (δUA-2S) in the heparinase preparation was removed by affinity chromatography with dermatan sulfate-bound AH-Sepharose 4B coated with dermatan sulfate. The heparitinase preparation separated by column chromatography on hydroxyla patite was purified by affinity chromatography with heparin-bound AH-Sepharose 4B coated with heparin. Sulfatase for 2-amino-2-deoxy-6-O-sulfo-D-glucose (GlcN-6S) and δ_4,5glycosiduronase for the unsaturated disaccharides obtained from heparin were removed by this chromatography. Sulfatase for 4-deoxy-α-L-threo-hex-4-enopyranosyluronic acid 2-sulfate (δUA-2S) remaining in the heparitinase preparation was finally removed by column chromatography on hydroxylapatite. The recoveries of the purified preparations of heparinase and heparitinase were estimated to be 39 and 50%, respectively, from the crude enzyme fractions obtained by the first column chromatography on hydroxyl- patite. The purified heparinase and heparitinase were free from all enzymes that could degrade the sulfated unsaturated disaccharides produced from heparin with heparinase.     

The purification and amino acid sequence of the lethal neurotoxin Tx1 from the venom of the Brazilian 'armed' spider Phoneutria nigriventer

A lethal neurotoxic polypeptide of Mr 8 kDa was purified from the venom of the South American 'armed' or wandering spider Phoneutria nigriventer by centrifugation, gel filtration on Superose 12, and reverse phase FPLC on columns of Pharmacia PepRPC and ProRPC. The purified neurotoxin Tx1 had an LD50 of 0.05 mg/kg in mice following intracerebroventricular injection. The complete amino acid sequence of the neurotoxin was determined by automated Edman degradation of the native and S-carboxymethylated protein in pulsed liquid and dual phase sequencers, and by the manual DABITC/PITC double coupling method applied to fragments obtained after digestions with the S. aureus V8 protease and trypsin. The neurotoxin Tx1 consists of a single chain of 77 amino acid residues, which contains a high proportion of cysteine. The primary structure showed no homology to other identified spider toxins.     

Characterization of Nontoxic-Nonhemagglutinin Component of the Two Types of Progenitor Toxin (M and L) Produced by Clostridium botulinum Type D CB-16

A 9.8-kbp DNA fragment which contained a neurotoxin gene and its upstream region was cloned from Clostridium botulinum type D strain CB-16. Nucleotide sequencing of the fragment revealed that genes encoding for hemagglutinin (HA) subcomponents and one for a nontoxic-nonhemagglutinin (NTNH) component were located upstream of the neurotoxin gene. This strain produced two toxins of different molecular size (approximately 300 kDa and 500 kDa) which were designated as progenitor toxins (M and L toxins). The molecular size of the NTNH component of L toxin was approximately 130 kDa on SDS-PAGE and its N-terminal amino acid sequence was M-D-I-N-D-D-L-N-I-N-S-P-V-D-N-K-N-V-V-I which agreed with that deduced from the nucleotide sequence. In contrast, the M toxin had a 115-kDa NTNH component whose N-terminal sequence was S-T-I-P-F-P-F-G-G-Y-R-E-T-N-Y-I-E, corresponding to the sequence from Ser141 of the deduced sequence. A 15-kDa fragment, which was found to be associated with an M toxin preparation, possessed the same N-terminal amino acid sequence as that of the 130-kDa NTNH component. Furthermore, five major fragments generated by limited proteolysis with V8 protease were shown to have N-terminal amino acid sequences identical to those deduced from the nucleotide sequence of 130-kDa NTNH. These results indicate that the 130-kDa NTNH of the L toxin is cleaved at a unique site, between Thr and Ser, leading to the 115-kDa NTNH of the M toxin.     

ADP-ribosylation of the rho/rac gene products by botulinum ADP-ribosyltransferase: identity of the enzyme and effects on protein and cell functions

1. Botulinum C1 toxin and C3 exoenzyme were purified from the culture filtrate of type C Clostridium botulinum strain 003-9, and specific antibodies were raised against each protein. Immunochemical analysis using these antibodies revealed the presence of minute amount of a C3-like molecule in C1 toxin preparation which tightly binds to the toxin component(s). This enzyme complex was separated from the major neurotoxin. Thus, the ADP-ribosyltransferases in C1 and D toxins and C3 exoenzyme appear to come from the same origin, and should be called together botulinum C3 enzyme. 2. Botulinum C3 enzyme ADP-ribosylates the rho and rac gene products, a family of small molecular weight GTP-binding proteins homologous to ras p21s. This ADP-ribosylation occurs at Asn41 of the rho products which is located in their putative effector domain, suggesting that it interferes interaction of these GTP binding proteins with their effector molecules. 3. When incubated with PC-12 cells, the enzyme inhibits cell growth and induces neurites and acetylcholine esterase. Several lines of evidence suggest that the ADP-ribosylation of the rho/rac proteins is responsible for these changes.     

Platelet derived growth factor is present in human placenta: purification from an industrially processed fraction

Platelet derived growth factor was purified from an industrially processed fraction of human placenta (EAP) donated by the Institut Merieux. We first demonstrated that EAP contains PDGF and the quantity of this growth factor was estimated by inhibition of its biological activity using antibodies against PDGF. According to this first estimation, 1 l of EAP (obtained from 125 kg of placenta) contains 10-1000 micrograms of PDGF. A purification procedure including fast flow chromatography (cationic S), heparin Sepharose affinity, chromatography on Cibacron Blue followed by a reverse phase on a C8 column gave a 6000-fold enrichment with a yield of 14%. This result suggests that the PDGF content in 1 l of EAP is between 10 and 30 micrograms. Mitogenic activity was measured on human fibroblast AG1523, Chinese hamster fibroblasts CCL39 and bovine epithelial cells BEC. Dose-response curves indicate that our preparation of purified PDGF from human placenta induces 50% of the maximal tritiated thymidine incorporation in CCL39 at a dose of 5 ng of PDGF/ml of culture medium.     

Ganglioside-Induced Adherence of Botulinum and Tetanus Neurotoxins to Adducin

Abstract: Preincubation of botulinum neurotoxin serotype A, B, or E with ganglioside GT1b was previously found to enhance adherence of botulinum neurotoxin to synapsin I and an ∼116-kDa bovine brain synaptosomal protein; in contrast, adherence to these two proteins by tetanus neurotoxin required preincubation with GT1b. We have now found that preincubation of the neurotoxins with ganglioside GD3 enhances their adherence to the ∼116-kDa protein more than that with GT1b. A purified preparation of the water-soluble ∼116-kDa protein was obtained from bovine brain synaptosomes by preparative column sodium dodecyl sulfate-polyacrylamide gel electrophoresis and two-dimensional gel electrophoresis. N-Terminal amino acid sequences were obtained for two tryptic fragments of the ∼116-kDa protein. These sequences matched with the data bank sequences for β-adducin, a cytoskeletal protein. The carboxy-terminal tail region of adducin, but not the head region, was adhered to by the neurotoxins. Adherence of the neurotoxin to adducin and synapsin I may facilitate presentation of the neurotoxin to its specific substrate(s).     

Purification of S1 muclease from Takadiastase by affinity chromatography on single-stranded DNA-acrylamide columns.

When S1 nuclease from Takadiastase was partially purified according to previously reported methods, it showed a 10 to 15 fold increase in specificactivity. Although such preparations were highly active on single-stranded DNA, they had traces of activity on native DNA and were contaminated by T1-RNase. The S1 enzyme was further purified by a single step of affinity chromatography on single-stranded DNA-acrylamide column to a final purification of 275-fold. This preparation was free of T1-RNase and had an absolute specificity for single-stranded DNA.     

Heat-stable protease from Pseudomonas fluorescens T16: purification by affinity column chromatography and characterization

A heat-stable extracellular protease from Pseudomonas fluorescens T16, a psychrotroph, was purified by affinity column chromatography on a carbobenzoxy-D-phenylalanine-triethylene tetramine-Sepharose-4B column. The purified enzyme is a monomer with a molecular weight of 38,905 +/- 2,000. In an analytical ultracentrifuge, the Schlieren profile revealed a single symmetrical peak. The sedimentation coefficient was estimated to be 3.93S. Alpha-casein was the preferred substrate, with a Km of 0.05 mM. Heating crude enzyme and purified enzyme in buffer at 50, 90, and 120 degrees C resulted in a rapid initial loss of more than 50% of the initial activity followed by a gradual inactivation which exhibited first-order kinetics. The activation energy for the hydrolysis of casein was calculated to be 3.2 kcal/mol (13.4 kJ/mol).