点击显示 收起
Department of Clinical Pharmacology, Research Institute, International Medical Center of Japan
Bioresources Research Laboratory, The Institute of Medical Chemistry, Hoshi University, Tokyo
PRESTO, Japan Science and Technology Agency
Department of Functional Materials Science, Saitama University, Saitama, Japan
Shiga toxin (Stx) is a major virulence factor of Stx-producing Escherichia coli. Recently, we developed a therapeutic Stx neutralizer with 6 trisaccharides of globotriaosyl ceramide, a receptor for Stx, in its dendrimer structure (referred to as "SUPER TWIG [1]6") to function in the circulation. Here, we determined the optimal structure of SUPER TWIG for it to function in the circulation and identified a SUPER TWIG with 18 trisaccharides, SUPER TWIG (2)18, as another potent Stx neutralizer. SUPER TWIGs (1)6 and (2)18 shared a structural similarity, a dumbbell shape in which 2 clusters of trisaccharides were connected via a linkage with a hydrophobic chain. The dumbbell shape was found to be required for formation of a complex with Stx that enables efficient uptake and degradation of Stx by macrophages and, consequently, for potent Stx-neutralizing activity in the circulation. We also determined the binding site of the SUPER TWIGs on Stx.
Shiga toxin (Stx)producing Escherichia coli (STEC), including O157:H7, cause diarrhea, hemorrhagic colitis, and, sometimes, potentially fatal systemic complications, such as hemolytic-uremic syndrome, in humans [14]. Stx produced by STEC in the gut traverses the epithelium and passes into the circulation, where it causes vascular damage in specific target tissues, such as the brain and the kidneys, resulting in systemic complications. Therefore, an effective Stx neutralizer that specifically binds to and inhibits Stx in the circulation would be expected to be a promising therapeutic agent.
Stx is classified into 2 closely related subgroups, Stx1 and Stx2. Epidemiological and experimental studies have suggested that Stx2 has greater clinical significance than does Stx1 [5, 6]. Both Stxs consist of a catalytic A-subunit that has RNA N-glycosidase activity and a pentameric B-subunit that is responsible for binding to the functional cell-surface receptor globotriaosyl ceramide (Gb3 [Gal{1-4}-Gal{1-4}-Glc1-ceramide]) [4, 7, 8]. Because it is known that multiple interactions of the B-subunit pentamer with the trisaccharide moiety of Gb3 are essential for high-affinity binding of Stx to its receptor, several Stx neutralizers containing the trisaccharide in multiple configurations have been developed [915].
In a previous study, we developed a therapeutic Stx neutralizer with 6 trisaccharides in its dendrimer structure (referred to as "SUPER TWIG [1]6") to function in the circulation [13]. This SUPER TWIG neutralized Stx in vivo by a dual mechanism: (1) it bound to Stx with high affinity and inhibited its Gb3-dependent incorporation into target cells, and (2) it induced active uptake and subsequent degradation of Stx by phagocytic macrophages present in the reticuloendothelium in vivo. In the present study, we synthesized several series of SUPER TWIGs and determined the optimal structures required for them to function most effectively in the circulation. As a result, we identified a SUPER TWIG with 18 trisaccharides, SUPER TWIG (2)18, as another potent Stx neutralizer in vivo.
We also determined the binding site of the SUPER TWIGs on Stx. Ling et al. identified, in the crystal structure of the Stx1 B-subunit in a complex with a trisaccharide receptor analogue, 3 receptor-binding sitesthat is, sites 1, 2, and 3, per B-subunit monomer [16]all of which were shown to be involved in the binding to Gb3 present under the physiological conditions [17]. Recently, analysis of the crystal structure of Stx2 also predicted the presence of the corresponding trisaccharide-binding sites on its B-subunit [18]. Here, we prepared a series of Stx B-subunit mutants and found that site 3 plays a pivotal role in high-affinity binding of SUPER TWIG (2)18 to either Stx1 or Stx2 B-subunits, demonstrating, for the first time, the significance of this site for development of an Stx neutralizer that functions in the circulation.
MATERIALS AND METHODS
Materials.
SUPER TWIGs (0)3, (1)6, and (1)12 were synthesized as described elsewhere [19]. The other SUPER TWIGs were synthesized as described elsewhere (K.M. and D.T., unpublished data). Recombinant Stx1 and Stx2 were prepared according to published methods [20]. Recombinant histidine-tagged Stx1 B-subunit (1BH) and Stx2 B-subunit (2BH), in which 6 histidine residues were added at the carboxy-termini, were prepared as described elsewhere [15]. Phospholipid vesicles containing Gb3 were prepared by use of phosphatidylcholine and Gb3 (molar ratio, 24 : 1). 125I-labeled Stx1 (125I-Stx1) and Stx2 (125I-Stx2) were prepared as described elsewhere [21]. Alexa Fluor 488labeled Stx2 (Alexa-Stx2) was prepared by use of an Alexa Fluor 488 Protein Labeling Kit (Molecular Probes), in accordance with the manufacturer's protocol.
Site-directed mutagenesis of 1BH and 2BH.
Site-directed mutagenesis of pET-28a1BH and pET-28a2BH was performed by use of a QuikChange Kit (Stratagene). The mutagenic oligonucleotides are listed in table 1. The presence of all of the mutations was confirmed by dideoxy sequencing analysis of the region of interest. All of the mutant B pentamers were obtained as described above and were characterized by gel-filtration column chromatography, to confirm protein integrity.
Cells.
Vero cells were cultured as described elsewhere [13]. U937, a human histiocytic lymphoma cell line, was maintained in RPMI 1640 supplemented with 10% fetal calf serum. To differentiate toward the macrophage lineage, U937 cells (2 × 105 cells/well in a 24-well plate) were incubated for 16 h with phorbol myristoyl acetate (50 ng/mL; Sigma) plus ionomycin (1 mol/L; Sigma).
Kinetic analysis of SUPER TWIG binding to immobilized B-subunits.
SUPER TWIG binding to immobilized 1BH, 2BH, and mutant B-subunits was quantified by use of a BIAcore system instrument (BIAcore), as described elsewhere [15]. Ni2+ was fixed on a nitrilotriacetic acid sensor chip (BIAcore), and recombinant 1BH or 2BH (10 g/mL) was injected into the system, to become immobilized on the chip. Various concentrations of compounds were injected (time 0) into the immobilized 1BH or 2BH at a flow rate of 20 L/min, to reach a plateau at 25°C. The resonance unit is an arbitrary unit used by the BIAcore system. The binding kinetics were analyzed by Scatchard plot, by use of BIAEVALUATION 3.0 (BIAcore) software.
125I-Stxbinding assay.
An 125I-Stxbinding assay was performed as described elsewhere [13]. Vero cells were treated with 125I-Stx1 or 125I-Stx2 (1 g/mL) in the absence or presence of the desired amount of a given SUPER TWIG for 30 min at 4°C. After extensive washing, the cells were dissolved in lysis solution (0.1 mol/L NaOH and 0.5% SDS). Recovered radioactivity was measured by use of a -counter (Packard).
Cytotoxicity assay.
Subconfluent Vero cells in a 96-well plate were treated with Stx1 or Stx2 (10 pg/mL) in the absence or presence of the desired amount of a given compound for 72 h. The relative number of living cells was determined by use of a WST-1 Cell Counting Kit (Wako Pure Industries).
Intravenous administration of Stx2 to mice.
A lethal dose of Stx2 (0.25 ng/g of body weight) was administered to 514 female ICR mice (1820 g; Japan SLC) through a tail vein, with or without the desired amount of a given SUPER TWIG. The animal-experimentation guidelines of the International Medical Center of the Japan Research Institute were followed in animal studies. The data were analyzed by Kaplan-Meier survival analysis or, when no mice had died by the end of the observation, by Fisher's exact test.
Uptake of 125I-Stx2 by macrophages.
U937 cellderived macrophages were incubated with 125I-Stx2 (1 g/mL) in the absence or presence of a given SUPER TWIG (10 g/mL) for 30 min at 37°C. After extensive washing, recovered radioactivity was measured.
Confocal microscopy.
U937 cellderived macrophages were incubated with Alexa-Stx2 (1 g/mL) and LysoTracker (0.2 mol/L; Molecular Probes), which was used as a lysosomal marker, in the absence or presence of a given SUPER TWIG (10 g/mL) for 1 h at 37°C. Confocal laser scanning microscopy was performed with an LSM510 confocal microscope (Carl Zeiss). Simultaneous double-fluorescence acquisitions were performed by use of the 488-nm and the 543-nm laser lines to excite Alexa Fluor 488 and LysoTracker, respectively.
RESULTS
Optimal structure of SUPER TWIG for it to function in the circulation.
In a previous study, we developed 3 SUPER TWIGs(0)3, (1)6, and (1)12with 3, 6, and 12 trisaccharides, respectively [19] (figure 1), and found, by use of glutathione S-transferasefused Stx1, that SUPER TWIGs (1)6 and (1)12 specifically bound to Stx1 B-subunit pentamer with very low dissociation constant (KD) values [13]. To optimize the structure of SUPER TWIG, we synthesized another series of SUPER TWIGs, with 4, 9, 18, and 36 trisaccharides (figure 1), and determined their KD values for binding to Stx B-subunits by use of 1BH and 2BH. As shown in table 2, the KD values of SUPER TWIGs (1)4, (1)6, (1)9, (1)12, (2)18, and (2)36, with respect to 1BH and 2BH, were in a similar range, suggesting that increasing the number of trisaccharides up to 36 in a single molecule did not significantly affect the KD value. In contrast, the KD values of SUPER TWIGs (0)3 and (0)4 were much higher than those of the others; however, SUPER TWIGs (0)4 and (1)4 have the same number of trisaccharides. These results clearly indicate the importance of trisaccharide grouping, rather than the number of trisaccharides, for high-affinity binding.
All of the newly synthesized SUPER TWIGs, except for SUPER TWIG (0)4, markedly inhibited the binding of 125I-Stx1 and 125I-Stx2 to Vero cells (figure 2A). The half-maximal IC50 values of SUPER TWIGs (1)4, (1)9, (2)18, and (2)36 were 0.43, 0.34, 0.21, and 0.21 mol/L, respectively, for 125I-Stx1 binding and 1.4, 11, 2.1, and 9.5 mol/L, respectively, for 125I-Stx2 binding. Corresponding to its high KD values, the IC50 values of SUPER TWIG (0)4 were much higher than those of the other SUPER TWIGs. SUPER TWIG (2)18 was the only compound whose IC50 values for both Stxs were lower than those of SUPER TWIG (1)6 (0.33 and 3.5 mol/L for 125I-Stx1 and 125I-Stx2 binding, respectively). SUPER TWIGs (1)4 and (2)18 markedly inhibited the cytotoxic activities of both Stx1 and Stx2 toward Vero cells (figure 2B). The IC50 values of SUPER TWIGs (1)4 and (2)18 were 0.19 and 0.18 mol/L, respectively, for Stx1 and 0.52 and 0.26 mol/L, respectively, for Stx2. In contrast, the IC50 values of SUPER TWIGs (1)9 and (2)36 were 26 and 17 mol/L, respectively, for Stx1 and 18 and 19 mol/L, respectively, for Stx2. Almost no inhibitory effect was observed with SUPER TWIG (0)4. These results indicate that SUPER TWIGs (1)4 and (2)18, as well as SUPER TWIG (1)6, all of which have 2 clusters of trisaccharide symmetrically located through their hydrophobic core structure, potently inhibited the biological activities of Stx1 and Stx2 in vitro.
Next, the inhibitory effect of each SUPER TWIG on the lethal effect of Stx2 intravenously administered to mice was investigated, because Stx2 is known to be more toxic than Stx1, both in vitro and in vivo, and clinically more significant [8, 22]. Only SUPER TWIG (2)18 completely suppressed the lethal effect of Stx2 when administered along with the toxin; in marked contrast, 100% of the control mice died within 4 days (mean ± SE survival period of mice without treatment with SUPER TWIG, 3.2 ± 0.2 days; P < .0001) (figure 2C). The SUPER TWIG (2)18treated mice survived >2 months without any pathological symptoms (data not shown). Compared with the strong effect of SUPER TWIG (2)18, only a slight inhibitory effect was observed when Stx2 was injected along with SUPER TWIG (1)4 (P = .0552); however, SUPER TWIG (1)4 was also an effective inhibitor in the in vitro assays.
To further optimize the structure, we synthesized 2 other sets of SUPER TWIGs on the basis of the structure of SUPER TWIG (1)6 (figure 3A). With respect to 1BH and 2BH, the KD values of SUPER TWIGs (1)2 and (1)3, which had 2 and 3 trisaccharides, respectively, were much higher than those of SUPER TWIGs (1)4, (1)5, and (1)6 (tables 2 and 3), suggesting that at least 4 trisaccharides are required for high-affinity binding. In the in vivo experiment, however, none of the SUPER TWIGs with <6 trisaccharides sufficiently suppressed the lethal effect of intravenously administered Stx2 in the mouse model (figure 3B). These results indicate that at least 6 trisaccharides are required for a SUPER TWIG to effectively function in the circulation.
The KD values of SUPER TWIGs (1)26, (1)46, and (1)56which have alkyl chains with 2, 4, and 5 carbons, respectively, between the central and the terminal siliconswere in a similar range, which was comparable to that of SUPER TWIG (1)6 (table 3). SUPER TWIG (1.5)6, with 4 silicons in the core, also had similar KD values. In contrast, SUPER TWIG (2)6, in which the relative distance between adjacent trisaccharides was longer than that of the others, had much higher KD values: 64 mol/L toward 1BH and 50 mol/L toward 2BH (table 3). SUPER TWIGs (1)46, (1)56, and (1.5)6 completely suppressed the lethal effect of intravenously administered Stx2 in the mouse model (P < .0001), whereas the inhibitory effect of SUPER TWIG (1)26 was slightly weaker (figure 3B). Corresponding to the high KD values, SUPER TWIG (2)6 did not show any inhibition of the lethal effect. These results indicate that the distance between the 2 terminal silicons present in the core should be at least 11 and that the terminal trisaccharides must be clustered in high density to function effectively in vivo.
SUPER TWIG (2)18dependent uptake of Stx2 by macrophages.
In a previous study, we found that SUPER TWIG (1)6 induced active uptake and subsequent degradation of Stx2 by phagocytic macrophages present in the reticuloendothelium in vivo, a step that has been shown to be involved in the Stx-detoxification mechanism of SUPER TWIG in a mouse model [13]. Of the SUPER TWIGs synthesized in the present study, only SUPER TWIG (2)18 markedly induced the uptake of 125I-Stx2 by U937 cellderived macrophages, and its efficiency was even better than that of SUPER TWIG (1)6 (figure 4A). Consistent with this observation, SUPER TWIG (2)18 induced the uptake of Alexa-Stx2 and its subsequent transfer to lysosomes for degradation in macrophages, as confirmed by the colocalization of Alexa-Stx2 with a lysosome marker (figure 4B). The SUPER TWIG (2)18dependent degradation of Stx2 was also confirmed by the active release of radioactive trichloroacetic acidsoluble degradation products into the culture medium after incorporation of 125I-Stx2 (data not shown). In contrast, a much weaker effect was observed with SUPER TWIG (2)36, although this SUPER TWIG bound to the Stx2 B-subunit with even higher affinity than did SUPER TWIG (2)18. Combined with the finding that only SUPER TWIG (2)18 completely suppressed the lethal effect of intravenously administered Stx2 in mice (figure 2C), these observations further confirm the significant role played by active uptake and degradation of Stx2 by macrophages, whose activities could be induced by a SUPER TWIG with the optimal structure determined here, such as SUPER TWIG (1)6 or (2)18.
SUPER TWIG (2)18binding sites on Stx B-subunit pentamer.
To understand how SUPER TWIG (2)18 binds to Stx B-subunits, the effect of amino acid substitution at each trisaccharide-binding site of the B-subunits was investigated. Amino acid substitutions of recombinant 1BH and 2BH were performed as described elsewhere [17], on the basis of the amino acid alignment of Stx1 and Stx2 B-subunits. To characterize the prepared B-subunit mutants, we determined the KD values of Gb3 present in synthetic phospholipid vesicles with respect to each B-subunit mutant. All of the single-point mutations at sites 1, 2, and 3 of 1BH, except for A56Y, markedly increased the KD values, suggesting that all of the trisaccharide-binding sites are involved in the binding of Gb3 under physiological conditions (table 4). This result is consistent with previous observations [17]. Also, the relatively mild inhibitory effect of A56Y, compared with that of G62A, at site 2 is in good agreement with previous data showing that the cytotoxicity of the Stx1 mutant with the A56Y substitution was 1000 times less than that of the mutant with the G62A substitution [17]. In contrast, single-point mutations at site 2 of 2BH did not affect the KD values at all, whereas all of the single-point mutations at sites 1 and 3 markedly increased the KD values, indicating the significant roles that sites 1 and 3, but not site 2, play in the physiological binding of Gb3 to the Stx2 B-subunit (table 4). The smaller contribution of site 2 present in Stx2 may be explained by the structural difference at this site between Stx1 and Stx2, which causes a different conformation in the disulfide-bridged loop involved in the binding to trisaccharides [18].
In terms of binding of SUPER TWIG (2)18, none of the single-point mutations of 1BH significantly affected the KD values, whereas all of the double- and triple-point mutations at the sites, including site 3, resulted in marked reductions in the KD values (table 4). Double-point mutants of 1BH, such as D17E+G62A and F30A+G62A, in which only site 3 was intact, bound to SUPER TWIG (2)18 with KD values similar to the value for wild-type (wt) 1BH. All of these results suggest that site 3 or site 1+2 present on the Stx1 B-subunit was involved in high-affinity binding of SUPER TWIG (2)18. In contrast, single-point mutations at site 3 of 2BH (W33A and D17E) markedly increased the KD values of the SUPER TWIG (table 4). Furthermore, double-point mutants of 2BH, with mutations at sites 1 and 2, effectively bound to the SUPER TWIG with KD values similar to the value for wt 2BH, although the other double- and triple-point mutants of 2BH did not bind to it at all. These results indicate that site 3 present on the Stx2 B-subunit is an essential and a sufficient site for high-affinity binding of SUPER TWIG (2)18. The binding sites of the other SUPER TWIGssuch as (1)6, (1)12, and (2)36, all of which were shown to bind with high affinity to both B-subunits (table 2)were also determined to be the same as those of SUPER TWIG (2)18 (K.N., K.M., D.T., and Y.N., unpublished data).
DISCUSSION
In the present study, we used several series of SUPER TWIGs and determined the following characteristics of the optimal structure for an Stx neutralizer to function in the circulation: (1) a dumbbell-shaped structure in which 2 clusters of trisaccharides are connected via a linkage with a hydrophobic core structure with a length of at least 11 is required; (2) when the dumbbell shape is present, at least 6 trisaccharides are required for in vivo activity, although 4 trisaccharides are sufficient for high-affinity binding to Stx B-subunits and for Stx-neutralizing activities in vitro; and (3) terminal trisaccharides with spacers must be branched from the same terminal silicon atom to be clustered in high density. Interestingly, the first and the second structural requirements, both of which are not necessarily required for Stx-neutralizing activity in vitro, were found to be essential to the marked induction of macrophage-dependent incorporation and degradation of Stx2, further supporting a pivotal role for this mechanism in the in vivo Stx-neutralizing activity of SUPER TWIGs with the optimal structure. A core length of at least 11 might be necessary for the terminal trisaccharides of the SUPER TWIG to embrace the site 3s in a multiple way and, consequently, to provide an adequate volume of hydrophobic region for recognition by macrophages, as described below. The third requirement is also essential to Stx-neutralizing activity in vivo, because SUPER TWIG (2)6, which satisfied the first and second requirements, had only very low affinities for both 1BH and 2BH, even in the in vitro assay. As a result, we identified SUPER TWIG (2)18, which satisfied all of these structural requirements, as another potent Stx neutralizer that functions in vivo.
Using single-, double-, and triple-point mutants of 1BH and 2BH, we found that SUPER TWIG (2)18 bound to the Stx1 B-subunit at trisaccharide-binding site 3 or site 1+2 and to Stx2 B-subunit exclusively at site 3. Although the crystal structure of the complex of the Stx2 B-subunit and a trisaccharide or its analogue has not yet been obtained, our present data showing that the 2 single-point mutants at site 3 of 2BH (W33A and D17E) markedly increased the KD values of Gb3 in lipid vesicles to a similar extent, with respect to SUPER TWIG (2)18, clearly demonstrate the presence of functional trisaccharide-binding activity at this site. Our present results demonstrate that site 3 can be an efficient target for development of an Stx neutralizer that functions in vivo, which is consistent with a recent report that demonstrated the pivotal role of site 3, as well as sites 1 and 2, in receptor binding of Stx1 [17]. In terms of Stx2, SUPER TWIG (2)18 is the first inhibitory compound whose binding site on the B-subunit has been determined to be exclusively site 3.
Our finding that there was no difference in the binding sites on B-subunits between SUPER TWIG (2)18 and the otherssuch as (1)6, (1)12, and (2)36suggests that the exclusive binding to site 3 present on the Stx2 B-subunit is essentially required for high-affinity binding of the SUPER TWIGs but is not sufficient to effectively suppress the toxicity of Stx2 in vivo. Because SUPER TWIGs (1)6 and (2)18, but not the others, induced active uptake of Stx2 by macrophages, it is likely that macrophages recognize the structural differences between the complexes of Stx2 and each SUPER TWIG for uptake and subsequent degradation of Stx2. In the complex of Stx2 and SUPER TWIG (1)6 or (2)18, the hydrophobic region of its core structure would be expected to be exposed irrespective of how the terminal trisaccharides of the SUPER TWIG bind to site 3 of the B-subunit. Such a characteristic structure of the complex, which would not be formed by binding with other SUPER TWIGs, may contribute to recognition by macrophages. In our preliminary data, maleylbovine serum albumin, which is well known to be a broad ligand for a series of scavenger receptors expressed on macrophages [23], partially inhibited incorporation of the complex of Stx2 and SUPER TWIG (1)6 or (2)18, suggesting that this scavenger-receptor pathway might be, at least in part, involved in this process (K.N., K.M., D.T., and Y.N., unpublished data). Although the precise mechanism remains to be elucidated, this type of neutralizer provides a new strategy for detoxification of Stx present in the circulation.
References
1. Karmali MA, Steele BT, Petric M, Lim C. Sporadic cases of hemolytic uremic syndrome associated with fecal cytotoxin and cytotoxin-producing Escherichia coli. Lancet 1983; 1:61920. First citation in article
2. Riley LW, Remis RS, Helgerson SD, et al. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N Engl J Med 1983; 308:6815. First citation in article
3. O'Brien AD, Holmes RK. Shiga and Shiga-like toxins. Microbiol Rev 1987; 51:20620. First citation in article
4. Paton JC, Paton AW. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin Microbiol Rev 1998; 11:45079. First citation in article
5. Ostroff SM, Tarr PI, Neill MA, Lewis JH, Hargrett-Bean N, Kobayashi JM. Toxin genotypes and plasmid profiles as determinants of systemic sequelae in Escherichia coli O157:H7 infections. J Infect Dis 1989; 160:9948. First citation in article
6. Tesh VL, Burris JA, Owens JW, et al. Comparison of the relative toxicities of Shiga-like toxins type I and type II for mice. Infect Immun 1993; 61:3392402. First citation in article
7. Karmali MA, Petric M, Lim C, Fleming PC, Arbus GS, Lior H. The association between idiopathic hemolytic uremic syndrome and infection by verotoxin-producing Escherichia coli. J Infect Dis 1985; 151:77582. First citation in article
8. Melton-Celsa AR, O'Brien AD. Structure, biology, and relative toxicity of Shiga toxin family members for cells and animals. In: Kaper JB, O'Brien AD, eds. Escherichia coli O157:H7 and other Shiga toxin-producing E. coli strains. Washington, DC: American Society for Microbiology, 1998:1218. First citation in article
9. Armstrong GD, Fodor E, Vanmaele R. Investigation of Shiga-like toxin binding to chemically synthesized oligosaccharide sequences. J Infect Dis 1991; 164:11607. First citation in article
10. Kitov PI, Sadowska JM, Mulvey G, et al. Shiga-like toxins are neutralized by tailored multivalent carbohydrate ligands. Nature 2000; 403:66972. First citation in article
11. Dohi H, Nishida Y, Mizuno M, et al. Synthesis of an artificial glycoconjugate polymer carrying Pk-antigenic trisaccharide and its potent neutralization activity against Shiga-like toxin. Bioorg Med Chem 1999; 7:205362. First citation in article
12. Paton AW, Morona R, Paton JC. A new biological agent for treatment of Shiga toxigenic Escherichia coli infections and dysentery in humans. Nature Med 2000; 6:26570. First citation in article
13. Nishikawa K, Matsuoka K, Kita E, et al. A therapeutic agent with oriented carbohydrates for treatment of infections by Shiga toxin-producing Escherichia coli O157:H7. Proc Natl Acad Sci USA 2002; 99:766974. First citation in article
14. Mulvey GL, Marcato P, Kitov PI, Sadowska J, Bundle DR, Armstrong GD. Assessment in mice of the therapeutic potential of tailored, multivalent Shiga toxin carbohydrate ligands. J Infect Dis 2003; 187:6409. First citation in article
15. Watanabe M, Matsuoka K, Kita E, et al. Oral therapeutic agents with highly clustered globotriose for treatment of Shiga toxigenic Escherichia coli infections. J Infect Dis 2004; 189:3608. First citation in article
16. Ling H, Boodhoo A, Hazes B, et al. Structure of the shiga-like toxin I B-pentamer complexed with an analogue of its receptor Gb3. Biochemistry 1998; 37:177788. First citation in article
17. Soltyk AM, MacKenzie CR, Wolski VM, et al. A mutational analysis of the globotriaosylceramide-binding sites of verotoxin VT1. J Biol Chem 2002; 277:53519. First citation in article
18. Fraser ME, Fujinaga M, Cherney MM, et al. Structure of shiga toxin type 2 (Stx2) from Escherichia coli O157:H7. J Biol Chem 2004; 279:275117. First citation in article
19. Matsuoka K, Terabatake M, Esumi Y, Terunuma D, Kuzuhara H. Synthetic assembly of trisaccharide moieties of globotriaosyl ceramide using carbosilane dendrimers as cores: a new type of functional glyco-material. Tetrahedron Lett 1999; 40:783942. First citation in article
20. Noda M, Yutsudo T, Nakabayashi N, Hirayama T, Takeda Y. Purification and some properties of Shiga-like toxin from Escherichia coli O157:H7 that is immunologically identical to Shiga toxin. Microb Pathog 1987; 2:33949. First citation in article
21. Goldstein JL, Basu SK, Brown MS. Receptor-mediated endocytosis of low-density lipoprotein in cultured cells. Methods Enzymol 1983; 98:24160. First citation in article
22. Takeda Y, Kurazono H, Yamasaki S. Vero toxins (Shiga-like-toxins) produced by enterohemorrhagic Escherichia coli (verocytotoxin-producing E. coli). Microbiol Immunol 1993; 37:5919. First citation in article
23. Goldstein JL, Ho YK, Basu SK, Brown MS. Binding site on macrophages that mediates uptake and degradation of acelylated low density lipoprotein, producing massive cholesterol deposition. Proc Natl Acad Sci USA 1979; 76:3337. First citation in article