Search

Epsilon‐toxin (ET) of Clostridium perfringens, which causes fatal enterotoxemia in ungulates, was previously shown to bind to and form a heptameric pore within the detergent‐resistant membranes (DRMs) of MDCK cells. Depletion of cholesterol has also been shown to decrease the cytotoxicity of ET and its heptamerization. In this study, we investigated the effects of changes in sphingolipids, other DRM components of MDCK cells, on the cells' susceptibility to ET. Treatment with fumonisin B1 and PDMP, inhibitors of sphingolipid and glycosphingolipid syntheses, respectively, increased the susceptibility, while D609, a sphingomyelin synthesis inhibitor, had the opposite effect. The exogenous addition of ganglioside GM1dramatically decreased the ET binding, heptamerization and cytotoxicity. These effects were shown not to be due to ET binding to GM1or to denaturation of ET. We also found that the ET cytotoxicity towards MDCK cells decreased with an increase in culture time. In accordance with the resistance observed for prolonged cultured cells, GM3, a major ganglioside component, increased and sialidase treatment increased their susceptibility. These results suggest that membrane‐anchored sialic acid of GM3within DRMs inhibits ET binding, leading to prevention of the heptamerization of ET and cell death. It is also suggested that sialidase produced by this organism aids the targeting of ET to MDCK cells.

Clostridium perfringens epsilon-toxin, which is responsible for enterotoxaemia in ungulates, forms a heptamer in rat synaptosomal and Madin-Darby canine kidney (MDCK) cell membranes, leading to membrane permealization. Thus, the toxin may target the detergent-resistant membrane domains (DRMs) of these membranes, in analogy to aerolysin, a heptameric pore-forming toxin that associates with DRMs. To test this idea, we examined the distribution of radiolabeled epsilon-toxin in DRM and detergent-soluble membrane fractions of MDCK cells and rat synaptosomal membranes. When MDCK cells and synaptosomal membranes were incubated with the toxin and then fractionated by cold Triton X-100 extraction and flotation on sucrose gradients, the heptameric toxin was detected almost exclusively in DRMs. The results of a toxin overlay assay revealed that the toxin preferentially bound to and heptamerized in the isolated DRMs. Furthermore, cholesterol depletion by methyl-beta-cyclodextrin abrogated their association and lowered the cytotoxicity of the toxin toward MDCK cells. When epsilon-protoxin, an inactive precursor able to bind to but unable to heptamerize in the membrane, was incubated with MDCK cell membranes, it was detected mainly in their DRMs. These results suggest that the toxin is concentrated and induced to heptamerize on binding to a putative receptor located preferentially in DRMs, with all steps from initial binding through pore formation completed within the same DRMs.

The hemgene cluster, which consists of hemA, cysGB, hemC, hemD, hemB, and hemLgenes, and encodes enzymes involved in the biosynthetic pathway from glutamyl‐tRNA to uroporphyrinogen III, has been identified by the cloning and sequencing of two overlapping DNA fragments from Clostridium perfringensNCTC8237. The deduced amino acid sequence of the N‐terminal region of C. perfringensHemD is homologous to those reported for the C‐terminal region of Salmonella typhimuriumCysG and Clostridium josuiHemD. C, perfringensCysGBis a predicted 220‐residue protein which shows homology to the N‐terminal region of S. typhimuriumCysG. Disruption of the cysGBgene in C. perfringensstrain 13 by homologous recombination reduced cobalamin (vitamin B12) levels by a factor of 200. When grown in vitamin B12‐deficient medium, the mutant strain showed a four‐fold increase in its doubling time compared with that of the wild‐type strain, and this effect was counteracted by supplementing the medium with vitamin B12. These results suggest that C. perfringensCysGBis involved in the chelation of cobalt to precorrin II as suggested for the CysGBdomain of S. typhimuriumCysG, enabling the synthesis of cobalamin.

The production of heat‐labile enterotoxin (LT) in 76 strains of human enterotoxigenic Escherichia coli(ETEC) varied by a factor of 100. Three ETEC strains that differ in the levels of LT production were chosen for the cloning of LT genes (toxAB)into plasmid pBR322, and the gene structure and expression were compared in E. coliHB101. The recombinant of the low LT‐producing strain produced LT at the same level as that of the moderate LT‐producing strain, but that of the high‐level producer continued to produce at a level 14–21 times higher than the others. The restriction maps of the coding regions of the cloned LT genes (toxAB)were identical, but the flanking regions were dissimilar. The content of LT mRNA per cell, examined by Northern blot analysis, was higher in the high producer than the others by 6 times. The promoter strengths of the recombinants were all alike. LT mRNA of the high producer was more stable than that of the moderate one by 1.3 times, but the difference was not large enough to explain the difference of the content of LT mRNA. It was shown that LT production can be controlled at a transcriptional step, and DNA structure of the flanking regions may be involved in the control of the LT gene expression.

Adherence to a HEp‐2 cell monolayer was tested for in four strains of Salmonella derbywhich were isolated from patients with diarrhea. One strain, SB1, was highly adherent and another strain, SB4, was nonadherent. The other two strains exhibited moderate adherence. Further in vitrostudy of invasion of HEp‐2 cells by S. derbyand its replication in murine peritoneal macrophages was carried out using SB1 and SB4. Thin section electron micrographs revealed that SB1 invaded HEp‐2 cells but SB4 did not. The number of viable bacteria within macrophages was determined at intervals after inoculation of bacteria. The result indicates that SB1 can replicate in the macrophages but SB4 cannot. Flagella and fimbriae were compared by electron microscopy between SB1 and SB4, and their lipopolysaccharides and outer membrane proteins were also compared with each other by SDS‐polyacrylamide gel electrophoresis. The presence of a 41 kDa protein in the outer membranes of SB1 was only the difference detected, suggesting that this protein could be a factor required for adherence of this serovar to epithelial cells.

The colHgene encoding 116‐kDa collagenase of Clostridium histolyticum(cColH) was cloned into an Escherichia coli‐Bacillus subtilisshuttle vector to develop a method for purification of recombinant collagenase (rColH). When plasmid pJCM310 containing the colHgene was introduced into B. subtilisDB104 and the transformant was grown in LB broth at 37 C, stability of the plasmid was not maintained. However, stability was partly improved by growing the transformant in a modified LB broth containing 0.5 Msodium succinate with gentle shaking at 35 C. When the transformant was grown to an optical density of 0.4 at 600 nm in this medium, pJCM310 was stable and rColH was produced in sufficient amounts. rColH was purified to homogeneity by ammonium sulfate precipitation, gel filtration and ion‐exchange chromatography. The yield of rColH from an 800‐ml culture was 0.53 mg and its specific activity was estimated to be 1,210 U per mg of protein. The purified rColH was capable of degrading native type‐I collagen fibril from bovine achilles tendon, as was demonstrated by zymography. A comparison of the N‐terminal amino acid sequence between cColH and rColH revealed that rColH has 10 extra N‐terminal amino acid residues. However, the peptide mapping of rColH with V8 protease was virtually identical to that of cColH. Furthermore, the molecular mass of rColH was estimated to be 112,999 Da by mass spectrometry, coinciding with the value of 112,977 Da, which was predicted from the nucleotide sequence of the colHgene. Therefore, the recombinant B. subtilisculture is capable of serving as a useful source for enzyme purification.

Clostridium perfringensKZ1340 isolated from Antarctic soil was first classified as Clostridium plagarumand later as a lecithinase‐negative variant of C. perfringens. Although the strain produced no detectable lecithinase (phospholipase C, PLC) activity in the culture supernatant, it was shown by Southern blot hybridization to possess a PLC‐encoding gene (plc). To determine the cause of the PLC deficiency, we cloned and sequenced the plcgene from KZ1340. The deduced amino acid sequence consists of 398 amino acid residues, coinciding with those of the plcgenes previously determined. Tyrosine was substituted for histidine at amino acid position 148, which is thought to bind a zinc ion essential for PLC activity. Northern blot analysis revealed that KZ1340 expressed the plcgene at an extremely low level. Furthermore, the plcgene cloned from C. perfringensstrain 13 into a plasmid was expressed weakly in KZ1340, compared to that in strain 13. This indicates that the former strain represses plcgene expression in trans. When a phylogenetic tree of plcgenes was constructed, the KZ1340 plcgene formed a monophyletic branch along with those of various other C. perfringensstrains, supporting the classification of the strain as a variant of C. perfringens.

The effect of γ‐toxin, a thermolysin‐like metalloprotease of Clostridium perfringens, on the inactive ε‐prototoxin produced by the same organism was examined. When the purified ε‐prototoxin was incubated with the purified γ‐toxin at 37 C for 2 hr, the 32.5‐kDa ε‐prototoxin was processed into a 30.5‐kDa polypeptide, as determined by SDS‐polyacrylamide gel electrophoresis. A mouse lethality test showed that the treatment activated the prototoxin: the 50% lethal doses (LD50) of the prototoxin with and without γ‐toxin treatment were 110 and 70,000 ng/kg of body weight, respectively. The lethal activity of the prototoxin activated by γ‐toxin was comparable to that with trypsin plus chymotrypsin and higher than that with trypsin alone: LD50of the prototoxin treated with trypsin and trypsin plus chymotrypsin were 320 and 65 ng/kg of body weight, respectively. The ε‐toxin gene was cloned and sequenced. Determination of the N‐terminal amino acid sequence of each activated ε‐prototoxin revealed that γ‐toxin cleaved between the 10th and 11th amino acid residues from the N‐terminus of the prototoxin, while trypsin and trypsin plus chymotrypsin cleaved between the 13th and 14th amino acid residues. The molecular weight of each activated ε‐prototoxin was also determined by matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry. The C‐terminus deduced from the molecular weight is located at the 23rd or 30th amino acid residue from the C‐terminus of the prototoxin, suggesting that removal of not only N‐terminal but also C‐terminal peptide is responsible for activation of the prototoxin.

ABSTRACTThe neurotoxicity of epsilon-toxin, one of the major lethal toxins produced by Clostridium perfringenstype B, was studied by histological examination of the rat brain. When the toxin was injected intravenously at a lethal dose (100 ng/kg), neuronal damage was observed in many areas of the brain. Injection of the toxin at a sublethal dose (50 ng/kg) caused neuronal damage predominantly in the hippocampus: pyramidal cells in the hippocampus showed marked shrinkage and karyopyknosis, or so-called dark cells. The dark cells lost the immunoreactivity to microtubule-associated protein-2, a postsynaptic somal and dendric marker, while acetylcholinesterase-positive fibers were not affected. Timm’s zinc staining revealed that zinc ions were depleted in the mossy layers of the CA3 subfield containing glutamate as a synaptic transmitter. The cerebral blood flow in the hippocampus was not altered significantly before or after administration of the toxin, as measured by laser-Doppler flowmetry, excluding the possibility that the observed histological change was due to a secondary effect of ischemia in the hippocampus. Prior injection of either a glutamate release inhibitor or a glutamate receptor antagonist protected the hippocampus from the neuronal damage caused by epsilon-toxin. These results suggest that epsilon-toxin acts on the glutamatergic system and evokes excessive release of glutamate, leading to neuronal damage.

We established a new method for detecting enteropathogenic Escherichia coliadhering to HEp‐2 cells. An essential part of the method is an assay of β‐galactosidase activity of adhered bacterial cells. It consisted of the following steps: (1) culture of bacterial cells in a medium containing isopropyl‐thio‐β‐d‐galactoside, an inducer of β‐galactosidase; (2) incubation of a bacterial culture with monolayered HEp‐2 cells in a 96‐well culture plate; (3) washing wells to remove bacterial cells which did not adhere to HEp‐2 cells, and (4) enzymic reaction for β‐galactosidase assay in wells. The strains used differed widely in their β‐galactosidase activities. However, a calibration curve for the enzyme activity, obtained from each bacterial sample, showed that 105bacteria per well permitted an accurate estimation. The enzyme activity of adhered bacteria to the monolayered cells showed that 107bacteria were appropriate for the adherence assay. The number of adhered bacteria thus obtained was in good agreement with a viable cell count. The result indicates that the new method is more reliable than a widely used method, counting the number of bacteria under a microscope. The present method also makes it easy to detect adherent strains of E. coliin large numbers of specimens.

The phospholipase C (α‐toxin) gene (plc) of Clostridium perfringenswas cloned into pUC19 and the effects of the upstream regions on expression of the plcgene were examined in Escherichia coliJM109. When the 0.7‐kb region just upstream of the putative — 35 site of the gene was deleted, production of phospholipase C increased approximately 10‐fold. Northern blot hybridization analysis of the plctranscript showed that the upstream region inhibited transcription from the plcpromoter. Nucleotide sequencing of this upstream region revealed that there are three periodically repeated (dA)5–6tracts between positions −66 and −40 of the plcgene. A fragment containing this sequence showed anomalously slow electrophoretic mobility at low temperature, indicating that the region immediately upstream of the plcpromoter is a locus of sequence directed DNA‐bending. Nested deletions of the upstream region were created from its 5′ end by exonuclease III and the effects of deletions on the expression of the plcgene were examined. When the 77‐bp fragment containing the two (dA)5–6tracts was deleted, phospholipase Cproduction increased markedly. These results indicate that the intrinsic DNA curvature upstream of the plcpromoter is involved in the negative regulation of the plcgene transcription.

The virulence of methicillin‐resistant Staphylococcus aureus(MRSA) was compared with that of methicillin‐sensitive S. aureus(MSSA), using 13 MRSA and 7 MSSA strains isolated from clinical specimens. The infectivity and lethality of the two groups were examined as to the inoculum required to infect 50% of guinea pigs (ID50) and to kill 50% of mice (LD50), respectively. The mean ID50[log10colony forming units (CFU)] for MRSA strains was 7.1 ± 0.60 standard deviation, which was 1.5 higher than that for MSSA strains (P< 0.001). The mean LD50(log10CFU) for MRSA strains was 9.0 ± 0.42, being 1.1 higher than that for MSSA strains (P= 0.001). Pretreatment of mice with cyclophosphamide decreased the mean LD50for MRSA strains more than that for MSSA strains, resulting in the difference in the mean LD50being insignificant (P= 0.502). These results indicate that MRSA is less virulent than MSSA in normal hosts, but that they are equally virulent in immunocompromised hosts. The growth of MRSA strains was much slower than that of MSSA strains in the lag phase, although their growth rates were almost the same in the exponential growth phase, suggesting that the difference in virulence between them may be at least partly due to such a difference in growth.