Phylogenetic evidence for horizontal transfer of mutS alleles among naturally occurring Escherichia coli strains.

mutS mutators accelerate the bacterial mutation rate 100- to 1,000-fold and relax the barriers that normally restrict homeologous recombination. These mutators thus afford the opportunity for horizontal exchange of DNA between disparate strains. While much is known regarding the mutS phenotype, the evolutionary structure of the mutS(+) gene in Escherichia coli remains unclear. The physical proximity of mutS to an adjacent polymorphic region of the chromosome suggests that this gene itself may be subject to horizontal transfer and recombination events. To test this notion, a phylogenetic approach was employed that compared gene phylogeny to strain phylogeny, making it possible to identify E. coli strains in which mutS alleles have recombined. Comparison of mutS phylogeny against predicted E. coli "whole-chromosome" phylogenies (derived from multilocus enzyme electrophoresis and mdh sequences) revealed striking levels of phylogenetic discordance among mutS alleles and their respective strains. We interpret these incongruences as signatures of horizontal exchange among mutS alleles. Examination of additional sites surrounding mutS also revealed incongruous distributions compared to E. coli strain phylogeny. This suggests that other regional sequences are equally subject to horizontal transfer, supporting the hypothesis that the 61.5-min mutS-rpoS region is a recombinational hot spot within the E. coli chromosome. Furthermore, these data are consistent with a mechanism for stabilizing adaptive changes promoted by mutS mutators through rescue of defective mutS alleles with wild-type sequences.

Promiscuous origin of a chimeric sequence in the Escherichia coli O157:H7 genome.

A novel sequence of 2.9 kb in the intergenic region between the mutS and rpoS genes of Escherichia coli O157:H7 and closely related strains replaces a sequence of 6.1 kb in E. coli K-12 strains. At the same locus in Shigella dysenteriae type 1, a sequence identical to that in O157:H7 is bounded by the IS1 insertion sequence element. Extensive polymorphism in the mutS-rpoS chromosomal region is indicative of horizontal transfer events.

Detection of mutator subpopulations in Salmonella typhimurium LT2 by reversion of his alleles.

Defects in the methyl-directed mismatch repair lead to both the hypermutability phenotype and removal of a barrier to genetic exchange between species. Mutator bacteria carrying such defects occur frequently among bacterial pathogens, suggesting that subpopulations of mutators are contained within pathogen clones and give rise to the genetic variants that are acted upon by selective forces to allow survival or successful infection. We report here on the detection of the mutator subpopulation in Salmonella typhimurium and determination of its frequency in laboratory cultures. The analysis involved screening for mutators among revertants of S. typhimurium histidine auxotrophs selected for the His+ phenotype, since the frequency of mutators is expected to be increased in the selected mutant population they helped to spawn. The increases in spontaneous reversion of histidine mutations were first measured in isogenic strains carrying mismatch repair-defective mutH, mutL, mutS, or uvrD alleles, relative to their mismatch repair-proficient counterparts. Screening for the mutator phenotype in nearly 12,000 revertants of repair-proficient strains carrying his mutations highly stimulated for reversion in mutator backgrounds, the base substitution in hisG428 and frameshift in hisC3076, yielded five mutator strains (0.04%). The his+ reversion mutations contained within the newly-arisen mutator strains were characteristic of the predominant nucleotide changes expected in such mutators, as assessed by comparison with the spectra for reversion events in wild-type and mismatch correction-defective backgrounds. The results show that subpopulations of mutators, residing in normal populations at a finite frequency, can be culled from the culture by strong selection for a required phenotype. We calculate that the frequency of mutators in the unselected population of S. typhimurium is 1-4x10-6, an incidence 10-fold lower than that expected based on studies of laboratory cultures of Escherichia coli.

High mutation frequencies among Escherichia coli and Salmonella pathogens.

Here it is reported that the incidence of mutators among isolates of pathogenic Escherichia coli and Salmonella enterica is high (over 1 percent). These findings counter the theory, founded on studies with laboratory-attenuated strains, that suggests mutators are rare among bacterial populations. Defects in methyl-directed mismatch repair underlie all mutator phenotypes described here. Of nine independently derived hypermutable strains, seven contained a defective mutS allele. Because these mutant alleles increase the mutation rate and enhance recombination among diverse species, these studies may help explain both the rapid emergence of antibiotic resistance and the penetrance of virulence genes within the prokaryotic community.

Simultaneous identification of strains of Escherichia coli serotype O157:H7 and their Shiga-like toxin type by mismatch amplification mutation assay-multiplex PCR.

Mismatch amplification mutation assay primers, specific for a unique base substitution in uidA of Escherichia coli O157:H7, was coupled with primers for the Shiga-like toxin I (SLT-I) and SLT-II genes in a multiplex PCR assay. Analysis of 108 bacteria showed that all Escherichia coli serotype O157:H7 strains were identified simultaneously with the SLT types encoded by these strains.

Rapid polymerase chain reaction method for detection of Vibrio cholerae in foods.

The polymerase chain reaction was used to selectively amplify sequences within the cholera toxin operon from Vibrio cholerae O1. Oysters, crabmeat, shrimp, and lettuce were seeded with V. cholerae and then homogenized or washed with alkaline peptone water, followed by short-term (6- to 8-h) enrichment. A detection limit of as few as 1 V. cholerae CFU per 10 g of food was obtained with amplification reactions from crude bacterial lysates. The method is extremely rapid and obviates the need for DNA isolation from a variety of complex food matrices.

Synthetic oligodeoxyribonucleotide probes to detect Kanagawa phenomenon-positive Vibrio parahaemolyticus.

Synthetic oligodeoxyribonucleotide probes were used in the colony hybridization test to examine the association between the Kanagawa phenomenon (KP) and the thermostable direct hemolysin gene (tdh) of Vibrio parahaemolyticus. Representative V. parahaemolyticus strains with a variety of KP reactions and 17 other Vibrio species were examined for homology with four synthetic oligodeoxyribonucleotide probes (19 to 21 bases long) representing different regions of the tdh structural gene. Under stringent conditions, two of the probes were capable of distinguishing KP-positive V. parahaemolyticus from KP-negative or KP weak-positive V. parahaemolyticus which possesses mutated tdh genes. Vibrio hollisae strains hybridized with all four probes under reduced stringency, suggesting that they have tdh-related genes which are homologous but not identical to the tdh gene in all the regions examined. The results suggest that the colony hybridization test with the synthetic oligonucleotide probes is more suitable for the definitive determination of KP-positive strains than the hybridization with the larger gene probe or immunological assays.

DNA colony hybridization method using synthetic oligonucleotides to detect enterotoxigenic Escherichia coli: collaborative study.

The genes that encode several of the enterotoxins produced by Escherichia coli have been cloned by recombinant DNA techniques. When the nucleotide sequence of these genes is determined, defined sequence oligonucleotides that include a part of these genes may be synthesized. A 22-base DNA hybridization probe was produced for each of 2 heat-stable E. coli enterotoxin (ST) genes: STH, from strains originally isolated from humans; and STP, from strains first found in pigs. For this study, 32P end-labeled DNA probes, sonicated calf thymus DNA, and 3 known and 20 unknown (10 ST-positive and 10 ST-negative) strains were sent to each of 23 collaborators. Cultures were spotted onto an agar-based medium and grown into colonies, which were transferred by blotting to cellulose filters, lysed by alkali and steam, and used for DNA colony hybridization with the ST DNA probes. Strains containing an ST gene were recognized as dark spots on an autoradiogram. Of the 460 samples analyzed, 440 (95.7%) were correctly classified by the collaborators. The method has been adopted official first action.

Synthetic oligodeoxyribonucleotide probes for detecting heat-stable enterotoxin-producing Escherichia coli by DNA colony hybridization.

DNA colony hybridization was used to identify and enumerate enterotoxigenic Escherichia coli strains in foods. The cells were identified and enumerated by using synthetic polynucleotide probes for the heat-stable enterotoxin genes. These 22-base oligonucleotides, made from known nucleotide sequences of the genes for the heat-stable enterotoxins of human and porcine strains of E. coli, contain two mismatches between the two heat-stable enterotoxins. Colonies were replicated from agar medium onto paper filters and lysed with alkali followed by steam; probes were end labeled. After overnight hybridization at 40 degrees C and washing at 50 degrees C, autoradiograms were exposed at -70 degrees C. Results were consistent with suckling-mouse tests for heat-stable enterotoxins. A stronger signal was obtained on paper filters than on nitrocellulose filters. Enterotoxigenic E. coli cells were detected when mixed with a 1,000-fold excess of nonenterotoxigenic E. coli cells. This procedure appears to be more acceptable for routine testing than the use of cloned DNA fragments, labeling by nick translation, and lysing colonies on nitrocellulose filters.

Probability of recovering pathogenic Escherichia coli from foods.

The probability of recovering pathogenic Escherichia coli from food by the Bacteriological Analytical Manual method was determined by the effects of several factors: the number of strains per food, the ability of pathogenic strains to survive enrichment, and the frequency of plasmid loss during enrichment. Biochemical patterns indicated the presence of about six E. coli strains per food sample. About half of the strains isolated from humans did not survive enrichment. Among those which grew, plasmid loss, as determined by gel electrophoresis and DNA colony hybridization, ranged from 20 to 95%. The combined effects of failure to survive enrichment and plasmid loss decreased the relative numbers of these strains and reduced the chance of detecting pathogens. To counteract this tendency and obtain a 90 to 95% probability off recovering a given pathogenic strain, 40 to 50 colonies per food sample should be picked during the routine testing of foods.

Genetic methods for the detection of microbial pathogens. Identification of enterotoxigenic Escherichia coli by DNA colony hybridization: collaborative study.

Enteropathogenic Escherichia coli strains may produce a cholera-like, heat-labile enterotoxin (LT) as a virulence factor. The gene that codes for LT can be purified by recombinant DNA techniques and used as a genetic probe for DNA hybridization. These probes detect enterotoxigenic strains as well as strains that may not manifest toxin production but carry the genetic information to do so. In this study, 13 laboratories tested 3 known and 25 unknown (10 positive and 15 negative) cultures of E. coli for the presence of the LT gene. The isolates had been tested and classified by the mouse Y-1 adrenal cell test and an enzyme-linked immunosorbent assay. Cultures were spotted on nitrocellulose filters on MacConkey agar and incubated. Colonies were lysed in situ and their DNA was hybridized to 32P-labeled, purified LT gene DNA (provided to the collaborators). Positive colonies were identified by autoradiography. Of 325 samples, 315 (96.9%) were identified correctly and 10 were misclassified; there were 6 false negative and 4 false positive identifications. Chi-square values indicated that the method agreed with the previous classification and was equally efficient in distinguishing positive and negative samples (95.7 and 98.1%, respectively). The method has been adopted official first action.

Detection and enumeration of virulent Yersinia enterocolitica in food by DNA colony hybridization.

A portion of a 44-megadalton plasmid found in Yersinia enterocolitica 8081 was used as a genetic probe to differentiate virulent and nonvirulent strains of the species. A DNA colony hybridization technique was employed. Three BamHI restriction endonuclease fragments labeled with 32P by nick translation were hybridized to lysed colonies of pure cultures, mixtures of virulent and nonvirulent cells, and portions of a food sample artificially contaminated with virulent Y. enterocolitica. The results of the colony hybridization test for virulence were the same as those obtained by the autoagglutination and suckling mouse tests. DNA colony hybridization detects pathogenic Y. enterocolitica in foods without the need for enrichment or pure cultures.

Foodborne enterotoxigenic Escherichia coli: detection and enumeration by DNA colony hybridization.

Four methods were compared for detecting heat-labile toxin production by Escherichia coli: DNA colony hybridization, two enzyme-linked immunosorbent assays, and the mouse Y-1 adrenal cell reaction. Although results of the methods were in general agreement, there were some differences in specificity and sensitivity. DNA colony hybridization was used to detect and enumerate enterotoxigenic E. coli isolates in artificially contaminated food without enrichment. Sensitivity level was 100 cells per g.

Microbiological Quality of Cocoa Powder, Dry Instant Chocolate Drink Mix, Dry Nondairy Coffee Creamer and Frozen Nondairy Topping Obtained at Retail Markets.

A national survey was conducted of the microbiological quality of three dry ingredients used in beverages and one frozen non-dairy topping obtained at retail markets. Geometric mean aerobic plate counts (APCs) of units examined at 35°C were as follows: 1,313 units of cocoa powder, 6,600 CFU/g; 1,552 units of dry instant chocolate drink mix, 290 CFU/g; 1,559 units of dry non-dairy coffee creamer, 37 CFU/g; and 1,532 units of frozen non-dairy topping, 34 CFU/g. At 30°C, the geometric mean APC was 34 CFU/g for frozen nondairy topping. Geometric means for most probable number determinations of coliform bacteria and Escherichia coli were <3/g for the four products. Geometric mean values for Staphylococcus aureus in three of the products were <10/g; no S. aureus was found in cocoa powder. Geometric mean values for yeasts and molds in dry instant chocolate drink mix and dry nondairy coffee creamer were 8 and 6 CFU/g, respectively.

Microbiological quality of macaroni and noodle products obtained at retail markets.

The microbiological quality of macaroni and noodle products was determined by a statistically based national survey at the retail level. Geometric means of aerobic plate counts for macaroni and noodle products were 520 and 1,400 per g, respectively. Means for yeast and mold counts were 72 per g for macaroni and 100 per g for noodles. Means for counts of coliforms and Staphylococcus aureus were less than 3 per g for both products. Escherichia coli was not found in macaroni but was present in 0.5% of the noodle samples and ranged from 3 to 93 per g.

Recovery of parasitic nematodes from fish by digestion or elution.

Two methods, digestion and elution, were used to recover parasitic nematodes from 470 flatfish belonging to species in the family Pleuronectidae. Samples of similar fish were collected from market lots; half of each sample was subjected to digestion, and half was subjected to elution (sedimentation). The edible (flesh) and the inedible (viscera) portions of each fish were analyzed separately. The total number of nematodes recovered by digestion was 1,110, which was not significantly greater than the 922 nematodes recovered by elution. However, digestion recovered 1,062 nematodes of the anisakine genera Anisakis and Phocanema, which are potentially pathogenic for human consumers of raw of semiraw fish. This number is significantly greater than the 608 pathogenic nematodes recovered by elution. Digestion also recovered 242 more nematodes from the edible flesh than did elution. Conversely, more nonpathogenic nematodes were recovered by elution. Approximately half the fish (240) had been collected in Boston markets, and the other half (230) had been collected in San Francisco markets. Fish from San Francisco each contained an average of eight nematodes, and those from Boston contained an average of less than one nematode per fish.

Nematodes in Fresh Market Fish of the Washington, D.C. Area.

For this study 1,010 fresh whole fish belonging to 14 families, 20 genera, and 23 species were bought in retail markets of the Washington, D.C. area. Most of the fish had been caught in Chesapeake Bay and adjacent waters. Examination for parasitic roundworms, i.e., nematodes, by dissection, candling or digestion of the flesh, and elution of nonedible viscera produced 6,547 nematodes, mostly alive. Among fish species of which 25 or more were examined, spotted hake ( Urophycis regius ) was the most wormy and white perch ( Morone americana ) the least. Only two of the nematodes were recovered from fish flesh; both were Anisakis sp. larvae. Recovered from fish viscera were nine Anisakis sp. larvae. 41 Porrocaecum sp. larvae, 3,221 Thynnascaris spp. larvae and adults, 21 Goezia sp. larvae, and 1,220 Raphidascaris acus larvae. All the aforementioned nematodes are considered to be anisakines; in addition, 225 anisakines were too damaged to be identified more precisely. The other nematodes recovered in the survey were 71 Spinitectus spp. adults and larvae, 114 Bulbodactinis sp. adults and larvae, 108 Metabronema sp. adults and larvae, 111 Spirurinae larvae, 662 Philometra sp. adults and larvae, one Capillaria sp. larva, 447 similar small larvae so undeveloped that they could not be identified, and 294 other nematodes too damaged even for general classification. Only the Anisakis sp. larvae are considered pathogenic to human consumers of raw or semiraw fish. The low incidence of pathogenic anisakines in these fish intermediate hosts is attributed to the absence of definitive hosts (marine mammals) from Chesapeake Bay and adjacent waters. Thirty genera of anisakine nematodes are distinguished morphologically.