Performance of a solid phase enzyme immunoassay for detection of group A streptococci in a pediatric office laboratory as refereed by a hospital laboratory.

We evaluated the performance of a new rapid solid phase enzyme immunoassay, SUDS Group A Strep (MUREX Corp., Norcross, GA) for the detection of Group A beta-hemolytic streptococci in a pediatric office practice. Duplicate throat swabs were obtained from 341 children with pharyngitis. One swab was used in the SUDS test and the other was cultured in the office laboratory. Office SUDS and culture (sheep blood agar plate, aerobic 24-hour incubation) were compared with culture using reference techniques (sheep blood agar plate, anaerobic 48-hour incubation) in a hospital laboratory. Compared with hospital laboratory culture, the sensitivity of office SUDS (73.8%) was superior to that of office culture (66.6%) at P = 0.05. Specificities were 93.1 and 98.6%, respectively; positive predictive values were 86.1 and 96.6%; and negative predictive values were 85.9 and 83.5%. The sensitivity and specificity of SUDS compared with office culture were 88.5 and 87.8%, respectively, but would have been 93 and 94% had hemolyzed media not been used on several occasions in the office culture procedure. We conclude that SUDS Group A Strep was significantly more sensitive than throat cultures as performed in a typical pediatric practice although the performance of office cultures could have been improved by standard quality control techniques.

Evaluation of a latex agglutination test for diagnosis of Clostridium difficile-associated colitis.

Current methods for diagnosis of Clostridium difficile-associated colitis (CAC) based on detection of cytotoxin B by a tissue culture assay (TCA) require technical expertise and up to 48 hours incubation. Recently, a latex agglutination (LA) test (Marion Laboratories) for rapid diagnosis of CAC has become available. Although early evaluations have been favorable, new evidence suggests that the LA reagent binds a soluble bacterial antigen that is not unique to toxigenic strains of C. difficile. The authors examined 201 stools received for CAC testing by LA and a reference TCA and investigated discrepant results. They obtained 29 LA(+)/TCA(+) and 155 LA(-)/TCA(-) results. Eleven patients had LA(+)/TCA(+) and 155 LA(-)/TCA(-) results. Eleven patients had LA(+)/TCA(-) results and 6 had LA(-)/TCA(+) results. The sensitivity and specificity of the LA were 83% and 93%, respectively, compared with TCA. The predictive values of positive and negative results obtained with the LA were 72% and 96%, respectively. Concentrated broth supernatants and live suspensions of three C. difficile isolates with LA(+)/TCA(-) results were tested in a rabbit ileal loop assay. All failed to demonstrate ability to produce an enterotoxin. The authors conclude that the LA method is suitable for rapid screening, but LA(+) results require confirmation by testing with other methods.

Diagnosis of infectious diarrheal diseases.

For many years, microbiologic examination of feces was focused on the isolation of two members of the family Enterobacteriaceae--Salmonellae and Shigellae. Over the past two decades, other enteric pathogens such as the various classes of diarrheagenic E. coli, Campylobacter, Vibrio spp., Yersinia enterocolitica, and Clostridium difficile have gained prominence. A newly recognized protozoan parasite, Cryptosporidium is now known to infect both immunocompetent and immunodeficient individuals. Added to these is the growing list of viruses incriminated in diarrheal diseases and gastritis. This article provides some of the basic information needed to allow for the isolation and identification of many of the currently recognized enteric pathogens. It also provides information on some of the currently available rapid test procedures that will speed up stool specimen testing and allow for more timely reporting to the physician.

Mutagenicity of 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) in the granuloma pouch assay.

The rat granuloma pouch assay was used to assess the in vivo mutagenic potential of 2-amino-3-methyl-imidazo[4,5-f]quinoline (IQ), a heterocyclic aromatic amine which is formed during the frying of meat and broiling of fish. The assay was performed with and without pre-induction by Aroclor. In the initial experiment IQ was injected directly into the pouch of non-induced rats. A 10-fold increase in mutation frequencies was obtained with the 2.0 mg/pouch dose of IQ with uninduced cell populations. In a second study IQ was injected intraperitoneally and into the pouch of rats that had been pre-induced with Aroclor. The dose of IQ administered varied from 0.1 to 2.0 mg/pouch. A 10-fold increase in mutation frequencies was obtained with the 2.0 mg/pouch dose of IQ with uninduced cell populations. Aroclor treatment produced no significant increase in mutation frequencies over uninduced animals. Its mutagenic effect is about 10-fold weaker than that of benzo[a]pyrene or N-methyl-N'-nitro-N-nitrosoguanidine (MNNG).

PHA production, from bacteria to plants.

The genes encoding the polyhydroxyalkanoate (PHA) biosynthetic pathway in Ralstonia eutropha (3-ketothiolase, phaA or bktB; acetoacetyl-CoA reductase, phaB; and PHA synthase, phaC) were engineered for plant plastid targeting and expressed using leaf (e35S) or seed-specific (7s or lesquerella hydroxylase) promoters in Arabidopsis and Brassica. PHA yields in homozygous transformants were 12-13% of the dry mass in homozygous Arabidopsis plants and approximately 7% of the seed weight in seeds from heterozygous canola plants. When a threonine deaminase was expressed in addition to bktB, phaB and phaC, a copolyester of 3-hydroxybutyrate and 3-hydroxyvalerate was produced in both Arabidopsis and Brassica.

Gastrointestinal infections--dietary interactions.

Considerable progress has been made in understanding the complexities involved in the production of bacterial diarrheal diseases. The general mechanisms of disease that have been recognized include enterotoxigenicity, enteroadherence, and invasiveness. The interplay of epithelial cell surface receptors with the surface components of the various bacterial pathogens or their toxins will be reviewed. Knowledge of the stereospecific interactions of bacterial ligands with the eukaryotic receptors has led to the development of new strategies for prevention and therapy. The presence of foodstuffs in the intestinal lumen can contribute by a number of mechanisms to interference with the invading organism's attack on the intestinal cell surfaces. The effects of milk fat and plant lectins on the colonization of the bowel by enteric organisms is discussed.

Mutations affecting aromatic amino acid transport in Escherichia coli and Salmonella typhimurium.

A genetic locus, aroT, located between chr and the trp operon in Salmonella typhimurium, and similar genes, aroR and aroS, near the trp locus of Escherichia coli, were found to be involved in the transport of aromatic amino acids. Genetic lesions at these loci cause a variable diminution in uptake and accumulation of aromatic amino acids, alanine and glycine compared with the wild type. The F'trp episome carries the aro R locus. Curing an E. coli strain of the F'trp episome which covers a chromosomal deletion from cysB through the trp operon and tonB regions, results in a 60 to 80% decrease in tryptophan uptake. The introduction of F'trp into a trp operon-deleted S. typhimurium of low transport ability restores transportability, suggesting that aroT in this organism may be homologous with aroR in E. coli. In E. coli, tryptophan accumulation is normally increased by prior growth in L-tryptophan, while in S. typhimurium it is repressed. In both genera, the trpR gene appears to have no effect on the tryptophan transport capabilities in response to changes in the concentration of L-tryptophan in the medium. Tryptophan transport in the S. typhimurium F'trp hybrid was subject to repression, while in the E. coli strain which carries F'trp covering the equivalent chromosomal delection, an increase in tryptophan accumulation was shown after growth in L-tryptophan supplemented medium.

Transfer of ampicillin resistance between strains of Haemophilus influenzae type B.

Haemophilus influenzae type B, strain W-2, is highly resistant to ampicillin (MIC, 12.5 mug/ml). The ampicillin resistance of strain W-2 was transferred to an antibiotic-sensitive strain TF-2 (RifR, SmR) during mixed incubation on membrane filters at 36 C(transfeer frequency, 4.6 times 10(-5) per donor). Resistance was also transferred from the primary recipient to a secondary one (TF-3, EryR, NovR). The transfer frequency between these derivative strains was 10(-4) after incubation for 30 min. Resistance in strain W-2 remained even after growth in the presence of ethidium bromide or at an elevated temperature, although ampicillin resistance was lost from 13%-25% of transcipient cells after growth in broth. Strain W-2 and transcipients of ampicillin resistance had equivalent levels of beta-lactamase activity, while sensitive segregants and recipient strains demonstrated little or no enzyme activity. Transfer of ampicillin resistance between strains of H. influenzae is probably mediated by conjugation since transfer (1) requires cell-to-cell contact, (2) remains unchanged in the presence of DNase I, and (3) occurs in the absence of demonstrable bacteriophage.

Humoral immune responses to Shiga-like toxins and Escherichia coli O157 lipopolysaccharide in hemolytic-uremic syndrome patients and healthy subjects.

Shiga-like-toxin (SLT)-producing Escherichia coli strains, especially serotype O157:H7, are important causes of bloody diarrhea and are associated with the development of hemolytic-uremic syndrome (HUS). Enzyme-linked immunosorbent assays (ELISAs) were developed for the serologic detection of immunoglobulin M (IgM), IgG, and IgA to Shiga toxin (ST) and SLT-I, IgG to SLT-II, and IgM and IgG reactive against E. coli O157 lipopolysaccharide (LPS). Serum samples were collected from 27 HUS patients (25 pediatric and 2 adult) and tested in the ELISAs. Of 27 patients, 10 (37%) were positive for at least one class of antibody to ST/SLT-I. None of the patients were positive for IgG antibody to SLT-II. Twenty-one of the 27 patients (78%) were positive for antibody to E. coli O157 LPS; 19 of 27 (70%) were positive for IgM, and 20 of 27 (74%) were positive for IgG. None of 48 control serum samples were positive in any of the toxin assays, and only 1 of 48 (2%) and 2 of 48 (4%) were positive for IgM and IgG, respectively, to E. coli O157 LPS. Twelve of the 24 patients (50%) from whom stool specimens were collected were positive by culture for E. coli O157. Overall, serology and culture produced confirmation of infection by SLT-producing organisms in 23 of 27 (85%) HUS patients. A combination of ELISA for antibodies to E. coli O157 LPS and culture provided evidence for 22 of 27 (82%) of these patients. The results indicate that while ELISAs for ST/SLT-I and SLT-II antibodies were of limited diagnostic value, the ELISAs for IgM and IgG to E. coli O157 LPS provided valuable and sensitive adjuncts to culture.

Genetic locus of a gene affecting leucine transport in Salmonella typhimurium.

By using a series of deletion mutations in the region of the tryptophan operon, it has been shown that a gene governing the transport of leucine maps on the side of the chr locus distal to the trp operon.

R factor-mediated antibiotic resistance in Serratia marcescens.

Nineteen of 39 multiresistant strains of Serratia marcescens isolated from clinical sources transferred antibiotic resistance to Escherichia coli or Klebsiella pneumoniae recipients. Marcesins and/or phage prevented effective resistance transfer to E. coli and attempts to select marcescin-resistant mutants of the E. coli recipient strain were unsuccessful. Transfer of resistance was demonstrated for all drugs tested except nalidixic acid. Approximately 90% of donors resistant to tobramycin, ampicillin, or carbenicillin transferred resistance to these drugs. High levels of transferred resistance (minimal inhibitory concentration, >2,500 mug/ml) were demonstrated particularly for ampicillin, carbenicillin, and kanamycin. Transmissibility of Serratia R factors was greatest between isogeneic strains of E. coli K-12. Comparative rates of spontaneous loss of R factor-mediated resistance indicated that Serratia R factors are less stable in E. coli and K. pneumoniae transcipients than in the indigenous hosts.

Serotypes of attachment pili of enterotoxigenic Escherichia coli isolated from humans.

Pili from enterotoxigenic Escherichia coli isolated from humans have been partially purified, and antisera have been prepared. These pili were initially attached to erythrocytes and then removed by thermal elution for purification. Three distinct antigenic types of pili have been identified. Antisera against these three pili types reacted with 60 of 106 (56%) enterotoxigenic E. coli isolated from humans but not with nontoxigenic, normal human fecal isolates of E. coli nor with enterotoxigenic E. coli strains isolated from animals. There was no correlation between pili serogroup and any of the following toxin production (heat labile, heat stable, or both), O antigenic type, geographical source of isolation, or mannose-resistant hemagglutination patterns of various erythrocyte types.

Hemagglutination and adhesiveness of toxigenic Escherichia coli isolated from humans.

Toxigenic strains of Escherichia coli isolated from humans were studied for adherence to human buccal mucosal epithelial cells. The E. coli strains were labeled with 3H-amino acids or fluorescein isothiocyanate. Toxigenic E. coli strains varied in their ability to adhere in the presence of mannose. Of 32 toxigenic strains examined, 52% bound to the buccal cells, whereas none of 8 control strains did so (Mann-Whitney U test, P =0.007). The control strains were nontoxigenic E. coli isolates from humans, enterotoxigenic E. coli isolates from animals, and E. coli K-12 containing the K88 or K99 plasmid; these strains exhibited only background-level adherence in this assay. Among the toxigenic E. coli strains that bound to human buccal mucosal cells, there was no correlation with mannose-resistant hemagglutination (MR-HA) of guinea pig and human erythrocytes. Screening 32 strains, we found the following phenotypes: (i) MR-HA+, buccal adherent; (ii) MR-HA+, buccal nonadherent; (iii) MR-HA-, buccal adherent. Presumably the third group represents strains with another type(s) of surface attachment components not involved in the MR-HA reaction. Our findings indicate that a number of bacterial surface structures can function in MR-HA and buccal adherence.

Enzymatically labelled nucleic acid (NA) probe assays for detection of Campylobacter spp. in human faecal specimens and in culture.

Seventy-two stock clinical isolates of C. jejuni were tested with the SNAP Culture Identification Test Kit (Molecular Biosystems, Inc., San Diego, CA), an enzyme labelled nucleic acid probe colony blot assay, and with the Campyslide (BBL Microbiology Systems, Cockeysville, MD), a latex agglutination test. The sensitivity and overall agreement of both the SNAP and Campyslide tests were 100% in comparison with standard culture and identification tests. In addition, 14 C. jejuni, two C. laridis, and one 'C. upsaliensis', and one C. hyointestinalis, all fresh isolates from acutely ill patients, were detected with both tests. Ninety-eight faecal specimens from acutely ill patients were tested within 48 h using the probe based SNAP Diagnostic Test Kit (Molecular Biosystems, Inc.). Results were compared with standard culture. The probe test detected 4/6 specimens positive by culture; sensitivity was 66%. However, the two culture positive specimens, undetected by the probe test, had pathogens present at levels well below the probe test sensitivity (10(5) CFU g-1 stool). Specificity of the probe test was 98% (90/92 negative specimens) compared to culture. The results of this study demonstrate the accuracy of enzyme labelled oligonucleotide probes in the identification of C. jejuni isolated on culture and suggest a potential role for such probes in the direct detection of campylobacter in clinical faecal specimens.

General method for site-directed mutagenesis in Escherichia coli O18ac:K1:H7: deletion of the inducible superoxide dismutase gene, sodA, does not diminish bacteremia in neonatal rats.

A defined deletion in the Escherichia coli K-12 sodA gene (encoding manganese-superoxide dismutase) linked to a nontransposable selectable marker was generated by transposon Tn5 insertion in combination with in vitro mutagenesis. This mutant allele was used to replace the wild-type sodA gene in an E. coli clinical isolate of serotype O18ac:K1:H7 by bacteriophage P1 transduction. The O18ac:K1:H7 sodA mutant contained no manganese-superoxide dismutase and no hybrid manganese-iron-superoxide dismutase. The sodA mutant was more sensitive to paraquat toxicity than were the parental strain and an isogenic mutant bearing an analogously constructed sodA+ Tn5 insertion allele. In a suckling rat model for bacteremia following oral inoculation of E. coli K1, the sodA mutant was undiminished in its capabilities both to colonize the gastrointestinal tract and, surprisingly, to cause bacteremia. In conjunction with the rat model for E. coli K1 pathogenesis, the method for site-directed mutagenesis described in this paper permits determination of the role played in colonization and bacteremia by any K1 gene which either has a homolog in E. coli K-12 or can be cloned and manipulated therein.

Comparison of two toxins produced by Clostridium difficile.

Clostridium difficile was shown to produce a toxin which could be biochemically separated from the previously described cytotoxin of the same organism. The two proteins differ in biological activity and physical properties. Antiserum prepared to the second toxin does not neutralize the biological activity of the cytotoxin, and immunological cross-reactivity could not be demonstrated. However, some relationship may exist between the two toxins, since the newly described toxin degrades on polyacrylamide electrophoresis into two molecules, one of which appears to migrate with the band of purified cytotoxin. We suggest that this newly described toxin be designated toxin A until its primary biological activity and physical relationship to cytotoxin is determined. This toxin is active in biological assays of enteric disease and may play an important role in C. difficile-induced colitis.