Differentiation of Shigella strains by plasmid profile analysis, serotyping and phage typing.

Ninety-one Shigella flexneri and 29 Shigella sonnei strains isolated during 1994 from sporadic cases of shigellosis and healthy carriers were analyzed for plasmid profile in order to compare the discriminating ability of this method with that of serotyping and phage typing. Our study revealed 10 plasmid profiles (PP) among S. sonnei strains. A total of 26 out of 29 (89%) S. sonnei isolates could be placed into two phage types (type 1 and 20) comprising four PP for phage type 1 and seven PP for type 20, respectively. Twenty-three different PP were identified among S. flexneri strains. Each serotype was associated with a specific predominant plasmid profile, except serotype 2a. This serotype, the most frequently isolated in Romania, was still rather homogeneous: 33 out of 39 isolates belonged to phage type 125, 27 of which could be placed into two related PP (F10 and F17). Comparison of plasmid patterns of epidemiologically independent S. flexneri serotype 2a isolates with those exhibited by 45 serotype 2a isolates associated to six independent outbreaks revealed the same homogeneity. Thirty-eight strains, representing 4 of 6 outbreaks, had F10 and F17 plasmid patterns. The discrimination indices (D) for plasmid profile analysis alone (D = 0.890) and for the combination of serotyping and phage typing (D = 0.841) indicate that both typing systems have a nearly similar ability of discriminating among S. flexneri strains. By combining the results of the three typing methods, a total of 42 types are distinguished and the D value is 0.942. Our data suggest that plasmid profile analysis can complement phenotyping methods resulting in a degree of discrimination that cannot be achieved by either system alone.

Serological identification of Shigella flexneri strains by the coagglutination reaction.

Several variants of reagents were prepared by coupling type (I, II III, IV, V and VI) and group (3, 4, 6 and 7, 8, 9) anti-Shigella flexneri sera with protein A--containing staphylococcal suspension. For most serum lots coagglutination led to a 1/10 minimum dilution. However, different efficiencies were reported between serum lots and even within the same type or group. No improvement by the coagglutination reaction could be obtained for one lot from group 3, 4 and 7, 8, 9 sera and for both type IV serum lots. The coagglutination reactions were specific both with collection strains and with recently isolated strains and the two staphylococcal suspension lots showed an identical behaviour. Coagglutination may be used for obtaining a better efficiency of type and group. Shigella flexneri sera but this varies in terms of the serum lot and the serum/Staphylococcus combination used.

Study of strains agglutinable with enteroinvasive anti E. coli polyvalent sera isolated from ill and healthy subjects for optimizing the laboratory diagnosis of diarrhoea.

369 E. coli strains that are agglutinated in A or B anti E. coli enteroinvasive (EIEC) polyvalent sera and inducing no keratoconjunctivitis in the guinea pig's eye have been selected. The biochemical reactions of E. coli strains that can be agglutinated in A or B EIEC polyvalent sera occupy an intermediary position between those of Shigella genus and the classical features of Escherichia, i.e.; 87% are motile bacteria, 88%--ferment lactose, 81% form acid from sodium mucate and lysine is not decarboxylated by 87% of the strains. As for the antigenic behaviour with 15 out of the 16 sera employed, the strains under consideration have agglutinated by slide test in one or several sera, with varying intensities. None of the strains agglutinated on slide in the O164 antiserum. Subsequently, agglutinations have been performed in tubes only with the strains that agglutinated on the slide with a +3 or +4 intensity. Out of the 369 strains under consideration only 25 have agglutinated in the tube, all the false positive reactions representing 6.7%. The data presented lead to the conclusion that such strains that have been isolated from ill or healthy patients cannot be considered as pathogenic diarrhoea agents, the differentiation from the invasive strains (in the absence of the guinea pig for the pathogenicity test) cannot be made on the basis of the biochemical reactions, but only by means of a set of monovalent sera and by performing the tube agglutination tests.(ABSTRACT TRUNCATED AT 250 WORDS)

Congo red binding test in enteroinvasive and nonpathogenic Escherichia coli strains.

Out of 6 variants the appropriate media to perform Congo red binding test for enteroinvasive E. coli strains were established (trypto-soy agar Eiken, T.S.A.--Cantacuzino Institute and B.T.S.D.). 12 E. coli strains belonging to enteroinvasive O-serogroups formed on Congo red agar red-coloured, non-coloured colonies or both; cultures from 59 red colonies and 61 white colonies were inoculated in guinea pig eyes. The correlation between positive Congo red binding test and positive Sereny test was 91% (out of 59 red colonies, 47 evoked keratoconjunctivitis in both infected eyes and 7 in only one eye). The negative Congo red binding test corresponds (98.4%) to the failure to induce illness in the guinea pigs' eye (only one out of 61 Crb = colonies was Sereny positive, evoking keratoconjunctivitis in only one of the two infected eyes of a guinea pig). Comparing in vivo lack of pathogenicity in 44 E. coli strains isolated from human normal intestinal flora and negative Congo red binding test, a correlation of 72.73% on B.T.S.D. and 65.91% on T.S.A. medium was found. Developing an appropriate method based on Crb test about 70% of the nonpathogenic E. coli colonies could be eliminated from the laborious agglutination with enteroinvasive O-serogroups E. coli antisera.

Correlation between Congo red binding as virulence marker in Shigella species and Sereny test.

Six variants of nutrient agar were tested in order to chose the suitable media for Congo red binding test. Trypto-soy Eiken, T.S.A - Cantacuzino Institute and B.T.S.D. (a medium prepared with Difco ingredients) are appropriate to distinguish between virulent Crb+ and avirulent Crb- strains. Congo red binding was compared with Sereny test using 25 Shigella strains. The strains were inoculated onto trypto-soy agar Eiken plates with 0.01% Congo red, incubated 24 hours at 37 degrees C. A number of each kind (Crb+ and Crb-) of colonies developed by every strain was subcultured on nutrient agar and Sereny test was performed with these cultures. As expected, all 84 Crb+ colonies in vivo tested, produced keratoconjunctivitis. In the case of Crb- colonies a proper correlation with Sereny negative test was observed in 57 out of 73 colonies (78.2%) to which 10.9% (8 out of 73) less virulent (evoking illness in only one of the two inoculated eyes) colonies may be added. As our results confirmed that loss of pigmentation was consistently accompanied by loss or diminishing of virulence, we consider that Congo red binding may be used as an alternative of in vivo test for establishing the virulence of Shigellae in the routine practice of microbiology laboratories which usually are not provided with cell cultures or animals. Its reduced cost is an important advantage, too.

Identification of Shiga-like toxin in Escherichia coli strains, etiological agents of diarrheal disease.

Shiga-like toxin presence, in 20 E. coli strains, etiological agents of diarrheal diseases, is studied by preparing extracts at +4 degrees C, in the presence of chloroform and by i.v. inoculation in mice. In 4 out of 20 strains, Shiga-like toxin in high titres was identified. Most of the strains presented an inconstant and variable production of Shiga-like toxin in comparison with Shigella dysenteriae type 1 (Shiga) reference strain. The authors also confirm the existence of Shiga-like toxin under 2 forms (neutralizable with Shiga antitoxic serum and non-neutralizable). The importance of the obtained results is further discussed from the point of view of pathogeny and diagnosis of the infections produced by these germs.