Search

Two bacteriophages isolated onEnterobacter cloacae (C2, C2F) and three isolated onErwinia herbicola (E3, E16P, E16B) were purified by D2O gradient centrifugation. Phage-containing fractions were negatively stained and examined by electron microscopy. Phages C2, C2F, E3, and E16P showed an elongated head 153×51 nm and a short noncontractile tail 12 nm long terminated by at least two short fibers. These phages correspond to the rate taxonomic group C3. Big capsomeres composing the phage head were evidenced when phage suspensions in D2O were stained. Phage E16B showed an elongated head 97×40.5 nm, and a contractile tail 89 nm long. This phage corresponds to the extremely rate group A3.

Eight strains isolated from birds, reptiles, and water constitute a new DNA hybridization group that is 37–58% related toEdwardsiella tarda and less than 10% related to other species of Enterobacteriaceae (SI nuclease method). This homogeneous group (78–100% relatedness within the group) constitutes a new species that is namedEdwardsiella hoshinae sp. nov. (type strain, CIP 78.56 ATCC 33379). Strains of this species produce acid fromd-mannitol, sucrose,d-trehalose, and salicin, and give a positive malonate test. Seven other strains that produced acid fromd-mannitol and sucrose (but not fromd-trehalose and salicin) and were malonate negative were found to belong to theEdwardsiella tarda DNA hybridization group. The base composition of the DNAs ofE. tarda andE. hoshinae is 55–58 mol% G+C.

Forty strains of Actinobacillus lignieresii, Actinobacillus equuli, Actinobacillus suis, Actinobacillus capsulatus, and Pasteurella ureae were 30 to 100% related as determined by deoxyribonucleic acid (DNA)-DNA hybridization (S1 nuclease method) and could be subdivided into five DNA hybridization groups. The DNA relatedness among A. suis, A. equuli, and P. ureae strains is indicative of some overlap among these three species. A comparison of DNA relatedness data with phenetic data showed that no single test can definitely separate A. suis from A. equuli. “Actinobacillus salpingitidis,” the Actinobacillus sp. described by Ross et al. (Int. J. Syst. Bacteriol. 22:39–46, 1972), “Actinobacillus seminis,” Actinobacillus actinomycetehicomitans, Pasteurella multocida, Pasteurella haemolytica, Pasteurella aerogenes, Pasteurella gallinarum, Pasteurella pneumotropica, “Pasteurella betti”, “Pasteurella anatipestifer,” and Pasteurella testudin is were only 0 to 11% related to P. ureae, A. suis, A. equuli, A. capsulatus, and A. lignieresii.

Seven biochemical groups were found among strains previously labeledSerratia liquefaciens (groups C1ab, C1c, C1d, EB, RB, RQ, and Adc). Comparison of biochemical data with DNA relatedness data allowed the definition (or redefinition) of threeSerratia species:Serratia liquefaciens sensu stricto (group C1ab),Serratia proteamaculans (groups C1c, EB, RB, and RQ), andSerratia grimesii sp. nov. (groups C1d and Adc). Biochemical group RQ, which is genomically related toS. proteamaculans at the borderline of species level, is proposed as a new subspecies ofS. proteamaculans (Serratia proteamaculans subsp.quinovora). Group Adc (3 strains) is also ambiguously related toS. grimesii, but no other proposal is made pending additional studies. The type strains of the newly named taxa,S. grimesii andS. proteamaculans subsp.quinovora, are respectively ATCC 14460 and strain 4364 (= CIP 8195 = ATCC 33765).

A defined agar medium (hereinafter designated caprylate-thallous [CT5 agar) containing 0.01% yeast extract, 0.1% caprylic (n-octanoic) acid, and 0.025% thallous sulfate is highly selective for all Serratia species and effectively discriminates against most non-Serratia strains likely to be in the same habitats. The selectivity of CT agar is demonstrated by the very high efficiency of colony formation (mean, 80.7% of that on a nonselective complex medium) on CT agar by known Serratia strains and the very low efficiency of colony formation (close to zero) on CT agar by bacterial strains known not to be Serratia. The utility of this medium in actual clinical laboratory practice is demonstrated by the more rapid and higher recovery of Serratia on this selective medium as compared to conventional procedures of in-tandem runs of 513 consecutive urine, feces, and sputum specimens. Pigmented and nonpigmented Serratia strains deliberately added to fecal specimens can be selectively and quantitatively recovered on CT agar. CT agar compares favorably with, or in some cases is an improvement over, other selective media which have been recommended for isolating Serratia. This selective CT agar medium could be quite useful in ecological surveys, especially those related to hospital-acquired infections.

A collection of 30 clones derived from 20 luminous and nonluminous Xenorhabdus strains was studied by using deoxyribonucleic acid (DNA) relatedness and electron microscopy, as well as growth factor requirement, nutritional ability, and other biochemical tests. Nonluminous Xenorhabdus cells each contained a crystal-like structure, whereas the cells of luminous strains each contained an accumulation of amorphous material associated with mesosomes and microtubules of unknown function. Two DNA relatedness groups of nonluminous strains were delineated, one of which corresponded to Xenorhabdus nematophilus, a species associated with the nematode Steinernema feltiae (= Neoplectana carpocapsae). The other group contained yellow-pigmented bacteria associated with Steinernema bibionis and new Steinernema species. No definite phenotypic characteristic (except pigmentation) separated the two DNA relatedness groups of nonluminous Xenorhabdus strains. Four DNA relatedness groups were delineated among the luminous clones studied. These four groups could be separated by biochemical tests. Unfortunately, different clones derived from the same strain belonged to different DNA relatedness groups. This happened with the type strain of Xenorhabdus luminescens.

Using the selective caprylate-thallous agar medium, the presence ofSerratia species was systematically examined in 623 plant samples. A total of 167Serratia strains was isolated from these plant samples and identified to species and biogroups. Uniform and characteristicSerratia populations were found in figs and coconuts: (i)Serratia ficaria was recovered from most figs collected in California, Tunisia, and France; various biotypes ofS. marcescens also were found in figs; (ii) onlyS. marinorubra was recovered from coconuts bought on two continents. From plants other than figs and coconuts, representatives were isolated of all eightSerratia species we presently recognize—with a large preponderance ofS. liquefaciens andS. proteamaculans. These other plant samples fell into threeSerratia-prevalence groups: (i) vegetables-mushrooms-mosses-decaying plant material (53.8% of these samples were positive forSerratia); (ii) grasses (23.7% positive); and (iii) trees and shrubs-small plants (8.4% positive). PigmentedS. marcescens biotypes were rarely isolated from plants (except from figs). Of theS. marcescens biogroups most frequently encountered in nosocomial and iatrogenic infections of man, A3 and A4 were isolated from plants in this study, but A5/8 and TCT were not.

Fourteen strains isolated from figs, caprifigs, and fig wasps collected in California and Tunisia, and from a small black ant in France, constitute a new DNA hybridization group that is 25–56% related toSerratia species, and 6–17% related to other species of Enterobacteriaceae. This homogeneous group (90% relatedness within the group) constitutes a new species that is namedSerratia ficaria sp. nov. (type strain, ICPB 4050, ATCC 33105). Strains of this species have a characteristic odor, similar to that ofS: odorifera andPseudomonas perolens. No strain ofS. ficaria has yet been recovered from clinical specimens.

Serratia liquefaciens andS. liquefaciens-like strains belonging to one of seven biotypes (C1ab, C1c, C1d, EB, RB, RQ, and Adc) were characterized by DNA relatedness (S1TCA method). These strains formed three distinct DNA relatedness groups: (i)S. liquefaciens sensu stricto (biotype C1ab); (ii)S. proteamaculans (biotypes C1c, EB, and RB); and (iii)S. grimesii (biotype C1d). Two biotypes were at the borderline of species level: Biotype RQ and biotype Adc were, respectively, related toS proteamaculans andS. grimesii. All of these strains were clearly distinct fromS. plymuthica and the otherSerratia species.

Seven new isolates of Serratia plymuthica (including two from a human source) and 24 isolates resembling unclustered Serratia strain 38 (Grimont et al., J. Gen. Microbiol 98:39-66, 1977) are described. Deoxyribonucleic acid relatedness studies, obtained with labeled reference deoxyribonucleic acid from S. plymuthica 392 confirm that S. plymuthica is a discrete species. Mean percent relatedness of S. plymuthica isolates to strain 392 was 88 ± 13.6 (standard deviation), whereas the mean percent relatedness of S. liquefaciens isolates (the closest species) to strain 392 was 52 ± 12.4. Strain 38 and twenty-four “38-like“ isolates constitute a new deoxyribonucleic acid hybridization group that is 28 to 43% related to Serratia species and to the “Citrobacter-like” group of Leclerc and Buttiaux, and 15 to 22% related to other known species of Enterobacteriaceae. This 38-like group constitutes a new species that is named Serratia odorifera sp. nov. (type strain, ICPB 3995). Two biotypes (1 and 2) are described. Strains of this species will grow on caprylate-thallous agar (selective for Serratia spp.), and they have a characteristic odor. Most strains of S. odorifera were recovered from clinical specimens.

Forty-eightSerratia strains associated with insects were, identified to species level and biotyped according to recent taxonomic schemes. Each strain was submitted to 36 biochemical tests, including 23 carbon source utilization tests. Twenty-eight strains were assigned to eight biotypes ofSerratia marcescens: A1a, A2a, and A6a (pigmented biotypes: 18 strains); and A3a, A3b, A4a, A5, and TCT (nonpigmented biotypes: 10 strains). However, biotypes A8a, A8b, and A8c, which are frequently involved in nosocomial infections, were not found in insects. Ninetten strains were identified asS. liquefaciens (S. proteamaculans) biotypes C1a (12 strains), C1c (4 strains), C1d (2 strains), and one atypicalS. liquefaciens strain. Only one strain was identified asS. marinorubra (a nonchitinolytic species). The recent emergence ofSerratia in human pathology calls for a reevaluation of the idea of usingSerratia to biologically control insects.

Deoxyribonucleic acid relatedness studies (S1 nuclease method) showed that 25 strains of Haemophilus ducreyi from diverse geographic origins form a homogeneous deoxyribonucleic acid hybridization group. A total of 17 type or reference strains of established or proposed Haemophilus species were 0 to 6% related to H. ducreyi at the optimal temperature. These findings do not support the assignment of H. ducreyi to the genus Haemophilus.

Deoxyribonucleic acid (DNA) was extracted from mycobacterial cells by gentle lysis (sodium dodecyl sulfate and pronase treatment) of wall-deficient forms obtained after D-cycloserine and glycine treatment. DNA from Mycobacterium fallax CIP 8139T(= ATCC 35219T) (T = type strain) was 3H-labeled by nick-translation. DNA relatedness studies (S1 nuclease method) showed that the seven M. fallax strains studied form a single DNA hybridization group which is less than 25% related to 16 other species of the genus Mycobacterium. The guanine-plus-cytosine content of M. fallax DNA was 68 mol%, and the guanine-plus-cytosine contents of Mycobacterium aurum, Mycobacterium chitae. Mycobacterium duvalii, Mycobacterium flavescens, Mycobacterium gilvum, Mycobacterium neoaurum, Mycobacterium parafortuitum, Mycobacterium thermoresistibile, and Mycobacterium vaccae DNAs were 64 to 65 mol%.

A deoxyribonucleic acid relatedness study of 13 Mycobacterium chelonae strains indicated that M. chelonae subsp. chelonae and M. chelonae subsp. abscessus are two different genomic species, both distinct from Mycobacterium fortuitum and “Mycobacterium peregrinum.” One strain of M. chelonae constitutes a third genomic species. The type strain of M. fortuitum and the reference strain of “M. peregrinum” (strain ATCC 14467) belong to two different genomic groups.

The correspondence between complete serotype and biotype (P.A.D. Grimont and F. Grimont, J. Clin. Microbiol. 8:73-83, 1978) of 474 Serratia marcescens strains was studied. Of 127 serotypes, 70 were represented by two or more strains of the same serotype belonged to one biotype. However, for 91% of serotypes, strains of the same serotype belonged to one biogroup--i.e., a group of closely related biotypes. Biogroups are A1 (A1a, A1b); A2/6 (A2a, A2b, A6a, A6b); A3 (A3a, A3b, A3c, A3d); A4 (A4a, A4b); A5/8 (A5, A8a, A8b, A8c); and TCT (TCT, TT). Only two serotypes were composed of a mixture of pigmented and nonpigmented biogroups. Pigmented biogroups (A1 and A2/6) were otherwise differentiated from nonpigmented biogroups (A3, A4, A5/8, and TCT) by serotyping. Some biogroups preferentially occurred in some O serogroups: A4 in 01; A2/6 in O6, O8, and O14; and A3 in O9, O12, and O15. Three H serogroups were found to be biochemically homogeneous: H1, H7, and H20 were respectively and uniquely composed of biogroups A4, TCT, and A3. A square matrix of O versus H serogroups, with the corresponding biogroup for each O X H combination, was used for comparisons between O groups and between H groups. Identical patterns of biogroups were shown by serogroups O6, O8, and O14. Taxonomical, ecological, and practical consequences of these findings are discussed.

Two deoxyribonucleic acid hybridization groups were found among Brochothrix strains isolated from various sources. One genospecies contained 165 strains and included the type strain of Brochothrix thermosphacta. The other genospecies (five strains) represented a new species, for which the name Brochothrix campestris sp. nov. is proposed. The type strain is strain S3 (= CIP 102920 = ATCC 43754). B. campestris differs from B. thermosphacta by growth in the presence of 8 and 10% NaCI and 0.05% (wt/vol) potassium tellurite, by hippurate hydrolysis, and by acid production from rhamnose.

Eighty-four strains of homofermentative lactobacilli isolated from meat showed a great diversity in carbohydrate fermentation patterns. DNA-DNA hybridization (S1 nuclease method) showed 13 representation strains to constitute a homogeneous genomic group 65 to 82 % related to Lactobacillus sake ATCC 15521, with ΔTm values below 1.5°C. The description of L. sake is emended.

Summary Haemophilus aegyptius should be considered a junior subjective synonym of Haemophilus influenzae . Both nomenspecies are indistinguishable by DNA/DNA hybridization (S1 nuclease method). No single phenotypic test can unambigously separate H. aegyptius from the other biotypes of H. influenzae .

Revision sur la base d'etudes des acides nucleiques. Le genre B ne contient qu'une seule espece, B. melitensis, avec differents biovars. Le determinisme genetique des differences pathogeniques est a preciser

Summary A total of 224 strains of the genus Pseudomonas sensu lato were studied by 99 carbon source utilization tests using BioMerieux Biotype-100 strips and 96 substrate oxidation tests using the Biolog GN Micro Plate system. Biotype-100 and Biolog data were computer analyzed using programs of the Taxotron package (Institut Pasteur) and the Bionum software (Biolog), respectively. Unweighted pair group method of averages (UPGMA) yielded 34 Biotype phenons and 36 Biolog phenons. A total of 18 phenons had identical composition in both systems. Six phenons were almost identical in structure whereas some phenons in one system were lumped in the other system. Some species (represented by at least three strains) constituted homogeneous phenons in both Biotype-100 and Biolog systems: Pseudomonas aeruginosa, P. tolaasii, P. mendocina, P. cichorii, P. viridif lava, P. fragi, P. stutzeri, P. agarici, P. alcaligenes, and P. pseudoalcaligenes . None of the two systems could discriminate P. fuscovaginae from P. asplenii, and P caricapapayae from P corona f aciens. P. corrugata was a tight subphenon in both systems. Strains labeled P. f luorescens were found in six phenons with either Biotype-100 or Biolog systems. Strains of P. putida were found in three (Biolog) or four (Biotype) phenons. Strains of P marginalis were distributed over five phenons in both systems. The use of Biotype-100 strips for Pseudomonas characterization was also validated by comparison with published carbon source utilization data.

In 1978, Grimont et al. (2) demonstrated the identity of Serratia Ziquefaciens biovar Clc with Erwinia proteamaculans ICPB XP 176 (,=ATCC 19323) and proposed the new combination Serratia proteamaculans. The taxonomic scheme in use at that time required that a choice be made between the epithets Ziquefaciens and proteamaculans. S. proteamaculans (Paine and Stansfield 1919) Grimont, Grimont, and Starr 1978 was then regarded as a senior subjective synonym of S. Ziquefaciens (Grimes and Hennerty 1931) Bascomb et al. 1971. Since the publication of this proposal (2), several new developments have occurred, which necessitate a reconsideration of the problem. The sequence of events was as follows. (i) Holmes (3) requested that the Judicial Commission of the International Committee on Systematic Bacteriology issue an Opinion conserving the specific epithet Ziquefaciens over proteamaculans in the scientific name Serratia Ziquefaciens (Grimes and Hennerty 1931) Bascomb et al. 1971. The main argument used by Holmes (3) was the worldwide acceptance enjoyed by the name S. Ziquefaciens as opposed to the little-known (among non-phytobacteriologists) epithet proteamaculans. (ii) The recently published Approved Lists of Bacterial Names (5) contain both the names S. Ziquefaciens and S. proteamaculans. The publication of the Approved Lists put an end to any taxonomic uncertainty concerning the type strain of S. Ziquefaciens, which is now definitely ATCC 27592, a member of biovar Clb (1). (iii) A systematic search of the Serratia species associated with plants (Grimont, Grimont, and Starr, manuscript in preparation) conducted in California and in France resulted in the isolation of a number of strains identical to S. proteamaculans (=S. Ziquefaciens biovar Clc [l]) with respect to biochemical properties and deoxyribonucleic acid (DNA) relatedness. Our present understanding is that the habitat of S. proteamaculans is the surfaces of plants, whereas S. Ziquefaciens “stricto sensu” (biovars Cla and Clb) is ubiquitous (unpublished data). (iv) The DNA relatedness among cold-tolerant serratias was re-investigated with 50 strains (P. A. D. Grimont, K. Irino, and F. Grimont, Conference on Taxonomy, Computer Identification of Bacteria and Diagnostic Methods, Liblice Castle, Czechoslovakia, 5-9 May 1980; P. A. D. Grimont, K. Irino, and F. Grimont, manuscript in preparation). With the DNAs from S. proteamaculans ATCC 19323 and S. Ziquefaciens ATCC 27592 as references, two DNA hybridization groups were identified among 24 strains, 13 strains of S. proteamaculans showed 86 f 6% relatedness (mean f standard deviation) to ATCC 19323 (Sl nuclease method, 60°C) and 56 7% relatedness to ATCC 27592, and 11 strains of S. Ziquefaciens stricto sensu showed 99 k 12% relatedness to ATCC 27592 and 45 f 7% relatedness to ATCC 19323. The thermal stability of hybrid DNA molecules was high within DNA hybridization groups (AT, [drop in thermal denaturation midpoint] below 4.4”C) and lower between groups (AT,, between 7.5 and 11.5”C). This experimental work will be published elsewhere (Grimont et al., manuscript in preparation). The International Code of Nomenclature of Bacteria (4) stipulates that a Request for an Opinion should be submitted to the Judicial Commission for vote after a period of 6 months after the date of publication of the Request. This 6-month time between publication of a Request for an Opinion and a decision of the Judicial Commission is too short for experimental work to be conducted and then published in the International Journal of Systematic Bacteriology, a quarterly. Because we now have evidence to prove that S. Ziquefaciens and S. proteamaculans are close but discrete genospecies, if one name or the other were to be rejected by the Judicial Commission, a new name would have to be given to replace the rejected name. This would further escalate the confusion that presently exists in the nomenclature of Serratia species. Therefore, we request that the Judicial Commission (i) conserve both names (S. Ziquefaciens and S. proteamaculans), (ii) drop any action

International audience; Background: Shigellosis or bacillary dysentery, an acute bloody diarrhoea, is a major public health burden in developing countries. In the absence of prompt and appropriate treatment, the infection is often fatal, particularly in young malnourished children. Here, we describe a new diagnostic test for rapid detection, in stool, at the bedside of patients, of Shigella flexneri 2a, the most predominant agent of the endemic form of the disease.Methodology/principal findings: The test is based on the detection of S.flexneri 2a lipopolysaccharide (LPS) using serotype 2a-specific monoclonal antibodies coupled to gold particles and displayed on one-step immunochromatographic dipstick. A concentration as low as 20 ng/ml of LPS is detected in distilled water and in reconstituted stools in under 15 minutes. The threshold of detection corresponds to a concentration of 5x10(7) CFU/ml of S. flexneri 2a, which provides an unequivocal positive reaction in three minutes in distilled water and reconstituted stools. The specificity is 100% when tested with a battery of Shigella and unrelated strains, in culture. When tested in Vietnam, on clinical samples, the specificity and sensitivity were 99.2 and 91.5%, respectively. A decrease of the sensitivity during the evaluation on stool samples was observed after five weeks at room temperature and was due to moistening of the dipsticks caused by the humidity of the air during the fifth week of the evaluation. This drawback is now overcome by improving the packaging and providing dipsticks individually wrapped in waterproof bags.Conclusion: This simple dipstick-bases test represents a powerful tool for case management and epidemiological surveys.

This work was conducted in Laboratoire de Contrôle des Eaux et Denrées Alimentaires, Institut Pasteur de Tunis, Tunisia.; International audience; Shiga toxin-producing Escherichia coli (STEC) strains were detected among 250 E. coli strains isolated from 204 food samples (meat and dairy products) obtained from butcher shops and retail stores in BEN AROUS (a Tunisian State/south east of the Tunisian capital). These 250 E. coli strains were isolated by applying culture techniques using sorbitol MacConkey (SMAC) agar after preenrichment of the homogenized food samples in buffered peptone water for 6 and 24 h, and violet red bile lactose (VRBL) agar. The highest percentage of E. coli recovery was from SMAC/6h preenrichment time and VRBL agar for meat and dairy products, respectively. Serotype O55:B5 was found to be the most prevalent type among E. coli isolates. E. coli isolates were screened for virulence factors: Shiga toxin 1 [stx1 (130 bp)], Shiga toxin 2 [stx2 (346 bp)], and eae gene [494 bp] by colony blot hybridization using three radioactively labelled DNA probes. Results obtained were confirmed by PCR. The three STEC isolates: one isolate-produced stx2, another isolate-produced stx1 and stx2, and the third isolate-expressed stx1, stx2, as well as eae genes. All three isolates produced EAST1. All Shiga toxin positive by PCR were found positive in the verocytotoxin test. In conclusion, this study indicates the presence of a possible bovine reservoir for STEC strains in Tunisia, signaling a potential risk for humans.