Previous Article | Next Article ![]()
Journal of Clinical Microbiology, August 2009, p. 2377-2380, Vol. 47, No. 8
0095-1137/09/$08.00+0 doi:10.1128/JCM.02512-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
,
Cyril Savin, and
Elisabeth Carniel*
Institut Pasteur, Yersinia Research Unit, Yersinia National Reference Laboratory, Paris, France
Received 31 December 2008/ Accepted 20 May 2009
|
|
|---|
-methyl-D-glucoside [
MG], and L-rhamnose). The collection of the French Yersinia Reference Laboratory (Institut Pasteur, Paris, France) contained 44 strains that were originally identified as Y. intermedia but whose characteristics did not fit into the biotyping scheme. These 44 strains were separated into two biochemical groups: variant 1 (positive for acid production from L-rhamnose and
MG and positive for Simmons citrate utlization) and variant 2 (positive for acid production from L-rhamnose and
MG). These atypical strains could correspond to new biotypes of Y. intermedia, to Y. frederiksenii strains having the atypical property of fermenting
MG, or to new Yersinia species. These strains did not exhibit growth or phenotypic properties different from those of Y. intermedia and Y. frederiksenii and did not harbor any of the virulence traits usually found in pathogenic species. DNA-DNA hybridizations performed between one strain each of variants 1 and 2 and the Y. intermedia and Y. frederiksenii type strains demonstrated that these variants do belong to the Y. intermedia species. We thus propose that Brenner's biotyping scheme be updated by adding two new biotypes: 9 (for variant 1) and 10 (for variant 2) to the species Y. intermedia. |
|
|---|
Y. intermedia was separated from Y. enterocolitica and defined as a new species by Brenner and colleagues in 1980 (6). Bacteria belonging to this species have been isolated from the environment (freshwater, sewage), various animals (fish, oysters, shrimps, snails, wild and domestic animals), food (milk, cream, meat), and sometimes, healthy and sick humans, mainly from their stools. This new species was named Y. intermedia because it has properties intermediate between those of Y. pseudotuberculosis and Y. enterocolitica. Indeed, this species shares with Y. pseudotuberculosis 45 to 55% DNA relatedness as well as the ability to ferment rhamnose and melibiose, but it also exhibits some of the biochemical characteristics of Y. enterocolitica (sucrose and cellobiose fermentation). Y. intermedia also shares several O antigens with Y. enterocolitica (27), of which O:4 and O:17 appear to be the prevailing serotypes. This species can further be subdivided into eight biotypes, on the basis of Simmons citrate utilization and acid production from
-methyl-D-glucoside (
MG), D-melibiose (Mel), D-raffinose (Raf), and L-rhamnose (Rha) (6).
Y. frederiksenii was also differentiated from Y. enterocolitica in 1980 (24). Y. frederiksenii and Y. intermedia have similar ecological niches, and they are phenotypically very close. Actually, the distinction between the two species relies on the simultaneous absence of acid production from Mel,
MG, and Raf in Y. frederiksenii, while at least one of these three sugars is acidified by the various biotypes of Y. intermedia (Table 1) .
|
View this table: [in a new window] |
TABLE 1. Scheme used to biotype Y. intermedia and compare the phenotypic characteristics of the two variants and Y. frederikseniia
|
MG. The objectives of this work were to analyze the main phenotypic and genetic characteristics of these atypical strains and to determine their taxonomic position.
|
|
|---|
Growth and biochemical properties.
All Yersinia strains were grown on Trypticase soy agar (TSA; Oxoid, France) and cefsulodin-irgasan-noviobiocin (CIN; Merck, Darmstadt, Germany) at 28 and 37°C. Biochemical tests were performed with API 20E and API 50CH strips (BioMérieux, Marcy l'Etoile, France), which were incubated at 28°C for 24 h and 48 h, respectively. In addition, biochemical tests for acid production from Mel, Rha,
MG, and Raf and the utilization of Simmons citrate, done for biotype determination according to the biotyping scheme of Brenner et al. (6), were performed in tubes at 28°C; and the results were read after 24 h to 96 h.
Serotyping and phage typing. O antigens were determined by slide agglutination with the 53 specific rabbit antisera against Y. enterocolitica and related species generated by the French Yersinia Reference Laboratory, according to the typing scheme of Wauters et al. (25, 27). Phage typing was carried out with a set of 12 lysogenic phages and 16 sewage phages, as described previously (15).
Antibiotic susceptibility testing. Antibiotic susceptibility was determined by the disc diffusion technique (4) on Mueller-Hinton II agar, and the results were interpreted according to the criteria of the Comité de l'Antibiogramme of the French Society of Microbiology (http://www.sfm.asso.fr). The antimicrobial drugs tested and their concentrations on the discs (Bio-Rad, Marnes La Coquette, France) were as follows: amoxicillin (amoxicilline) (25 µg), amoxicillin-clavulanic acid (20 µg/10 µg), cefalotin (30 µg), cefoxitin (30 µg), ceftriaxone (30 µg), ciprofloxacin (5 µg), nalidixic acid (30 µg), sulfonamides (200 µg), ticarcillin (75 µg), tetracycline (30 IU), and trimethoprim (5 µg).
Phenotypic virulence markers. Pyrazinamidase activity was tested at 28°C, as described previously (13). The strains were examined for the presence of the virulence plasmid by using three phenotypic tests: calcium dependency at 37°C (10), Congo red binding on Congo red acid morpholine propanesulfonic acid pigmentation agar incubated for 48 h at 25°C (16), and autoagglutination at 37°C in Trypticase soy broth (Biokar Diagnostics, Beauvais, France) (13).
Detection of virulence genes by PCR. Genomic DNA was extracted with an IsoQuick kit (Orca Research Inc., Bothell, WA). The PCR conditions were those previously reported for the detection of inv (18), ail (23), yst (12), and virF (28). Y. enterocolitica strain IP28643 (bioserotype 4/O:3), which harbors all genes for which tests were conducted, was used as a positive control. The PCR products were subjected to agarose gel electrophoresis and were visualized under UV light after the gels were stained with ethidium bromide.
DNA-DNA hybridizations. Bacteria grown for 24 h at 28°C on brain heart infusion agar were harvested and sent to the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany, for DNA-DNA hybridizations. Briefly, the bacteria were broken in a French pressure cell (Thermo Spectronic, Waltham, MA). The DNA was purified by chromatography on hydroxyapatite (8). DNA-DNA hybridizations (9, 11) were carried out with a Cary 100 Bio UV/visible spectrophotometer equipped with a Peltier-thermostated multicell changer (six by six cells) and a temperature controller with an in situ temperature probe (Varian, Palo Alto, CA).
|
|
|---|
MG, Mel, Raf, and Rha), the species Y. intermedia can be divided into eight biotypes (Table 1). We identified 44 isolates in the strain collection of the French Yersinia Reference Laboratory (see Table S1 in the supplemental material) which were originally classified as Y. intermedia but whose biochemical characteristics did not match those of any of the known biotypes. These 44 atypical strains could be separated into two groups: variant 1 (8 strains), which gave positive reactions for acid production from Rha and
MG and the utilization of Simmons citrate but not for acid production from Mel and Raf, and variant 2 (38 strains), which displayed positive reactions only for acid production from Rha and
MG (Table 1). Most of these strains were isolated from the environment, essentially from vegetables and water, in various countries between 1977 and 2005 (see Table S1 in the supplemental material). They had all the characteristics of the Enterobacteriaceae family (gram negative, oxidase negative, catalase positive, facultatively anaerobic fermentative rods that reduced nitrates to nitrites). Because they were able to ferment Rha, they could be differentiated from Y. enterocolitica sensu stricto and Y. kristensenii (5), and their sucrose-positive reaction distinguished them from Y. aldovae. These two groups of strains could be new biotypes of Y. intermedia, Y. frederiksenii strains that had acquired the ability to ferment
MG (Table 1), or new Yersinia species. To determine the taxonomic position of these two atypical groups of strains, they were further characterized.
Growth and biochemical properties of variants 1 and 2.
The two groups of strains grew at 28°C and 37°C on both TSA and CIN media. Only half of the atypical strains were motile at 28°C. Like other Yersinia species, the two variants formed colonies of 0.5 to 1 mm in diameter on TSA and CIN media after 24 h of incubation at 28 and 37°C. All variant 1 and 2 strains gave positive reactions for urea, indole, sucrose, L-arabinose, ribose, D-xylose, galactose, D-glucose, D-fructose, D-mannose, mannitol, arbutine, cellobiose, maltose, trehalose, gluconate, and nitrate reductase and negative reactions for arginine dihydrolase, lysine decarboxylase, H2S, tryptophan deaminase, gelatin, D-arabinose, L-xylose, adonitol, β-methyl-xyloside, dulcitol, lactose, melibiose, inuline, melezitose, xylitol, D-turanose, D-lyxose, D-tagatose, L-arabitol, D-fucose, and starch (data not shown). The results of these biochemical tests for the two variants were similar to those for the Y. intermedia IP3953 and Y. frederiksenii IP6175 type strains. However, not all strains of variants 1 and 2 gave a homogeneous positive or negative response for a few other tests (listed in Table S2 in the supplemental material). In most instances, the unusual biochemical character was limited to a small number of isolates (
13%). The only marked exception was D-arabitol fermentation, for which the Y. frederiksenii type strain and 63% and 58% of the variant 1 and 2 strains, respectively, were positive and for which the Y. intermedia type strain was negative (see Table S2 in the supplemental material). However, the analysis of 290 Y. intermedia strains from the French Yersinia Reference Laboratory collection indicated that only 67% of them fermented D-arabitol, a percentage similar to the percentages for variants 1 and 2. In contrast, almost all (99%) 262 Y. frederiksenii strains from the collection analyzed gave a positive D-arabitol reaction. Thus, the variability in D-arabitol fermentation is more similar to the properties of Y. intermedia than to those of Y. frederiksenii.
Of note, the Simmons citrate reaction was sometimes hardly detectable with the API 20E strip, and we recommend that this test be performed in a tube incubated at 28°C.
Serotypes and phage types of variant 1 and 2 strains. Three of the 44 atypical strains were not slide agglutinable (see Table S1 in the supplemental material). The other strains belonged to a wide variety of serotypes (27 different specificities), arguing against the clonality of these variants. The same serotypes were found in both variant 1 and variant 2 strains (see Table S2 in the supplemental material), indicating that there was no specific association between the biochemical properties and the O-antigen specificities of the two groups of strains. Conventional Y. intermedia and Y. fredericksenii strains also exhibited a wide variety of serotypes, including those found for the two variants.
All 44 strains belonged to two phage types (types Xo and Xz; see Table S1 in the supplemental material), and similar to the findings for the serotypes, there was no preferential association between one phage type and one variant. These two phage types are also the most common phage types found in conventional Y. intermedia and Y. frederiksenii strains.
Antibiotic susceptibilities of variants 1 and 2. All variant 1 and 2 strains were resistant to ticarcillin, amoxicillin, cefalotin, and amoxicillin-clavulanic acid and were susceptible to tetracycline, ciprofloxacin, sulfonamide, trimethoprim, nalidixic acid, and ceftriaxone. Moreover, the susceptibilities to cefoxitin of the variant 1 and 2 strains ranged from resistant to intermediate. These susceptibility profiles are similar to those of the type strains Y. intermedia IP3953 and Y. frederiksenii IP6175. They are also similar to the antibiotic susceptibility profiles of Y. intermedia and Y. frederiksenii reported in the literature (2, 3, 22).
Virulence traits of variants 1 and 2. Several virulence factors have been identified in pathogenic yersiniae. The virulence plasmid (pYV) plays a key role, but additional chromosomal genes, such as those mediating cell invasion (inv), resistance to complement (ail), or the production of an enterotoxin (yst), are also required. Furthermore, strains possessing the high-pathogenicity island (irp2), which encodes the siderophore yersiniabactin, are highly pathogenic (7). All 44 strains exhibited negative results by phenotypic tests for the presence of the virulence plasmid (data not shown). In addition, none of the strains harbored the pYV-borne virF gene or the chromosomal virulence genes inv, ail, irp2, and yst. Furthermore, while pathogenic Yersinia almost systematically lack pyrazinamidase activity (13), all variant 1 and 2 strains were pyrazinamidase positive. These results indicate that, as for conventional Y. intermedia and Y. frederiksenii strains, variants 1 and 2 do not possess the virulence genes usually found in pathogenic Yersinia species. This is consistent with the fact that all except one of the isolates were isolated from the environment (see Table S1 in the supplemental material). The only strain of human origin was isolated from a diarrheal patient in Nigeria (1). However, its presence in stools does not prove its virulence capacity but may simply reflect a transient and asymptomatic carriage of a saprophytic microorganism in the human intestinal tract.
Genetic relatedness between the two variants, Y. intermedia, and Y. frederiksenii.
To determine whether strains of variants 1 and 2 belong to the Y. intermedia or Y. fredericksenii species or to a new species, DNA-DNA hybridization reactions between one representative each of variants 1 and 2 and the type strains Y. frederiksenii IP6175 and Y. intermedia IP3953 were performed. As shown in Table 2, variants 1 and 2 were genetically closely related and were also closely related to the Y. intermedia type strain. Therefore, variant 1 and 2 strains do belong to the species Y. intermedia. Since they are members of this species but have biochemical characteristics that do not match those of the eight defined biotypes, we propose that two new biotypes be added to the Y. intermedia biotyping scheme: biotype 9 (positive for acid production from Rha and
MG and the utilization of Simmons citrate and negative for acid production from Mel and Raf), which corresponds to the variant 1 strains, and biotype 10 (positive for acid production from Rha and
MG and negative for the utilization of Simmons citrate and acid production from Mel and Raf), which corresponds to the variant 2 isolates (Table 1).
|
View this table: [in a new window] |
TABLE 2. Percentage of DNA-DNA relatednessa between one representative each of variant 1 and variant 2 and the Y. intermedia and Y. frederiksenii type strains
|
This study was funded in part by the Institut de Veille Sanitaire (InVS).
Published ahead of print on 3 June 2009. ![]()
Supplemental material for this article may be found at http://jcm.asm.org/. ![]()
Present address: Institut Pasteur, CNR Listeria, 25-28 rue du Docteur Roux, Paris 75724 cedex 15, France. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2010 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»