WO2008046664A1

WO2008046664A1 - Method and device for the selective isolation of serotypes of shiga toxin producing e. coli

Info

Publication number: WO2008046664A1
Application number: PCT/EP2007/009173
Authority: WO
Inventors: Bjorn Posse; Lieven De Zutter; Lieve Herman; Marc Heyndrickx
Original assignee: Universiteit Gent; Instituut Voor Landbouw- En Visserijonderzoek
Priority date: 2006-10-19
Filing date: 2007-10-19
Publication date: 2008-04-24
Also published as: GB0620787D0

Abstract

This invention provides a method for selectively identifying pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes on two or more plating media, comprising the step of determining the STEC serotype based on at least 4 different STEC serotype specific colours formed onto at least one of said plating media. The invention further provides a plating medium for the selective identification of pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes comprising a combination of at least one chromogenic marker and at least two metabolic markers, wherein said combination allows for at least three different STEC serotype-specific colours.

Description

METHOD AND DEVICE FOR THE SELECTIVE ISOLATION OF SEROTYPES OF SHIGA TOXIN PRODUCING E . COLI

The present invention relates to the field of identification and isolation of serotypes. In particular the invention relates to methods and devices, i.e. plating media, for selectively identifying and isolating pathogenic Shigatoxin producing Escherichia coli serotypes.

Background of the invention Escherichia coli (E. coli) is widely recognized as an enteric commensal, but the species also includes a significant pathogenic group frequently associated with severe human infection, i.e. Shigatoxin producing E. coli (hereinafter referred as STEC). Shiga toxigenic Escherichia coli (STEC) is an important cause of gastrointestinal disease in humans, particularly since such infections may result in life-threatening sequelae such as hemolytic-uremic syndrome (HUS) and thrombotic thrombocytopenic purpura. The morbidity and mortality associated with several recent large outbreaks of STEC disease have highlighted threats that these organisms pose to public health.

Although E. coli 0157 is currently most widely recognised, more than 100 other serotypes have been implicated in cases of human disease. Among these other serotypes, 026, 0103, 0111 and 0145 have been increasingly frequently isolated from clinical cases. Only few laboratories test for these non-0157 STEC due to the lack of a specific standardised method or the need for complex and expensive equipment. Non-0157 STEC strains are a heterogeneous group which display a broad range of both genotypic and phenotypic differences.

The 0157 serotype has some unique phenotypic features, easing its isolation from food and faeces: strains belonging to this serotype fail to ferment sorbitol and are referred to as sorbitol negative 0157 E. coli. More recently however, sorbitol positive strains have been increasingly isolated from clinical cases, stressing the importance of screening food products for both sorbitol positive and negative 0157. Sorbitol MacConkey agar (hereinafter referred as SMAC) has been developed for isolating sorbitol negative 0157 strains. However, non-0157 serotypes and sorbitol positive 0157 strains cannot be isolated using this method. Modifications of SMAC have also been published and mainly enhance selectivity for sorbitol negative 0157 E. coli by using inhibitory components and/or antibiotics: CT-SMAC wherein potassium tellurite and cefixime are added to the SMAC medium, and CR-SMAC wherein cefixime and rhamnose are added to the SMAC medium to improve 0157 isolation efficiency. Previously reported methods describing non-0157 STEC isolation media use one specific carbohydrate to indicate the presence of a certain serotype (e.g. RMAC and SMAC). Such isolation media have proven to lack specificity when applied to a larger collection of strains because of the limited metabolic variation comprised within the method. Several other isolation media have been developed using different mixtures of chromogenic compounds in order to enhance contrast, but most of them still focus on 0157 isolation. The use of a mixture of chromogenic compounds dramatically increases the cost per plate when using such media: e.g. TBX, PTX, Rainbowagar, Chromagar, and so on. Isolation media making use of enterohaemolytic properties shared by most

STEC to facilitate their isolation were also described (EHEC-agar, modified EHEC agar).

Most available isolation media are designed to isolate sorbitol negative 0157 STEC strains. However, many other serotypes and sorbitol positive 0157 strains have been identified for inflicting disease in humans. Currently, no standard isolation medium is available to selectively isolate these severe pathogenic strains from food products or other samples such as faeces, blood or urine due to the lack of specificity of the analysing methods currently available. A culture medium and a method for the differentiation and identification of different species belonging to the family Enterobacteriaceae are known in the art, wherein said culture medium comprises a chromogenic beta- galactosidase substrate in combination with a member of the group consisting of (a) a decarboxylase substrate, (b) a deaminase substrate, (c) a urease substrate, (d) a hydrogen sulfide detecting system, or (e) a carbohydrate fermentation system.

A method for the isolation and detection of sorbitol negative 0157 STEC using a plating medium comprising one carbohydrate, a pH indicator and X- gal chromogen is also known in the art.

There is an increasing demand for improved and more cost-effective diagnostic procedures for the selective identification (detection) and selective isolation of different STEC serotypes in patient samples, e.g. faeces, blood or urine and, in particular, in food such as meat and dairy products. It has been recognised for a number of years that STEC strains causing human disease may belong to a very broad range of pathogenic O sero-groups. Hence, there is a need in the art for a method that allows laboratories to routinely perform a fast and relatively easy identification method, within a small budget, in order to determine whether a patient is infected, or food is contaminated, with STEC belonging to anyone of the pathogenic sero-groups 026, 0103, 0111 , 0145 or 0157.

Summary of the invention An object of the present invention is to provide an improved method for the selective identification and, if necessary, isolation of pathogenic Shigatoxin producing Escherichia coli serotypes, in particular E. coli serotypes 026, 0103, 0111 , 0145 and sorbitol negative, as well as sorbitol positive 0157. The improved identification method of this invention uses a combination of two or more specifically designed plating media and achieves a highly reliable detection result in particular by reducing the percentage of false positive determinations compared to other methods known in the art.

It is a first advantage of the present invention to provide a method for the selective identification and, if necessary, isolation of STEC, in particular E. coli serotypes 026, 0103, 0111 , 0145, and sorbitol negative 0157 as well as sorbitol positive 0157, that allows for an improved and highly reliable identification and, if necessary, isolation (and purification) of two or more of said STEC substantially at the same time and on the same plating medium. It is a further advantage of the present invention that the novel identification and, if necessary, isolation method is highly selective for each of said serotypes against other STEC serotypes as well as other microbiological organisms, in particular commensal E. coli serotypes. It is a further economical advantage of the present invention that the novel selective identification and, if necessary, isolation method achieves a reliable result within a significantly short time, and at a significantly reduced cost, especially within a shorter time and/or at a lower cost than alternative methods known in the art. Additionally the present invention provides a specifically designed plating medium, as well as a combination of plating media, which are economical in construction and are suitable for use in performing the above-referred method for the selective identification and, if necessary, isolation of pathogenic Shigatoxin producing Escherichia coli serotypes.

Definitions

As used herein, and unless otherwise stated, the terms "plating medium " or " medium " refers to a nutrient system for the artificial cultivation of cells or organisms and especially bacteria. This system may be a simple substance but is more commonly a complex of inorganic and organic materials in a base that may be fluid or rendered more or less solid by coagulation or by the addition of gelling agents such as but not limited to gelatin or agar.

For the purpose of the present invention, and unless otherwise stated, the terms " isolation medium " and " discriminating medium " interchangeably refer to a plating medium that by the addition of specifically selected nutrients allows for a selective identification and, if necessary, isolation of certain microbiological species.

As used herein, and unless otherwise stated, the term "marker " refers to a distinctive indicator of a process, event, or condition. As used herein, and unless otherwise stated, the term "chromogenic marker " refers to forming a distinctly coloured product facilitating discrimination amongst other bacteria. For instance it can refer to non- inducing substrates for specific enzymes being present or suspected to be present within the group of bacteria to be isolated. When the target specific enzyme is indeed present, the component is hydrolysed.

As used herein, and unless otherwise stated, the term " metabolic marker " refers to specific nutrients which may or may not be metabolised by the targeted species and therefore allow for discrimination between metabolizers and non-metabolizers.

Brief description of the drawing

The single figure schematically shows an illustrative embodiment of the improved method of identification and, if necessary, isolation according to the present invention.

Detailed description of the invention

It is a first aspect of the present invention to provide an improved method for selectively identifying and, if necessary, isolating pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes on a combination of two or more plating media. More specifically herein, STEC serotypes are determined on the basis of at least 4 different STEC serotype specific colours formed on at least one of said plating media. E. coli serotypes that can be identified by the method of the present invention comprise, but are not limited to, STEC 026, 0103, 0111 , 0145 and 0157, more particularly wherein said 0157 STEC may be sorbitol negative or may be sorbitol positive. The improved method of the present invention allows for the rapid and efficient discrimination of STEC serotypes from other commensal E. coli serotypes and for the rapid and efficient discrimination of the different STEC serotypes against each other. Samples that need analysing for STEC and are therefore candidates for the practice of the present invention comprise, but are not limited to, food samples, preferably prior to food intake to prevent infection of the mammal for which the food was intended, and samples taken from a mammal, in particular a human being, a livestock animal, a pet animal or a captive animal suffering from infection such as, but not limited to, faeces, urine, blood or saliva. In a suitable but non-limiting embodiment of the method of the present invention, said at least two plating media may be used in consecutive steps, i.e. one or more plating media may be used after a first plating medium. In this way, the incidence of false positive results on a first discriminating medium is reduced by consecutive use of a confirmation medium. Such an embodiment of the present invention allows for a more appropriate choice of a second plating medium, based on the detection result obtained from the first plating medium in the previous step of the method. This particular embodiment allows for a substantial reduction in the occurrence of false positive results, i.e. achieves an improved accuracy and reliability of the method of the present invention, as compared to previously known methods. A first selective identification and, if necessary, isolation of STEC serotypes is thus achieved on a first plating medium and then, based on this resulting preliminary STEC serotype determination, the content of at least one further plating medium for confirmation and, if necessary, further purification of the STEC serotype is defined. Said preferred embodiment is illustrated by the global scheme in the single attached figure schematically representing a selective purification of STEC serotypes according to the present invention.

For the purpose of the present invention, the first selective identification and, if necessary, isolation step of the method may be preceded by enrichment of the E. coli strains present in the sample being analysed, using one or more techniques well known in the art.

Different types of E. coli may differ in their general appearance or morphology and also differ in their metabolic activity, on the basis of which different E. coli can be selectively or differentially identified and, if necessary, isolated and/or purified according to the improved method of the present invention. All STEC strains usually have a similar morphology. Colonies are round and have a solid edge. Commensal E. coli strains can have similar morphological traits, although many commensal strains have a branched morphology. For the purpose of the present invention, differences in metabolic activity can be demonstrated and evidenced in a cost-effective, reliable and accurate manner by the use of certain types of markers allowing for a suitable assessment of whether or not said metabolic activity is present in the sample to be analysed. Metabolic activity, as used herein, broadly refers to any activity pertaining to metabolism, i.e. the sum of all the physical and chemical processes by which a living organism is produced and maintained and also the transformation by which energy is made available for the uses of this living organism. Within the above broad concept, metabolic activity mostly refers, according to a preferred embodiment of this invention, to enzymatic activity.

In one embodiment, markers useful for the performance of the identification method of the present invention may be chromogenic markers such as, but not limited to, 5-bromo-4-chloro-3-indolyl-beta-D- galactopyranoside (hereinafter referred as X-gal). X-gal is typically used in combination with isopropyl-beta-D-thiogalacto-pyranoside (hereinafter referred to as IPTG) that induces the beta-galactosidase activity. In another embodiment, markers useful for the performance of the improved identification method of the present invention may also be metabolic markers such as, but not limited to, certain carbohydrates. Different E. coli serotypes are known to ferment different carbohydrates. Suitable carbohydrates for the purpose of carrying out the identification method of the present invention may be selected from the group consisting of, but are not limited to, sorbitol, sucrose, sorbose, L-rhamnose, dulcitol, D-arabinose, and D-raffinose. Carbohydrate fermentation usually results in acidification of the medium, thereby inducing a pH-indicator comprised within said medium to change colour after a significant pH decrease has occurred. Suitable pH-indicators for detecting carbohydrate fermentation, in particular when one of the above-referred carbohydrates is involved, are well known in the art and may be, but are not limited to, neutral red or phenol red.

It is another aspect of the present invention to provide a combination of suitable markers in a plating medium that allows for the formation of different colours indicating the presence of different STEC serotypes, in particular a combination of one or more chromogenic markers with one or more, preferably two metabolic markers.

In a first suitable but non-limiting embodiment of this aspect of the present invention, at least one chromogenic marker is combined with at least two carbohydrates. In this embodiment of the present invention, due to the combination of at least one chromogenic marker with at least two carbohydrates, at least 4 different colours may be formed on the at least two plating media. In another suitable but non-limiting embodiment of the present invention, at least 3 different STEC serotype specific colours may be formed on a plating medium at a first stage of serotype isolation. A selective isolation medium, e.g. a suitable medium for the selective identification and, if necessary, isolation of STEC serotypes such as, but not limited to, 026, 0103, O111 and 0145, according to this invention can exhibit or evidence up to 7 different colour shades (6 coloured and translucent colony types) based on one chromogenic marker for beta-galactosidase activity and carbohydrate fermentation. Since all STEC are capable of hydrolysing the chromogenic marker X-gal, all STEC colonies display a dark coloured centre. Blending the colours originating from X-gal hydrolyzation and carbohydrate fermentation may result into up to 7 different colours on said selective identification and, if necessary, isolation medium after incubation under suitable conditions (incubation time and temperature, in particular). For instance, E. coli 026 strains provide a typical purple colour after suitable incubation, due to fermentation of both sucrose and sorbose. E. coli 0103 and 0111 strains can easily be recognised by their blue colour, strains belonging to these serotypes fail to ferment sorbose while sucrose can be fermented. E. coli 0145 strains which are not able to ferment either sucrose or sorbose, therefore only show the dark centre indicating beta-D-galactosidase activity, these colonies show a typical green colour.

It is another aspect of the invention to provide a plating discriminating medium being especially useful for the selective identification and, if necessary, isolation of pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes. According to one embodiment of the invention, this plating medium comprises a combination of at least one chromogenic marker and at least two metabolic markers, wherein said combination allows for at least 3 different STEC serotype-specific colours. A plating medium for the selective identification and, if necessary, isolation of STEC serotypes such as, but not limited to, 026, 0103, 0111 and 0145 may display up to 7 different colour shades or more. In one embodiment of the plating medium of the present invention, the at least one chromogenic marker may be 5-bromo-4-chloro-3- indolyl-beta-D-galactopyranoside. In another embodiment of the plating discriminating medium of the present invention, the metabolic markers may be carbohydrates and, for instance, may be suitably selected from the group consisting of, but are not limited to, sorbitol, sucrose, sorbose, L-rhamnose, dulcitol, D-arabinose, and D-raffinose.

It is within the basic knowledge of the person skilled in the art to determine, starting from the essential components described herein-before, the full detailed composition of a plating discriminating medium according to the present invention. Said plating discriminating medium may for instance comprise, in addition to the components stated herein-above, a medium base which may be selected from the group consisting of, but is not limited to, MacConkey agar base, broth base or agar. The medium base may be further supplemented with a range of nutriments and/or inhibitory agents, as is standard in the art. Suitable inhibitory agents may be selected from the group consisting of, but are not limited to, antibiotics such as novobiocine, bile salts or potassium tellurite. For the purpose of the present invention, the pH of the plating discriminating medium should preferably be within the range of about 7.0 to about 7.8, more preferably within the range of about 7.2 to about 7.6, most preferably about 7.4.

It is within the knowledge of the person skilled in the art to determine the specific incubation conditions (such as, but not limited to, incubation time and incubation temperature) suitable for performing the method of the present invention or for using the plating discriminating media of the present invention for selectively identifying and, if necessary, isolating STEC serotypes. In a suitable but non-limiting embodiment of the present invention, a plating discriminating medium described herein may be incubated at most about 48 hours, preferably no longer than about 36 hours, more preferably about 24 hours. In another suitable but non-limiting embodiment of the present invention, a plating discriminating medium described herein may be incubated at least about 12 hours, preferably at least about 18 hours, more preferably about 24 hours. All incubation times specified herein are normally suitable for achieving an accurate and selective determination of STEC serotypes. The following examples are provided as an illustration, without any limiting intention, of certain embodiments of the identification method and the plating media according to the present invention.

Example 1 - selective isolation of STEC 026, 0103, 0111 and 0145

A total of 127 STEC isolates were obtained of which 85 strains belonging to several STEC serotypes were isolated from hospitalised patients diagnosed with (bloody) diarrhoea, haemorrhagic colitis (HC) or haemolytic uremic syndrome (HUS) according to the following numerical distribution: 15 (026), 14 (0103), 15 (0111 ), 15 (0145), 11 (0157 sorbitol negative) and 15 (0157 sorbitol positive) strains.

42 commensal E. coli strains were isolated from beef carcasses. Commensal strains were isolated using a rapid E. coli medium (REC, commercially available from Bio-Rad) and were considered as commensal strains since all tested PCR negative for both vt1 and vt2 genes.

Metabolic characteristics, in terms of carbohydrate fermentation profiles, of the E. coli strains 026, 0103, 0111 , 0145 and 0157, used in this and the following examples, were examined using API50 CH tests and CHB/E medium (both commercially available from Bio-Merieux, France). The following carbohydrates were evaluated in these tests: Glycerol, Erythrol, D-Arabinose, L-Arabinose, D-Ribose, D-Xylose, L-Xylose, D-Adonitol, Methyl-D- xylopyranoside, D-Galactose, D-Glucose, D-Fructose, D-Mannose, L- Sorbose, L-Rhamnose, Dulcitol, Inositol, D-Mannitol, D-Sorbitol, Methyl-D- mannopyranoside, Methyl-D-glucopyranoside, N-Acetylglucosamine, Amygdalin, Arbutin, Esculin ferric citrate, Salicin, D-Cellobiose, D-Maltose, D- Lactose, D-Melibiose, D-Saccharose, D-Trehalose, Inulin, D-Melezitose, D- Raffinose, Amidon (starch), Glycogen, Xylitol, Gentiobiose, D-Turanose, D- Lyxose, D-Tagatose, D-Fucose, L-Fucose, D-Arabitol, L-Arabitol, K- Gluconate, K-2-Ketogluconate and K-5-Ketogluconate. The testing tubes were inoculated with pure culture suspension, grown overnight in TSB at 37°C, and sealed using mineral oil according to the manufacturer's instructions. Tests were incubated for 24 hours or 48 hours at 37°C, depending upon the specific carbohydrate, and then carbohydrate fermentation was evaluated upon pH indicator (phenol red) colour change. Fermentation scores for each carbohydrate and each strain were analysed and the results were processed using Ward cluster analysis wherein the similarity coefficient was set to Cosine correlation. The metabolic profile of the pathogenic E. coli strains for the most important carbohydrates, i.e. differentiating carbohydrates, for the purpose of the present invention are summarized in Table 1. All carbohydrates mentioned in table 1 were fermented within 24 hours, except D-Arabinose (48 hours).

Table 1

026 O103 0111 0145 0157 sorb- 0157 sorb +

D-Arabinose 66% - - + + - -

L-Sorbose + - - - 55% + +

L-Rhamnose - + + + + -

Dulcitol - - + 86% - + +

D-Sorbitol + + + + - +

D-Saccharose + + + - + +

D-Raffinose + + + - + +

Enrichment of strains

Prior to being used for evaluation of selective isolation media, E. coli strains were maintained on tryptic soy agar. E. coli strains were propagated by aerobic growth at 37⁰C in modified tryptic soy broth, modifications (g/L) being as follows: peptone from casein 17.0; peptone from soy meal 3.0; sodium chloride 5.0; di-basic potassium phosphate 2.5; glucose 2.5; novobiocine 0.008; rifampicine 0.002; vancomycin 0.016; bile salts 1.5; potassium tellurite 0.001. Suitable concentrations of selective agents were defined after a minimal inhibitory concentration (MIC) study for pathogenic STEC serotypes. MIC values (expressed in mg/L unless otherwise stated) resulting from this preliminary study are listed in Table 2. Table 2

Serotype K-tellurite Novobiocine Rifampicine Vancomycine bile salts

(g/L)

026 >256 256-128 16 >256 > 40

0103 >256 256-128 16-8 >256 > 40

0111 >256 128-64 8 >256 > 40

0145 >256 256-128 16-8 >256 > 40

0157 256-64 256 32-8 >256 > 40

Subsequently, two different media, i.e. a discriminating medium and a confirmation medium, were used consecutively to respectively identify (and, if necessary, isolate) and confirm strains belonging to E. coli serotypes 026, 0103, 0111 and 0145.

Selective identification and isolation The composition and pH of the discriminating medium used for 026,

0103, 0111 and 0145 STEC are presented in Table 3.

Table 3

Components g/L

MacConkey agar base 40 potassium tellurite 0.0025 sucrose 6 sorbose 6

5-bromo-4-chloro-3-indolyl-beta-D-galactopyranoside 0.05 isopropyl-beta-D-thiogalactopyranoside 0.05

Novobiocine 0.008 bile salts 3.5 final pH at 25°C 7.4 MacConkey agar base (commercially available from BD Biosciences under the trade reference 281810) comprises (g/L): peptone 17.0; proteose peptone 3.0; bile salts 1.5; sodium chloride 5.0; agar 13.5; neutral red 0.03; and crystal violet 0.001. Since MacConkey medium already contains 1.5 g/L of bile salts, 3.5 g/L bile salts were added, as indicated in table 3, in order to obtain a final bile salts concentration of 5 g/L in the discriminating (and isolation) medium. All components added to the MacConkey agar base, except 5-bromo-4- chloro-3-indolyl beta-D-galactopyranoside (X-gal), isopropyl-beta-D- thiogalactopyranoside (IPTG)₁ novobiocine and bile salts, were suspended in distilled water and autoclaved for 15 minutes at 121⁰C. All other components were filter sterilised and added to the medium after autoclaving and cooling to 5O⁰C.

Inoculated selective plates were placed in an incubator at 37°C for 24 hours. Then, suspected colonies were identified based on their general appearance and colour as follows:

- the 0145 STEC strain, due to its inability to ferment sucrose and sorbose, results in a colony with light green morphology;

- the O103 and 0111 STEC strains ferment one of these carbohydrates, resulting in a dark orange colour of the medium, and additional X-gal hydrolyzation colours the colonies themselves into blue-purple; and

- 026 STEC strains ferment both sucrose and sorbose, thus resulting in a pH shift that changes their colour from pale yellow into dark red, which change in turn is enhanced by X-gal hydrolyzation turning the colonies into a bright purple to red colour.

Confirmation medium

Suspected colonies were picked from the discriminating (and isolation) medium and streaked onto a second elective medium for confirmation and strain purification. Compositions, pH and incubation conditions of this confirmation (and purification) medium are shown in Table 4 below. Table 4 components g/L incubation

Phenol red broth base 20 agar 15

L-rhamnose 6 24 hours at 37 ⁰C

Dulcitol 6 24 hours at 37 ⁰C

D-arabinose 10 24 hours at 37 ⁰C

D-raffinose 6 (anaerobic)

24 hours at 37 ⁰C final pH at 25 ⁰C 7.4

Each medium comprises a phenol red broth base and specific carbohydrates. Phenol red Broth base (commercially available from Sigma under the reference P8976) has the following composition (g/L): proteose peptone 10.0; beef extract 1.0; sodium chloride 5.0; phenol red 0.018. The carbohydrate composition of this confirmation medium is dependent upon the serotype aimed in the test.

As indicated in Table 5 below: - suspected 026 strains are confirmed on a L-rhamose or dulcitol comprising medium whereby neither carbohydrate is fermented,

- suspected 0103 strains are confirmed on a dulcitol or D-arabinose comprising medium whereby neither carbohydrate is fermented,

- suspected 0111 strains are confirmed on a dulcitol or D-arabinose confirmation medium whereby both carbohydrates are fermented, and

- suspected 0145 strains are confirmed on a D-raffinose or D-arabinose comprising medium whereby D-arabinose is and D-raffinose is not fermented.

Calculation of false positive ratios (FP %) Carbohydrate fermentation properties of the tested strains were determined using API50 test strips (commercially available from Bio-Merieux, France). False positive ratios (FP %) were determined by calculating chance rates for isolating commensal strains expressing the exact same carbohydrate fermentation pattern for certain carbohydrates as the STEC strains - all data used for these calculations were generated by commensal and STEC strains present in the collection described herein-above. FP % (I) indicates the false positive ratio when using the first discriminating (and isolation) medium only. FP % (C) indicates the false positive ratio using both the discriminating and confirmation media.

Table 5

Table 5 thus demonstrates that the additional use of a confirmation medium in the method of this invention provides the advantage of significantly reducing the false positive ratio with respect to a method performed with a discriminating medium only.

Example 2 - Selective identification and isolation of sorbitol positive and sorbitol negative Q157 strains Isolation of 85 strains from a total of 127 STEC isolates, and enrichment of the strains were performed as in example 1.

Subsequently two different media, i.e. a discriminating medium and a confirmation medium, were used consecutively to identify, isolate and confirm strains belonging to sorbitol positive and sorbitol negative 0157 STEC, according to the following procedure.

Selective identification and isolation The composition and pH of the discriminating medium used for identifying and isolating 0157 strains, both sorbitol negative and sorbitol positive, is presented in table 6 below.

Table 6

Component g/L

MacConkey agar base 40 sorbitol 10

X-gal 0.05

IPTG 0.05

Novobiocine 0.016 bile salts 3.5 final pH at 25°C 7.4

Selective plates were placed in an incubator at 37⁰C for 24 hours. After incubation green colonies indicate sorbitol negative colonies while purple colonies indicate sorbitol positive colonies.

Confirmation In order to decrease the incidence of false positive results, suspected colonies were streaked on a second elective medium for strain confirmation and purification, containing L-rhamnose (6 g/L) and plates were incubated for 24 hours at 37° C.

Calculation of false positive ratios

Carbohydrate fermentation properties of the tested strains were determined using API50 test strips and are shown in Table 7 below. False positive ratios (hereinafter referred as FP %) were determined as described for example 1 - all data used for these calculations were generated by commensal and STEC strains present in the collection described herein- above.

Table 7

Table 7 thus demonstrates that the additional use of a confirmation medium in the method of this invention provides the advantage of significantly reducing the false positive ratio for sorbitol positive 0157 strains with respect to a method performed with a discriminating medium only.

Example 3 - Evaluation of identification selectivity

Discriminating media described for examples 1 and 2 were inoculated with different mixtures of STEC and commensal strains. Ten randomly selected commensal strains and one randomly selected strain from each STEC serotype mentioned in example 1 were grown overnight in TSB. Cultures were diluted to 10⁴ and different mixtures of strains were prepared as shown in Table 8. 100 μl of these mixtures were used to inoculate isolation media and 100 μl were plated onto TSA and incubated at 37 ⁰C for 24 hours for CFU calculation. After incubation for 24 hours at 37 ⁰C, suspected colonies were transferred from the discriminating (identification and isolation) media to the confirmation media and incubated for 24 or 48 hours at 37 ⁰C prior to PCR confirmation.

Genes targeted using PCR were verotoxin encoding genes (stx1, stx2, stx2c, stx2d, stx2e and stx2f), intimin encoding genes (eaeA, eae- alpha, eae- beta, eae-gamma, eae-theta, eae-epsilon and eae-zeta), shiga-toxin auto agglutinating adhesin gene saa, enterohaemolysin gene NyA, translocated intimin receptor gene tir, katalase antiperoxidase gene katP, and extra cellular serine protease espP. From all mixtures indicated in Table 8, the STEC strain(s) that were added could be isolated using discriminating and confirmation media as described in example 1.

Table 8

Mixture commensal mixture (CFU) STEC strain (CFU)

M1 10 strains (10² each) 026, 0103, 0111 , 0145 (10² each)

M2 10 strains (10² each) 026 (10²)

M3 10 strains (10² each) 0103 (55)

M4 10 strains (10² each) 0111 (55)

M5 10 strains (10² each) 0145 (71 )

Example 4 - identification of STEC serotypes in artificially contaminated food products

A range of discriminating and confirmation plating media for a selection of non-0157 Shigatoxin producing E. coli serotypes (026. 0103. 0111. 0145), sorbitol negative 0157 and sorbitol positive 0157 was evaluated by using artificially contaminated dairy products (raw milk, cheese made from raw milk and cheese made from pasteurised milk) and meat products (ground beef, salami sausage). Naturally occurring background bacteria present in the samples prior to contamination were reduced by applying a double enrichment procedure comprising a 6 hours pre-enrichment at 37 ⁰C and an 18 hours selective enrichment at 42° C. For inoculated raw milk samples and inoculated cheese made from pasteurized milk, isolation efficiency was 100 % and 83.3 % respectively for an inoculum size of ≤ 50 CFU / 25 g. An inoculum size of ≤ 25 CFU / 25 g resulted in an isolation efficiency of 86.4 % for cheese made from raw milk, 63.3 % for ground beef and 55.5 % for salami sausage. As shown below, the consecutive use of discriminating and confirmation media limited the incidence of false positive isolates to 0.0 % for raw milk, salami sausage and cheese made from pasteurised milk, 2.1 % for cheese made from raw milk, and 8.9 % for ground beef. Sample preparation

The selective identification and isolation method of the invention for non-0157 serotypes, sorbitol negative 0157 and sorbitol positive 0157 was evaluated on the following samples: raw milk (RM), cheese made from pasteurized milk (CPM), cheese made from raw milk (CRM)₁ salami sausage (SS), and ground beef (GB). All samples were stored at 4°C and were treated within 24 hours. For all experiments, one or more 25 g (or 25 ml for raw milk samples) sub-sample was taken and diluted to a 1/10 ratio using Tryptone Soy Broth (TSB from Oxoid Ltd., London, United Kingdom) (37°C) as described below. Diluted samples were homogenised for 2 minutes prior to artificial contamination.

From each sample, total count was determined by plating 100 Fl of serial dilutions (10^"1 to 10^{' 6}) on Tryptone Soy Agar (TSA from Oxoid Ltd., London, United Kingdom) and incubation for 24 hours at 37°C; coliform counts were determined using violet red bile agar (VRB from Oxoid Ltd., London, United Kingdom) by using the same dilution and incubation conditions. Artificial contamination

All STEC strains used for artificial contamination have been described in the previous examples. Strains were stored at -8O⁰C using cryovials (from Pro-Lab Microbank, Ontario, Canada) according to the manufacturer's instructions. Strains were cultured on TSA at 37°C for 24 hours. One colony was transferred into TSB and incubated at 37°C for 24 hours to reach a stationary phase culture (109 CFU/ml). Ten-fold serial dilutions were made in Buffered Pepton Water (BPW from Oxoid Ltd., London, United Kingdom) and an appropriate volume of the diluted bacterial culture was added to the homogenized sample. Hundred Fl of the diluted bacterial culture was also streaked onto duplicate TSA plates and incubated for 24 hours at 37°C for total count. Only samples contaminated with ≥ 1 CFU per 25 g of sample were taken into account. From each sample, an uncontaminated sub-sample was also evaluated. Selective Enrichment

After artificial contamination of the 1/10 diluted and homogenised sample. 8 mg/l novobiocine and 16 mg/l vancomycin were added to the enrichment medium. After 6 hours of pre-enrichment at 37°C, 2 mg/l rifampicine, 1.5 g/l bile salts and 1.0 mg/l potassium tellurite were added to the enrichment medium and incubated further for 18 hours at 42°C (selective enrichment). Selective identification and confirmation

After the two sample enrichment stages (pre-enrichment and selective enrichment), 100 Fl of a decimal dilution (10^~3 to 10^'5) of the enriched sample was plated onto the selective discriminating medium for non-0157 STEC or 0157 STEC described in the previous examples. On this medium, 026 strains produced purple colonies. O103 and 0111 strains produced bluish coloured colonies and 0145 strains produced green colonies. All colonies showed dark coloured centres as a result of X-gal hydrolyzation. The selective differential medium for 0157 strains produced two suspected colours:

- purple colonies indicated 0157 sorbitol positive strains, and

- bluish-green colonies indicate 0157 sorbitol negative strains. Inoculated spread plates were incubated for 24 hours at 37°C. All colonies present on the differential medium were evaluated: suspected colonies were identified based on colour and morphological features and were transferred to one or more confirmation media according to the previous examples, except for 0157 sorbitol positive strains.

Thirty one experiments made use of serotype specific immunomagnetic beads (from Dynal, Oslo, Norway) according to the manufacturer's instructions. Briefly, one millilitre of each enriched homogenate was added to 20 Fl of E. coli 026. O103. 0111. 0145 or 0157 specific beads depending on the serotype targeted during the experiment. Suspension was mixed for 10 minutes and placed onto an immunomagnetic separator (IMS from Dynal, Oslo, Norway) prior to removing the supernatant. The beads were washed three times using 1 ml of 10 mM PBS containing 0.05 % (vol/vol) Tween 20 (PBS-T from Sigma-Aldrich, Poole, United Kingdom) and re-suspended in 100 Fl PBS-T. This suspension was then plated onto the selective identification medium and incubated for 24 hours at 37 ⁰C.

Confirmation and identification

A number of colonies from the selective discriminating media and the confirmation media were subsequently confirmed using conventional PCR and RAPD analysis. Colonies expressing suspected morphology and colour on both discriminating and confirmation media were confirmed as the strain used for artificial contamination by PCR-based Shigatoxin screening. In case of doubt, e.g. colonies with suspected phenotype on both the discriminating and confirmation media which appeared Sfx-negative, an additional RAPD analysis was performed.

Colonies which could not be confirmed, both suspected and non- suspected, were identified using API20 tests (from Biomerieux. France) according to the manufacturer's instructions. Results Dairy samples

The total count and the amount of conforms present in dairy products varied significantly between sample types. No conforms were present in raw milk samples except for one batch of samples with 1.0*10⁴ CFU conforms, while total count was between 1.60*10 ⁴ and 4.60^*10⁷ (batch of samples stored at 4°C for 10 days). Coliform counts between < 10.0 and 2.20^*10⁶ and total counts between 2.05*10⁶ and 2.30^*10⁸ were found for cheese made from pasteurized milk. Coliform counts between 9.60^*10⁶ and 2.54^*10⁷ and total counts between 1.85*10⁷ and 5.82*10⁷ were found for cheese made from raw milk.

Overall, 24 samples of raw milk (RM) were artificially contaminated with single strains belonging to different serotypes. All 24 experiments resulted in PCR confirmed isolates. All samples inoculated with high levels (≥ 125 CFU /

25 ml) or with low levels (< 125 CFU / 25 ml) yielded confirmed isolates. Only

1 out of 95 suspected isolates (1.0 %) on the discriminating medium could not be confirmed by PCR as the pathogenic strain added to the sample; subsequent evaluation of this strain on the confirmation media identified this strain as non-suspected, resulting in a FP (D+C) of 0.0 %. API20E test indicated this isolate was a non-pathogenic E. coli. 50 samples of cheese made from pasteurised milk (CPM) were artificially contaminated with single strains belonging to different serotypes. Overall success rate was 82 %: 100 % PCR confirmed experiments were obtained after artificial sample contamination (inoculum size between 25 and 475 CFU / 25 g) using 0111 , 0145, 0157 sorbitol positive and 0157 sorbitol negative strains; 81.8 % (18/22) of 026 contaminated samples yielded confirmed isolates. One sample of CPM presented a coliform count of 5.00*10⁵: all four experiments using this sample failed due to overgrowth by Hafnia alvei, Enterobacter cloacae, E. coli, Klebsiella pneumoniae and Kluyvera ssp. Only one out of six 0103 contaminated samples (16.7 %) was successful: both experiments with an 0103 inoculum size of 44 CFU / 25 g were unsuccessful. API20E results indicated these experiments suffered overgrowth from Proteus vulgaris. 8 false positives were identified out of 135 colonies evaluated (5.9 %) originating from the non-0157 discriminating medium. Subsequent use of the confirmation media reduced this number to 0.0 % for isolates originating from 026 inoculated experiments. 4 false positives out of 19 colonies evaluated (21.1 %) were isolated from the 0157 differential medium: all were subsequently identified as non-pathogenic E. coli.

38 samples of cheese made from raw milk (CRM) were artificially contaminated with single strains belonging to different serotypes: overall success rate was 86.8%. 28 samples were contaminated with inoculum sizes below 125 CFU / 25 g: 24 samples yielded confirmed isolates. All samples contaminated with 0103, 0111 or 0145 strains resulted in confirmed isolates, while only 87.5 % (7/8), 80 % (8/10) and 83.3 % (10/12) of samples resulted in confirmed isolates after contamination with O26, 0157 sorbitol negative and 0157 sorbitol positive strains respectively. Both experiments using a sample inoculated with 25 CFU of strain PH73 were negative: plates were overgrown with non-pathogenic E. coli and Hafnia alvei. The incidence of false positive isolates on the non-0157 discriminating medium was 2.1 % (1/47). These isolates remained suspected after subsequent plating on the confirmation media and were identified using API20E as a non-pathogenic E. coli.

Meat samples

None of the fermented salami sausage samples evaluated had detectable levels of coliforms (< 10 CFU), while some samples of ground beef presented coliform loads up to 2.5*10³. Total counts were between 7.65*10⁵ and 4.95*10⁸ for salami sausages and between 1.40^*10³ and 3.85^*10⁶ for ground beef samples.

Overall, artificially contaminated ground beef samples were analysed during 84 experiments with inoculum sizes between 1 and 6400 CFU / 25 g of sample. In 61 experiments no IMS was applied. These experiments resulted in 47.4 % (9/19), 66.7 % (6/9), 50 % (5/10), 90 % (18/20) and 100 % (3/3) detection rates for 026, O103, O111 , 0145 and 0157 S-contaminated samples respectively. Overall isolation success without application of IMS was 67.2 % (41/61 ). From the 61 contaminated samples without IMS treatment, 276 colonies were isolated from the non-0157 differential medium. 36 out of 276 were identified as a false positive result on the discriminating medium alone: 18.2 % (12/66), 22.2 % (10/45), 0.0 % (0/41 ) and 11.3 % (14/124) originated from 026, O103, O111 and 0145 contaminated samples. After subsequent streaking on the confirmation media, 23 out of 257 (8.9 %) colonies with suspected morphology were identified as false positives: 22 % (10/45) of presumed O111 isolates failed PCR confirmation, while 11.5 % (7/61 ), 0.0 % (0/41 ) and 5.5 % (6/110) of suspected isolates were identified as false positives from samples contaminated with 026, 0103 and 0145 strains. All false positive isolates were confirmed as non-pathogenic E. coli using API20E. 23 out of 84 experiments were subjected to IMS treatment. While the isolation rate was 47.4 %, 66.7 % and 50.0 % for 026, 0103 or 0111 contaminated samples, application of IMS did allow isolation of strains belonging to these serotypes in all experiments conducted. 37 experiments were conducted using salami sausage as sample matrix, without application of IMS treatment: 31 had an inoculation size equal or below 125 CFU / 25 g. Overall isolation efficiency was 62.2 % (23/37); isolation efficiency was 50 %, 50 %, 33 %, 84.6 % and 75 % for experiments using strains belonging to 026, O103, 0111 , 0145 and 0157 S-contaminated respectively. Out of a total of 104 suspected isolates on the non-0157 discriminating medium, three (2.9 %) were identified as false positives. No false positives were isolated from 026, 0111 , 0145 or 0157 S- contaminated samples. After subsequent streaking on the confirmation media, the overall ratio dropped to 0.0 % (0/101 ), as all 3 isolates from 0103 inoculated experiments were identified as non-suspected on the confirmation media. During three experiments IMS treatment was successfully applied to increase isolation efficiency of 026 contaminated samples.

Claims

1. A method for selectively identifying pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes on two or more plating media, comprising the step of determining the STEC serotype based on at least 4 different STEC serotype specific colours formed onto at least one of said plating media.

2. A method according to claim 1 , wherein said two or more plating media are used consecutively for the said determination.

3. A method according to claim 1 or claim 2, wherein a plating medium onto which at least 3 different STEC serotype specific colours are formed is used in the first place for the said determination.

4. A method according to any of claims 1 to 3, wherein at least one of said two or more plating media comprise at least one chromogenic marker in combination with at least two metabolic markers.

5. A method according to any of claims 1 to 4, wherein the plating medium onto which at least 3 different STEC serotype specific colours are formed comprises at least one chromogenic marker in combination with at least two metabolic markers.

6. A method according to claim 4 or claim 5, wherein said metabolic markers comprise carbohydrates.

7. A method according to claim 4 or claim 5, wherein said at least one chromogenic marker is δ-bromo^-chloro-S-indolyl-beta-D-galactopyranoside.

8. A method according to any of claims 1 to 7, wherein said STEC serotypes are independently selected from the group consisting of 026, 0103, 0111 , 0145, sorbitol negative 0157 and sorbitol positive 0157.

9. A method according to any of claims 1 to 8, being performed on a sample suspected of containing said STEC serotypes, wherein said sample is selected from the group consisting of blood, saliva, faeces and urine.

10. A method according to any of claims 1 to 8, being performed on a sample suspected of containing said STEC serotypes, wherein said sample is food.

11. A plating medium for the selective identification of pathogenic Shigatoxin producing Escherichia coli (STEC) serotypes comprising a combination of at least one chromogenic marker and at least 2 metabolic markers, wherein said combination allows for at least 3 different STEC serotype-specific colours.

12. A plating medium according to claim 11 , having a pH of about 7.4 at a temperature of 20⁰C.

13. A plating medium according to claim 11 or claim 12, wherein said metabolic markers comprise carbohydrates.

14. A plating medium according to claim 13, wherein said carbohydrates are selected from the group consisting of sucrose and sorbose.

15. A plating medium according to any of claims 11 to 14, wherein said at least one chromogenic marker is δ-bromo^-chloro-S-indolyl-beta-D-galacto- pyranoside.

16. A plating medium according to any of claims 11 to 15, further comprising one or more compounds selected from the group consisting of MacConkey agar base, potassium tellurite, Novobiocine and bile salts.

17. A plating medium according to any of claims 11 to 16, in combination with a second plating medium comprising a carbohydrate as a metabolic marker.

18. A plating medium according to claim 17, wherein said carbohydrate in said second plating medium is selected from the group consisting of sorbitol, sucrose, sorbose, L-rhamnose, dulcitol, D-arabinose and D-raffinose.