Chapter 15 Isolation of yersinia enterocolitica from foods

Handbook of Culture Media for Food Microbiology, J.E.L. Corry et al. (Eds.) 9 2003 Elsevier Science B.V. All rights reserved

215

Chapter 15 Isolation of Yersinia enterocolitica from foods E. d e B o e r Inspectorate for Health Protection, PO Box 202, 7200 AE Zutphen, The Netherlands

Many selective enrichment and plating media for the isolation of Yersinia enterocolitica from foods have been described. Use of many of these results in the isolation of non-pathogenic Yersinia strains. At present no single isolation procedure is available for the recovery of all pathogenic strains of Y. enterocolitica. Cold enrichment in phosphate-buffered saline plus 1% sorbitol and 0.15% bile salts (PBSSB) and two-step enrichment with tryptone soy broth (TSB) and bile oxalate sorbose (BOS) broth are very efficient methods for the recovery of a wide spectrum of Y enterocolitica serotypes. Enrichment in irgasan ticarcillin chlorate (ITC) broth is the most efficient method for recovery of strains of serotype 0:3, the most prevalent clinical serotype of Y enterocolitica in Europe. Postenrichment alkali treatment often results in higher isolation rates. Cefsulodin irgasan novobiocin (CIN) agar and Salmonella-Shigella deoxycholate calcium chloride (SSDC) agar are the most frequently used plating media. For the recovery of serotype 0:8 strains, the common clinical isolates in North America, enrichment in BOS and plating on CIN agar seems the most efficient procedure. Selection of the proper isolation procedure will depend on the bio/serogroups of Yersinia spp. sought and on the type of food to be examined. Use of more than one medium for both enrichment and plating will result in higher recovery rates of Yersinia spp. from foods. Serotyping, biotyping and virulence testing is essential for differentiation between pathogenic and environmental Yersinia strains. The International Standard Organization method for the detection of presumptive pathogenic Y enterocolitica includes parallel use of the following two isolation procedures: (1) Enrichment in peptone, sorbitol and bile salts (PSB) broth for 2-3 days at 22-25~ with agitation or 5 days without agitation; plating on CIN agar directly and after alkaline treatment and incubation for 24 h at 30~ (2) Enrichment in ITC for 2 days at 24~ plating on SSDC agar and incubation for 2 days at 30~

Introduction Yersinia enterocolitica is a Gram-negative, facultatively anaerobic coccobacillus belonging to the genus Yersinia in the Enterobacteriaceae family. The organism is recognised as a foodborne pathogen and some large food-associated outbreaks of yersiniosis have been reported (Bottone, 1997). In developed countries, Y. enterocolitica can be isolated from 1 - 2 % of all h u m a n cases of acute enteritis (Kapperud, 1991). Yersiniosis in the United States has been characterized by foodborne outbreaks, whereas in Europe and Japan it is endemic (Schiemann and Wauters, 1992). The disease can range in severity from self-limiting gastroenteritis to pseudoappendicitis, septicaemia in neonates and immunodeficient patients and postinfectious complica-

216 tions like reactive arthritis and myocarditis (Bottone, 1997). Though Y. enterocolitica was considered an emerging pathogen after the large outbreaks in the seventies and eighties, epidemiological data do not indicate an increase in cases and outbreaks of yersiniosis in the past decade. Yersinia enterocolitica and related species have been isolated from many types of both raw and processed foods (De Boer et al., 1986; De Boer, 1995). The majority of these food isolates differ in biochemical and serological characteristics from typical clinical strains and are usually classified as 'non-pathogenic' or 'environmental' Yersinia strains. These strains, which include the 'related species' Y. frederiksenii, Y. kristensenii, Y. intermedia, Y. aldovae, Y. rohdei, Y. mollaretti and Y. bercovieri, are ubiquitous and probably have no clinical significance with the exception of a few atypical cases (Kapperud, 1991). Pathogenic Yersinia strains harbour a virulence plasmid, as well as chromogenic virulence genes, such as the enterotoxin gene yst and the invasion-associated gene ail (Burnens et al., 1996). Strains associated with human disease mainly belong to the serogroups 0:3, 0:5,27, 0:8 and 0:9 of Y. enterocolitica sensu stricto (Stolk-Engelaar and Hoogkamp-Korstanje, 1996). The epidemiology of Y. enterocolitica infections is, for the greater part, not understood. There is an association with consumption of contaminated foods, especially pork. The oral cavity and the intestinal tract of healthy pigs have been found to be important reservoirs of pathogenic serotypes of Y. enterocolitica and reduction of the contamination of pig carcasses at the slaughterhouse and, consequently, of raw pork and pork products has been suggested as a means of prevention of foodborne yersiniosis (De Boer and Nouws, 1991; De Boer et al., 1998; Verhaegen et al., 1998). The increasing interest in Y. enterocolitica infections and the role of foods in some outbreaks of yersiniosis has led to the development of improved procedures for the isolation of this organism from foods during the last 15-20 years. As the numbers of Y. enterocolitica organisms in foods are usually low and there is often a great variety of background flora, direct isolation on selective plating media is seldom successful. Isolation methods usually involve enrichment of the sample followed by plating onto selective agar media, confirmation of typical colonies and testing for virulence properties of isolated strains.

Enrichment

As a psychrotrophic organism Y. enterocolitica is able to multiply at 4~ and enrichment at this temperature for 2-4 weeks is widely used. At this temperature, the growth rate of competitive bacteria is slowed sufficiently to enable Y. enterocolitica to multiply to numbers necessary for isolation on plating media. Media used for this 'cold enrichment' include simple buffers like phosphate-buffered saline (PBS), PBS modified by addition of 1% sorbitol and 0.15% bile salts (PBSSB) (Mehlman et al., 1978), PBS supplemented with 1% mannitol (Schiemann, 1979a), PBS with 0.5% peptone (Weagant and Kaysner, 1983), PBS with peptone and cycloheximide (Vidon and Delmas, 1981), tryptone soy broth (Van Pee and Stragier, 1979), trypticase soy broth (Schiemann, 1983a), tryptic soy broth plus polymyxin and

217 novobiocin (TSPN) (Landgraf et al., 1993) and tris-buffered peptone water, pH 8.0 (Greenwood and Hooper, 1989). The long period required for cold enrichment is often unacceptable for quality assurance of foods. Schiemann and Olson (1984) showed than incubation at 15~ for 2 days was as efficient as enrichment at 4~ for some weeks. Doyle and Hugdahl (1983) incubated PBS for 1-3 days at 25~ while tris-buffered peptone water was incubated at 9~ for 11-14 days (Greenwood and Hooper, 1989) or at 21~ with subculturing after 4-7 days and 11-14 days of incubation (Greenwood, 1993). Landgraf et al (1993) recommended incubation at 18~ for 3 days. Several other enrichment procedures involving incubation at higher temperature for shorter periods and using selective media have been proposed. Modified Rappaport broth (Wauters, 1973) has been used for many years as the first-choice medium for the isolation of the major pathogenic serotypes of Y. enterocolitica in Europe. This medium has been shown to inhibit the common North American serotype 0:8 strains of Y. enterocolitica (Schiemann, 1983a; Walker and Gilmour, 1986). Carbenicillin in modified Rappaport broth was shown to inhibit the growth of certain serotype 0:3 strains (Schiemann, 1982). However, Wauters et al. (1988a) stated that serogroup 0:3 strains are not inhibited by carbenicillin and that omission of this antibiotic results in a decrease in selectivity of the enrichment medium. Lee et al. (1980) described two modified selenite media that effectively recovered certain strains of Y. enterocolitica from meats. They found it critical to limit the sample size of the blended meat suspension to 0.2 g per 100 ml enrichment medium to restrict the growth of competitive bacteria. Otherwise the slower growing Y. enterocolitica would be overgrown by the faster growing normal bacterial flora of the meat. Schiemann (1982) developed a two-step enrichment procedure for recovery of Y. enterocolitica from food. In this procedure pre-enrichment for 9 days at 4~ in yeast extract rose bengal broth was followed by selective enrichment with bile oxalate sorbose (BOS) broth at 22~ for 5 days. As the pre-enrichment medium was less selective, it allowed multiplication of small inocula and repair of injured cells. BOS broth was found especially useful for the isolation of serotype 0:8 strains, but strains of serotype 0:5,27 were more difficult to recover (Schiemann, 1983a). Wauters et al. (1988a) developed a new enrichment broth, named ITC broth, derived from modified Rappaport broth and based on the selective agents irgasan, ticarcillin and potassium chlorate. In comparative studies ITC broth was especially effective for the recovery of Y. enterocolitica serotype 0:3 from pork and procine tonsils, while cold and two-step enrichments yielded better results for non-pathogenic strains (Wauters et al., 1988a; Kwaga et al., 1990; De Boer and Nouws, 1991). However, it was found that ITC broth had some short-comings for the enrichment of Y. enterocolitica serotype 0:9 from meat, because of the sensitivity of this serotype to chlorate. De Zutter et al. (1994) suggested omitting chlorate in ITC broth and reducing the concentrations of magnesium chloride and malachite green to 80% of the original concentrations for optimal recovery of serotype 0:9 strains. For the inoculation of ITC broth a filtrate or supernatant of a meat homogenate should preferably be used, allowing the examination of 1 g of meat (De Zutter et al., 1995). A study by Toora et al. (1994) indicated that the addition of ticarcillin and chlorate to ITC did not improve its efficiency

218 and that irgasan alone functioned as a better selective supplement.

Alkaline treatment Aulisio et al. (1980) found that strains of Y. enterocolitica were more tolerant of alkaline solutions than other Gram-negative bacteria. By treating food enrichments with potassium hydroxide (KOH) solutions before plating, the background flora was markedly reduced, thus improving separation of target colonies from similar colonies on the isolation medium. However, Schiemann (1983b) found that various factors, including medium, temperature and growth phase, influence tolerance of alkaline conditions and reduce the effectiveness of this treatment. Weagant and Kaysner (1983) concluded that a specific treatment time cannot be recommended. They found that streaking three to four successive plates from the KOH rinse at 10-s intervals enhanced the probability of obtaining isolated colonies of Y. enterocolitica even when growing in the presence of numerous organisms. Direct KOH treatment of meat samples proved to be a valuable rapid method for direct isolation of Yersinia from meat contaminated with more than 102 cells per g (Fukushima, 1985). Based on results of comparative experiments the use of 0.125% KOH with an exposure time of 5 min was recommended for the direct detection of Y. enterocolitica in foods (Schraft and Untermann, 1989). Direct plating after KOH treatment was only suitable for strongly positive material (Wauters et al., 1988a).

Plating media Different plating media have been used to isolate Y. enterocolitica from clinical specimens and food (Table 1). Initially media like MacConkey, Salmonella-Shigella, desoxycholate citrate and bismuth sulphite agars, designed for the isolation of enteropathogens, were used. Since Y. enterocolitica ferments lactose slowly, colonies are colourless on media such as MacConkey agar, which contains lactose as an indicator substrate. Lee (1977) modified MacConkey agar by addition of Tween 80 to improve differentiation of Yersinia colonies from other lactose-negative colonies. However, lipolytic Yersinia strains, easily recognized on this medium as white wrinkled colonies surrounded by a sheen, are usually non-pathogenic (De Boer and Seldam, 1987). Salmonella-Shigella agar was made more selective for Y. enterocolitica by addition of sodium deoxycholate and CaC12 (SSDC) (Wauters, 1973; Wauters et al., 1988a). Colonies of Y. enterocolitica on this medium are small, round and colourless. Some species of Morganella, Proteus, Serratia and Aeromonas are also able to develop on SSDC and differentiation of Yersinia from these competing organisms may be difficult. Schiemann (1979b) developed cefsulodin-irgasan-novobiocin (CIN) agar, a selective and differential agar medium for Y. enterocolitica. Organisms capable of fermenting mannitol, including yersiniae, produce red coloured 'bullseye' colonies on this medium. CIN agar was found inhibitory to Pseudomonas aeruginosa, Escherichia

219 Table 1 Selective agents in enrichment and plating media for Yersinia enterocolitica. Mediuma

Enrichment MRB Selenite media PBSSB BOS ITC

Plating CIN VYE SSDC MacConkey

Selective agents magnesium chloride (2.8%), malachite green (0.0013 %), carbenicillin (0.00025%) sodium selenite (0.15 or 0.25%), malachite green (0.002%), carbenicillin (0.001%) bile salts (0.15 %) sodium oxalate (0.5%), bile salts (0.2%), irgasan (0.0004%), sodium furadantin (0.001%) magnesium chloride (6%), malachite green (0.001%), irgasan (0.00001%), ticarcillin (0.00001%), potassium chlorate (0.1%) sodium deoxycholate (0.05%), crystal violet (0.0001%), irgasan (0.0004%), cefsulodin (0.0015%), novobiocin (0.00025%) sodium deoxycholate (0.1%), crystal violet (0.0001%), irgasan (0.0004%), cefsulodin (0.0004%), oleandomycin (0.001%), josamycin (0.002%) sodium deoxycholate (0.85 %) bile salts (0.15%), crystal violet (0.0001%)

a MRB, modified Rappaport broth (Wauters, 1973); Selenite media (Lee et al., 1980); PBSSB. phosphate-buffered saline with sorbitol and bile salts (Mehlman et al., 1978); BOS, bile oxalate sorbose broth (Schiemann, 1982; Part 2 of this volume); ITC, irgasan ticarcillin chlorate (Wauters et al., 1988a; Part 2 of this volume); CIN, cefsulodin irgasan novobiocin agar (Schiemann, 1979b; Part 2 of this volume); VYE, virulent Yersinia enterocolitica agar (Fukushima, 1987); SSDC, Salmonella-Shigella deoxycholate calcium chloride agar (Wauters et al., 1988a; Part 2 of this volume); MacConkey agar no.3 (Oxoid CM115). coli, Klebsiella pneumoniae and Proteus mirabilis, but some Enterobacter, Aeromonas and Proteus strains showed a colonial appearance similar to that of Yersinia (De Boer and Seldam, 1987). The addition of 1 lag/ml of streptomycin improved the selectivity of CIN agar, but resulted in smaller Yersinia colonies (Schiemann, 1987). As the recovery rate and the colony size of Y. enterocolitica diminishes during storage of the medium, it is recommended to use CIN medium within 14 days of preparation (Petersen, 1985). On CIN agar colonies of pathogenic and environmental Yersinia strains appear similar. Fukushima (1987) developed a selective agar medium for isolation of pathogenic (virulent) Y. enterocolitica (VYE agar). Pathogenic strains form red colonies on this medium, the result of mannitol fermentation and aesculin nonhydrolysis, while most environmental Yersinia strains form dark red colonies with a dark peripheral zone as a result of mannitol fermentation and aesculin hydrolysis. Another medium for the direct detection and isolation of plasmid-bearing virulent serotypes of Y. enterocolitica was described by Bhaduri and Cottrell (1997). This medium, low-calcium - Congo red - Brain Heart Infusion - agarose (CR-BHO), is incubated at 37~ for 24 h and plasmid-bearing virulent Y. enterocolitica strains appear as red pinpoint colo-

220

Table 2

Yersinia

Differential characteristics of

and related genera.

Characteristics

Y. enterocolitica

Hafnia

Serratia

Citrobacter

Enterobacter

Escherichia

Klebsiella

Proteus

Urease

+"

-

d

d

d

Motility

+

+

+

+

+

-

d

+

+

-

+

-

+

+

+

+

d

-

+

-

-

-

+

d

-

-

-

-

+

d

-

d

+

d

-

at 2 5 ~ Motility at 3 7 ~ Arginine dihydrolase Lysine decarboxylase Phenylalanine

.

.

.

.

.

.

.

+

deaminase H2S p r o d u c t i o n +, p o s i t i v e ; - ,

-

-

+

d

-

-

-

d

n e g a t i v e ; d, d i f f e r e n t r e a c t i o n s .

Table 3 Biochemical

differentiation within the genus E

Test

E

E

Yersinia

E

( f r o m W a u t e r s et al., 1 9 8 8 b ) .

E

E

E

E

E

entero- inter- frede- kristen- aldovae rhodei mollaretii bercovieri pseudocolitica media riksenii senii tuberculosis Indole

d a

+

+

d

.

Voges-

d

+

d

-

+

. .

.

.

-

+

d

-

d

+

-

-

-

L-Omithine

+

+

+

+

+

+

+

+

-

Mucate, acid

-

d

d

-

d

-

+

+

-

Pyrazinamidase

d

+

+

+

+

+

+

+

-

Sucrose

+

+

+

-

-

+

+

+

-

Cellobiose

+

+

+

+

-

+

+

+

-

L-Rhamnose

-

+

+

-

+

-

-

-

+

Melibiose

-

+

-

-

-

d

-

-

+

L-Sorbose

d

+

+

+

-

ND

+

-

-

L-Fucose

d

d

+

d

d

ND

-

+

-

.

. .

.

Proskauer Citrate (Simmons)

a d, d i f f e r e n t r e a c t i o n s " +, p o s i t i v e ; - ,

negative; ND, not determined.

nies.

I d e n t i f i c a t i o n

For the differentiation be used.

Table

3 shows

of

Yersinia

from

the characteristics

related

genera

differentiating

the tests listed

in Table

2 may

the foodborne

species

within

221 Table 4 Biotypes of Yersinia enterocolitica (Wauters et al., 1987). Test

Lipase (Tween-esterase) Aesculin hydrolysis (24 h) Salicin (acid production in 24 h) Indole production Xylose (acid production) Trehalose (acid production) Nitrate reduction Pyrazinamidase

Biogroups 1A

1B

+" + + + + + + +

+ . . + + + + .

2

3

. . .

. . .

d + + + .

.

. . . + + + .

4

5

+ +

-

. . .

.

+, positive;-, negative; d, different reactions. the genus Yersinia. Some biochemical activities (cellobiose, raffinose, indole, ONPG hydrolysis, ornithine decarboxylase, Voges-Proskauer) of Yersinia strains are temperature-dependent (Bercovier and Mollaret, 1984). These test are preferably incubated at 25 or 30~ rather than at 37~ as yersiniae are more active biochemically at these lower temperatures. Miniaturised identification kits like the API 20E (bioM6rieux) and the Crystal E/NF system (Becton Dickinson) have proven to be valuable for rapid identification of Y. enterocolitica strains (Peele et al., 1997; Neubauer et al., 1998; Linde et al., 1999), though for definitive species identification additional testing is often necessary. Serotyping, biotyping and virulence testing are essential for differentiation between pathogenic and environmental Yersinia strains. The serotyping scheme of Y. enterocolitica and related Yersinia species now involves 67 major O factors and 44 H factors (Wauters et al., 1991). Table 4 shows the biotyping scheme of Y. enterocolitica as proposed by Wauters et al. (1987). In Europe, Y. enterocolitica strains of serotypes O: 3 (biogroup 4), 0:9 (biogroup 2) and 0:5,27 (biogroup 2) are the most frequently isolated human pathogenic strains. In North America, strains causing human yersiniosis usually belong to biogroup 1B (serotypes 0:4, 0:8, O:13a,13b, O:18, 0:20). Though biotype 1A strains are considered to be non-pathogenic, some of these strains may be human-adapted and therefore potentially pathogenic (Burnens et al., 1996; Grant et al., 1998). Several in vitro tests have been described to determine the potential virulence of Yersinia isolates and many of these tests are easy to perform in routine laboratories (Farmer et al., 1992) (Table 5). Virulence plasmid expression in Yersinia mostly occurs at temperatures greater than 30~ As most of these tests are plasmid-dependent, it must be realized that plasmids may easily be lost during subculture. DNA-based tests for the identification of potentially pathogenic Y. enterocolitica strains include polymerase chain reaction (PCR) tests for detection of the heat stable enterotoxin gene (yst)(Ibrahim et al., 1997), the attachment and invasion gene plasmid (ail)(Blais and Phillippe, 1995), the invasin gene locus (inv)(Rasmussen et al., 1994) and the virulence plasmid (yadA)(Blais and Phillippe, 1995).

222 Table 5 Pathogenicity testing of Yersinia enterocolitica strains. Test

Reference

Aesculin hydrolysis, 24h, 25~ (negative) Salicin fermentation, 24h, 37~ (negative) Calcium-dependent growth at 37~ on magnesium oxalate agar Congo red uptake at 25 or 37~ Pyrazinamidase activity at 25~ (absent) Autoagglutination at 37~ Crystal violet binding at 37~

Schiemann and Devenish, 1982 Schiemann and Devenish, 1982 Gemski et al., 1980 Prpic et al., 1983 Kandolo and Wauters, 1985 Laird and Cavanaugh, 1980 Bhaduri et al., 1987

In studies on the molecular epidemiology of pathogenic Y enterocolitica strains a number of DNA-based techniques have been utilized, of which pulsed-field gel electrophoresis (PFGE) seems very efficient for differentiation of bioserotype 4/0:3 strains (Fredriksson-Ahomaa et al., 1999).

Comparative studies Reporting a comparison of pre-enrichment media, Schiemann (1983a) proposed tryptone soy broth as the medium of choice. Comparison of the two-step enrichment against cold enrichment and modified Rappaport broth showed improved recovery of human pathogenic strains of Y. enterocolitica from inoculated foods using pre-enrichment in tryptone soy broth followed by enrichment in BOS broth (Schiemann, 1982; 1983a; Walker and Gilmour, 1986). In a study on the occurrence of Y. enterocolitica in foods Delmas and Vidon (1985) evaluated seven enrichment procedures. Most positive samples obtained within the shortest time were found using enrichment in PBSSB and BOS, with alkali treatment before plating onto CIN agar. However, only environmental serogroups of Y. enterocolitica were isolated. The enrichment medium ITC was found superior to enrichment in BOS for isolation of pathogenic Y. enterocolitica from pork products, although enrichment in BOS isolated more non-pathogenic strains (Wauters et al., 1988a; Kwaga et al., 1990; De Boer and Nouws, 1991). In different studies cold enrichment in PBSSB gave high isolation rates for Yersinia spp., which proved to be mainly non-pathogenic environmental strains (De Boer et al., 1982; 1986; De Boer and Seldam, 1987). In these studies MRB was found a suitable medium for the isolation of serogroup 0:3 and 0:9 strains, but much less efficient for the isolation of environmental strains. Alkali treatment often resulted in a marked increase in the number of Yersinia isolations. This occurred especially when high numbers of competing organisms were present, as was usually the case for PBSSB, and when a plating medium with low selectivity, such as MacConkey agar, was used (De Boer et al., 1982; De Boer and Seldam, 1987).

223 MacConkey agar proved to be a very productive medium for Y. enterocolitica (Mehlman et al., 1978; Schiemann, 1979b; De Boer and Seldam, 1987). However, Yersinia colonies are hard to recognize on this medium because of its low selectivity. In several comparative studies CIN agar was found to be the most selective isolation medium for Yersinia spp. (Aldova et al., 1990; Cox et al., 1990 ; Schiemann, 1983a ; Walker and Gilmour, 1986). Fukushima (1987) found that biotype 3B, serogroup O: 3 strains were inhibited in CIN agar. Differentiation of Yersinia colonies from other colonies on CIN agar is not always easy (Fukushima, 1987; De Boer and Seldam, 1987). Growth kinetics of Y. enterocolitica in CIN broth were influenced by selective agents, incubation temperature and virulence plasmid carriage (Logue et al., 2000). Plating ITC enrichments onto SSDC isolated more serogroup 0:3 strains than onto CIN agar (Wauters et al., 1988a; De Boer and Nouws, 1991). SSDC proved to be unsuitable after alkali treatment (Wauters et al., 1988a). In a recent study on the occurrence of pathogenic Y. enterocolitica in faeces and tonsils of pigs at slaughterhouses the enrichment media peptone, sorbitol and bile salts (PSB) broth, ITC broth and Yersinia Selective Enrichment broth (YSEB)(Merck) and the plating media CIN agar and SSDC agar were compared. ITC broth was found to be the most productive enrichment medium, whereas much lower isolation percentages were obtained with YSEB and PSB broths. Results with CIN agar were slightly better than those obtained with SSDC agar (De Boer et al., 1998).

Discussion Addition of selective agents including magnesium chloride, malachite green, bile salts, irgasan, and the antibiotics carbenicillin, ticarcillin and cefsulodin to Yersinia isolation media results in growth inhibition of Gram-positive and some Gram-negative flora (Table 1). The colonial appearance of Yersinia spp. on currently used isolation media is not always characteristic, thus confirmation of presumptive colonies is always necessary. At present, no single isolation procedure is available for the recovery of all pathogenic strains of Y. enterocolitica from foods. Cold enrichment in PBSSB, two-step enrichment with PBS and BOS, and enrichment in ITC are the most commonly used enrichment procedures. CIN and SSDC agars are the most frequently used plating media and are also commercially available. These enrichment and plating media are not particularly selective for Y. enterocolitica as they support the growth of several other members of the Enterobacteriaceae. This makes the isolation of low numbers of Yersinia in products containing many other contaminants rather difficult. Moreover, non-pathogenic environmental Yersinia strains are very common in many raw foods and may greatly hinder the isolation of pathogenic Yersinia strains from these products. Cold enrichment and two-step enrichment are very efficient methods for the recovery of a wide spectrum of Y. enterocolitica serotypes. However, usually only pathogenic serotypes are of interest. Since the colonial appearance of pathogenic and environmental strains on CIN and SSDC agar is similar, selection of pathogenic

224 strains on these media requires much effort. This highlights the need for the development and evaluation of agar media selecting for pathogenic serotypes, like VYE agar (Fukushima, 1985). Enrichment in ITC was found to be the most efficient method for the recovery of strains of serotype 0:3, which is the most common clinical serotype of Y. enterocolitica in Europe; however for other pathogenic serotypes this method was less efficient, as was also the case for SSDC agar. For the recovery of the North American serotype 0:8 strains, enrichment in BOS and plating on CIN agar seems the most efficient procedure. Selection of a suitable enrichment procedure will depend on the bio/serotypes of Yersinia spp. sought and on the type of food to be examined. The use of more than one medium for both enrichment and plating will obviously result in higher recovery rates of Yersinia spp. from foods. The International Standard Organization method for the detection of presumptive pathogenic Y. enterocolitica (ISO, 1994) includes parallel use of the following two isolation procedures: (1) Enrichment in peptone, sorbitol and bile salts (PSB) broth for 2-3 days at 22-25~ with agitation or 5 days without agitation; plating on CIN agar directly and after alkaline treatment and incubation for 24 h at 30~ (2) Enrichment in ITC for 2 days at 24~ plating on SSDC agar and incubation for 2 days at 30~ Higher sensitivity and specificity for the detection of pathogenic Y. enterocolitica is obtained by using PCR methods directed at virulence genes (Harnett et al., 1996; Nilsson et al., 1998; Fredriksson-Ahomaa et al., 2000).

References

Aldova, E., Svandova, E., Votypka, J. and Sourek, J. (1990) Comparative study of culture methods to detect Yersinia enterocolitica serogroup 0:3 on swine tongues. Zbl. Bakteriol. 272, 306-312. Aulisio, C.C.G., Mehlman, I.J. and Sanders, C. (1980) Alkali method for rapid recovery of Yersinia enterocolitica and Yersinia pseudotuberculosis from foods. Appl. Environ. Microbiol. 39, 135-140. Bercovier, H. and Mollaret, H.H. (1984) Yersinia. In: Krieg, N.R. and Holt, J.G. (Eds.), Bergey's Manual of Systematic Bacteriology, Vol. 1, Williams and Wilkins, Baltimore and London. Bhaduri, S., Conway, L.K. and Lachica, R.V. (1987) Assay of crystal violet binding for rapid identification of virulent plasmid-bearing clones of Yersinia enterocolitica. J. Clin. Microbiol 25, 1039-1042. Bhaduri, S. and Cottrell, B. (1997) Direct detection and isolation of plasmid-bearing virulent serotypes of Yersinia enterocolitica from various foods. Appl. Environ. Microbiol. 63, 49524955. Blais, B.W. and Phillippe, L.M. (1995) Comparative analysis of yadA and ail polymerase chain reaction methods for virulent Yersinia enterocolitica. Food Control 4, 211-214. Bottone, E.J. (1997) Yersinia enterocolitica: The charisma continues. Clin. Microbiol. Rev. 10, 257-276. Burnens, A.E, Frey, A. and Nicolet, J. (1996) Association between clinical presentation, biogroups and virulence attributes of Yersinia enterocolitica strains in human diarrhoeal disease. Epidemiol. Infect. 116, 27-34.

225 Cox, N.A., Bailey, J.S., Del Corral, E, Shotts, E.B. (1990) Comparison of enrichment and plating media for isolation of Yersinia. Poultry Sci. 69, 686-693. De Boer, E., Hartog, B.J. and Oosterom, J. (1982) Occurrence of Yersinia enterocolitica in poultry products. J. Food Prot. 45,322-325. De Boer, E., Seldam, W.M. and Oosterom, J. (1986) Characterization of Yersinia enterocolitica and related species isolated from foods and porcine tonsils in the Netherlands. Int. J. Food Microbiol. 3, 217-224. De Boer, E. and Seldam, W.M. (1987) Comparison of methods for the isolation of Yersinia enterocolitica from porcine tonsils and pork. Int. J. Food Microbiol. 5, 95-101. De Boer, E. and Nouws, J.EM. (1991) Slaughter pigs and pork as a source of human pathogenic Yersinia enterocolitica. Int. J. Food Microbiol. 12, 375-378. De Boer, E. (1995) Isolation of Yersinia enterocolitica from foods. Contrib. Microbiol. Immunol. 13, 71-73. De Boer, E., Hulleman, A. and Kleverwal, M. (1998) Comparison of culture media for the isolation of Yersinia enterocolitica from porcine faeces and tonsils. Ned. Tijdschrift voor Medische Microbiologie 6, $46-47 (Abstracts 7th International Congress on Yersinia, Nijmegen, The Netherlands). De Zutter, L., Le Mort, L., Janssens, M. and Wauters, G. (1994) Short-comings of irgasan ticarcillin chlorate broth for the enrichment of Yersinia enterocolitica biotype 2, serotype 9 from meat. Int. J. Food Microbiol. 23, 231-237. De Zutter, L., Janssens, M. and Wauters, G. (1995) Detection of Yersinia enterocolitica serogroup 0:3 using different inoculation methods of the enrichment medium Irgasan Ticarcillin Chlorate. Contrib. Microbiol. Immunol. 13, 123-125. Delmas, C.L. and Vidon, D.J.M. (1985) Isolation of Yersinia enterocolitica and related species from foods in France. Appl. Environ. Microbiol. 50, 767-771. Doyle, M.P. and Hugdahl, M.B. (1983) Improved procedure for recovery of Yersinia enterocolitica from meats. Appl. Environ. Microbiol. 45, 127-135. Farmer, J.J., Carter, G.P., Miller, V.L., Falkow, S. and Wachsmuth, I.K. (1992) Pyrazinamidase, CRMOX agar, salicin fermentation-esculin hydrolysis, and D-xylose fermentation for identifying pathogenic serotypes of Yersinia enterocolitica. J. Clin. Microbiol. 30, 2589-2594. Fredriksson-Ahomaa, M., Autio, T. and Korkeala, H. (1999) Efficient subtyping of Yersinia enterocolitica bioserotype 4/0:3 with pulsed-field gel electrophoresis. Lett. Appl. Microbiol. 29, 308-312. Fredriksson-Ahomaa, M., Korte, T. and Korkeala, H. (2000) Contamination of carcasses, offals, and the environment with yadA-positive Yersinia enterocolitica in a pig slaughterhouse. J. Food Protect. 63, 31-35. Fukushima, H. (1985) Direct isolation of Yersinia enterocolitica and Yersinia pseudotuberculosis from meat. Appl. Environ. Microbiol. 50, 710-712. Fukushima, H. (1987) New selective agar medium for isolation of virulent Yersinia enterocolitica. J. Clin. Microbiol. 25, 1068-1073. Gemski, P., Lazere, J.R. and Casey, T. (1980) Plasmid associated with pathogenicity and calcium dependency of Yersinia enterocolitica. Infect. Immun. 27, 682-685. Grant, T., Bennett-Wood, V. and Robins-Browne, R.M. (1998) Identification of virulence-associated characteristics in clinical isolates of Yersinia enterocolitica lacking classical virulence markers. Infect. Immun. 66, 1113-1120. Greenwood, M.H. and Hooper, W.L. (1989) Improved methods for the isolation of Yersinia species from milk and foods. Food Microbiol. 6, 99-104. Greenwood, M.H. (1993) Comparison of enrichment at 9~ and 21~ for recovery of Yersinia species from food and milk. Food Microbiology 10, 23-30. Harnett, N., Lin, Y.P. and Krishnan, C. (1996) Detection of pathogenic Yersinia enterocolitica using the multiplex polymerase chain reaction. Epidemiol. Infect. 117, 59-67.

226 Ibrahim, A., Liesack, W., Griffiths, M.W. and Robins-Browne, R.M. (1997) Development of a highly specific assay for rapid identification of pathogenic strains of Yersinia enterocolitica based on PCR amplification of the Yersinia heat-stable enterotoxin gene (yst). J. Clin. Microbiol. 35, 1636-1638. ISO. (1994) Microbiology- General guidance for the detection of presumptive pathogenic Yersinia enterocolitica. ISO 10273. International Standard Organization, Geneva. Kandolo, K. and Wauters, G. (1985) Pyrazinamidase activity in Yersinia enterocolitica and related organisms. J. Clin. Microbiol. 21,980-982. Kapperud, G. (1991) Yersinia enterocolitica in food hygiene. Int. J. Food Microbiol. 12, 53-66. Kwaga, J., Iversen, J.O. and Saunders, J.R. (1990) Comparison of two enrichment protocols for the detection of Yersinia in slaughtered pigs and pork products. J. Food Protect. 53, 1047-1049. Laird, W.J. and Cavanaugh, D.C. (1980) Correlation of autoagglutination and virulence of yersinae. J. Clin. Microbiol. 11,430-432. Landgraf, M., Iaria, S.T. and Falc~o, D.E (1993) An improved enrichment procedure for the isolation of Yersinia enterocolitica and related species from milk. J. Food Protect. 56, 447-450. Lee, W.H. (1977) Two plating media modified with Tween 80 for isolating Yersinia enterocolitica. Appl. Environ. Microbiol. 33,215-216. Lee, W.H., Harris, M.E., McClain, D., Smith, R.E. and Johnston, R.W. (1980) Two modified selenite media for recovery of Yersinia enterocolitica from meats. Appl. Environ. Microbiol. 39, 205-209. Linde, H.J., Neubauer, H., Meyer, H., Aleksic, S. and Lehn, N. (1999) Identification of Yersinia species by the VITEK GNI card. J. Clin. Microbiol. 37, 211-214. Logue, C.M., Sheridan, J.J., McDowell, D.A., Blair, I.S. and Hegarty, T. (2000) The effect of temperature and selective agents on the growth of Yersinia enterocolitica serotype 0:3 in pure culture. J. Appl. Microbiol. 88, 1001-1008. Mehlman, I.J., Aulisio, C.C.G. and Sanders, A.C. (1978) Problems in the recovery and identification of Yersinia from food. J. Assoc. Off. Anal. Chem. 61, 761-771. Neubauer, H., Sauer, T., Becker, H., Aleksic, S. and Meyer, H. (1998) Comparison of systems for identification and differentiation of species within the genus Yersinia. J. Clin. Microbiol. 36, 3366-3368. Nilsson, A., Lambertz, S.T., St~lhandske, E, Norberg, E and Danielsson-Tham, M.-L. (1998) Detection of Yersinia enterocolitica in food by PCR amplification. Lett. Appl. Microbiol. 26, 140-144.

Peele, D., Bradfield, J., Pryor, W. and Vore, S. (1997) Comparison of identifications of human and animal source Gram-negative bacteria by AP120E and Crystal E/NF systems. J. Clin. Microbiol. 35, 213-216. Petersen, T. (1985) Keeping quality of cefsulodin-irgasan-novobiocin (CIN) medium for detection and enumeration of Yersinia enterocolitica. Int. J. Food Microbiol. 2, 49-54. Prpic, J.K., Robins-Browne, R.M. and Davey, R.B. (1983) Differentiation between virulent and avirulent Yersinia enterocolitica isolates by using Congo red agar. J. Clin. Microbiol. 18, 486-490. Rasmussen, H.N., Rasmussen, O.E, Andersen, J.K. and Olsen, J.E. (1994) Specific detection of pathogenic Yersinia enterocolitica by two-step PCR using hot-start and DMSO. Molecular and Cellular Probes 8, 99-108. Schiemann, D.A. (1979a) Enrichment methods for recovery of Yersinia enterocolitica from foods and raw milk. Contr. Microbiol. Immunol. 5, 212-217. Schiemann, D.A. (1979b) Synthesis of a selective agar medium for Yersinia enterocolitica. Can. J. Microbiol. 25, 1298-1304. Schiemann, D.A. (1982) Development of a two-step enrichment procedure for recovery of Yersinia enterocolitica from food. Appl. Environ. Microbiol.43, 14-27. Schiemann, D.A. and Devenish, J.A. (1982) Relationship of HeLa cell infectivity to biochemical,

227 serological and virulence characteristics of Yersinia enterocolitica. Infect. Immun. 35,497-506. Schiemann, D.A. (1983a) Comparison of enrichment and plating media for recovery of virulent strains of Yersinia enterocolitica from inoculated beef stew. J. Food Protect. 46, 957-964. Schiemann, D.A. (1983b) Alkalotolerance of Yersinia enterocolitica as a basis for selective isolation from food enrichments. Appl. Environ. Microbiol. 46, 22-27. Schiemann, D.A. and Olson, S.A. (1984) Antagonism by Gram-negative bacteria to growth of Yersinia enterocolitica in mixed cultures. Appl. Environ. Microbiol. 48, 539-544. Schiemann, D.A. (1987) Yersinia enterocolitica in milk and dairy products. J. Dairy Sci. 70, 383-391. Schiemann, D.A. and Wauters, G. (1992) Yersinia. In: Vanderzant, C. and Splitttoesser (eds.), Compendium of methods for the microbiological examination of foods, p. 433-450. American Public Health Association, Washington, D.C. Schraft, H. and Untermann, E (1989) Use of KOH treatment for direct detection of Yersinia enterocolitica in foods. 10th WAVFH Int. Symp., Stockholm, Abstracts, p.31. Stolk-Engelaar, V.M.M. and Hoogkamp-Korstanje, J.A.A. (1996) Clinical presentation and diagnosis of gastrointestinal infections by Yersinia enterocolitica in 261 Dutch patients. Scand. J. Infect. Dis. 28, 571-575. Toora, S., Budu-Amoako, E., Ablett, R.E and Smith, J. (1994) Evaluation of different antimicrobial agents used as selective supplements for isolation and recovery of Yersinia enterocolitica. J. Appl. Bacteriol. 77, 67-72. Van Pee, W. and Stragier, J. (1979) Evaluation of some cold enrichment and isolation media for the recovery of Yersinia enterocolitica. Antonie van Leeuwenhoek J. Microbiol. Serol. 45, 465-477. Verhaegen, J., Charlier, J., Lemmens, P., Delm6e, M., Van Noyen, R., Verblist, L. and Wauters, G. (1998) Surveillance of human Yersinia enterocolitica infections in Belgium: 1967-1996. Clin. Infect. Dis. 27, 59-64. Vidon, D.J.M. and Delmas, C.L. (1981) Incidence of Yersinia enterocolitica in raw milk in Eastern France. Appl. Environ. Microbiol. 41,355-359. Walker, S.J. and Gilmour, A. (1986) A comparison of media and methods for the recovery of Yersinia enterocolitica and Yersinia enterocolitica-like bacteria from milk containing simulated raw milk microfloras. J. Appl.. Bacteriol. 60, 175-183. Wauters, G. (1973) Improved methods for the isolation and the recognition of Yersinia enterocolitica. Contr. Microbiol. Immunol. 2, 68-70. Wauters, G., Kandolo, K. and Janssens, M. (1987) Revised biogrouping scheme of Yersinia enterocolitica. Contr. Microbiol. Immunol. 9, 14-21. Wauters, G., Goossens, V., Janssens, M., and Vandepitte, J. (1988a) New enrichment method for isolation of pathogenic Yersinia enterocolitica serogroup O :3 from pork. Appl. Environ. Microbiol. 54, 851-854. Wauters, G., Janssens, M., Steigerwalt, A.G. and Brenner, D.J. (1988b) Yersinia mollaretii sp.nov. and Yersinia bercovieri sp.nov., formerly called Yersinia enterocolitica biogroups 3A and 3B. Int. J. Syst. Bacteriol. 38,424-429. Wauters, G., Aleksic, S., Charlier, J. and Schulze, G. (1991) Somatic and flagellar antigens of Yersinia enterocolitica and related species. Contrib. Microbiol. Immunol. 12, 239-243. Weagant, S.D. and Kaysner, C.A. (1983) Modified enrichment broth for isolation of Yersinia enterocolitica from nonfood sources. Appl. Environ. Microbiol. 45,468-471.