- Email: [email protected]
Evaluation of culture media for enrichment and isolation of Shigella sonnei and S. flexneri
International Journal of Food Microbiology 70 Ž2001. 255–265 www.elsevier.comrlocaterijfoodmicro
Evaluation of culture media for enrichment and isolation of Shigella sonnei and S. flexneri M. Uyttendaele ) , C.F. Bagamboula, E. De Smet, S. Van Wilder, J. Debevere Laboratory of Food Microbiology and Food PreserÕation, Department of Food, Technology and Nutrition, UniÕersity of Gent, Coupure Links 653, 9000 Ghent, Belgium Received 10 October 2000; received in revised form 22 April 2001; accepted 18 May 2001
Abstract The performance of Gram-negative ŽGN. broth with Ž10 mgrml. and without novobiocin, Shigella broth ŽSB. with 0.5 and 3.0 mgrml novobiocin, all incubated at 37 8C ŽSB with 3.0 mgrml novobiocin also at 42 8C. and buffered brilliant green bile glucose ŽEE. broth with 1.0 mgrml novobiocin incubated at 37 and 42 8C were evaluated for resuscitation and growth of Shigella sonnei and S. flexneri Žeight strains; unstressed, chill-stressed and acid-stressed. and non-shigellae Ž11 strains.. GN broth with or without novobiocin supported significantly less growth of Shigella sp. No significant differences in growth of shigellae were obtained between the other culture media. Performance depended more on the Shigella strain used. None of the tested media were significantly superior for suppressing the competitive flora. Electivity and selectivity of MacConkey agar ŽMAC., tergitol-7 agar ŽT7., desoxycholate citrate agar ŽDCA., xylose lysine desoxycholate agar ŽXLD., Salmonella Shigella agar and Hektoen enteric agar ŽHEA. were determined by ecometric testing. HEA confirmed to be a high selective medium for both non-shigellae and stressed Shigella sp. Klebsiella sp., Enterobacter sp., Citrobacter sp., Salmonella sp. and the Escherichia strains can mask the presence of shigellae. In vitro competition experiments and experiments with artificially contaminated foods showed higher resistance of S. sonnei than S. flexneri towards the stress imposed by the food matrix and its indigenous flora. Reliable detection, however, of shigellae in foods with the current enrichment and isolation media was not achieved. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Shigella sonnei; Shigella flexneri; Detection; Isolation; Enrichment
1. Introduction
)
Corresponding author. Tel.: q32-9-264-6178; fax: q32-9225-5510. E-mail address: [email protected] ŽM. Uyttendaele..
Shigella spp. are the causative agents of bacillary dysentery Žshigellosis.. Although the number of reported cases of shigellosis arising from food is much lower than that of salmonellosis or campylobacteriosis, it is an important cause of foodborne bacterial infection in developed countries ŽMead et al., 1999..
0168-1605r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 Ž 0 1 . 0 0 5 4 9 - 9
256
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
The species involved are mostly Shigella sonnei Žabout 70%. and S. flexneri Žapproximately 25%. with the other types seldom implicated ŽMossel et al., 1995b.. The foods most often implicated are potato salads and salads containing chicken or fish, raw vegetables and seafoods. The contamination usually occurs due to poor personal hygiene by an infected food handler ŽLampel et al., 2000; Smith and Buchanan, 1992.. The reliable detection of shigellae in foods is rather difficult. They are easily overgrown by other bacteria present in the food. In particular, acids produced by other Enterobacteriaceae during the fermentation of carbohydrates are toxic for shigellae ŽMehlman et al., 1985.. An effective selective enrichment procedure for all shigellae is not yet available. The current culture method in the US FDA Ž1992. bacteriological analytical manual ŽBAM. selects for Shigella by enrichment in a low carbohydrate medium Ž Shigella broth. and incubating in an anaerobic atmosphere at an elevated temperature in the presence of the antibiotic novobiocin Ž0.5 mgrml for S. sonnei, 3.0 mgrml for other shigellae. ŽJune et al., 1993; Lampel et al., 2000.. Other enrichment broths are recommended for isolation of shigellae, such as Gram-negative ŽGN. broth with novobiocin Ž10 mgrml. ŽBeckers and Soentoro, 1989. or without novobiocin ŽNKML, 1995. and enrichment in brilliant green bile glucose ŽEE. broth with 1 mgrml novobiocin ŽMossel and Jacobs-Reitsma, 1990.. Many culture media for isolation of shigellae from foods have been suggested: solid media with a low ŽMacConkey agar ŽMAC. and tergitol-7 ŽT7.., intermediate Žxylose lysine desoxycholate agar ŽXLD. and desoxycholate citrate agar ŽDCA.. or high selectivity Ž Salmonella Shigella agar ŽSSA. and Hektoen enteric agar ŽHEA.. can be used Žde Boer, 1998; Lampel et al., 2000.. At least two to three different types of selective agar media should be used to increase the chance of recovery of shigellae from foods. These detection methods are recommended in different regions of the world but have never been extensively compared and evaluated in their performance for isolation of shigellae from foods. The present study was undertaken to provide data on the electivity and selectivity of the culture media for
enrichment and isolation of S. sonnei and S. flexneri using unstressed, chill-stressed and acid-stressed cells.
2. Material and methods 2.1. Bacterial strains S. sonnei CIP8249 and S. flexneri CIP8248 were obtained from the Institut Pasteur-Paris ŽFrance.. In addition, two clinical isolates of S. sonnei Žstrains 1 and 2. and four clinical isolates of S. flexneri Žtwo of serotype six Žstrains 6 and 7., two of serotype 2a Žstrains 11 and 12. were obtained from the Scientific Institute of Public Health ŽIPH. in Brussels ŽBelgium.. Eleven non-shigella strains were also included in the study. Aeromonas hydrophila LMG 3771, Citrobacter freundii LMG 3246, Enterobacter aerogenes LMG 2094, Escherichia coli LMG 8223, E. hermanii LMG 7867, Klebsiella oxytoca LMG 3055, Salmonella enteritidis LMG 10395, Proteus Õulgaris LMG 2096, ProÕidencia rettgeri LMG 3259 and Pseudomonas fluorescens LMG 1794 were obtained from the LMGrBCCMe Culture Collection ŽUniversity of Gent, Ghent, Belgium.. Erwinia carotoÕora was an own isolate obtained from vegetables. 2.2. Culture media Tryptone soy broth ŽTSB; Oxoıd, ¨ Basingstoke, Hampshire, England. was used as a control medium. Selective enrichment media included in the study were Gram-negative ŽGN. broth according to Hajna Žin Beckers and Soentoro, 1989. with 10 mgrml novobiocin ŽSigma, St. Louis, MO. ŽGN q 10.0. and without novobiocin ŽGN., Shigella broth prepared according to the US FDA Ž1992. BAM with 0.5 mgrml novobiocin ŽSB q 0.5. or 3.0 mgrml novobiocin ŽSB q 3.0. and EE broth ŽOxoıd ¨ . with 1.0 mgrml novobiocin ŽEE q 1.0.. Plating media included in the study were tryptone soy agar ŽTSA; Oxoid. as a control medium and the selective media: MacConkey agar ŽMAC; Merck, Darmstadt, Germany., tergitol-7 agar ŽT7; Oxoıd ¨ ., desoxycholate citrate agar ŽDCA; Diagnostics Pasteur, Marnes la Coquette, France., xylose lysine
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
desoxycholate agar ŽXLD; Oxoıd ¨ ., Salmonella Shigella agar ŽSSA; Oxoıd ¨ ., Hektoen enteric agar ŽHEA; Diagnostics Pasteur.. 2.3. Inoculum From a stock culture, a subculture was placed in brain heart infusion ŽBHI; Oxoıd ¨ . broth and incubated for 18–20 h at 37 8C Žunstressed culture.. For the chill-stress, this BHI culture was stored for a further 6 days at 4 8C. For the acid-stress, 1 ml of an 8-h culture in BHI at 37 8C was transferred to BHI acidified to pH 4.0 with HCl and incubated for 16–18 h at 37 8C. 2.4. EÕaluation of enrichment media A 0.1-ml volume of appropriate dilutions of the BHI cultures of shigellae Žunstressed, chill-stressed and acid-stressed. and non-shigella species were inoculated into 10 ml of the various enrichment media Žin duplicate. to obtain approximately 10 3 cfurml. Dilutions of the BHI cultures were plated on TSA ŽOxoıd ¨ . to determine the exact inoculum. After incubation for 4 h Žunstressed and chill-stressed inoculum. or 6 h Žacid-stressed inoculum. at 37 8C for GN, GN q 10.0, SB q 0.5, SB q 3.0 and EE q 1.0 and incubation at 42 8C for SB q 3.0 and EE q 1.0, numbers of colony-forming units were determined by spiral plating of appropriate dilutions on TSA, counting colonies after 24 h of incubation at 37 8C. 2.5. EÕaluation of isolation media The various selective media were tested for their productivity and selectivity towards Shigella spp. and non-shigellae by ecometric evaluation. Tryptone soy agar ŽTSA, Oxoid. was used as the control medium. Ecometric evaluation relies on inoculating activated cultures on solid media applying a carefully standardized, sequential streaking technique to ever decreasing numbers of colonies ŽMossel et al., 1995a.. Unstressed cultures of Shigella sp. and nonshigellae as well as chill-stressed cultures of Shigella sp. Žapproximately 10 8 cfurml. were inoculated on TSA, MAC, T7, XLD, DCA, HEA and SSA. The streaking was done on 9-cm diameter petri dishes.
257
The four quadrants were inoculated in succession with a series of five parallel lines with a single loop, the first one at the perimeter. Finally, a last line was drawn through the centre. For the inoculation, 1-ml plastic disposable loops were used ŽNunc, Life Technologies, Merelbeke, Belgium.. After incubation for 24 h at 37 8C, the final streak that led to substantial colony formation was assessed. The reading was expressed in absolute growth index ŽAGI. as explained by Mossel et al. Ž1995a,b.. The results were then converted to a relative growth index ŽRGI. defined as the ratio of the AGIs on the test medium to those on the reference medium ŽTSA.. Additionally, the size and colour of the colonies on the different media were examined. For each of the test organisms, this procedure was performed in duplicate. 2.6. Effect of non-shigellae on the enrichment and isolation of Shigella sp. EE q 1.0 and SB q 3.0 were inoculated with appropriate dilutions of an unstressed culture of Shigella sp. and a mixture of non-shigella competitors Ž E. coli, E. hermanii, Ent. aerogenes, C. freundii and K. oxytoca. to obtain a ratio of Shigella sp.rnon-shigella of 10:1, 1:1, 1:10 and 1:100 using a Shigella inoculum of 10 3 cfurml. A control with shigellae with no competitive flora was also included. The inoculated enrichment media were incubated for 18–20 h at 42 8C, before streaking onto MAC, XLD, SSA and HEA, incubating for a further 24 h at 37 8C. Typical colonies were streaked for purity on nutrient agar ŽNA; Oxoıd ¨ . and further tested for fermentation of glucose and lactose, H 2 S and indole production and motility by inoculation of KIA and SIM ŽOxoıd ¨ . tubes. The experiment was performed in duplicate for S. sonnei CIP8249, S. flexneri CIP8248, S. sonnei 1, and S. flexneri 2ar12. 2.7. Determination of the effectiÕeness of the selected culture method for the recoÕery of S. sonnei and S. flexneri from foods Samples Ž25 g. of six foods Žchicken salad, potato salad, pre-cut iceberg lettuce, Chinese cabbage, strawberries and scampi. were weighed into sterile
258
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
stomacher bags Žfour portions of each food. and inoculated with appropriate dilutions of an unstressed BHI culture of S. sonnei 1 or S. flexneri 2ar12 to obtain ca. 5–50, 10–100, 50–500 and no Žcontrol. cfurg. Dilutions of the BHI culture were plated on TSA to determine the exact inoculum. The pH was measured with an Ingold electrode ŽMGDX K57, Urdorf. connected to a Knick pH meter Žtype 763, Knickn, Berlin.. The inoculated food was kept at 4 8C for 6 h before 225 ml SB q 3.0 were added to each food sample. The content was homogenized and incubated for 18–20 h at 42 8C, streaked onto MAC, XLD and SSA after enrichment, then incubated for 24 h at 37 8C. Typical colonies were streaked for purity on NA and further tested as described above.
3. Results 3.1. EÕaluation of enrichment media The performance of the enrichment media was evaluated by determination of the increase or decrease in numbers of colony-forming units of Shigella spp. and non-shigella after a short incubation period in the selective broths. The incubation period was restricted to cover the lag phase and the onset of the exponential growth but avoiding multiplication to the stationary growth phase. Enumeration after a short incubation period enabled the response of the bacterial strains to stress exerted by the selective broth andror the ability of these media to resuscitate chillor acid-damaged shigellae to be determined. Results for the various S. sonnei and S. flexneri strains Žunstressed, chill- and acid-stressed. in the various enrichment media are shown in Table 1. Results were statistically treated ŽScheffe analysis by SPSS for Windows. to determine the mean increase in numbers over all Shigella strains for each enrichment medium included in the study and to determine the mean increase in numbers for each of the Shigella strains over all enrichment media tested. GN and GN q 10.0 showed a significantly lower outgrowth, respectively 1.7 and 1.4 log units, of unstressed shigellae than TSB Ž3.2 log. or the other enrichment media Ž2.7–3.1 log.. The GN media
were not used in subsequent experiments with stressed Shigella sp. and non-shigellae. If Shigella spp. were subjected to a chill-stress, the increase in numbers of shigellae Ž2.6 log in TSB and 2.0–2.4 log in the other media. was lower compared to unstressed Shigella spp. Prior exposure to acid Ž16–20 h at pH 4.0. prolonged the lag time of the shigellae and extension of the incubation time to 6 h was necessary to establish multiplication in the enrichment media Ž1.7 log in TSB and 0.7–1.5 log in the selective enrichment media. although numbers were still lower than those achieved after 4 h with unstressed or chill-stressed Shigella spp. No significant differences in increase in numbers for the stressed shigellae were obtained between the different enrichment media tested. Only with acidstressed shigellae in EE q 1.0 and EE q 1.0–42 and with chill-stressed shigellae in SB q 3.0 was outgrowth reduced by more than 0.5 log unit compared to TSB ŽTable 1.. Variation in the performance of the media depended more upon the Shigella strain used than on the enrichment media or incubation temperature ŽTable 1. and this was more pronounced if the strains were stressed. While outgrowth of some strains was unaffected or only slightly affected by prior environmental stress, others showed significantly lower or no increase in numbers independent of the selective enrichment medium applied. Based upon the results obtained, strains can be classified as stress resistant, e.g. S. sonnei CIP8249, S. sonnei 1, S. flexneri CIP8248, S. flexneri 2ar12 or stress sensitive, e.g. S. sonnei 2, S. flexneri 2ar11, S. flexneri 6r6, S. flexneri 6r7. The increase in numbers of competitive flora in the various enrichment media was determined ŽFig. 1.. The incubation temperature rather than the selective agents was probably more inhibitory or lethal for the competitive flora. Reduction of numbers of P. fluorescens was probably due to its inability to grow at 37 8C. Similarly, 42 8C was lethal for Aeromonas and Erwinia. Inhibition of the majority of the competitive flora was accomplished by the selective enrichment media; however, none of the tested media were significantly superior than the others for suppressing competitive flora. EE q 1.0 at 37 8C suppressed growth of Proteus sp. fairly good, while SB q 3.0 at 37 8C suppressed multiplication of
Table 1 Change in numbers Žmean log 10 cfurml of duplicate determinations. of pure cultures of Shigella strains Žunstressed, chill- or acid-stressed. in the various enrichment media, counting on tryptone soya agar S. sonnei 1
S. sonnei 2
S. flexneri CIP
Unstressed (4 h) TSB Ž37 8C. GN Ž37 8C. GNq10.0 Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.74"0.06 1.51"0.30 2.02"0.07 3.47"0.24 3.48"0.02 3.86"0.08 3.57"0.06 3.73"0.02 3.16"0.47
2.58"0.06 2.05"0.04 1.38"0.02 2.32"0.46 2.32"0.11 2.48"0.08 2.05"0.21 2.23"0.38 2.15"0.21
3.15"0.05 0.38"0.28 1.46"0.34 3.04"0.09 2.96"0.02 1.71"0.04 3.05"0.01 2.24"0.15 2.23"0.53
3.38"0.20 1.78"0.10 y0.63"0.16 3.36"0.16 3.76"0.03 2.07"0.04 2.85"0.11 2.51"0.12 2.38"0.72
Chill-stressed (4 h) TSB Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.13"0.04 2.89"0.03 2.87"0.11 3.21"0.04 2.94"0.08 3.05"0.01 3.01"0.09
2.05"0.03 1.98"0.08 1.79"0.00 2.0"0.03 1.71"0.06 1.98"0.01 1.92"0.08
3.25"0.04 3.05"0.03 2.81"0.10 2.40"0.06 3.06"0.04 2.20"0.03 2.80"0.25
Acid-stressed (6 h) TSB Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.65"0.03 3.48"0.04 3.33"0.00 3.43"0.10 3.35"0.08 3.33"0.15 3.42"0.08
2.58"0.08 2.37"0.03 1.71"0.05 2.00"1.11 1.47"0.01 2.23"0.04 1.84"0.37
1.52"0.49 0.33"0.00 0.74"0.43 0.35"1.14 0.03"0.92 y0.97"0.00 0.15"0.58
S. flexneri 2ar11
S. flexneri 2ar12
S. flexneri 6r6
S. flexneri 6r7
Mean increase for all Shigella strains
3.07"0.00 2.16"0.03 2.04"0.05 3.13"0.14 3.26"0.00 3.17"0.03 3.11"0.01 3.21"0.12 2.89"0.25
2.69"0.12 1.56"0.07 1.31"0.01 2.58"0.09 2.84"0.10 2.85"0.05 2.70"0.02 2.15"0.00 2.33"0.31
3.13"0.02 1.98"0.02 1.50"0.04 2.97"0.03 3.27"0.00 3.17"0.16 3.05"0.00 3.11"0.16 2.77"0.34
3.38"0.06 2.00"0.01 2.04"0.07 2.89"0.02 2.86"0.05 3.13"0.05 2.89"0.02 2.83"0.04 2.75"0.25
3.14"0.20 1.67"0.31 1.38"0.45 2.94"0.22 3.09"0.23 2.80"0.36 2.90"0.23 2.74"0.31
2.40"0.09 2.40"0.08 1.81"0.01 2.50"0.04 2.24"0.04 2.16"0.04 2.25"0.15
2.03"0.01 2.07"0.04 1.36"0.04 0.50"0.02 1.89"0.02 1.75"0.02 1.60"0.36
2.48"0.04 2.44"0.07 1.98"0.03 1.94"0.12 2.18"0.07 1.89"0.14 2.15"0.16
2.49"0.00 2.53"0.01 1.81"0.02 2.40"0.01 2.35"0.01 2.54"0.04 2.35"0.17
2.67"0.01 2.20"0.03 1.54"0.14 2.28"0.07 2.36"0.00 2.33"0.03 2.23"0.23
2.56"0.23 2.44"0.19 1.99"0.29 2.15"0.40 2.34"0.24 2.23"0.21
2.17"0.02 2.01"0.01 1.58"0.01 2.00"0.04 1.73"0.04 1.74"0.00 1.87"0.14
0.22"0.08 0.29"0.07 0.14"0.03 y0.01"0.07 y0.33"0.09 y0.49"0.14 0.03"0.20
2.90"0.02 2.75"0.01 2.81"0.03 2.73"0.04 2.86"0.01 2.18"0.02 2.71"0.16
0.72"0.01 0.74"0.02 0.59"0.07 0.28"0.02 0.59"0.07 y0.30"0.23 0.43"0.25
0.14"0.03 0.19"0.02 0.18"0.02 y0.09"0.06 0.19"0.03 y0.41"0.03 0.03"0.15
1.72"0.68 1.52"0.66 1.37"0.62 1.27"0.75 1.19"0.73 0.75"0.70
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
S. sonnei CIP
259
260 M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
Fig. 1. Outgrowth of non-shigellae competitive bacteria in various enrichment media: Ža. TSB, Žb. EEq1.0, Žc. EEq1.0–42, Žd. SBq0.5, Že. SBq3.0, Žf. SBq3.0–42.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
E. hermanii. Incubation at 42 8C instead of 37 8C also promoted inhibition of the competitive flora. Nevertheless, important competitors for shigellae in EE q 1.0–42 and SB q 3.0–42 are Citrobacter sp., Enterobacter sp., Klebsiella sp., E. coli, E. hermani, Salmonella sp. and ProÕidencia sp. 3.2. EÕaluation of isolation media Electivity and selectivity of the isolation media were determined by ecometric testing. Results for chill-stressed Shigella strains and the competitive flora are shown in Tables 2 and 3, respectively. Although SSA is generally thought to be a highly selective medium, it supports the growth of chillstressed shigellae well ŽRGI ) 0.8. while it suppresses the growth of important competitors such as Citrobacter sp. and E. coli. HEA is most inhibitory towards the competitive flora but 4r8 chill-stressed Shigella strains showed an RGI - 0.8. MAC supports growth of shigellae well Žonly one strain with RGI - 0.8. but the competitive flora is difficult to differentiate because this is solely based upon the fermentation of lactose. XLD is moderately selective towards Shigella Žtwo strains with RGI - 0.8. that enables part of the competitive flora to be inhibited, while the other competitors can be differentiated on the basis of colony characteristics. From the tested competitive flora, primarily Klebsiella sp., Enterobacter sp. Citrobacter sp., Salmonella sp. and also the Escherichia strains seem to be the most important competitors for the isolation of Shigella sp. from the selective media.
Table 2 Econometric test of Shigella sp. on Shigella isolation media Žchill-stressed culture. Shigella strains
Relative growth index Žisolation mediarTSA. MAC T7
S. sonnei CIP S. sonnei 1 S. sonnei 2 S. flexneri CIP S. flexneri 2ar11 S. flexneri 2ar12 S. flexneri 6r6 S. flexneri 6r7
DCA
XLD
SSA
4.5r5 4.5r5 3.5r5 5r5 4.5r5 4.5r5 4r5 4r5 5r5 5r5 4.5r4 3r4 0r4 1.5r4 4.5r4 3r5 3.5r5 5r5 5r5 5r5 4r5 3r5 5r5 5r5 4r5 4r5 5r5 2r5 3.5r5 5r5 5r5 4r5 5r5 5r5 4r5 4.5r5 5r5 5r5 5r5 4.5r5
261
Table 3 Econometric test of non-shigellae on the Shigella isolation media Žunstressed culture. Competitive flora
Relative growth index Žisolation mediarTSA. MAC T7 DCA XLD SSA HEA
Aeromonas sp. Citrobacter sp. Enterobacter sp. Erwinia sp. E. coli E. hermani Klebsiella sp. Proteus sp. ProÕidencia sp. Pseudomonas sp. Salmonella sp.
0r2 3.5r5 4r5 1r4 3r4 2.5r4 1.5r3 1r4 1.5r4 1r3 1.5r4
0.5r2 4.5r5 5r5 1r4 4r4 2r4 2.5r3 1r4 2.5r4 1r3 2r4
2r2 1r2 3r2 2.5r2 4r5 2.5r5 1.5r5 1.5r5 3r5 5r5 4r5 5r5 1r4 2r4 4r4 1r4 0r4 0r4 0r4 0r4 1.5r4 0r4 3r4 0r4 2.5r3 3.5r3 2.5r3 1.5r3 0r4 0r4 0r4 0r4 2.5r4 1r4 1.5r4 1r4 1r3 1.5r3 1r3 1r3 1.5r4 3r4 3.5r4 1.5r4
3.3. Effect of non-shigellae on the enrichment and isolation of Shigella sp. From the results mentioned above, SB q 3.0 and EE q 1.0 incubated at 42 8C and MAC, XLD, SSA and HEA incubated at 37 8C were selected as enrichment and isolation media, respectively for further testing. Shigella sp. and non-shigellae were inoculated at different ratios in the enrichment media and subsequently plated on the four isolation media ŽTable 4.. It can be noticed that S. sonnei is a better competitor than S. flexneri. In addition, the CIP strains of S. sonnei and S. flexneri showed better recovery than the clinical isolates. S. flexneri 2ar12 was never recovered while S. flexneri CIP was only isolated after enrichment in SB q 3.0 incubated at 42 8C. The superiority of SB q 3.0 over EE q 1.0 as enrichment medium was also confirmed with the S. sonnei 1 inoculum. No major differences in the performance of the four selected isolation media were observed.
HEA 3.5r5 4.5r5 1r4 4.5r5 3.5r5 4.5r5 3.5r5 4r5
3.4. Determination of the effectiÕeness of enrichment in SB q 3.0 incubated at 42 8C for the recoÕery of S. sonnei and S. flexneri from foods Results are shown in Table 5. The present experiment confirmed better resistance of S. sonnei than S. flexneri towards stress imposed by the food and its indigenous flora. S. flexneri was suppressed either
262
SBq3.0 Ž42 8C. MAC
XLD
SSA
HEA
EEq1.0 Ž42 8C. MAC
XLD
SSA
HEA
S. sonnei CIP Control 1:1 1:10 1:100
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh.rCF CFqSh.rCF CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh.rCF CFqSh.rCF CF Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
S. sonnei 1 Control 10:1 1:1 1:10 1:100
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCF CFrCFqSh. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
S. flexneri CIP Control 1:1 1:10 1:100
Sh. Ž2=. CFrCFqSh. CFrCFqSh. CFrCFqSh.
Sh. Ž2=. CFrCFqSh. CFrCFqSh. CFrCFqSh.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
S. flexneri 2ar12 Control 10:1 1:1 1:10 1:100
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Shigella strain Ž Shigellar non-shigellae.
Sh.: isolation of Shigella sp.; CF: competitive flora on the isolation medium; –: not determined.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
Table 4 Effect of non-shigellae on the enrichment and isolation of Shigella sp.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
263
Table 5 Determination of the effectiveness of SB q 3.0 Ž42 8C. for the recovery of S. sonnei and S. flexneri from foods Food matrix
Chicken salad ŽpH 4.70.
Potato salad ŽpH 4.68.
Iceberg lettuce ŽpH 6.32.
Chinese cabbage ŽpH 5.69.
Strawberries ŽpH 3.83.
Scampis ŽpH 7.39.
Inoculum Žcfurg.
control 5–50 10–100 50–500 control 5–50 10–100 control 5–50 10–100 control 5–50 10–100 control 5–50 10–100 50–500 control 5–50 10–100 50–500
S. sonnei 1
S. flexneri 2ar12
MAC
XLD
SSA
MAC
XLD
SSA
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF Sh.q CF Sh.q CF Sh.q CF
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF Sh.q CF Sh.q CF Sh.q CF
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF CF Sh.q CF Sh.q CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
Sh.: isolation of Shigella sp.; CF: competitive flora on the isolation medium; –: not determined.
by acid foods Žchicken salad, potato salad, strawberries. or by the competitive flora in the food, which could grow and develop in the enrichment and on the isolation media Žiceberg lettuce, Chinese cabbage, scampi.. S. sonnei was also suppressed in some cases by acid Žstrawberries. or competitive flora ŽChinese cabbage, scampi.; however, it was isolated from chicken salad, potato salad and scampi using an inoculum of 10–100 cfurg.
4. Discussion The classical techniques and media used for detection, isolation and enumeration of shigellae are neither very specific nor very sensitive. Presence of injured cells due to sublethal stress in foods is common. In the present study, the performance of the usual enrichment and isolation media was tested using both non-stressed and acid- and chill-stressed shigellae. It was noted that even for non-stressed shigellae, the GN broth supported significant lower outgrowth of shigellae than SB and EE broth. Never-
theless, GN is the recommended enrichment medium in the NMKL Ž1995. method no. 151. Smith and Dell Ž1990. reported GN broth to be more inhibitory in the repair of heat-injured cells compared to SB. GN broth contains sodium deoxycholate. Sodium deoxycholate and also bile salts are selective agents that are commonly included as selective agents in culture media for Enterobacteriaceae. However, both sodium deoxycholate and bile salts were shown to be inhibitory to heat-stressed shigellae ŽTollison and Johnson 1985.. Shigella broth, the recommend medium by the US FDA, contains neither sodium deoxycholate nor bile salts. Although EE broth contains bile salts, resuscitation of shigellae was not negatively influenced. It was noted that the different species of shigellae vary in their ability to grow on any particular medium. This means that the characteristics of the Shigella strain are more important than the type of selective enrichment medium used to recover shigellae from a stressing environment like foods. Although the US FDA Ž1992. BAM method recommends addition of only 0.5 mg novobiocinrml to
264
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
SB for S. sonnei and 3.0 mg novobiocinrml for other shigellae, no significant differences in multiplication of unstressed cultures of either S. sonnei or S. flexneri were noticed for the two concentrations of novobiocin tested. If stressed Shigella cultures were used, increase in numbers in SB q 0.5 was in general higher than in SB q 3.0, but this was observed for both the S. sonnei and the S. flexneri strains. Various members of the Enterobacteriaceae antagonize Shigella spp. growth. S. flexneri did not compete well with strains of Klebsiella, Proteus or E. coli in vitro ŽSmith, 1987.. None of the tested enrichment media ŽSB or EE broth. were significantly superior to others for suppressing competitive flora. On the basis of the present study, a combination of MAC, XLD and SSA was preferred for isolation of Shigella spp. HEA is confirmed to be a highly selective medium but can fail to detect stressed shigellae. Smith and Buchanan Ž1992. also particularly recommended use of MAC, XLD and SSA agars. The competition with non-shigellae can negatively influence the detection of shigellae from food, depending on the composition of the competitive flora and the numbers present. Isolation of shigellae from different types of foods using SB q 3.0 mgrml novobiocin incubated at 42 8C and isolation on MAC, XLD and SSA was negatively influenced either by presence of suboptimal intrinsic factors of the food Že.g. low pH. and high level of competitive flora. Few studies are available on the performance of classical culture media for isolation of shigellae from foods. Rafii and Lunsford Ž1997. followed survival of S. flexneri in vegetables and commercially prepared salads, however, using high inoculation levels Ž10 5 –10 6 cfurg.. The recovery of S. sonnei at an inoculum level of 1–10 cfur25 g by the US FDA BAM culture method Ženrichment in SB q 0.5 for 20 h at 44 8C and isolation on MAC. from potato salad, chicken salad, cooked salad shrimp and lettuce was described by June et al. Ž1993., although the same authors found the US FDA BAM method to be relatively ineffective for the recovery of S. sonnei from raw ground beef and raw oysters. In the present study, it was found that, in general, S. sonnei has a better resistance towards stress by suboptimal intrinsic factors of the food and the competitive flora than S. flexneri.
The reliable detection of shigellae in foods is rather difficult. The current detection methods are reasonably effective for S. sonnei. However, an effective detection procedure for all shigellae is not yet available. PCR offers an alternative for detection of shigellae in foods. The detection of a low inoculation level Ž10 cfurg. of S. flexneri in iceberg lettuce, milk, shrimps and blue cheese using nested PCR was demonstrated by Lindqvist Ž1999.. Analysis of enriched samples with a single PCR may sometimes be sufficient for successful detection of Shigella sp. in the food. However, if only a single PCR is used, enrichment must lead to a sufficient increase in the number of Shigella bacteria to achieve the detection limit of the PCR reaction. This will depend upon the intrinsic attributes of the food matrix, its indigenous flora and the characteristics of the Shigella strain.
Acknowledgements Mieke Uyttendaele is grateful to the Fund of Scientific Research-Flanders for a position as Postdoctoral Research Assistant. Catherine Flore Bagam´ boula is grateful to Belgian Technical Cooperation for obtaining a PhD scholarship.
References Beckers, H.J., Soentoro, P.S.S., 1989. Method for detection of Shigella from foods. Zb. Bakt. Hyg. B 187, 261–265. de Boer, E., 1998. Update on media for isolation of Enterobacteriaceae from foods. Int. J. Food Microbiol. 45, 43–53. June, G.A., Sherrod, P.S., Amaguana, R.M., Andrews, W.H., Hammack, T.S., 1993. Effectiveness of the bacteriological analytical manual culture method for the recovery of Shigella sonnei from selected foods. J. AOAC Int. 76, 1240–1248. Lampel, K.A., Sandlin, R.C., Formal, S., 2000. Shigella: introduction and detection by classical cultural techniques. In: Robinson, R.K., Batt, C.A., Patel, P.D. ŽEds.., Encyclopedia of Food Microbiology. Academic Press, London, pp. 2015–2020. Lindqvist, R., 1999. Detection of Shigella spp. in food with a nested PCR method-sensitivity and performance compared with a conventional culture method. J. Appl. Microbiol. 86, 971–978. Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M., Tauxe, R.V., 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5, 1–21.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265 Mehlman, I.J., Romero, A., Wentz, B.A., 1985. Improved enrichment for recovery of Shigella sonnei from foods. J. AOAC 68, 552–555. Mossel, D.A.A., Jacobs-Reitsma, W.F., 1990. Grensreactie op Shigella. Microbiologisch Onderzoek van Levensmiddelen. Keesing Noordervliet, Zeist ŽNetherlands., pp. 94–95. Mossel, D.A.A., Corry, J.E.L., Struijk, C.B., Baird, R.M., 1995a. The ecometric monitoring technique. Essentials of the Microbiology of Foods. Wiley, Chichester ŽEngland., pp. 278– 281. Mossel, D.A.A., Corry, J.E.L., Struijk, C.B., Baird, R.M., 1995b. Shigellosis. Essentials of the Microbiology of Foods. Wiley, Chichester ŽEngland., p. 131. NKML, 1995. Shigella bacteria. Detection in foods Žproposed method., NMKL No. 151. Oslo, Norway: National Veterinary Institute, Department of Food and Feed Hygiene. Rafii, F., Lunsford, P., 1997. Survival and detection of Shigella
265
flexneri in vegetables and commercially prepared salads. J. AOAC Int. 80, 1191–1197. Smith, J.L., 1987. Shigella as a foodborne pathogen. J. Food Prot. 50, 788–801. Smith, J.L., Dell, B.J., 1990. Capability of selective media to detect heat-injured Shigella flexneri. J. Food Prot. 53, 141– 144. Smith, J.L., Buchanan, R.L., 1992. Shigella. In: Vanderzant, C., Splittstoesser, D.F. ŽEds.., Compendium of Methods for the Microbiological Examination of Foods. APHA, Washington, USA, pp. 423–431. Tollison, S.B., Johnson, M.G., 1985. Sensitivity to bile salts of Shigella flexneri sublethally heat stressed in buffer or broth. Appl. Environ. Microbiol. 50, 337–341. US Food and Drug Administration ŽUS FDA., 1992. Bacteriological Analytical Manual. 7th edn. AOAC International, Arlington, VA.
Evaluation of culture media for enrichment and isolation of Shigella sonnei and S. flexneri M. Uyttendaele ) , C.F. Bagamboula, E. De Smet, S. Van Wilder, J. Debevere Laboratory of Food Microbiology and Food PreserÕation, Department of Food, Technology and Nutrition, UniÕersity of Gent, Coupure Links 653, 9000 Ghent, Belgium Received 10 October 2000; received in revised form 22 April 2001; accepted 18 May 2001
Abstract The performance of Gram-negative ŽGN. broth with Ž10 mgrml. and without novobiocin, Shigella broth ŽSB. with 0.5 and 3.0 mgrml novobiocin, all incubated at 37 8C ŽSB with 3.0 mgrml novobiocin also at 42 8C. and buffered brilliant green bile glucose ŽEE. broth with 1.0 mgrml novobiocin incubated at 37 and 42 8C were evaluated for resuscitation and growth of Shigella sonnei and S. flexneri Žeight strains; unstressed, chill-stressed and acid-stressed. and non-shigellae Ž11 strains.. GN broth with or without novobiocin supported significantly less growth of Shigella sp. No significant differences in growth of shigellae were obtained between the other culture media. Performance depended more on the Shigella strain used. None of the tested media were significantly superior for suppressing the competitive flora. Electivity and selectivity of MacConkey agar ŽMAC., tergitol-7 agar ŽT7., desoxycholate citrate agar ŽDCA., xylose lysine desoxycholate agar ŽXLD., Salmonella Shigella agar and Hektoen enteric agar ŽHEA. were determined by ecometric testing. HEA confirmed to be a high selective medium for both non-shigellae and stressed Shigella sp. Klebsiella sp., Enterobacter sp., Citrobacter sp., Salmonella sp. and the Escherichia strains can mask the presence of shigellae. In vitro competition experiments and experiments with artificially contaminated foods showed higher resistance of S. sonnei than S. flexneri towards the stress imposed by the food matrix and its indigenous flora. Reliable detection, however, of shigellae in foods with the current enrichment and isolation media was not achieved. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Shigella sonnei; Shigella flexneri; Detection; Isolation; Enrichment
1. Introduction
)
Corresponding author. Tel.: q32-9-264-6178; fax: q32-9225-5510. E-mail address: [email protected] ŽM. Uyttendaele..
Shigella spp. are the causative agents of bacillary dysentery Žshigellosis.. Although the number of reported cases of shigellosis arising from food is much lower than that of salmonellosis or campylobacteriosis, it is an important cause of foodborne bacterial infection in developed countries ŽMead et al., 1999..
0168-1605r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 Ž 0 1 . 0 0 5 4 9 - 9
256
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
The species involved are mostly Shigella sonnei Žabout 70%. and S. flexneri Žapproximately 25%. with the other types seldom implicated ŽMossel et al., 1995b.. The foods most often implicated are potato salads and salads containing chicken or fish, raw vegetables and seafoods. The contamination usually occurs due to poor personal hygiene by an infected food handler ŽLampel et al., 2000; Smith and Buchanan, 1992.. The reliable detection of shigellae in foods is rather difficult. They are easily overgrown by other bacteria present in the food. In particular, acids produced by other Enterobacteriaceae during the fermentation of carbohydrates are toxic for shigellae ŽMehlman et al., 1985.. An effective selective enrichment procedure for all shigellae is not yet available. The current culture method in the US FDA Ž1992. bacteriological analytical manual ŽBAM. selects for Shigella by enrichment in a low carbohydrate medium Ž Shigella broth. and incubating in an anaerobic atmosphere at an elevated temperature in the presence of the antibiotic novobiocin Ž0.5 mgrml for S. sonnei, 3.0 mgrml for other shigellae. ŽJune et al., 1993; Lampel et al., 2000.. Other enrichment broths are recommended for isolation of shigellae, such as Gram-negative ŽGN. broth with novobiocin Ž10 mgrml. ŽBeckers and Soentoro, 1989. or without novobiocin ŽNKML, 1995. and enrichment in brilliant green bile glucose ŽEE. broth with 1 mgrml novobiocin ŽMossel and Jacobs-Reitsma, 1990.. Many culture media for isolation of shigellae from foods have been suggested: solid media with a low ŽMacConkey agar ŽMAC. and tergitol-7 ŽT7.., intermediate Žxylose lysine desoxycholate agar ŽXLD. and desoxycholate citrate agar ŽDCA.. or high selectivity Ž Salmonella Shigella agar ŽSSA. and Hektoen enteric agar ŽHEA.. can be used Žde Boer, 1998; Lampel et al., 2000.. At least two to three different types of selective agar media should be used to increase the chance of recovery of shigellae from foods. These detection methods are recommended in different regions of the world but have never been extensively compared and evaluated in their performance for isolation of shigellae from foods. The present study was undertaken to provide data on the electivity and selectivity of the culture media for
enrichment and isolation of S. sonnei and S. flexneri using unstressed, chill-stressed and acid-stressed cells.
2. Material and methods 2.1. Bacterial strains S. sonnei CIP8249 and S. flexneri CIP8248 were obtained from the Institut Pasteur-Paris ŽFrance.. In addition, two clinical isolates of S. sonnei Žstrains 1 and 2. and four clinical isolates of S. flexneri Žtwo of serotype six Žstrains 6 and 7., two of serotype 2a Žstrains 11 and 12. were obtained from the Scientific Institute of Public Health ŽIPH. in Brussels ŽBelgium.. Eleven non-shigella strains were also included in the study. Aeromonas hydrophila LMG 3771, Citrobacter freundii LMG 3246, Enterobacter aerogenes LMG 2094, Escherichia coli LMG 8223, E. hermanii LMG 7867, Klebsiella oxytoca LMG 3055, Salmonella enteritidis LMG 10395, Proteus Õulgaris LMG 2096, ProÕidencia rettgeri LMG 3259 and Pseudomonas fluorescens LMG 1794 were obtained from the LMGrBCCMe Culture Collection ŽUniversity of Gent, Ghent, Belgium.. Erwinia carotoÕora was an own isolate obtained from vegetables. 2.2. Culture media Tryptone soy broth ŽTSB; Oxoıd, ¨ Basingstoke, Hampshire, England. was used as a control medium. Selective enrichment media included in the study were Gram-negative ŽGN. broth according to Hajna Žin Beckers and Soentoro, 1989. with 10 mgrml novobiocin ŽSigma, St. Louis, MO. ŽGN q 10.0. and without novobiocin ŽGN., Shigella broth prepared according to the US FDA Ž1992. BAM with 0.5 mgrml novobiocin ŽSB q 0.5. or 3.0 mgrml novobiocin ŽSB q 3.0. and EE broth ŽOxoıd ¨ . with 1.0 mgrml novobiocin ŽEE q 1.0.. Plating media included in the study were tryptone soy agar ŽTSA; Oxoid. as a control medium and the selective media: MacConkey agar ŽMAC; Merck, Darmstadt, Germany., tergitol-7 agar ŽT7; Oxoıd ¨ ., desoxycholate citrate agar ŽDCA; Diagnostics Pasteur, Marnes la Coquette, France., xylose lysine
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
desoxycholate agar ŽXLD; Oxoıd ¨ ., Salmonella Shigella agar ŽSSA; Oxoıd ¨ ., Hektoen enteric agar ŽHEA; Diagnostics Pasteur.. 2.3. Inoculum From a stock culture, a subculture was placed in brain heart infusion ŽBHI; Oxoıd ¨ . broth and incubated for 18–20 h at 37 8C Žunstressed culture.. For the chill-stress, this BHI culture was stored for a further 6 days at 4 8C. For the acid-stress, 1 ml of an 8-h culture in BHI at 37 8C was transferred to BHI acidified to pH 4.0 with HCl and incubated for 16–18 h at 37 8C. 2.4. EÕaluation of enrichment media A 0.1-ml volume of appropriate dilutions of the BHI cultures of shigellae Žunstressed, chill-stressed and acid-stressed. and non-shigella species were inoculated into 10 ml of the various enrichment media Žin duplicate. to obtain approximately 10 3 cfurml. Dilutions of the BHI cultures were plated on TSA ŽOxoıd ¨ . to determine the exact inoculum. After incubation for 4 h Žunstressed and chill-stressed inoculum. or 6 h Žacid-stressed inoculum. at 37 8C for GN, GN q 10.0, SB q 0.5, SB q 3.0 and EE q 1.0 and incubation at 42 8C for SB q 3.0 and EE q 1.0, numbers of colony-forming units were determined by spiral plating of appropriate dilutions on TSA, counting colonies after 24 h of incubation at 37 8C. 2.5. EÕaluation of isolation media The various selective media were tested for their productivity and selectivity towards Shigella spp. and non-shigellae by ecometric evaluation. Tryptone soy agar ŽTSA, Oxoid. was used as the control medium. Ecometric evaluation relies on inoculating activated cultures on solid media applying a carefully standardized, sequential streaking technique to ever decreasing numbers of colonies ŽMossel et al., 1995a.. Unstressed cultures of Shigella sp. and nonshigellae as well as chill-stressed cultures of Shigella sp. Žapproximately 10 8 cfurml. were inoculated on TSA, MAC, T7, XLD, DCA, HEA and SSA. The streaking was done on 9-cm diameter petri dishes.
257
The four quadrants were inoculated in succession with a series of five parallel lines with a single loop, the first one at the perimeter. Finally, a last line was drawn through the centre. For the inoculation, 1-ml plastic disposable loops were used ŽNunc, Life Technologies, Merelbeke, Belgium.. After incubation for 24 h at 37 8C, the final streak that led to substantial colony formation was assessed. The reading was expressed in absolute growth index ŽAGI. as explained by Mossel et al. Ž1995a,b.. The results were then converted to a relative growth index ŽRGI. defined as the ratio of the AGIs on the test medium to those on the reference medium ŽTSA.. Additionally, the size and colour of the colonies on the different media were examined. For each of the test organisms, this procedure was performed in duplicate. 2.6. Effect of non-shigellae on the enrichment and isolation of Shigella sp. EE q 1.0 and SB q 3.0 were inoculated with appropriate dilutions of an unstressed culture of Shigella sp. and a mixture of non-shigella competitors Ž E. coli, E. hermanii, Ent. aerogenes, C. freundii and K. oxytoca. to obtain a ratio of Shigella sp.rnon-shigella of 10:1, 1:1, 1:10 and 1:100 using a Shigella inoculum of 10 3 cfurml. A control with shigellae with no competitive flora was also included. The inoculated enrichment media were incubated for 18–20 h at 42 8C, before streaking onto MAC, XLD, SSA and HEA, incubating for a further 24 h at 37 8C. Typical colonies were streaked for purity on nutrient agar ŽNA; Oxoıd ¨ . and further tested for fermentation of glucose and lactose, H 2 S and indole production and motility by inoculation of KIA and SIM ŽOxoıd ¨ . tubes. The experiment was performed in duplicate for S. sonnei CIP8249, S. flexneri CIP8248, S. sonnei 1, and S. flexneri 2ar12. 2.7. Determination of the effectiÕeness of the selected culture method for the recoÕery of S. sonnei and S. flexneri from foods Samples Ž25 g. of six foods Žchicken salad, potato salad, pre-cut iceberg lettuce, Chinese cabbage, strawberries and scampi. were weighed into sterile
258
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
stomacher bags Žfour portions of each food. and inoculated with appropriate dilutions of an unstressed BHI culture of S. sonnei 1 or S. flexneri 2ar12 to obtain ca. 5–50, 10–100, 50–500 and no Žcontrol. cfurg. Dilutions of the BHI culture were plated on TSA to determine the exact inoculum. The pH was measured with an Ingold electrode ŽMGDX K57, Urdorf. connected to a Knick pH meter Žtype 763, Knickn, Berlin.. The inoculated food was kept at 4 8C for 6 h before 225 ml SB q 3.0 were added to each food sample. The content was homogenized and incubated for 18–20 h at 42 8C, streaked onto MAC, XLD and SSA after enrichment, then incubated for 24 h at 37 8C. Typical colonies were streaked for purity on NA and further tested as described above.
3. Results 3.1. EÕaluation of enrichment media The performance of the enrichment media was evaluated by determination of the increase or decrease in numbers of colony-forming units of Shigella spp. and non-shigella after a short incubation period in the selective broths. The incubation period was restricted to cover the lag phase and the onset of the exponential growth but avoiding multiplication to the stationary growth phase. Enumeration after a short incubation period enabled the response of the bacterial strains to stress exerted by the selective broth andror the ability of these media to resuscitate chillor acid-damaged shigellae to be determined. Results for the various S. sonnei and S. flexneri strains Žunstressed, chill- and acid-stressed. in the various enrichment media are shown in Table 1. Results were statistically treated ŽScheffe analysis by SPSS for Windows. to determine the mean increase in numbers over all Shigella strains for each enrichment medium included in the study and to determine the mean increase in numbers for each of the Shigella strains over all enrichment media tested. GN and GN q 10.0 showed a significantly lower outgrowth, respectively 1.7 and 1.4 log units, of unstressed shigellae than TSB Ž3.2 log. or the other enrichment media Ž2.7–3.1 log.. The GN media
were not used in subsequent experiments with stressed Shigella sp. and non-shigellae. If Shigella spp. were subjected to a chill-stress, the increase in numbers of shigellae Ž2.6 log in TSB and 2.0–2.4 log in the other media. was lower compared to unstressed Shigella spp. Prior exposure to acid Ž16–20 h at pH 4.0. prolonged the lag time of the shigellae and extension of the incubation time to 6 h was necessary to establish multiplication in the enrichment media Ž1.7 log in TSB and 0.7–1.5 log in the selective enrichment media. although numbers were still lower than those achieved after 4 h with unstressed or chill-stressed Shigella spp. No significant differences in increase in numbers for the stressed shigellae were obtained between the different enrichment media tested. Only with acidstressed shigellae in EE q 1.0 and EE q 1.0–42 and with chill-stressed shigellae in SB q 3.0 was outgrowth reduced by more than 0.5 log unit compared to TSB ŽTable 1.. Variation in the performance of the media depended more upon the Shigella strain used than on the enrichment media or incubation temperature ŽTable 1. and this was more pronounced if the strains were stressed. While outgrowth of some strains was unaffected or only slightly affected by prior environmental stress, others showed significantly lower or no increase in numbers independent of the selective enrichment medium applied. Based upon the results obtained, strains can be classified as stress resistant, e.g. S. sonnei CIP8249, S. sonnei 1, S. flexneri CIP8248, S. flexneri 2ar12 or stress sensitive, e.g. S. sonnei 2, S. flexneri 2ar11, S. flexneri 6r6, S. flexneri 6r7. The increase in numbers of competitive flora in the various enrichment media was determined ŽFig. 1.. The incubation temperature rather than the selective agents was probably more inhibitory or lethal for the competitive flora. Reduction of numbers of P. fluorescens was probably due to its inability to grow at 37 8C. Similarly, 42 8C was lethal for Aeromonas and Erwinia. Inhibition of the majority of the competitive flora was accomplished by the selective enrichment media; however, none of the tested media were significantly superior than the others for suppressing competitive flora. EE q 1.0 at 37 8C suppressed growth of Proteus sp. fairly good, while SB q 3.0 at 37 8C suppressed multiplication of
Table 1 Change in numbers Žmean log 10 cfurml of duplicate determinations. of pure cultures of Shigella strains Žunstressed, chill- or acid-stressed. in the various enrichment media, counting on tryptone soya agar S. sonnei 1
S. sonnei 2
S. flexneri CIP
Unstressed (4 h) TSB Ž37 8C. GN Ž37 8C. GNq10.0 Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.74"0.06 1.51"0.30 2.02"0.07 3.47"0.24 3.48"0.02 3.86"0.08 3.57"0.06 3.73"0.02 3.16"0.47
2.58"0.06 2.05"0.04 1.38"0.02 2.32"0.46 2.32"0.11 2.48"0.08 2.05"0.21 2.23"0.38 2.15"0.21
3.15"0.05 0.38"0.28 1.46"0.34 3.04"0.09 2.96"0.02 1.71"0.04 3.05"0.01 2.24"0.15 2.23"0.53
3.38"0.20 1.78"0.10 y0.63"0.16 3.36"0.16 3.76"0.03 2.07"0.04 2.85"0.11 2.51"0.12 2.38"0.72
Chill-stressed (4 h) TSB Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.13"0.04 2.89"0.03 2.87"0.11 3.21"0.04 2.94"0.08 3.05"0.01 3.01"0.09
2.05"0.03 1.98"0.08 1.79"0.00 2.0"0.03 1.71"0.06 1.98"0.01 1.92"0.08
3.25"0.04 3.05"0.03 2.81"0.10 2.40"0.06 3.06"0.04 2.20"0.03 2.80"0.25
Acid-stressed (6 h) TSB Ž37 8C. SBq0.5 Ž37 8C. SBq3.0 Ž37 8C. SBq3.0 Ž42 8C. EEq1.0 Ž37 8C. EEq1.0 Ž42 8C. Mean increase for all enrichment media
3.65"0.03 3.48"0.04 3.33"0.00 3.43"0.10 3.35"0.08 3.33"0.15 3.42"0.08
2.58"0.08 2.37"0.03 1.71"0.05 2.00"1.11 1.47"0.01 2.23"0.04 1.84"0.37
1.52"0.49 0.33"0.00 0.74"0.43 0.35"1.14 0.03"0.92 y0.97"0.00 0.15"0.58
S. flexneri 2ar11
S. flexneri 2ar12
S. flexneri 6r6
S. flexneri 6r7
Mean increase for all Shigella strains
3.07"0.00 2.16"0.03 2.04"0.05 3.13"0.14 3.26"0.00 3.17"0.03 3.11"0.01 3.21"0.12 2.89"0.25
2.69"0.12 1.56"0.07 1.31"0.01 2.58"0.09 2.84"0.10 2.85"0.05 2.70"0.02 2.15"0.00 2.33"0.31
3.13"0.02 1.98"0.02 1.50"0.04 2.97"0.03 3.27"0.00 3.17"0.16 3.05"0.00 3.11"0.16 2.77"0.34
3.38"0.06 2.00"0.01 2.04"0.07 2.89"0.02 2.86"0.05 3.13"0.05 2.89"0.02 2.83"0.04 2.75"0.25
3.14"0.20 1.67"0.31 1.38"0.45 2.94"0.22 3.09"0.23 2.80"0.36 2.90"0.23 2.74"0.31
2.40"0.09 2.40"0.08 1.81"0.01 2.50"0.04 2.24"0.04 2.16"0.04 2.25"0.15
2.03"0.01 2.07"0.04 1.36"0.04 0.50"0.02 1.89"0.02 1.75"0.02 1.60"0.36
2.48"0.04 2.44"0.07 1.98"0.03 1.94"0.12 2.18"0.07 1.89"0.14 2.15"0.16
2.49"0.00 2.53"0.01 1.81"0.02 2.40"0.01 2.35"0.01 2.54"0.04 2.35"0.17
2.67"0.01 2.20"0.03 1.54"0.14 2.28"0.07 2.36"0.00 2.33"0.03 2.23"0.23
2.56"0.23 2.44"0.19 1.99"0.29 2.15"0.40 2.34"0.24 2.23"0.21
2.17"0.02 2.01"0.01 1.58"0.01 2.00"0.04 1.73"0.04 1.74"0.00 1.87"0.14
0.22"0.08 0.29"0.07 0.14"0.03 y0.01"0.07 y0.33"0.09 y0.49"0.14 0.03"0.20
2.90"0.02 2.75"0.01 2.81"0.03 2.73"0.04 2.86"0.01 2.18"0.02 2.71"0.16
0.72"0.01 0.74"0.02 0.59"0.07 0.28"0.02 0.59"0.07 y0.30"0.23 0.43"0.25
0.14"0.03 0.19"0.02 0.18"0.02 y0.09"0.06 0.19"0.03 y0.41"0.03 0.03"0.15
1.72"0.68 1.52"0.66 1.37"0.62 1.27"0.75 1.19"0.73 0.75"0.70
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
S. sonnei CIP
259
260 M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
Fig. 1. Outgrowth of non-shigellae competitive bacteria in various enrichment media: Ža. TSB, Žb. EEq1.0, Žc. EEq1.0–42, Žd. SBq0.5, Že. SBq3.0, Žf. SBq3.0–42.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
E. hermanii. Incubation at 42 8C instead of 37 8C also promoted inhibition of the competitive flora. Nevertheless, important competitors for shigellae in EE q 1.0–42 and SB q 3.0–42 are Citrobacter sp., Enterobacter sp., Klebsiella sp., E. coli, E. hermani, Salmonella sp. and ProÕidencia sp. 3.2. EÕaluation of isolation media Electivity and selectivity of the isolation media were determined by ecometric testing. Results for chill-stressed Shigella strains and the competitive flora are shown in Tables 2 and 3, respectively. Although SSA is generally thought to be a highly selective medium, it supports the growth of chillstressed shigellae well ŽRGI ) 0.8. while it suppresses the growth of important competitors such as Citrobacter sp. and E. coli. HEA is most inhibitory towards the competitive flora but 4r8 chill-stressed Shigella strains showed an RGI - 0.8. MAC supports growth of shigellae well Žonly one strain with RGI - 0.8. but the competitive flora is difficult to differentiate because this is solely based upon the fermentation of lactose. XLD is moderately selective towards Shigella Žtwo strains with RGI - 0.8. that enables part of the competitive flora to be inhibited, while the other competitors can be differentiated on the basis of colony characteristics. From the tested competitive flora, primarily Klebsiella sp., Enterobacter sp. Citrobacter sp., Salmonella sp. and also the Escherichia strains seem to be the most important competitors for the isolation of Shigella sp. from the selective media.
Table 2 Econometric test of Shigella sp. on Shigella isolation media Žchill-stressed culture. Shigella strains
Relative growth index Žisolation mediarTSA. MAC T7
S. sonnei CIP S. sonnei 1 S. sonnei 2 S. flexneri CIP S. flexneri 2ar11 S. flexneri 2ar12 S. flexneri 6r6 S. flexneri 6r7
DCA
XLD
SSA
4.5r5 4.5r5 3.5r5 5r5 4.5r5 4.5r5 4r5 4r5 5r5 5r5 4.5r4 3r4 0r4 1.5r4 4.5r4 3r5 3.5r5 5r5 5r5 5r5 4r5 3r5 5r5 5r5 4r5 4r5 5r5 2r5 3.5r5 5r5 5r5 4r5 5r5 5r5 4r5 4.5r5 5r5 5r5 5r5 4.5r5
261
Table 3 Econometric test of non-shigellae on the Shigella isolation media Žunstressed culture. Competitive flora
Relative growth index Žisolation mediarTSA. MAC T7 DCA XLD SSA HEA
Aeromonas sp. Citrobacter sp. Enterobacter sp. Erwinia sp. E. coli E. hermani Klebsiella sp. Proteus sp. ProÕidencia sp. Pseudomonas sp. Salmonella sp.
0r2 3.5r5 4r5 1r4 3r4 2.5r4 1.5r3 1r4 1.5r4 1r3 1.5r4
0.5r2 4.5r5 5r5 1r4 4r4 2r4 2.5r3 1r4 2.5r4 1r3 2r4
2r2 1r2 3r2 2.5r2 4r5 2.5r5 1.5r5 1.5r5 3r5 5r5 4r5 5r5 1r4 2r4 4r4 1r4 0r4 0r4 0r4 0r4 1.5r4 0r4 3r4 0r4 2.5r3 3.5r3 2.5r3 1.5r3 0r4 0r4 0r4 0r4 2.5r4 1r4 1.5r4 1r4 1r3 1.5r3 1r3 1r3 1.5r4 3r4 3.5r4 1.5r4
3.3. Effect of non-shigellae on the enrichment and isolation of Shigella sp. From the results mentioned above, SB q 3.0 and EE q 1.0 incubated at 42 8C and MAC, XLD, SSA and HEA incubated at 37 8C were selected as enrichment and isolation media, respectively for further testing. Shigella sp. and non-shigellae were inoculated at different ratios in the enrichment media and subsequently plated on the four isolation media ŽTable 4.. It can be noticed that S. sonnei is a better competitor than S. flexneri. In addition, the CIP strains of S. sonnei and S. flexneri showed better recovery than the clinical isolates. S. flexneri 2ar12 was never recovered while S. flexneri CIP was only isolated after enrichment in SB q 3.0 incubated at 42 8C. The superiority of SB q 3.0 over EE q 1.0 as enrichment medium was also confirmed with the S. sonnei 1 inoculum. No major differences in the performance of the four selected isolation media were observed.
HEA 3.5r5 4.5r5 1r4 4.5r5 3.5r5 4.5r5 3.5r5 4r5
3.4. Determination of the effectiÕeness of enrichment in SB q 3.0 incubated at 42 8C for the recoÕery of S. sonnei and S. flexneri from foods Results are shown in Table 5. The present experiment confirmed better resistance of S. sonnei than S. flexneri towards stress imposed by the food and its indigenous flora. S. flexneri was suppressed either
262
SBq3.0 Ž42 8C. MAC
XLD
SSA
HEA
EEq1.0 Ž42 8C. MAC
XLD
SSA
HEA
S. sonnei CIP Control 1:1 1:10 1:100
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh.rCF CFqSh.rCF CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh.rCF CFqSh.rCF CF Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
S. sonnei 1 Control 10:1 1:1 1:10 1:100
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCF CFrCFqSh. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CFrCFqSh. CFrCFqSh. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCFqSh. CF Ž2=. CF Ž2=. CFr–
S. flexneri CIP Control 1:1 1:10 1:100
Sh. Ž2=. CFrCFqSh. CFrCFqSh. CFrCFqSh.
Sh. Ž2=. CFrCFqSh. CFrCFqSh. CFrCFqSh.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=. CFqSh. Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
Sh. Ž2=. CF Ž2=. CF Ž2=. CF Ž2=.
S. flexneri 2ar12 Control 10:1 1:1 1:10 1:100
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Sh. Ž2=. –rCF CF Ž2=. CF Ž2=. CFr–
Shigella strain Ž Shigellar non-shigellae.
Sh.: isolation of Shigella sp.; CF: competitive flora on the isolation medium; –: not determined.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
Table 4 Effect of non-shigellae on the enrichment and isolation of Shigella sp.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
263
Table 5 Determination of the effectiveness of SB q 3.0 Ž42 8C. for the recovery of S. sonnei and S. flexneri from foods Food matrix
Chicken salad ŽpH 4.70.
Potato salad ŽpH 4.68.
Iceberg lettuce ŽpH 6.32.
Chinese cabbage ŽpH 5.69.
Strawberries ŽpH 3.83.
Scampis ŽpH 7.39.
Inoculum Žcfurg.
control 5–50 10–100 50–500 control 5–50 10–100 control 5–50 10–100 control 5–50 10–100 control 5–50 10–100 50–500 control 5–50 10–100 50–500
S. sonnei 1
S. flexneri 2ar12
MAC
XLD
SSA
MAC
XLD
SSA
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF Sh.q CF Sh.q CF Sh.q CF
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF Sh.q CF Sh.q CF Sh.q CF
– Sh. Sh. Sh. – – Sh. – – Sh. CF CF CF – – – – CF CF Sh.q CF Sh.q CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
– – – – – – – CF CF CF CF CF CF – – – – CF CF CF CF
Sh.: isolation of Shigella sp.; CF: competitive flora on the isolation medium; –: not determined.
by acid foods Žchicken salad, potato salad, strawberries. or by the competitive flora in the food, which could grow and develop in the enrichment and on the isolation media Žiceberg lettuce, Chinese cabbage, scampi.. S. sonnei was also suppressed in some cases by acid Žstrawberries. or competitive flora ŽChinese cabbage, scampi.; however, it was isolated from chicken salad, potato salad and scampi using an inoculum of 10–100 cfurg.
4. Discussion The classical techniques and media used for detection, isolation and enumeration of shigellae are neither very specific nor very sensitive. Presence of injured cells due to sublethal stress in foods is common. In the present study, the performance of the usual enrichment and isolation media was tested using both non-stressed and acid- and chill-stressed shigellae. It was noted that even for non-stressed shigellae, the GN broth supported significant lower outgrowth of shigellae than SB and EE broth. Never-
theless, GN is the recommended enrichment medium in the NMKL Ž1995. method no. 151. Smith and Dell Ž1990. reported GN broth to be more inhibitory in the repair of heat-injured cells compared to SB. GN broth contains sodium deoxycholate. Sodium deoxycholate and also bile salts are selective agents that are commonly included as selective agents in culture media for Enterobacteriaceae. However, both sodium deoxycholate and bile salts were shown to be inhibitory to heat-stressed shigellae ŽTollison and Johnson 1985.. Shigella broth, the recommend medium by the US FDA, contains neither sodium deoxycholate nor bile salts. Although EE broth contains bile salts, resuscitation of shigellae was not negatively influenced. It was noted that the different species of shigellae vary in their ability to grow on any particular medium. This means that the characteristics of the Shigella strain are more important than the type of selective enrichment medium used to recover shigellae from a stressing environment like foods. Although the US FDA Ž1992. BAM method recommends addition of only 0.5 mg novobiocinrml to
264
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265
SB for S. sonnei and 3.0 mg novobiocinrml for other shigellae, no significant differences in multiplication of unstressed cultures of either S. sonnei or S. flexneri were noticed for the two concentrations of novobiocin tested. If stressed Shigella cultures were used, increase in numbers in SB q 0.5 was in general higher than in SB q 3.0, but this was observed for both the S. sonnei and the S. flexneri strains. Various members of the Enterobacteriaceae antagonize Shigella spp. growth. S. flexneri did not compete well with strains of Klebsiella, Proteus or E. coli in vitro ŽSmith, 1987.. None of the tested enrichment media ŽSB or EE broth. were significantly superior to others for suppressing competitive flora. On the basis of the present study, a combination of MAC, XLD and SSA was preferred for isolation of Shigella spp. HEA is confirmed to be a highly selective medium but can fail to detect stressed shigellae. Smith and Buchanan Ž1992. also particularly recommended use of MAC, XLD and SSA agars. The competition with non-shigellae can negatively influence the detection of shigellae from food, depending on the composition of the competitive flora and the numbers present. Isolation of shigellae from different types of foods using SB q 3.0 mgrml novobiocin incubated at 42 8C and isolation on MAC, XLD and SSA was negatively influenced either by presence of suboptimal intrinsic factors of the food Že.g. low pH. and high level of competitive flora. Few studies are available on the performance of classical culture media for isolation of shigellae from foods. Rafii and Lunsford Ž1997. followed survival of S. flexneri in vegetables and commercially prepared salads, however, using high inoculation levels Ž10 5 –10 6 cfurg.. The recovery of S. sonnei at an inoculum level of 1–10 cfur25 g by the US FDA BAM culture method Ženrichment in SB q 0.5 for 20 h at 44 8C and isolation on MAC. from potato salad, chicken salad, cooked salad shrimp and lettuce was described by June et al. Ž1993., although the same authors found the US FDA BAM method to be relatively ineffective for the recovery of S. sonnei from raw ground beef and raw oysters. In the present study, it was found that, in general, S. sonnei has a better resistance towards stress by suboptimal intrinsic factors of the food and the competitive flora than S. flexneri.
The reliable detection of shigellae in foods is rather difficult. The current detection methods are reasonably effective for S. sonnei. However, an effective detection procedure for all shigellae is not yet available. PCR offers an alternative for detection of shigellae in foods. The detection of a low inoculation level Ž10 cfurg. of S. flexneri in iceberg lettuce, milk, shrimps and blue cheese using nested PCR was demonstrated by Lindqvist Ž1999.. Analysis of enriched samples with a single PCR may sometimes be sufficient for successful detection of Shigella sp. in the food. However, if only a single PCR is used, enrichment must lead to a sufficient increase in the number of Shigella bacteria to achieve the detection limit of the PCR reaction. This will depend upon the intrinsic attributes of the food matrix, its indigenous flora and the characteristics of the Shigella strain.
Acknowledgements Mieke Uyttendaele is grateful to the Fund of Scientific Research-Flanders for a position as Postdoctoral Research Assistant. Catherine Flore Bagam´ boula is grateful to Belgian Technical Cooperation for obtaining a PhD scholarship.
References Beckers, H.J., Soentoro, P.S.S., 1989. Method for detection of Shigella from foods. Zb. Bakt. Hyg. B 187, 261–265. de Boer, E., 1998. Update on media for isolation of Enterobacteriaceae from foods. Int. J. Food Microbiol. 45, 43–53. June, G.A., Sherrod, P.S., Amaguana, R.M., Andrews, W.H., Hammack, T.S., 1993. Effectiveness of the bacteriological analytical manual culture method for the recovery of Shigella sonnei from selected foods. J. AOAC Int. 76, 1240–1248. Lampel, K.A., Sandlin, R.C., Formal, S., 2000. Shigella: introduction and detection by classical cultural techniques. In: Robinson, R.K., Batt, C.A., Patel, P.D. ŽEds.., Encyclopedia of Food Microbiology. Academic Press, London, pp. 2015–2020. Lindqvist, R., 1999. Detection of Shigella spp. in food with a nested PCR method-sensitivity and performance compared with a conventional culture method. J. Appl. Microbiol. 86, 971–978. Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M., Tauxe, R.V., 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5, 1–21.
M. Uyttendaele et al.r International Journal of Food Microbiology 70 (2001) 255–265 Mehlman, I.J., Romero, A., Wentz, B.A., 1985. Improved enrichment for recovery of Shigella sonnei from foods. J. AOAC 68, 552–555. Mossel, D.A.A., Jacobs-Reitsma, W.F., 1990. Grensreactie op Shigella. Microbiologisch Onderzoek van Levensmiddelen. Keesing Noordervliet, Zeist ŽNetherlands., pp. 94–95. Mossel, D.A.A., Corry, J.E.L., Struijk, C.B., Baird, R.M., 1995a. The ecometric monitoring technique. Essentials of the Microbiology of Foods. Wiley, Chichester ŽEngland., pp. 278– 281. Mossel, D.A.A., Corry, J.E.L., Struijk, C.B., Baird, R.M., 1995b. Shigellosis. Essentials of the Microbiology of Foods. Wiley, Chichester ŽEngland., p. 131. NKML, 1995. Shigella bacteria. Detection in foods Žproposed method., NMKL No. 151. Oslo, Norway: National Veterinary Institute, Department of Food and Feed Hygiene. Rafii, F., Lunsford, P., 1997. Survival and detection of Shigella
265
flexneri in vegetables and commercially prepared salads. J. AOAC Int. 80, 1191–1197. Smith, J.L., 1987. Shigella as a foodborne pathogen. J. Food Prot. 50, 788–801. Smith, J.L., Dell, B.J., 1990. Capability of selective media to detect heat-injured Shigella flexneri. J. Food Prot. 53, 141– 144. Smith, J.L., Buchanan, R.L., 1992. Shigella. In: Vanderzant, C., Splittstoesser, D.F. ŽEds.., Compendium of Methods for the Microbiological Examination of Foods. APHA, Washington, USA, pp. 423–431. Tollison, S.B., Johnson, M.G., 1985. Sensitivity to bile salts of Shigella flexneri sublethally heat stressed in buffer or broth. Appl. Environ. Microbiol. 50, 337–341. US Food and Drug Administration ŽUS FDA., 1992. Bacteriological Analytical Manual. 7th edn. AOAC International, Arlington, VA.