Persistence strategies of Bacillus cereus spores ...

32 downloads 0 Views 2MB Size Report
Dec 1, 2009 - Survival of Bacillus cereus spores of dairy silo tank origin was investigated under conditions simulating those in operational dairy silos.
Food Microbiology 27 (2010) 347e355

Contents lists available at ScienceDirect

Food Microbiology journal homepage: www.elsevier.com/locate/fm

Persistence strategies of Bacillus cereus spores isolated from dairy silo tanks Ranad Shaheen a, Birgitta Svensson b, Maria A. Andersson a, Anders Christiansson c, Mirja Salkinoja-Salonen a, * a

Department of Applied Chemistry and Microbiology, University of Helsinki, P.O. Box 56, Biocenter 1, Viikinkaari 9, FIN-00014 Helsinki, Finland Tetra Pak, Development & Engineering, Packaging Technology, Ruben Rausings gata, SE-221 86 Lund, Sweden c Swedish Dairy Association, Research and Development, Scheelevägen 17, Ideon Science Park, S-22370 Lund, Sweden b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 August 2009 Received in revised form 28 October 2009 Accepted 1 November 2009 Available online 1 December 2009

Survival of Bacillus cereus spores of dairy silo tank origin was investigated under conditions simulating those in operational dairy silos. Twenty-three strains were selected to represent all B. cereus isolates (n ¼ 457) with genotypes (RAPD-PCR) that frequently colonised the silo tanks of at least two of the sampled eight dairies. The spores were studied for survival when immersed in liquids used for cleaningin-place (1.0% sodium hydroxide at pH 13.1, 75  C; 0.9% nitric acid at pH 0.8, 65  C), for adhesion onto nonliving surfaces at 4  C and for germination and biofilm formation in milk. Four groups with different strategies for survival were identified. First, high survival (log 15 min kill 1.5) in the hot-alkaline wash liquid. Second, efficient adherence of the spores to stainless steel from cold water. Third, a cereulide producing group with spores characterised by slow germination in rich medium and well preserved viability when exposed to heating at 90  C. Fourth, spores capable of germinating at 8  C and possessing the cspA gene. There were indications that spores highly resistant to hot 1% sodium hydroxide may be effectively inactivated by hot 0.9% nitric acid. Eight out of the 14 dairy silo tank isolates possessing hotalkali resistant spores were capable of germinating and forming biofilm in whole milk, not previously reported for B. cereus. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Bacillus cereus Spores Dairy silo tank Adherence Biofilm Cereulide Psychrotrophic Ribopattern Alkali tolerance

1. Introduction Bacillus cereus is a spore forming bacterium commonly contaminating raw milk and considered a major microbiological problem in the dairy industry (Andersson et al., 1995). Heat stable spores of B. cereus in milk are a source of contamination for milk derived products, such as milk powder, infant food formulas (Becker et al., 1994; Shaheen et al., 2006) and many food commodities (Wijnands et al., 2006). It is known that B. cereus spores occur in low numbers (102e103 per liter) in farm collected milk (Banyko and Vyletelova, 2009; Bartoszewicz et al., 2008; Christiansson et al., 1999; Svensson et al., 2004, 2006; Vissers et al., 2007). Studies by global typing methods (fatty acid profiling, biochemical typing, RAPD (random polymorphic DNA)-PCR, rep-PCR fingerprinting) have shown that the distribution of genotypes in the dairy and its products differed from that in raw milk (Bartoszewicz et al., 2008; Lin et al., 1998; Svensson et al., 1999, 2004; Te Giffel et al., 1997, 2002). Thus the farms are not * Corresponding author. Tel.: þ358 9 19159300, þ358 40 5739049 (mobile); fax: þ358 9 19159301, þ358 9 19159331 (secr). E-mail address: mirja.salkinoja-salonen@helsinki.fi (M. Salkinoja-Salonen). 0740-0020/$ e see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.fm.2009.11.004

the sole source of B. cereus in dairy milk. Additional contamination of milk occurs after the arrival to the dairy plant. A modern dairy plant is not an easy environment for B. cereus to colonise. The incoming milk is stored at cold temperature, heat treated, and the equipment is washed with hot, highly alkaline (pH > 13) and acid (pH < 1) liquids. It has been shown that certain genotypes of B. cereus recurr in dairy silo tanks (Svensson et al., 2004, 2006) but the phenotypic properties enabling the persistence under the dairy conditions are not understood. Spore adhesion to nonliving surfaces at cold temperature has rarely been studied. The aim of this study was to identify phenotypic features of the recurrent B. cereus dairy silo genotypes to explain their frequent presence in the silo tanks. We report here on the spore survival properties of 23 B. cereus strains from dairy silo tanks, selected to represent the isolates (n ¼ 457) with RAPD-PCR genotypes frequent in the silo tanks of several dairy plants. We exposed the spores to conditions simulating those in an operational dairy, including highly alkaline and acid liquids at high temperature applied during the cleaning-in-place (CIP) procedures. We also inspected the ability of the spores to adhere in cold environments to stainless steel and other nonliving materials, their germination at cold temperature and ability to form biofilm.

348

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

2. Materials and methods 2.1. Bacterial strains and their characterisations The dairy silo tank isolates of B. cereus (sensu lato) in this study originated from a study conducted in milk silos of eight different dairies over a period of two years (Svensson et al., 2004). The isolates indicated as psychrotrophs (marked P in tables/figures) grew (Svensson et al., 2004) on TSA at 8  C in 7e10 days and were PCR positive (using the primers of Francis et al., 1998) for the CspA gene. The 23 strains selected for the present study represented all (n ¼ 453) isolates that belonged to the frequently detected RAPDPCR patterns (Nilsson et al., 1998; Svensson et al., 2004) and had been found from the silos of more than one dairy. Strains UM 218, GO 282, SU 160, SU 226, GR 177 were deposited to the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig DE). Ribopattern analysis was done as described by Apetroaie et al. (2005). The reference libraries were the commercially available library of Qualicon (RiboExplorer, release 2008, software v.2.1.4216.0, Dupont, Wilmington NJ) and the in-house (Mirja Salkinoja-Salonen, Helsinki University) library of 120 well characterised strains of B. cereus sensu lato. Emetic toxin (cereulide) production was detected by the bioassay based on loss of sperm motility as described by Andersson et al. (2004) and confirmed by the chemical assay based on cereulide specific mass ions as described by Jääskeläinen et al. (2003). Spores were prepared as described by Magnusson et al. (2006). The spore count was determined on plate count agar and the spores were diluted to a concentration of 1e5109 cfu ml1 and stored in sterile saline at 4  C until use. 2.2. Reagents and media Reagents. The neutralizing reagent was 67 mM Sörensen NaeK phosphate (KH2PO4eNa2HPO4) buffer according to Sörensen, pH 7.0. Fixation reagent for electron microscopy was 2.5% (w/v) glutaraldehyde (J.T. Baker Chemical Co.) in 0.1 M NaeKephosphate buffer pH 7.2. Acridine orange staining solution contained 100 mg of acridine orange ml1 (Molecular Probes Europe, Leiden, The Netherlands) in water. The Live/Dead stain was Syto 9 (3.34 mM) with propidium iodide (PI, 1.5 mg ml1), Molecular Probes Europe (Leiden, The Netherlands) Media. Plate Count Agar (PCA) and R2 were prepared as described in Eaton et al. (2005), tryptic soy broth (TSB) and skim milk medium were from Difco (Detroit, MD, USA). The whole milk medium was pasteurized milk (3.5% fat) purchased from a local store and autoclaved (15 min, 121  C). The electric conductances of the media used for the spore adhesion experiments were measured as follows (mS cm1): skim milk, 5.3, tryptic soy broth 13.6, R2 broth 0.87, drinking water 0.15. 2.3. Alkali and acid tolerance of the spores To prepare the test suspension of spores, 100 ml of the stock (1e5  109 cfu ml1) was diluted in 100 ml of sterile water and the spore count determined (in triplicate, concentration at time zero) following dilution 1:10 in the neutralizing reagent. For measurement of hot-alkaline resistance, 100 ml of 1% (w/v) aqueous NaOH (pH 13.1) in a 200 ml flask was heated at 77  C in a water bath. When the temperature in a parallel flask containing 100 ml of water (measured with a thermocouple) reached 75  C, 100 ml of the spore test suspension was added into the flask, kept under gentle mixing with a magnetic stirrer. This mixture (temperature measured was 74.2e75.2  C) was sampled (1 ml) at intervals of 1 min up to 15 min. Each sample was diluted in 9 ml of the neutralizing reagent and the

viable spores counted on Plate Count Agar (PCA) read after 24 h at 30  C or 7e9 days at 8  C. Resistance of the spores to hot acid was measured similarly, except that 0.9 %w/v aqueous HNO3 (pH 0.8) was used instead of 1% NaOH, the temperatures of the water bath was 66.7  C and that of the test flasks from 65.3 to 65.8  C. D-values were calculated by linear regression using SYSTATÒ 9 (Systat Software Inc., Chicago, USA) from the log-transformed count of viable spores (cfu on PCA) versus heating time. The initial linear parts of the killing curves were used to calculate the reciprocals of the regression coefficients. 2.4. Assays of spore adherence and biofilm formation on nonliving surfaces Adherence of the spores to polystyrene and to glass was measured using 96-well plates NunclonÔD (with hydrophobic optical bottom or with optical bottom with cover glass base), Nunc F96 MicroWellÔ (with untreated or with hydrophilic cell culture treated polystyrene bottom). The media used to fill the wells (200 ml per well) were: sterilized drinking water, R2 medium, tryptic soy broth, skim milk medium and whole milk medium. The microplate wells were inoculated with 5  106e5  107 spores per well, covered with a lid and incubated on a rotary shaker (160 rpm, 4  C) for 2 days. The wells were emptied, washed three times with drinking water to remove unadhered spores. The adhered spores were stained with 300 ml of aqueous acridine orange (100 mg ml1) for 3 min and then washed three times under running water. The cumulative fluorescence emission of the wells was measured using a scanning fluorometer (Fluoroskan Ascent, Thermofisher Scientific, Vantaa, Finland) with band pass filter of 450e480 nm for excitation and a long pass filter of 520 nm for emission. Adhesion of the spores to stainless steel (AISI 304) was assayed with coupons of stainless steel (AISI304, w1 cm2), cleaned before use with 1% w/v detergent (Nelli soap, Farmos, Turku, Finland), disinfected with ethanol (96 vol %) and autoclaved. The coupons were aseptically mounted into the wells of a 6-well polystyrene plate (Nunc multidish) with 4 ml of sterilized drinking water per well. Spores, 5  107 (or 5  106, strain UM 169) cfu or none (background) per well were added, the plate covered with a lid and incubated with shaking (160 rpm) for 2 d at 4  C. The coupons were then washed with drinking water, stained with acridine orange for 3 min, rinsed with water and the fluorescence emission from the whole surface area of each coupon was measured with the scanning fluorometer at 485 nm (excitation) and 520 nm (emission). Background fluorescence (wells with no added spores) was subtracted from each reading. Capacity of the B. cereus spores to form biofilm in milk was measured as wall growth in 96-microplate wells. Polystyrene microplates were filled with 200 ml of full-fat milk or 1:10 diluted milk, inoculated with w5  107 spores per well and incubated on a rotary shaker (160 rpm, 21  C) for 2 days. The wells were emptied, washed three times with running water to remove the unadhered materials and then Live-Dead stained, 300 ml per well for 20e30 min, then washed three times under running water. The cumulative fluorescence of the biofilm was measured using the scanning fluorometer with the filter pairs 485/538 nm (Syto 9). Biofilms on the microplates were also observed using an epifluorescence light microscope. 2.5. Microscopy Microscopy of the microplate biofilms and adhered spores on the steel coupons was done using epifluorescence microscope (Nikon Eclipse E800, Tokyo, Japan) with filters 485 nm (excitation) and 520 nm (emission).

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

For field emission scanning electron microscope (FESEM) analysis, steel coupons were fixed with phosphate-buffered glutaraldehyde for 3e5 h at 22  1  C. The coupons were washed with the buffer twice and incubated with the buffer for 30 min. The coupons were dehydrated in ethanol series (50, 70, 96 vol %, 5 min each and 99.9 vol %, 10 min). The coupons were dried in hexamethyldisilazane (Fluka, Buchs, Switzerland) for 20 min and observed with FESEM (Hitachi S-4380, Tokyo, Japan) operated at 1 kV. Nonadhering spores were pipetted as a suspension on carbon tape and prepared for FESEM as above. 3. Results The dairy silo isolates of B. cereus for the present study were selected to represent RAPD-PCR patterns prevalent among those repeatedly found in the silos of more than one dairy (Table 1). Of the twenty-three selected strains 18 were mesophiles and 5 psychrotrophic (based on the PCR assay and growth test at 8  C). Three mesophilic isolates possessed the typical emetic RAPD type and were found to produce the emetic toxin (cereulide) both in the bioassay (sperm test) and the chemical assay (LCeMS). 3.1. Tolerance of the spores towards solutions of aqueous hot alkali and acid Spore suspensions were prepared from each of the 23 strains in Table 1 and investigated for survival when mixed into 1% w/v sodium hydroxide (pH 13.1) at 75  C and into 0.9% w/v nitric acid Table 1 Bacillus cereus isolates from dairy silo tanks. Strainsa,b

RAPD-group

Mesophilic/psychrotrophicc

Cereulide production

KA 111 UM 169 JO 59 GO 159

Ungrouped Ungrouped Ungrouped Ungrouped

M M M M

e e e e

UM 218* GO 95 SU 119

Ungrouped Ungrouped Ungrouped

P P P

e e e

GR 117 JO 273 SU 285 UM 284 GO 282*

1 1 1 1 1

M M M M M

e e e e e

GR 225 SU 160* VI 104

2 2 2

M M M

e e e

KA 155 SU 226*

3 3

M M

e e

UM 98

4

M

e

VI 172 GR 53

8 8

P P

e e

GR 177d* GR 651 mjA1e

E E E

M M M

þ þ þ

a

The strains were selected from the collection representing all RAPD groups that were found in more than one dairy. RAPD groups 1, 2, 3, 4, 8 and E represented 28%, 28%, 23%, 7.4%, 5.0% and 1.1%, respectively, of the isolates from the silo tanks (ca. 100 000 L each) at eight different dairies, used to store the milk from farms (Svensson et al., 2004, 2006). b The strains marked with * were deposited at the German Collection of Microorganisms and Cell Cultures (DSMZ) under accession codes DSM 22652 (UM 218); DSM 22648 (GO 282), DSM 22650 (SU 160), DSM 22651 (SU 226). DSM 22649 (GR 177). c M, mesophilic; P, psychrotrophic as judged by the PCR-method (cspA) and growth at 8  C (Svensson et al., 1999, 2004). d SDA strain also included in the studies of Carlin et al. (2006). e mjA1 was isolated from milk collected from a dairy farm (Svensson et al., 2006).

349

(pH 0.8) at 65  C, typically used for cleaning of dairy silos. Initial Dvalues (time for the first 1 log kill of the spores) and 15 min log kill values of the spores in the hot-alkali solution are shown in Table 2. Initial D-values ranged from 1.6 min to 67 min. The 15 min log kill value was measured to model the duration of the hot washing of the silo tanks in the dairy. Twelve (52%) of the 23 isolates had log 15-min kill values R2 broth > drinking water, from 13.6 to 0.15 mS cm1. The adhered spores on the microplates stained with acridine orange were inspected with epifluorescence microscope (Fig. 1) and the retained fluorescence measured by a scanning fluorescence reader (Fig. 2). The spores adhered significantly to the highly hydrophobic polymer (Nunclon D, Fig. 2A) and to glass (Fig. 2D) whereas the polystyrene plates (“tissue culture treated” or not treated, Fig. 2B and C) attracted less spores. It is also evident from Fig. 2 that the spores adhered more efficiently from drinking water or a dilute culture medium (R2) than from rich broth (skim milk, tryptic soy broth). Only few (103 mm2) spores adhered to the polystyrenes

in skim milk but more (105 spores mm2) adhered in drinking water (examples shown in Fig. 1A and B). The results in Fig. 2 demonstrate that the adherence of the spores at 4  C varied from strain-to-strain with no connection to RAPD group (Table 1) or longevity in hot alkali (Table 2). The ability of the spores to adhere to steel surface was tested with coupons made of steel similar to that used for dairy silo tanks. The results (Fig. 3) showed that the spores immersed at 4  C into drinking water adhered to steel similarly or slightly more than to glass or the plastic surfaces (Fig. 2). Of the eight strains with significant (RFU > 5 in Fig. 3) adherence to steel three were psychrotrophs (marked with P in Fig. 3). The hot-alkali resistant spores (marked with arrows in Fig. 3) were not more adhesive than the alkali sensitive spores. To assess if the spores possessed specific organelles responsible for the adherence (as suggested by Stalheim and Granum, 2001), the ultrastructures of spores were inspected by field emission scanning microscope (FESEM). Fig. 4 shows images of spores representing those that adhered effectively (UM 98) and those adhering poorly (GR 117). Small spike-like organelles are visible on the spores of UM98 adhered to steel as well as on the nonadhered spores of GR 117, offering no explanation for the different ability to adhere. 3.3. Generation of biofilm by spores in milk Ability to adhere and to form biofilm was investigated with spores in whole milk and its 10 dilution with water. Nonadhering cells were removed by vigorous washing as before. It was found that in whole milk the spores of nine dairy silo isolates, UM 169, GR 117, JO 273, DU 285, UM 284, GO 282, SU 226, UM 98, GR53, germinated and grew into a biofilm within 48 h at 21  C (Fig. 5).

Fig. 1. Visualisation of adherence of spores of a dairy silo representative of B. cereus to different substrata. Spores adhering at 4  C onto the wells of the microplates were stained with acridine orange and inspected with epifluorescence microscope. The figure shows the spores of the strain GR 177 (RAPD group E) adhering from skim milk medium to the hydrophobic polymer (A); from sterilized drinking water to the hydrophobic polymer (B); from skim milk medium to glass (C); from sterilized drinking water to glass (D) or to stainless steel (E). Stainless steel with no added spores (F). Measure bars indicate the scale.

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

351

Fig. 3. Adherence of spores of the 23 dairy silo tank representatives of B. cereus onto stainless steel. The spores suspended in sterile drinking water into the wells of 6-well microplates with a stainless steel coupon horizontally inserted into each well and incubated under shaking at þ4  C. After 2 d the plates were washed, rinsed and stained with acridine orange. Each column shows the fluorescence reading from the entire surface of the steel coupon (average of three parallels, error bars indicate S.D.) Background fluorescence (no added spores) was subtracted from the readings. The strains resistant to hot alkali (Table 2), are marked with arrows and the psychrotrophic strains (Table 1) with P.

polystyrene (Fig. 2B) when tested at 4  C in drinking water, skim milk or laboratory media (R2 broth, TSB broth). 3.4. Connections between the survival phenotypes and genotypes Fig. 2. The adherence of spores of the 23 dairy silo representatives of Bacillus cereus to nonliving surfaces at cold temperature. Spores of the dairy silo tank isolates dispersed in different media (water, R2 broth, tryptic soy broth or skin milk medium) were dispensed in the wells of microplates with bottoms made of hydrophobic polymer (panel A), polystyrene (panel B), cell culture treated polystyrene (panel C) or glass (panel D). After 2 d shaking at þ4  C the wells were emptied, washed and stained with acridine orange. The adhering spores were scored by a scanning fluorometer. The four slices of fluorescence readings stacked in each column indicate the spores adhered onto that microplate surface from each of the four media (three replicate assays, S.D. marked with bars). The strains resistant to hot alkali (Table 2), are marked with arrows.

Eight (marked with arrows in Fig. 5) of these nine isolates were among the most resistant to hot 1% NaOH (Table 2). It is interesting that only one (GR 53) of the five psychrotrophs and none of the emetic toxin producing strains (GR 177, GR 651, mjA1) formed biofilm within 48 h in milk. The phenomenon was very clear-cut: either the biofilm growth was plentiful or it was none. When the test for biofilm formation was executed in milk diluted (10) with water, none of the 23 isolates produced any biofilm (result not shown). The results thus showed that spores of the nonemetic, but not of the emetic, B. cereus silo isolates with remarkable resistance to hot NaOH (Table 2), possessed capacity to grow into tenacious (i.e. not removable by mechanical washing) biofilm when immersed into whole milk (Fig. 5). The same spores did not significantly adhere to

The 23 strains were genotyped by RAPD-PCR (Table 1) and by riboprinting analysis (Fig. 6). It is seen in Fig. 6 that all strains within the RAPD group 1 (n ¼ 5) yielded identical ribopatterns by the restriction enzymes EcoR1 and PvuII. The ribopatterns of the isolates of RAPD groups 2 (n ¼ 3) and 3 (n ¼ 2) also were identical within the RAPD group, as well as those of the RAPD group E (n ¼ 3, emetic toxin producing). For colonising dairy silo tanks it is important to survive when immersed into hot NaOH. Out of the 12 isolates with high survival (log 15 min kill 1.5, Table 2) at pH > 13 at 75  C six isolates (GR 117, UM 284, JO 273, SU 285, GO 282, GR 53) showed identical EcoR1 and Pvu2 ribopatterns, and four (SU 160, GR 225, Vi 104, UM 98) shared at least the EcoR1 patterns, although the isolates came from different silos and isolation times. It thus appears from Tables 1, 2 and Fig. 6 that resistance to hot alkali of the mesophilic dairy silo isolates affiliated to certain ribopatterns, indicating strain-level relatedness. It is also interesting that of the nine isolates with spores that rapidly germinated and grew into biofilm in whole milk (Fig. 5) seven (GO 159, GR 117, UM 284, JO 273, SU 285, GO 282, GR 53) shared identical ribopatterns (EcoR1 and Pvu2, Fig. 6) although they grouped differently by RAPD (Table 1). The psychrotrophic strains (n ¼ 5) showed no similarity to each other in the ribopatterns. None of the dairy silo isolates showed ribopatterns matching to any of the Bacillus thuringiensis strains (n ¼ 29) in the ribopattern reference libraries used in this study.

352

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

Fig. 5. Germination and outgrowth to a biofilm in whole milk of spores of the dairy silo tank isolates of Bacillus cereus. The spores were suspended into the wells of polystyrene microplates with whole milk. After 48 h of shaking at 21  C the cells not or loosely adhered were removed by washing. The green fluorescence (Y-axis) by the live-stained (Syto 9) cells proves that these are germinated spores as adhered, nongerminated spores do not fluoresce green. Background with no added spores is subtracted from the emissions; error bars show the S.D. of three independent assays. The strains resistant to hot alkali (Table 2) are marked with arrows and the psychrotrophic strains (Table 1) with P.

4. Discussion

Fig. 4. Field emission scanning electron micrographs (FESEM) of steel adhered spores of B. cereus isolated from dairy silo tanks. The spores were incubated with the steel coupons at þ4 for 24 h in drinking water, UM 98 (A, B); GR 117 (C). Panel D shows nonadhered spores of the strain UM 98 collected from water suspension onto carbon tape. The arrow marks a spike-like appendage. Measure bars indicate the magnification.

The spore survival properties of B. cereus isolates from dairy silo tanks were investigated under conditions simulating those in operational dairy silos, with extremes of temperatures (from 4  C to þ75  C) and of pH (from 13). The B. cereus isolates for the present study were selected to represent the 457 isolates representing the RAPD groups that occurred in several of the earlier investigated eight dairies (Svensson et al., 2004, 2006). The results revealed more than one strategy for survival of the B. cereus spores. Four of the 23 dairy silo isolates had spores with extremely high D value, >40 min in hot alkali (pH > 13) and a low 15 min kill (log < 0.3). This extreme alkali tolerance may be the highest reported for B. cereus. Earlier Langsrud et al. (2000) reported for B. cereus (ATCC 9139) 3 log reduction of the spore viability when suspended in 1% NaOH at 80  C for 20 min. Nine of the 23 representative dairy silo isolates studied in this paper had initial D-values (time needed for viability reduction by 90%) of 15e67 min. In 15 min of hot alkali washing the spore count was reduced by 1 log unit or less (0.1e1). Such spores cannot be eradicated from the silo tanks by hot alkaline washing. Therefore it is interesting that the spores with highest resistance to hot alkali were effectively killed by hot acid washing: the spore viability was reduced by 4e5 log units during 15 min. The results suggest that hot acid washing could be used to effectively reduce colonization of dairy silo tanks by hot-alkali resistant B. cereus spores. The four silo isolates with extreme tolerance to hot alkali were identical in ribopatterns to the human gut isolate B. cereus UB 0962, connected to unexplained gastrointestinal illness of an infant (Apetroaie et al., 2005). The similarity suggests this genotype may be competitive also in other environments. Another strategy for silo survival was represented by isolates with spores that readily adhered from water to steel at cold

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

353

Fig. 6. Ribopatterns obtained with PvuII or with EcoRI for the 23 dairy silo tank isolates of B. cereus. The psychrotrophic strains (Table 1) are marked with P. The closest matches (for both enzymes, marked Ref) are also shown: F3371/93 from diarrheal food poisoning in UK (Pirttijärvi et al., 1999; Apetroaie et al., 2005); ATCC 4342 from milk, USA (Helgason et al., 2000; Apetroaie et al., 2005); AB11A, an endophyte of potato from cold soil (Virtanen et al., 2008); B402 from food (not connected to illness) Finland; UB 0962, faeces of a child with unexplained digestive tract symptoms (Apetroaie et al., 2005); CIF2 and MHI87, infant food formulas from Finland and Germany (Shaheen et al., 2006); F4810/72, vomit of an emetic food poisoning patient, UK (Pirttijärvi et al., 1999).

temperature. Interestingly, these same isolates were the most sensitive to hot alkali. These isolates adhered from cold water to steel better than did the alkali resistant isolates. It is known that adhered bacteria are more difficult to kill by surface disinfection or by heat than are bacteria suspended in aqueous medium (Te Giffel et al., 1995; Faille et al., 2001; Simmonds et al., 2003; Ryu and Beuchat, 2005; Hornstra et al., 2007). Adhesion of the spores to dairy equipment of stainless steel may thus represent a mechanism for survival. It is also possible that strong adhesion is of advantage for spreading of spores with rinse water from one location to another. Emetic strains (GR 177, GR 651, mjA1) represented a third strategy for survival. These strains were highly or moderately resistant to hot alkali. They displayed unique ribopatterns (both EcoRI and PvuII) that link them to the major clade of known cereulide producing strains (Pirttijärvi et al., 1999; Apetroaie et al., 2005). This clade (Ehling-Schulz et al., 2005), is known for producing spores germinating extremely slowly (1 log in 7 d) in rich medium (at 7  C or at 30  C) and retain viability upon exposure to 120 min heating at 90  C (Carlin et al., 2006). Slow germination is likely to promote chances for survival in dairies where most milk is

heat treated (pasteurized) resulting to inactivation of germinated spores. The ribopatterns of the emetic toxin (cereulide) producing dairy strains GR 177, GR 651 and mjA1 in this paper, were identical to those of emetic toxin producing isolates from dried infant formulas (Shaheen et al., 2006). Thus dairies could be one possible source for cereulide producing B. cereus in industrial infant food. The five psychrotrophic isolates all possessed the cspA gene (Francis et al., 1998), suggested to determine psychrotolerance (Bartoszewicz et al., 2009), and grew at þ8  C but otherwise shared little in common. Ability to germinate and grow at low temperature may have been the main strategy of the psychrotrophic strains to colonise the dairy silos. The ribopattern of the acid tolerant psychrotrophic dairy silo isolate (UM 218) was identical to that of the endophytic B. cereus strains AB11A and BVG1A (Fig. 6) isolated from the interior of cold soil potatoes in Finland (Virtanen et al., 2008), suggesting this genotype may be successful in cold environments. The present study revealed a new class of B. cereus, representing 40% of the dairy silo tank isolates. These highly alkali resistant isolates germinated and formed biofilm in whole milk. Biofilm formation by B. cereus in milk appears to have received little

354

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355

attention, possibly because it may be rare in environments other than dairy silos. In species other than B. cereus negative effects of whole milk on biofilm formation have been observed (Helke et al., 1993; Flint et al., 1997; Wong, 1998). Hsueh et al. (2006) reported that biofilm formation in B. cereus is enhanced by low nutrient conditions and that coating the polystyrene plates with surfactin increased the ability of B. cereus strain ATCC 14579 to form biofilm. The presence of biosurfactant was shown to trigger biofilm formation also in Bacillus subtilis (Lopez et al., 2009; Branda et al., 2005). Whole milk contains natural surfactants and also phospholipid is a surface active compound and found in the fat globule of milk. Forming biofilm by some B. cereus strains in whole milk and not in water diluted milk suggests that the surface active compound found in the whole milk might work as a surfactant needed for the biofilm of certain strains of B. cereus. Chemical CIP (cleaning-in-place procedure) sanitizers were reported less effective against B. cereus cells in biofilm formed on milk presoiled than on nonsoiled stainless steel chip (Peng et al., 2002). Spore adhesion of B. cereus onto stainless steel has been studied (Hornstra et al., 2007; Wijman et al., 2007) but adhesive properties of spores of isolates from dairy plants appear not to have been studied. The spores of the 23 strains in this study adhered to steel more efficiently from water than from TSB or skim milk. We found large strain-to-strain variations in the adhesion properties of the B. cereus spores both at þ4  C and at the temperature allowing growth and formation of biofilms. The more uniform behaviour of spores of the strains studied by Wijman et al. (2007) may reflect the fact that their isolates originated from food and clinical environments, which are less extreme compared to the dairies where the temperatures during cleanup may range from þ4  C to >70  C and the pH from 0.8 to 13.1. The small, spike-like organelles, observed by electron microscopy on the surfaces of many but not all spores of several isolates, did not reflect the spore adhesion properties of the isolates in any consistent manner. They may represent extrusions of the exosporium or glycoprotein naps observed earlier by several authors on B. cereus spores (Kulikovsky et al., 1975; Tauveron et al., 2006; Stalheim and Granum, 2001). Summarising, the 23 B. cereus isolates selected to represent the most prevalent RAPD genotypes of a large number of primary isolates (n ¼ 2297, Svensson et al., 2004) in the farm milk receiving silo tanks of eight different dairies (Svensson et al., 2004, 2006) showed several properties explaining their successful colonization of the silo tanks. In addition, these isolates revealed features previously unknown to the species B. cereus, e.g. extreme resistance to hot alkali (pH > 13) and hot acid (pH < 1), ability adhere to nonliving surface at cold temperature and to form biofilm in whole milk. Acknowledgements This work was supported by funding from the European Commission (QLK1-CT-2001-00854) and by Academy of Finland grant to the CoE Photobiomics (118637). Douwe Hoornstra and Camelia Apetroaie-Constantin are thanked for constructing the ribopattern database used in this study. We want to thank Viikki Science Library for the excellent information service, the Faculty Instrument Centre for technical service and Leena Steininger, Hannele Tukiainen and Tuula Suortti for many kinds of help. References Andersson, A., Rönner, U., Granum, P., 1995. What problems does the food industry have with the spore-forming pathogens Bacillus cereus and Clostridium perfringens? International Journal of Food Microbiology 28, 145e155. Andersson, M.A., Jääskeläinen, E.L., Shaheen, R., Pirhonen, T., Wijnands, L.M., Salkinoja-Salonen, M.S., 2004. Sperm bioassay for rapid detection of cereulide

producing Bacillus cereus in food and related environments. International Journal of Food Microbiology 94 (5), 1453e1457. Apetroaie, C., Andersson, M.A., Spröer, C., Tsitko, I., Shaheen, R., Jääskeläinen, E.L., Wijnands, L.M., Heikkilä, R., Salkinoja-Salonen, M.S., 2005. Cereulide producing strains of Bacillus cereus show diversity. Archives in Microbiology 184, 141e151. Banyko, J., Vyletelova, M., 2009. Determining the source of Bacillus cereus and Bacillus licheniformis isolated from raw milk, pasteurized milk and yoghurt. Letters in Applied Microbiology 48, 318e323. Bartoszewicz, M., Hansen, B.M., Swiecicka, I., 2008. The members of the Bacillus cereus group are commonly present contaminants of fresh and heat-treated milk. Food Microbiology 25, 588e596. Bartoszewicz, M., Dideshi, D.K., Kraszewska, A., Modzelewska, E., Swicicka, I., 2009. Natural isolates of Bacillus thuringiensis display genetic and psychrotrophic properties characteristic of Bacillus weihenstephanensis. Journal of Applied Microbiology 106, 1967e1975. Becker, H., Schaller, G., Wiese, W.v., Terplan, G., 1994. Bacillus cereus in infant foods and dried milk products. International Journal of Food Microbiology 23, 1e15. Branda, S.S., Vik, Å., Friedman, L., Kolter, R., 2005. Biofilms: the matrix revisited. Trends in Microbiology 13 (1), 20e26. Carlin, F., Fricker, M., Pielaat, A.-M., Shaheen, R., Salkinoja Salonen, M., Svensson, B., Nguyen-the, C., Ehling-Schulz, M., 2006. Emetic toxin-producing strains of Bacillus cereus show a distinct phenotype within the Bacillus cereus group. International Journal of Food Microbiology 109, 132e138. Christiansson, A., Bertilsson, J., Svensson, B., 1999. Bacillus cereus spores in raw milk: factors affecting the contamination of milk during the grazing period. Journal of Dairy Science 82, 305e314. Eaton, A.D., Clesceri, L.S., Rice, E.W., Greenberg, A.E., 2005. Microbiological Examination. In: Standard Methods for the Examination of Water and Wastewater, 21st ed. American Public Health Association, Washington, DC, pp. 9e36. Part 9000, 9215. Ehling-Schulz, M., Svensson, B., Guinebretiere, M.-H., Lindbäck, T., Andersson, M., Schulz, A., Fricker, M., Schirsitnasson, A., Granum, P.E., Märtlbauer, E., NguyenThe, C., Salkinoja-Salonen, M.S., Scherer, S., 2005. Emetic toxin formation of Bacillus cereus is restricted to a single evolutionary lineage of closely related strains. Microbiology 151, 183e197. Faille, C., Fontaine, F., Bénézech, T., 2001. Potential occurrence of adhering living Bacillus spores in milk product processing lines. Journal of Applied Microbiology 90, 892e900. Flint, S.H., Bremer, P.J., Brooks, J.D., 1997. Biofilms in dairy manufacturing plant description, current concerns and methods of control. Biofouling 11, 81e97. Francis, K.P., Mayr, R., von Stetten, F., Stewart, G.S., Scherer, S., 1998. Discrimination of psychrotrophic and mesophilic strains of the Bacillus cereus group by PCR targeting of major cold shock protein genes. Applied and Environmental Microbiology 64 (9), 3525e3529. Helgason, E., Økstad, O.A., Caugant, D.A., Johansen, H.A., Fouet, A., Mock, M., Hegna, I., Kolstø, A.B., 2000. Bacillus anthracis, Bacillus cereus and Bacillus thuringiensis e one species on the basis of genetic evidence. Applied and Environmental Microbiology 66 (6), 2627e2630. Helke, D.M., Somers, E.B., Wong, A.C.L., 1993. Attachment of Listeria monocytogenes and Salmonella typhimurium to stainless steel and buna-N in the presence of milk and individual milk components. Journal of Food Protection 56, 479e484. Hornstra, L.M., de Leeuw, P.L.A., Moezelaar, R., Wolbert, E.J., De Vries, Y.P., De Vos, W. M., Abee, T., 2007. Germination of Bacillus cereus spores adhered to stainless steel. International Journal of Food Microbiology 116, 367e371. Hsueh, Y.H., Somers, E.B., Lereclus, D., Wong, A.C.L., 2006. Biofilm formation by Bacillus cereus is influenced by PlcR a pleiotropic regulator. Applied and Environmental Microbiology 72, 5089e5092. Jääskeläinen, E.L., Häggblom, M.M., Andersson, M.A., Vanne, L., SalkinojaSalonen, M.S., 2003. Potential of Bacillus cereus for producing an emetic toxin, cereulide, in bakery products: quantitative analysis by chemical and biological methods. Journal of Food Protection 66, 1047e1054. Kulikovsky, A., Pankratz, H.S., Sadoff, H.L., 1975. Ultrastructural and chemical changes in spores of Bacillus cereus after action of disinfectants. Journal of Applied Bacteriology 38, 39e46. Langsrud, S., Baardsen, B., Sundheim, G., 2000. Potentiation of the lethal effect of peroxygen on Bacillus cereus spores by alkali and enzyme wash. International Journal of Food Microbiology 56, 81e86. Lin, S., Schraft, H., Odumeru, J.A., Griffiths, M.W., 1998. Identification of contamination sources of Bacillus cereus in pasteurized milk. International Journal of Food Microbiology 43, 159e171. Lopez, F., Fischbach, M.A., Chu, F., Losick, R., Kolter, R., 2009. Structurally diverse natural products that cause potassium leakage trigger multicellularity in Bacillus subtilis. Proceedings of the National Academy of Sciences of the USA 106, 280e285. Magnusson, M., Christiansson, A., Svensson, B., Kolstrup, C., 2006. Effect of different premilking manual teat-cleaning methods on bacterial spores in milk. Journal of Dairy Science 89, 3866e3875. Nilsson, J., Svensson, B., Ekelund, K., Christiansson, A., 1998. A RAPD-PCR method for large-scale typing of Bacillus cereus. Letters in Applied Microbiology 27, 168e172. Peng, J.S., Tsai, W.C., Chou, C.C., 2002. Inactivation and removal of Bacillus cereus by sanitizer and detergent. International Journal of Food Microbiology 77, 11e18. Pirttijärvi, T.S.M., Andersson, M.A., Scoging, A.C., Salkinoja-Salonen, M., 1999. Evaluation of methods for recognizing strains of the Bacillus cereus group with food poisoning potential among industrial and environmental contaminants. Systematic and Applied Microbiology 22, 133e144.

R. Shaheen et al. / Food Microbiology 27 (2010) 347e355 Ryu, J.H., Beuchat, L., 2005. Biofilm Formation and sporulation by Bacillus cereus on a stainless steel surface and subsequent resistance of vegetative cells and spores to chlorine, chlorine dioxide and a peroxyacetic acid-based sanitizer. Journal of Food Protection 68, 2614e2622. Shaheen, R., Andersson, M.A., Apetroaie, C., Schulz, A., Ehling-Schulz, M., Ollilainen, V.-M., Salkinoja-Salonen, M.S., 2006. Potential of selected infant food formulas for production of Bacillus cereus emetic toxin, cereulide. International Journal of Food Microbiology 107, 287e294. Simmonds, P., Mossel, B.L., Intaraphan, T., Deeth, H.C., 2003. Heat resistance of Bacillus spores when adhered to stainless steel and its relationship to spore hydrophobicity. Journal of Food Protection 66 (11), 2070e2075. Stalheim, T., Granum, P.E., 2001. Characterization of spore appendages from Bacillus cereus strains. Journal of Applied Microbiology 91, 839e845. Svensson, B., Eneroth, Å., Brendehaug, J., Christiansson, A., 1999. Investigation of Bacillus cereus contamination sites in a dairy plant with RAPD-PCR. International Dairy Journal 9, 903e912. Svensson, B., Ekelund, K., Ogura, H., Christiansson, A., 2004. Characterisation of Bacillus cereus isolated from milk silo tanks at eight different dairy plants. International Dairy Journal 14, 17e27. Svensson, B., Monthan, A., Shaheen, R., Andersson, A., Salkinoja-Salonen, M., Christiansson, A., 2006. Occurrence of emetic toxin producing Bacillus cereus in the dairy production chain. International Dairy Journal 16 (7), 740e749. Tauveron, G., Slomianny, C., Henry, C., Faille, C., 2006. Variability among Bacillus cereus strains in spore surface properties and influence on their ability to

355

contaminate food surface equipment. International Journal of Food Microbiology 100, 254e262. Te Giffel, M.C., Beumer, R.R., Van Dam, W.F., Slaghuis, B.A., Rombouts, F.M., 1995. Sporicidal effect of disinfectants on Bacillus cereus isolated from the milk processing environment. International Biodeterioration & Biodegradation 36, 421e430. Te Giffel, M.C., Beumer, R.R., Granum, P.E., Rombouts, F.M., 1997. Isolation and characterisation of Bacillus cereus from pasteurised milk in household refrigerators in the Netherlands. International Journal of Food Microbiology 34, 307e318. Te Giffel, M.C., Wagendorp, A., Herrewegh, A., Driehuis, F., 2002. Bacterial spores in silage and milk. Antonie Van Leeuwenhoek 81, 625e630. Vissers, M.M.M., Te Giffel, M.C., Driehuis, F., De Jong, P., Lankveld, J.M.G., 2007. Predictive modeling of Bacillus cereus spores in farm tank milk during grazing and housing periods. Journal of Dairy Science 90, 281e282. Virtanen, S.M., Roivainen, M., Andersson, M.A., Ylipaasto, P., Hoornstra, D., Mikkola, R., Salkinoja-Salonen, M.S., 2008. In vitro toxicity of cereulide on porcine pancreatic Langerhans islets. Toxicon 51, 1029e1037. Wijman, J.G.E., de Leeuw, P.P.L.A., Moezelaar, R., Zwietering, M.H., Abee, T., 2007. Aireliquid interface biofilms of Bacillus cereus: formation, sporulation, and dispersion. Applied and Environmental Microbiology 73, 1481e1488. Wijnands, L.M., Dufrenne, J.B., Rombouts, F.M., in t'Veld, P.H., Van Leusden, F.M., 2006. Prevalence of potentially pathogenic Bacillus cereus in food commodities in the Netherlands. Journal of Food Protection 69 (11), 2587e2594. Wong, A.C.L., 1998. Biofilms in food processing environments. Journal of Dairy Science 81, 2765e2770.