1271

Journalof Food Protection, Vol. 67, No. 6, 2004, Pages 1271– 1277 Copyright q,International Association forFood Protection

Research Note PCRDetection of Bacillus and Staphylococcus inV ariousFoods

SHIGERU NAKANO, * TORU KOBAYASHI, KENICHI FUNABIKI, ATSUSHI MATSUMURA, YASUHIRO NAGAO, AND TOSHIHIRO YAMADA

FoodSafety Research Institute,Nissin Food Products Co. Ltd., 2247, Noji-cho, Kusatsu, Shiga 525-0055, Japan

MS03-403:Received 10September 2003/ Accepted 25January 2004 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 ABSTRACT

Abroad-rangePCR assayfor the detection of bacteria belonging to Bacillus and Staphylococcus generawas developed. Primerstargeting the bacterial 16S rRNA genewere newly designed and used in a PCR assay.T odeterminethe speciŽ city ofthe assay, 81 different bacterial strains (of 50 genera), 2 fungi,3 animals,and 4 plantswere tested. Results were positive forevery tested Bacillus,Staphylococcus, or Aerococcus strain.In addition, the result for Listeriagrayi waspositive with lowerPCR product.For all other bacterial strains and eukaryotes tested, results were negative. Bacterial DNA wasprepared withthe use of achromopeptidase and Chelex 100 resin from culture after growth in brain heart infusion medium. T otestthe sensitivityof this PCR assayfor Bacillus or Staphylococcus genus,either Bacilluscereus or Staphylococcusaureus was inoculatedinto various foods with undetectable levels of endogenous microbial contamination as an indicator .Inoculationof bacteriaat 10 to 30 CFU/ goffood was followed by a 5-henrichment culture step after which the PCR assayallowed the detectionof bacterial cells. When the inoculation ( B. cereus or S. aureus)of10 to 90 CFU/ gintonoodle foods containing endogenousmicro ora (10 3 to 105 CFU/g)wasfollowed by a6-henrichment culture step, the PCR assaydetected the bacteria. Includingthe enrichment culture step, the entire PCR detectionprocess can be completed within 8.5 h.

Inthe food industry, bacteria of the Bacillus or Staph- positiveand gram-negative bacterial species on thebasis of ylococcus genushave been notable because many strains of thebacterial 16S rRNA gene (5, 11, 12) havebeen reported thosebacteria can cause food spoilage or because several inrecent years. T oinspectfoods comprehensively and rap- strainsof those bacteria can produce toxins that are haz- idly,a universalor broad-range PCR assaywould be help- ardousto the health of the consumer (1,14, 17, 19, 20). fulin detecting targeted bacterial groups. Althoughsteps to detect the foodborne pathogens Bacillus Weattemptedto develop a rapidand speciŽ c PCRas- cereus and Staphylococcusaureus infoodshave been taken sayfor thedetection of bacteriaof the Bacillus and Staph- insome food industries, several approved traditional meth- ylococcus generain foods. W edesigneda setof broad- odsfor thedetection of these bacteria in foods usually re- rangePCR primersthat is based on thebacterial 16S rRNA quire1 or2 daysto complete (1, 14). For processedfood- gene;these primers are speciŽ c for thedetection of bacteria stuffs,such as whole milk, ham, and boiled Chinese noodle, of the Bacillus and Staphylococcus genera.In this report, therapid detection of contaminationnot only by foodborne we describehow the analysis of PCR-ampliŽed DNA frag- pathogenicbacteria but also by putrefactive bacteria is im- mentscould be used for thesensitive and rapid detection portant;therefore, a rapidprimary screening assay for the ofbacteria of the Bacillus and Staphylococcus genera in detectionof these bacterium groups has been sought by severalfood products after a shortenrichment period in manufacturersof foods. brainheart infusion (BHI) mediumbefore the PCR assay. PCRtechnologyhas proved to besensitiveand speciŽ c tothe rapid detection of pathogenic bacteria contaminating MATERIALS AND METHODS variousfoods (3,6– 8, 20, 22). Theuse of conventional, Preparationof DNA frombacterial strains and eukary- single-targetampliŽ cation methods are relatively costly be- oticorganisms. Thebacterial strains used in this study are sum- causeseveral sets of PCRprimersare usually necessary for marizedin T able1. Each strain was cultured in either BHI me- thedetection or identiŽcation of eachpathogenic bacterium; dium(Nacalai T esqueInc., Kyoto, Japan), BHI mediumsupple- however,targetingmultiple microbial organisms simulta- mentedwith 3% NaCl,buffered charcoal yeast extract agar (Bec- neouslyon asinglePCR reactionis moretime-efŽ cient and tonDickinson, Franklin, N.J.), chocolate agar (Nissui Pharmaceutical cost-effective.For clinicaldiagnosis, universal PCR prim- Co.Ltd., T okyo,Japan), blood agar (Nissui), mannitol salt agar ers thatare capable of detecting various eubacteria by the withegg yolk (Nissui), or lactobacilli deMan Rogosa Sharpe agar conservedregions of the bacterial 16S rRNA gene (15, 18) (Difco,Becton Dickinson, Sparks, Md.). Bacterial DNA wasex- orthatcan use broad-range PCR for theseparation of gram- tractedand puriŽ ed with the use of a PuregeneY eastand Gram- PositiveDNA Isolationkit (Gentra Systems Inc., Minneapolis, *Authorfor correspondence. T el: 81-77-561-9114;Fax: 81-77-561-9140; Minn.)or DNA ExtractionIsoplant II kit(Nippon Gene Co. Ltd., E-mail: [email protected]. Tokyo,Japan). The DNA offungi( Saccharomycescerevisiae IFO 1272 NAKANO ETAL. J.FoodProt., Vol. 67, No. 6

TABLE 1. Summaryof ampliŽcation results obtained with prbac TABLE 1. Continued primersand BS primersfor puriŽ ed DNA fromvarious bacteria a PCR result PCR result with primer with primer Bacterium Strain prbac BS Bacterium Strain prbac BS L.monocytogenes ATCC 7644 1 2 Agrobacteriumradiobacter ATCC 19358T 1 2 L. innocua NCIMB 13450 1 2 Alcaligenesfaecalis JCM 1474T 1 2 Lactobacilluscasei JCM 1134T 1 2 Chromobacteriumviolaceum JCM 1249T 1 2 L. gasseri JCM 1131T 1 2 Neisseriameningitidis ATCC 13077T 1 2 L.bulgaricus JCM 1002T 1 2 Xanthomonasmaltophilia JCM 1975T 1 2 Enterococcusfaecalis JCM 5803T 1 2 Legionellapneumophila JCM 7571T 1 2 Streptococcusmutans JCM 5705T 1 2 Pseudomonasaeruginosa ATCC 27843T 1 2 S.salivarius JCM 5707T 1 2

Moraxellaantipestifer JCM 9532T 1 2 S.agalactiae JCM 5671T 1 2 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 Acinetobacterbaumannii JCM 6841T 1 2 Collinsellaaerofaciens JCM 10188T 1 2 VibriovulniŽ cus JCM 3725T 1 2 Actinomycesnaeslundii JCM 8349T 1 2 Photobacteriumleiognathi ATCC 25521T 1 2 Micrococcusluteus JCM 1464T 1 2 Aeromonashydrophila ATCC 7966T 1 2 Microbacteriumlacticum JCM 1379 1 2 Enterobacteragglomerans JCM 1236T 1 2 Corynebacteriumkutscheri JCM 9385T 1 2 Citrobacterfreundii JCM 1657T 1 2 C. xerosis JCM 1971T 1 2 Escherichiacoli JCM 1649T 1 2 Propionibacteriumacnes JCM 6425T 1 2 Hafniaalvei JCM 1666T 1 2 BiŽdobacterium longum JCM 1217T 1 2 Klebsiellapneumoniae JCM 1662T 1 2 B.adolescentis JCM 1275T 1 2 Plesiomonasshigelloides ATCC 14029T 1 2 Bacteroidesvulgatus JCM 5826T 1 2 Proteusmirabilis JCM 1669T 1 2 Porphyromonasgingivalis ATCC 33277T 1 2 Salmonella Typhimurium IFO 13245 1 2 Prevotellaintermedia ATCC 25611T 1 2 SerratiaŽ caria JCM 1241T 1 2 Flavobacteriumjohnsoniae JCM 8514T 1 2 Shigella exneri ATCC 29903T 1 2 Cytophagaarvensicola JCM 2836T 1 2 Yersiniaenterocolitica ATCC 9610T 1 2 Fusobacteriumnucleatum ATCC 25586T 1 2 Haemophilusin uenzae ATCC 33391T 1 2 a Campylobactercoli ATCC 33559T 1 2 Fiftypicograms of puriŽed DNA fromeach bacterium was used Helicobacterpylori ATCC 43504T 1 2 asthe template. 1, positive; 2, negative; 1L,positive(the Clostridiumperfringens JCM 1290T 1 2 amountof amplicon with size targeted was lower). b Eubacteriumalactolyticum b JCM 6480T 1 2 Thisstrain was reclassiŽ ed in Pseudoramibacteralactolyticym Veillonellaalcalescens ATCC 27215 1 2 (10). Brevibacillusbrevis IFO 15304 1 2 Bacilluscereus IFO 15305T 1 1 0282 and Aspergillusoryzae IFO 4206),plant (wheat, maize, po- B. cereus ATCC 10876 1 1 tato,and canola), and animal (cow, chicken, and salmon) was B. cereus ATCC 11950 1 1 preparedor purchased as described previously (16). DNA was B. cereus ATCC 10987 1 1 quantiŽed by measuring the absorbance at 260nm, and the DNA B. cereus DSM 4312 1 1 samples,diluted at the indicated concentration, were prepared. B. cereus DSM 4313 1 1 B.thuringiensis ATCC 10792T 1 1 PCR primers. Asetof broad-range PCR primers(BS primers) B.thuringiensis ATCC 13366 1 1 wasconstructed from the conserved regions of the 16S rRNA gene B.thuringiensis ATCC 33679 1 1 ofmembers of the Bacillus and Staphylococcus generabut was B. mycoides NCIMB 13305T 1 1 designednot to react with chloroplast 16S rRNA genes (21). The B.weihenstephanensis DSM 11821 1 1 sequenceof the forward primer (positions 83 through 107 in the B.licheniformis JCM 2505T 1 1L Escherichiacoli 16SrDNA) was5 9-CTTGCTCCTCTGAAGT - B.megaterium JCM 2506T 1 1 TAGCGGCG-39,andthe sequence of the reverse primer (positions B. pumilus JCM 2508T 1 1 422through 454 in the E. coli 16SrDNA) was5 9-TGTTCTTCCC- B.circulans IFO 13626 1 1 TAATAACAGAGTTTTACGACCCG -3 9.BS primersare expected B. subtilis IFO 13719T 1 1 toamplify a PCR productof 380to 382 bp (in the case of B. cereus, B.sphaericus IFO 15095 1 1 381bp; in the case of S. aureus, 380bp). Primers prbac1 and B. lentus IFO 15655 1 1 prbac2(prbac primers), reported to be eubacteria-speciŽ c universal Staphylococcusaureus IFO 3060 1 1 PCR primers,were used to conŽ rm the presence of bacterial DNA S. aureus JCM 2874 1 1 (18), andthe PCR conditionswere as described in the ‘ ‘PCR pro- S.epidermidis JCM 2414T 1 1 cedures’’ sectionbelow. Prbac primers are expected to lead to the S.intermedius JCM 2422T 1 1 formationof an amplicon of 296to 300 bp (in the case of E. coli, S. hyicus subsp. hyicus JCM 2423T 1 1 296bp). These primers were synthesized and puriŽ ed with an oli- S. simulans JCM 2424T 1 1 gonucleotidepuriŽ cation cartridge (Sawady T echnologyCo. Ltd., Aerococcusviridans IFO 12219T 1 1 Tokyo,Japan). A. urinae IFO 15544T 1 1 PCR procedures. Onemicroliter of each DNA samplewas Listeriagrayi ATCC 19120T 1 1L usedin the PCR assay,and a Žnalvolume of 20 mlwasused for J.FoodProt., Vol. 67, No. 6 PCRDETECTIONOF BACILLUS AND STAPHYLOCOCCUS 1273

PCR ampliŽcation. On the basis of avolumeof 19 mlperreaction roomtemperature, and 1 mlofthe supernatant was used for PCR mixture (20 ml),the master mixture containing 2 mM MgCl 2, ampliŽcation. B. cereus and S. aureus cellswere counted after 0 0.005%of SYBR GreenI (FMC Bioproducts,Rockland, Maine), to6 hofenrichment culturing on NGKG agarwith egg yolk 250nM forwardprimer ,and250 nM reverse primer was prepared (Nissui)and mannitol salt agar with egg yolk, respectively, ac- withthe R-PCR versionof the T akaraEx Taq kit(T akaraBio cordingto the standard method. Inc.,Otsu, Japan), which contained 10 3 R-PCR buffer(Mg 21 free),a dNTP mixture(dA TP,dTTP,dCTP,anddGTP), and Taq PCR detectionof indicative bacteria ( B. cereus or S. au- DNA polymerase(hot start type). The PCR wasperformed with reus)in noodlesamples containing endogenous micro ora. Af- aLightCycler(Roche Diagnostics GmbH, Mannheim,Germany). ter5 to6hofenrichment culturing of noodlefood samples (listed TheampliŽ cation conditions were as follows: one cycle of de- inT able3) in the presence and absence of the addition of indic- naturationat 95 8Cfor1 minand 30 cycles of heating at 20 8C/s ativebacteria ( B. cereus or S. aureus),DNA sampleswere pre- to 958C,holdingat 95 8Cfor5 s,coolingat 20 8C/s to 598C, hold- paredfrom the culture and used for PCR ampliŽcation according ing at 598Cfor10 s, heating at 10 8C/s to 728C,andholding at tothe procedures described above. Noodle samples (homogenate 728Cfor20 s. Fluorescent product was collected at the last step inBHI medium)were contaminated with serial dilution of B. ce-

2 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 ofeach cycle. After ampliŽ cation, a meltingcurve was obtained reus or S. aureus rangingfrom 0 to10 CFU/ml.Endogenous byheating at 20 8C/s to 958C,coolingat 20 8C/s to 608C, holding microora counts for noodle samples were determined according at 608Cfor10 s, and slowly heating at 0.2 8C/s to 978C with tothe standard method on BHI agar. uorescencecollection at 0.2 8Cintervals.For easier visualization RESULTS AND DISCUSSION ofthe melting temperature ( Tm)ofPCR product, uorescence meltingpeaks were derived from the initial  uorescencemelting ThespeciŽ city of BS primerswas conŽrmed by PCR curves( uorescence [ F]versustemperature [ T])byplotting the assaysof the 81 bacterial strains listed in T able1. Results negativederivative of  uorescenceover temperature versus tem- were positivefor everytested strain belonging to the Staph- dF dT T T perature (2 / versus ).Thevalues of m obtainedfrom the ylococcus or Bacillus genus;the amount of PCR product meltingcur veswere used to estimate the speciŽ city of the PCR for B.licheniformis was low.The reason appears to be a assay (2): observed Tm valuesfor most PCR productsampliŽ ed withBS primerswere 91.1 6 1.38C.ThePCR products(4 ml) mismatchingof forward primersequence (of theBS prim- wereseparated by electrophoresis in 1.8% (wt/ vol)agarose gels, ers) tothe bacterial DNA sequencetargeted. In addition, andthe separated gels were stained with ethidium bromide and results for Aerococcusviridans, Aerococcus urinae, and visualizedwith UV light. FX174 HaeIIIdigestionDNA (Takara) Listeriagrayi were positive;the amount of L. grayi PCR wasused as a molecularmass marker .DNAsequencingof productwas low.The bacteria of the Aerococcus genus PCR productswas carried out with an Applied Biosystems 310 havebeen rarely detected in commercial noodle foods, but GeneticAnalyzer (Foster City, Calif.) and sequence similarity atleastone strain ( A. urinae)ispathogenic (23);therefore, searcheswere carried out with a BLAST searcher(http:/ / we considerthe Aerococcus-positivereaction a permissible www.ncbi.nlm.nih.gov/BLAST/).Allexperiments described be- resultfrom thepoint of view of detecting foodborne path- low,including PCR andbacterial enumeration, were conducted in ogenicor putrefactive bacteria. The positive reaction with duplicate. lowerPCR productfor L. grayi (nonpathogenicbacteria) Preparationof DNA fromartiŽ cially contaminated bac- was anunexpected false-positive reaction. For the L. grayi teriain foodsamples. Freshfoods were purchased from local PCRproductampliŽ ed with BS primers,we sequencedthe foodstores and plated on BHI agarto conŽ rm the absence of product,and the sequence obtained was homologousto that bacterialcontamination; the detection limit for this plating pro- of the L. grayi 16SrRNA gene(e.g., accession numbers cedurewas 10 CFU/ goffood. Sterilized whole milk was artiŽ - X98526and X56150; data not shown). BS primersseemed ciallycontaminated directly and used for the enrichment culture notto cross-react well with the L. grayi 16SrRNA gene; step.Samples of all of the other foods (10 g each)in 90 ml of however,we willhave to becareful in thecase of thisfalse- BHI mediumwere aseptically homogenized with a stomacher positivereaction. For allof theother bacterial strains tested, (Stomacher400, Seward, London, UK) athigh speed for 1 min. resultswere negative.Meanwhile, all of thetested bacterial Thehomogenate was then artiŽ cially contaminated and used for theenrichment culture step. Milk and other food samples listed DNAs formedamplicons with the prbac primers that were inT able2 wereartiŽ cially inoculated with serial dilutions of B. reportedto be eubacteria-speciŽ c universalPCR primers cereus or S. aureus, rangingfrom 0 to10 2 CFU/ml.After 0 to6 targetedto thebacterial 16S rRNA gene (18). Theseresults hofculturingat 35 8C(enrichmentculture step), 1.3 ml of culture demonstratethat PCR with the use of BSprimersis aprac- wascentrifuged at 100 3 g for1 min,and 1 mlof the supernatant ticalmethod for thedetection of bacteria belonging to the wascentrifuged at 13,000 3 g for3 minat room temperature. Bacillus and Staphylococcus genera,although BS primers Thepellets were washed with 1 mlof Tris-EDTA buffer(10 mM alsoreact with bacteria belonging to the Aerococcus genus Tris-HCl[pH 8.0] and 0.1 mM EDTA) andwere resuspended in and with L. grayi. 100 mlofachromopeptidasesolution (250 U/ mlachromopeptidase WhenPCR isused for thedetection of bacteria, it is [WakoPure Chemical Industries Ltd., Osaka, Japan] in 10 mM desirableto use PCR primersthat do not react with DNA Tris-HCl[pH 8.0] and 0.1 mM EDTA). Afterincubating at 55 8C derivedfrom theorganisms being tested, especially in the for15 min, 100 mlofChelex solution (10% [wt/ vol]Chelex 100 resin[Bio-Rad Laboratories Inc., Hercules, Calif.] in 10 mM Tris- caseof PCRprimerstargeted to the gene-encoding bacterial HCl [pH8.0] and 0.1 mM EDTA) wasadded to the resuspension, 16SrRNA. As shownin Figure 1, the BS primersof this incubated,and mixed by vigorousshaking for 10 s. After heating PCRproceduredetected bacterial DNA ( B. cereus and S. at 998Cfor5 minand cooling on ice, the resin and cell debris aureus)butdid not detect four plants (wheat, maize, potato, wereprecipitated by centrifugation at 13,000 3 g for 1 min at andcanola), three animals (cow, chicken, and salmon), or 1274 NAKANO ETAL. J.FoodProt., Vol. 67, No. 6

TABLE 2. PCR detectionof Bacilluscereus and Staphylococcusaureus ininoculated and uninoculated food samples with undetectable levelsof endogenous microbial contamination after an enrichment culture step and incubation for 0 to6 h

PCRresultand cell count(CFU/ ml ofculture) b Inoculum Sample levela 0 h 4 h 5 h 6 h

Sterilizedwhole milk c 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 100 (B) 2 (1.0 3 100) 1 (1.9 3 103) 1 (3.6 3 104) 1 (9.4 3 105) 100 (S) 2 (1.0 3 100) 2 (1.5 3 102) 1 (1.0 3 103) 1 (8.6 3 103) Saltedbutter 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (B) 2 (1.0 3 100) 1 (2.6 3 103) 1 (4.0 3 104) 1 (4.3 3 105) 101 (S) 2 (1.0 3 100) 2 (1.7 3 102) 1 (1.3 3 103) 1 (1.2 3 104) Heavywhipping cream 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 1 0 3 4 5 10 (B) 2 (1.0 3 10 ) 1 (3.3 3 10 ) 1 (2.9 3 10 ) 1 (5.6 3 10 ) Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 101 (S) 2 (1.5 3 100) 2 (2.5 3 102) 1 (1.8 3 103) 1 (1.8 3 104) Ham 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (1.0 3 100) 2 (4.0 3 101) 1 (3.3 3 102) 1 (1.4 3 103) Wiener(Japanese type) 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (1.5 3 100) 2 (1.6 3 102) 1 (1.2 3 103) 1 (1.7 3 104) BoiledŽ shpaste 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (2.0 3 100) 2 (1.9 3 102) 1 (9.5 3 102) 1 (3.2 3 103) TubularŽ shsausage 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (2.0 3 100) 2 (1.1 3 102) 1 (9.4 3 102) 1 (4.2 3 103) Shao-mai(Japanese type) 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (B) 2 (1.0 3 100) 1 (2.1 3 103) 1 (2.1 3 104) 1 (1.0 3 105) 101 (S) 2 (1.0 3 100) 2 (1.7 3 102) 1 (1.5 3 103) 1 (1.2 3 104) Pudding 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (1.5 3 100) 2 (1.0 3 102) 1 (1.6 3 103) 1 (1.2 3 104) Mayonnaise 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (1.0 3 100) 2 (1.1 3 102) 1 (4.2 3 102) 1 (1.5 3 103) BoiledChinese noodle 0 2 (,1) 2 (,10) 2 (,10) 2 (,10) 101 (S) 2 (3.0 3 100) 2 (1.6 3 102) 1 (4.7 3 102) 1 (6.8 3 103) a Inoculumlevels are expressed in terms of cells per milliliter for milk and cells per gram for other samples. B, Bacilluscereus IFO 15305T; S, Staphylococcusaureus IFO 3060. b 1, positive; 2, negative. c ArtiŽcially contaminated milk samples were directly cultured without medium dilution. two fungi (S.cerevisiae and A. oryzae).Theseresults in- allowfor thedetection of these bacteria because of the dicatethat the possibility of false-positivereactions to DNA higherdetection limit for L. grayi. originatingfrom thefoods themselves would be low when Todetectcontaminated bacteria in foods by PCR, sev- BSprimersare used. On the other hand, when prbac prim- eralinvestigators have used DNA isolationand puriŽ cation ers were used,PCR productsof thetargeted size were am- methods (3, 20) andenrichment culture methods (8, 16, 22) pliŽed notonly for bacteriabut also for thefour plants (data toovercome two serious problems: food matrices can in- notshown); we previouslyreported that the sequences am- hibitthe PCR, and the PCR cannotdistinguish the DNA of pliŽed with prbac primers were homologousto chloroplast livingcells from thatof dead cells (6). Theenrichment 16SrRNA gene (16). culturemethod could, to some extent, reduce the inhibitory Astudyof the detection sensitivity of the PCR pro- substancethrough medium dilution of food and could in- cedurewith genomic DNA puriŽed from B. cereus IFO creasea ratioof living cells to dead cells through propa- 15305T showedthat the detection limit was 0.1pg per as- gationof bacteria. Therefore, we implementedan enrich- say,which corresponds to approximately 15 cellsper assay; mentculture step with BHI brothbefore the PCR assayto theamplicon was detectedfor 0.1pg per assay, but no detectbacteria of the Bacillus and Staphylococcus genera ampliconwas detectedfor 0.01pg per assay (data not infood products. Bacterial DNA was preparedby a DNA shown).Similar results were obtainedfor S. aureus IFO extractionmethod with the use of Chelex 100 resin (3, 13, 3060.The detection limit of L. grayi ATCC 19120T was 16) toreduce the potential inhibitory effects of PCR by 50pg per assay (data not shown), which corresponds to foodmatrices and with achromopeptidase (4, 9) for cell approximately10 4 cellsper assay; therefore, our PCR assay lysis.Brie y, cells were harvestedfrom thesupernatant of withthe enrichment culture step described below might not theculture centrifuged at weak force (100 3 g), which J.FoodProt., Vol. 67, No. 6 PCRDETECTIONOF BACILLUS AND STAPHYLOCOCCUS 1275

TABLE 3. PCR detectionwith BS primersof Bacilluscereus and Staphylococcusaureus ininoculated noodle samples containing endogenousmicro ora after 5-h and 6-h enrichment culture steps

Endogenous PCR assayc Sample micro ora Inoculum no. Sample type levela levelb 5 h 6 h

1 Steamedchow mein 6.2 3 103 0 2 2 6.0 3 101 (B) 1 1 5.0 3 101 (S) 1 1 2 Steamedchow mein 1.6 3 104 0 2 2 3.5 3 101 (B) 1 1 3.0 3 101 (S) 2 1 3 Steamedchow mein 1.2 3 103 0 2 2

2.0 3 101 (B) 1 1 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 1.5 3 101 (S) 2 1 4 Boiledwheat noodle 3.5 3 103 0 2 2 2.0 3 101 (B) 1 1 1.0 3 101 (S) 2 1 5 Boiledwheat noodle 2.2 3 104 0 2 2 1.5 3 101 (B) 1 1 2.0 3 101 (S) 2 1 6 Boiledwheat noodle 1.3 3 105 0 2 2 6.0 3 101 (B) 1 1 9.0 3 101 (S) 1 1 7 BoiledChinese noodle 3.6 3 103 0 1 1 2.0 3 101 (B) 1 1 2.5 3 101 (S) 1 1 8 BoiledChinese noodle 1.1 3 104 0 1 1 6.0 3 101 (B) 1 1 5.0 3 101 (S) 1 1 9 BoiledChinese noodle 7.3 3 103 0 1 1 2.0 3 101 (B) 1 1 1.0 3 101 (S) 1 1 a Endogenousmicro ora level is expressed in terms of CFU pergram of sample before culturing, as determined by BHI platecount. b Inoculumlevel is expressed in terms of CFU pergram of sample before culturing. B, Bacilluscereus IFO 15305T; S, Staphylococcus aureus IFO 3060. c 1, positive; 2, negative. resultedin the precipitation of a largefraction of the food 2,andrepresentative PCR results(of sterilizedwhole milk) matrix,but not the cells. The harvested cells were treated areshown in Figure 2. Before the PCR assay,inoculation withachromopeptidase to degrade the bacterial cell walls, ofbacteria at 10 to 30 CFU/ goffood was followedby a andbacterial DNA from theenzyme-treated cells was ex- 4-to 5-h enrichment culture step, and the PCR assayde- tractedwith Chelex 100 resin. At thispoint, no signiŽ cant tectedbacterial cells in several foods: in the case of B. differencebetween the supernatant and the culture (con- cereus, thePCR assaydetected cells after a 4-henrichment tainingbacteria of 10 2 to 103 CFU/ml) withrespect to cell culturestep; in the case of S. aureus, thePCR assayde- countswas observed:the cell count for the B. cereus IFO tectedcells after a 5-henrichment culture step. For steril- 15305T supernatantwas 88%of that for theculture, and izedwhole milk, either B. cereus or S. aureus was inocu- thecell count for the S. aureus IFO 3060supernatant was latedinto milk and directly incubated without BHI medium 97%of thatfor theculture (data not shown). Moreover ,no dilution.When bacteria were inoculatedinto milk at 1to2 false-positivereactions from tracesof bacterial DNA that CFU/ml,the PCR assaydetected bacteria after a 4-to 5-h canoccur in the achromopeptidase were observedon the enrichmentculture step: in the case of B. cereus, the PCR PCRassaywith BS primers(data not shown). assaydetected bacteria after a 4-henrichment culture step; Totestthe ability of the PCR assaydescribed above to inthe case of S. aureus, thePCR assaydetected bacteria detect Bacillus and Staphylococcus genera,either B. cereus aftera 5-henrichment culture step. Meanwhile, when the IFO 15305T or S. aureus IFO 3060was inoculatedinto enrichmentculture was testedfrom 0to6 hwithoutthe variousfoods (with undetectable levels of endogenousmi- inoculationof bacterial cells into food samples, the PCR crobialcontamination) as indicators and incubated at 35 8C assaydid not detect the cells, indicating that no false-pos- for 0to6h;theresults of thesetests are presented in T able itivereaction caused by dead cells was observedin these 1276 NAKANO ETAL. J.FoodProt., Vol. 67, No. 6

FIGURE 1. AmpliŽcation results obtained with BS primersfor puriŽed DNA fromvarious bacteria and other organisms. Lane 3, Escherichiacoli JCM 1649T; lane 4, Salmonella Typhimurium IFO 13245;lane 5, Staphylococcusaureus IFO 3060;lane 6, Bacilluscereus IFO 15305T; lane 7, Triticumaestivum (wheat); PCR detectionof ininoculated and lane 8, Zea mays (maize);lane 9, Solanumtuberosum (potato); FIGURE 2. Bacilluscereus uninoculatedsterilized whole milk samples after an enrichment lane 10, Brassicanapus (canola);lane 11, Bos taurus (cow); lane

culturestep. Samples of milk (with undetectable levels of endog- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 12, Gallusgallus (chicken);lane 13, Oncorhynchus sp.(salmon); enousmicrobial contamination) after artiŽ cial contamination with lane 14, Saccharomycescerevisiae IFO 0282;lane 15, Aspergillus B. cereus IFO 15305T weredirectly cultured without dilution, and oryzae IFO 4206.For lanes 3 through6, 5 pgof template was used;for lane 7, 8 ngof template was used; for lanes 8 through PCR analysiswas carried out after 0 to6 hofculturing. Lanes 12,3 ngof template was used; for lanes 13 through 15, 1 ngof 1and6, molecular mass marker ( FX174 HaeIIIdigestionDNA); templatewas used. Lanes 1 and16, molecular mass marker lanes2 through5, ampliŽ cation products after 0, 4, 5, and 6 h ofculturingwithout inoculation, respectively; lanes 7 through10, (FX174 HaeIIIdigest).Lane 2, negative control (without template ampliŽcation products after 0, 4, 5, and 6 hofculturing with DNA). inoculation(1 to 2 cellsper ml ofmilk), respectively. foodsamples tested. Because the PCR assaycannot dis- criminatebetween DNA ofliving cells and that of dead CFU/gbythis PCR assaymethod. Noodle samples 7 cells,if food samples contain dead cells, this assay cannot through9 (boiledChinese noodle) had endogenous micro- berecommendedfor thedetection of viable cells. However , oraranging from 10 3 to 104 CFU/g.The three samples ifthe in uence of the dead cells is somewhat weak, a rel- withoutthe bacterial inoculation tested positive by thePCR ativecomparison between the amounts of PCR products assay,indicating that they were contaminatedwith bacteria beforeand after the enrichment culture step could be con- (probably Bacillus,Staphylococcus, or Aerococcus genus) sidereda possiblesolution to such a problem. detectablewith BS primers. Somefood samples containing endogenous micro ora Resultspresented in this article indicate that BS prim- willbe encounteredin a practicalassay, and many bacteria ers canbe used for thespeciŽ c detectionof Bacillus and ofendogenous micro ora would grow during enrichment Staphylococcus withoutcross-reactivity to eukaryotic or- culturingwith BHI medium.The presence of thecompeting ganismsand that the PCR assaypresented here with BS microora might cause problems for thisPCR assay:the primersand an enrichment culture step involving BHI me- competingbacteria could outgrow Bacillus or Staphylococ- diumcould be used to detect the bacteria belonging to the cus duringenrichment culturing, or the competing micro- Bacillus and Staphylococcus generain several processed oracould interfere with the PCR assay.T oinvestigatethe foods.Therefore, this PCR assaywould be helpful in mon- effectof competing micro ora on this PCR assay,several itoring Bacillus and Staphylococcus bacteriaand in com- noodlesamples with endogenous micro ora were incubated prehensivelyand rapidly inspecting manufactured foods. at 358Cfor 5to6 hinthe presence or absenceof either B. However,apreliminarytest for thisPCR assaywill be re- cereus or S. aureus (as anindicator). The results of these quiredfor eachtype of food product because targeted bac- testsare presented in T able3. Noodle samples 1 through6 teriamight not grow well in BHI mediumcontaining the (steamedchow mein or boiled wheat noodle) had endoge- foodculture, DNA samplesprepared by thissimple method nousmicro ora ranging from 10 3 to 105 CFU/g.The six mighthave unknown inhibitory substances, or the food sampleswithout the bacterial inoculation tested negative mightcontain excessive bacterial dead cells that are de- aftera 5-to 6-h enrichment culture step; the samples with tectablewith the BS primers.The PCR productsampliŽ ed thebacterial inoculation (10 to 90 CFU/ g)tested positive bythis assay can also be detected by measurement of the aftera 5-to 6-h enrichment culture step. In the case of B. uorescencesignal and by melting curve analysis because cereus inoculation,the PCR assaydetected bacteria after a thePCR assayis performed with a real-timePCR device 5-henrichment culture step; in the case of S. aureus inoc- andSYBR Green I. Althoughwe haveshown in thisarticle ulation,a 6-henrichment culture step was neededfor the thepresence of PCR productsby band separation on aga- unambiguousdetection of theindicative bacteria. These re- rosegels, analyses of PCR-ampliŽed productsby measure- sultsindicate that the endogenous micro ora (10 3 to 105 mentof the  uorescencesignal and by melting curve anal- CFU/g)in the noodle samples might interfere somewhat ysisare faster thanby agarose gels. Hence, instead of an- withbacterial growth (of S. aureus especially)during en- alyzingPCR-ampliŽ ed products by agarose gels, we are richmentculturing and that a 6-henrichment culture step currentlyusing this PCR assayexperimentally in a food isrequired to detect the Bacillus or Staphylococcus genus factoryby measuring the  uorescencesignal and by ana- innoodle foods with a detectionsensitivity of 10 to 90 lyzingmelting curves to monitor bacteria of the Bacillus J.FoodProt., Vol. 67, No. 6 PCRDETECTIONOF BACILLUS AND STAPHYLOCOCCUS 1277 and Staphylococcus generathat might contaminate noodle gram-positiveand gram-negative bacteria. J.Clin.Microbiol. 40: productsor their sources. 4304–4307. 12.Klausegger ,A.,M.Hell,A. Berger,K. Zinober,S. Baier,N.Jones, W.Sperl,and B. Koer. 1999.Gram type–speciŽ c broad-rangePCR REFERENCES ampliŽcation for rapid detection of 62 pathogenic bacteria. J. Clin. 1.Bennet, R. W.,andN. Belay.2001. Bacilluscereus, p. 311–316. In Microbiol. 37:464–466. F.P.Downesand K. Ito(ed.), Compendium of methodsfor the mi- 13.Kobayashi, K., M.Taguchi,and K. Seto.1994. Polymerase chain reactionmethod for the detection of Salmonella byusing Chelex crobiologicalexamination of foods, 4th ed. American PublicHealth resinafter enrichmentculture. KansenshogakuZasshi. 68:1203– Association,Washington, D.C. 1210. 2.Bohling, S. D.,T .C.King,C. T.Wittwer,and K. S.J.Elenitoba- 14.Lancette, G.A.,andR. W.Bennett.2001. Staphylococcusaureus Johnson.1999. Rapid simultaneous ampliŽ cation and detection of andstaphylococcal enterotoxins, p. 387– 403. In F.P.Downesand theMBR/ JHchromosomaltranslocation by  uorescence melting K.Ito(ed.), Compendium of methods for the microbiological ex- curveanalysis. Am.J. Pathol. 154:97–103. aminationof foods, 4th ed. American PublicHealth Association, 3.Brasher ,C.W.,A.DePaola,D. D.Jones,and A. K.Bej.1998. Washington,D.C. Detectionof microbial pathogens in shellŽ sh with multiplex PCR. 15.Lu, J. J.,C. L.Perng,S. Y.Lee, andC. C.Wan.2000. Use ofPCR Curr.Microbiol. 37:101–107. withuniversal primers andrestriction endonuclease digestions for Downloaded from http://meridian.allenpress.com/jfp/article-pdf/67/6/1271/1677289/0362-028x-67_6_1271.pdf by guest on 01 October 2021 4.Carroll, K. C.,R. B.Leonard,P .L.Newcomb-Gayman, andD. R. detectionand identiŽ cation of common bacterial pathogensin cere- Hillyard.1996. Rapid detection of the staphylococcal mecA gene brospinal uid. J.Clin.Microbiol. 38:2076–2080. fromBACTEC blood culture bottles by the polymerase chain reac- 16.Nakano, S., T.Kobayashi,K. Funabiki,A. Matsumura,Y .Nagao, tion. Am.J. Clin.Pathol. 106:600–605. andT .Yamada. 2003.Development of a PCRassay fordetection of 5.Greisen, K.,M. Loeffelholz,A. Purohit,and D. Leong.1994. PCR Enterobacteriaceae in foods. J.FoodProt. 66:1798–1804. primers andprobes for the 16S rRNA geneof mostspecies ofpath- 17.Phelps, R. J.,andJ. L.McKillip.2002. Enterotoxin production in ogenicbacteria, includingbacteria foundin cerebrospinal  uid. J. naturalisolates of Bacillaceae outside the Bacilluscereus group. Clin.Microbiol. 32:335–351. Appl.Environ. Microbiol. 68:3147–3151. 6.Hill, W .E.1996.The polymerase chain reaction: applications for 18.Rupf, S., K.Merte,and K. Eschrich.1999. QuantiŽ cation of bacteria thedetection of foodborne pathogens. Crit.Rev. FoodSci. Nutr. 36: inoral samples bycompetitive polymerase chain reaction. J. Dent. 123–173. Res. 78:850–856. 7.Hough, A. J.,S. A.Harbison,M. G.Savill,L. D.Melton,and G. 19.Stevenson, K. E.,andW .P.Segner.2001.Mesophilic aerobic spo- Fletcher.2002. Rapid enumeration of Listeriamonocytogenes in ar- reformers, p.223– 227. In F.P.Downesand K. Ito(ed.), Compen- tiŽcially contaminated cabbage using real-time polymerasechain re- diumof methods for the microbiological examination of foods, 4th action. J.FoodProt. 65:1329–1332. ed.American PublicHealth Association,Washington, D.C. 20.Tamarapu, S., J.L.McKillip,and M. Drake. 2001.Development of 8.Hsu, S. C.,and H. Y.Tsen.2001. PCR primers designedfrom malic amultiplexpolymerase chain reaction assay fordetection and dif- acid dehydrogenasegene and their use fordetection of Escherichia ferentiationof Staphylococcusaureus indairy products. J. Food coli inwater andmilk samples. Int.J. FoodMicrobiol. 64:1–11. Prot. 64:664–668. 9.Imafuku, Y .,A.Nozawa, H.Okamura,H. Yoshida,M. Saito,F .Tan- 21.Tohdoh, N., and M. Sugiura.1982. The complete nucleotidese- abe,and M. Ogata.1994. Method of DNA extractionand mecA, quenceof 16Sribosomal RNA genefrom tobacco chloroplasts. Gene femA and PBP29 inmulti-drug resistant bacterial strains. Rinsho By- 17:213–218. ori. 42:669–675. 22.Wang, R. F.,W.W.Cao,and C. E.Cerniglia.1997. A universal 10.Jousimies-Somer ,H.1997.Recently described clinically important protocolfor PCR detectionof 13species offoodborne pathogens in anaerobicbacteria: taxonomicaspects andupdate. Clin.Infect. Dis. foods. J.Appl.Microbiol. 83:727–736. 25:S78–S87. 23.Zhang, Q., C. Kwoh,S. Attorri,and J. E.ClarridgeIII. 2000. Aero- 11.Klaschik, S., L.E.Lehmann,A. Raadts,M. Book,A. Hoeft,and F . coccus urinae inurinary tract infections. J. Clin.Microbiol. 38: Stuber.2002.Real-time PCRfordetection and differentiation of 1703–1705.