Recent advances in quantitative PCRapplications in food microbiology(1).pdf

ReviewRecent advances in quantitative PCR(qPCR)applications in food microbiologyFlorence Postolleca,*,Hlne Falentinb,c,Sonia Pavana,Jrme Combrissond,Danile SohieraaADRIA Dveloppement-UMT 08.3PHYSIOpt,Creach Gwen,29196 Quimper,FrancebINRA,UMR 1253,Science et Technologie du Lait et de luf,65 rue de St Brieuc,35000 Rennes,FrancecAgrocampus Ouest,UMR1253,Science et Technologie du Lait et de lOeuf,35000 Rennes,FrancedDanone Research,91767 Palaiseau Cedex,Francea r t i c l ei n f oArticle history:Received 17 November 2010Received in revised form14 February 2011Accepted 21 February 2011Available online 1 April 2011Keywords:Quantitative PCRReverse-transcription real-time PCR(RT-qPCR)Food microbiologyPopulation dynamicsa b s t r a c tMolecular methods are being increasingly applied to detect,quantify and study microbial populations infood or during food processes.Among these methods,PCR-based techniques have been the subject ofconsiderable focus and ISO guidelines have been established for the detection of food-borne pathogens.More particularly,real-time quantitative PCR(qPCR)is considered as a method of choice for the detectionand quantification of microorganisms.One of its major advantages is to be faster than conventionalculture-based methods.It is also highly sensitive,specific and enables simultaneous detection ofdifferent microorganisms.Application of reverse-transcription-qPCR(RT-qPCR)to study populationdynamics and activities through quantification of gene expression in food,by contrast with the use ofqPCR,is just beginning.Provided that appropriate controls are included in the analyses,qPCR andRT-qPCR appear to be highly accurate and reliable for quantification of genes and gene expression.Thisreview addresses some important technical aspects to be considered when using these techniques.Recent applications of qPCR and RT-qPCR in food microbiology are given.Some interesting applicationssuch as risk analysis or studying the influence of industrial processes on gene expression and microbialactivity are reported.?2011 Elsevier Ltd.All rights reserved.1.IntroductionIn the last two decades,culture-independent molecular app-roaches have undergone considerable development in microbialecology.Techniques enabling analyses of total microbial commu-nities have greatly improved our understanding of their composi-tion,dynamics and activity(e.g.Wilmes and Bond,2009;Zoetendalet al.,2008).A few years ago,a system based on quantitative PCRamplification of specific sequences was developed to rapidlyquantify human intestinal bacteria(Yif-Scan proprietary system,Yakult Honsha Co,Ltd).In food microbiology,the first culture-independent application of molecular methods to a fermented foodmatrix was described in 1999(Ampe et al.,1999).Nowadays,PCR-based methods,in particular quantitative PCR,are used predomi-nantly to detect,identify and quantify either pathogens or benefi-cial populations such as fermenting microbes or probiotics(LeDran et al.,2010;Malorny et al.,2008;Masco et al.,2007).ISOstandards have also been established and provide guidelines toqualitatively detect food-borne pathogens by PCR(ISO 22174:2005,ISO/TS 20836:2005,ISO 20837:2006,ISO 20838:2006).However,incomparisonwith environmental microbiology,the use of moleculartools applied to the study of population dynamics and geneexpression in food is only starting(Falentin et al.,2010;Juste et al.,2008;Smith and Osborn,2009).Recent publications have shownthe possibility to follow the growth and activity of microbial pop-ulations in complex environments and highlight the potential ofmolecular approaches in assisting to control industrial processes(Hagi et al.,2010;Nakayama et al.,2007).Compared with culture-based methods,PCR is faster,moresensitive and more specific and enables detection of sub-dominantpopulations,even in the absence of a selective enrichment mediumand in the presence of other(dominant)populations.Moreover,itallows detection of dead cells or viable but non-cultivable cells.Real-time PCR(thereafter named qPCR for quantitative PCR)offersthepossibilitytoquantifymicrobialpopulationsthroughmeasurement of gene numbers.Combined with reverse transcrip-tion(RT),qPCR can also estimate transcript amounts,thereforeproviding data on microbial activity.Currently,qPCR and RT-qPCRhave become the methods of choice to quantify genes and geneexpression,respectively(Nolan et al.,2006).Nucleic acid isolation*Corresponding author.Tel.:33 298101812;fax 33 298101808.E-mail address:florence.postollecadria.tm.fr(F.Postollec).Contents lists available at ScienceDirectFood Microbiologyjournal homepage: see front matter?2011 Elsevier Ltd.All rights reserved.doi:10.1016/j.fm.2011.02.008Food Microbiology 28(2011)848e861and qPCR preparation can be automated and,depending on thedetection system,the molecular method can be relatively inex-pensive and suitable for routine analysis.Compared with end-pointPCR,qPCR and RT-qPCR(these two techniques will be hereafterdefined with the single abbreviation(RT-)qPCR)do not requirepost-amplificationmanipulations,hencelimitingtheriskofcontamination.In addition,they are more sensitive,and accuratetemplate quantification is allowed over a wide dynamic range(7e8 log)(Bustin et al.,2005).However,due to very high sensi-tivity,(RT-)qPCRexperimentsshouldbecarefullydesigned.Provided that proper controls are carried out,this techniqueappears to be the most accurate and reliable for genes or transcriptsquantification(Bustin,2009).qPCRandRT-qPCRtechnologieshavebeenextensivelydescribed in other reviews(e.g.Heid et al.,1996;Kubista et al.,2006;VanGuilder et al.,2008;Wong and Medrano,2005).Inbrief,similarly to end-point PCR,qPCR consists in a succession ofamplification cycles in which the template nucleic acid is dena-tured,annealedwithspecificoligonucleotideprimers,andextended to generate a complementary strand using a thermo-stable DNA polymerase.This results in exponential increase ofamplicons(amplification products)that,in contrast with end-pointPCR,can be monitored at every cycle(in real time)using a fluo-rescent reporter.The increase in fluorescence is plotted against thecycle number to generate the amplification curve,from whicha quantification cycle Cq(often described as Ct for cycle threshold)value can be determined.Cq corresponds to the number of cyclesfor which the amount of fluorescence(hence,of template)issignificantly higher than the background fluorescence.Therefore,the Cq value can be linked to the initial concentration of targetnucleic acid and serves as a basis for absolute or relative templatequantification(see below).Several detection chemistries are nowavailable with well-described protocols(Wong and Medrano,2005).As each of them is displaying specific characteristics,theirchoice will depend on the application.Currently in food microbi-ology,the two most popular detection systems are the DNA bindingdye technologyand the 50nuclease assay.While the first one is verywell adapted to low-cost routine analyses(among other charac-teristics),the second technology enables the screening of multipletarget genes within a single reaction(multiplex PCR).First,this review highlights some important technical aspects toconsider in food microbiology when designing or using(RT-)qPCRor when analyzing the results,with respect to the current scientificknowledge and also to our own field experience.In a second part,recent applications of(RT-)qPCR to quantify genes or transcripts infood samples are presented,with the aim to provide an overviewabout the possible range of applications of these methods.2.Technical considerations for(RT-)qPCR implementation infood microbiologyIn this section,we would like to point out some aspects of(RT-)qPCR protocolsthat are notoften raised in technical papers and thatare essential in food microbiology.Other“basic”aspects such asprimer design,choice of reagents,etc.are not discussed here.2.1.Quality of nucleic acid extractsNucleic acid extraction is the first step in the analysis processand sample quality is probably the most important component toensure reproducibility of the analysis and to preserve the biologicalmeaning(Bomjen et al.,1996;Bustin and Nolan,2004).Nowadays,it is easy to isolate DNA with very high qualitative and quantitativeyields.Most procedures employ commercial extraction kits,used assuch or with some adaptations depending on the food matrix,withsatisfactory results.By contrast with DNA,intact RNA extraction ismore laborious,especially from complex or fatty food matrices.Some extraction methods compatible with subsequent RT-qPCRhave been developed for various foods(de Wet et al.,2008;Hierroet al.,2006;Rantsiou et al.,2008;Ulve et al.,2008).Due to fastdegradation,RNA should be quickly analyzed.Currently automatedcapillary-electrophoresis equipment(e.g.Bioanalyzer 2100,Agi-lent)is the most appropriate to determine sample quality.A RNAintegrity number(RIN)can be calculated(Schroeder et al.,2006)todetermine suitability of samples for RT-qPCR analysis(Fleige andPfaffl,2006).In spite of these technical breakthroughs,upstreamsteps of the detection procedure,i.e.sampling and sample prepa-ration,often remain overlooked in comparison with the analyticalpart(Brehm-Stecher et al.,2009).2.2.Detection chemistriesSeveral reporter systems are available.A description of theirmode of action,advantages and limitations can be found elsewhere(e.g.Wong and Medrano,2005).In food microbiology,essentiallytwo detection chemistries are commonly used:the DNA bindingdye assay using SYBR?Greenas a fluorophore(Wittweret al.,1997),and the hydrolysis probe method(or 50nuclease assay)(Gibsonet al.,1996)mostly employing the TaqMan?probe(Applied Bio-systems)assay.As SYBR?Green binding is not specific for a targetsequence this system can be readily used for different gene assays,is flexible,inexpensive,and accurate results can be obtainedprovided validation of the specificity by melt curve(or dissociationcurve)analysis.The TaqMan?chemistry is more expensive thanDNA binding dye assays,but presence of the hydrolysis probeensures that only specific amplicons is measured.In addition,multiplexing reactions are possible,although their set up requiresan important optimization phase.2.3.Quantification methodsAccurate quantification is of prime importance for most foodmicrobiology applications.Absolute quantification is based oncomparison of Cq values with a standard curve generated fromamplification of known amounts of the target gene.This methodrequires similar amplification efficiencies(see below)for allsamples and standards.Therefore,the standard curve templatemust be carefully chosen(Dhanasekaran et al.,2010;Leong et al.,2007;Malorny et al.,2008;Whelan et al.,2003).Relative quanti-fication is used to estimate changes in gene expression.It is basedon the use of an external standard or a reference sample.Thequantification results are expressed as a target/reference ratio.Several mathematical models have been set up(see for review(Wong and Medrano,2005).Depending on the quantificationmethod chosen different results can be observed(Cikos et al.,2007).Compared to absolute quantification,relative quantifica-tion is simpler as it does not necessitate setting up a reliablestandard to be included in every PCR.However,it can be appliedonly to the samples run within the same PCR.To compare differentPCRs,a reference control must be included in every run(Wong andMedrano,2005).Amplification efficiency is important to consider when relativequantification is performed,as many PCR do not display idealefficiency(presence of inhibitors,nucleotide variability).It is rec-ommended to calculate and report amplification efficiency valuesfor each amplicon(Smith and Osborn,2009;Tuomi et al.,2010),especially when Cq values are to be compared between differentsamples originating from different food matrices,or when differentstrains are quantified.F.Postollec et al./Food Microbiology 28(2011)848e861849Table 1Some applications of(RT)-qPCR in food microbiology.MicroorganismTarget geneApplicationTest characteristicsFood matrixReferenceqPCR studiesSalmonella spp.invADetectionEnrichmentqPCR?TaqMan?,IACaDLb:?2.5 CFU/25 g salmon and mincedmeat,5 CFU/25 g chicken meat,5 CFU/25 ml milkArtificially contaminatedchicken meat,minced meat,salmon,raw milk(Hein et al.,2006)Salmonella spp.invADetectionEnrichmentqPCR?LightCycler?hybridization probes,IACDL:5 cells/25 gArtificially contaminated fish,minced beef,raw milkNaturally contaminated rawmilk and meat(Perelle et al.,2004)Salmonella spp.invADetectionEnrichmentqPCR?TaqMan?DL:0.08 or 0.2 CFU/g(24 h-enrichmentor 48 h-enrichment)Artificially contaminatedmashed potatoes,soft cheese,chilli powder,chocolate,eggs,sprouts,apple juice,fish,shrimp,ground beef,ground chicken(Cheng et al.,2009)Salmonella spp.ssrADetectionEnrichmentqPCR?TaqMan?,IACDL:1e10 CFU/cm2Artificially contaminated freshmeat carcasses(McGuinnesset al.,2009)Salmonella spp.iagADetectionEnrichmentqPCR?Molecular BeaconDL:4 CFU/25 gArtificially contaminated cantaloupe,mixed-salad,cilantro,alfalfa sprouts(Liming andBhagwat,2004)Salmonella entericainvADetectionEnrichmentqPCR?TaqMan?DL:10 CFU/mlArtificially contaminated chickenNatural chicken samples(Hong et al.,2007)Pectinatus,Megasphaera,Selenomonas,Zymophilus species16S rRNADetectionqPCR?SYBR?GreenDL:1e103CFU/25 mlArtificially contaminated beerReal brewery samples(Juvonen et al.,2008)Yersinia pestisPlasmid sequences(four sets of primers),cnp60DetectionQuantificationqPCR?TaqMan?DL:101e103CFU/ml(milk),102e105CFU/g(beef)Artificially contaminated milk,ground beef(Amoako et al.,2010)Genera and species of spore-formingfood bacteria16S rRNAspecific genes(commercial biochip)DetectionEnrichmentmultiparametricqPCR,TaqMan?DL:?1 spore/25 g B.cereus(variabledepending on complexity of food matrix)Artificially contaminated andnatural samples of cream cheese,curd,milk powder,fishsoup,sausage-lentils,couscous,pasteurized whole liquid egg,egg white,whole eggpowder(Postollec et al.,2010)Clostridium tyrobutyricum sporesflaDetectionQuantificationqPCR?TaqMan?,IACQL:25 spores/25 mlArtificially contaminatedraw milk,heat-treated milk(Lopez-Enriquezet al.,2007)Campylobacter,Salmonella spp.16S rRNA,invADetectionQuantificationMultiplex qPCR?hybridization probesDL:3?103CFU/mlArtificially contaminatedchicken skin rinses(Wolffs et al.,2007)(continued on next page)F.Postollec et al./Food Microbiology 28(2011)848e861851Table 1(continued)MicroorganismTarget geneApplicationTest characteristicsFood matrixReferenceEscherichia coli O157:H7,Salmonella spp.,Staphylococcus aureusuidA,nuc,oriCDetectionMultiplex qPCR?SYBR?Greenmeltingcurve analysis,TaqMan?DL:103CFU/g for each pathogen(TaqMan?);104CFU/g for E.coliand Salmonella,103for S.aureus(SYBR?Green)Artificially contaminated lettuce(Elizaquivel andAznar,2008)Salmonella spp.,Listeria monocytogenes,Escherichia coli O157:H7invA,hlyA,rfbEDetectionEnrichmentmultiplexqPCR?TaqMan?,IACDL:18 CFU/10 gArtificially contaminatedground beefNatural beef