A-novel-ril-suitable-for-molecular-traceability_2015_Food-Chemistry

Analytical MethodsA novel reliable method of DNA extraction from olive oil suitable for moleculartraceabilityKatia Raieta,Livio Muccillo,Vittorio Colantuoni ⇑Department of Sciences and Technologies,University of Sannio,82100Benevento,Italya r t i c l e i n f o Article history:Received 19February 2014Received in revised form 4August 2014Accepted 17September 2014Available online 28September 2014Keywords:Olive oilDNA extraction SSR analysisCultivar authentication Traceabilitya b s t r a c tExtra virgin olive oil production has a worldwide economic impact.The use of this brand,however,is of great concern to Institutions and private industries because of the increasing number of fraud and adulteration attempts to the market products.Here,we present a novel,reliable and not expensive method for extracting the DNA from commercial virgin and extra virgin olive oils.The DNA is stable overtime and amenable for molecular analyses;in fact,by carrying out simple sequence repeats (SSRs)markers analysis,we characterise the genetic profile of monovarietal olive oils.By comparing the oil-derived pattern with that of the corresponding tree,we can unambiguously identify four cultivars from Samnium,a region of Southern Italy,and distinguish them from reference and more widely used varieties.Through a parentage statistical analysis,we also identify the putative pollinators,establishing an unprecedented and powerful tool for olive oil traceability.Ó2014Elsevier Ltd.All rights reserved.1.IntroductionOlive tree (Olea europaea L.)is one of the oldest and most important crops in the Mediterranean region and includes both cultivated (var.sativa )and wild type (var.oleaster )cultivars.These latter are fully interfertile with the domesticated varieties and are interconnected by sporadic,spontaneous hybridization (Zohary &Spiegel-Roy,1975).Olive is a vegetatively propagated plant tree and more than 1000varieties under cultivation have been selected over the centuries for their adaptability to the microclimate and to the local soils (Rugini &Baldoni,2005).This,however,is only an underestimate due to the lack of data on minor local cultivars in many olive-growing areas.Olive oil production is of great economic impact,for which rea-son olive oil industries,nurseries and olive growers are interested into methods that allow accurate identification of olive varieties.The choice of a specific variety significantly contributes to the quality of the resulting oil.Some olive oil cultivars,in fact,are recognised of high quality as they derive from well-defined geographical areas and consequently they impose market prices.Consistently,many international Institutions are concerned with agricultural productions that combine safety and quality attributes with a clear authentication.To this goal,also the European Union has issued specific rules about olive oil production and its market through the EC Regulation No.865/2004on the Protected Designa-tion of Origin (PDO)mark and Protected Geographical Indication (PGI)mark,(No.510/2006)(Official Journal of European Communities,2004,2006).On the basis of these considerations,it is becoming mandatory to establish extra virgin olive oil (EVOO)authenticity and detect possible adulterations.To this goal,several analytical techniques have been set up:most chemical analyses,however,are of limited impact due to the high variability of oil components such as fatty acids and secondary metabolites in relation to environmental conditions.The European Commission (Regulations No.2568/91,1991)and the International Olive Oil Council (IOOC)(COI/T.20/Doc.No.19/Rev 2,2008)have established official methods based on of physical–chemical parameters (free acidity,peroxide value,fatty acids and specific ultraviolet (UV)absorptions)measurement to characterise EVOO and guarantee its quality according to maxi-mum values that cannot be exceeded.These analyses,nonetheless,do not provide reliable results for oil traceability (Ben-Ayed,Grati-Kamoun,Moreau,&Rebai,2009;Christopoulou,Lazakari,Komaitis,&Kaselimis,2004).DNA-based technologies are gaining increasing attention in the food authenticity field (Rasmussen &Morrissey,2008).DNA-based markers,in fact,have been success-fully applied to overcome problems associated with differences due to environmental growth conditions or to verify food authen-ticity and traceability (Consolandi et al.,2008;Martins-Lopes,/10.1016/j.foodchem.2014.09.1010308-8146/Ó2014Elsevier Ltd.All rights reserved.⇑Corresponding author at:Department of Sciences and Technologies,University of Sannio,Via Port’Arsa,11,82100Benevento,Italy.Tel.:+39(0)824305102;fax:+39(0)824305147.E-mail address:colantuoni@unisannio.it (V.Colantuoni).Gomes,Santos,&Guedes-Pinto,2008;Montemurro,Pasqualone, Simeone,Sabetta,&Blanco,2008;Pafundo,Busconi,Agrimonti, Fogher,&Marmiroli,2010).Their employment in olive oil charac-terisation offers a number of advantages over biochemically-based markers as they provide increased specificity,sensitivity and a reli-able performance with highly processed samples(Montealegre, Marina Alegre,&Garcia-Ruiz,2010).Simple sequence repeats (SSR),amplified fragment length polymorphism(AFLP),random amplified polymorphic DNA(RAPD),inter-simple sequence repeats (ISSR)and single nucleotide polymorphisms(SNP)have been pro-posed to identify olive cultivars.The reliability and reproducibility of the profiles obtained with molecular markers appear,however, to be strictly dependent on the quality of the DNA extracted from oil(Ben-Ayed et al.,2009;Breton,Claux,Metton,Skorski,& Bervillé,2004;Muzzalupo,Pellegrino,&Perri,2007;Testolin& Lain,2005).Therefore,extraction and recovery of DNA from olive oil is a key and hard step in the process as addressed by Agrimonti,Vietina,Pafundo,and Marmiroli(2011).The amount of DNA isolated with the available methods is usually low,the quality is poor and it is often degraded by nucleases(Ben-Ayed et al.,2009; Muzzalupo&Perri,2002;Pafundo,Agrimonti,Maestri,&Marmiroli, 2007;Testolin&Lain,2005).Additionally,phenolic compounds and residual polysaccharides can interfere with subsequent analy-ses inhibiting the activity of DNA polymerase and providing non-specific Polymerase Chain Reaction(PCR)amplifications (Testolin&Lain,2005).The aim of this study was to set up a new procedure to extract DNA from virgin and extra virgin olive oils.The method presented here is simple,reproducible,not expensive and capable to recover stable and high-molecular weight DNA that can be analysed by SSRs markers to achieve a genetic characterisation and distinguish different olive cultivars.Filtered and not-filtered commercial oils packed in any type of container were used as source of DNA yield-ing similar results.In particular,the new protocol allowed the identification of four major varieties from the Samnium area (Campania,Italy).Finally,the genetic profile of the DNA from EVOO was compared with that from embryos,pulps and leaves establish-ing a new highly reproducible standard for the authentication and traceability of the produced oil.2.Materials and methods2.1.SamplesEight olive(O.europeaea L.)cultivars were sampled:Ortice, Ortolana,Racioppella and Ravé(native from Samnium,Southern Italy);Frantoio,Leccino and Pendolino(selected as pollinators and reference cultivars)are spread all-over Italy;Olivastro is a pollinating variety generally used for table olives but not for oil production.All samples came from private nurseries and from the depositoryfield of Campania germplasms(Azienda Sperimen-tale Improsta,Eboli,Salerno).Five accessions were analysed for each cultivar,with the total sample number of forty.Monovarietal filtered oils obtained for each of the four Samnitic cultivars inves-tigated were collected from different oil mills along with the corre-sponding drupes.Five independent oil samples for each cultivar were analysed,twenty oil samples overall.Leaves and drupes were dried and then stored in the dark at room temperature.Also oil samples were kept at room temperature until DNA extraction. 2.2.DNA extraction from olive leaves,pulps and embryosAfter collection,all materials were lyophilized.The DNA was extracted from young leaves using a protocol based on that previ-ously described with minor modifications(Sefc,Lopes,Mendoncxa,Machado,&Da Camara Machado,2000).Briefly,80mg of plant material were frozen in liquid nitrogen and grinded using steel beads and a Tissuelyser,(QIAGENÒ,Germany).Each sample was added with800l l of CTAB(Cetyltrimethylammonium Bromide) buffer(100mM Tris/HCl pH8.0;1.4M NaCl;20mM EDTA(Ethyl-enediaminetetraacetic Acid);2%CTAB;1%PVP(Polyvinylpyrroli-done);0.2%b-mercaptoethanol),incubated at65°C for30min and then added with800l l chloroform–isoamyl alcohol(24:1). The DNA was precipitated from the recovered aqueous phase with 0.7volume of isopropanol and0.1volume of3M sodium acetate and incubated atÀ20°C for15min.The pellet was resuspended in300l l TE(10mM Tris pH8.0,1mM EDTA)buffer and incubated with40l g of Proteinase K(Lifetech,CA,USA)at56°C for2h.After digestion,the supernatant was extracted with phenol–chloroform–isoamyl alcohol(25:24:1).The recovered aqueous phase was added with0.4volume of10M ammonium acetate and4volumes of ethanol and incubated for5min atÀ80°C.After ethanol precip-itation,DNA was resuspended in50l l TE buffer.The same protocol was used for extracting DNA from pulps and embryos with slight modifications.In these cases,the material was added with400l l of CTAB buffer,incubated at65°C for30min,crushed by Tissuely-ser and added with the remaining400l l of CTAB buffer.Then the extraction was carried out as described before.2.3.DNA extraction from olive oilHerein we describe a novel and optimised protocol based on the CTAB–hexane–chloroform method.The extraction was carried out in a2ml microtube by adding600l l of pre-warmed2%CTAB buf-fer to a mixture of600l l of hexane and600l l of olive oil(v/v/v 1:1:1).After mixing by inverting the tube,the sample was centri-fuged(1h,20°C,14,000rpm).All further centrifugation steps were carried out under the same conditions:15min,20°C,14,000rpm. The oily phase was discarded while the aqueous phase was incu-bated at65°C for30min and centrifuged.The pellet was resus-pended in400l l of0.5%CTAB buffer(100mM Tris/HCl pH8.0;1.4M NaCl;20mM EDTA;0.5%CTAB;1%PVP;0.2%b-mercap-toethanol),added with one volume of phenol and centrifuged. The aqueous phase was recovered,added with one volume of chloroform:isoamyl alcohol(24:1)and centrifuged(15min,4°C, 14,000rpm).For DNA precipitation,to the sample was added2.5 volumes of cold ethanol,0.1volume of10M ammonium acetate, and20l g/ml of glycogen,gently mixed by inverting the tube and kept atÀ20°C overnight.After centrifugation,the supernatant was discarded and the pellet washed twice with500l l of70%v/v ethanol/water,centrifuged again and dried.After resuspension in 30l l of TE buffer,the DNA was run in TAE Buffer1Â(40mM Tris, 20mM acetic acid,and1mM EDTA)on a1%agarose gel stained with ethidium bromide(1l g mlÀ1)and visualised under UV light. The integrity of the DNA was evaluated by the lack or presence of low amounts of smeared DNA(degraded DNA).The bands corre-sponding to not degraded DNA were cut with a sterile scalpel under UV light and introduced into a dialysis bag(Spectra/Por Dial-ysis Membrane MWCO:12–14,000/Spectrum Laboratories Inc.) containing TAE buffer1Â.Electroelution was carried out using a field strength of$4–5V cmÀ1for1h until the DNA completely ran out of the band,as checked under UV light.A current in inverse orientation was applied for30s in order to detach the DNA from the dialysis bag and redistribute into the running buffer.The sam-ple recovered was added with1volume of chloroform:isoamyl alcohol(24:1),centrifuged at4°C and precipitated atÀ80°C for 30min adding2.5volumes of cold ethanol,100l l of3M sodium acetate,and20l g/ml of glycogen.The pellet was then rinsed twice with500l l of70%v/v ethanol/water,centrifuged,dried and resus-pended in TE buffer.Eight independent extractions of olive oil were carried out for each oil sample and the total DNA of the replicatesK.Raieta et al./Food Chemistry172(2015)596–602597was quantitated by spectrophotometric measurement(Jenway Genova Plus Spectrophotometer–Bibby Scientific Ltd.,Dunmow, UK)and pooled in a single tube for subsequent analyses.This pro-cedure has extensively been reproduced and DNA recovered from each extraction in a concentration range of200–600ng/ml of oil. DNA was extracted also from commercially available olive oils packed either in cans or glass or PET bottles.Starting from the same amount of oil,we recovered not degraded DNA in the same concentration range reported for our samples.2.4.SSR analysisFive SSR markers were selected amongst those available in the literature.The loci used were three(GAPU17,GAPU59,GAPU71A) from Carriero,Fontanazza,Cellini,and Giorio(2002),and two (UDO099028,UDO099039)from Cipriani,Marrazzo,Marconi, Cimato,and Testolin(2002).Amplification of SSR markers by the PCR was performed in a20l l reaction volume containing40ng of DNA,0.5l M forward and reverse primers(forward primers were labelled at50-end with either the FAM,or HEX,or ATTO 565fluorescent dyes),50l M of each dNTPs,0.25U AmpliTaq Gold DNA Polymerase(Applied Biosystems,USA),1Âsupplied PCR buf-fer,1.5mM MgCl2.The following PCR cycle was used on a T gradi-ent ThermoCycler(Biometra,Germany):95°C for2min for1cycle; 95°C for1min,annealing temperatures for1min,and72°C for 1min for40cycles;then72°C for7min.Triplicates of each PCR were carried out for each DNA extraction.The amplification prod-ucts were detected with the genetic analyser ABI PRISM3130 (Applied Biosystems,Foster City,CA,USA)using the GeneScan internal size standard LIZ500(PE Applied Biosystems).Data were collected and allele size determined using the Peak Scanner™v.1.2 software(Applied Biosystems).2.5.Data analysisThe genetic relationships amongst genotypes and Principal Coordinates Analysis(PCoA)were estimated by the Nei index (Nei&Roychoudhury,1972)using the GenAlEx 6.41software (Peakall&Smouse,2006;.au/BoZo/GenAlEx/). As for the genetic analysis,the allele number and frequencies, the expected(He)and observed(Ho)heterozygosity,null alleles frequencies,Polymorphic Index Content(PIC),probability of non exclusion and Delta LOD score(Delta Logarithm ODds Score) were calculated using the CERVUS3.0.3software(http://www.fi/pages/home.jsp,Marshall1998–2007).The Power of Discrimination(PD)was evaluated by the formula proposed by Kloosterman,Budowle,and Daselaar(1993).A dendrogram was generated based on the Unweighted Pair Group Method of analysis(UPGMA)with the SHAN algorithm applied to a Nei matrix using R package‘‘Cluster’’v.1.15.2 (Maechler,Rousseeuw,Struyf,Hubert,&Hornik,2014)(http:// /,CopyrightÓ2011The R Foundation for Statistical Computing).3.Results and discussion3.1.DNA extraction and isolation from extra virgin olive oilOne of the main goals of this study was to set up an efficient and not expensive protocol for the extraction of high molecular weight DNA from extra virgin olive oil with a good yield and of high purity for subsequent analyses.To this end many efforts have been made, all aimed at isolating a DNA that could be amplified by PCR and tar-geted by molecular markers.Positive results have been claimed by using commercial kits,in particular QIamp DNA stool kit(QIagen),Gene Elute Plant Kit(Sigma),Nucleospin Food and Nucleospin Plant(Macherey Nagel).A series of problems,however,have been reported concerning DNA degradation,low yield and presence of impurities and,hence,reproducibility of the results in subsequent assays(Agrimonti et al.,2011).In preliminary experiments,we employed some of the reported procedures(Busconi et al.,2003; Consolandi et al.,2008;Costa,Mafra,Amaral,&Oliveira,2010; De la Torre,Bautista,Cánovas,&Claros,2004;Doyle&Doyle, 1987;Giménez,Pistón,Martín,&Atienza,2010)and one of the commercially available kits(Qiagen QIAmp DNA stool kit)as it is considered to give the most reproducible results by several authors (Ben-Ayed et al.,2009;Muzzalupo et al.,2007;Rabiei,Enferandi, Saidi,Patui,&Vannozzi,2010;Spaniolas,Bazakos,Spano,Zoghby, &Kalaitzis,2010;Spaniolas et al.,2008;Testolin&Lain,2005)to extract the DNA from our samples(data not shown).However, we experienced the same problems reported in the literature;spe-cifically,we did not obtain DNA that could be detected either by spectrophotometer and/or agarose gel electrophoresis analysis. Alternatively,the recovered DNA was degraded and not appropri-ate for further molecular analyses;even in those cases in which the DNA was not detectable under UV light upon staining with ethidium bromide we performed PCRs but we did not obtain reli-able amplicons.For all these reasons,we focused our attention on setting up a novel extraction method with a better yield starting from the published CTAB–hexane–chloroform protocol(Giménez et al.,2010).The DNA recovered with our new protocol is of high molecular weight,only slightly fragmented and suitable for subsequent SSR analysis.Possible explanations for this result are the purification steps employed and the excision of the DNA major band from the agarose gel.This latter step is a definitive advantage over exist-ing methods as it removes most,if not all,the residues of organic components(polysaccharides and polyphenols)or additional con-taminants that could inhibit Taq polymerase and hence PCR ampli-fications or interfere with subsequent molecular analyses.The methodology is extremely reproducible as we obtain DNA from each olive oil extraction as documented by the detection of DNA by spectrophotometric and agarose gel determinations.An inter-esting feature of this procedure is that it can be successfully employed with commercially available olive oils,eitherfiltered and notfiltered,both mono and multivarietal,independently of the container type used(i.e.can or glass or PET bottles).The amount of recovered DNA in all cases was in the same concentra-tion range obtained with our samples.Consistent with these results,the DNA was successfully amplified in all subsequent PCR reactions.This is a relevant achievement and an important step forward with respect to the OLIV-TRACK project,one conclusion of which is that at least three replicate extractions should be per-formed to obtain at least one positive sample in PCR amplification (Agrimonti et al.,2011;Marmiroli,Maestri,Pafundo,&Vietina, 2009).According to the literature(Costa,Mafra,&Oliveira,2012; Pafundo et al.,2010),olive oil has to be processed as fresh as pos-sible(until a maximum period of one month)in order to avoid DNA damage processes.We were able to get non-degraded DNA from EVOO even one year from milling(Fig.1A).Moreover,as the DNA extracted is stable for long time when stored at4°C,it can success-fully be used as template in PCR reactions in the following months (Fig.1B).3.2.SSR analysis for cultivar identificationTo test the efficacy of our procedure,we established the genetic profile of eight olive varieties.To this goal,we genotyped forty samples of DNA isolated from olive leaves by using carefully selected SSR primer pairs atfive loci,in the same range(4–6)598K.Raieta et al./Food Chemistry172(2015)596–602reported by others (Delgado-Martinez,Amaya,Sánchez-Sevilla,&Gomez-Jimenez,2012;Pérez-Jiménez,Besnard,Dorado,&Hernandez,2013;Roubos,Moustakas,&Aravanopoulos,2010).A total of twenty alleles were generated.Specifically,the number of alleles per locus ranges from 3(GAPU17and GAPU71A)to 5(UDO99028and GAPU59),with an average of 4alleles per locus.The Ho ranges from 0.625(UDO99028)to 0.875(UDO99039),while the He ranges from 0.598(GAPU71A)to 0.710(UDO99039)(Table 1).Our data indicate that Ho is higher than He and this is due to a high genetic variability amongst the cultivars studied;this is confirmed by PD values that range from 0.594(GAPU71A)to 0.813(GAPU59).The PIC assayed within loci ranges from 0.520(GAPU71A)to 0.645(UDO9939);both PD and PIC indexes exceed the threshold value of 0.5pointing to a high polymorphic degree.No intra-cultivar variability was detected.All these data indicate that the chosen primer pairs are highly polymorphic,conforming the previously results published by Muzzalupo and Lombardo (2006),and thus have the ability to discriminate and,hence,unambiguously identify our cultivars.An alternative approach has recently been reported by Pérez-Jiménez et al.(2013)who used 9plastid SSRs to characterise 17varieties of Spanish olive oils.They used chloroplast DNA as it is more abundant and more resistant than nuclear DNA and their genetic analysis was based on the strength of 6cp SSRs while 3others showed somewhat discordant results.This experimental approach will likely be employed in the future and give reliable results.Here we used a more traditional and consolidated approach and the results we report seem to confirm the strength of nuclear DNA SSR loci in olive oil identification issue.Based on the Nei distance and applying the R Studio,a dendro-gram can be obtained showing four main clusters:(1)Ortolana;(2)Ortice and Ravé;(3)Racioppella;(4)Leccino,Frantoio,Pendolino and Olivastro (Fig.2).Collectively,our data clearly indicate that genotyping the cultivars using SSR loci allows the identification of each variety and definitely distinguish those of the Samnium region from the reference ones.3.3.SSR analysis in olive fruit,embryo and The same SSR markers were then used ity.The molecular profiles obtained on leaf,drupe and embryo were compared amplification.(A)Genomic DNA extraction example from oil (O)and leaf (L)of Ortice cultivar six months purity as assessed by the 260/280and 260/230spectrophotometric ratios.(B)Agarose gel electrophoresis from the olive cultivar Ortice obtained at different times after oil nes 1–2:leaf sample (control);3–4:storage of DNA at 4°C;7–8:oil sample at second extraction;9–10:oil sample at third extraction;11:negative sizes of the marker bands are given in base pairs).The olive oil first DNA extraction was carried out two months,milling.Table 1SSR statistical analysis obtained on eight olive cultivars using five microsatellites markers.For each locus the number of total alleles (k ),the number of analysed samples (N ),Observed Heterozygosity (HObs),Expected Heterozygosity (HExp),Polymorphic Index Content (PIC),Power of Discrimination (PD)and Allelic Size Range were evaluated.Locus k N HObs HExp PIC PD Allelic size range UDO990285240.6250.6540.5840.813124–169UDO990394240.8750.7100.6450.656105–181GAPU173240.8750.5980.5200.594213–224GAPU595240.8750.6940.6410.781204–228GAPU71A3240.8750.5980.5200.594209–223Dendrogram.Unweighted Pair Group Method using the Nei index shows genetic amongst the following olive cultivars:Ravé(RV),Frantoio (FR),Leccinoelectropherograms.(A)The locus UDO099039of the Ortice cultivar was amplified using as template the DNA indicates the181bp allele,while the star the171bp allele.(B)The electropherogram shows the profile obtained pollinator(Olivastro cultivar).Leaf and drupe DNA samples display completely overlapping peaks corresponding some additional alleles that are slightly detectable also in the oil profile.This additional peak was found⁄).The orange peaks correspond to DNA ladder as an internal standard while the remaining peaks from embryos to as the‘‘stutter’’bands(Walsh,Fildes,&Teynolds,1996).(For interpretation of the references to colourarticle.)Analysis(PCoA).PCoA analysis of SSR data shows the clustering of the analysed samples.Each cultivar is indicated Ortice,OV=Olivastro,PE=Pendolino,RC=Racioppella,RV=Ravé.The DNA source is indicated as-Dthe respective cultivars.samples to generate the PCoA by the GenAlex tool.Each variety gives rise to a well-defined cluster,which includes leaf,drupe and oil(Fig.4).While the genetic profiles of drupe and leaf DNA are similar,only slight differences are observed with the oil profile in each cluster.Frantoio,Leccino,Olivastro and Pendolino(putative pollinators and reference cultivars)are all placed in the left side of the graph.All embryos’profiles differ and are closer to the axes ori-gins with respect to the corresponding cluster of leaf,drupe and oil profiles of the Samnium cultivars,likely due to the putative pollin-ators’alleles.In support of these considerations,leaf and drupe DNA samples display completely overlapping peaks corresponding to the same alleles,while embryo DNAs display some additional alleles that are slightly evident also in the oil profiles(Fig.3).3.4.Parentage analysisBecause oil DNA bears alleles that appear to be specifically orig-inated from the embryo,their identification is mandatory to prove the efficacy of the genetic analysis in olive oil traceability.Olive oil DNA may derive from two different sources:the cells of the pulp and those of the embryo in the kernel(seed).The genome of the drupe is in large part represented by pulp specific alleles that,in turn,are identical to those of the olive tree from which it derives. The genome of the embryo contains alleles derived from the mother plant and the pollinator.As indicated above,the genetic profiles of leaf and pulp DNA cluster together and are very similar to the oil one;the seed profile appears to be different and may be influenced by the pollinator.Detection of the pollinator alleles in oil,then, causes a distancing between the reference profile and the resulting oil,especially when the patterns are different from those of the mother plant.Taking into account all these remarks,we carried out a parentage analysis by using the CERVUS3.0algorithm and processing oil samples as offspring(Kalinowski,Taper,& Marshall,2007;Marshall,Slate,Kruuk,&Pemberton,1998).For each analysis,the program simulates datasets and calculates the expected distribution of their statistic tests and the Delta LOD score (D LOD score)that usually indicates the difference in likelihood ratios between the two males most likely to be the father of the off-spring.From the distribution of the statistic tests in simulation,the program establishes a critical value that produces a desired level of confidence(95%)in parentage assignment.Such a simulation of paternity uses an offspring replication of10,000and is based on multilocus genotyping of16trees(4of which are putative pollina-tors),4drupes samples,4embryos and20oils(Meagher,1986).We performed four separate analyses and for each cultivar we consid-ered the DNA from leaf,pulp,embryo,oil,and putative pollina-tor(s).All information acquired through interviews with nurseries’owners about the putative pollinators was confirmed by genetic testing on seeds.In detail,we were able to assign the fol-lowing pollinators(à)to respective cultivars producing the monova-rietal EVOOs:(1)Olivastro(à)and Racioppella(à)for Ortolana,2) Olivastro(à)for Ortice,3)Pendolino(à)for Racioppella,4)Ortol-ana(à)and Ortice(à)for Ravé.Moreover,all cultivars sampled may be subject to self-pollination,as previously indicated by farmers. We calculated probability of non-exclusion-first parent(NE-1P), probability of non-exclusion-second parent(NE-2P),probability of non-exclusion-parent pair(NE-PP)for each cultivar in order to get useful and sufficient information for obtaining a D LOD score.For Ortice NE-1P ranged from0.825to0.858,NE-2P ranged from 0.705to0.762,and NE-PP ranged from0.562to0.582.For Ortolana NE-1P ranged from0.791to0.978,NE-2P ranged from0.638to 0.894,and NE-PP ranged from0.202to0.636.For Racioppella NE-1P ranged from0.804to0.970,NE-2P ranged from0.651to0.876, and NE-PP ranged from0.485to0.793.For RavéNE-1P ranged from 0.838to0.957,NE-2P ranged from0.724to0.849,and NE-PP ran-ged from0.590to0.738.As reported in Table2,the D LOD scores for Ortice,Ortolana,Racioppella and Ravé,considering father alone, were 1.47, 1.40, 1.47and 1.32respectively.Considering father given known mother,the D LOD scores were2.10,1.95,2.07and 1.86,listed in the same order as above.The simulation of the pater-nity analysis test failed in two out of four cases(Ortolana and Ravé) when we evaluated the parentage relationship considering father alone(D CV⁄).By contrast,all values exceeded the self-generated critical threshold(D CV⁄⁄)by almost threefold when we considered the father given known mother.On the basis of these results,we can unambiguously assign the oil to the corresponding leaf profile witha reassuring confidence level.4.ConclusionsWe report here a new procedure for isolating DNA from extra virgin olive oil that is reliable and not expensive.Mono and multi-varietal commercially available olive oils packed in different con-tainers can be processed with positive and reliable results.The recovered DNA is devoid of contaminations that could interfere with subsequent molecular analyses and,importantly,can be stored for months with no degradation.The opportunity of having high molecular weight DNA is the key element for assessing the genetic profile of different olive cultivars and to discriminate them on the basis of objective and reliable parameters.Moreover,the method is highly sensitive as we can appreciate trace levels of exogenous alleles as those from the pollinators.The possibility to provide a market product with a geneticfingerprint is undoubtedly an added value,especially when considering the high level of food counterfeits spread all-over the world.These results are also in line with the last rules issued by the European Union and International Institutions aimed at saving the Protected Designation of Origin (PDO)and other niche products.Conflict of interestAll authors have seen and approved the manuscript being submitted;they furthermore declare no competingfinancial or personal interest.AcknowledgementsSome of the results reported here were obtained thanks to financial support from the P.R.U.S.S.T.‘‘C A L I D O N E’’–Programma di Riqualificazione Urbana e.Sviluppo Sostenibile del Territorio.We thank Mrs.Carmela Ferravante(Azienda Ferravante, Paduli)and Mr.Antonio Parente(Azienda Olivicola Titerno,San Lorenzello)for supplying monocultivar oil samples and plant materials,as well as Azienda Sperimentale Improsta(Eboli,Sal-erno)and Mr.Angelo Lo Conte for kindly providing plant materials. ReferencesAgrimonti,C.,Vietina,M.,Pafundo,S.,&Marmiroli,N.(2011).The use of food genomics to ensure the traceability of olive oil.Trends in Food Science& Technology,22(5),237–244.Table2Parentage analysis carried out on eight olive cultivars.For each cultivar the meandelta LOD score(D LOD m)and critical value of delta LOD score(D CV)were assessedconsidering father alone(⁄)and father given mother(⁄⁄).Cultivars D CV⁄D LOD m⁄D CV⁄⁄D LOD m⁄⁄Ortice0.83 1.470.67 2.10Ortolana 1.70 1.400.70 1.95Racioppellla0.98 1.470.67 2.07Rave’ 1.74 1.320.70 1.86K.Raieta et al./Food Chemistry172(2015)596–602601。