lable at ScienceDirect

Food Microbiology 27 (2010) 973e978

Contents lists avai

Food Microbiology

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

Strain typing of acetic acid bacteria responsible for vinegarproduction by the submerged elaboration method

Rocío Fernández-Pérez 1, Carmen Torres, Susana Sanz, Fernanda Ruiz-Larrea*,1

Department of Food and Agriculture, Faculty of Science, University of La Rioja, Av. Madre de Dios 51, 26006 Logroño, Spain

a r t i c l e i n f o

Article history:Received 20 January 2010Received in revised form18 May 2010Accepted 20 May 2010Available online 1 June 2010

Keywords:Acetic acid bacteriaStrain identificationPFGE (pulsed field gel electrophoresis)ERIC-PCRVinegarSubmerged elaboration method

* Corresponding author. Tel.: þ34 941 299749; fax:E-mail address: fernanda.ruiz@unirioja.es (F. Ruiz-

1 Presentaddress: ICVV(UR,CSIC,CAR),Av.Madrede

0740-0020/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.fm.2010.05.020

a b s t r a c t

Strain typing of 103 acetic acid bacteria isolates from vinegars elaborated by the submerged method fromciders, wines and spirit ethanol, was carried on in this study. Two different molecular methods wereutilised: pulsed field gel electrophoresis (PFGE) of total DNA digests with a number of restrictionenzymes, and enterobacterial repetitive intergenic consensus (ERIC) e PCR analysis. The comparativestudy of both methods showed that restriction fragment PFGE of SpeI digests of total DNA was a suitablemethod for strain typing and for determining which strains were present in vinegar fermentations.Results showed that strains of the species Gluconacetobacter europaeus were the most frequent leaderstrains of fermentations by the submerged method in the studied vinegars, and among them strain R1was the predominant one. Results showed as well that mixed populations (at least two different strains)occurred in vinegars from cider and wine, whereas unique strains were found in spirit vinegars, whichoffered the most stressing conditions for bacterial growth.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Avariety of vinegars is produced inmostMediterranean countriesand extensively used as a condiment and as an efficient acidifyingagent for food preservation. Vinegar is produced by twowell-definedmethods: a slow surface process, in which acetic acid bacteria (AAB)are placed on the aireliquid interface in direct contact with atmo-spheric oxygen, and a fast submerged process, in which AAB aresubmerged in the acetifying liquid and a continuous strong aerationis applied to provide the necessary oxygen for acetic fermentation totake place. Generally, the surface process is employed for elaboratingtraditional vinegars and the submerged process is employed for theelaboration of most commercial vinegars of major consumption.

AAB are the microorganisms responsible for the transformationof ethanol into acetic acid, and it should be pointed out that notall the strains of a certain species have the same ability to oxidizeethanol into acetic acid (Gullo and Giudici, 2008). Therefore,discriminating among AAB strains of the same species is importantto determine how many strains are involved in a fermentationprocess and which one is leading the transformation. DNA-basedtyping methods that have been successfully used for identificationto strain level of AAB from a variety of origins are the following:enterobacterial repetitive intergenic consensus sequence

þ34 941 299721.Larrea).Dios51, 26006Logroño, Spain.

All rights reserved.

amplification (ERIC-PCR) (González et al., 2004, 2005; Gullo andGiudici, 2008; Gullo et al., 2009; Nanda et al., 2001), randomamplified polymorphic DNA (RAPD-PCR) analysis (Nanda et al.,2001; Prieto et al., 2007; Trcek et al., 1997), and pulsed field gelelectrophoresis (PFGE) of genomic restriction fragments applied toawide range of bacterial isolates (López et al., 2007). Microbiologicalstudies reported in the last years have focused on characterisation ofAAB from traditional balsamic vinegars (De Vero et al., 2006; Gulloet al., 2006, 2009; Gullo and Giudici, 2008; Ilabaca et al., 2008),rice vinegar (Haruta et al., 2006) and some other vegetable productssuch as grapes or cocoa beans (De Vuyst et al., 2008; Prieto et al.,2007), and very few reports can be found on strain characteriza-tion of AAB from vinegars produced by the submerged method(Callejón et al., 2008; Schüller et al., 2000; Trcek et al., 2000).

The objectives of the present study were searching for appro-priate and efficient DNA-based molecular methods to type AABstrains, and to characterize up to strain level AAB isolates fromvinegars elaborated by the submerged method from cider, wineand spirit ethanol.

2. Materials and methods

2.1. Vinegar sampling

Vinegar sampleswere aseptically taken from30,000 l bioreactors(Frings Xuzhou Bio- and Chemical Technology Co., Ltd.) optimized

R. Fernández-Pérez et al. / Food Microbiology 27 (2010) 973e978974

for the submerged production of vinegar, of the company Vina-grerías Riojanas S. A., containing either wine, cider or spirit vinegarsin full fermentation (fermentation rate ¼ 0.17e0.25 acetic degrees/h). A total of 58 samples (34 wine vinegars, 20 cider vinegars and 4spirit vinegars) were collected during the period from November2007 to December 2008. Samples of 25 ml were collected in 50 mlsterile tubes and transported with continuous agitation, which fav-oured aerobic conditions. Samples were rapidly submitted (within10 min) to microbiological analysis in the laboratory.

2.2. Culture media and growth conditions

25 ml samples of wine and cider vinegars were centrifuged at2100�g (Megafuge Heraeus, Thermo Scientific, Wilmington, USA)for 10min. Cell pellets of approximately 1ml volumewere collected,and 100 ml of each sample were cultivated for 5 days on GY agarplates [5% glucose (Panreac Química S.A., Barcelona, Spain), 1% yeastextract (Scharlau Chemie S A, Barcelona, Spain) and 1.5% agar (Bec-toneDickinson, Madrid, Spain)] at 30 �C under aerobic conditions.25 ml samples of spirit ethanol vinegars were centrifuged and cellpellets were cultivated in GY broth for 48 h at 30 �Cwith continuousand vigorous agitation in order to adapt AAB cells to growing in theculture medium. These samples were subsequently cultivated on GYagar plates following the same procedure as described for wineand cider vinegars. Colonies were submitted to gram staining andmorphological analysis by optical microscopy. Three colonies wererandomly taken from each vinegar sample. Isolates thus selectedwere considered to represent the numerically dominant strainspresent in the vinegar samples. AAB isolates were sub-cultured topurity on GYagar plates andwere typed to species level by sequenceanalysis of the amplicon obtained by PCR of the 16Se23S intergenicregion in a previous study (Fernández-Pérez et al., submitted forpublication). A total of 103 pure isolates were recovered and storedin 20% sterile skim milk (BectoneDickinson) at �20 �C.

2.3. Strain typing by ERIC (enterobacterial repetitiveintergenic consensus) -PCR

A total of 90 AAB pure isolates were grown onto GYagar plates at30 �C for 3 days under aerobic conditions for DNA extraction. TheDNA extraction method that was carried out was as follows. Cellsfrom this fresh culture were suspended in 200 ml of lysis buffer(50 mM Tris/HCl, pH 8; 10 mM b-mercaptoethanol). Samples werevigorously vortexed and reposed for 15 min at room temperature.They were incubated at 100 �C for 10 min, vortexed and frozen to�80 �C. Samples were unfrozen and submitted to precipitation(Sambrook et al., 1989). 200 ml of solution II (0.2 N NaOH; 1% SDS)was added over 100 ml of sample, mixed manually and kept on icefor 3 min. 150 ml of solution III (3 M potassium; 5 M acetate) wasadded to the mix and agitated for 10 s and kept on ice for 3e5 min.The sample was centrifuged at 14,800�g (Biofuge Heraeus, ThermoScientific) for 5 min at 4 �C. The supernatant was recovered ina sterile microtube and DNA was precipitated with 2 vol of ethanol(96%) at 4 �C and resting on ice for 2 min. The sample was centri-fuged at 12,000�g at 4 �C for 5 min and the supernatant waseliminated. The DNA was dissolved in 50 ml of TE buffer (10 mMTris/Cl pH 8; 1 mM EDTA pH 8) and DNAses were inactivated byheating in a water bath at 85 �C for 15 min. DNA quantification wasperformed with the apparatus NanoDrop (Thermo Scientific) andadjusted between 80 and 150 ng/ml. The oligonucleotide primersused for the amplification of the ERIC-PCR were those described byVersalovic et al. (1991): ERIC1 (50ATCGAAGCTCCTGGGGATTCAC30)and ERIC2 (50AAGTAAGTGACTGGGGTGAGCG30) (synthesized bySigma Aldrich, Madrid, Spain). PCR amplification was carried out ina final volume of 50 ml following the conditions described by

González et al. (2004). The amplification products were resolved byelectrophoresis in 1.5% (w/v) agarose gels, separated at 80 V for 1 hand 45 min, stained with ethidium bromide and photographed.ERIC-PCR patterns were classified as indistinguishable, closelyrelated or unrelated when they were identical, differed in 1e3bands, or differed in more than 3 bands respectively.

2.4. Strain typing by restriction fragment PFGEof total DNA digested with SpeI

AAB pure isolates were cultured for 24 h in GY broth withvigorous and continuous shaking to reach an optical density of0.8e1.2 at 660 nm. Three ml samples of these fresh cultures werecentrifuged and cells were washed once with 3 ml sterile salinesolution. Cells were resuspended in 100 ml of storage buffer (10 mMTris/HCl pH 8, 10 mM EDTA pH 8). The suspension was warmed at50 �C and 100 ml of 1% pulse field certified agarose (D-5 PronadisaHispanlab S.A., Madrid, Spain) in TBE buffer (45 mM Tris, 45 mMboric acid, 1 mM EDTA, pH 8) at the same temperature was added.The suspension was allowed to solidify in molds and they weretreated following the method described by López et al. (2007) forcell lysis and DNA isolation under immobilised conditions. Beforerestriction enzyme digestion, agarose blocks were cut (slices1e2 mm) and balanced for 30 min at room temperature in 100 ml ofthe appropriate restriction enzyme buffer. The following restrictionendonucleases were tested: SfiI, XbaI, NotI, AluI, SmaI and SpeI.Digestions with SpeI (New England Biolabs, Beverly, MA) were themost effective and DNAs were incubated with this enzyme forpattern comparison. SpeI digestions were performed overnight at37 �C in a 100 ml total volume of the specific buffer with 5 U ofrestriction enzyme. Before loading, gel blocks were washed with1 ml of TBE for 8 min at 52 �C. DNA fragments were separated in 1%(wt/vol) agarose (D-5 Pronadisa) in TBE buffer with a CHEF DR IIsystem (Bio-Rad Laboratories, CA, USA). A total of 77 pure isolateswere analysed by this PFGEmethod. Electrophoresis was performedat 14 �C at a constant voltage of 4.5 V/cmwith a switch time rampedfrom 5 to 45 s over 24 h period. Gels were stained with ethidiumbromide (0.5 mg/ml) and photographed under UV light. Lambdaladder PFG marker (New England Biolabs) was used as molecularsize standard. PFGE patterns were classified, like in the case of ERIC-PCR analyses, as indistinguishable, closely related or unrelatedwhentheywere identical, differed in 1e3 bands, or differed inmore than 3bands respectively, according to published criteria for bacterialstrain typing (Tenover et al., 1995).

2.5. Reproducibility study

To determine the percentage of similarity necessary for straindiscrimination, reproducibility studies were carried out accordingto López et al. (2008). The level of similarity obtained betweenrepeats of the same isolate when included within the dendrogramfor all strains, established the discriminatory threshold belowwhich patterns were considered to be different.

2.6. Numerical analysis of gel images

The GelCompar 2.5 software (Applied Maths, Kortrijd, Belgium)was used for conversion, normalization, and further processing ofimages. Comparison of the obtained restriction fragment PFGE andERIC-PCR patterns was performed with Dice coefficient and theUnweighted Pair GroupMethod using Arithmetic averages (UPGMA).The cophenetic correlation value was calculated for the dendro-grams. This parameter is a measure of the reliability of the calculateddistances in the dendrogram (Sokal and Rohlf, 1962).

R. Fernández-Pérez et al. / Food Microbiology 27 (2010) 973e978 975

The index of strain diversity was evaluated by calculating thefollowing percentage: 100 � (number of different patterns/totalnumber of isolates)

3. Results

From the total 58 vinegar samples in full fermentation that werecollected in this study, a total of 103 AAB isolates were recoveredand submitted to genetic typing up to strain level. All the isolateswere shown to be gram negative and ellipsoidal to rod shaped.

3.1. AAB strain typing by ERIC-PCR

Different protocols were assayed for obtaining total DNAfrom AAB isolates, and results indicated that the method for DNAextraction was a determinant step for the subsequent ERIC-PCRanalysis. Results showed that the method described in Section 2.3,which included two steps for DNA extraction, rendered consistentERIC-PCR patterns. The time required for the whole analysis of anAAB isolate by this method was one day, in contrast with four daysrequired for the restriction fragment PFGE analysis. Ninety of the 103AAB isolates were analysed by ERIC-PCR and Table 1 and Fig. 1 showthe patterns obtained for isolates recovered from the different typesof vinegars, as well as their corresponding species. Table 1 alsoshows that two closely related (CR) patterns (E3a CR to E3b, and E5aCR to E5b) were obtained for a number of isolates recovered fromwhite wine vinegars. The reproducibility study was carried out by

Table 1ERIC-PCR patterns obtained from a total of 90 analysed AAB vinegar isolates.

Patternname

Closelyrelatedpatterns

No. ofisolates

Species Origin of vinegar

E1 7 Ga. europaeus White wineE3a 13 Ga. europaeus White wine

E3b 2 Ga. europaeus White wineE5a 2 Ga. europaeus White wine

E5b 3 Ga. europaeus Red wineE15 4 Ga. europaeus Red wineE18 3 Ga. europaeus White wineE19 1 Ga. europaeus White wineE27 1 Ga. europaeus White wineE28 1 A. pasteurianus CiderE30 1 A. pasteurianus CiderE32 5 Ga. europaeus White wineE36 2 Ga. xylinus CiderE38 1 A. hansenii CiderE39 2 Ga. europaeus CiderE45 1 Ga. europaeus White wineE46 2 Ga. xylinus CiderE48 1 Ga. europaeus CiderE49 2 Ga. xylinus CiderE51 1 Ga. europaeus White wineE52 2 Ga. europaeus White wineE60 2 Ga. europaeus White wineE61 5 Ga. europaeus White wine

and spiritE62 4 Ga. europaeus White wineE65 1 Ga. europaeus White wineE81 2 Ga. europaeus White wineE90 2 A. pasteurianus CiderE109 9 Ga. xylinus CiderE110 1 Ga. xylinus CiderE111 2 Ga. xylinus CiderE119 3 Ga. xylinus CiderE125 1 Ga. xylinus CiderE156 1 Ga. europaeus Red wine31 patterns 2 closely

relatedpatterns

90 isolates

duplicate with 10 isolates, and the resulting discriminatorythreshold for this method was 81%. Cluster analysis using thisthreshold value and visual inspection of the ERIC-PCR imagesrendered the dendrogram shown in Fig. 1, which includes a total of31 unrelated patterns for the 90 AAB isolates of this study. Asdeduced from the dendrogram, the percentage of similarity betweenunrelated profiles varied from 31% to 81%. Patterns were classifiedinto two groups with similarity percentages >30% (represented inFig. 1 with a discontinuous line) within each group. The resultingindexof diversity for this typingmethodwas 37% and the copheneticcorrelation value calculated for the dendrogram was �60%. Isolatesof cider vinegarswere not clustered together in the same groupwiththis typing method, which differed from the results obtained withthe restriction fragment PFGE typingmethod shown in Fig. 2. Table 1shows the unrelated (n ¼ 31) patterns with the number of isolatesthat presented each pattern, their species identification and the typeof vinegar from which isolates were recovered in this series of 90AAB isolates submitted to ERIC-PCR typing method.

3.2. AAB strain typing by restriction fragmentPFGE of SpeI digests of total DNA

Seventy seven out of the 103 AAB isolates were typed byrestriction fragment PFGE. Among the six tested endonucleases (SfiI,XbaI,NotI, AluI, SmaI and SpeI) SpeI rendered themost discriminatingrestriction fragments. AAB total DNA digested with SpeI yieldedapproximately 14 bands in the 25e600 kb size range, suitable fornumeration, whereas the other endonucleases yielded either toofew bands (NotI, SmaI, AluI) or too many fragments (XbaI) (data notshown). Fig. 2 shows SpeI restriction patterns of the isolatessubmitted to this analysis. The reproducibility of restriction frag-ment PFGE analysis was determined analysing by duplicate themacrorestriction patterns of 15 AAB isolates, and a value of 85% forDice’s coefficient of similarity was obtained, which was a highervalue than that obtained for ERIC-PCR analysis. This value wasconsidered as the limit percentage of similarity below which themacrorestriction patterns were considered different and corre-sponding to distinct strains or genotypes, and above which thegrouped isolates were considered identical, multiple copies of thesame strain. Cluster analysis using this criteria and visual inspectionof the restriction fragment PFGE images rendered the dendrogramshown in Fig. 2, which shows 22 distinct macrorestriction patternsamong the 77 AAB isolates of this study. As deduced from thedendrogram, the percentage of similarity between unrelated andclosely related profiles varied from 45% to 85% (cophenetic correla-tion value calculated for the dendrogram �60%). Table 2 shows theunrelated (n ¼ 17) and closely related (n ¼ 5) patterns with thenumber of isolates that presented each pattern, the species identi-fication and the type of vinegar fromwhich isolates were recoveredin this series of 77AAB isolates. AABpatternswere classified into twogroups which are shown in Fig. 2, with coefficients of similarity�45% (this value is indicated in Fig. 2 with a discontinuous line)within each group, Group 1 with coefficient of similarities�45% andGroup 2 with �49% of similarity among their correspondingAAB isolates. All the isolates clustered in Group 1were of the speciesGluconacetobacter europaeus and it included all the isolates fromspirit vinegars as well as isolates fromwine vinegars. All the isolatesfrom cider vinegar were clustered together in Group 2; this grouppresented the highest variability of species: Acetobacter pasteur-ianus, Ga. europaeus, Ga. hansenii and Ga. xylinus. The average indexof strain diversity calculated from this PFGE typing method,expressed as the ratio of number of different patterns versus totalnumber of analysed isolates, was 28%.

As shown in Table 2, themost abundant strainwas Ga. europaeusof PFGE pattern R1, which was identified in 25 isolates of different

Fig. 1. ERIC-PCR distinct patterns and UPGMA dendrogram obtained from 90 AAB isolates.

R. Fernández-Pérez et al. / Food Microbiology 27 (2010) 973e978976

types of wine vinegars (red and white wine vinegars) and spiritvinegar (frequency of appearance: 32% among the total 77 analysedisolates). This strain was found in vinegar samples during a periodof seven months. The second most frequently found strain was Ga.europaeus of PFGE pattern R3, which appeared in 12 isolatesobtained fromwhite wine vinegars during a period of five months.

Spirit vinegars showed pure cultures, and single strains of thespecies Ga. europaeus were present in these vinegars. It is impor-tant to note that isolates from spirit vinegars were difficult tocultivate in the laboratory and required an adaptation periodof growing in GY broth before plating out and isolation. Strains ofPFGE pattern R1 and R75 that appeared in spirit vinegar sampleswere also present in wine vinegar samples, which indicates thesame origin and an adaptation of these two strains to growing inthe highly stressing growth conditions that spirit vinegars posses(14% ethanol). Most wine vinegar samples (73%) showed mixedcultures of at least two different strains, and strain R1 appeared asthe leader strain in seven wine vinegar samples and the uniquestrain in one spirit vinegar, followed by strain R3which appeared asthe leader strain in four wine vinegars.

In contrast to spirit vinegars, all cider vinegars showed mixedcultures of different strains (at least two different strains in eachvinegar sample) either of the same species or of different species asdetermined by sequencing of the 16Se23S intergenic region. Cidervinegars showed as well a variety of leader strains, which were notrepeated along the time span of this study andwhich appeared onlyin one sample. The index of strain diversity in cider vinegars was75%, whereas the diversity index found in wine vinegars was 17%,which indicates that cider vinegars showed the highest diversityamong their strains.

4. Discussion

Very few studies on AAB strain typing from vinegars have beenreported to date, and genetic differentiation of strains withinspecies has always been a challenge (Bartowsky and Henschke,2008). A number of authors used ERIC-PCR as the typing methodfor AAB, and thus this method has been used for strain typing of A.pasteurianus isolates from traditional balsamic vinegars (Gulloet al., 2009) and AAB isolates from rice vinegars (Nanda et al.,2001). Some other authors used RAPD-PCR analysis for straintyping, and thus Bartowsky et al. (2003) typed A. pasteurianusisolates from spoiled bottled red wine, later they reported thistechnique for typing isolates from different sources, includingvinegar, rice vinegar, and spoiled cider (Bartowsky and Henschke,2008), and more recently this random amplified PCR techniquewas used to type AAB isolates from Chilean vineyards (Prieto et al.,2007). Some authors have used ERIC-PCR analysis in combinationwith other techniques, and thus ERIC and REP-PCR were applied toinvestigate species and strain evolution of the AAB populationduring wine production (González et al., 2005). REP-PCR has beenalso applied to identify at strain level AAB isolates from Ghanaianfermented cocoa beans using (GTC)5-REP-PCR fingerprinting(De Vuyst et al., 2008).

PFGE of genomic restriction fragments of AAB was reported asa method for strain typing (Sievers and Swings, 2005) and it hadbeen reported as an appropriate tool for strain typing of a widenumber of bacterial isolates (López et al., 2008). Our results showedthat PFGE of total DNA restriction fragments with the enzyme SpeIgave the most discriminating analysis when compared with otherrestriction enzymes. Sievers and Swings (2005) based their analysis

Fig. 2. SpeI PFGE distinct patterns and UPGMA dendrogram obtained from 77 AAB isolates.

Table 2PFGE patterns obtained from SpeI digests of a total of 77 analysed AAB vinegarisolates.

Patternname

Closelyrelatedpatterns

No. ofisolates

Species Origin of vinegar

R1 25 Ga. europaeus White and redwine and spirit

R3 12 Ga. europaeus White wineR6 1 Ga. europaeus White wineR28 1 A. pasteurianus CiderR30 1 A. pasteurianus CiderR33a 1 Ga. europaeus White wine

R33b 3 Ga. europaeus White wineR33c 1 Ga. europaeus White wineR33d 5 Ga. europaeus White wine and spirit

R35 1 Ga. xylinus CiderR36 2 Ga. xylinus CiderR39 2 Ga. europaeus CiderR41 1 Ga. hansenii CiderR46 2 Ga. xylinus CiderR48 4 Ga. europaeus Cider and white wineR49 2 Ga. xylinus Cider and white wineR54a 2 Ga. europaeus White wine

R54b 6 Ga. europaeus White wineR55 2 Ga. europaeus White wineR70 1 Ga. europaeus White wineR90a 1 A. pasteurianus Cider

R90b 1 A. pasteurianus White wine17 patterns 5 closely

relatedpatterns

77isolates

R. Fernández-Pérez et al. / Food Microbiology 27 (2010) 973e978 977

on the restriction fragments obtained with the enzyme XbaI,nevertheless, our results showed that the fragments obtained withSpeI were larger that those obtained with XbaI and gave theappropriate number of PFGE bands (around 14 bands) for straintyping, and that the reproducibility of the analysis was high (85%).Many reports agree with our results in that PFGE typing by usingthe appropriate restriction endonucleases offers the best discrim-inatory power to differentiate strains of the same species (Lópezet al., 2008; Cleenwerck and De Vos, 2008).

Our results showed that the DNA extraction method used forERIC-PCR analysis was a rapid and efficientmethod, suitable for thistype of analysis. As mentioned above, other authors reported ERIC-PCR analysis as an appropriate method for strain typing of AAB(Nanda et al., 2001; González et al., 2005; Gullo et al., 2009).Nevertheless, our results showed that the clusters generated by theERIC-PCR patterns did not correlate either with the isolate originor the isolate species, whereas both clusters generated from therestriction PFGE patterns correlated with the origin of isolates(Group 1 included isolates fromwine and spirit vinegars, belongingexclusively to the species Ga. europaeus, and Group 2 includedisolates from cider and wine belonging to a variety of species).These results suggest that PFGE restriction fragment analysis is themethod of choice to type AAB strains, nevertheless, this method ismore time-consuming than ERIC-PCR, which could be an appro-priate method for following the fermentation process of a biore-actor that has been inoculated with a specific AAB strain.

The index of diversity shown by the PFGEmethod (28%) revealedthe diversity among the AAB isolates of this study. It should bepointed out that bioreactors (17,000 l of total reaction volume) wereinoculated with 12,000 l of white wine vinegar in full fermentationprocess from an operational tank, therefore, the diversity amongisolates was due to the different type and origin of the wines and

ciders used for vinegar elaboration. Strains R1 and R75 wereobtained from wine vinegars (red and white wine) and from spiritvinegars. This result indicated that both strains were well adapted(specially R1) to survive under the different conditions of the

R. Fernández-Pérez et al. / Food Microbiology 27 (2010) 973e978978

acetifying liquid and could constitute excellent starter cultures forthese vinegars. Cider vinegars showed the highest variability ofspecies (A. pasteurianus, Ga. xylinus, Ga. hansenii and Ga. europaeus)among its isolates, species identification that was determined bysequencing the 16Se23S intergenic region, and they showed as wellthe highest diversity of clones and mixed populations of strains(n � 2 strains) growing simultaneously in all cider vinegar samples,as mentioned in Results section. This was most probably due to thehigh sugar content (4%) (Del Campo et al., 2008) and the low contentin alcohol (6%) of ciders, whereas wine vinegars (12% ethanol) andspirit vinegars (14% ethanol) offer more stressing conditions forbacterial growth.

Summarizing, this study demonstrates that strains of thespecies Ga. europaeus are the most frequent leader strains ofvinegar fermentations by the submerged method in a variety ofvinegars (wine, cider and spirit vinegars) and among them strainR1 was predominant. It shows as well that mixed populations(at least two different strains) occur in vinegars from cider andwine, whereas unique strains grow in spirit vinegars, which offerthe most stressing conditions for bacterial growth. Efficient straintyping tools are essential for identification of particular strains andfor preparing defined starter cultures, and our study shows thatPFGE of total DNA restriction fragments with the enzyme SpeI isa reliable and discriminating method for strain typing.

Acknowledgments

This workwas supported by grant OTEM070621 by the companyVinagrerías Riojanas S.A. and grant AGL2007-60504 of the Ministryof Research and Science of Spain and FEDER of the EuropeanCommunity. Rocío Fernández-Pérez was supported by a researchfellowship associated to OTEM070621.

References

Bartowsky, E.J., Henschke, P.A., 2008. Acetic acid bacteria spoilage of bottled redwine; a review. Int. J. Food Microbiol. 125, 60e70.

Bartowsky, E.J., Xia, D., Gibson, R.L., Fleet, G.H., Henschke, P.A., 2003. Spoilage ofbottled red wine by acetic acid bacteria. Lett. Appl. Microbiol. 36, 307e314.

Callejón, R.M., Tesfaye, W., Torija, M.J., Mas, A., Troncoso, A.M., Morales, M.L., 2008.HPLC determination of amino acids with AQC derivatization in vinegars alongsubmerged and surface acetifications and its relation to the microbiota. Eur.Food Res. Technol. 227, 93e102.

Cleenwerck, I., De Vos, P., 2008. Polyphasic taxonomy of acetic acid bacteria: anoverview of the currently applied methodology. Int. J. Food Microbiol. 125, 2e14.

De Vero, L., Gala, E., Gullo, M., Solieri, L., Landi, S., Giudici, P., 2006. Application ofdenaturing gradient gel electrophoresis (DGGE) analysis to evaluate acetic acidbacteria in traditional balsamic vinegar. Food Microbiol. 23, 809e813.

De Vuyst, L., Camu, N., De Vinter, T., Vandemeulebroecke, K., Van de Perre, V.,Vacanneyt, M., De Vos, P., Cleenwerck, I., 2008. Validation of the (GTG)(5)-REP-PCR fingerprinting technique for rapid classification and identification of aceticacid bacteria, with a focus on isolates from Ghanaian fermented cocoa beans.Int. J. Food Microbiol. 125, 79e90.

Del Campo, G., Berregi, I., Santos, J.I., Dueñas, M., Irastorza, A., 2008. Development ofalcoholic and malolactic fermentations in highly acidic and phenolic applemust. Bioresour. Technol., 2854e2863.

Fernández-Pérez, R., Torres, C., Sanz, S., Ruiz-Larrea, F. Rapid molecular methods forenumeration and taxonomical identification of acetic acid bacteria responsiblefor submerged vinegar production, submitted for publication.

González, A., Hierro, N., Poblet, M., Mas, A., Guillamón, J.M., 2005. Application ofmolecular methods to demonstrate species and strain evolution of acetic acidbacteria population during wine production. Int. J. FoodMicrobiol. 102, 295e304.

González, A., Hierro, N., Poblet, M., Rozes, N., Mas, A., Guillamón, J.M., 2004.Application of molecular methods for the differentiation of acetic acid bacteriain a red wine fermentation. J. Appl. Microbiol. 96, 853e860.

Gullo, M., Caggia, C., De Vero, L., Giudici, P., 2006. Characterisation of acetic acidbacteria in “traditional balsamic vinegar”. Int. J. Food Microbiol. 106, 209e212.

Gullo, M., De Vero, L., Giudici, P., 2009. Succession of selected strains of Acetobacterpasteurianus and other acetic acid bacteria in traditional balsamic vinegar. Appl.Environ. Microbiol. 75, 2585e2589.

Gullo, M., Giudici, P., 2008. Acetic acid bacteria in traditional balsamic vinegar: pheno-typic traits relevant for starter cultures selection. Int. J. Food Microbiol. 125, 46e53.

Haruta, S., Ueno, S., Egawa, I., Hashiguchi, K., Fujii, A., Nagano, M., Ishii, M.,Igarashi, Y., 2006. Succession of bacterial and fungal communities duringa traditional pot fermentation of rice vinegar assessed by PCR-mediateddenaturing gradient gel electrophoresis. Int. J. Food Microbiol. 109, 79e89.

Ilabaca, C., Navarrete, P., Mardones, P., Romero, J., Mas, A., 2008. Application ofculture-independent molecular biology based methods to evaluate aceticacid bacteria diversity during vinegar processing. Int. J. Food Microbiol. 126,245e249.

López, I., Tenorio, C., Zarazaga, M., Dizy, M., Torres, C., Ruiz-Larrea, F., 2007. Evidenceof mixed wild populations of Oenococcus oeni strains during wine spontaneousmalolactica fermentations. Eur. Food Res. Technol. 226, 215e223.

López, I., Torres, C., Ruiz-Larrea, F., 2008. Genetic typification by pulsed field gelelectrophoresis (PFGE) and randomly amplified polymorphic DNA (RAPD) ofwild Lactobacillus plantarum and Oenococcus oeni wine strains. Eur. Food Res.Technol. 227, 547e555.

Nanda, K., Taniguchi, M., Ujike, S., Ishihara, N., Mori, H., Ono, H., Murooka, Y., 2001.Characterization of acetic acid bacteria in traditional acetic acid fermentation ofrice vinegar (komesu) and unpolished rice vinegar (kurosu) produced in Japan.Appl. Environ. Microbiol. 67, 986e990.

Prieto, C., Jara, C., Mas, A., Romero, J., 2007. Application of molecular methods foranalysing the distribution and diversity of acetic acid bacteria in Chileanvineyards. Int. J. Food Microbiol. 115, 348e355.

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. In: Molecular Cloning. A LaboratoryManual, vol. 1. CSH Press, pp. 1.26e1.28.

Schüller, G., Hertel, C., Hammes, W.P., 2000. Gluconacetobacter entanii sp. nov.,isolated from submerged high-acid industrial vinegar fermentations. Int. J. Syst.Evol. Microbiol. 50, 2013e2020.

Sievers, M., Swings, J., 2005. Family Acetobacteraceae. In: Garrity, G.M. (Ed.)Bergey’sManual of Systematic Bacteriology, second ed., vol. 2. Springer, New York, pp.41e95 (Part C).

Sokal, R.R., Rohlf, F.J., 1962. The comparison of dendrograms by objective methods.Taxon 11, 33e40.

Tenover, F.C., Arbeit, R.D., Goering, R.V., Mickelsen, P.A., Murray, B.E., Persing, D.H.,Swaminathan, B., 1995. Interpreting chromosomal DNA restriction patternsproduced by pulsed-field gel electrophoresis: criteria for bacterial strain typing.J. Clin. Microbiol. 33, 2233e2239.

Trcek, J., Ramus, J., Raspor, P., 1997. Phenotypic characterization and RAPD-PCRprofiling of Acetobacter sp. isolated from spirit vinegar production. Food Tech-nol. Biotechnol. 35, 63e67.

Trcek, J., Raspor, P., Teuber,M., 2000.Molecular identification of Acetobacter isolates fromsubmerged vinegar production, sequence analysis of plasmidpJK2-1 andapplicationin the development of a cloning vector. Appl. Microbiol. Biotechnol. 53, 289e295.

Versalovic, J., Koeuth, T., Lupski, J.R., 1991. Distribution of repetitive DNA sequencesin eubacteria and application to fingerprinting of bacterial genomes. NucleicAcids Res. 19, 6823e6831.