Available online at www.sciencedirect.

com

International Journal of Food Microbiology 125 (2008) 79 90

www.elsevier.com/locate/ijfoodmicro

Validation of the (GTG)5-rep-PCR fingerprinting technique for rapid

classification and identification of acetic acid bacteria, with a
focus on isolates from Ghanaian fermented cocoa beans
Luc De Vuyst a,, Nicholas Camu a , Tom De Winter a , Katrien Vandemeulebroecke b ,
Vincent Van de Perre a , Marc Vancanneyt b , Paul De Vos b , Ilse Cleenwerck b
a
Research Group of Industrial Microbiology and Food Biotechnology, Department of Applied Biological Sciences and Engineering, Vrije Universiteit Brussel,
Pleinlaan 2, B-1050 Brussels, Belgium
b
BCCM/LMG Bacteria Collection, Ghent University, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium

Abstract

Amplification of repetitive bacterial DNA elements through the polymerase chain reaction (rep-PCR fingerprinting) using the (GTG)5 primer,
referred to as (GTG)5-PCR fingerprinting, was found a promising genotypic tool for rapid and reliable speciation of acetic acid bacteria (AAB).
The method was evaluated with 64 AAB reference strains, including 31 type strains, and 132 isolates from Ghanaian, fermented cocoa beans, and
was validated with DNA:DNA hybridization data. Most reference strains, except for example all Acetobacter indonesiensis strains and Gluco-
nacetobacter liquefaciens LMG 1509, grouped according to their species designation, indicating the usefulness of this technique for identification
to the species level. Moreover, exclusive patterns were obtained for most strains, suggesting that the technique can also be used for
characterization below species level or typing of AAB strains. The (GTG)5-PCR fingerprinting allowed us to differentiate four major clusters
among the fermented cocoa bean isolates, namely A. pasteurianus (cluster I, 100 isolates), A. syzygii- or A. lovaniensis-like (cluster II, 23
isolates), and A. tropicalis-like (clusters III and IV containing 4 and 5 isolates, respectively). A. syzygii-like and A. tropicalis-like strains from
cocoa bean fermentations were reported for the first time. Validation of the method and indications for reclassifications of AAB species and
existence of new Acetobacter species were obtained through 16S rRNA sequencing analyses and DNA:DNA hybridizations. Reclassifications
refer to A. aceti LMG 1531, Ga. xylinus LMG 1518, and Ga. xylinus subsp. sucrofermentans LMG 18788T.
2007 Elsevier B.V. All rights reserved.

Keywords: Acetic acid bacteria; Cocoa; Rep-PCR; (GTG)5 primer

1. Introduction acetic acid in neutral and acidic (pH 4.5) media. These
characteristics make that they are involved in the production
Acetic acid bacteria (AAB) are Gram-negative, ellipsoidal to of fermented foods, either in a beneficial (chocolate products,
rod-shaped, obligate aerobic bacteria that oxidize alcohols or coffee, vinegar, nata de coco, and specialty beers) or detrimental
sugars incompletely, leading to the accumulation of organic (spoilage of beers, wines, and ciders) way, and in the production
acids as end-products (De Ley et al., 1984; Swings et al., 1992; of commercially important fine chemicals (Kersters et al., 2006).
Sievers and Swings, 2005). They are widespread in nature and Today, little is known about AAB associated with cocoa bean
most of them are capable of oxidizing ethanol as substrate to fermentation, the first step in cocoa and chocolate production
(Ostovar and Keeney, 1973; Carr et al., 1979; Passos and
Passos, 1985; Thompson et al., 1997; Ardhana and Fleet, 2003;
Schwan and Wheals, 2004.). During this spontaneous fermen-
Corresponding author. Mailing address: Research Group of Industrial
tation, AAB oxidize ethanol, initially produced by yeasts, to
Microbiology and Food Biotechnology (IMDO), Vrije Universiteit Brussel
(VUB), Pleinlaan 2, B-1050 Brussels, Belgium. Tel.: +32 2 6293245; fax: +32 2 acetic acid. This volatile acid diffuses into the beans and this, in
6292720. combination with heat produced by this exothermic bioconver-
E-mail address: [email protected] (L. De Vuyst). sion, causes the death of the seed embryo as well as the end
0168-1605/$ - see front matter 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijfoodmicro.2007.02.030
80 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

of fermentation. In turn biochemical changes in the beans are primer, referred to as (GTG)5-PCR fingerprinting, has been
initiated, leading to the formation of precursor molecules that found a promising genotypic tool for rapid and reliable
are necessary for the development of a characteristic aroma, speciation and typing of lactobacilli (Gevers et al., 2001) and
flavor, and color of the beans (Hansen et al., 1998). These enterococci (vec et al., 2005). In this study, (GTG)5-PCR
properties are further developed during drying, roasting, and fingerprinting was evaluated for AAB, using reference strains of
final processing of well-fermented cocoa beans (Thompson AAB as well as AAB isolates from Ghanaian, fermented cocoa
et al., 2001). The activity of AAB is thus essential for the beans. In addition, validation of the method and indications for
production of high-quality cocoa. Strains of Acetobacter and reclassifications of AAB species from culture collections and
Gluconobacter are mostly found (Schwan and Wheals, 2004). existence of new Acetobacter species from fermented cocoa
The taxonomy of AAB has undergone many changes in beans were obtained through 16S rRNA gene sequencing
recent years. Several genera and species of AAB have been analyses and DNA:DNA hybridizations.
newly described. AAB are currently classified into ten genera
and 44 species, namely Acetobacter (16 species), Glucono- 2. Materials and methods
bacter (5 species), Acidomonas (1 species), Gluconacetobacter
(15 species), Asaia (3 species), Kozakia (1 species), Sacchari- 2.1. Strains and growth conditions
bacter (1 species), Swaminathania (1 species), Neosaia
(1 species), and Granulibacter (1 species), in the family Two sets of AAB were included in this study. The first group
Acetobacteraceae as a branch of the acidophilic bacteria in the consisted of AAB reference strains (64 in total, including 31
-subdivision of the Proteobacteria (Dellaglio et al., 2005; type strains) and is listed in Table 1. These AAB were obtained,
Dutta and Gachhui, 2006; Greenberg et al., 2006; Jojima et al., together with the type strain of Lactobacillus plantarum LMG
2004; Lisdiyanti et al., 2006; Loganathan and Nair, 2004; 6907T, from the BCCM/LMG Bacteria Collection (Ghent
Sievers and Swings, 2005; Silva et al., 2006; Tanasupawat et al., University, Gent, Belgium). All reference strains were grown
2004; Yukphan et al., 2004b,c, 2005). according to the provider's specifications (http://www.belspo.
As AAB are involved in the production of food, their be/bccm/), unless indicated otherwise. The second group
identification is important information for technologists, who consisted of 132 AAB isolates from Ghanaian, fermented
try to control the bioprocesses involved. The identification of cocoa beans. They were maintained frozen at 80 C in MYP
AAB to species level has traditionally been performed by medium (2.5% [wt/vol] D-mannitol, 0.5% [wt/vol] yeast extract,
studying physiological and chemotaxonomic properties such as and 0.3% [wt/vol] bacteriological peptone [Oxoid, Basingstoke,
growth at low pH and growth in the presence of 0.35% (vol/vol) UK]), supplemented with 25% (vol/vol) glycerol as cryopro-
acetic acid, production of acetic acid and gluconic acid from tectant, and recovered by incubation at 30 C in MYP medium
ethanol and glucose, respectively, the presence of unique under aerobic conditions for 14 days.
ubiquinones and cellular fatty acids, and protein electrophoretic
patterns (De Ley et al., 1984; Swings et al., 1992). However, the 2.2. Phenotypic tests
identification methods based on phenotypic characteristics of
AAB are not reliable and very time-consuming. Therefore, they All isolates were tested for their Gram reaction, cell shape,
have been complemented or replaced by different molecular cell size, and mobility, from cells grown at 30 C in MYP
techniques, in particular DNA:DNA hybridizations (Cleen- medium under aerobic conditions for 14 days. Catalase activity
werck et al., 2002; Dellaglio et al., 2005; Lisdiyanti et al., 2000, was detected by the appearance of oxygen gas bubbles from 20%
2001) and PCR-based genomic fingerprinting techniques such (vol/vol) hydrogen peroxide solution by colonies on MYP
as Restriction Fragment Length Polymorphism (RFLP) analysis agar (MYP medium supplemented with 1.5% agar, wt/vol).
of PCR-amplified 16S rRNA (Gonzlez et al., 2004, 2006a,b; L. plantarum LMG 6907 T (catalase-negative lactic acid
Gullo et al., 2005; Poblet et al., 2000; Ruiz et al., 2000) or 16S bacterium) and Acetobacter aceti LMG 1504T (catalase-positive
23S rRNA intergenic spacer regions (Gonzlez et al., 2006a; acetic acid bacterium) were used as controls.
Ruiz et al., 2000; Trek, 2005; Trek and Teuber, 2002; The production of acetic acid from ethanol and gluconic acid
Yukphan et al., 2004a), Randomly Amplified Polymorphic from glucose was tested following growth of the strains at 30 C
DNA (RAPD) fingerprinting (Bartowsky et al., 2003; Nanda for 4 days in basal medium [0.05% yeast extract and 0.3%
et al., 2001; Trek et al., 1997; Trek and Raspor, 1999), and vitamin-free casamino acids (Difco, Detroit, MI); wt/vol] plus
PCR amplification of repetitive bacterial DNA elements (rep- ethanol (0.3%, wt/vol) and GY medium (5% D-glucose and
PCR) using the REP (Repetitive Extragenic Palindromic 0.5% yeast extract, wt/vol), respectively. The production of 2-
sequences) or ERIC (Enterobacterial Repetitive Intergenic and 5-keto-gluconic acid from ethanol and glucose was tested
Consensus sequences) primers (Gonzlez et al., 2004; Nanda following growth in basal medium plus ethanol and GY
et al., 2001). Many of these methods were tested on a limited medium, respectively, at 30 C for 4 days. Residual substrates
number of strains and failed to discriminate between closely and metabolites were quantified by high pressure liquid
related AAB species or were discriminating below species level. chromatography (HPLC) with refractive index detection for
The aim of this study was to obtain a method for fast and glucose and ethanol, and high pressure anion exchange
possibly high-throughput classification and identification of a chromatography (HPAEC) with ion suppression and conduc-
broad range of AAB. Rep-PCR fingerprinting using the (GTG)5 tivity detection for acetic acid, gluconic acid, 2-keto-gluconic
 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990 81

acid, and 5-keto-gluconic acid, basically following the proto- temperature was 46 1 C (Acetobacter) or 49 1 C (Gluco-
cols of Van der Meulen et al. (2006). nacetobacter). Reciprocal reactions (e.g. A B and B A) were
All tests were done in duplicate. performed and their variation was within the limits of this
method (Goris et al., 1998). The DNA binding values reported
2.3. DNA preparation are the means of a minimum of four hybridization experiments,
including the reciprocal reactions.
Total genomic DNA was extracted from the reference strains
by the method of Wilson (1987), with minor modifications 2.6. 16S rRNA gene sequencing
(Cleenwerck et al., 2002). For the isolates it was extracted either
as described by Gevers et al. (2001), except that mutanolysin A fragment of the 16S rRNA gene (corresponding with the
was substituted by proteinase K (VWR International, Darm- positions 81541 in the Escherichia coli numbering system)
stadt, Germany) in an amount of 2.5 mg per ml of TrisHCl was amplified by PCR using the conserved primers pA (5 AGA
(10 mM)EDTA (1 mM) buffer (pH 8.0), or by the method of GTT TGA TCC TGG CTC AG 3) and pH (5 AAG GAG GTG
Wilson (1987), with minor modifications (Cleenwerck et al., ATC CAG CCG CA 3). PCR-amplified 16S rRNA genes
2002). All isolates were grown in MYP medium for 14 days, were purified using the NucleoFast 96 PCR Cleanup Kit
except for some strains that grew better in GY medium. Asaia (Macherey-Nagel, Dren, Germany). Sequencing reactions
siamensis and Asaia bogorensis were grown in M17 medium were performed using the BigDye Terminator Cycle Sequenc-
(5% glucose, 3% calcium carbonate, 1% yeast extract, wt/vol). ing Kit (Applied Biosystems, Foster City, CA) and purified
using the Montage SEQ96 Sequencing Reaction Cleanup Kit
2.4. Rep-PCR genomic fingerprinting (Millipore, Bedford, MA). Sequencing was performed using an
ABI Prism 3100 Genetic Analyzer (Applied Biosystems).
A set of 64 reference strains and 132 isolates was subjected Nearly complete sequences were determined using the eight
to rep-PCR fingerprinting with the oligonucleotide primer pair sequencing primers listed in Coenye et al. (1999), except for
REP1R-I (5-IIIICGICGICATCIGGC-3) and REP2-I (5- R-29337 where the primer BKL1 (position according to the
IIICGNCGNCATCNGGC-3) and with the single oligonucle- E. coli 16S rRNA gene sequence numbering: 536516;
otide primer (GTG)5 (5-GTGGTGGTGGTGGTG-3) (Versa- sequence: 5 GTA TTA CCG CGG CTG CTG GCA 3) was
lovic et al., 1994; Gevers et al., 2001). The reproducibility of used instead of the primer PD. Sequence assembly was performed
(GTG)5-PCR was tested by amplifying DNA from twelve using the program AutoAssembler (Applied Biosystems).
randomly chosen strains several times. In addition, each PCR Sequence similarities between the consensus sequences and
reaction was controlled for reproducibility by inclusion of the small ribosomal subunit sequences collected from the interna-
type strain of L. plantarum LMG 6907T. Optimal PCR tional nucleotide sequence library EMBL (EMBL-EBI, Hinxton,
conditions for each of the primer sets used were as described Cambridge, UK) were calculated pairwise using an open gap
by Versalovic et al. (1994). PCR amplifications were performed penalty of 100% and a unit gap penalty of 0% with the software
with a DNA T3 thermal cycler (Biometra, Westburg, The package BioNumerics version 4.5 (Applied Maths).
Netherlands) using Taq DNA polymerase (Roche Diagnostics
GmbH, Mannheim, Germany). The PCR products were 3. Results
electrophoresed in a 1.5% (wt/vol) agarose gel (15 20 cm)
for 16 h at a constant voltage of 55 V in 1 TAE buffer (40 mM 3.1. Phenotypic characterization of isolates from Ghanaian,
TrisAcetate, 1 mM EDTA, pH 8.0) at 4 C. The rep-PCR fermented cocoa beans
profiles were visualized under UV light after staining of the gel
with ethidium bromide, and digital image capturing was done Cells of all isolates were motile or non-motile rods, Gram-
using the Gel Doc EQ system (Biorad, Hercules, CA). The negative, and catalase-positive. All isolates grew in basal
resulting fingerprints were analyzed using the BioNumerics medium plus ethanol and in GY medium. They all oxidized
version 4.0 software package (Applied Maths, Sint-Martens- ethanol to acetic acid and produced gluconic acid from glucose
Latem, Belgium). The similarity among digitized profiles was in the respective media. Most of the isolates produced 2-keto-
calculated using the Pearson correlation, and an average linkage gluconic acid from glucose in very small amounts and did not
(UPGMA or unweighted pair group method with arithmetic produce 5-keto-gluconic acid.
averages) dendrogram was derived from the profiles.
3.2. Rep-PCR genomic fingerprinting: method validation
2.5. DNA:DNA hybridizations
The (GTG)5 primer as well as the REP1R-I and REP2-I
Only high-molecular-mass DNA with A260/A280 and A234/ primer set generated DNA fragments of 300 to 4000 bp for
A260 absorption (A) ratios of 1.82.0 and 0.400.60, respec- AAB reference strains and isolates (Fig. 1A). The (GTG)5
tively, was used for DNA:DNA hybridizations. DNA:DNA primer resulted in banding patterns containing generally
hybridizations were performed according to a modification between 10 and 30 visualized PCR products, while the
(Cleenwerck et al., 2002; Goris et al., 1998) of the microplate REP1R-I and REP2-I primer set generated banding patterns
method described by Ezaki et al. (1989). The hybridization containing generally less than 10 bands (Fig. 1A and B). The
 82
L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990
Table 1
Acetic acid bacteria reference strains and DNADNA relatedness of Acetobacter peroxydans LMG 1635T and LMG 1633 and of Gluconacetobacter strains determined in the frame of this study
LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG LMG
1633 1635T 7602 7603T 7971 8065 8067 18890T 1524 1527T 1528 1348 1381T 1509 1728 1510 1515T 1518 18788T
Acetobacter aceti LMG 1496
A. aceti LMG 1504T
A. aceti LMG 1531
A. aceti LMG 1535
A. cerevisiae LMG 1599
A. cerevisiae LMG 1625T
A. cerevisiae LMG 1682
A. cerevisiae LMG 1699
A. cibinongensis LMG 21418T
A. estunensis LMG 1572
A. estunensis LMG 1580
A. estunensis LMG 1626T
A. indonesiensis LMG 1571
A. indonesiensis LMG 1588
A. indonesiensis LMG 19824T
A. lovaniensis LMG 1617T
A. malorum LMG 1746T
A. oeni LMG 21952T
A. orientalis LMG 21417T
A. orleanensis LMG 1583T
A. pasteurianus LMG 1262T
A. pasteurianus LMG 1555
A. pasteurianus LMG 1629
A. pasteurianus LMG 1630
A. pasteurianus LMG 1658
A. pasteurianus LMG 1659
A. pasteurianus LMG 1686
A. peroxydans LMG 1633 100
A. peroxydans LMG 1635T 63 100
A. pomorum LMG 18848T
A. syzygii LMG 21419T
A. tropicalis LMG 1663
A. tropicalis LMG 1754
A. tropicalis LMG 19825T
A. tropicalis LMG 19826
Gluconacetobacter azotocaptans LMG 21311T
Ga. diazotrophicus LMG 7602 100
Ga. diazotrophicus LMG 7603T 96 100
Ga. diazotrophicus LMG 7971 93 74 100
Ga. diazotrophicus LMG 8065 95 75 97 100
Ga. diazotrophicus LMG 8067 86 94 89 92 100
Ga. europaeus LMG 18494
Ga. europaeus LMG 18890T 100
Ga. hansenii LMG 1524 100

L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

Ga. hansenii LMG 1527T 97 100
Ga. hansenii LMG 1528 100 100 100
Ga. intermedius LMG 18809T
Ga. johannae LMG 21312T
Ga. liquefaciens LMG 1348 100
Ga. liquefaciens LMG 1381T 100 100
Ga. liquefaciens LMG 1509 91 82 100
Ga. liquefaciens LMG 1728 69 68 69 100
Ga. oboediens LMG 18849T
Ga. rhaeticus LMG 22126T
Ga. sacchari LMG 19747T
Ga. swingsii LMG 22125T
Ga. xylinus subsp. xylinus LMG 1510 100
Ga. xylinus subsp. xylinus LMG 1515T 50 77 100
Ga. xylinus subsp. xylinus LMG 1518 73 61 42 100
Ga. xylinus subsp. sucrofermentans LMG 18788T 25 57 56 20 100
Acidomonas methanolica LMG 1668T
Asaia bogorensis LMG 21650T
Asaia siamensis LMG 21651T
Gluconobacter oxydans LMG 1408T

83
84 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

Fig. 1. Rep-PCR banding patterns of AAB reference strains and isolates. (A) Picture of a rep-PCR gel for the (GTG)5 primer (left part) and the REP primer pair (right
part) on DNA of AAB isolates; lane assignments 1 to 16 correspond with isolates 103, 109, 110, 117, 119, 121, 122, 123, 126, 127, 128, 129, 131, 133, 138, and 139,
respectively, while the lanes marked with M are molecular size markers (mixture of a 100-bp and 5000-bp ladder; Biorad). (B) Digitized (GTG)5-PCR (left part) and
REP-PCR (right part) fingerprints of AAB reference strains.

(GTG)5 primer was preferred for further evaluation for rapid times, ranged between 89 and 96% (results not shown). The
classification, identification, and typing of AAB. inclusion of the type strain of L. plantarum LMG 6907T in each
The similarity between the (GTG)5-PCR patterns, obtained PCR reaction and banding pattern maintained a high reproduc-
through amplification of DNA from twelve strains several ibility (results not shown).
 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990 85

3.3. Classification and identification of AAB with (GTG)5-PCR A. peroxydans (LMG 1633 and LMG 1635T) and A. indonesienis
fingerprinting (LMG 1571, LMG 1588, and LMG 19824T) strains that were
dispersed, and A. aceti LMG 1531, Ga. liquefaciens LMG 1509
3.3.1. Reference strains and LMG 1728, Ga. xylinus LMG 1518 and Ga. xylinus subsp.
The results of the numerical analysis of the generated (GTG)5- sucrofermentans LMG 18788T, which did not cluster with the
PCR banding patterns of a total of 64 reference AAB strains are other strains from their respective taxa.
shown in Fig. 2. All reference strains grouped in separate clusters To evaluate the (GTG)5-PCR technique for species identi-
according to their respective taxonomic designations, except for fication, it is obligatory to know the exact species identity of the

Fig. 2. (GTG)5-PCR banding patterns of a total of 64 reference AAB strains, representing all validly described Acetobacter species (35 strains), 12 Gluconacetobacter
species (25 strains) and other AAB, namely Asaia bogorensis LMG 21650T, Asaia siamensis LMG 21651T, Acidomonas methanolica LMG 1668T, and Glucono-
bacter oxydans LMG 1408T. The dendrogram was generated after cluster analysis of the digitized fingerprints and was derived from UPGMA linkage of Pearson
correlation coefficients. LMG: BCCM/LMG Bacteria Collection (Laboratory of Microbiology, Ghent University, Gent, Belgium, http://www.belspo.be/bccm); T: type
strain.
86 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

Fig. 3. (GTG)5-PCR banding patterns of the AAB reference strains and 132 fermented cocoa bean isolates. The dendrogram, derived from UPGMA linkage of Pearson
correlation coefficients, shows four different clusters of isolates, represented as clusters I (100 isolates), II (23 isolates), III (4 isolates), and IV (5 isolates).
 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990 87

reference AAB used. All Acetobacter (35 strains) and some ness at species level amongst each other and with the type strain
Gluconacetobacter (9 strains) reference strains were previously of A. pasteurianus of 6880% and below species level with the
subjected to DNA:DNA hybridizations, for all using the method type strain of A. pomorum of 4247%. These results prove that
of Ezaki et al. (Cleenwerck et al., 2002; Dellaglio et al., 2005; cluster I consisted of A. pasteurianus strains.
Ezaki et al., 1989; Silva et al., 2006). In the frame of this study To verify if clusters II, III, and IV represented possible new
additional hybridizations were performed between A. perox- AAB species, representative strains of each of them were
ydans LMG 1635T and LMG 1633 and on the remaining Glu- subjected to 16S rRNA gene sequence analysis and the
conacetobacter strains (Table 1). Comparing the results phylogenetic closest neighbors were determined. The 16S
obtained by (GTG)5-PCR fingerprinting (Fig. 2) with the rRNA gene sequence similarities obtained showed that 384 and
DNA:DNA hybridization results (Table 1), in most cases strains 444B (cluster II) were most closely related to A. syzygii (99.7%)
that did not cluster with their respective taxa showed less or and A. lovaniensis (99.5%), while 108B and 420A (cluster III)
approximately 70% DNA homology: this is the generally and 426 and 434 (cluster IV) were most closely related to
accepted limit for species delineation (Stackebrandt et al., A. tropicalis (99.9%). The EMBL accession numbers for the
2002). For example, Ga. liquefaciens LMG 1728, A. perox- 16S rRNA gene sequences of 108B, 384, 420A, 426, 434, and
ydans LMG 1633, Ga. xylinus LMG 1518, and A. aceti LMG 444B are DQ887339, DQ887338, DQ887340, DQ887342,
1531 showed a DNA homology of 6869%, 63%, 4261%, and DQ887341, and DQ887337, respectively.
5359%, respectively, with other strains from Ga. liquefaciens, DNA:DNA hybridizations between 444B and A. syzygii
A. peroxydans, Ga. xylinus, and A. aceti, respectively. The LMG 21419T and A. lovaniensis LMG 1617T revealed a DNA
DNA relatedness between A. peroxydans LMG 1635T and homology value of 46% and 47%, respectively. DNA:DNA
LMG 1633 determined in this study was slightly lower than the hybridizations between 108B, 420A, and A. tropicalis LMG
one reported previously (Cleenwerck et al., 2002). For Ga. 19825T revealed a high DNA homology value between both
xylinus LMG 1518, the (GTG)5-PCR fingerprint was visually isolates (75%) and intermediate values between these isolates
very similar to these of Ga. europaeus strains LMG 18890T and and A. tropicalis LMG 19825T (5458%).
LMG 18894. Hybridizations showed a DNA relatedness of 77%
with Ga. europaeus LMG 18890T. Gluconacetobacter xylinus 4. Discussion
subsp. sucrofermentans LMG 18788T showed a DNA homol-
ogy of 56% with Ga. xylinus subsp. xylinus LMG 1515T. For a Highly conserved repetitive DNA elements, such as Repetitive
few strains, such as for all A. indonesiensis strains (sharing Extragenic Palindromic (REP) elements, Enterobacterial Repet-
N 80% DNA homology with each other) and Ga. liquefaciens itive Intragenic Consensus (ERIC) elements and BOX elements,
LMG 1509, (GTG)5-PCR fingerprints generated none or not seem to be widely distributed in the genomes of various bacterial
enough group-specific bands, in comparison with the profiles of groups. The amplification of the sequences between each of these
the other AAB strains. repetitive elements has been used to generate DNA fingerprints of
several Gram-negative and Gram-positive bacterial species
3.3.2. Fermented cocoa bean isolates (Beyer et al., 1998; De Urraza et al., 2000; Gonzlez et al.,
To evaluate the applicability of (GTG)5-PCR for identifica- 2004; Guinebretier et al., 2001; Lanoot et al., 2004; Pooler et al.,
tion of unknown isolates, the 132 fermented cocoa bean isolates 1996; Sander et al., 1998; Wieser and Busse, 2000). Also, (GTG)5
were subjected to (GTG)5-PCR fingerprinting (Fig. 3). The sequences seem to be widely distributed in the genomes of various
(GTG)5-PCR banding patterns of the isolates were clustered bacterial groups, as the amplification of the sequences between
with those of the AAB reference strains and were found to be these repetitive elements has generated typical DNA fingerprints
dispersed over four major clusters, with patterns sharing more of several lactobacilli (Gevers et al., 2001) and enterococci (vec
than 36% similarity and visually quite similar (Fig. 3). The et al., 2005), allowing their rapid and reliable speciation and
biggest group of isolates (cluster I, containing 100 isolates) typing. We tested the usefulness of this technique with reference
clustered with A. pasteurianus reference strains and was strains of most of the species of AAB and found that it generated
therefore assigned to that species. The 32 remaining isolates complex banding patterns in comparison with the REP primer
were dispersed over three clusters, with 23 (cluster II), 4 (cluster pair. The reproducibility of these (GTG)5-PCR fingerprints from
III), and 5 (cluster IV) isolates, respectively. These clusters did AAB was similar to that obtained by Gevers et al. (2001) with
not contain any of the reference strains and therefore these (GTG) 5 -PCR fingerprints from lactobacilli. Moreover,
strains could not be identified to species level. most reference strains grouped according to their species
Isolates 150, 165D, and 406 of cluster I were subjected to designation and exclusive patterns were obtained for most
16S rRNA gene sequencing analysis and DNA:DNA hybridiza- strains, indicating the usefulness of this technique for identifica-
tions to test the validity of their identification as A. pasteurianus. tion to the species level and characterization below species level
They showed 99.599.8% 16S rRNA gene sequence similarity or typing of AAB strains.
with the type strains of A. pomorum and A. pasteurianus, and To evaluate the (GTG)5-PCR technique for species identi-
less than 98.1% with the type strains of other Acetobacter fication, (GTG)5-PCR fingerprinting data were compared with
species (the EMBL accession numbers for the 16S rRNA gene DNA:DNA hybridization data. The latter data are recognized as
sequences of isolates 150, 165D, and 406 are DQ887334, the data needed for species delineation (Stackebrandt et al.,
DQ887336, and DQ887335, respectively) and a DNA related- 2002) and for accurate identification of Acetobacter and
88 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

Gluconacetobacter strains (Cleenwerck et al., 2002; Dellaglio et al., 2001). A. syzygii and A. tropicalis have never been
et al., 2005; Lisdiyanti et al., 2000). Perfect matches occurred associated with cocoa bean fermentation before. Moreover, A.
between these data and the (GTG)5-PCR clusters, the clustering syzygii mainly occurs in flowers and fruits and is seldom
being in line with the 70% DNA relatedness limit, which is isolated from fermented foods, in contrast with A. tropicalis that
generally accepted for species delineation (Stackebrandt et al., is found in fruits and fermented foods (Lisdiyanti et al., 2001,
2002). The results presented in this paper thus show that 2003). All these new species were not discovered at the time of
(GTG)5-PCR DNA fingerprinting is useful for identification the microbiological analyses of fermented cocoa beans
and classification of AAB to the species level. performed before (Ardhana and Fleet, 2003; Carr et al., 1979;
The additional hybridizations performed in the present paper Ostovar and Keeney, 1973; Passos and Passos, 1985; Schwan
underline some taxonomic problems. For A. aceti LMG 1531 and Wheals, 2004; Thompson et al., 1997). Interestingly, the
the current classification in the species A. aceti can be results of the present paper indicate that clusters II, III, and IV
discussed. Strain LMG 1531 has been assigned to A. aceti represent three novel Acetobacter species. Further studies will
subsp. aceti and cannot be differentiated from other A. aceti be carried out to describe these new species.
strains phenotypically (Gossel et al., 1983). For this reason To summarize, as far as we know this is the first study in which
strain LMG 1531 has never been removed from A. aceti, despite a fast molecular method for AAB was validated by DNA:DNA
its low DNA relatedness with other A. aceti strains. Also, for hybridization data. The (GTG)5-PCR fingerprinting technique
Ga. xylinus LMG 1518 several arguments exist that the strain presented in this paper offers significant advantages over
has been misclassified. Based on its (GTG)5-PCR fingerprint identification methods based on phenotypic characteristics of
and its DNA relatedness to Ga. europaeus (77%), strain LMG AAB, but also over many currently used molecular techniques,
1518 should be reclassified as Ga. europaeus. For Ga. xylinus hence enabling its implementation as high-throughput method-
subsp. sucrofermentans LMG 18788T the DNA relatedness ology. Manual (GTG)5-PCR fingerprinting allows identification
(56%) to Ga. xylinus subsp. xylinus is comparable to the 58.2% of AAB in one working day and can be used for classification and
reported by Toyosaki et al. (1995), but the (GTG)5-PCR identification of AAB at the species level, as well as for
fingerprints of both subspecies are different. Therefore, the characterization of AAB below species level (typing). (GTG)5-
question can be raised whether Ga. xylinus subsp. sucrofer- PCR fingerprinting can quickly increase our knowledge of the
mentans should be elevated to the species level. Also, for A. ecology and biodiversity of AAB and help us to more accurately
peroxydans LMG 1633 the current classification can be determine their growth behavior and beneficial or undesirable role
questioned on the basis of the results presented in this paper. during various stages of food fermentation such as vinegar
For a few strains, such as all A. indonesiensis strains (sharing production, vinification, and cocoa bean fermentation.
N 80% DNA homology with each other) and Ga. liquefaciens
LMG 1509, (GTG)5-PCR fingerprinting does not seem to be Acknowledgements
adequate for species identification. It has been shown before
that rep-PCR may yield whether or not species-specific banding This research was funded by the Research Council of the
patterns when analyzing different species of a genus (Pooler Vrije Universiteit Brussel (OZR, GOA, and IOF projects), the
et al., 1996; Wieser and Busse, 2000). In many cases, the Fund for Scientific Research-Flanders, the Institute for the
(GTG)5-PCR fingerprints generated for AAB were strain- Promotion of Innovation through Science and Technology in
specific and (GTG)5-PCR fingerprinting can therefore be used Flanders (IWT Project 040043), the Federal Research Policy
for typing of AAB as well. Alternatively, rep-PCR with the (Action for the promotion of and co-operation with the Belgian
REP1R-I and REP2-I primer pair can also be used for typing. Coordinated Collections of Micro-organisms, BCCM), and
The (GTG)5-PCR fingerprinting allowed us to differentiate Barry Callebaut N.V.
four major clusters among the 132 isolates from fermented
cocoa beans tested, with decreasing number of isolates
References
encompassing A. pasteurianus (cluster I), A. syzygii- or A.
lovaniensis-like (cluster II), and A. tropicalis-like (clusters III
Ardhana, M., Fleet, G., 2003. The microbial ecology of cocoa bean
and IV). The presence of A. pasteurianus in cocoa bean fermentations in Indonesia. International Journal of Food Microbiology
fermentations has been reported before (Ardhana and Fleet, 86, 8799.
2003; Passos and Passos, 1985). Moreover, previous studies Bartowsky, E.J., Xia, D., Gibson, R.L., Fleet, G.H., Henschke, P.A., 2003.
have shown that members of the genus Acetobacter are found Spoilage of bottled red wine by acetic acid bacteria. Letters in Applied
more frequently than members of the genus Gluconobacter Microbiology 36, 307314.
Beyer, W., Mukendi, F.M., Kimming, P., Bhm, R., 1998. Suitability of
(Schwan and Wheals, 2004), probably reflecting their prefer- reproductive-DNA-sequence based PCR fingerprinting for characterizing
ence for ethanol and sugars, respectively. In general, Aceto- epidemic isolates of Salmonella enterica serovar Saintpaul. Journal of
bacter and Gluconacetobacter strains are mainly isolated from Clinical Microbiology 36, 15491554.
fermented foods, while Gluconobacter strains are mainly Carr, J.G., Davis, P.A., Dougan, J., 1979. Cocoa Fermentation in Ghana and
isolated from fruits and flowers (Lisdiyanti et al., 2003). The Malaysia I. Natural Resources Institute, Chatham, UK.
Cleenwerck, I., Vandemeulebroecke, K., Janssens, D., Swings, J., 2002. Re-
major species found in wine fermentation are A. aceti, A. examination of the genus Acetobacter, with descriptions of Acetobacter
pasteurianus, and G. oxydans (Gonzlez et al., 2005). A. cerevisiae sp. nov. and Acetobacter malorum sp. nov. International Journal
pasteurianus is also involved in rice vinegar production (Nanda of Systematic and Evolutionary Microbiology 52, 15511558.
 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990 89

Coenye, T., Falsen, E., Vancanneyt, M., Hoste, B., Govan, J.R.W., Kersters, K., Kersters, K., Lisdiyanti, P., Komagata, K., Swings, J., 2006. The family Ace-
Vandamme, P., 1999. Classification of Alcaligenes faecalis-like isolates tobacteraceae: the genera Acetobacter, Acidomonas, Asaia, Gluconaceto-
from the environment and human clinical samples as Ralstonia gilardii sp. bacter, Gluconobacter, and Kozakia, In: Dworkin, M., Falkow, S.,
nov. International Journal of Systematic Bacteriology 49, 405413. Rosenberg, E., Schleifer, K.-H., Stackebrandt, E. (Eds.), The Prokaryotes:
De Ley, J., Gillis, M., Swings, J., 1984. Family VI. Acetobacteraceae, In: Krieg, An Evolving Electronic Resource for the Microbiological Community, 3rd
N.R., Holt, J.G. (Eds.), 9th ed. Bergey's Manual of Systematic Bacteriology, ed. Springer-Verlag, New York, NY. http://141.150.157.117:8080/prokPUB/
vol. 1. Williams and Wilkins Co, Baltimore, MD, pp. 267278. index.htm.
Dellaglio, F., Cleenwerck, I., Felis, G.E., Engelbeen, K., Janssens, D., Marzotto, Lanoot, B., Vancanneyt, M., Dawyndt, P., Cnockaert, M., Zhang, J., Huang, Y.,
M., 2005. Description of Gluconacetobacter swingsii sp. nov., and Gluco- Liu, Z., Swings, J., 2004. BOX-PCR fingerprinting as a powerful tool to
nacetobacter rhaeticus sp. nov., isolated from Italian apple fruit. Interna- reveal synonymous names in the genus Streptomyces. Emended descriptions
tional Journal of Systematic and Evolutionary Microbiology 55, 23652370. are proposed for the species Streptomyces cinereorectus, S. fradiae, S.
De Urraza, P.J., Gomez-Zavaglia, A., Lozano, M.E., Romanoswki, V., De tricolor, S. colombiensis, S. filamentosus, S. vinaceus and S. phaeopurpur-
Antoni, G.L., 2000. DNA fingerprinting of thermophilic lactic acid bacteria eus. Systematic and Applied Microbiology 27, 8492.
using repetitive sequence-based polymerase chain reaction. Journal of Dairy Lisdiyanti, P., Kawasaki, H., Seki, T., Yamada, Y., Uchimura, T., Komagata, K.,
Research 67, 381392. 2000. Systematic study of the genus Acetobacter with descriptions of Ace-
Dutta, D., Gachhui, R., 2006. Novel nitrogen-fixing Acetobacter nitrogenifigens tobacter indonesiensis sp. nov., Acetobacter tropicalis sp. nov., Acetobacter
sp. nov., isolated from Kombucha tea. International Journal of Systematic orleanensis (Henneberg 1906) comb. nov., Acetobacter lovaniensis (Frateur
and Evolutionary Microbiology 56, 18991903. 1950) comb. nov., and Acetobacter estunensis (Carr 1958) comb. nov.
Ezaki, T., Yashimoto, T., Yabuuchi, E., 1989. Fluorometric deoxyribonucleic Journal of General and Applied Microbiology 46, 147165.
acid-deoxyribonucleic acid hybridisation in microdilution wells as an Lisdiyanti, P., Kawasaki, H., Seki, T., Yamada, Y., Uchimura, T., Komagata, K.,
alternative to membrane filter hybridization in which radioisotopes are used 2001. Identification of Acetobacter strains isolated from Indonesian sources,
to determine genetic relatedness among bacterial strains. International and proposals of Acetobacter syzygii sp. nov., Acetobacter cibinongensis sp.
Journal of Systematic and Evolutionary Microbiology 39, 224299. nov., and Acetobacter Orientalis sp. nov. Journal of General and Applied
Gevers, D., Huys, G., Swings, J., 2001. Applicability of rep-PCR fingerprinting Microbiology 47, 119131.
for differentiation of Lactobacillus species. FEMS Microbiology Letters Lisdiyanti, P., Katsura, K., Potacharoen, W., Navarro, R.R., Yamada, Y., Uchimura,
205, 3136. T., Komagata, K., 2003. Diversity of acetic acid bacteria in Indonesia,
Gonzlez, ., Hierro, N., Poblet, M., Rozs, N., Mas, A., Guillamn, J.M., Thailand, and the Philippines. Microbiology and Culture Collections 19,
2004. Application of molecular methods for the differentiation of acetic acid 9198.
bacteria in a red wine fermentation. Journal of Applied Microbiology 96, Lisdiyanti, P., Navarro, R.R., Uchimura, T., Komagata, K., 2006. Reclassifi-
853860. cation of Gluconacetobacter hansenii strains and proposals of Gluconace-
Gonzlez, ., Hierro, N., Poblet, M., Mas, A., Guillamn, J.M., 2005. tobacter saccharivorans sp. nov. and Gluconacetobacter nataicola sp. nov.
Application of molecular methods to demonstrate species and strain International Journal of Systematic and Evolutionary Microbiology 56,
evolution of acetic acid bacteria population during wine production. 21012111.
International Journal of Food Microbiology 102, 295304. Loganathan, P., Nair, S., 2004. Swaminathania salitolerans gen. nov., sp. nov., a
Gonzlez, ., Guillamn, J.M., Mas, A., Poblet, M., 2006a. Application of salt-tolerant, nitrogen-fixing and phosphate-solubilizing bacterium from
molecular methods for routine identification of acetic acid bacteria. wild rice (Porteresia coarctata Tateoka). International Journal of Systematic
International Journal of Food Microbiology 108, 141146. and Evolutionary Microbiology 54, 11851190.
Gonzlez, ., Hierro, N., Poblet, M., Mas, A., Guillamn, J.M., 2006b. Nanda, K., Taniguchi, M., Ujike, S., Ishihara, N., Mori, H., Ono, H., Murooka,
Enumeration and detection of acetic acid bacteria by real-time PCR and Y., 2001. Characterization of acetic acid bacteria in traditional acetic acid
nested PCR. FEMS Microbiology Letters 254, 123128. fermentation of rice vinegar (komesu) and unpolished rice vinegar (kurosu)
Goris, J., Suzuki, K., De Vos, P., Nakase, T., Kersters, K., 1998. Evaluation of a produced in Japan. Applied and Environmental Microbiology 67, 986990.
microplate DNADNA hybridization method compared with the initial Ostovar, K., Keeney, P.G., 1973. Isolation and characterization of microorgan-
renaturation method. Canadian Journal of Microbiology 44, 11481153. isms involved in the fermentation of Trinidad's cacao beans. Journal of Food
Gossel, F., Swings, J., Kersters, K., Pauwels, P., De Ley, J., 1983. Numerical Science 38, 611617.
analysis of phenotypic features and protein gel electropherograms of a wide Passos, F.M.L., Passos, F.J.V., 1985. Descrio e classificao de bactrias
variety of Acetobacter strains. Proposal for the improvement of the acticas isoladas da fermentao do cacau, com base em uma analise
taxonomy of the genus Acetobacter Beijerinck 1898, 215. Systematic and numrica. Revista Microbiologica 16, 290298.
Applied Microbiology 4, 338368. Poblet, M., Rozs, N., Guillamn, J.M., Mas, A., 2000. Identification of acetic
Greenberg, D.E., Porcella, S.F., Stock, F., Wong, A., Conville, P.S., Murray, P.R., acid bacteria by restriction fragment length polymorphism analysis of a
Holland, S.M., Zelazny, A.M., 2006. Granulibacter bethesdensis gen. nov., PCR-amplified fragment of the gene coding for 16S rRNA. Letters in
sp. nov., a distinctive pathogenic acetic acid bacterium in the family Aceto- Applied Microbiology 31, 6367.
bacteraceae. International Journal of Systematic and Evolutionary Micro- Pooler, M.R., Ritchie, D.F., Hatung, J.S., 1996. Genetic relationships among
biology 56, 26092616. strains of Xanthomonas fragariae based on random amplified polymorphic
Guinebretier, M.H., Berge, O., Normand, P., Morris, C., Carlin, F., Nguyen-The, DNA PCR, repetitive extragenic palindromic PCR and enterobacterial
C., 2001. Identification of bacteria in pasteurized Zucchini pures stored at repetitive intergenic consensus PCR data and generation of multiplexed PCR
different temperatures and comparison with those found in other pasteurized primers useful for the identification of this phytopathogen. Applied and
vegetable pures. Applied and Environmental Microbiology 67, Environmental Microbiology 62, 31213127.
45204530. Ruiz, A., Poblet, M., Mas, A., Guillamn, J.M., 2000. Identification of acetic
Gullo, M., Caggia, C., De Vero, L., Giudici, P., 2005. Characterization of acetic acid bacteria by RFLP of PCR-amplified 16S rDNA and 16S23S rDNA
acid bacteria in traditional balsamic vinegar. International Journal of Food intergenic spacer. International Journal of Systematic and Evolutionary
Microbiology 106, 209212. Microbiology 50, 19811987.
Hansen, G.E., del Olmo, M., Burns, C., 1998. Enzyme activities in cocoa beans Sander, A., Ruess, M., Bereswill, S., Schuppler, M., Steinbrueckner, B., 1998.
during fermentation. Journal of the Science of Food and Agriculture 77, Comparison of different DNA fingerprinting techniques for molecular
273281. typing of Bartonella henselae isolates. Journal of Clinical Microbiology 36,
Jojima, Y., Mihara, Y., Suzuki, S., Yokozeki, K., Yamanaka, S., Fudou, R., 2004. 29732981.
Saccharibacter floricola gen. nov., sp. nov., a novel osmophilic acetic acid Schwan, R.F., Wheals, A.E., 2004. The microbiology of cocoa fermentation and
bacterium isolated from pollen. International Journal of Systematic and its role in chocolate quality. Critical Reviews in Food Science and Nutrition
Evolutionary Microbiology 54, 22632267. 44, 205221.
90 L. De Vuyst et al. / International Journal of Food Microbiology 125 (2008) 7990

Sievers, M., Swings, J., 2005. Family Acetobacteraceae. In: Garrity, M. (Ed.), Trek, J., Raspor, P., 1999. Molecular characterization of acetic acid bacteria
Bergey's Manual of Systematic Bacteriology, vol. 2. Springer, New York, isolated from spirit vinegar. Food Technology and Biotechnology 37, 113116.
NY, pp. 4195. Trek, J., Teuber, M., 2002. Genetic and restriction analysis of the 16S23S
Silva, L.R., Cleenwerck, I., Rivas, R., Swings, J., Trujillo, M.E., Willems, A., rDNA internal transcribed spacer regions of the acetic acid bacteria. FEMS
Velzquez, E., 2006. Acetobacter oeni sp. nov., isolated from spoiled red Microbiology Letters 208, 6975.
wine. International Journal of Systematic and Evolutionary Microbiology Trek, J., Ramus, J., Raspor, P., 1997. Phenotypic characterization and RAPD-
56, 2124. PCR profiling of Acetobacter sp. isolated from spirit vinegar production.
Stackebrandt, E., Frederiksen, W., Garrity, G.M., Grimont, P.A.D., Kampfer, P., Food Technology and Biotechnology 35, 6367.
Maiden, M.C.J., Nesme, X., Rossello-Mora, R., Swings, J., Truper, H.G., Van der Meulen, R., Makras, L., Verbrugghe, K., Adriany, T., De Vuyst, L.,
Vauterin, L., Ward, A.C., Whitman, W.B., 2002. Report on the ad hoc committee 2006. In vitro kinetic analysis of oligofructose consumption by Bacteroides
for the re-evaluation of the species definition in bacteriology. International and Bifidobacterium spp. indicates different degradation mechanisms.
Journal of Systematic and Evolutionary Microbiology 52, 10431047. Applied and Environmental Microbiology 72, 10061012.
vec, P., Vancanneyt, M., Seman, M., Snauwaert, C., Lefebvre, K., Sedlek, I., Versalovic, J., Schneider, M., de Bruijn, F.J., Lupski, J.R., 1994. Genomic
Swings, J., 2005. Evaluation of (GTG)5-PCR for identification of Entero- fingerprinting of bacteria using repetitive sequence-based polymerase chain
coccus spp. FEMS Microbiology Letters 247, 5963. reaction. Methods in Molecular and Cellular Biology 5, 2540.
Swings, J., Gillis, M., Kersters, K., 1992. Phenotypic identification of acetic acid Wieser, M., Busse, H.J., 2000. Rapid identification of Staphylococcus
bacteria. In: Board, R.G., Jones, D., Skinner, F.A. (Eds.), Identification epidermidis. International Journal of Systematic and Evolutionary Micro-
Methods in Applied and Environmental Microbiology. Technical Series Nr biology 50, 10871093.
29. Society for Applied Bacteriology. Blackwell Scientific Publication, Wilson, K., 1987. Preparation of genomic DNA from bacteria, p. 241.245. In F.M.
London, UK, pp. 103110. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and
Tanasupawat, S., Thawai, C., Yukphan, P., Moonmangmee, D., Itoh, T., Adachi, K. Struhl (ed.), Current Protocols in Molecular Biology. Green Publishing and
O., Yamada, Y., 2004. Gluconobacter thailandicus sp. nov., an acetic acid Wiley-Interscience, New York, NY.
bacterium in the -Proteobacteria. Journal of General and Applied Yukphan, P., Potacharoen, W., Nakagawa, Y., Tanticharoen, M., Yamada, Y.,
Microbiology 50, 159167. 2004a. Identification of strains assigned to the genus Gluconobacter Asai
Thompson, S.S., Miller, K.B., Lopez, A., 1997. Cocoa and coffee. In: Doyle, M.P., 1935 based on the sequence and the restriction analyses of the 16S23S
Beuchat, L.R., Montville, T.J. (Eds.), Food Microbiology: Fundamentals and rDNA internal transcribed spacer regions. Journal of General and Applied
Frontiers. ASM Press, Washington, DC, pp. 649661. Microbiology 50, 915.
Thompson, S.S., Miller, K.B., Lopez, A.S., 2001. Cocoa and coffee, In: Doyle, Yukphan, P., Potacharoen, W., Tanasupawat, S., Tanticharoen, M., Yamada, Y.,
M.P., Beuchat, L.R., Montville, T.J. (Eds.), Food Microbiology: Funda- 2004b. Asaia krungthepensis sp. nov., an acetic acid bacterium in the -
mentals and Frontiers, 2nd ed. ASM Press, Washington, DC, pp. 721733. Proteobacteria. International Journal of Systematic and Evolutionary
Toyosaki, H., Kojima, Y., Tsuchida, T., Hoshino, K.I., Yamada, Y., Yoshinaga, Microbiology 54, 313316.
F., 1995. The characterization of an acetic acid bacterium useful for Yukphan, P., Takahashi, M., Potacharoen, W., Tanasupawat, S., Nakagawa, Y.,
producing bacterial cellulose in agitation cultures: the proposal of Aceto- Tanticharoen, M., Yamada, Y., 2004c. Gluconobacter albidus (ex Kondo
bacter xylinum subsp. sucrofermentans subsp. nov. Journal of General and and Ameyama 1958) sp. nov., nom. rev., an acetic acid bacterium in the -
Applied Microbiology 41, 307314. Proteobacteria. Journal of General and Applied Microbiology 50, 235242.
Trek, J., 2005. Quick identification of acetic acid bacteria based on nucleotide Yukphan, P., Malimas, T., Potacharoen, W., Tanasupawat, S., Tanticharoen, M.,
sequences of the 16S23S rDNA internal transcribed spacer region and of Yamada, Y., 2005. Neoasaia chiangmaiensis gen. nov., sp. nov., a novel
the PQQ-dependent alcohol dehydrogenase gene. Systematic and Applied osmotolerant acetic acid bacterium in the -Proteobacteria. Journal of
Microbiology 28, 735745. General and Applied Microbiology 51, 301311.