International Journal of Food Microbiology 143 (2010) 2631

Contents lists available at ScienceDirect

International Journal of Food Microbiology

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j f o o d m i c r o

A real-time PCR assay for detection and quantication of 2-branched

(1,3)--Dglucan producing lactic acid bacteria in cider
Idoia Ibarburu a, Rosa Aznar b,c, Patricia Elizaquvel b, Nieves Garca-Quintns d, Paloma Lpez d,
Arantza Munduate a, Ana Irastorza a, Mara Teresa Dueas a,
a

Facultad de Ciencias Qumicas, Universidad del Pas Vasco (UPV/EHU), Paseo Manuel de Lardizabal 3, 20018, San Sebastin, Spain
Departamento de Microbiologa y Ecologa, Universitat de Valencia, Burjassot, Valencia, Spain
Instituto de Agroqumica y Tecnologa de Alimentos (IATA), Consejo Superior de Investigaciones Cientcas (CSIC), Burjassot, Valencia, Spain
d
Centro de Investigaciones Biolgicas (CIB), Consejo Superior de Investigaciones Cientcas (CSIC) Ramiro de Maeztu 9, 28040 Madrid, Spain
b
c

a r t i c l e

i n f o

Article history:
Received 10 February 2010
Received in revised form 6 July 2010
Accepted 16 July 2010
Keywords:
(1,3)(1,2)--D-glucan
Lactic acid bacteria
Real-time PCR
Spoilage
Cider

a b s t r a c t
Ropiness in natural cider is a relatively frequent alteration, mainly found after bottling, leading to consumer
rejection. It is derived from the production of exopolysaccharides (EPS) by some lactic acid bacteria most of
which synthesize a 2-branched (1,3)--D-glucan and belong to the genera Pediococcus, Lactobacillus and
Oenococcus. This polysaccharide synthesis is controlled by a single transmembrane glycosyltransferase (GTF).
In this work, a method based on quantitative PCR (qPCR) and targeting the gtf gene was developed for
detection and quantication of these bacteria in cider. The newly designed primers GTF3/GTF4 delimit a
151 bp fragment within the 417 bp amplicon previously designed for conventional PCR. The inclusivity and
exclusivity of the qPCR assay were assessed with 33 cider isolates belonging to genus Lactobacillus,
Oenoccocus and Pedioccocus, together with reference strains of 16 species and ve genera including -glucan,
-glucan and heteropolysaccharide (HePS) producing strains and non-EPS producers. The qPCR assay,
followed by the melting curve analysis, conrmed the generation of a single PCR product from the -glucan
producers with a Tm of 74.28 0.08 and CT values (10 ng DNA) ranging between 8.46 and 16.88 (average
12.67 3.5). Some EPS LAB strains rendered CT values ranging from 28.04 to 37.75 but they were
signicantly higher (P(CT b 28.54) = 0.05) than those of the -glucan producers. The assay showed a wide
quantication range of 5 log units using calibrated cell suspensions of Pediococcus parvulus 2.6 and
Oenococcus oeni I4. The linearity was extended over 7 log orders when calibration curves were obtained from
DNA. The detection limit for -glucan producing LAB in articially contaminated cider was about 3 102 CFU
per ml. The newly developed qPCR assay was successfully applied to monitor the cidermaking process, in 13
tanks from two cider factories, revealing a decrease in CT values derived from an increase in -glucan
producing LAB populations. In addition, 8 naturally spoiled bottled cider were tested for the quantication of
these organisms using the ve standard curves constructed: P. parvulus 2.6 genomic DNA and gtf amplicon
(417 bp), calibrated cell suspensions of Pediococcus parvulus 2.6, Lactobacillus diolivorans G77 and Oenococcus
oeni I4 and results were compared to LAB total counts on MRS. Levels obtained from the different approaches
were within a log range and showed no signicant differences. Therefore, the amplicon-derived standard
curve is proposed for the routine estimation of gtf+ populations in cider.
2010 Elsevier B.V. All rights reserved.

1. Introduction
Lactic acid bacteria (LAB) are able to produce a wide variety of
exopolysaccharides (EPSs) which can be used as a starter to improve the
texture and stability of some dairy products (Ruas-Madiedo et al., 2008).
However, these EPSs have deleterious effects on the organoleptic
properties of alcoholic beverages. In cider (Fernndez et al., 1995) and
Corresponding author. Facultad de Qumicas, Universidad del Pas Vasco (UPV/
EHU), Paseo 18 Manuel de Lardizabal 3, 20018, San Sebastin, Spain. Tel.: +34
943018170; fax: +34 943015270.
E-mail address: [email protected] (M.T. Dueas).
0168-1605/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijfoodmicro.2010.07.023

wine (Llaubres et al., 1990), EPS-producing LAB are responsible for an

alteration, called ropiness or oiliness, characterized by a viscous,
thick texture and oily feel, which although not appreciably altering the
taste, renders the products unpleasant to the palate.
In the Basque Country (North Spain), natural ciders are produced
in small factories according to traditional methods and the usual
oenological procedures (sulfur dioxide treatment, clarication or
correction of acidity) are not applied (Garai et al., 2006). As natural
ciders are not microbiologically stabilized before bottling, ropy bottled
ciders can be encountered and refused by the consumers, resulting in
considerable nancial loss to cidermakers. Therefore, early detection of
these spoiling bacteria, not only in bottled ciders but also during cider

I. Ibarburu et al. / International Journal of Food Microbiology 143 (2010) 2631

making, would lead to processing decisions (i.e. sulting) to overcome

this drawback.
Most of the EPS-producing bacteria isolated from ropy cider and
wine synthesize an identical 2-branched (1,3)--D-glucan and
belong to the genera Pediococcus (Llaubres et al., 1990; DueasChasco et al., 1997), Lactobacillus (Dueas-Chasco et al., 1998) and
Oenococcus (Ibarburu et al., 2007; Dols-Lafargue et al., 2008). This
polysaccharide synthesis is controlled by a single transmembrane
glycosyltransferase (GTF), which belongs to the COG1215 membrane-bound glycosyltransferase family, and polymerizes glycosyl
residues from UDP-glucose (Walling et al., 2005; Werning et al.,
2006). It is encoded by the gtf gene (Werning et al., 2008), which
has a different genomic location in cider and wine-spoiling LAB
strains, as it is present on plasmids in pediococci and lactobacilli and
on the chromosome in O. oeni (Werning et al., 2006; Dols-Lafargue
et al., 2008). Determination of its nucleotide sequence in Lactobacillus
diolivorans G77, Pediococcus damnosus and Oenococcus oeni I4 revealed that it possesses a 100, 99.9 and 98.8% identity, respectively,
with their counterparts in Pediococcus parvulus 2.6 (Werning et al.,
2006; Dols-Lafargue et al., 2008).
In a previous study, a PCR assay was developed for the detection of
(1,3)(1,2)--D-glucan producing LAB with primers targeted to the
coding sequences of the putative glycosiltransferase domain and the
fth transmembrane segment of the GTF, respectively (Werning et al.,
2006). It allowed the detection of (1,3)(1,2)--D-glucan producing
Pediococcus, L. diolivorans and O. oeni to date reported as cider spoilers.
PCR based methods have quickly been replacing more traditional assays
in the microbiological analysis of food since they are rapid and specic
detection systems. Nowadays, real-time or quantitative PCR (qPCR) is
one of the most promising tools in food control. It is based on the
detection of a uorescent signal and allows the automated detection of
amplicons without post-PCR manipulation, thus reducing the risk of
cross-contamination (McKillip and Drake, 2004). qPCR has successfully
been applied to detect and quantify the presence of Aspergillus
carbonarius in wine grapes (Selma et al., 2008), and yeasts (Hierro et al.,
2007; Tessonnire et al., 2009), acetic (Gonzlez et al., 2006) and lactic
acid bacteria (Neeley et al., 2005) in wine. In addition, Delaherche et al.
(2004) developed a real-time PCR method for specic detection and
quantication in spoiled wine of -glucan producing P. damnosus
strains. However, qPCR procedures have not yet been tested in cider. In
this work a qPCR procedure has been developed for the detection and
quantication of (1,3)(1,2)--D-glucan producing bacteria in cider.
Further, it has been tested in naturally contaminated cider following a
DNA purication step.
2. Material and methods
2.1. Bacterial strains, culture media and growth conditions
Bacterial strains used in this work include 33 cider isolates belonging
to genus Lactobacillus, Oenococcus and Pediococcus together with
reference strains of 16 species and ve genera. Both strains and sources
are listed in Table 1. Cider isolates were obtained from our culture
collection at the Department of Applied Chemistry, University of Basque
Country (CUPV). They had been isolated from spoiled cider along a large
period (from 1994 to 2007) and some of the ropy strains had been
biochemically identied and genetically characterized (Werning et al.,
2006; Ibarburu et al., 2007; Garai-Ibabe et al., 2010). Reference cultures
were supplied by the Spanish Type Culture Collection (CECT), the
National Collection of Industrial, Marine and Food Bacteria (NCIMB), the
ARS Culture Collection (NRRL) and the Belgian Co-ordinated Collections
of Micro-organisms (BCCM/LMG). LAB strains were stored at 80 C in
Man Rogosa Sharpe (MRS) broth (Pronadisa, Madrid, Spain), containing
20% (v/v) glycerol. Before experimental use, bacteria were propagated in
MRS broth at 28 C in an atmosphere containing 5% CO2. To isolate and
select ropy strains from cider, aliquots of the different samples were

27

spiked on modied MRS (Pronadisa, Madrid, Spain) with 10 g/l of

fructose and tomato juice (10% v/v).
2.2. DNA isolation
DNA from pure cultures and cider samples was isolated using the
DNeasy Blood and Tissue Kit (Qiagen GmbH, Hilden, Germany). One
millilitre aliquots were taken and following centrifugation at 8000 g
for 10 min, pellets were washed twice with 1 ml of Ringer's solution
(Oxoid, Hampshire, England) and centrifuged at 8000 g for 5 min.
Pellets were resuspended in 180 l of lysozyme (20 mg/ml) in TE
buffer (10 mM TrisHCl; 1 mM EDTA, pH 8). After 30 min at 37 C,
samples were homogenized with 200 l of lysis buffer and proteinase
K (600 mAU/ml). The homogenate was incubated at 70 C for 30 min.
DNA was puried through the column using two cleaning buffers
supplied in the kit. DNA was eluted in 100 l of ultra pure water
(Sigma) and 5 l was used for PCR amplication.
2.3. Conventional PCR
Amplication of the gtf gene by conventional PCR (Gene Amp PCR
System 2400, Perkin Elmer, USA) was carried out using primers GTFF
and GTFR, which delimit a 417 bp fragment, and according to the
method described by Werning et al. (2006).
2.4. qPCR SYBR Green I-based assay
2.4.1. Primer design
Two primers were designed with the Beacon Designer software (BioRad, Spain) targeting the gtf gene encoding for the GTF glicosiltransferase (Werning et al., 2006). Primer sequences were checked against
sequences available in the GenBank database, using the Basic Local
Alignment Search Tool (BLAST) from the National Center for Biotechnology Information (NCBI, http://www.ncbi.nlm.nih.gov/blast/Blast.
cgi). The sequence and target position (P. parvulus 2.6 gtf sequence,
GenBank accession number AY551933, Werning et al., 2006) for each
primer were GTF3 (5-ATCAAGTCAAAGACCATAAGTCTCTATC-3, 2365
2392 in the putative carboxyl glycosytransferase domain of GTF protein)
and GTF4R (5TAAATAATTGTGTTACTAGTGGAATGTGC-3, 25152486,
in the fourth transmembrane segment of GTF protein). They delimit a
151 bp fragment. Oligonucleotides were synthesized by Eurons MWG/
operon (Ebersberg, Germany).
2.4.2. Set up of the qPCR reaction: amplication conditions
qPCR reactions were performed using SYBR Green I Core Reagents
(Applied Biosystems, Madrid, Spain). Reactions were done in triplicate for
each strain. Amplication mixtures for qPCR, contained in a nal volume of
20 l, 1x buffer (SYBR Green I PCR Buffer), 200 M each dATP, dCTP, dGTP,
and 400 M dUTP; 1U of AmpErase uracil N-glycosidase; 1 U of AmpliTaq
Gold DNA polymerase; 3.5 mM of MgCl2; 200 nM of each primer and 5 l
(10 ng) of template DNA. Different concentrations of MgCl2 (1.5, 3 and
4.5 mM) and primer (100, 200 and 300 nM of each) were assayed. qPCR
assays were carried out in a 7500 Real-Time PCR System (Applied
Biosystems, Foster City, Calif.) programmed to hold at 50 C for 5 min, to
hold at 95 C for 10 min, and to complete 40 cycles of 95 C for 15 s and
55 C for 1 min. PCR results were given as the increase in the uorescence
signal of the reporter dye detected and visualized by the 7500 System SDS
Software provided with the version 1.4 (Applied Biosystems). CT values
(threshold cycle) represent the PCR cycle in which uorescence rst
increased, over a dened threshold (set to a uorescence value of 0.09), for
each amplication plot. Melting curve analysis was determined according
to manufacturers' instructions (SDS software 1.4, Applied Biosystems).
2.4.3. Quantication assays
Standard curves were calculated for quantication purposes using:
i) Ten-fold dilutions of genomic DNA extracted from 1 ml of a log

28

I. Ibarburu et al. / International Journal of Food Microbiology 143 (2010) 2631

Table 1
Bacterial strains used in this study.
Species

Lactobacillus brevis
Lactobacillus buchneri
L. collinoides

Lactobacillus delbrueckii subsp bulgaricus

L. diolivorans
Lactobacillus helveticus
Lactobacillus hilgardii
Lactobacillus mali
Lactobacillus plantarum
L. suebicus

Leuconostoc mesenteroides
Leu. mesenteroides subsp mesenteroides
P. parvulus

P. damnosus
Pediococcus pentosaceus
O. oeni

Streptococcus thermophilus

Strain and sourcea

CECT 216
CECT 4111
CECT 922T
CUPVd 231
CUPVe 234, 235
NCIMB 702772
LMG 19667T
CUPVd G77
NCIMB 700766
CECT 4786T,
CECT 4681
CECT 4149
CUPVe 241, 242, 243
CECT 5917
CUPV 221d
CUPVd 225, 226
NRRL B742
CECT 394
CECT 7350
CUPV 2.6d
CUPVd 1, 2, 22, 23, 24, 26
CECT 793,
CECT 4694
CECT 4695T
CECT 217T
CECT 218
CUPV I4d
CUPVe 302309, 3010, 3011,
30133015, 30173020
NCIMB 700859

6
2

17

qPCRb

EPS phenotype/type of EPSc

CT

TM

N40
35.06 0.32
N40
8.46 0.12
30.49 0.09
N 40
37.76 0.24
10.96 0.16
39.59 0.48
N 40
29.23 0.49
31.15 0.09
30.56 0.31
28.04 0.07
16.88 0.66
38.64 0.14
38.25 0.32
36.98 0.28
35.49 0.15
16.80 0.037
10.31 0.36
36.90 0.20
37.38 0.08
35.19 0.18
28.34 0.26
34.40 0.16
12.71 0.44
31.18 1.65

60.7 0
74.7 0.23
60.7 0
74.4 0.17
74.4 0.15
60.9 0
72.3 0.49
74.1 0
60.6 0
60.7 0
74.7 0
74.4 0
74.4 0.15
74.7 0
74.5 0
60.6 0
74.2 0.28
74.2 0.28
74.2 0
74.1 0
74.4 0.19
60.7 0
74.4 0
74.7 0
74.7 0
74.7 0
74.2 0.17
74.4 0.26

37.93 0.25

74.2 0

+/-D-glucan (Ibarburu, 2009)

+/HePS (Grobben et al., 1997)

+/- and -D-glucan (Dueas-Chasco et al., 1998)

+/HePS (De Vuyst et al., 2001)

+/-D-glucan (Ibarburu, 2009)

+/HePS (Ibarburu, 2009)
+/-D-glucan (Monsan et al., 2001)
+/-D-glucan (Monsan et al., 2001)

+/-D-glucan (Dueas-Chasco et al., 1997)

+/-D-glucan

+/D-glucan and HePS (Ibarburu et al., 2007)

+/HePS

Institutional names: CUPV, Coleccin de la Universidad del Pas Vasco (Spain); CECT, Coleccin Espaola de Cultivos Tipo (Spanish Type Culture Collection, University of Valencia,
Burjassot, Spain); NCIMB, National Collection of Industrial and Marine Bacteria (Aberdeen Scotland, UK); NRRL, Agricultural Research Service (NRRL) Culture Collection (Peoria, Illinois, USA).
b
qPCR parameters: CT SD and TM SD.
c
EPS phenotype (+ or ) assigned by visual examination. EPS type was previously established by determination of the structures in the corresponding references.
d
Isolated from Basque Country ropy cider.
e
Isolated from Basque Country non-ropy cider.
T
Type strain.

phase culture of P. parvulus 2.6, covering the range from 1 to 105 CFU/
reaction (determined by plate count on MRS); ii) Calibrated cell
suspensions prepared from ten-fold dilutions of a log phase culture of
each strain P. parvulus 2.6, and Lactobacillus diolivorans G77 in MRS, and
Oenococcus oeni I4 in MLO (Ibarburu et al., 2007), covering the range
from 1 to 105 CFU/ml (determined by plate count on MRS). In both
assays, 1 ml from each dilution was subjected to DNA extraction using
the DNeasy Blood and Tissue Kit (Qiagen). Puried DNA was recovered
in 100 l of ultra pure water (Sigma) and 5 l of DNA solution was used
as template for qPCR amplication. PCR amplication reactions were
done in triplicate; and iii) Ten-fold dilutions of a 417 bp PCR
amplication product, obtained by conventional PCR from strain
P. parvulus 2.6 as previously described (Werning et al., 2006). This
amplicon includes the targets for qPCR primers.

2.4.4. Sensitivity of qPCR assays for (1,3)(1,2)--D-glucan producing

bacteria in an articially contaminated cider
Sensitivity assays were carried out from calibrated cell suspensions
prepared from ten-fold dilutions in cider of a log phase culture (36 h) of
P. parvulus 2.6, obtained also by incubation in cider, and covering the
range from 0.1 to 105 CFU/ml (determined by plate count on MRS). 1 ml
aliquots were taken for DNA extraction using the DNeasy Blood and
Tissue Kit (Qiagen) as described above. Puried DNA was recovered in
100 l of ultra pure water (Sigma) and 5 l of DNA solution was used as
template for qPCR amplication. Reactions were carried out in triplicate
per dilution.

2.5. Analysis of ropy and non-ropy cider samples by qPCR

The qPCR procedure was applied for the quantitative detection
of gtf+ LAB population during cidermaking in 13 tanks from two
cider factories with frequent incidence of ropiness. Samples were
collected during three consecutive months: during the simultaneous alcoholic and malolactic fermentations (October sample)
and during maduration period, before bottling (November and
December samples). Density (g/l) and pH measurements were
determined as described by Dueas et al. (1995). DNA was
extracted from 1 ml aliquots using the DNeasy Blood and Tissue
Kit (Qiagen) as described above. Puried DNA was recovered in
100 l of ultra pure water (Sigma) and 5 l of DNA solution was
used as template for qPCR amplication. Reactions were carried out
in triplicate. The absence of PCR inhibitors in the samples matrix
was tested using as template puried DNA from P. parvulus 2.6
together with 5 l DNA from the non-inoculated negative control.
This reaction performed in triplicate was considered an internal
amplication control.

2.6. Statistical analysis

Statistical analysis was performed using SPSS 16.0 software (SPPS
Inc., Chicago, Illionois, USA). With CT values obtained from EPS
negative strains, a lower limit of the unilateral condence interval for
a signicance level of 5% was determined.

I. Ibarburu et al. / International Journal of Food Microbiology 143 (2010) 2631

3. Results and discussion

3.1. Optimization of the qPCR reaction and specicity
In this work, a SYBR Green based qPCR procedure was developed by
designing specic primers targeting the gtf gene present in all -glucan
producers (Werning et al., 2006). When tested in silico they showed
100% homology with glycosyltransferases encoding genes for (1,3)
(1,2)--D-glucan synthesis in P. damnosus IOEB8801, O. oeni IOEB 0205,
L. diolivorans G77 and O. oeni I4 sequences. Specicity was further tested
in vitro by conventional PCR and the expected 151 bp amplicon, was
only obtained in the (1,3)(1,2)--D-glucan producers (Table 1).
Improved specicity and qPCR reaction efciency was obtained
using 200 nM of each oligonucleotide and 4.5 mM of MgCl2. Primer
specicity was analyzed by qPCR using the optimized reaction
conditions and puried DNA (10 ng/reaction) from the 54 strains
listed in Table 1. This assay, followed by the melting curve analysis,
conrmed the generation of a single PCR product from the glucan producers, with Tm values between 74.1 and 74.5. CT values
obtained ranged between 8.46 and 16.88 (average 12.67 3.5).
Despite amplication was only obtained in EPS producers by
conventional PCR, the qPCR approach showed a slight amplication
signal in some of the EPS LAB strains, both in reference strains
and in cider isolates, with CT values ranging from 28.04 to 37.76.
These results indicate a low efciency in amplication that could be
attributed to either the presence of these targets in low numbers or to
the occurrence of similar but not identical targets. From the statistical
analysis of the CT data of these EPS strains, a critical value of 28.54
with the signicant level set at 0.05 was established, in order to assess
the detection of (1,3)(1,2)--D-glucan producing strains. They rendered CT values signicantly higher (P(CT b 28.54)= 0.05) than those of
the -glucan producers. This CT (28.54) value was considered the
threshold to discriminate (1,3)(1,2)--D-glucan producing strains from
non producers.
3.2. Standard curves and detection limits of the qPCR
As shown in previous studies, the gtf gene is present in most of
the naturally occurring LAB species in cider, such as O. oeni,
Lactobacillus suebicus, Lactobacillus collinoides, P. parvulus, and
L. diolivorans (Werning et al., 2006). Of them, P. parvulus, and

29

L. diolivorans are the predominant species meanwhile O. oeni

became the most abundant microbiota during malolactic fermentation. In order to approach quantication of -glucan producers,
standard curves were constructed using P. parvulus 2.6, O. oeni I4
and L. diolivorans G77 as representative producer strains. Standard
curves derived from both, puried DNA and calibrated cell
suspensions of P. parvulus 2.6, showed a linear correlation between
Log10 input DNA and CT in the range from 101 to 105 CFU/reaction.
Slope values were 3.58 and 3.31, respectively, very close to the
theoretical optimum of 3.32 (Higuchi et al., 1993) and R2 (square
correlation coefcient after the linear regression) values were above
0.98, indicating that the SYBR Green PCR assay was highly linear
(Table 2). Quantication limit, dened as the lowest concentration in
which the linearity is maintained, was established in 0.43 and 26 CFU/
reaction (5.2 102 CFU/ml), using puried DNA and cell suspensions,
respectively. Moreover, detection was possible in both cases at the
lowest level assayed, reaching 0.43 and 0.26 CFU/reaction (5.2 CFU/
ml), respectively. A high correspondence was found between CT values
obtained from puried DNA and calibrated cell suspensions of P.
parvulus 2.6 for equivalent concentrations (CFU/reaction), which
indicates that the DNA extraction procedure used was efcient.
Standard curves obtained from calibrated cell suspensions of the
(1,3)(1,2)--D-glucan producing O. oeni I4 and L. diolivorans G77
strains showed slope values similar to the one corresponding to the cell
suspensions of P. parvulus 2.6 (Table 2). The quantication limit was set
at 58 and 393 CFU/reaction for O. oeni I4 and L. diolivorans G77,
respectively. Therefore, the quantication range is between 10 and
105 CFU/reaction, similar to that of P. parvulus 2.6, except for L.
diolivorans G77 that ranged from 102 to 105 CFU/reaction.
Due to the biodiversity of -glucan producing strains and the
lack of information about the copy number of gtf gene in each
species, a third approach was tested to construct a universal
standard curve to approach quantication of gtf gene in cider, using
a ten-fold serially diluted PCR product from P. parvulus 2.6, that
included the qPCR primer sequences. Linear regression analysis of
the CT values showed good linearity (R2 0.998) between 3.19 and
3.19 107 molecules/reaction (Table 2).
In summary, CT values obtained from calibrated cell suspensions of
P. parvulus 2.6, and O. oeni I4 showed a linear relationship over 5
orders of magnitude, and a quantication range between 10 and
105 CFU/reaction. However, when using P. parvulus 2.6 DNA (either

Table 2
Standard curves, amplication efciency and CT values obtained from P. parvulus 2.6 genomic, DNA, DNA extracted from calibrated cell suspensions of P. parvulus 2.6, O. oeni I4 or L.
diolivorans G77, and PCR amplication product from P. parvulus 2.6.

R2
sc
AE
n

DNA

Cell suspensions

P. parvulus 2.6

P. parvulus 2.6

O. oeni I4

L. diolivorans G77

PCR product
P. parvulus 2.6

0.998
y = 3.58x + 35.68
0.9
7

0.984
y = 3.31x + 31.28
1
5

0.977
y = 3.41x + 36.20
0.96
5

0.987
y = 4.66x + 35.37
0.64
4

0.998
y = 3.64x + 33.99
0.88
7

CFU/reaca

CT SDb

CFU/reacc

CT SDb

CFU/reacc

CT SDb

CFU/reacc

CT SDb

molecules/reacd

CT SDb

4.3 105
4.3 104
4.3 103
4.3 102
43
4.3
0. 43

14.70 0.138
17.82 0.16
21.33 0.11
25.06 0.05
28.40 0.15
32.24 0.16
35.69 0.32

2.6 105
2.6 104
2.6 103
2.6 102
26
2.6
0.26

14.26 0.58
18.74 0.16
22.01 0.04
24.20 0.69
28.19 0.15
27.47 0.45
26.17 1.23

5.83 105
5.83 104
5.83 103
5.83 102
58.3
5.83

14.76 0.24
18.22 0.12
23.09 0.33
27.27 0.42
30.33 0.08
31.51 0.24

3.93 105
3.93 104
3.93 103
3.93 102
39.3
3.93

16.38 0.31
21.44 0.98
26.99 0.49
30.06 0.38
31.95 0.41
31.97 0.61

3.19 106
3.19 105
3.19 104
3.19 103
3.19 102
31.9
3.19

12.14 0.13
15.61 0.05
19.30 0.21
22.99 0.27
26.39 0.23
30.00 0.41
34.52 0.40

R2, square correlation coefcient, sc, standard curve created by plotting CT value versus log concentration. Equation describing data obtained by linear regression, AE, Amplication
efciency, n, number of points on the standard curve corresponding to the number of dilutions of DNA.
a
CFU/reaction calculated from DNA concentrations (4.2 ng corresponds to 4.3 105 CFU).
b
Mean CT values standard deviation (SD) from three independent cultures, each performed in triplicate.
c
Mean values of three separate cell cultures expressed as CFU/reaction.
d
Molecules/reaction calculated from PCR amplication product.

Values that are not linear and they were not considered for the linear regression analysis.

30

I. Ibarburu et al. / International Journal of Food Microbiology 143 (2010) 2631

Table 3
Quantication of -glucan producing LAB populations in ropy ciders using SYBR Green qPCR.
Sample

S1
S2
S3
S4
S5
S6
S7
S8
a
b
c

CT DSa

18.31 0.61
18.46 0.46
19.33 0.47
17.67 0.13
15.14 0.46
15.34 0.13
25.11 0.62
21.55c 0.14

LAB plating
(Log CFU/ml)b

Quantication from standard curves (Log CFU/ml)

Amplicon

DNA

P. parvulus 2.6 cell

suspensions in MRS

P. parvulus 2.6 cell

suspensions in cider

O. oeni I4 cell
suspensions in MRS

L. diolivorans G77cell
suspensions in MRS

5.38
5.41
6.16
7.16
7.09
5.00
7.41
6.23

6.19
6.15
5.91
6.36
7.5
7
3.72
4.70

6.75
6.71
6.47
6.93
7.64
7.58
4.25
5.25

5.82
5.77
5.51
6.01
6.78
6.72
3.16
4.24

6.14
6.10
5.88
6.31
6.98
6.92
3.76
4.69

7.15
7.10
6.85
7.33
8.08
8.02
4.55
5.59

6.91
6.86
6.61
7.09
7.81
7.76
4.36
5.38

Mean CT value standard deviation (SD).

Total counts on MRS agar.
Value corresponding to the dilution of the sample.

total genomic DNA or the 417 bp amplicon) linearity was extended

over 7 log orders.

3.4. Detection of (1,3)(1,2)--D-glucan producing bacteria along

cidermaking process and ropy bottled ciders

3.3. Sensitivity of the qPCR assay on articially contaminated cider samples

As sulting is not a usual practice in Basque Country cidermaking,

alcoholic and malolactic fermentations take place simultaneously
and, thereafter, LAB population became always dominant during the
maturation period (Dueas et al., 1995). In addition, natural ciders
are not microbiologically stabilized before bottling and, as a
consequence, bottled ciders are prone to spoilage by -glucan
producers. In this context, the developed qPCR method was used
for quantication of the (1,3)(1,2)--D-glucan producers along
cidermaking in 13 vats from two cider factories, with frequent
incidence of ropiness (Table 3). In some tanks, a statistically
signicant decrease of CT values was found (P b 0.05), with Tm values
of the single amplication product specic for the -glucan
producing LAB, ranging from 74.1 to 74.5, as described above,. All
these results indicated a clear increase in LAB gtf + populations
during the maturation period, after alcoholic and malolactic
fermentations.
The qPCR procedure was also used to estimate the gtf+
populations in 8 ropy ciders and they were compared to the levels
of total lactic acid bacteria estimated by total counts on MRS.
Results are shown in Table 4. Amplication was detected in all
samples and melting curves analysis showed that Tm values ranged
from 74.1 to 74.4. The CT values of six ciders (S1 to S6) ranged
between 15.14 and 19.33 and were signicantly lower than those
found throughout the cidermaking process (Table 3), demonstrating a clear enrichment of gtf+ LAB populations, in comparison to the

The qPCR procedure developed was tested for gtf+ LAB detection
using cider contaminated with serial ten-fold dilutions of P. parvulus 2.6.
The regression analysis showed a good linearity for concentrations
between 16 and 1.65 105 CFU/reaction (R2 = 0.97) and the slope was
of 3.81. These values indicate that linearity of the amplication
reaction was also kept in the inoculated cider. However, amplication
efciency (AE= 0.83) was slightly lower than that obtained from cell
suspensions in MRS (AE = 1). Therefore, the sensitivity of qPCR
evaluated on calibrated cell suspensions of P. parvulus 2.6, in cider or
in MRS broth, showed detection limits around one log lower than the
values obtained from puried DNA. This result can be explained by the
DNA loss produced during the extraction step (Alarcn et al., 2006;
Fernndez et al., 2006). Furthermore, the standard curves obtained in
both, MRS or cider, showed a linear relationship over 5 magnitude
orders, allowing quantication in a range from 10 to 105 CFU/reaction,
but the amplication efciency was lower in the samples from
inoculated cider. This decrease could be explained by the presence of
inhibitory compounds such as polyphenols in the beverage. Similar
inhibitory effect on PCR reaction has been reported in wines by some
authors (Delaherche et al., 2004; Martorell et al., 2005). With respect to
quantication and detection limits, they are in the same range than the
ones reported for other LAB i.e. Lactobacillus sakei (Martn et al., 2006)
and Leuconostoc mesenteroides (Elizaquvel et al., 2008).

Table 4
Evolution of physico-chemical parameters and -glucan producing bacteria along cidermaking process.
Sample

pH
a

I5
I7
I8
I9
I10
I12
I13
I14
A2
A4
A5
A6
A10

qPCR SYBR Green (CT SD)b

Density (g/l)

1a

3.77
3.75
3.77
3.72
3.71
3.65
3.68
3.77
3.99
4.18
3.92
4.15
3.89

3.95
4.01
3.99
4.03
4.03
4.02
4.04
3.99
3.97
4.07
4.12
4.11
3.95

3.93
3.96
3.92
3.96
3.95
4.00
4.04
3.95
3.94
4.02
4.03
nd
3.88

1.027
1.029
1.027
1.038
1.001
1.022
1.027
1.011
1.025
1.027
1.035
1.023
1.010

0.997
0.995
0.998
0.999
0.997
0.995
0.999
0.997
0.997
0.998
1.000
0.998
1.001

0.995
0.994
0.996
0.996
0.995
0.995
0.997
0.996
0.995
0.995
0.998
0.996
0.997

32.59 0.64
35.21 0.68
29.68 0.40
32.65 0.49
35.44 0.30
33.27 0.26
36.5 0.31
34.56 0.75
40
32.79 0.87
35.58 0.60
40
40

24.54 0.37
31.81 0.63
26.07 0.44
28.35 0.01

24.19 0.30
31.88 0.16
28.48 0.26

nd, not determined.

a
Sampling event: 1, October; 2, November; 3, December.
b
Mean CT value standard deviation (SD). Reactions were performed in triplicate.
These samples were considered positive for the presence of -glucan producers with P b 0.05.

33.23 0.32
33.96 0.25
33.18 0.01
30.47 0.25
27.26 0.41
34.54 0.46
26.07 0.24
31.53 0.45
40

29.93 0.61
26.17 0.08
30.93 0.67
37.61 0.67
30.33 0.37
27.13 0.22
30.11 0.09
nd
30.46 0.25
40

I. Ibarburu et al. / International Journal of Food Microbiology 143 (2010) 2631

levels found along this period. Sample S7 showed a CT value of 25.11

and sample S8 did not show amplication until it was diluted onehalf (external amplication control indicated absence of PCR
inhibitors). Quantication of gtf+ populations approached by
interpolating CT values in the different standard curves, showed
values within a log range, which corroborates the adequate
performance of the developed qPCR method as a rapid procedure
to estimate -glucan producing LAB in cider. Quantication of gtf+
populations derived from the different standard curves showed no
signicant differences. Therefore, the amplicon-derived standard
curve is proposed for the routine estimation of gtf+ populations in
cider. Although cell numbers were probably overestimated because
of the qPCR inability to differentiate between living and dead cells
(Neeley et al., 2005), our results indicate that in most of spoiled
ciders the gtf+ population is the predominant microbiota, ranging
from 105 to 107 CFU/ml.
On the basis of these results, we conclude that the SYBR Green
I assay developed in this study is suitable for a sensitive and rapid
detection of the (1,3)(1,2)--D-glucan LAB producers and provides
a good tool for its early detection during cider making process.
This knowledge will allow cidermakers to make decisions to
control and avoid spoilage of ciders by sulting or by other
microbiological stabilization method. This constitutes a great
advantage in comparison to current situation, in which cider
contamination by -glucan producing lactic acid bacteria is not
detected until occurrence of ropiness and the beverage is refused
by the consumer.
Acknowledgments
This work was supported by the Ministerio de Educacin y Ciencia
(projects AGL2006-11932, AGL-2009-12998 and CSD2007-00063),
the Universidad del Pas Vasco (UPV/EHU) (EHU08/37) and the
Diputacin Foral de Gipuzkoa (Programa Red Gipuzkoana de Ciencia,
Tecnologa e Innovacin, co-nanced by the European Union), and
the Generalitat Valenciana (GVACOMP2009-257). Idoia Ibarburu
acknowledges the Universidad del Pas Vasco for the predoctoral
fellowship.
References
Alarcn, B., Vicedo, B., Aznar, R., 2006. PCR-based procedures for detection and
quantication of Staphylococcus aureus and their application in food. Journal of
Applied Microbiology 100, 352364.
Delaherche, A., Claisse, O., Lonvaud-Funel, A., 2004. Detection and quantication of
Brettanomyces bruxellensis and ropy Pediococcus damnosus strains in wine by
real-time polymerase chain reaction. Journal of Applied Microbiology 97,
910915.
De Vuyst, L., De Vin, F., Vaningelmem, F., Degeest, B., 2001. Recent developments in the
biosynthesis and applications of heteropolysaccharides from lactic acid bacteria.
International Dairy Journal 11, 687707.
Dols-Lafargue, M., Lee, H.Y., Le Marrec, C., Heyraud, A., Chambat, G., Lonvaud-Funel, A.,
2008. Characterization of gtf, a glucosyltransferase gene in the genomes of
Pediococcus parvulus and Oenococcus oeni, two bacterial species commonly found
in wine. Applied and Environmental Microbiology 74, 40794090.
Dueas, M.T., Irastorza, A., Fernndez, K., Bilbao, A., 1995. Heterofermentative Lactobacilli
causing ropiness in Basque country ciders. Journal of Food Protection 58, 7680.
Dueas-Chasco, M.T., Rodrguez-Carvajal, M.A., Tejero-Mateo, P., Franco-Rodrguez, G.,
Espartero, J.L., Irastorza-Iribas, A., Gil-Serrano, A.M., 1997. Structural analysis of the
exopolysaccharide produced by Pediococcus damnosus 2.6. Carbohydrate Research
303, 453458.

31

Dueas-Chasco, M.T., Rodrguez-Carvajal, M.A., Tejero-Mateo, P., Espartero, J.L.,

IrastorzaIribas, A., Gil-Serrano, A.M., 1998. Structural analysis of the exopolysaccharides produced by Lactobacillus spp. G-77. Carbohydrate Research 307, 125133.
Elizaquvel, P., Chenoll, E., Aznar, R., 2008. A TaqMan-based real-time PCR assay for the
specic detection and quantication of Leuconostoc mesenteroides in meat
products. FEMS Microbiology Letters 278, 6271.
Fernndez, K., Dueas, M., Irastorza, A., Bilbao, A., del Campo, G., 1995. Characterization
and DNA plasmid analysis of ropy Pediococcus damnosus spp. strains isolated from
Basque Country ciders. Journal of Food Protection 59, 3540.
Fernndez, M., Del Ro, B., Linares, M.D., Martn, M.C., Alvarez, M.A., 2006. Real-time
polymerase chain reaction for quantitative detection of histamine-producing
bacteria: use in cheese production. Journal of Dairy Science 89, 37633769.
Garai, G., Dueas, M.T., Irastorza, A., MartnAlvarez, P.J., Moreno-Arribas, M.V., 2006.
Biogenic amines in natural ciders. Journal of Food Protection 69, 30063012.
Garai-Ibabe, G., Areizaga, J., Aznar, R., Elizaquivel, P., Prieto, A., Irastorza, A., Dueas,
M.T., 2010. Screening and selection of 2-branched (1, 3)--D-glucan producing lactic
acid bacteria and exopolysaccharide characterization. Journal of Agricultural and
Food Chemistry 58, 61496156.
Gonzlez, A., Hierro, N., Poblet, M., Mas, A., Guillamn, J.M., 2006. Enumeration and
detection of acetic acid bacteria by real-time PCR and nested PCR. FEMS
Microbiological letters 254, 123128.
Grobben, G.J., van Casteren, W.H.M., Schools, A., Oosterveld, A., Sala, G., Smith, M.R.,
Sikkema, J., de Bont, J.A.M., 1997. Analysis of exopolysaccharides produced by Lb.
delbrueckii subsp. bulgaricus NCFB 2772 grown in continuous culture on glucose
and fructose. Applied Microbiology and Biotechnology 48, 516521.
Hierro, N., Esteve-Zarzoso, B., Mas, A., Guillamn, J.M., 2007. Monitoring of
Saccharomyces and Hanseniaspora populations during alcoholic fermentation by
real-time quantitative PCR. FEMS Yeast Research 7, 13401349.
Higuchi, R., Fockler, C., Dollinger, G., Watson, R., 1993. Kinetic PCR analysis: real-time
monitoring of DNA amplication reactions. Biotechnology 11, 10261030.
Ibarburu, I. 2009. Bacterias lcticas productoras de exopolisacridos: anlisis estructural de polisacridos y deteccin molecular de estirpes que sintetizan -(1,3)
(1,2)-D-glucanos. PhD thesis, Universidad del Pas Vasco (UPV/EHU), Donostia-San
Sebastin, Spain.
Ibarburu, I., Soria-Daz, M.E., Rodrguez-Carvajal, M.A., Velasco, S.E., Tejero-Mateo, P.,
Gil-Serrano, A.M., Irastorza, A., Dueas, M.T., 2007. Growth and exopolysaccharide
(EPS) production by Oenococcus oeni I4 and structural characterization of their
EPSs. Journal of Applied Microbiology 103, 477486.
Llaubres, R.M., Richard, B., Lonvaud, A., Dubourdieu, D., Fournet, B., 1990. Structure of
an exocellular -D-glucan from Pediococcus sp., a wine lactic bacteria. Carbohydrate
Research 203, 103107.
McKillip, J.L., Drake, M., 2004. Real-time nucleic acid-based detection methods for
pathogenic bacteria in food. Journal of Food Protection 67, 823832.
Martn, B., Jofr, A., Garriga, M., Pla, M., Aymerich, T., 2006. Rapid quantitative detection
of Lactobacillus sakei in meat and fermented sausages by real-time PCR. Applied and
Environmental Microbiology 72, 60406048.
Martorell, P., Querol, A., FernndezEspinar, M.T., 2005. Rapid identication and
enumeration of Saccharomyces cerevisiae cells in wine by Real-Time PCR. Applied
and Environmental Microbiology 71, 68236830.
Monsan, P., Bozonnet, S., Albenne, C., Joucla, G., Willemot, R.M., RemaudSimeon, M.,
2001. Homopolysaccharides from lactic acid bacteria. International Dairy Journal
11, 675685.
Neeley, E.T., Phister, T.G., Mills, D.A., 2005. Differential real-time PCR assay for enumeration
of lactic acid bacteria in wine. Applied and Environmental Microbiology 71, 89548957.
Ruas-Madiedo, P., Abraham, A., Mozzi, F., de los Reyes-Gaviln, C.G., 2008. Functionality
of exopolysaccharides produced by lactic acid bacteria. In: Mayo, B., Lpez, P.,
Prez-Martn, G. (Eds.), Molecular aspects of lactic acid bacteria for traditional and
new applications. Research Signpost, Kerala, India, pp. 137166.
Selma, M.V., Martnez-Culebras, P.V., Aznar, R., 2008. Real-time PCR based procedures
for detection and quantication of Aspergillus carbonarius in wine grapes.
International Journal of Food Microbiology 122, 126134.
Tessonnire, H., Vidal, S., Barnavon, L., Alexandre, H., Remize, E., 2009. Design and performance testing of a real-time PCR assay for sensitive and reliable direct quantication
of Brettanomyces in wine. International Journal of Food Microbiology 129, 237243.
Walling, E., Gindreau, E., Lonvaud-Funel, A., 2005. A putative glucan synthase gene dps
detected in exopolysaccharide-producing Pediococcus damnosus and Oenococcus
oeni strains isolated from wine and cider. International Journal of Food
Microbiology 98, 5362.
Werning, M.L., Ibarburu, I., Dueas, M.T., Irastorza, A., Navas, J., Lpez, P., 2006.
Pediococcus parvulus gtf gene encoding the GTF glycosyltransferase and its
application for specic PCR detection of -D-glucan-producing bacteria in foods
and beverages. Journal of Food Protection 69, 161169.
Werning, M.L., Corrales, M.A., Prieto, A., Fernndez de Palencia, P., Navas, J., Lpez, P.,
2008. Heterologous expression of a 2-substituted-1, 3)--D-glucan in Lactococcus
lactis. Applied and Environmental Microbiology 74, 52595262.