Food Chemistry 129 (2011) 1037–1044

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Investigation of the factors influencing the survival of Bifidobacterium longum

in model acidic solutions and fruit juices
Sawaminee Nualkaekul a, Ivan Salmeron b, Dimitris Charalampopoulos a,⇑
a
Department of Food and Nutritional Sciences, The University of Reading, P.O. Box 226, Reading RG6 6AP, UK
b
Facultad de Ciencias Quimicas, Universidad Autonoma de Chihuahua, Chihuahua, Chih., Mexico C.P. 31125, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: The survival of Bifidobacterium longum NCIMB 8809 was studied during refrigerated storage for 6 weeks
Received 12 January 2011 in model solutions, based on which a mathematical model was constructed describing cell survival as a
Received in revised form 12 March 2011 function of pH, citric acid, protein and dietary fibre. A Central Composite Design (CCD) was developed
Accepted 18 May 2011
studying the influence of four factors at three levels, i.e., pH (3.2–4), citric acid (2–15 g/l), protein (0–
Available online 25 May 2011
10 g/l), and dietary fibre (0–8 g/l). In total, 31 experimental runs were carried out. Analysis of variance
(ANOVA) of the regression model demonstrated that the model fitted well the data. From the regression
Keywords:
coefficients it was deduced that all four factors had a statistically significant (P < 0.05) negative effect on
Probiotics
Bifidobacteria
the log decrease [log10N0 weeklog10N6 week], with the pH and citric acid being the most influential ones.
Storage Cell survival during storage was also investigated in various types of juices, including orange, grapefruit,
Survival blackcurrant, pineapple, pomegranate and strawberry. The highest cell survival (less than 0.4 log
Predictive modeling decrease) after 6 weeks of storage was observed in orange and pineapple, both of which had a pH of about
Fruit juices 3.8. Although the pH of grapefruit and blackcurrant was similar (pH 3.2), the log decrease of the former
was 0.5 log, whereas of the latter was 0.7 log. One reason for this could be the fact that grapefruit con-
tained a high amount of citric acid (15.3 g/l). The log decrease in pomegranate and strawberry juices was
extremely high (8 logs). The mathematical model was able to predict adequately the cell survival in
orange, grapefruit, blackcurrant, and pineapple juices. However, the model failed to predict the cell sur-
vival in pomegranate and strawberry, most likely due to the very high levels of phenolic compounds in
these two juices.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction suggested that the minimum concentration of live probiotic bacte-

Bifidobacteria are natural habitants of the human gastrointesti-
nal tract and can exert several beneficial effects to the host (Julio
et al., 2010). A number of health benefits have been associated
and are under investigation for bifidobacteria including protection
against diarrhoea, relief of lactose intolerance, resistance to
microbial infections, cancer prevention, treatment of inflammatory
bowel disease, improvement of constipation, immune stimulation,
and lowering of cholesterol (Leahy, Higgins, Fitzgerald, & Sinderen,
2005; Miyauchi, Xiao, & Tanabe, 2010; Williams, Ha, & Ciorba,
2010).
As a result, they have been incorporated into a variety of food
products, mainly dairy, such as fermented milks and yogurts
(Sanchez, de los Reyes-Gavilan, Margolles, & Gueimonde, 2009).
To this end, the viability of the selected strain in the product during
its manufacture and storage is very important. It has been
⇑ Corresponding author. Tel.: +44 118 3788216; fax: +44 118 931 0080.
E-mail address: d.charalampopoulos@reading.ac.uk (D. Charalampopoulos).
ria, such as lactobacilli or bifidobacteria, at the expiry date of the
product should be above 107 CFU/ml (Homayouni, Azizi, Ehsani,
Yarmand, & Razavi, 2008; Krasaekoopt, Bhandari, & Deeth, 2003),
as it is likely that the cell concentration will decrease during pas-
sage through the stomach and the intestine (Shah, 2000).
Fruit juices have a big market sector as functional drinks, as
they are good sources of vitamins and minerals and are consumed
frequently and loyally by consumers (Luckow & Delahunty, 2004).
They have also the advantage that they are suitable for consumers
that are lactose intolerant (Prado, Parada, Pandey & Soccol, 2008).
Although significantly less amount of research has been conducted
on the factors that influence the viability of probiotics during stor-
age in fruit juices compared to dairy products, it has been shown
that the storage temperature and time (Saarela, Virkajarvi, Alak-
omi, Mattila, & Matto, 2006), the presence of dietary fibre in the
product (Saarela, Virkajarvi, Nohynek, Vaari, & Matto, 2006), and
the type of container used (Champagne, Raymond, & Gagnon,
2008) are likely to have an important role. However, there is very
little information on the effect of the chemical composition of the
fruit juices on probiotic survival.

0308-8146/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2011.05.071
1038 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044
In this study, a Bifidobacterium longum strain was used as a
model probiotic microorganism. This particular strain was
previously shown to have good survival under acidic conditions
(Clark, Cotton, & Martin, 1993; Sanchez et al., 2007), good bile tol-
erance (Clark & Martin, 1994), and modulate the host’s immune
system (Medina, Izquierdo, Ennahar, & Sanz, 2007). The aim of
the study was to develop a mathematical model describing the sur-
vival of the B. longum strain during refrigerated storage in model
solutions, as a factor of the pH, and the citric acid, protein and die-
tary fibre contents. These results were then used to explain the sur-
vival of B. longum in various types of fruit juices taking into account
their composition.
2. Materials and methods
2.1. Preparation of bacterial culture
B. longum NCIMB 8809 (National Collection of Industrial and
Marine Bacteria, UK), isolated from nursling stools, was used in this
study. The strain was preserved in 2 ml cryovials containing 40%
glycerol, and was stored at 80 °C. The culture was revived by
spread plating on Wilkins-Chalgren agar (Oxoid, UK). A single col-
ony was then used to inoculate a Hungate tube containing 10 ml of
Trypticase-Phytone-Yeast Extract Broth (TPY), which consisted
(per litre) of: trypticase (10 g), phytone (5 g), glucose (15 g), yeast
extract (2.5 g), tween 80 (1 ml), cysteine hydrochloride (0.5 g),
di-potassium hydrogen phosphate (2 g), magnesium chloride-
hexahydrate (0.5 g), zinc sulphate heptahydrate (0.25 g), calcium
chloride (0.15 g), ferric chloride (5 drop, 0.005%w/v) (Rosaria
D’Aimmo, Modesto & Biavati, 2007). The cells were cultivated at
200 rpm, 37 °C, 16 h, under anaerobic conditions, and were har-
vested by centrifugation at 3200g, for 15 min. The pellets were
washed once with 0.1 M PBS (Oxoid, UK) and resuspended in
3 ml PBS, resulting in a 10-fold concentration. The cell concentra-
tion in the PBS suspension was approximately 2  109 CFU/ml.
2.2. Experimental design
The experiments were designed on the basis of a Central Com-
posite Design (CCD). Four factors (pH, citric acid, protein and die-
tary fibre) were studied at three different levels, pH (3.2–4),
citric acid (2–15 g/l), protein (0–10 g/l), and dietary fibre (0–8 g/
l). In total, thirty-one experimental runs were performed (Table 1).
Multiple regression analysis was carried out in order to fit a sec-
ond order polynomial equation, described below, to the data:
Y ¼ b0 þ b1 X 1 þ b2 X 2 þ b3 X 3 þ b4 X 4 þ b11 X 21 þ b22 X 22 þ b33 X 23
þ b44 X 24 þ b12 X 1 X 2 þ b13 X 1 X 3 þ b14 X 1 X 4 þ b23 X 2 X 3 þ b24 X 2 X 4
þ b34 X 3 X 4
where Y is the response log decrease [log10N0 weeklog10N6 week], b0,
b1, b2, b3,..., b34 are the regression coefficients, and X1, X2, X3, X4 are
the pH, and the citric acid, protein and dietary fibre concentrations,
respectively. The modeling was conducted by the multiple regres-
sion method using SPSS Statistics (Ver. 18) (SPSS, USA). The data
were statistically treated by analysis of variance (ANOVA). Three-
and higher-order interactions were neglected.
2.3. Preparation of model solutions
The model solutions contained 50 g/l sucrose, 25 g/l glucose and
25 g/l fructose (Fisher Scientific, UK), and variable amounts of citric
acid (Fisher Scientific, UK), protein (Vegetable peptone No.1, Oxoid,
UK) and pectin as dietary fibre (Fisher Scientific, UK), according to
Table 1. Their pH values were adjusted with 2 M NaOH (Sigma,
UK); then the solutions were filter-sterilised (0.2 lm) into 60 ml
sterile bottles and separated into small bottles under anaerobic
conditions. Appropriate amounts of the PBS/cell suspension were
added so that the initial cell concentration was approximately
1  108 CFU/ml. The solutions were stored at 4 °C for 6 weeks.
Samples were collected every week and analysed for pH, composi-
tional changes and for the concentrations of live and dead cells.
2.4. Bacterial enumeration
The LIVE/DEADÒ BacLight TM Bacterial Viability Kit (Invitrogen,
USA) was used to enumerate live and dead cells. The kit consists of
two dyes; the SYTO 9 stain generally labels all bacteria in a popu-
lation, and propidium iodide penetrates only bacteria with dam-
aged membranes. For the experiment, 1 ll of the dye mixture
was added in 100 ll of sample, and the cell suspension was incu-
bated in the dark at room temperature for 15 min. The number
of viable and dead cells was then counted using a fluorescence
microscope (Nikon Microphot-SA, Japan).
2.5. Chemical analyses
Sugar concentrations (sucrose, glucose and fructose) were
determined by high performance liquid chromatography (HPLC)
using an Agilent HPLC 1100 system, consisting of a Rezex RCM-
Monosaccharide Ca + column linked to a Carbo-Ca (4  3.0 mm)
guard column (Phenomenex, USA), (300  7.80 mm), and a refrac-
tive index detector (Shodex RI-71). The eluent was HPLC grade
water with a flow rate of 0.5 ml/min; the temperature was set at
84.5 °C.
Citric, lactic and acetic acids were determined using an Agilent
1100 HPLC system, consisting of a Prevail™ organic acid column
(Alltech, USA), (4.6  250 mm) and a 210 nm UV detector (Hewlett
Packard Interface 35900). The eluent was 25 mM KH2PO4 (adjusted
to pH 2.5 with phosphoric acid) with a flow rate of 1.0 ml/min; the
analysis was conducted at room temperature.
The total dietary fibre in the juice samples was determined by a
combination of enzymatic and gravimetric methods (AOAC, 1997),
using a total dietary fibre assay kit (Sigma, UK). One gramme of
freeze-dried sample of each juice was dissolved into 50 ml of
0.08 M sodium phosphate buffer (pH 6.0). a-Amylase (0.1 ml)
was then added and the contents mixed in a water bath at 95 °C
for 15 min. After cooling, the pH of the solutions was adjusted to
7.5 by adding 10 ml of 0.275 M NaOH; then 0.1 ml of protease solu-
tion was added, and the solutions were incubated at 60 °C for
30 min. Subsequently, the pH of the solutions was adjusted be-
tween pH 4.0 and 4.6 using 0.325 M HCl; 0.1 ml of amyloglucosi-
dase was then added and the solutions were incubated at 60 °C
for 30 min. Four volumes of 95% ethanol were added into each
solution; these were left overnight at room temperature to allow
complete precipitation. The solution was filtered and the residue
was washed thrice with 20 ml of 78% ethanol, twice with 10 ml
of 95% ethanol, and twice with 10 ml acetone, and was then air-
dried. After drying, the residue was weighed and the sample split
in two. One half was used for analysing the protein content, using
the bicinchoninic acid protein assay kit (Sigma, UK) and the other
half the ash content. The total dietary fibre was calculated as the
weight of the residue minus the weights of protein and ash.
The total phenol content of juices was determined using the
Folin–Ciocalteu assay (Lukanin, Gunko, Bryk, & Nigmatullin,
2003). The juices were initially clarified by centrifugation
(10000g, 10 min, 4 °C). Then, 0.5 ml of each clarified juice was
mixed with 0.5 ml of Folin–Ciocalteu reagent and 5 ml of 20%
Na2CO3 solution. The solutions were diluted to 50 ml with distilled
water and mixed thoroughly. The assay mixtures were allowed to
stand for about 30 min at room temperature. The absorbance was
 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044 1039

Table 1
Central Composite Design (CCD) studying the survival of B. longum as a function of pH, citric acid, protein and dietary fibre.

Exp. pH Citric acid Protein Dietary fibre pH Citric acid(g/l) Protein(g/l) Dietary fibre(g/l)
1 1 1 1 1 3.2 2 0 0
2 1 1 1 1 4 2 0 0
3 1 1 1 1 3.2 15 0 0
4 1 1 1 1 4 15 0 0
5 1 1 1 1 3.2 2 10 0
6 1 1 1 1 4 2 10 0
7 1 1 1 1 3.2 15 10 0
8 1 1 1 1 4 15 10 0
9 1 1 1 1 3.2 2 0 8
10 1 1 1 1 4 2 0 8
11 1 1 1 1 3.2 15 0 8
12 1 1 1 1 4 15 0 8
13 1 1 1 1 3.2 2 10 8
14 1 1 1 1 4 2 10 8
15 1 1 1 1 3.2 15 10 8
16 1 1 1 1 4 15 10 8
17 1 0 0 0 3.2 8.5 5 4
18 1 0 0 0 4 8.5 5 4
19 0 1 0 0 3.6 2 5 4
20 0 1 0 0 3.6 15 5 4
21 0 0 1 0 3.6 8.5 0 4
22 0 0 1 0 3.6 8.5 10 4
23 0 0 0 1 3.6 8.5 5 0
24 0 0 0 1 3.6 8.5 5 8
25 0 0 0 0 3.6 8.5 5 4
26 0 0 0 0 3.6 8.5 5 4
27 0 0 0 0 3.6 8.5 5 4
28 0 0 0 0 3.6 8.5 5 4
29 0 0 0 0 3.6 8.5 5 4
30 0 0 0 0 3.6 8.5 5 4
31 0 0 0 0 3.6 8.5 5 4

2.8. Statistical analysis

Table 2
Composition of the model solutions used for the validation of the mathematical
model. The statistical significance of the differences in the values of the
various measurements between week 0 and week 6 of storage, was
Solution pH Citric acid (g/l) Protein (g/l) Dietary fibre (g/l)
analysed by two-tailed t-test using SPSS Statistics (Ver. 18) (SPSS,
1 3.2 6 5 3
USA). A P-value below 0.05 (presented as P < 0.05) was considered
2 3.3 15 2 7
3 3.5 12 9 0 statistically significant.
4 3.7 5 10 1
5 3.9 6 5 3
3. Results

3.1. Cell survival in model solutions

measured at 630 nm; the blank was distilled water. The total phe-
nol concentration was calculated using a calibration curve, pre-
pared with various concentrations of gallic acid.
2.6. Model validation
In order to validate the model, five conditions (Table 2) were
randomly selected within the ranges for pH, citric acid, protein
and dietary fibre that were used to construct the model, and the
cell survival monitored for 6 weeks. The observed log decreases
[log10N0 weeklog10N6 week] were then compared to that predicted
by the model.
2.7. Cell survival in fruit juices
Six commercial fruit juices (orange, grapefruit, blackcurrant,
pineapple, pomegranate and strawberry juice) were purchased
from a local supermarket. The PBS/cell suspension was added into
40 ml of each juice, so that the initial cell concentration was
approximately 1  108 CFU/ml, and the juices stored at 4 °C for
6 weeks. Samples were collected weekly and analysed for pH, cell
concentration and composition.
Table 3 presents the evolution of the cell concentration of B. lon-
gum during storage in the model solutions and the log decrease be-
tween week 0 and week 6 [log10N0 weeklog10N6 week]. In certain
solutions, such as those with very low pH (pH 3.2), the cell concen-
tration started to decrease immediately after the first week of stor-
age, whereas in the majority of the rest of the solutions, the cell
concentration started to decrease after week 4. With the exception
of solutions 4, 8, 12, 16, 18, 20 and 22, the cell concentration in all
solutions was significantly reduced (P < 0.05) after 6 weeks of stor-
age compared to week 0. The greatest reduction in cell concentra-
tion (>0.8 log) viable cells was observed for solutions 1, 5, 9, 13 and
19. However, the cell concentration in most cases was still higher
than 107 CFU/ml (the only exception were solutions 1 and 9). The
general observation from these results was that a high pH as well
as high concentrations of citric acid, protein and dietary fibre
seemed to result to very high cell survival.
3.2. Model development
ANOVA analysis of the stepwise regression model demonstrated
that a second order polynomial model fitted well the data. The R2
1040 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044

Table 3
Survival of B. longum in model solutions during 6 weeks of storage at 4 °C.

Exp. log10N (0 week) log10N (1 week) log10N (2 week) log10N (3 week) log10N (4 week) log10N (5 week) log10N (6 week) log10N0 weeklog10N6 week

1 8.02 ± 0.04 7.15 ± 0.02 7.18 ± 0.20 6.52 ± 0.15 6.33 ± 0.28 6.34 ± 0.15 6.45 ± 0.06 1.57
2 8.03 ± 0.06 7.92 ± 0.07 7.82 ± 0.09 7.72 ± 0.07 7.41 ± 0.22 7.37 ± 0.03 7.00 ± 0.29 1.03
3 8.01 ± 0.03 7.34 ± 0.07 7.44 ± 0.11 7.40 ± 0.02 7.28 ± 0.09 7.24 ± 0.60 7.26 ± 0.28 0.75
4 8.01 ± 0.02 8.04 ± 0.03 7.99 ± 0.01 7.94 ± 0.03 7.97 ± 0.02 7.98 ± 0.00 7.87 ± 0.08 0.14
5 8.07 ± 0.03 7.41 ± 0.08 7.30 ± 0.09 7.44 ± 0.21 7.43 ± 0.03 7.42 ± 0.42 7.21 ± 0.20 0.86
6 8.03 ± 0.01 7.81 ± 0.07 7.92 ± 0.20 7.86 ± 0.04 7.63 ± 0.16 7.72 ± 0.07 7.59 ± 0.12 0.44
7 8.03 ± 0.02 7.73 ± 0.04 7.68 ± 0.08 7.89 ± 0.15 7.56 ± 0.23 7.67 ± 0.08 7.51 ± 0.13 0.51
8 8.03 ± 0.02 8.05 ± 0.04 8.00 ± 0.02 7.95 ± 0.03 7.96 ± 0.01 7.99 ± 0.02 8.02 ± 0.01 0.00
9 8.01 ± 0.08 7.11 ± 0.05 7.04 ± 0.10 6.99 ± 0.14 6.90 ± 0.24 6.96 ± 0.04 6.56 ± 0.30 1.46
10 8.01 ± 0.02 7.80 ± 0.07 7.70 ± 0.03 7.77 ± 0.05 7.66 ± 0.07 7.73 ± 0.05 7.47 ± 0.19 0.55
11 8.09 ± 0.05 7.83 ± 0.16 7.61 ± 0.15 7.59 ± 0.02 7.40 ± 0.13 7.37 ± 0.02 7.48 ± 0.01 0.61
12 8.03 ± 0.01 8.05 ± 0.03 8.01 ± 0.01 7.97 ± 0.03 7.99 ± 0.01 7.97 ± 0.01 8.03 ± 0.02 0.00
13 8.02 ± 0.04 7.73 ± 0.22 7.43 ± 0.08 7.44 ± 0.38 7.32 ± 0.04 7.33 ± 0.28 7.16 ± 0.13 0.86
14 8.02 ± 0.02 7.75 ± 0.11 7.90 ± 0.06 7.88 ± 0.01 7.82 ± 0.05 7.77 ± 0.03 7.64 ± 0.10 0.37
15 8.00 ± 0.02 7.77 ± 0.07 7.68 ± 0.02 7.71 ± 0.02 7.70 ± 0.00 7.80 ± 0.07 7.60 ± 0.15 0.41
16 8.02 ± 0.00 8.03 ± 0.02 8.00 ± 0.01 7.95 ± 0.04 7.99 ± 0.03 7.98 ± 0.01 8.02 ± 0.02 0.01
17 8.03 ± 0.07 7.68 ± 0.04 7.74 ± 0.21 7.58 ± 0.11 7.44 ± 0.10 7.42 ± 0.06 7.62 ± 0.17 0.41
18 8.02 ± 0.07 7.98 ± 0.00 7.99 ± 0.05 7.98 ± 0.01 7.99 ± 0.01 7.99 ± 0.00 7.98 ± 0.02 0.04
19 8.03 ± 0.01 7.90 ± 0.05 7.84 ± 0.21 7.85 ± 0.01 7.54 ± 0.22 7.64 ± 0.07 7.22 ± 0.12 0.81
20 8.03 ± 0.03 7.97 ± 0.02 7.95 ± 0.05 7.86 ± 0.06 7.87 ± 0.01 7.95 ± 0.05 8.01 ± 0.03 0.01
21 8.03 ± 0.05 8.04 ± 0.11 7.88 ± 0.19 7.82 ± 0.04 7.61 ± 0.15 7.64 ± 0.02 7.55 ± 0.08 0.47
22 8.03 ± 0.07 7.89 ± 0.04 7.95 ± 0.12 7.91 ± 0.03 7.78 ± 0.09 7.85 ± 0.12 7.90 ± 0.09 0.13
23 8.01 ± 0.01 7.86 ± 0.07 7.96 ± 0.29 7.88 ± 0.06 7.55 ± 0.23 7.70 ± 0.21 7.47 ± 0.27 0.54
24 8.01 ± 0.03 8.03 ± 0.14 7.84 ± 0.02 7.80 ± 0.02 7.87 ± 0.05 7.88 ± 0.01 7.81 ± 0.06 0.20
25 8.03 ± 0.07 7.71 ± 0.08 7.82 ± 0.00 7.84 ± 0.02 7.82 ± 0.02 7.81 ± 0.01 7.84 ± 0.00 0.19
26 8.01 ± 0.07 7.77 ± 0.00 7.77 ± 0.03 7.80 ± 0.02 7.81 ± 0.01 7.78 ± 0.02 7.81 ± 0.02 0.20
27 8.02 ± 0.01 7.79 ± 0.02 7.82 ± 0.00 7.81 ± 0.01 7.83 ± 0.01 7.76 ± 0.05 7.82 ± 0.03 0.19
28 8.02 ± 0.03 7.76 ± 0.07 7.86 ± 0.06 7.83 ± 0.02 7.78 ± 0.04 7.76 ± 0.02 7.79 ± 0.01 0.23
29 8.02 ± 0.03 7.76 ± 0.04 7.81 ± 0.01 7.85 ± 0.07 7.79 ± 0.08 7.78 ± 0.01 7.82 ± 0.01 0.20
30 8.03 ± 0.04 7.75 ± 0.04 7.80 ± 0.01 7.88 ± 0.06 7.81 ± 0.05 7.78 ± 0.02 7.88 ± 0.05 0.15
31 8.01 ± 0.03 7.75 ± 0.02 7.78 ± 0.01 7.89 ± 0.08 7.79 ± 0.07 7.77 ± 0.01 7.82 ± 0.03 0.19

Standard deviation (±S.D.) calculated with 95% confidence.

values of the model was R2 = 0.97. Moreover, the residual plots did Table 4
Regression coefficients, their 95% confidence intervals and the corresponding P values
not show any trend in the distribution of the residuals around the
for the log decrease [log10N0 weeklog10N6 week]. The regression coefficients are based
zero line, further confirming the goodness of fit. Based on the on the coded values of the controlling factors. The regression coefficients of Eq. (1) are
regression coefficient estimates (Table 4), it was deduced that all based on the uncoded values of the controlling factors.
the controlling factors, i.e., pH, and citric acid, protein and dietary
Term Coefficient Confidence intervals P value
fibre concentrations had a significant effect on the model.
Constant 0.209 (0.158, 0.260) 0.000⁄
The resulting polynomial mathematic model describing the log
pH 0.271 (0.312, 0.231) 0.000⁄
decrease [log10N0 weeklog10N6 week] as a function of the four fac- Citric acid 0.307 (0.348, 0.267) 0.000⁄
tors was: Protein 0.167 (0.208, 0.127) 0.000⁄
Dietary fibre 0.077 (0.118, 0.037) 0.001⁄
½log10 N0week  log10 N6weeks  ¼ 4:051  0:703½pH pH  pH 0.004 (0.110, 0.103) 0.941
Citric acid  Citric acid 0.181 (0.075, 0.288) 0.002⁄
 0:152½citric acid Protein  Protein 0.071 (0.035, 0.178) 0.176
Dietary fibre  Dietary fibre 0.141 (0.035, 0.248) 0.013⁄
 0:151½protein
pH  Citric acid 0.013 (0.030, 0.056) 0.526
þ 0:005½citric acid½citric acid pH  Protein 0.052 (0.009, 0.095) 0.021⁄
pH  Dietary fibre 0.022 (0.065, 0.021) 0.296
þ 0:009½dietary fibre Citric acid  Protein 0.093 (0.050, 0.136) 0.000⁄
Citric acid  Dietary fibre 0.017 (0.026, 0.060) 0.417
 ½dietary fibre þ 0:026½pH
Protein  Dietary fibre 0.043 (0.000, 0.086) 0.049⁄
 ½protein  0:026½pH *
Significant (P < 0.05).
 ½dietary fibre
þ 0:003½citric acid½protein ð1Þ juices, whereas the survival curves are depicted in Fig. 2. The cell
concentration in orange, grapefruit, blackcurrant and pineapple
In order to validate the models, five independent storage runs
juices decreased by less than 0.8 log, with the highest cell survival
were conducted under various conditions. There was a good corre-
being obtained in orange and pineapple juice. The log decrease
lation (R2 = 0.90) between the observed and the predicted values of
[log10N0 weeklog10N6 week] in grapefruit was only 0.5 log, despite
the log decrease [log10N0 weeklog10N6 week] (Fig. 1), which indi-
the fact that its pH was very low (pH 3.21); however, it contained a
cated that the model had a good predictive ability.
high concentration of citric acid (15.3 g/l). The log decrease
[log10N0 weeklog10N6 week] in pomegranate and strawberry juices
3.3. Cell survival in fruit juices was extremely high (8 logs). The cell concentrations in pome-
granate decreased down to undetectable levels within one week,
Table 5 shows the log decrease [log10N0 weeklog10N6 week] in whereas within four weeks in strawberry (Fig. 2). The developed
the cell concentration after 6 weeks of storage in various fruit mathematical model could predict well the cell survival in grape-
 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044 1041

lated juices compared to their respective controls (non inoculated

juices).
Fig. 3 presents the concentrations of protein, dietary fibre and
total phenol in the various juices. The highest concentrations of
dietary fibre were found in strawberry (8 g/l) followed by orange
and grapefruit (in both cases between 3 and 3.5 g/l). Similarly, the
highest concentrations of protein were found in strawberry fol-
lowed by orange and grapefruit. Very high total phenol concentra-
tion was found in pomegranate ( 6.2 g/l) and strawberry (3.3 g/l).
In the rest of the juices the total phenol concentration was less
than 1 g/l.

Fig. 1. Linear regression of the observed versus the predicted values (derived by
from Eq. (1)) for the log decrease [log10N0 weeklog10N6 week] values in five different
solutions (). Solution 1: pH, 3.2 citric acid (CA) 6 g/l, protein 5 g/l, dietary fibre (DF)
3 g/l; Solution 2: pH 3.3, CA 15 g/l, protein 2 g/l, DF 7 g/l; Solution 3: pH 3.5, CA
12 g/l, protein 9 g/l, DF 0 g/l; Solution 4: pH 3.7, CA 5 g/l protein 10 g/l, DF 1 g/l;
Solution 5: pH 3..9, CA 6 g/l, protein 5 g/l, DF 3 g/l.
fruit and pineapple and to an adequate level in orange and black-
currant. It was however unable to predict cell survival in pome-
granate and strawberry juice.
3.4. Chemical changes in the fruit juices during storage
Table 5 presents the chemical changes that took place in the
fruit juices after 6 weeks of storage. The pH and citric acid did
not change significantly (P > 0.05) in the juices. Small amounts of
lactic acid (from 0.1 to 0.8 g/l) and acetic acid (from 0.1 to 1.0 g/
l), were produced, depending on the juice.The concentration of
the total sugar (expressed as the sum of sucrose, glucose and
fructose concentrations) increased in most cases between 1 and
13 g/l. However, smaller decreases were observed for the inocu-
4. Discussion
In order for probiotics to reach the intestine in sufficient num-
bers, they need to be able to survive during processing and storage
and be present in the product at the time of consumption at con-
centrations higher than 107 cells/ml (Champagne & Gardner,
2005; Rivera-Espinoza & Gallardo-Navarro, 2010). In the case of
non-dairy products, such as fruit juices, it has been suggested that
their dietary fibre content affects probiotic storage (Saarela, Alak-
omi, Puhakka, & Matto, 2009), however, up to now there was no
systematic study looking at the likely effect of the chemical com-
position of the juices, and in particular the influence of low pH val-
ues, high levels of organic acids and dietary fibre, and of moderate
to low amounts of total protein.
According to the results from model solutions and the devel-
oped mathematical model it was deduced that all four factors,
i.e., pH, citric acid, protein and dietary fibre significantly influenced
cell survival during 6 weeks of storage (Table 4). Fig. 4 shows the
contour plots of the log decrease [log10N0 weeklog10N6 week] as a
function of two factors, with the other being kept constant at the
middle point values. Taking into account the plots, it can be

Table 5
Chemical and microbiological changes in fruit juices after 6 weeks of storage at 4 °C.

Juice Sample pH Citric acid (g/l) Total sugar (g/l) Lactic acid (g/l) Acetic acid (g/l) [log10N0 weeklog10N6 week]

Predicteda Experimental
Orange Week 0 3.81 ± 0.01 10.1 ± 0.2 80.4 ± 1.4 0 0 0.01 0.35
Week 6 3.80 ± 0.01 10.3 ± 0.4 83.1 ± 1.0 0.8 ± 0.2 0.5 ± 0.2
Controlb 3.78 ± 0.01 10.3 ± 0.1 85.9 ± 0.2 0 0
P value 0.423 0.634 0.139 0.019* 0.034*
Grapefruit Week 0 3.21 ± 0.01 15.3 ± 1.0 68.6 ± 0.1 0 0 0.31 0.49
Week 6 3.19 ± 0.01 14.0 ± 0.3 78.6 ± 0.1 0.8 ± 0.2 1.0 ± 0.3
Control 3.19 ± 0.01 14.1 ± 0.1 81.4 ± 0.8 0 0
P value 0.130 0.212 0.000* 0.028* 0.026*
Blackcurrant Week 0 3.20 ± 0.04 2.8 ± 0.1 84.3 ± 0.1 0 0 1.27 0.73
Week 6 3.12 ± 0.04 2.9 ± 0.1 83.6 ± 0.4 0.7 ± 0.2 0.3 ± 0.1
Control 3.15 ± 0.01 2.8 ± 0.1 86.5 ± 0.2 0 0
P value 0.151 0.585 0.109 0.018* 0.013*
Pineapple Week 0 3.71 ± 0.03 9.0 ± 0.4 89.7 ± 0.5 0 0 0.21 0.36
Week 6 3.63 ± 0.04 9.3 ± 0.1 93.1 ± 0.6 0.5 ± 0.1 0.7 ± 0.2
Control 3.62 ± 0.01 8.8 ± 0.4 96.7 ± 0.5 0 0
P value 0.057 0.414 0.001* 0.008* 0.036*
Pomegranate Week 0 3.33 ± 0.02 21.3 ± 0.5 133.4 ± 3.2 0 0 0.69 8.00
Week 6 3.34 ± 0.01 20.2 ± 0.9 135.8 ± 2.9 0 0
Control 3.33 ± 0.00 19.9 ± 1.2 135.4 ± 0.7 0 0
P value 0.225 0.116 0.007* – –
Strawberry Week 0 3.45 ± 0.01 6.7 ± 0.4 84.0 ± 0.8 0 0 0.38 8.06
Week 6 3.43 ± 0.01 7.0 ± 0.6 83.1 ± 1.4 0 0.3 ± 0.1
Control 3.41 ± 0.01 6.8 ± 0.5 91.4 ± 0.8 0 0
P value 0.057 0.366 0.534 – 0.006*

Standard deviation (±S.D.) and t-test between week 0 and week 6 calculated with 95% confidence.
a
[log10N0 weeklog10N6 week] as predicted by the mathematical model (Eq. (1)).
b
The control was uninoculated juice (after 6 weeks of storage).
*
Significant (P < 0.05).
1042 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044
Fig. 2. Survival curves of B. longum in fruit juices during 6 weeks of storage at 4 °C.
Fig. 3. Concentration of dietary fibre, protein and total phenol in the various fruit
juices.
derived that good survival of B. longum (log decrease <0.4 log) was
obtained in the model solutions were the pH was above 3.7, the cit-
ric acid concentration 8–15 g/l, the protein concentration above
3 g/l and the dietary fibre concentration 3–7 g/l. Moreover, from
Fig. 4 it can be observed that pH, citric acid and protein seemed
to have a bigger effect on the log decrease than dietary fiber did.
It has been well established that the pH is an important factor
influencing the survival of probiotic bacteria in food products
(Champagne & Gardner, 2005). When the cells are present in an
environment of low external pH, the energy consumption, which
is required for maintenance of the intracellular pH, is increased.
As a result, other crucial cellular functions are depressed of ATP,
and the cells cannot survive (Corcoran, Stanton, Fitzerald, & Ross,
2005; Shabala, McMeekin, Budde, & Siegumfeldt, 2006). It is inter-
esting also to note that according to the model coefficients (Table
4) there was a significant (P < 0.05) although small interaction be-
tween pH and the protein levels. However, such interaction was
not observed for citric acid and dietary fibre, indicating that the ef-
fects of the latter two was not influenced by the pH.
Regarding the role of citric acid, up to now, there has been no
information in the literature on the role of organic acids, such as
citric acids in the survival of lactic acid bacteria, in particular food
products during storage. It is expected that organic acids would
have a negative effect on cell survival. However, according to the
results from the model solutions, citric acid was seen to have a sig-
nificant (P < 0.05) positive effect, especially when present at con-
centrations between 8 and 15 g/l (Fig. 4). Possible explanations
for this could be that citric acid was either metabolised by the cells
during storage, although HPLC analysis did not show any signifi-
cant (P < 0.05) decreases (data not shown), or that there was some
kind of buffering effect by citric acid inside the cells.
Regarding the role of protein on cell survival, there is very little
information in the literature. In a study by Dave and Shah (1998) it
was shown that various nitrogen sources, such as whey powder,
whey protein concentrate, acid casein hydrolysate and tryptone
improved the survival of a mixed yogurt starter culture, which in-
cluded Streptococcus thermophilus, Lactobacillus acidophilus, and
bifidobacteria, during refrigerated storage for 35 days. The protec-
tive effect of protein was confirmed in this study, as the model
indicated a significant (P < 0.05) effect (Table 4). Vegetable peptone
was used in the model solutions in order to resemble the type of
protein that is likely to be found in juices, although it must be
emphasised that each type of fruit juice will probably contain dif-
ferent types of proteins, peptides and free amino acids.
Regarding dietary fibre, it has been suggested that it can protect
probiotic cells during processing and storage by a mechanism that
involves the physical immobilization of the cells onto the fibre
(Saarela, Virkajarvi, Nohynek, et al., 2006). Examples of such fibres
include oat b-glucan, the addition of which into yogurt resulted in
the improved survival of B. animalis subsp. lactis during 4 weeks
storage at 4 °C (Vasiljevic, Kealy, & Mishra, 2007). Other research-
ers have reported that oat flour with 20% b-glucan was able to pro-
tect L. rhamnosus during storage in apple juice (Saarela, Virkajarvi,
Nohynek, et al., 2006). The specific effect of pectin, the most com-
mon dietary fibre found in juices, on cell survival has not been
investigated. However, Oliveira et al. (2007) used a casein/pectin
complex as an encapsulating material to improve the survival of
Bifidobacterium lactis and L. acidophilus. The microencapsulated
cells were produced by spray drying and were shown to survive
well during storage for four months at 7–37 °C, and subsequently
in acidic conditions (pH 1 and pH 3). The results from this study
confirm that pectin is likely to protect the bacterial cells during
storage, possibly via a mechanism involving physical protection.
The model predicted the cell survival of the Bifidobacterium
strain in grapefruit and pineapple well and adequately in orange
and blackcurrant. However, it failed to predict the cell survival in
pomegranate and strawberry. This suggests that although the four
controlling factors that were studied in this work, namely pH, citric
acid, protein and dietary fibre, are important factors influencing
the cell survival in fruit juices, additional ones, for example the lev-
els of phenols in the juices, are likely to be involved. These points
will be discussed in the following paragraphs.
Looking in more detail at the composition of the juices, the or-
ange and pineapple juice had similar pH (3.8), citric acid (10
and 9 g/l, respectively) but different (by more than 30%) levels of
protein (6.3 and 3.5 g/l), dietary fibre (3.2 and 2.2 g/l) and total
phenol (0.6 and 1 g/l). The fact that the pH and citric acid levels
were similar and the total phenol levels were low in both cases
is probably the reason that the log decrease [log10N0 weeklog10N6
week] was similar (0.35 log) for both juices. On the hand, it was
shown that the cells survived better in grapefruit compared to
blackcurrant despite the fact the pH and the total phenol concen-
trations were similar (pH 3.2 and total phenol 1.2 g/l). The main
reason was probably the fact that grapefruit had higher citric acid,
protein and dietary fibre consents than blackcurrant, and in partic-
ular citric acid (15.3 g/l vs 2.8 g/l). This further confirms the protec-
tive role of these compounds.
In the case of pomegranate and strawberry juices the pH were
similar (pH 3.3) but there were considerable differences in the
levels of citric acid (21.3 and 6.7 g/l), protein (0.5 and 7.5 g/l), die-
tary fibre (0.6 and 8 g/l) and total phenol (6.2 and 3.3 g/l), respec-
tively. The very high concentration of total phenol in pomegranate
juice probably explains the fact that the cells died within a week of
storage compared to four weeks in the case of strawberry (Fig. 2).
The polyphenols in pomegranate juice were most likely tannins,
which have been shown to inhibit the growth of certain intestinal
bifidobacteria, as well as pathogenic clostridia and Staphyloccocus
aureus (Bialonska, Kasimsetty, Schrader, & Ferreira, 2009). Straw-
berry juice has also been shown to contain high amounts of total
 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044
Fig. 4. Contour plots describing the dependence of the log decrease [log10N0 weeklog10N6 week] as a function of two controlling factors. A): pH and citric acid (CA), protein 5 g/
l, dietary fibre (DF) 4 g/l; B): pH and protein, CA 8.5 g/l, DF 4 g/l; C): pH and DA, CA 8.5 g/l, protein 5 g/l; D): CA and protein, pH 3.6, DF 4 g/l.
phenols, including both hydrolyzable and condensed tannins
(Herken & Guzel, 2010). Simlarily to pomegranate, this could be
the main reason for the fact that the cells died within 4 weeks of
storage in strawberry juice, despite the fact that as shown in this
study (Fig. 3) and another study (Ramulu & Rao, 2003) strawberry
juice contains high amounts of fibre, both soluble and insoluble.
Regarding the chemical changes taking place during storage, al-
most in all cases, the total sugar (sucrose, glucose and fructose)
levels increased significantly after storage. This might be due to
the hydrolysis of polysaccharides in the juices through the action
of pectinases (Mohammad, Andres, & Klaus, 2010). It was notewor-
thy the fact that the sugar content of the controls (uninoculated
juices) increased more than the inoculated juices after 6 weeks of
storage, indicating that some of the free sugar was consumed by
the cells. This hypothesis was backed up by the significant
(P < 0.05) increases in the concentrations of lactic and acetic acid,
the total amount of which in some cases exceeded 1 g/l. The pH
of the juices, however, did not change significantly, indicating that
the buffering capacities of the juices were high.
5. Conclusions
Based on the storage experiments of B. longum in model solu-
tions and the mathematical model that was developed it was
shown that high levels of pH, citric acid, protein and dietary fibre
enhanced the cell survival during refrigerated storage. The
mathematical model was able to predict adequately cell survival
in orange, grapefruit, blackcurrant and pineapple juices. However,
1043
the model failed to predict cell survival in pomegranate and
strawberry, in which the cell viability declined rapidly, most
likely due to the very high levels of phenolic compounds in these
two juices.
References
AOAC, 1997. Official Methods of Analysis AOAC. Association of Official Analytical
Chemists, 16th ed. Washington, DC.
Bialonska, D., Kasimsetty, S. G., Schrader, K., & Ferreira, D. (2009). The effect of
pomegranate (Punica granatum L.) by-products and ellagitannins on the growth
of human gut bacteria. Journal of Agricultural and Food Chemistry, 57,
8344–8349.
Champagne, C. P., & Gardner, N. J. (2005). Challenges in the addition of probiotic
cultures to foods. Critical Reviews in Food Science and Nutrition, 45, 61–84.
Champagne, C. P., Raymond, Y., & Gagnon, R. (2008). Viability of Lactobacillus
rhamnosus R0011 in an apple-based fruit juice under simulated storage
conditions at the consumer level. Food Microbiology and Safety, 73, 221–226.
Clark, P. A., & Martin, J. H. (1994). Selection of bifidobacteria for use as dietary
adjuncts in cultured dairy foods: III—Tolerance to simulated bile concentrations
of human small intestines. Cultured Dairy Products Journal, 29, 18–21.
Clark, P. A., Cotton, L. N., & Martin, J. H. (1993). Selection of bifidobacteria for use as
dietary adjuncts in cultured dairy foods: II—Tolerance to simulated pH of
human stomachs. Cultured Dairy Products Journal, 28, 11–14.
Corcoran, B. M., Stanton, C., Fitzerald, G. F., & Ross, R. P. (2005). Survival of probiotic
lactobacilli in acidic environments is enhanced in the presence of metabolizable
sugars. Applied and Environmental Microbiology, 71, 3060–3067.
Dave, R. I., & Shah, N. P. (1998). Ingredient supplementation effects on viability of
probiotic bacteria in yogurt. Journal of Dairy Science, 81, 2804–2816.
Herken, E. N., & Guzel, S. (2010). Total antioxidant capacity and total phenol
contents of selected commercial fruit juices in Turkey. International Journal of
Food Properties, 10, 1373–1379.
Homayouni, A., Azizi, A., Ehsani, M. R., Yarmand, M. S., & Razavi, S. H. (2008). Effect
of microencapsulation and resistant starch on the probiotic survival and
sensory properties of synbiotic ice cream. Food Chemistry, 111, 50–55.
1044 S. Nualkaekul et al. / Food Chemistry 129 (2011) 1037–1044
Julio, A., Patricia, A., Fabienne, B., Monique, Z., Marie-Christine, C. V., & Marie-Jose, B.
(2010). Proteomic comparison of the cytosolic proteins of three Bifidobacterium
longum human isolates and B. longum NCC2705. BMC Microbiology, 10, 29.
Krasaekoopt, W., Bhandari, B., & Deeth, H. (2003). Evaluation of encapsulation
techniques of probiotics for yoghurt. International Dairy Journal, 13, 3–13.
Leahy, S. C., Higgins, D. G., Fitzgerald, G. F., & Sinderen, D. V. (2005). Getting better
with bifidobacteria. Journal of Applied Microbiology, 98, 1303–1315.
Luckow, T., & Delahunty, C. (2004). Which juice is healthier? A consumer study of
probiotic non-dairy juice drinks. Food Quality and Preference, 15, 751–759.
Lukanin, E. S., Gunko, S. M., Bryk, M. T., & Nigmatullin, R. R. (2003). The effect of
content of apple juice biopolymers on the concentration by membrane
distillation. Journal of Food Engineering, 60, 275–280.
Medina, M., Izquierdo, E., Ennahar, S., & Sanz, Y. (2007). Differential
immunomodulatory properties of Bifidobacterium longum strains: relevance to
probiotic selection and clinical applications. Clinical and Experimental
Immunology, 150, 531–538.
Miyauchi, E., Xiao, J.-z., & Tanabe, S. (2010). Interaction of bifidobacteria with the
Intestinal Mucosa with a Focus on Immunomodulating Effects, Caister
Academic Press.
Mohammad, G. A., Andres, D. H., & Klaus, D. K. (2010). Isolation of polysaccharides
from pineapple fruit pulp and their enzymatic liquefaction. International Food
Research Journal, 17, 193–203.
Oliveira, A. C., Moretti, T. S., Boschini, C., Baliero, J. C. C., Freitas, O., & Favaro-
Trindade, C. S. (2007). Stability of microencapsulated B. lactis (Bl 01) and L.
acidophilus (LAC 4) by complex coacervation followed by spray drying. Journal of
Microencapsulation, 24, 685–693.
Prado, F. C., Parada, J. L., Pandey, A., & Soccol, C. R. (2008). Trends in non-dairy
probiotic beverages. Food Research International, 41, 111–123.
Ramulu, P., & Rao, P. U. (2003). Total, insoluble and soluble dietary fiber contents of
Indian fruits. Journal of Food Composition and Analysis, 16, 677–685.
Rivera-Espinoza, Y., & Gallardo-Navarro, Y. (2010). Non-dairy probiotic products.
Food Microbiology, 27, 1–11.
Rosaria D’Aimmo, M., Modesto, M., & Biavati, B. (2007). Antibiotic resistance of
lactic acid bacteria and Bifidobacterium spp. isolated from dairy and
pharmaceutical products. International Journal of Food Microbiology, 115, 35–42.
Saarela, M., Alakomi, H. L., Puhakka, A., & Matto, J. (2009). Effect of the fermentation
pH on the storage stability of Lactobacillus rhamnosus preparations and
suitability of in vitro analyses of cell physiological functions to predict it.
Journal of Applied Microbiology, 106, 1204–1212.
Saarela, M., Virkajarvi, I., Alakomi, H. L., Mattila, P. S., & Matto, J. (2006a). Stability
and functionality of freeze-dried probiotic Bifidobacterium cells during storage
in juice and milk. International Dairy Journal, 16, 1477–1482.
Saarela, M., Virkajarvi, I., Nohynek, L., Vaari, A., & Matto, J. (2006b). Fibres as carriers
for Lactobacillus rhamnosus during freeze-drying and storage in apple juice and
chocolate-coated breakfast cereals. International Journal of Food Microbiology,
112, 171–178.
Sanchez, B., Champomier-Verges, M. C., Collado, M. D., Anglade, P., Baraige, F., Sanz,
Y., et al. (2007). Low-pH adaptation and the acid tolerance response of
Bifidobacterium longum biotype longum. Applied and Environmental
Microbiology, 73, 6450–6459.
Sanchez, B., de los Reyes-Gavilan, C. G., Margolles, A., & Gueimonde, M. (2009).
Probiotic fermented milks: Present and future. International Journal of Dairy
Technology, 62, 472–483.
Shabala, L., McMeekin, T., Budde, B. B., & Siegumfeldt, H. (2006). Listeria innocua and
Lactobacillus delbrueckii subsp bulgaricus employ different strategies to cope
with acid stress. International Journal of Food Microbiology, 110, 1–7.
Shah, N. P. (2000). Probiotic bacteria: selective enumeration and survival in dairy
foods. Journal of Dairy Science, 83, 894–907.
Vasiljevic, T., Kealy, T., & Mishra, V. K. (2007). Effects of b-glucan addition to a
probiotic containing yogurt. Journal of Food Science, 72, 405–411.
Williams, M. D., Ha, C. Y., & Ciorba, M. A. (2010). Probiotics as therapy in
gastroenterology a study of physician opinions and recommendations. Journal
of Clinical Gastroenterology, 44, 631–636.