Optimization of the Viability of Probiotics in a Fermented Milk Drink by the Response Surface Method

Growth promoters were added to skim milk to retain the viability of Lactobacillus acidophilus and Bifidobacterium longum to help the product meet the “therapeutic minimum” at the time of consumption. The experiments were divided into two parts. The first part of the study used chicory inulin, isomalto-oligosaccharides and sucrose to investigate the effects of sugars on the activity of L. acidophilus and B. longum. The results indicated that the addition of isomalto-oligosaccharides stimulated growth of L. acidophilus and B. longum, resulting in a higher level of the probiotics after one month storage and yielded better β-galactosidase activity during fermentation. The second part studied the effects of three growth promoters on the viability of the probiotic cultures and the response surface method was employed to find the optimal ratio for addition of the growth promoters. The optimal ratio for added calcium gluconate, sodium gluconate and N-acetylglucosamine in fermented milk drinks were established. The response surface method proved to be a very effective way of optimizing the activity of probiotic cultures when developing a new fermented milk drink. (Asian-Aust. J. Anim. Sci. 2004. Vol 17, No. 5 : 705-711)


INTRODUCTION
Recently, a number of novel fermented dairy products have been developed and are being marketed under the concept of probiotic products.Due to their increasing popularity among consumers, the number of probiotic dairy products on the Taiwan, Japan, Scandinavia, Netherlands, France and United States markets have increased tremendously during the last few years.Several health benefits have been claimed to be associated with the consumption of fermented milk products.Although yogurt microflora (Streptococcus thermophilus and Lactobacillus delbrueckii ssp.bulgaricus) have been found to be beneficial for human health and nutrition, recent emphasis have focused on development of fermented milk products with added Lactobacillus acidophilus and bifidobacteria (known as probiotic or AB-cultures).This is due to the ability of these cultures to tolerate acid and bile, which enables them to implant in intestinal tract.The benefits derived by the consumption of AB products are well documented and have been reviewed by several workers (Kanbe, 1992;Scheinbach, 1998;Ziemer and Gibson, 1998;Lourens-Hattingh and Viljoen, 2001;Liu et al., 2002).
L. acidophilus and bifidobacteria have to retain viability and activity in the food carrier to meet the suggested 'therapeutic minimum' at the time of consumption (Playne, 1994).Several factors have been claimed to affect the viability of probiotic cultures in fermented milk products.
The culture conditions, the chemical composition of the fermentation medium (e.g.carbohydrate source), the final acidity, growth promoter and inhibitors, incubation temperature, fermentation time and storage temperature have all been identified as effecting on the stability of probiotic cultures during storage (Hamman and Marth, 1983;Kneifel et al., 1993;Lankaputhra and Shah, 1994;Young and Nelson, 1978).
Bifidobacteria prefer an anaerobic environment and a low redox potential (Klaver et al., 1993), as well as exhibit a weak growth in milk.Therefore, the addition of bifidogenic factors to achieve the desired levels of growth is necessary.The purpose of this study was to study the addition of growth promoters to milk to optimize, by the response surface method (RSM), the activity and viability of the probiotics.

Preparation of fermented milk drinks
The milk used in the experiments was reconstituted from high heat treated whole milk powder (12% total solids, Anchor, New Zealand), heat treated at 85°C for 30 min.A series of pretests were conducted and the results indicated that milk inoculated with 1% of L. acidophilus and 2% of B.longum and fermented for 10 h would produce the highest viability of probiotic cultures.Therefore, experimental samples were prepared using skim milk inoculated with 1% of L. acidophilus and 2% of B. longum.After fermentation for 10 h at 37°C (final pH about 5.75), the samples were then mixed with vegetable juice (V8 100% Vegetable Juice, Campbell Soup Company, NJ, USA) in the ratio of 1:1 to produce a new fermented milk drink (final pH about 5.4).

Cultures and medium performance
Pure lyophilized cultures of B. longum (CCRC 14605) and L. acidophilus (CCRC 14079) were purchased from the Culture Collection and Research Center, Taiwan, ROC.Lactobacilli MRS (deMan, Rogosa and Sharp) was used as the selective media for L. acidophilus and Lithium propionate MRS agar (LP-MRS) was used as the selective media for B. longum (Lapierre et al., 1992).

Activity determination
For determination of the viabilities of the probiotic cultures, the populations of B. longum and L. acidophilus were measured as colony-forming units (CFU) and by the amount of β-galactosidase that was produced by them.
The suitability of the media was tested by plating decimal dilutions of the probiotic cultures.Thus, a sample of 1 g of each pure lyophilized cultures was decimally diluted into sterile peptone water (0.1%) and 0.1 ml aliquot dilutions were then plated onto the different media, in triplicate.Plates of MRS agar were incubated aerobically for 72 h at 37°C to inhibit bifidobacteria.Plates of LP-MRS agar were incubated anaerobically (72 h at 37°C, GasPak System-Oxoid, Basingstoke, Hampshire, England) to count the number of B. longum.The population in colony-forming units (CFU) and the characteristics of the colonies were recorded for each medium.
β-galactosidase activity was measured by determining the rate of hydrolysis of ο-nitrophenol-β-galactopyranoside as described by Yu et al. (1987). 1 ml of 15 mM οnitrophenol-β-galactopyranoside solution prepared in 0.02 M phosphate buffer (pH 7.0) was added to samples, vortexed for 1 s, and incubated for 5 to 20 min at 37°C until a faint yellow tint was observed.The reaction was terminated by the addition of 1 ml of 0.5 M Na 2 CO 3 .Hydrolysis of this substrate resulted in the release of οnitrophenol, a highly chromogenic compound that was detected spectrophotometrically at 420 nm.The quantities of ο-nitrophenol in samples were determined by use of a standard curve.One unit of enzyme activity released 1 µM of ο-nitrophenol/min.

Experiment design
Six growth promoters were used to investe the effects on the activity of L. acidophilus and B. longum.The sugars are the sweeteners that could affect the taste of the product, so the experiments were divided into two parts.In first part selected chicory inulin, isomalto-oligosaccharides and sucrose were used.The second part studied the effects of three growth promoters on the viability of the probiotics and the response surface method (RSM) was employ to find the optimal ratio for addition of the growth promoters.

Effects of sugars
Three different samples for experiments were created by adding 4% chicory inulin, isomalto-oligosaccharides and sucrose to milk, respectively, and then following the same procedures described in the section of "Preparation of fermented milk drinks".The viability of probiotic cultures were determined by measuring the populations of B. longum and L. acidophilus as well as β-galactosidase activities on those three samples and one control (without any sugar added).Finally, the storage tests were conducted by holding samples at 5°C for a period of 4 weeks.During the storage period, viable counts of probiotic cultures were determined.
The experiments were repeated three times and the results were analyzed using Analysis of variance (ANOVA) from the SAS software package (SAS Institute 1990), with least-significant-difference (LSD) method as a multiplecomparison technique for significance to detect differences among means (Montgomery, 1991;Kleinbaum et al., 1998).
Center points : Data from the center points provides estimates of pure error and estimates of curvature.
For our studies with three factors, the total number of runs is equal to 20, 2 3 (two-level factorial)+2×3 (axial points)+6(center points)=20.The coded and uncoded factors, and their respective levels are presented in Table 1.

Response surface method (RSM)
The RSM procedure of the Design-Expert  software package (Stat-Ease, Inc., USA, 2000) was used to fit the experimental data to polynomial equations of order 1 through 3 to obtain coefficients.The following relationship achieved this.
where Y 1 , Y 2 and Y 3 are the observed numbers of L. acidophilus, B. longum and β-galactosidase activity, respectively.ƒ 1 , ƒ 2 and ƒ 3 represent the modeled response surfaces.X 1 , X 2 and X 3 , defined as natural (uncoded) factors, are the concentrations of N-acetylglucosamine, Cagluconate and Na-gluconate, respectively.ε 1 , ε 2 , ε 3 are the errors in each model.With RSM, it is convenient to transform the natural factors to coded factors ξ 1 , ξ 2 and ξ 3 , which are defined as dimensionless, with mean zero and the same spread or standard deviation: An optimization of viability of AB cultures in terms of calcium gluconate, sodium gluconate and Nacetylglucosamine concentrations was calculated using the predictive equation (eq.1).

Verification of model
Five optimal producing conditions were suggested by Design-Expert software (Stat-Ease, Inc., USA, 2000) and were verified by additional independent experiments.The experimental samples were prepared using skim milk mixed with 4% IMO and the prescribed growth promoters, then following the same procedures described in the section of "Preparation of fermented milk drinks".

Statistical analysis
The models selection was complicated by the sequential model sum of squares and by lack of fit tests from the Design-Expert  software package.The model verification experiments were repeated three times and the results were analyzed using ANOVA from the SAS software package, with least-significant-difference (LSD) method as a multiple-comparison technique for significance to detect differences among means.

Effect of sugars on the activities of AB cultures
Sugars are not only sweeteners but also used as growth promoters for the probiotics.Chicory inulin, isomaltooligosaccharides (IMO) and sucrose were selected to investigate the effects of sugars on the activity of L. acidophilus and B. longum.Among the various sugars tested, supplementation of IMO yielded the highest (p<0.05)viable count of L. acidophilus (7.92 CFU/ml) after 10 h of fermentation (Figure 1(a)), whereas in case of B.  longum, a significantly higher (p<0.05)viable count was noted in the IMO and chicory inulin supplemented fermented milk than in the sucrose and control (milk without the added supplement).The β-galactosidase activity (Figure 1(c)) was greater (p<0.05) in IMO supplemented milk, while the β-galactosidase activity was lower (p<0.05) in sucrose supplemented milk at the end of fermentation.
Among the sugars tested, IMO was the best (p<0.05)growth promoting substance for the AB culture and showed a higher population and β-galactosidase activity than other sugars (Figure 1).IMO has been shown to promote the growth of bifidobacteria (Kohmoto et al., 1988).Fooks et al. (1999) studied probiotics in an in vitro model of the human gut and showed that IMO stimulate lactic microflora as well as elevating production of butyrate.Chen et al. (2003) also confirmed that IMO improve the growth rate of probiotics.Sucrose is the most popular sweetener for dairy products.In our studies, 4% sucrose did not improve the growth of AB cultures and even inhibit the enzyme activities during fermentation.Vinderola et al. (2002) studied the influence of sugars associated with fermented dairy products and proved that some strains of bifidobacteria were inhibited by 15 or 20% of sucrose.
During storage, chicory inulin maintained the population of L. acidophilus at up to 7.27 log CFU/ml which was the same (p>0.05)as the IMO (Figure 2(a)), however inulin decreased (p<0.05) the viability of B. longum to 6.01 log CFU/ml during storage (Figure 2(b)).This could be explained by the length of polymers and enzyme activity.Chicory inulin is made up of fructoseoligosaccharides (FOS) and contains oligosaccharides with a size of 2-30 degree of polymeration.Marx et al. (2000) studied metabolization of FOS and showed that inulin polymers of molecular mass greater than ca.4500 were not utilized by bifidobacteria, because hydrolysis of long-chain FOS is inhibited by steric hindrance at the active site of the hydrolytic enzymes.During fermentation, B. longum uses short chain FOS as a carbohydrate source for growth and the growth slowed or ceased once the short-chain oligosaccharides were consumed.
The results indicated that adding IMO stimulated growth of L. acidophilus and B. longum, allowed retention  of a higher level of the probiotics during a month storage, and yielded better β-galactosidase activity during fermentation.

Effect of the prebiotics
Fitting the model : Three growth promoters (calcium gluconate, sodium gluconate and N-acetylglucosamine) were mixed with milk in order to improve the activity of L. acidophilus and B. longum.The responses as linear, quadratic and cubic functions of the factors were tested for adequacy and fitness by analysis of variance.Table 2(a) examines the probability (p>F) to see if it falls below 0.05.The highest order polynomial that is significant is selected.The "Lack of Fit Tests" (Table 2(b)) compares the residual error to the pure error from replicated design points.If there is a significant lack of fit, as indicated by a low probability value (p>F), the response predictor should be discarded.The model with insignificant lack-of-fit is selected.The model analysis results (Table 2(a)) showed that the quadratic models for all three responses that were significant (p<0.05).The lack of fit tests (Table 2(b)) indicated that the quadratic models appeared to be the most accurate for B. longum andβ-galactosidase activity, with no significant lack of fit (p>0.05).However, the quadratic models had a significant lack of fit for L. acidophilus.The second order polynomial equations (3) were fitted to the experimental data using the Design Expert procedure: where f 1 , f 2 and f 3 are the three responses and β 0 , β i , β ii and β ij are constant coefficients and X 1 , X 2 and X 3 are the uncoded independent factors.The equations of prediction for three responses were found as the following: • L. acidophilus : f 1 ＝7.56-0.38X 1 -(2.000E-03)X 2 -0.24X 3 +0.87X 1 2 -0.12 X 2 2 +0.058X 3 2 -0.16 X 1 X 2 +0.24 X 1 X 3 +0.040X 2 X 3 • B. longum: f 2 ＝7.68+1.24X 1 +0.39 X 2 -0.61 X 3 -0.56X 1 2 -0.30X 2 2 +0.45 X 3 2 -0.70 X 1 X 2 +0.16 X 1 X 3 +0.030X 2 X 3 • β-galactosidase: f 3 ＝269.32+58.25 X 1 +26.45X 2 +8.28 X 3 -14.00X 1 2 -56.00 X 2 2 -50.10 X 3 2 -55.00 X 1 X 2 -10.00 X 1 X 3 +39.10X 2 X 3

CONCLUSION
In the current study, the sugar tests indicated that adding IMO stimulated growth of L. acidophilus and B. longum, allowed retention of a higher level of the probiotics during a month storage.The response surface method proved to be a very effective way of optimizing the activity of probiotic cultures when developing a new fermented milk drink.

Figure 1 .
Figure 1.The effects of the addition of sugars on the activity of L. acidophilus and B. longum in fermented milk drinks during fermentation.

Figure 2 .
Figure 2. The effects of addition of sugars on the activity of L. acidophilus in fermented milk drinks during storage.

Table 1 .
Process factors and their levels for the three factors-three levels response surface design

Table 2 .
Analysis of variance for the factors as linear, quadratic and cubic terms and their interactions in a response surface model Lack of fit tests-want the selected model to have insignificant lack-of-fit.
* Significant at 5% level; ** Significant at 1% level.a Model analysis-select the highest order polynomial where the additional terms are significant.b SS-Sum of squares.c DF-Degree of freedoms.d

Table 3 .
Analysis of variance showing the significance of the response variables p>F Independent factors df L. acidophilus (log CFU/ml) * Significant at 5% level, ** Significant at 1% level.