Journal of Dairy Research (2014) 81 59–64. doi:10.1017/S0022029913000733
© Proprietors of Journal of Dairy Research 2013
59
Optimisation of medium composition for probiotic biomass production using response surface methodology Masumeh Anvari1*, Gholam Khayati2 and Shora Rostami3 1
Department of Microbiology, Faculty of Sciences, Islamic Azad University, Rasht Branch, P.O. Box 41335-3516, Rasht, Iran Department of Chemical Engineering, Faculty of Eng., Guilan University, P. O. Box 3756-41635, Rasht, Iran 3 Department of Microbiology, Sciences and Research Branch, Islamic Azad University, Guilan, Iran 2
Received 25 April 2013; accepted for publication 16 October 2013
This study was aimed to optimise lactose, inulin and yeast extract concentration and also culture pH for maximising the growth of a probiotic bacterium, Bifidobacterium animalis subsp. lactis in apple juice and to assess the effects of these factors by using response surface methodology. A second-order central composite design was applied to evaluate the effects of these independent variables on growth of the microorganism. A polynomial regression model with cubic and quadratic terms was used for analysis of the experimental data. It was found that the effects involving inulin, yeast extract and pH on growth of the bacterium were significant, and the strongest effect was given by the yeast extract concentration. Estimated optimum conditions of the factors on the bacterial growth are as follows: lactose concentration = 9·5 g/l; inulin concentration = 38·5 mg/l; yeast extract concentration = 9·6 g/l and initial pH = 6·2. Keywords: Bifidobacterium animalis subsp. Lactis, probiotic, response surface methodology, apple juice.
Probiotics are defined as viable microorganisms which, upon ingestion in sufficient amounts, exert health benefits to the host beyond inherent basic nutrition (Guarner & Schaafsma, 1998). They have been used for the treatment of various types of diarrhoea (Szymanski et al. 2006), urogenital infections (Reid et al. 2003), and gastrointestinal diseases such as Crohn’s disease (Bousvaros et al. 2005) and pouchitis (Kuehbacher et al. 2006), although there is still no consensus about their effectiveness (Lin, 2003). Probiotic bacteria, often belonging to the Lactobacillus and Bifidobacterium genera (Weinbreck et al. 2010). Lactic acid bacteria are commercialised mainly as food supplements with dairy products being the most often used vehicle (Heller, 2001; Lourens-Hattingh & Viljoen, 2001). However, recent studies have suggested fruit juices as an alternative vehicle for the incorporation of probiotics (Mousavi et al. 2011; Pereira et al. 2011; Fonteles et al. 2012; Costa et al. 2013). Fruit juices are rich in nutrients and do not contain starter cultures that compete for nutrients with probiotics. Furthermore, fruit juices contain high amounts of sugars, which could encourage probiotic growth (Ding & Shah, 2008). The optimal growth of probiotic bacteria is affected by fermentation conditions such as pH, temperature, medium
*Corresponding author; e-mail: [email protected]
formulation and the others. Study of the individual and interactive effects of these factors will help in efforts to optimise biomass production of the probiotic microorganism (Du Toit et al. 2011). According to Oliveira & Damin (2003), to evaluate the growth of lactic acid bacteria, it is necessary to know the substrates applied for the microbial growth, as well as, the optimal temperature and pH values because these factors are the most important for the microbial development. Response surface methodology (RSM) is a useful model for studying the effect of several factors influencing the responses by varying them simultaneously and carrying out a limited number of experiments. In addition, response surface methodology is an efficient strategic experimental tool by which the optimal conditions of a multivariable system may be determined (Khayati & Kiyani, 2012; Khayati, 2013). Lactobacilli are also extensively used as probiotics, but no information is available on the growth of the species Bifidobacterium in apple juice. The objective of this study was to determine the suitability of apple juice as a raw material for production of probiotic by Bifidobacterium spp.. Thus, the use of apple juice as substrate to produce a probiotic was studied herein. So in this paper, the growth of probiotic Bifidobacterium animalis subsp. lactis in apple juice (as a basement medium) with a function of four affecting parameters including lactose (g/l), inulin (mg/l) and yeast
60
M Anvari and others
extract concentration (g/l) and initial pH was studied by RSM. Then these bacterial strains can be successfully manufactured and incorporated into highly acceptable dairy food products where they can retain their viability and functionality. (Alander & Mattila-Sandholm, 2000).
Materials and methods Microorganisms and media Bifidobacterium animalis subsp. lactis PTCC 1736 was prepared from the Iranian Research Organization for Science and Technology (IROST). All chemicals used were of analytical grade. The strain was maintained in MRS agar medium containing 1% lithium chloride, 0·3% sodium propionate and 0·5% L-cysteine at 4 °C. Subcultures (1% inoculum, incubated for 10 h at 37 °C in anaerobic jars) were prepared immediately before the culture was used experimentally. Fermentation conditions In the design orthogonal array, each row consists of a number of conditions depending on the levels assigned to each factor. Submerged fermentation experiments were carried out in cotton plugged 100 ml Erlenmeyer flasks containing 30 ml apple juice in different conditions. pH adjusted by adding 0·02 N HCl prior to sterilisation (15 min, 121 °C). Lactose and yeast extract were sterilised separately. Cultures were inoculated with mentioned above pre-cultures (approximate concentration of 103 CFU/ml of the probiotic strain), and incubated for 36 h at 37 °C in an anaerobic atmosphere (85% N2/10% H2/5% CO2). Microbiological analysis The number of viable cells was determined as colony forming units (CFU). One millilitre of sample was diluted with 9 ml 0·1% sterile peptonated water. Serial decimal dilutions of each sample were plated in triplicate onto MRSLP agar; afterwards, bacteria were counted applying the pour plate technique (Kodaka et al. 2005). Plates were incubated at 37 °C for 72 h under anaerobic conditions (Gas-Pak plus system). CFU were enumerated in plates containing 30–300 colonies. The selectivity of culture media was confirmed by microscopic examination of cells in the colonies. Results were expressed as log10 CFU/ml. Experimental design and statistical analysis In this experiment, the response was the growth of Bifido. animalis subsp. lactis PTCC 1736, represented by log10 (number of viable cells/ml; CFU/ml). The effect of four independent variables: lactose (x1), inulin (x2) and yeast extraction concentration (x3) also initial pH (x4) on the response variable (Y, the log10 (number of viable cells/ml)) was evaluated using central composite design (CCD)
Table 1. Independent variables and their coded and actual levels Actual values of coded levels Independent variables (factors) Lactose concentration (g/l) Inulin concentration (mg/l) Yeast extract concentration (g/l) Initial pH
Symbol α
1
0
+1
+ α†
x1
0
5
10
15
20
x2
5
15
25
35
45
x3
2
4
6
8
10
x4
5·5
6·0
6·5
7·0
7·5
† α = 2·0 (star point for orthogonal CCD for the case of 4 independent variables)
Table 2. Central composite experimental design and responses Independent variables† Run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
x1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0
x2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0
x3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 2 2 0 0 0 0 0 0 0 0 0
x4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0
Response (Y )‡ 4·354 5·123 5·636 6·388 8·691 9·134 10·268 9·813 3·508 3·782 3·172 4·131 8·227 7·308 8·662 8·986 6·195 7·044 7·745 7·483 3·151 9·236 8·452 4·041 8·109 8·562 8·770 8·461 8·398 8·801 8·369
† x1: lactose concentration (g/l); x2: inulin concentration (mg/l); x3: yeast extract concentration (g/l); x4: initial pH ‡ Y: log10 (CFU/ml)
(Table 1). The five coded levels of each variable were incorporated into the design and were analysed in 31 experimental trails (Table 2). The central point of the design was repeated seven times to calculate the reproducibility of the method (Montgomery, 2001). For each experimental trail
Medium for probiotic biomass production using RSM
61
a
a
10
10
Log (CFU/ml)
Log (CFU/ml)
5
0
0 0
15
9 6 3 45 yeast extract con. (g/l)
30
inulin con. (mg/l)
0
30
7.5
10
7.0
8
initial pH
7 6 8
5
45 5
6
7
8
initial pH
5
10
9
yeast extract con. (g/l)
15
inulin con. (mg/l)
b
b
5
6
7
6.5
6
9
6.0
4 3
7
4
2 5
10
15
20
25
30
35
40
45
5.5 5
8
10
15
20
25
30
35
40
45
inulin con. (mg/l)
inulin con. (mg/l)
Fig. 1. Response surface plot (a) and its contour plot (b) for the effects of inulin concentration and yeast extract concentration on log10 (CFU/ml) at its centre level.
Fig. 2. Response surface plot (a) and its contour plot (b) for the effects of inulin concentration and initial pH on log10 (CFU/ml) at its centre level.
of the independent variables in the experimental design, the dependent parameter (the log10 (number of viable cells/ml)) was determined. The effect of these independent variables x1, x2, x3 and x4 on the response Y was investigated using the second-order polynomial regression equation. This equation, derived using RSM for the evaluation of the response variable, is as follows:
statistical calculations to generate response surface and contour maps from the regression models. The analysis of data and the optimising process were generated using Minitab statistical software version 15.
Y ¼ β0 þ
k X i¼1
βi xi þ
k X i¼1
βii x2 þ
k X
βij xi xj
ð1Þ
i=j
where β0 is defined as the constant, βi the linear coefficient, βii the quadratic coefficient and βij the interaction coefficient. xi and xj are the independent variables while k equals to the number of the tested factors (k = 4). The analysis of variance (ANOVA) tables were generated and the effect and regression coefficients of individual linear, quadratic and interaction terms were determined. The significances of all terms in the polynomial were judged statistically by computing the P < 0·05. The regression coefficients were then used to make
Results and discussion Effects of independent variables on responses Interest in the incorporation of Bifidobacteria into fermented products has developed considerably over recent years. In many studies reporting human health benefits associated with the consumption of these bacteria so they have received attention as probiotic (Doleyres & Lacroix, 2005). Response surface was used to illustrate the effect of lactose, inulin and yeast extract concentration and initial pH on the growth of Bifido. animalis subsp. lactis in apple juice. The response surfaces and contour plots of the bacterial growth conditions are presented in Figs. 1–3. Also, the effect of inulin and yeast extraction concentrations on the response is shown in Fig. 1.
M Anvari and others
62
a
Table 3. Significance of regression coefficients of the fitted secondorder polynomial model for response (Y ) 10
5
0
3
6
9
5
yeast extract con. (g/l)
b
7.5
6
7
8
initial pH
6
initial pH
7.0
10
6.5
SE
β0 Linear β1 β2 β3 β4
8·49570
0·2283
< 0·001
0·16014 0·26682 1·96532 0·85218
0·1233 0·1233 0·1233 0·1233
0·212 0·046 < 0·001 < 0·001
Quadratic β11 β22 β33 β44
0·46394 0·21532 0·57043 0·55723
0·1129 0·1129 0·1129 0·1129
0·001 0·075 < 0·001 < 0·001
Interaction β12 β13 β14 β23 β23 β34
0·06326 0·21009 0·05449 0·11321 0·16738 0·13655
0·1510 0·1510 0·1510 0·1510 0·1510 0·1510
0·681 0·183 0·723 0·464 0·284 0·379
Coef.
P-value
Table 4. Analysis of variance (ANOVA) of the regression parameters for the response surface model
6.0
4
5.5 2
Regression Coef.
8
3
4
5
6
7
8
9
10
yeast extract con. (g/l)
Fig. 3. Response surface plot (a) and its contour plot (b) for the effects of yeast extract concentration and initial pH on log10 (CFU/ ml) at its centre level.
The growth of microorganism was increased with the yeast extract concentration increasing to a certain value, thereafter was constant. It is due to the low content of free amino acids and small peptides in the apple juice medium (Elbert & Esselen, 1959) and could be improved by adding the yeast extract. The content of both small peptides and vitamins improved the bacterial growth. Avonts et al. (2004) have shown that addition of yeast extract (0·3–1·0% w/v) to milk medium enhanced both growth and bacteriocin production for all strains and no growth took place in milk medium without that. An inverse effect was observed to that of pH value (Figs. 2 and 3). As pH increased the response decreased. These results were also confirmed by Doleyres et al. (2002). They reported lower growth of Bifido. longum with increased pH. The results showed that the growth of Bifido. animalis subsp. lactis was dramatically decreased when initial pH of culture media increased to 7·5 (Fig. 2). According to Poolman & Konings (1988), the amino acid or peptide transport, which is one of the growth-rate-determining steps, depends on the pH of the culture medium. The optimum pH for the growth of bifidobacteria has been reported to be between 6·5 and
Source
D.F.
Sum of square
Mean square
Model Residual Lack of fit Pure error
14 16 10 6
134·517 5·837 5·487 0·350
9·6083 0·3648 0·5487 0·0583
Total
30
140·354
Log (CFU/ml) [Cal.]
Log (CFU/ml)
Term
F-value
P-value
26·34
< 0·001
9·42
0·06
y = 0.9584x + 0.2951 R² = 0.9584
Log (CFU/ml) [Expt.]
Fig. 4. The relationship between the calculated the growth of probiotic Bifido. animalis subsp.lactis and experimental data.
7·0 (Rozada-Sánchez et al. 2008). Thus, the low nutrient consumption and consequently, the low growth rate of microorganism during incubation in the culture at the initial pH = 7·5, could be related to a limitation in nutrient transport. The inulin supplementation of apple juice had a significant influence on the growth of Bifido. animalis subsp.
Medium for probiotic biomass production using RSM lactis (Figs. 1 and 3). The effect of inulin has already been reported to stimulate Bifidobacterium spp. metabolism (Shin et al. 2000; Bruno et al. 2002; Akalın et al. 2004). Also the results showed that in compared with yeast extract the prebiotic growth development effects of inulin is weaker. The effects of growth conditions on the log10 (number of viable cells/ml) by the regression coefficients of fitted second-order polynomial are presented in Table 3. It was evident that the linear terms except for lactose concentration were significant (P < 0·05), whereas all the interaction terms were not significant (P > 0·05). The results indicated that the pH and yeast extraction concentration were the major contributing factors to growth of Bifido. lactis (Table 3). Rozada-Sánchez et al. (2008) also showed that yeast extract is a strong growth promoter for Bifidobacterium spp. Most strains of Bifidobacteria are unable to grow in a totally synthetic medium and require complex nitrogenous substrates such as bovine casein hydrolysates, milk whey or yeast extract (Poch & Bezkorovainy, 1988; Petschow & Talbott, 1990).
63
compared with the experimental values (Fig. 4). The figure showed that was good agreement between model (Eq. 2) and experimental data. Conclusion The response surface methodology was successfully employed to the growth of Bifido. animalis subsp. lactis. The second-order polynomial model gave a satisfactory description of the experimental data. Yeast extract concentration and initial pH culture media were the most important factors affecting on the growth of the microorganism, whereas lactose concentration had no significant effects (P > 0·05). Predicted and experimental results showed high similarity, which reflected the accuracy and applicability of RSM to process optimisation of probiotic production. The present study introduced the new production conditions for Bifidobacterium strains that are potentially useful to the production of novel dairy foods for human health benefit. References
Fitting the model and Response surface analysis The experimental results of the growth of Bifido. animalis subsp. lactis were presented in Table 2. The log10 (number of viable cells/ml) was analysed to get a regression model. The following mathematical model was used to express response as a function of the independent variables: Y ¼849 þ 026x2 þ 196x3 085x4 046x21 057x23 055x24
ð2Þ
where Y is the log10 (number of viable cells/ml), whereas x1, x2, x3 and x4 are the coded variables for lactose, inulin and yeast extract concentration and initial pH respectively. If the model exhibits an unsuitable fit, there may be poor or misleading results from a fitted response surface (LiyanaPathirana & Shahidi, 2005). An ANOVA analysis was performed in Table 4. The P-value of the model was less than 0·001 (Table 4). This confirmed that the model fitness was good and acceptable. For Eq. (2), lack of fit P-value of 0·06 implied that the model of number of viable cells developed was insignificance. The non-significant lack of fit indicates good predictability of the model (Khayati 2013). Coefficient of determination (R2) is defined as the ratio of the explained variation to the total variation and used to measure the degree of fitness (Nath & Chattopadhyay, 2007). The closer the R2 value to unity, the better the empirical models fits the actual data (Sin et al. 2006). On the other hand, the smaller the R2 value the less relevance the dependent variables in the model have in explaining the behaviour of variations (Lee et al. 2006). By analysis of variance, the R2 value of this model was determined to be 0·958. Therefore, the developed model could adequately represent the real relationship among the factors chosen. Regression model was used for the log10 (number of viable cells/ml) predicted values calculation and the results
Akalın AS, Fenderya S & Akbulut N 2004 Viability and activity of Bifidobacteria in yoghurt containing fructo oligosaccharide during refrigerated storage. International Journal of Food Science and Technology 39 613–621 Alander M & Mattila-Sandholm T 2000 Functional foods for EU-Health in 2000, in 4th Workshop, FAIR CT96-1028, PROBDEMO, VTT Symposium 198, Rovaniemi, Finland Avonts L, Van Uytven E & De Vuyst L 2004 Cell growth and bacteriocin production of probiotic Lactobacillus strains in different media. International Dairy Journal 14 947–955 Bousvaros A, Guandalini S, Baldassano RN, Botelho C, Evans J, Ferry GD, Goldin B, Hartigan L, Kugathasan S, Levy J, Murray KF, OlivaHemker M, Rosh JR, Tolia V, Zholudev A, Vanderhoof JA & Hibberd PL 2005 A randomized, double-blind trial of Lactobacillus GG versus placebo in addition to standard maintenance therapy for children with Crohn’s disease. Inflammatory Bowel Disease 11 833–839 Bruno A, Lankaputhra WEV & Shah NP 2002 Growth viability and activity of Bifidobacterium spp. in skim milk containing prebiotics. Journal of Food Science 67 2740–2744 Costa MGM, Fonteles TV, de Jesus ALT & Rodrigues S 2013 Sonicated pineapple juice as substrate for L. casei cultivation for probiotic beverage development: Process optimisation and product stability. Food Chemistry 139 261–266 Ding WK & Shah NP 2008 Survival of free and microencapsulated probiotic bacteria in orange and apple juices. International Food Research Journal 15 219–232 Doleyres Y & Lacroix C 2005 Technologieswith free and immobilised cells for probiotic bifidobacteria production and protection. International Dairy Journal 15 973–988 Doleyres Y, Fliss I & Lacroix C 2002 Quantitative determination of the spatial distribution of pure- and mixed-strain immobilized cells in gel beads by immunofluorescence. Applied Microbiology and Biotechnology 59 297–302 Du Toit M, Engelbrecht L, Lerm E & Krieger-Weber S 2011 Lactobacillus: The next generation of malolactic fermentation starter cultures—An overview. Food Bioprocess and Technology 4 876–906 Elbert EM & Esselen WB 1959 Free Amino Acids and Thiamine of Apple Juice and their relation of Fermentation. Journal of Food Science 24 352–361 Fonteles TV, Costa MG, de Jesus ALT & Rodrigues S 2012 Optimization of the fermentation of cantaloupe juice by Lactobacillus casei NRRL B-442. Food and Bioprocess Technology 5 2819–2826
64
M Anvari and others
Guarner F & Schaafsma GJ 1998 Probiotics. International Journal of Food Microbiology 39 237–238 Heller KJ 2001 Probiotic bacteria in fermented foods: product characteristics and starter organisms. American Journal of Clinical Nutrition 73 374–379 Khayati G 2013 Optimization of Propionic Acid Extraction by Aqueous Two-Phase System Using Response Surface Methodology. Chemical Engineering Communications 200 667–677 Khayati G & Kiyani F 2012 A statistical approach for optimization of lipase production by using rice straw: analysis of different inducers and nitrogen sources effect. Minerva Biotecnologica 24 83–89 Kodaka H, Mizuochi S, Teramura H & Nirazuka T 2005 Comparison of the compact dry TC method with the standard pour plate method (AOAC Official Method 966·23) for determining aerobic colony counts in food samples: performance-tested method SM. Journal of AOAC International 88 1702–1713 Kuehbacher T, Ott SJ, Helwig U, Mimura T, Rizzello F, Kleessen B, Gionchetti P, Blaut M, Campieri M, Folsch UR, Kamm MA & Schreiber S 2006 Bacterial and fungal microbiota in relation to probiotic therapy (VSL#3) in pouchitis. Gut 55 833–841 Lee WC, Yusof S, Hamid NSA & Baharin BS 2006 Optimizing conditions for hot water extraction of banana juice using response surface methodology (RSM). Journal of Food Engineering 75 473–479 Lin DC 2003 Probiotics as functional foods. Nutrition and Clinical Practice 18 497–506 Liyana-Pathirana C & Shahidi F 2005 Optimization of extraction of phenolic compounds from wheat using response surface methodology. Food Chemistry 93 47–56 Lourens-Hattingh A & Viljoen BC 2001 Yogurt as probiotic carrier food. International Dairy Journal 11 1–17 Montgomery DC 2001 Design and Analysis of Experiments. 5th edition. New York: John Wiley and Sons Mousavi ZE, Mousavi SM, Razavi SH, Emam-Djomeh Z & Kiani H 2011 Fermentation of pomegranate juice by probiotic lactic acid bacteria. World Journal of Microbiology and Biotechnology 27 123–128 Nath A & Chattopadhyay PK 2007 Optimization of oven toasting for improving crispness and other quality attributes of ready to eat potato–soy snack using response surface methodology. Journal of Food Engineering 80 1282–1292
Oliveira MN & Damin MR 2003 Effect of total solids and sucrose contents on acidity, firmness and viability of yogurt and probiotic bacteria in fermented milk. Ciência e Tecnologia de Alimentos 23 172–176 Pereira ALF, Maciel TC & Rodrigues S 2011 Probiotic beverage from cashew apple juice fermented with Lactobacillus casei. Food Research International 44 1276–1283 Petschow BW & Talbott RD 1990 Growth promotion of Bifidobacterium species by whey and casein fractions from human and bovine milk. Journal of Clinical Microbiology 28 287–292 Poch M & Bezkorovainy A 1988 Growth-enhancing supplements for various pieces of the genus Bifidobacterium. Journal of Dairy Science 71 3214–3221 Poolman B & Konings WN 1988 Relation of growth of Streptococcus lactis and Streptococcus cremoris to amino acid transport. Journal of Bacteriology 170 700–707 Reid G, Charbonneau D, Erb J, Kochanowski B, Beuerman D, Poehner R & Bruce AW 2003 Oral use of Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 significantly alters vaginal flora: randomized, placebo-controlled trial in 64 healthy women. FEMS Immunology and Medicinal Microbiology 35 131–134 Rozada-Sánchez R, Sattur AP, Thomas K & Pandiella SS 2008 Evaluation of Bifidobacterium spp. for the production of a potentially probiotic malt-based beverage. Process Biochemistry 43 848–854 Shin HS, Lee JH, Pestka JJ & Ustunol Z 2000 Growth and viability of commercial Bifidobacterium spp. in skim milk containing oligosaccharides and inulin. Journal of Food Science 65 885–887 Sin HN, Yusof S, Hamid NSA & Rahman RA 2006 Optimization of enzymatic clarification of sapodilla juice using response surface methodology. Journal of Food Engineering 73 313–319 Szymanski H, Pejcz J, Jawien M, Chmielarczyk A, Strus M & Heczko PB 2006 Treatment of acute infectious diarrhoea in infants and children with a mixture of three Lactobacillus rhamnosus strains—a randomized, doubleblind, placebo-controlled trial. Alimentary Pharmacology & Therapeutics 23 247–253 Weinbreck F, Bodnár I & Marco ML 2010 Can encapsulation lengthen the shelf-life of probiotic bacteria in dry products? International Journal of Food Microbiology 136 364–367
© Proprietors of Journal of Dairy Research 2013
59
Optimisation of medium composition for probiotic biomass production using response surface methodology Masumeh Anvari1*, Gholam Khayati2 and Shora Rostami3 1
Department of Microbiology, Faculty of Sciences, Islamic Azad University, Rasht Branch, P.O. Box 41335-3516, Rasht, Iran Department of Chemical Engineering, Faculty of Eng., Guilan University, P. O. Box 3756-41635, Rasht, Iran 3 Department of Microbiology, Sciences and Research Branch, Islamic Azad University, Guilan, Iran 2
Received 25 April 2013; accepted for publication 16 October 2013
This study was aimed to optimise lactose, inulin and yeast extract concentration and also culture pH for maximising the growth of a probiotic bacterium, Bifidobacterium animalis subsp. lactis in apple juice and to assess the effects of these factors by using response surface methodology. A second-order central composite design was applied to evaluate the effects of these independent variables on growth of the microorganism. A polynomial regression model with cubic and quadratic terms was used for analysis of the experimental data. It was found that the effects involving inulin, yeast extract and pH on growth of the bacterium were significant, and the strongest effect was given by the yeast extract concentration. Estimated optimum conditions of the factors on the bacterial growth are as follows: lactose concentration = 9·5 g/l; inulin concentration = 38·5 mg/l; yeast extract concentration = 9·6 g/l and initial pH = 6·2. Keywords: Bifidobacterium animalis subsp. Lactis, probiotic, response surface methodology, apple juice.
Probiotics are defined as viable microorganisms which, upon ingestion in sufficient amounts, exert health benefits to the host beyond inherent basic nutrition (Guarner & Schaafsma, 1998). They have been used for the treatment of various types of diarrhoea (Szymanski et al. 2006), urogenital infections (Reid et al. 2003), and gastrointestinal diseases such as Crohn’s disease (Bousvaros et al. 2005) and pouchitis (Kuehbacher et al. 2006), although there is still no consensus about their effectiveness (Lin, 2003). Probiotic bacteria, often belonging to the Lactobacillus and Bifidobacterium genera (Weinbreck et al. 2010). Lactic acid bacteria are commercialised mainly as food supplements with dairy products being the most often used vehicle (Heller, 2001; Lourens-Hattingh & Viljoen, 2001). However, recent studies have suggested fruit juices as an alternative vehicle for the incorporation of probiotics (Mousavi et al. 2011; Pereira et al. 2011; Fonteles et al. 2012; Costa et al. 2013). Fruit juices are rich in nutrients and do not contain starter cultures that compete for nutrients with probiotics. Furthermore, fruit juices contain high amounts of sugars, which could encourage probiotic growth (Ding & Shah, 2008). The optimal growth of probiotic bacteria is affected by fermentation conditions such as pH, temperature, medium
*Corresponding author; e-mail: [email protected]
formulation and the others. Study of the individual and interactive effects of these factors will help in efforts to optimise biomass production of the probiotic microorganism (Du Toit et al. 2011). According to Oliveira & Damin (2003), to evaluate the growth of lactic acid bacteria, it is necessary to know the substrates applied for the microbial growth, as well as, the optimal temperature and pH values because these factors are the most important for the microbial development. Response surface methodology (RSM) is a useful model for studying the effect of several factors influencing the responses by varying them simultaneously and carrying out a limited number of experiments. In addition, response surface methodology is an efficient strategic experimental tool by which the optimal conditions of a multivariable system may be determined (Khayati & Kiyani, 2012; Khayati, 2013). Lactobacilli are also extensively used as probiotics, but no information is available on the growth of the species Bifidobacterium in apple juice. The objective of this study was to determine the suitability of apple juice as a raw material for production of probiotic by Bifidobacterium spp.. Thus, the use of apple juice as substrate to produce a probiotic was studied herein. So in this paper, the growth of probiotic Bifidobacterium animalis subsp. lactis in apple juice (as a basement medium) with a function of four affecting parameters including lactose (g/l), inulin (mg/l) and yeast
60
M Anvari and others
extract concentration (g/l) and initial pH was studied by RSM. Then these bacterial strains can be successfully manufactured and incorporated into highly acceptable dairy food products where they can retain their viability and functionality. (Alander & Mattila-Sandholm, 2000).
Materials and methods Microorganisms and media Bifidobacterium animalis subsp. lactis PTCC 1736 was prepared from the Iranian Research Organization for Science and Technology (IROST). All chemicals used were of analytical grade. The strain was maintained in MRS agar medium containing 1% lithium chloride, 0·3% sodium propionate and 0·5% L-cysteine at 4 °C. Subcultures (1% inoculum, incubated for 10 h at 37 °C in anaerobic jars) were prepared immediately before the culture was used experimentally. Fermentation conditions In the design orthogonal array, each row consists of a number of conditions depending on the levels assigned to each factor. Submerged fermentation experiments were carried out in cotton plugged 100 ml Erlenmeyer flasks containing 30 ml apple juice in different conditions. pH adjusted by adding 0·02 N HCl prior to sterilisation (15 min, 121 °C). Lactose and yeast extract were sterilised separately. Cultures were inoculated with mentioned above pre-cultures (approximate concentration of 103 CFU/ml of the probiotic strain), and incubated for 36 h at 37 °C in an anaerobic atmosphere (85% N2/10% H2/5% CO2). Microbiological analysis The number of viable cells was determined as colony forming units (CFU). One millilitre of sample was diluted with 9 ml 0·1% sterile peptonated water. Serial decimal dilutions of each sample were plated in triplicate onto MRSLP agar; afterwards, bacteria were counted applying the pour plate technique (Kodaka et al. 2005). Plates were incubated at 37 °C for 72 h under anaerobic conditions (Gas-Pak plus system). CFU were enumerated in plates containing 30–300 colonies. The selectivity of culture media was confirmed by microscopic examination of cells in the colonies. Results were expressed as log10 CFU/ml. Experimental design and statistical analysis In this experiment, the response was the growth of Bifido. animalis subsp. lactis PTCC 1736, represented by log10 (number of viable cells/ml; CFU/ml). The effect of four independent variables: lactose (x1), inulin (x2) and yeast extraction concentration (x3) also initial pH (x4) on the response variable (Y, the log10 (number of viable cells/ml)) was evaluated using central composite design (CCD)
Table 1. Independent variables and their coded and actual levels Actual values of coded levels Independent variables (factors) Lactose concentration (g/l) Inulin concentration (mg/l) Yeast extract concentration (g/l) Initial pH
Symbol α
1
0
+1
+ α†
x1
0
5
10
15
20
x2
5
15
25
35
45
x3
2
4
6
8
10
x4
5·5
6·0
6·5
7·0
7·5
† α = 2·0 (star point for orthogonal CCD for the case of 4 independent variables)
Table 2. Central composite experimental design and responses Independent variables† Run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
x1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0
x2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0
x3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 2 2 0 0 0 0 0 0 0 0 0
x4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0
Response (Y )‡ 4·354 5·123 5·636 6·388 8·691 9·134 10·268 9·813 3·508 3·782 3·172 4·131 8·227 7·308 8·662 8·986 6·195 7·044 7·745 7·483 3·151 9·236 8·452 4·041 8·109 8·562 8·770 8·461 8·398 8·801 8·369
† x1: lactose concentration (g/l); x2: inulin concentration (mg/l); x3: yeast extract concentration (g/l); x4: initial pH ‡ Y: log10 (CFU/ml)
(Table 1). The five coded levels of each variable were incorporated into the design and were analysed in 31 experimental trails (Table 2). The central point of the design was repeated seven times to calculate the reproducibility of the method (Montgomery, 2001). For each experimental trail
Medium for probiotic biomass production using RSM
61
a
a
10
10
Log (CFU/ml)
Log (CFU/ml)
5
0
0 0
15
9 6 3 45 yeast extract con. (g/l)
30
inulin con. (mg/l)
0
30
7.5
10
7.0
8
initial pH
7 6 8
5
45 5
6
7
8
initial pH
5
10
9
yeast extract con. (g/l)
15
inulin con. (mg/l)
b
b
5
6
7
6.5
6
9
6.0
4 3
7
4
2 5
10
15
20
25
30
35
40
45
5.5 5
8
10
15
20
25
30
35
40
45
inulin con. (mg/l)
inulin con. (mg/l)
Fig. 1. Response surface plot (a) and its contour plot (b) for the effects of inulin concentration and yeast extract concentration on log10 (CFU/ml) at its centre level.
Fig. 2. Response surface plot (a) and its contour plot (b) for the effects of inulin concentration and initial pH on log10 (CFU/ml) at its centre level.
of the independent variables in the experimental design, the dependent parameter (the log10 (number of viable cells/ml)) was determined. The effect of these independent variables x1, x2, x3 and x4 on the response Y was investigated using the second-order polynomial regression equation. This equation, derived using RSM for the evaluation of the response variable, is as follows:
statistical calculations to generate response surface and contour maps from the regression models. The analysis of data and the optimising process were generated using Minitab statistical software version 15.
Y ¼ β0 þ
k X i¼1
βi xi þ
k X i¼1
βii x2 þ
k X
βij xi xj
ð1Þ
i=j
where β0 is defined as the constant, βi the linear coefficient, βii the quadratic coefficient and βij the interaction coefficient. xi and xj are the independent variables while k equals to the number of the tested factors (k = 4). The analysis of variance (ANOVA) tables were generated and the effect and regression coefficients of individual linear, quadratic and interaction terms were determined. The significances of all terms in the polynomial were judged statistically by computing the P < 0·05. The regression coefficients were then used to make
Results and discussion Effects of independent variables on responses Interest in the incorporation of Bifidobacteria into fermented products has developed considerably over recent years. In many studies reporting human health benefits associated with the consumption of these bacteria so they have received attention as probiotic (Doleyres & Lacroix, 2005). Response surface was used to illustrate the effect of lactose, inulin and yeast extract concentration and initial pH on the growth of Bifido. animalis subsp. lactis in apple juice. The response surfaces and contour plots of the bacterial growth conditions are presented in Figs. 1–3. Also, the effect of inulin and yeast extraction concentrations on the response is shown in Fig. 1.
M Anvari and others
62
a
Table 3. Significance of regression coefficients of the fitted secondorder polynomial model for response (Y ) 10
5
0
3
6
9
5
yeast extract con. (g/l)
b
7.5
6
7
8
initial pH
6
initial pH
7.0
10
6.5
SE
β0 Linear β1 β2 β3 β4
8·49570
0·2283
< 0·001
0·16014 0·26682 1·96532 0·85218
0·1233 0·1233 0·1233 0·1233
0·212 0·046 < 0·001 < 0·001
Quadratic β11 β22 β33 β44
0·46394 0·21532 0·57043 0·55723
0·1129 0·1129 0·1129 0·1129
0·001 0·075 < 0·001 < 0·001
Interaction β12 β13 β14 β23 β23 β34
0·06326 0·21009 0·05449 0·11321 0·16738 0·13655
0·1510 0·1510 0·1510 0·1510 0·1510 0·1510
0·681 0·183 0·723 0·464 0·284 0·379
Coef.
P-value
Table 4. Analysis of variance (ANOVA) of the regression parameters for the response surface model
6.0
4
5.5 2
Regression Coef.
8
3
4
5
6
7
8
9
10
yeast extract con. (g/l)
Fig. 3. Response surface plot (a) and its contour plot (b) for the effects of yeast extract concentration and initial pH on log10 (CFU/ ml) at its centre level.
The growth of microorganism was increased with the yeast extract concentration increasing to a certain value, thereafter was constant. It is due to the low content of free amino acids and small peptides in the apple juice medium (Elbert & Esselen, 1959) and could be improved by adding the yeast extract. The content of both small peptides and vitamins improved the bacterial growth. Avonts et al. (2004) have shown that addition of yeast extract (0·3–1·0% w/v) to milk medium enhanced both growth and bacteriocin production for all strains and no growth took place in milk medium without that. An inverse effect was observed to that of pH value (Figs. 2 and 3). As pH increased the response decreased. These results were also confirmed by Doleyres et al. (2002). They reported lower growth of Bifido. longum with increased pH. The results showed that the growth of Bifido. animalis subsp. lactis was dramatically decreased when initial pH of culture media increased to 7·5 (Fig. 2). According to Poolman & Konings (1988), the amino acid or peptide transport, which is one of the growth-rate-determining steps, depends on the pH of the culture medium. The optimum pH for the growth of bifidobacteria has been reported to be between 6·5 and
Source
D.F.
Sum of square
Mean square
Model Residual Lack of fit Pure error
14 16 10 6
134·517 5·837 5·487 0·350
9·6083 0·3648 0·5487 0·0583
Total
30
140·354
Log (CFU/ml) [Cal.]
Log (CFU/ml)
Term
F-value
P-value
26·34
< 0·001
9·42
0·06
y = 0.9584x + 0.2951 R² = 0.9584
Log (CFU/ml) [Expt.]
Fig. 4. The relationship between the calculated the growth of probiotic Bifido. animalis subsp.lactis and experimental data.
7·0 (Rozada-Sánchez et al. 2008). Thus, the low nutrient consumption and consequently, the low growth rate of microorganism during incubation in the culture at the initial pH = 7·5, could be related to a limitation in nutrient transport. The inulin supplementation of apple juice had a significant influence on the growth of Bifido. animalis subsp.
Medium for probiotic biomass production using RSM lactis (Figs. 1 and 3). The effect of inulin has already been reported to stimulate Bifidobacterium spp. metabolism (Shin et al. 2000; Bruno et al. 2002; Akalın et al. 2004). Also the results showed that in compared with yeast extract the prebiotic growth development effects of inulin is weaker. The effects of growth conditions on the log10 (number of viable cells/ml) by the regression coefficients of fitted second-order polynomial are presented in Table 3. It was evident that the linear terms except for lactose concentration were significant (P < 0·05), whereas all the interaction terms were not significant (P > 0·05). The results indicated that the pH and yeast extraction concentration were the major contributing factors to growth of Bifido. lactis (Table 3). Rozada-Sánchez et al. (2008) also showed that yeast extract is a strong growth promoter for Bifidobacterium spp. Most strains of Bifidobacteria are unable to grow in a totally synthetic medium and require complex nitrogenous substrates such as bovine casein hydrolysates, milk whey or yeast extract (Poch & Bezkorovainy, 1988; Petschow & Talbott, 1990).
63
compared with the experimental values (Fig. 4). The figure showed that was good agreement between model (Eq. 2) and experimental data. Conclusion The response surface methodology was successfully employed to the growth of Bifido. animalis subsp. lactis. The second-order polynomial model gave a satisfactory description of the experimental data. Yeast extract concentration and initial pH culture media were the most important factors affecting on the growth of the microorganism, whereas lactose concentration had no significant effects (P > 0·05). Predicted and experimental results showed high similarity, which reflected the accuracy and applicability of RSM to process optimisation of probiotic production. The present study introduced the new production conditions for Bifidobacterium strains that are potentially useful to the production of novel dairy foods for human health benefit. References
Fitting the model and Response surface analysis The experimental results of the growth of Bifido. animalis subsp. lactis were presented in Table 2. The log10 (number of viable cells/ml) was analysed to get a regression model. The following mathematical model was used to express response as a function of the independent variables: Y ¼849 þ 026x2 þ 196x3 085x4 046x21 057x23 055x24
ð2Þ
where Y is the log10 (number of viable cells/ml), whereas x1, x2, x3 and x4 are the coded variables for lactose, inulin and yeast extract concentration and initial pH respectively. If the model exhibits an unsuitable fit, there may be poor or misleading results from a fitted response surface (LiyanaPathirana & Shahidi, 2005). An ANOVA analysis was performed in Table 4. The P-value of the model was less than 0·001 (Table 4). This confirmed that the model fitness was good and acceptable. For Eq. (2), lack of fit P-value of 0·06 implied that the model of number of viable cells developed was insignificance. The non-significant lack of fit indicates good predictability of the model (Khayati 2013). Coefficient of determination (R2) is defined as the ratio of the explained variation to the total variation and used to measure the degree of fitness (Nath & Chattopadhyay, 2007). The closer the R2 value to unity, the better the empirical models fits the actual data (Sin et al. 2006). On the other hand, the smaller the R2 value the less relevance the dependent variables in the model have in explaining the behaviour of variations (Lee et al. 2006). By analysis of variance, the R2 value of this model was determined to be 0·958. Therefore, the developed model could adequately represent the real relationship among the factors chosen. Regression model was used for the log10 (number of viable cells/ml) predicted values calculation and the results
Akalın AS, Fenderya S & Akbulut N 2004 Viability and activity of Bifidobacteria in yoghurt containing fructo oligosaccharide during refrigerated storage. International Journal of Food Science and Technology 39 613–621 Alander M & Mattila-Sandholm T 2000 Functional foods for EU-Health in 2000, in 4th Workshop, FAIR CT96-1028, PROBDEMO, VTT Symposium 198, Rovaniemi, Finland Avonts L, Van Uytven E & De Vuyst L 2004 Cell growth and bacteriocin production of probiotic Lactobacillus strains in different media. International Dairy Journal 14 947–955 Bousvaros A, Guandalini S, Baldassano RN, Botelho C, Evans J, Ferry GD, Goldin B, Hartigan L, Kugathasan S, Levy J, Murray KF, OlivaHemker M, Rosh JR, Tolia V, Zholudev A, Vanderhoof JA & Hibberd PL 2005 A randomized, double-blind trial of Lactobacillus GG versus placebo in addition to standard maintenance therapy for children with Crohn’s disease. Inflammatory Bowel Disease 11 833–839 Bruno A, Lankaputhra WEV & Shah NP 2002 Growth viability and activity of Bifidobacterium spp. in skim milk containing prebiotics. Journal of Food Science 67 2740–2744 Costa MGM, Fonteles TV, de Jesus ALT & Rodrigues S 2013 Sonicated pineapple juice as substrate for L. casei cultivation for probiotic beverage development: Process optimisation and product stability. Food Chemistry 139 261–266 Ding WK & Shah NP 2008 Survival of free and microencapsulated probiotic bacteria in orange and apple juices. International Food Research Journal 15 219–232 Doleyres Y & Lacroix C 2005 Technologieswith free and immobilised cells for probiotic bifidobacteria production and protection. International Dairy Journal 15 973–988 Doleyres Y, Fliss I & Lacroix C 2002 Quantitative determination of the spatial distribution of pure- and mixed-strain immobilized cells in gel beads by immunofluorescence. Applied Microbiology and Biotechnology 59 297–302 Du Toit M, Engelbrecht L, Lerm E & Krieger-Weber S 2011 Lactobacillus: The next generation of malolactic fermentation starter cultures—An overview. Food Bioprocess and Technology 4 876–906 Elbert EM & Esselen WB 1959 Free Amino Acids and Thiamine of Apple Juice and their relation of Fermentation. Journal of Food Science 24 352–361 Fonteles TV, Costa MG, de Jesus ALT & Rodrigues S 2012 Optimization of the fermentation of cantaloupe juice by Lactobacillus casei NRRL B-442. Food and Bioprocess Technology 5 2819–2826
64
M Anvari and others
Guarner F & Schaafsma GJ 1998 Probiotics. International Journal of Food Microbiology 39 237–238 Heller KJ 2001 Probiotic bacteria in fermented foods: product characteristics and starter organisms. American Journal of Clinical Nutrition 73 374–379 Khayati G 2013 Optimization of Propionic Acid Extraction by Aqueous Two-Phase System Using Response Surface Methodology. Chemical Engineering Communications 200 667–677 Khayati G & Kiyani F 2012 A statistical approach for optimization of lipase production by using rice straw: analysis of different inducers and nitrogen sources effect. Minerva Biotecnologica 24 83–89 Kodaka H, Mizuochi S, Teramura H & Nirazuka T 2005 Comparison of the compact dry TC method with the standard pour plate method (AOAC Official Method 966·23) for determining aerobic colony counts in food samples: performance-tested method SM. Journal of AOAC International 88 1702–1713 Kuehbacher T, Ott SJ, Helwig U, Mimura T, Rizzello F, Kleessen B, Gionchetti P, Blaut M, Campieri M, Folsch UR, Kamm MA & Schreiber S 2006 Bacterial and fungal microbiota in relation to probiotic therapy (VSL#3) in pouchitis. Gut 55 833–841 Lee WC, Yusof S, Hamid NSA & Baharin BS 2006 Optimizing conditions for hot water extraction of banana juice using response surface methodology (RSM). Journal of Food Engineering 75 473–479 Lin DC 2003 Probiotics as functional foods. Nutrition and Clinical Practice 18 497–506 Liyana-Pathirana C & Shahidi F 2005 Optimization of extraction of phenolic compounds from wheat using response surface methodology. Food Chemistry 93 47–56 Lourens-Hattingh A & Viljoen BC 2001 Yogurt as probiotic carrier food. International Dairy Journal 11 1–17 Montgomery DC 2001 Design and Analysis of Experiments. 5th edition. New York: John Wiley and Sons Mousavi ZE, Mousavi SM, Razavi SH, Emam-Djomeh Z & Kiani H 2011 Fermentation of pomegranate juice by probiotic lactic acid bacteria. World Journal of Microbiology and Biotechnology 27 123–128 Nath A & Chattopadhyay PK 2007 Optimization of oven toasting for improving crispness and other quality attributes of ready to eat potato–soy snack using response surface methodology. Journal of Food Engineering 80 1282–1292
Oliveira MN & Damin MR 2003 Effect of total solids and sucrose contents on acidity, firmness and viability of yogurt and probiotic bacteria in fermented milk. Ciência e Tecnologia de Alimentos 23 172–176 Pereira ALF, Maciel TC & Rodrigues S 2011 Probiotic beverage from cashew apple juice fermented with Lactobacillus casei. Food Research International 44 1276–1283 Petschow BW & Talbott RD 1990 Growth promotion of Bifidobacterium species by whey and casein fractions from human and bovine milk. Journal of Clinical Microbiology 28 287–292 Poch M & Bezkorovainy A 1988 Growth-enhancing supplements for various pieces of the genus Bifidobacterium. Journal of Dairy Science 71 3214–3221 Poolman B & Konings WN 1988 Relation of growth of Streptococcus lactis and Streptococcus cremoris to amino acid transport. Journal of Bacteriology 170 700–707 Reid G, Charbonneau D, Erb J, Kochanowski B, Beuerman D, Poehner R & Bruce AW 2003 Oral use of Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 significantly alters vaginal flora: randomized, placebo-controlled trial in 64 healthy women. FEMS Immunology and Medicinal Microbiology 35 131–134 Rozada-Sánchez R, Sattur AP, Thomas K & Pandiella SS 2008 Evaluation of Bifidobacterium spp. for the production of a potentially probiotic malt-based beverage. Process Biochemistry 43 848–854 Shin HS, Lee JH, Pestka JJ & Ustunol Z 2000 Growth and viability of commercial Bifidobacterium spp. in skim milk containing oligosaccharides and inulin. Journal of Food Science 65 885–887 Sin HN, Yusof S, Hamid NSA & Rahman RA 2006 Optimization of enzymatic clarification of sapodilla juice using response surface methodology. Journal of Food Engineering 73 313–319 Szymanski H, Pejcz J, Jawien M, Chmielarczyk A, Strus M & Heczko PB 2006 Treatment of acute infectious diarrhoea in infants and children with a mixture of three Lactobacillus rhamnosus strains—a randomized, doubleblind, placebo-controlled trial. Alimentary Pharmacology & Therapeutics 23 247–253 Weinbreck F, Bodnár I & Marco ML 2010 Can encapsulation lengthen the shelf-life of probiotic bacteria in dry products? International Journal of Food Microbiology 136 364–367
