Evaluation of Listeria monocytogenes Survival in Ice Cream Mixes Flavored with Herbal Tea Using Taguchi Method

Abstract

In this study, the effects of the incorporation of some herbal teas at different concentrations into the ice cream mix on the population of Listeria monocytogenes were studied using Taguchi method. The ice cream mix samples flavored with herbal teas were prepared using green tea and sage at different concentrations. Afterward, fresh culture of L. monocytogenes was inoculated into the samples and the L. monocytogenes was counted at different storage periods. Taguchi method was used for experimental design and analysis. In addition, some physicochemical properties of samples were examined. Results suggested that there was some effect, although little, on the population of L. monocytogenes when herbal tea was incorporated into the ice cream mix. Additionally, the use of herbal tea caused a decrease in the pH values of the samples and significant changes in the color values.

Introduction

T he Listeria genus has important pathogenic species that can cause disease in both human and animals and exist in foods (McLauchlin et al., 2004). They can reproduce in the conditions of pH 6.0–9.0 and a _w 0.90–0.97. Although being psychrotropic bacteria, they can optimally develop in the temperature range of 30°C–37°C and also survive in a temperature range of 1°C–45°C (Lovett, 1989; Martin and Fisher, 1999). Ice cream is a dairy product consumed worldwide, particularly during summer time. As dairy products are rich sources of various nutritional elements required for the growth of microorganisms, they are easily deteriorated by microorganisms (Warke et al., 2000; Kanbakan et al., 2004). Therefore, the microbial risk in dairy products is rather high. Although ice cream is stored at frozen conditions, the psychrotropic bacteria can still grow in the ice cream mix (Miettinen et al., 1999). Listeria monocytogenes can be inactivated by heat treatment above 70°C, but when the product is contaminated, it may grow at refrigerated conditions (Miettinen et al., 1999; Wagner and McLauchlin, 2008). Being a psychrotrophic bacterium, L. monocytogenes may pose a high risk in the ice cream (Warke et al., 2000; Wagner and McLauchlin, 2008) and reproduce in the ice cream mix as a result of cross contamination.

There have been various studies in the literature about the antioxidative and antimicrobial properties of certain plants, vegetables, and fruit extracts (Kahkonen et al., 1999; Sagdic et al., 2002, 2003). Green tea (Camellia sinensis L.) is commonly consumed as a hot beverage in the Far East countries such as China (Hamilton-Miller, 1995). Green tea contains a high amount of natural polyphenols, of which catechins are the major components (Leung et al., 2001). Green tea was incorporated into ice cream to flavor the product (Kubo et al., 1992). It has been shown that green tea extracts or polyphenols had antimicrobial activity against some bacteria, protozoa, and viruses (Hamilton-Miller, 1995; Sakanaka et al., 2000). Sage (Salvia officinalis) is a popular herb commonly consumed as hot beverage. Sage is a wild shrub that usually grows in the Mediterranean region. There are ∼500 Salvia species in the nature (Gali-Muhtasib, 2006). Salvia species contain essential oils and phenolic compounds with antioxidative, anti-inflammatory, anticancer, and antimicrobial activities (Sagdic et al., 2002, 2003; Gali-Muhtasib, 2006; Ozkan et al., 2010).

Taguchi method, developed by Genichi Taguchi, is a highly effective statistical method specifically devised for the improvement of industrial products in Japan (Taguchi, 1986). Although initially implemented in various fields of engineering, it could be utilized for the optimization of any complex process (Dawson and Barnes, 1992). The effect of factors on characteristic properties and the optimum conditions of factors can be determined using Taguchi design. Taguchi design has certain advantages over the conventional statistical experimental designs in terms of determination of the optimum experimental conditions. The experimental condition with the least variability can be determined as the optimum condition. The signal-to-noise (S/N) ratio is used for the expression of variability in Taguchi method. Variability characteristics are inversely proportional to the S/N ratio and the optimum condition is the experimental condition with the maximum S/N ratio (Oztop et al., 2007). The higher the S/N, the better will be the quality of product. Also, the S/N ratio characteristic is commonly divided into three categories: smaller-the-better; larger-the-better; and nominal-the-best. In this study, the larger-the-better approach with the following equation was utilized: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*} \frac { S } { N } = - 10 \log \frac { 1 } { n } \left( \sum \limits_ { i = 1 } ^n \frac { 1 } { { y_i }^2 } \right)\end{align*} \end{document}

where n ≥ 2 is the number of observations in a sample size n and y is the observed data. Orthogonal arrays are often employed for determination of effect of many different parameters on the quality performance characteristics in a designed product set of experiments. Once the parameters affecting a product that can be controlled have been determined, the levels at which these parameters should be varied must be determined. The proper orthogonal array can be selected after determining the number of parameters and the number of levels.

The aim of the present study was to eliminate or reduce the potential risk of L. monocytogenes by incorporation of green tea and Salvia fructicosa extracts into the ice cream mix. Additionally, the experimental results of the study were evaluated using Taguchi method for decision of optimum values.

Materials and Methods

Preparation of the ice cream mix

Ingredients including UHT milk, nonfat milk powder, sugar, salep, emulsifier, and gelatine were purchased from a local market. Cream (70% fat) and herbal teas (sage: S. fructicosa Mill., and green tea: Camellia sinensis L.) were obtained from Unal Dairy and Ozselamoglu Food Ind., respectively. The ice cream mixes were prepared according to the method described by Karaman (2009). Teas at different concentration (2%, 5%, and 10%) were incorporated into the milk at 80°C and brewed for 15 minutes. Afterward, milk flavored with tea was used for the preparation of ice cream mix. Also, a control sample containing only L. monocytogenes with no herbal tea extract was prepared in the experimental design. Having been inoculated with 10¹⁰ cfu/mL L. monocytogenes 1/2b bacterial fresh culture at 0.5% concentration, the prepared ice cream mixes were aged at 4°C for different ageing times, namely 0, 5, 10, and 20 hours.

Enumeration of Listeria monocytogenes

Ten mL of ice cream mix samples was homogenized using a magnetic stirrer (Yellowline) after addition of 90 mL sterile solution of 0.85% (w/v) sodium chloride (Merck). Decimal dilutions were prepared in 9 mL sterile NaCl (0.85%).

At 0, 5, 10, and 20 hours of ageing of the ice cream mix, L. monocytogenes was enumerated on Oxford Listeria Selective Agar (Merck) using the pour-plate method, after incubation at 37°C for 24 hours (Akkaya et al., 2009). Bacterial counts were expressed as log₁₀ cfu/mL.

Physicochemical analysis

The color values of the samples were determined using an automatic colorimeter (Lovibond RT Series Reflactance Tintometer). A benchtop pH meter (WTW Inolab, Terminal Level 3, Weilheim) was utilized for determination of pH values. Titratable acidity was calculated according to the method described by Karaman (2009). Each analysis was repeated three times with two replications.

Taguchi method, design of experiment, and analysis

In the present study, for design of experiment, there were three factors and a mixed level design with two, four, and four levels, respectively. All experiments were carried out using Minitab 14 statistical software. The mixed level design used was L ₃₂ (2¹ × 4²) orthogonal array. The levels of each factor were represented by “1,” “2,” “3,” or “4” in the matrix. The factors and levels are presented in Table 1 and the L ₃₂ (2¹ × 4²) orthogonal array is shown in Table 2.

Table 1.

Factors and Levels Used in the Experiment

	Level
Factors	1	2	3	4
F ₁—herbal tea	Green tea	Sage	—	—
F ₂—concentration (%)	0	2	5	10
F ₃—time (hour)	0	5	10	20

Table 2.

L ₃₂ (2¹ × 4²) Orthogonal Array Table with Factors F ₁, F ₂, and F ₃ and Experimental Results for the Microbial Population and Their Corresponding Signal-to-Noise Ratios

	Factors				Measured parameter	Computed S/N ratio
Experimental run	F ₁	F ₂	F ₃	Designation	Listeria monocytogenes count (log cfu/mL)	Microbial population
1	1	1	1	F ₁₁ F ₂₁ F ₃₁	6.61314	16.4082
2	1	1	2	F ₁₁ F ₂₁ F ₃₂	6.61819	16.4148
3	1	1	3	F ₁₁ F ₂₁ F ₃₃	6.77574	16.6191
4	1	1	4	F ₁₁ F ₂₁ F ₃₄	6.76870	16.6101
5	1	2	1	F ₁₁ F ₂₂ F ₃₁	6.51809	16.2824
6	1	2	2	F ₁₁ F ₂₂ F ₃₂	6.54183	16.3140
7	1	2	3	F ₁₁ F ₂₂ F ₃₃	6.40581	16.1315
8	1	2	4	F ₁₁ F ₂₂ F ₃₄	6.26279	15.9354
9	1	3	1	F ₁₁ F ₂₃ F ₃₁	6.45733	16.2011
10	1	3	2	F ₁₁ F ₂₃ F ₃₂	6.18250	16.0119
11	1	3	3	F ₁₁ F ₂₃ F ₃₃	6.54270	16.1969
12	1	3	4	F ₁₁ F ₂₃ F ₃₄	6.61179	16.4064
13	1	4	1	F ₁₁ F ₂₄ F ₃₁	6.31825	16.0119
14	1	4	2	F ₁₁ F ₂₄ F ₃₂	6.55857	16.3362
15	1	4	3	F ₁₁ F ₂₄ F ₃₃	6.29402	15.9786
16	1	4	4	F ₁₁ F ₂₄ F ₃₄	6.37458	16.0890
17	2	1	1	F ₁₂ F ₂₁ F ₃₁	6.61314	16.4082
18	2	1	2	F ₁₂ F ₂₁ F ₃₂	6.61819	16.4148
19	2	1	3	F ₁₂ F ₂₁ F ₃₃	6.77574	16.6191
20	2	1	4	F ₁₂ F ₂₁ F ₃₄	6.76870	16.6101
21	2	2	1	F ₁₂ F ₂₂ F ₃₁	6.47406	16.2235
22	2	2	2	F ₁₂ F ₂₂ F ₃₂	6.40186	16.1261
23	2	2	3	F ₁₂ F ₂₂ F ₃₃	6.57306	16.3554
24	2	2	4	F ₁₂ F ₂₂ F ₃₄	6.58585	16.3722
25	2	3	1	F ₁₂ F ₂₃ F ₃₁	6.33805	16.0391
26	2	3	2	F ₁₂ F ₂₃ F ₃₂	6.63908	16.4422
27	2	3	3	F ₁₂ F ₂₃ F ₃₃	6.48856	16.2430
28	2	3	4	F ₁₂ F ₂₃ F ₃₄	6.66330	16.4738
29	2	4	1	F ₁₂ F ₂₄ F ₃₁	6.49829	16.2560
30	2	4	2	F ₁₂ F ₂₄ F ₃₂	6.47406	16.2235
31	2	4	3	F ₁₂ F ₂₄ F ₃₃	6.57526	16.3583
32	2	4	4	F ₁₂ F ₂₄ F ₃₄	6.57306	16.3554

S/N, signal-to-noise ratio.

Results and Discussion

Experimental results and data analysis

The objective of this experiment was to minimize the microbial load of ice cream mix with incorporation of herbal teas, and the “larger-the-better” approach was used to evaluate the microbial load of samples. Table 2 shows the actual data for the microbial population along with its computed S/N ratio. Table 3 shows the mean S/N ratio for each levels of the microbial population. These data were then plotted as shown in Figure 1 and also the interaction S/N ratios were plotted as shown in Figure 2.

FIG. 1.

The larger-the-better signal-to-noise ratio (S/N) graphs for the microbial population factors.

FIG. 2.

Interaction S/N ratio graphs for microbial population factors. Herbal tea 1: sage; herbal tea 2: green tea; concentration 1: control; concentration 2: 2% of the herbal tea; concentration 3: 5% of the herbal tea; concentration 4: 10% of the herbal tea; time 1: 1^st hour; time 2: 5^th hour; time 3: 10^th hour; time 4: 20^th hour.

Table 3.

Response Table for Average Signal-to-Noise Ratio for the Microbial Population Factors

	Mean S/N ratio
Factors	Level 1	Level 2	Level 3	Level 4	Delta	Rank
F ₁—herbal tea	16.25	16.35	—	—	0.10	3
F ₂—concentration (%)	16.51	16.22	16.25	16.20	0.31	1
F ₃—time (hour)	16.23	16.29	16.31	16.36	0.13	2

Conceptual S/N ratio approach

Taguchi suggests evaluating the means and S/N ratio using conceptual approach that involves plotting the effects and visually identifying the factors that are significant (Phadke, 1989).

The average S/N ratios for the larger-the-better approach for the microbial population and significant interactions are shown in Figures 1 and 2, respectively. Figure 1 suggests that the type of herbal tea, concentration, and time factors were significant. The sample containing green tea at the concentration of 2% and aged for 5 hours appears to be the best choice to get low value of the microbial load. Therefore, the optimum combination to get low value of the microbial population was F ₁₁ F ₂₂ F ₃₄ within the tested range. Further, as shown in the interaction graphs of S/N ratios in Figure 2, it could be claimed that the addition of green tea was more effective than sage in the reduction of L. monocytogenes.

In the present study, two different techniques have been used for the analysis of data. Both techniques gave similar conclusions. The use of S/N ratio indicated that the middle concentration of herbal tea was the level to obtain a microbiologically safe ice cream mix. The results also revealed that there was no effect of aging time on microbial count of ice cream mix. For example, the sample containing 2% sage had an initial microbial load of 6.47 log cfu/mL, and then the microbial load became 6.40, 6.57, and 6.58 log cfu/mL at the time of 5, 10, and 20 hours, respectively.

Physicochemical properties

The color, pH, and acidity values of ice cream mix samples are given in Table 4. The use of herbal tea affected the color of ice cream mix. The L* values decreased while a* and b* values increased for all samples as expected. It was observed that L* value of control sample was 63.36, but when the mix was incorporated with sage at 10%, L* value became 55.71. Similar trend was also observed for samples containing green tea, in which the L* value decreased with increasing concentration of herbal tea. Incorporation of herbal tea decreased the pH value of the ice cream mixes. The decrease in the pH value increased with increasing the amount of herbal tea incorporated into the mix. The pH value was measured as 6.43 for control sample, whereas it was 6.22 and 6.18 for samples 4 and 7, respectively. It could be stated that a decrease in pH value could be related to the reduction of L. monocytogenes in the samples. Titratable acidity of the samples increased with increasing herbal tea amount in the ice cream mix. The highest acidity value was determined in the samples flavored with incorporation of green tea at high concentration.

Table 4.

Physicochemical Properties of Ice Cream Mixes

		Color values
Sample	Herbal tea concentration	L*	a*	b*	pH	Titratable acidity (%)
1	Control	63.36 ± 0.58	−2.63 ± 0.01	6.13 ± 0.06	6.43 ± 0.01	0.02 ± 0.001
2	Sage—2%	60.47 ± 0.15	−2.53 ± 0.01	13.13 ± 0.05	6.38 ± 0.01	0.03 ± 0.000
3	Sage—5%	58.71 ± 0.13	−2.02 ± 0.01	15.62 ± 0.04	6.33 ± 0.02	0.03 ± 0.002
4	Sage—10%	55.71 ± 0.11	−0.11 ± 0.01	20.17 ± 0.04	6.22 ± 0.03	0.03 ± 0.001
5	Green tea—2%	61.27 ± 0.15	−1.81 ± 0.01	9.06 ± 0.05	6.40 ± 0.01	0.03 ± 0.002
6	Green tea—5%	58.38 ± 0.72	−1.13 ± 0.01	16.13 ± 0.26	6.29 ± 0.01	0.04 ± 0.003
7	Green tea—10%	59.90 ± 0.10	−1.10 ± 0.01	12.91 ± 0.03	6.18 ± 0.00	0.05 ± 0.000

Conclusion

The results obtained from conceptual S/N ratio approach for aged ice cream mixes flavored with herbal tea and the physicochemical properties of products indicated that Taguchi's robust design method could be suitable to analyze the microbial load of the ice cream samples in this study. Both conceptual S/N ratio and physicochemical properties of products for data analysis draw similar conclusions.

The microbial load of ice cream is very important for consumer health. Some psychrotrophic bacteria such as L. monocytogenes can grow in ice cream mix during ageing. It could be concluded that incorporation of herbal tea into the ice cream mix decreased the count of L. monocytogenes. The ice cream mix flavored with herbal tea may be acceptable for consumers. Further studies could be recommended to determine the acceptability of ice cream flavored with herbal tea extracts for consumers.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Akkaya

, Telli

, Sagdic

. Growth-death kinetics of Listeria monocytogenes in strained yogurt. Int J Food Prop, 2009; 12:705–712.

Dawson

, Barnes

. A new approach to the statistical optimization of catalyst preparation. Appl Catal A, 1992; 90:217–231.

Gali-Muhtasib

. Anticancer and medicinal properties of essential oil and extracts of East Mediterranean sage (Salvia triloba) Adv Phytomed, 2006; 2:169–180.

Hamilton-Miller

JMT

. Antimicrobial properties of tea (Camellia sinensis L.) Antimicrob Agents Chemother, 1995; 39:2375–2377.

Kahkonen

, Hopia

, Vuorela

et al. Antioxidant activity of plant extracts containing phenolic compounds. J Agric Food Chem, 1999; 47:3954–3962.

Kanbakan

, Çon

, Ayar

. Determination of microbiological contamination sources during ice cream production in Denizli, Turkey. Food Control, 2004; 15:463–470.

Karaman

. Investigation of Ice Cream Production Possibilities Aromatized with Tea or Some Herbal Teas [M.Sc. Thesis] Erciyes University, Graduate School of Natural and Applied Sciences, 2009 Kayseri, Turkey.

Kubo

, Muroi

, Himejima

. Antimicrobial activity of green tea flavor components and their combination effects. J Agric Food Chem, 1992; 40:240–245.

Leung

, Su

, Chen

, Zhang

, Huang

, Chen

. Theaflavins in black tea and catechins in green tea are equally effective antioxidants. J Nutr, 2001; 131:2248–2251.

10.

Lovett

. Listeria monocytogenes. Foodborne Bacterial Pathogens. Doyle

. New York: Marcel Dekker, 1989; 735.

11.

Martin

, Fisher

. Listeria monocytogenes. Encylopedia of Food Microbiology. Robinson

, Batt

, Pate

. San Diego, San Francisco, New York, Boston, London, Sydney, Tokyo: Academic Press, 1999; 1228–1251.

12.

McLauchlin

, Mitchell

, Smerdon

et al. Listeria monocytogenes and listeriosis: a review of hazard characterisation for use in microbiological risk assessment of foods. Int J Food Microbiol, 2004; 9:15–33.

13.

Miettinen

, Björkroth

, Korkeala

. Characterization of Listeria monocytogenes from an ice cream plant by serotyping and pulsed-field gel electrophoresis. Int J Food Microbiol, 1999; 46:187–192.

14.

Ozkan

, Sagdic

, Gokturk

et al. Study on chemical composition and biological activities of essential oil and extract from Salvia pisidica Boiss. & Heldr. ex Bentham. LWT Food Sci Technol, 2010; 43:186–190.

15.

Oztop

, Sahin

, Sumnu

. Optimization of microwave frying of potato slices by using Taguchi technique. J Food Eng, 2007; 79:83–91.

16.

Phadke

. Quality Engineering Using Robust Design. Englewood Cliffs, NJ: Prentice-Hall, 1989.

17.

Sagdic

, Karahan

, Ozcan

et al. Effect of some spice extracts on bacterial inhibition. Food Sci Technol Int, 2003; 9:353–356.

18.

Sagdic

, Kuscu

, Ozcan

et al. Effects of Turkish spice extracts at various concentrations on the growth of Escherichia coli O157:H7. Food Microbiol, 2002; 19:473–480.

19.

Sakanaka

, Juneja

, Taniguchi

. Antimicrobial effects of green tea polyphenols on thermophilic spore-forming bacteria. J Biosci Bioeng, 2000; 90:81–85.

20.

Taguchi

. Introduction to Quality Engineering. Tokyo: Asian Productivity Organization, 1986.

21.

Wagner

, McLauchlin

. Section 1: Biology. Handbook of Listeria monocytogenes. Liu

. New York: CRC Press, Taylor and Francis Group, 2008; 554.

22.

Warke

, Kamat

et al. Incidence of pathogenic psychrotrophs in ice creams sold in some retail outlets in Mumbai, India. Food Control, 2000; 11:77–83.