Sensory Evaluation of Non-Dairy Probiotic Beverages

This study measured the survival of Lactobacillus rhamnosus GR-1 in non-dairy probiotic beverages over a 28-day storage period and to determine which sample was most preferred on measures of appearance, consistency, flavour, texture, and overall acceptability. Three non-dairy samples and one control milk sample were prepared. The non-dairy samples were soy (1:3 product-to-water ratio), almond (1:3), and peanut (1:5). L. rhamnosus GR-1 remained viable (10 CFU/ml) in all samples over 28 days storage period and changes over time were dependent on the sample (P=0.03). The results of the sensory evaluation (n=90) showed that the soy and peanut samples were significantly different (P<0.03) from the control milk sample in appearance, consistency, flavour, texture, and overall acceptability. The almond sample was not rated significantly different (P>0.05) from the milk control in all categories. The results of this study suggest that probiotic almond milk may be a feasible substitute for conventional probiotic milk beverages, particularly for vegetarians who cannot consume dairy products and individuals with lactose intolerance.


Background and Rationale
Approximately 8% of Canadians are vegetarian (Iposos-Reid, 2004).Vegetarianism is a growing trend, which can be attributed to animal rights and ethics courses in post-secondary education, websites, magazines, newsletters, vegetarian cookbooks, and arrival of immigrants from countries where vegetarianism is commonly practiced (American Dietetic Association, 2003).Environmental concern, animal welfare, economic reasons, ethical considerations, world-hunger issues, and religious beliefs are additional reasons for consuming a vegetarian diet (American Dietetic Association, 2003;Fox & Ward, 2008).Plant-based diets are gathering growing appreciation and are recommended because of the many benefits that they confer, including lower levels of saturated fat, cholesterol, and animal protein, while offering higher carbohydrate, fibre, magnesium, boron, folate, antioxidant, carotenoid, and phytochemical intake (American Dietetic Association, 2003).
Fermented dairy products are the generic carrier matrices of probiotic microorganisms, with yogurt and fermented milk as the most commonly marketed products (Martins et al., 2013).Probiotic bacteria are beneficial bacteria that provide therapeutic effects on their host when ingested, and probiotic food products should contain at least 10 6 CFU/100g to transfer beneficial effects to the host (Hekmat & Reid, 2006;Rybka & Kailaspathy, 1995).L. rhamnosus GR-1 have been proven to help maintain a favourable microbial balance in the intestine and can survive in the intestinal tract without induction of systemic immune or inflammatory responses (Gardiner et al., 2002).The increasing consumption of probiotics parallels the growing trend in vegetarianism and diets that promote health and wellness (Martins et al., 2013).Therefore, the growing number of vegetarian individuals reinforces the importance of the need to develop non-dairy probiotic alternatives.Three non-dairy probiotic beverages were developed from soybeans, almond, and peanut.
Soybeans are composed of 40% high-quality protein and 20% oil (Liu, 2004).Soy contains high levels of a number of phytochemicals and is specially noted for the cholesterol-lowering effects of its protein (Liu, 2004).Among the numerous health-promoting compounds of soy, isoflavones play a key role, including the prevention and treatment of cardiovascular disease, cancer, osteoporosis, premenstrual and postmenstrual symptoms, and many more (Liu, 2004;Messina, 2003;Zubik & Meydani, 2003).Fermentation and heat treatment are methods to increase the digestibility of soy proteins (Liu, 2004;Han, Rombouts, & Nout, 2001).
To the authors' knowledge, there has been no research conducted on the survival of probiotic bacteria in non-dairy soy, almond, or peanut beverages nor were sensory evaluations conducted on these beverages.Isanga and Zhang (2007) conducted a study investigating the production and sensory properties of peanut milk stirred yogurt.Samples were prepared from a blend of 70% peanut milk with 30% reconstituted whole milk, 70% peanut milk and 30% reconstituted whole milk with yogurt flavouring, and pure reconstituted whole milk control (Isanga & Zhang, 2007).The results of this study indicate that peanut milk-based yogurt had a good sensory texture, appearance, and flavour (Isanga &Zhang, 2007).
The purpose of this study was to evaluate the survival of Lactobacillus rhamnosus GR-1 in soy, almond, and peanut milk over a 28-day period and to evaluate which sample was most preferred on measures of appearance, consistency, flavour, texture, and overall acceptability.

Milk (Control)
6% (wt/vol) sucrose was added to 1% (milk fat) milk.The milk and sucrose mixture was autoclaved at 15psi for 15 minutes.

Soymilk
The different probiotic samples were prepared using the authors' adaption of conventional methods for preparing nut milks.Dried soybeans (Glycine max) were soaked overnight.The rehydrated soybeans wereblended with water in a 1:3 ratio, respectively, for 2 minutes.The resulting slurry was strained through a double-lined cheesecloth to render soymilk.6% (wt/vol) sucrose was added to the soymilk, and the soymilk was autoclaved at 15psi for 15 minutes.

Almond Milk
Unblanched and unroasted almonds (Prunus amygdalus) were soaked overnight.After soaking, the skin of the almonds was manually removed.The hydrated almonds were blended with water in a 1:3 ratio, respectively, for 2 minutes.The resulting slurry was strained through a double-lined cheesecloth to render almond milk.6% (wt/vol) sucrose was added to the almond milk, and the almond milk was autoclaved at 15psi for 15 minutes.

Peanut Milk
Unroasted, blanched, de-skinned peanuts (Arachis hypogaea) were soaked overnight.Although a 1:3 ratio was ideally preferred to be consistent with the other samples, this could not be achieved without undesirably over-thickening the resulting peanut milk.Therefore, after soaking, the hydrated peanuts were blended with water in a 1:5 ratio, respectively, for 2 minutes.The resulting slurry was strained through a double-lined cheesecloth to render peanut milk.6% (wt/vol) sucrose was added to the peanut milk, and the peanut milk was autoclaved at 15psi for 15 minutes.

Probiotic Sample Production
After the milk (control), soymilk, almond milk, and peanut milk have been cooled to 37 °C, they were individually inoculated with 1% (wt/vol) L. Rhamnosus GR-1 stock solution and were incubated at 37 °C in an anaerobic environment for 24 hours.After incubation, the probiotic samples were stored at 4 °C.

Microbial Analysis
Enumeration of L. rhamnosus GR-1 from all samples was conducted on days 1, 14, and 28 of storage at 4 °C using serial dilution and subsequent plating.Two microbial analyses were performed using two different sets of the aforementioned samples.All four samples were diluted in sterile saline (0.85% wt/vol NaCl) to 10 -1 , 10 -3 , 10 -5 , 10 -7 , and 10 -8 dilution factors.0.1 mL of 10 -5 , 10 -7 , and 10 -8 sample dilutions were plated, on two separate MRSagar plates for each sample, with a calibrated pipette.Agar plates were prepared using 1.5% (wt/vol) agar (EMD Laboratories), 5.22% (wt/vol) MRS, and 1.5x10 -5 % (wt/vol) fusidic acid (Sigma Laboratories).The plates were inverted and incubated anaerobically at 37 °C for 48 hours.Viable numbers were determined and recorded as colony-forming units (CFU) per mL, based on colony counts.An average was taken from the two duplicate plates, for each replication, for analysis.

Panelists and Recruitment
For hedonic tests, typically 80-100 panelists are recommended (European Sensory Network, 2011); therefore, 110 participants were recruited from Western University and Brescia University College to fulfill both the study recommendations as well as to account for any potential drop-outs and/or unusable data.93 lay panelists participated in this study, and the data of 90 participants was used.Participants were recruited via email, word-of-mouth, and posters.Participants 18 to 55 years old were invited to participate in this study.Individuals who are lactose intolerant, pregnant, under 18 or over 55, diabetic, and/or undergoing chemotherapy and individuals who have a tree-nut allergy, and/or peanut allergy, and/or soy allergy were not eligible to participate.

Sample Preparation for Sensory Panel
All four samples (milk, soy, almond, peanut) were produced on the same date and were stored at 4 °C.Each sample was distributed in 20 g portions in plastic cups and was presented in a balanced random order to reduce order bias.Three-digit codes were used to mask the identity of the samples.

Sensory Evaluation
Each panelist was seated in individual booths in the sensory testing area, separate from the preparation room.All panelists received a tray of four coded samples at 4 °C in balanced random order.Each tray also included a napkin, a glass of water, a $5 Tim Card (token of appreciation), and the accompanying questionnaire.Before evaluating, panelists were instructed to evaluate each sample from left to right and to cleanse their palate with water between each sample.Panelists were also instructed to not speak to other panelists as they completed their questionnaire.

Data Collection
Panelists were asked to rate each sample based on the characteristics of appearance, consistency, flavour, texture, overall acceptability using a nine-point hedonic scale.The hedonic scale ranged from one to nine, where one corresponded with "dislike extremely," and nine corresponded with "like extremely."

Statistical Analysis
Statistical analysis was conducted using SAS Version 9.3.A one-way repeated measures analysis of variance (ANOVA) and Tukey-Kramer adjustment (P<0.05) were used to analyze viable bacteria between samples and between days.A one-way repeated measures ANOVA and Tukey-Kramer adjustment were also used to compare differences amongst samples for each characteristics of appearance, consistency, flavour, texture, and overall acceptability.

Bacterial Analysis
All three non-dairy milk samples (soy, almond, peanut) were successful at yielding a probiotic, fermented beverage.Table 1 shows the descriptive statistics for the enumeration of L. rhamnosus GR-1 in these samples and milk (control) over 28 days storage period.After 28 days of storage, the soymilk sample had the highest average mean count (22.1 × 10 7 CFU mL -1 ) amongst the non-dairy samples.The almond milk sample was significantly different when compared to soy, almond, and peanut samples (p=0.023).The almond sample was significantly different on Days 1 and 14 (p=0.035)and Days 1 and 28 (p=0.025).When comparing between time, Day 1 (p<0.001)and Day 14 (p=0.002)were significantly different.For Day 1, there is a significant difference between almond and soy (p=0.032),almond and milk (p=0.016),soy and milk (p=0.001),almond and peanut (p=0.004), and milk and peanut (p<0.001).On Day 14, there is a significant difference between almond and milk (p=0.004),soy and milk (p=0.005), and milk and peanut (p=0.004).

Appearance
There was a significant overall difference (p<0.001)amongst participants' perception of appearance for all the samples.Intra-sample comparisons reveal that there was a significant difference between all samples except between milk and almond (p=0.726).Mean score for the milk sample was the highest (5.9 ± 2.2), and soy scored the lowest (2.5 ± 1.4).

Consistency
There was a significant overall difference (p<0.001)amongst participants' perception of consistency for all the samples.Intra-sample comparisons reveal that there was a significant difference between all samples except between milk and almond (p=0.687).Mean scores for the consistency of milk was the highest (5.0 ± 2.6), and soy scored the lowest (2.4 ± 1.5).

Flavour
There was a significant overall difference (p<0.001)amongst participants' perception of flavour for all the samples.Intra-sample comparisons reveal that there was a significant difference between all samples except between milk and almond (p=0.067) and soy and peanut (p=0.523).Mean scores for the flavour of almond was the highest (5.1 ± 2.1), and soy scored the lowest (3.1 ± 1.9).

Texture
There was a significant overall difference (p<0.001)amongst participants' perception of texture for all the samples.Intra-sample comparisons reveal that there was a significant difference between all samples except between milk and almond (p=0.562).Mean scores for the texture of almond was the highest (5.4 ± 2.1), and soy scored the lowest (2.3 ± 1.3).

Overall Acceptability
There was a significant overall difference (p<0.001)amongst participants' perceived overall acceptability for all the samples.Intra-sample comparisons reveal that there was a significant difference between all samples except between milk and almond (p=0.434).Mean scores for the overall acceptability of almond was the highest (5.2 ± 2.1), and soy scored the lowest (2.7 ± 1.5).Table 2. Scoring results of the three non-dairy probiotic samples and milk (control) sample based on nine-point hedonic scale ratings, where 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like or dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, and 9 = like extremely.

Discussion
The primary objective of this study was to measure the survival of L. rhamnosus GR-1 in non-dairy probiotic beverages over 28-day storage period.This study showed that almond, soy, and peanut milks are viable matrices that support L. rhamnosus GR-1 survival greater than 10 6 CFU mL -1 during 28 days of storage.
The secondary objective of this study was to determine which non-dairy probiotic beverage was most preferred on measure of appearance, consistency, flavour, texture, and overall acceptability.Probiotic almond beverage was the most similar to probiotic milk beverage (control) on measures of appearance, consistency, flavour, texture, and overall acceptability.Mean scores for almond was the highest for consistency, flavour, texture, and overall acceptability when compared to probiotic soy, peanut, and milk (control) beverages.The results of this study suggest that probiotic almond beverage is a feasible alternative and potentially a better sensory substitute to probiotic milk beverage.This study also shows the potential in the marketability of probiotic almond beverages for vegetarians and individuals with lactose intolerance.
A limitation to this research was the number of repetitions conducted (3 in total).If more repetitions were to be conducted, then statistical results for the bacterial enumeration component of this study would be less affected by outliers, as evidenced by the large standard deviation values in Table 1.Another limitation to this study was the inability to ensure that the bacteria grown on plates were exclusively L. rhamnosus GR-1, as it was impossible to ensure a 100% sterile storage and plating environment.Future research on these non-dairy samples may benefit from more repetitions.

Conclusion
Soy, almond, and peanut milk samples successfully support the growth of L. rhamnosus GR-1 beyond 10 6 CFU mL -1 , each yielding a probiotic beverage sample.Sensory analysis of these samples reveals that probiotic almond beverage was most liked on measures of consistency, flavour, texture, and overall acceptability.This study shows promise in the potential marketability of probiotic almond beverages for vegetarians and individuals with lactose intolerance.

Table 2 .
Scoring results of probiotic samples