Storage Stability of Rebaudioside A in Various Buffer Solutions by Qianyun Gong A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama August 4, 2012 Keywords: rebaudioside A, stevia, stevioside, stability, kinetics Copyright 2012 by Qianyun Gong Approved by Leonard N. Bell, Chair, Professor, Department of Poultry Science Shelly McKee, Associate Professor, Department of Poultry Science Tung-shi Huang, Associate Professor, Department of Poultry Science ii Abstract Rebaudioside A is a non-caloric high intensity sweetener extracted from Stevia rebaudiana. For it to be used in the food industry, rebaudioside A needs to be stable during processing and storage. Kinetic data on its long term stability as affected by solution composition are lacking. The primary objective of this study was to evaluate the storage stability of rebaudioside A in various buffer solutions as a function of pH, buffer type, buffer concentration and temperature. The effect of light exposure on rebaudioside A stability was also evaluated. Rebaudioside A solutions were prepared in 0.02 and 0.1 M phosphate and citrate buffers at pH 3, 5 and 7. Duplicate samples were stored at 20, 30 and 40 ?C. Some samples were stored at room temperature under light or dark conditions. Aliquots were removed nine times for approximately nine months. The concentrations of rebaudioside A were analyzed and pseudo- first-order rate constants with 95% confidence intervals were calculated for the loss of rebaudioside A. In phosphate buffer, the degradation of rebaudioside A was generally faster at higher pH values. The pH effect on rebaudioside A stability was generally reversed in citrate buffer. Rebaudioside A broke down significantly faster in phosphate buffer than in citrate buffer at pH 5 and 7; degradation rates were similar at pH 3. Higher buffer concentrations promoted faster degradation. Rebaudioside A degradation was accelerated by the elevation of temperature. The exposure of light did not have an obvious effect in phosphate buffer at pH 7 while it lowered the stability of rebaudioside A in citrate buffer at pH 3. iii For optimum stability of beverages containing rebaudioside A, lower temperatures and lower buffer concentrations are preferred. If the product has a pH value of 5 or 7, citrate buffer is more preferred than phosphate. Dark environments help stabilize rebaudioside A in beverage at pH 3. iv Acknowledgments The author would like to express tremendous gratitude to Dr. Leonard Bell for his help and support throughout these two years. He dedicated numerous time for conducting the experiment and developing of this thesis. She also wants to express appreciation to her thesis committee, Dr. Shelly McKee and Dr. Tung-shi Huang, for review and recommendations of the thesis as well as their continuous advice and support. Then the author would like to thank her family and friends for loving and supporting her to chase her dream for further study in food science. v Table of Contents Abstract ......................................................................................................................................... ii Acknowledgments........................................................................................................................ iv List of Tables .............................................................................................................................. vii List of Figures ............................................................................................................................ viii Chapter 1 Introduction ................................................................................................................. 1 Chapter 2 Literature Review ......................................................................................................... 2 Introduction ...................................................................................................................... 2 Extraction methods ......................................................................................................... 3 Analysis methods .............................................................................................................. 4 Metabolism ....................................................................................................................... 7 Toxicity study ................................................................................................................... 8 Health effects .................................................................................................................. 11 Food uses and approvals ................................................................................................. 13 Stability ........................................................................................................................... 14 Objective ......................................................................................................................... 17 Chapter 3 Material and Methods................................................................................................. 18 Reagents .......................................................................................................................... 18 Sample Preparation ......................................................................................................... 18 Sampling Procedure ........................................................................................................ 20 vi Sample Analysis.............................................................................................................. 20 Data Analysis .................................................................................................................. 22 Chapter 4 Results and Discussion ............................................................................................... 23 Effect of pH..................................................................................................................... 28 Effect of buffer type and concentration .......................................................................... 29 Effect of temperature ...................................................................................................... 31 Effect of light .................................................................................................................. 33 Comparison to other sweeteners ..................................................................................... 34 Chapter 5 Summary and Conclusion .......................................................................................... 35 References ................................................................................................................................. 36 Appendix ................................................................................................................................... 44 vii List of Tables Table 4.1 Pseudo-first rate constants (d-1) with confidence limits for rebaudioside A stored at 20?C ........................................................................................................................................... 26 Table 4.2 Pseudo-first rate constants (d-1) with confidence limits for rebaudioside A stored at 30?C ........................................................................................................................................... 26 Table 4.3 Pseudo-first rate constants (d-1) with confidence limits for rebaudioside A stored at 40?C ........................................................................................................................................... 27 Table 4.4 Time for 10% rebaudioside A concentration decrease (days) .................................... 27 Table 4.5 Rate constants (d-1) calculated from the data presented by Prakash and others (2012) ..................................................................................................................................................... 29 Table 4.6 Activation Energy (kcal/mol) for rebaudioside A degradation in solution ................ 32 Table 4.7 Pseudo-first rate constants with 95% confidence limits (d-1) for rebaudioside A stored under light or dark at room temperature ..................................................................................... 33 Table 4.8 Predicted rebaudioside A loss (%) in 0.1 M buffer solutions after 1 week at room temperature ................................................................................................................................ 34 viii List of Figures Figure 2.1 Structures of rebaudioside A and stevioside .............................................................. 2 Figure 2.2 Degradation of rebaudioside A at pH 2.6 and 100 ?C modeled using pseudo zero order kinetics. Data from Chang and Cook (1983) .............................................................................. 15 Figure 2.3 Degradation of rebaudioside A at pH 2.6 and 100 ?C modeled using pseudo first order kinetics. Data from Chang and Cook (1983) .............................................................................. 15 Figure 3.1 Chromatograph of rebaudioside A in 0.1 M phosphate buffer at pH 7 and room temperature after 285 days of dark storage (Rebaudioside A eluted at 7.56 min)...................... 21 Figure 4.1 Degradation of rebaudioside A in different buffer solutions at pH 3 and 40?C ........ 24 Figure 4.2 Degradation of rebaudioside A in different buffer solutions at pH 7 and 40?C ........ 24 Figure 4.3 Degradation of rebaudioside A at different temperatures in 0.02 M phosphate buffer at pH 7 ........................................................................................................................................... 25 Figure 4.4 Degradation of rebaudioside A at different temperatures in 0.1 M citrate buffer at pH 3................................................................................................................................................... 25 Figure 4.5 Rate constants of rebaudioside A degradation in 0.1 M phosphate and citrate buffer at 30 ?C and 40 ?C as a function of pH .......................................................................................... 28 Figure 4.6 Arrhenius plots of rebaudioside A degradation in buffer .......................................... 32 1 Chapter 1: Introduction Rebaudioside A is a natural non-caloric sweetener found in Stevia rebaudiana, a plant originating in South America and now commercially cultivated in East Asia (Carakostas and others 2008). This steviol glycoside is 200-300 times sweeter than sucrose (Soejarto and others 1983). Although many steviol glycoside can be extracted from Stevia rebaudiana, rebaudioside A is most commonly used in foods. According to numerous toxicity studies, rebaudioside A does not have reproductive toxicity (Curry and Roberts 2008), genotoxicity (Matsui and others 1996), mutagencity (Nakajima 2000) or carcinogenicity (Xili and others 1992). FDA recognized rebaudioside A as generally recognized as safe (GRAS) in a letter dated December 2008 (Tarantino 2008). Rebaudioside A is starting to be incorporated into beverages and other foods. However, its stability has only been partially investigated. Limited data on this stability indicate pH, temperature and light are important variables for limiting degradation. A systematic study on the effects at buffer type, buffer concentration, pH and temperature has not been reported. Kinetic data (i.e., rate constants) have not been determined. Thus, the objective of this study was to evaluate the long term storage stability of rebaudioside A in different solutions at different storage conditions. 2 Chapter 2: Literature Review Introduction Rebaudioside A is a non-caloric natural sweetener classified chemically as a steviol glycoside, which is extracted and purified from Stevia rebaudiana (bertoni). Stevia rebaudiana originated from South America and has now been cultivated in Asia (Carakostas and others 2008). Stevia rebaudiana extracts consist of 5-10% stevioside, 2-4% rebaudioside A, 1-2% rebaudioside C and other steviol glycosides, like steviolbioside, dulcoside A and rebaudiosides B, D and E (Chabot and Beaulieu 2012). The structures of rebaudioside A and stevioside are shown in Figure 2.1. Rebaudioside A Stevioside Figure 2.1 Structures of rebaudioside A and stevioside 3 Rebaudioside A is a white, crystalline, odorless powder that is freely soluble in water (Carakostas and others 2008). Several steviol glycosides provide sweet tastes but stevioside and rebaudioside A are the predominant sweeteners in Stevia rebaudiana. Rebaudioside A is approximately 200 to 300 times sweeter than sucrose when consumed as a 0.4% solution (Soejarto and others 1983). According to some experts, stevioside and rebaudioside C have some bitterness and unpleasant aftertastes while rebaudioside A has a clean aftertaste (Chen and others 1999). Compared to aspartame, rebaudioside A may have wider usage since it can be consumed by people with the metabolic disease phenylketonuria (Grenby 1991). Phenylketonuria (PKU) is a disease where people cannot metabolize phenylalanine, a component of aspartame. Rebaudioside A does not contain phenylalanine and is therefore safe for people with PKU. Extraction methods There are ten kinds of steviol glycosides in Stevia rebaudiana, with stevioside, rebaudioside A and rebaudioside C being the predominant ones. Rebaudioside A only makes up 2-4% of the dry weight of Stevia rebaudiana. Stevia glycosides are extracted mostly with water or methanol. The extraction procedure itself is not complicated, however the separation and purification of rebaudioside A from stevioside, which has a higher concentration and similar chemical structure as rebaudioside A, is more time and energy consuming. Prakash and others (2007) used hot water (50-60 ?C) to extract steviol glycosides from Stevia rebaudiana leaves followed by filtration. Adsorption resins with food grade methanol or ethanol were used to retain steviol glycosides in the extracted solution and then it was washed by water. The products were then dried, typically by spray or vacuum drying. The limitation for this extraction method is that it does not intend to separate rebaudioside A and stevioside so further purification is necessary. 4 Jaitak and others (2008) extracted steviol glycosides with 80% MeOH and 20% H2O (v/v) for 12 h at room temperature three times. The extract was concentrated under reduced pressure at 50 ?C. This method could not extract rebaudioside A alone and further purification was required. In a later study, Jaitak and others (2009) extracted rebaudioside A and stevioside together from the dry leaves of Stevia rebaudiana by ultrasound and microwave-assisted extraction to speed up the process. Microwave-assisted extraction was shown to be rapid and efficient at 50 ?C and a power level of 80 W with a high breakage of analyte-matrix bonds yet not too powerful to make rebaudioside A and stevioside adsorb on the raw material surface. Microwave assisted extraction with methanol:water (80:20) only took 1 minute at the optimum condition with almost twice the yield of rebaudioside A compared to cold water extraction for 12 hours at 25 ?C and ultrasound extraction for 30 minutes at 35?5 ?C. Like the method of Prakash and others (2007), the primary extraction could not separate rebaudioside A from stevioside. Chen and others (1999) studied the effect of methanol and ethanol used as solvents for the selectivity and enrichment of rebaudioside A. Pyridyl sorbent exhibits higher adsorptive selectivity toward stevioside than rebaudioside A and make the effluent rich in rebaudioside A. Ethanol showed better eluting ability and efficiency but worse selectivity than methanol. They combined the selective adsorption with dynamic chromatographic resolution, whose efficiency was improved by slowing the flow rate and increasing the column length. The concentration of rebaudioside A was enriched by a factor of four under the optimum conditions. Analysis methods The concentration of rebaudioside A has typically been determined using HPLC (high- performance liquid chromatography) in recent years. The effects of several parameters, like 5 column type, column temperature, mobile phase composition and flow rate were studied by many researchers to find the most efficient analytical method. Kitada and others (1989) used the NH2 column at a temperature of 50 ?C with acetonitrile/water (80:20, v/v) as the mobile phase at a flow rate of 0.8 mL/min. The retention time was 14 minutes for rebaudioside A in pickled radish with 93.2% to 100% recoveries. The retention time was shorter at 36 ?C, which means higher column temperatures could shorten the retention time. Kolb and others (2001) also used a NH2 column and acetonitrile/water (80:20, v/v) mobile phase at pH 5 after fast extraction of rebaudioside A by EtOH: H2O (70:30, w/w). The flow rate was set as 2.0 mL/min. The precision of this method was the same as traditional gradient HPLC method, which is 200 mL CHCl3 for 3 h and 200 mL MeOH for 5 h, with less sample preparation time and analysis time. Fan and others (2007) studied the effect of mobile phase composition and NH2 column temperature on retention time. The flow rate was set as 1.0 mL/min and detection wavelength was 205 nm. With acetonitrile/water (80:20, 82:18, 78:22, v/v) as the mobile phase, they indicated that the retention time was longer with the higher organic composition. According to their study, column temperature (43 ?C, 45 ?C, 47 ?C) did not affect the retention time and peak shape. Liquid chromatography was also studied to better separate the several stevia glycosides. Comprehensive two-dimensional liquid chromatography was better than single dimension liquid chromatography. The best separation with a single column was obtained by using a C18 column. A combination of a C18 column followed by a NH2 column could separate all the stevia glycosides from the matrix. The flow rates should be slow for first-dimension separation 6 (maximum 0.1 mL/min), and as fast as possible while not causing too high pressure in the second dimension columns (Pol and others 2007). Wolwer-Rieck and others (2010a) used a Luna HILIC analytical column with a mobile phase of acetonitrile/water (85:15, v/v) and a NH2 column with mobile phase as acetonitrile/water (75:25, v/v). The same flow rate of 1 mL/min and column temperature of 36 ?C were applied. Both of the columns showed the same retention pattern and were suitable for the detection of rebaudioside A. The retention times were 9.7 minutes and 6.6 minutes for the HILIC column and NH2 column, respectively. To extract the rebaudioside A and stevioside, they used another solvent instead of water. Ground stevia leaves were extracted three times in boiling acetonitrile and water (8:2 v/v) for 30 min and centrifuged after cooling to room temperature. To better separate these two compounds, solid-phase extraction was attempted. Wolwer-Rieck and others (2010b) used a Luna HILIC column at 36 ?C with a mobile phase of acetonitrile/water (80:20 v/v) for their HPLC analysis of rebaudioside A in soft drinks. The absorption wavelength for detection was set for 210 nm. The injection volume was 20 ?L and the flow rate was 1.0 mL/min. The retention time was 10.5 minutes and the recovery ranged from 95.9% to 109.2%. To separate rebaudioside A from other steviol glycosides, Liu and others (2011) and Li and others (2012) both used mixed-mode macroporous adsorption resins (MAR). In Liu and others? study, four tyrene divinyl-benzenes with different polarity, particle size and specific surface area, pore size and moisture content were tested. For a single MAR, a larger pore size gave a higher purity of rebaudioside A because stevioside diffused easily into the pores. A larger specific surface area lowered the recovery of rebaudioside A. However, the ideal purity and recovery of rebaudioside A could not be obtained using a single MAR, but a mixture of them 7 could increase the purity of rebaudioside A from 40.77% to 60.53% after one single run. In Li and others? (2012) study, 19 kinds of tyrene divinyl-benzenes were tested. Some combinations of MAR were able to increase the purity of the obtained product from 60% to 97%. Metabolism In anaerobic conditions, like the human digestive system, rebaudioside A degraded into stevioside and then to steviol; bacterial enzymes are not able to cleave it into further products. Steviol can be quickly converted to its glucuronide, and is excreted by the kidneys in humans (Carakostas and others 2008). An in vitro experiment showed that stevioside could be degraded completely to steviol after 10 hours under strict anaerobic conditions. Rebaudioside A could also be completely degraded to steviol, but it would take longer, 24 hours, in rats? intestinal microflora (Wingard and others 1980). Using human intestinal microflora (37 ?C for 72 h) from different volunteers, rebaudioside A and stevioside were degraded to steviol; no other derivatives were found. After an initial lag phase of 6-7 h, rebaudioside A was hydrolyzed to steviolbioside and this was rapidly converted to steviol, which remained unchanged during the 72 h incubation. Rebaudioside A showed a weak inhibitory activity on aerobic bacteria and particularly on coliforms (Gardana and others 2003). In the gastrointestinal track, none of the digestive enzymes from humans or the acidic environment of the stomach were able to degrade stevioside into steviol, and steviol was the only metabolite found in the feces. Very low stevioside concentrations were found in blood plasma (Geuns and others 2007). In the cecum, stevioside was metabolized to steviol by the bacterial flora. Steviol was found in the blood with its maximum concentration occurring after 8 h. In an experiment with rats, little or no stevioside was absorbed into the blood (Nakayama and others 8 1986). In the colon, steviol was metabolized by bacteria similarly in rats and humans, although the rate of metabolism and uptake in rats appeared to be slightly faster (Koyama and others 2003a,b). Roberts and Renwick (2008) confirmed the previous work on stevioside and demonstrated that rebaudioside A is metabolized in the same pathway as stevioside in both rats and humans. Although the metabolic pathways of stevioside and rebaudioside A are similar in humans and rats, the excretion of steviol from them is different. The excretion of steviol glucuronide in rats occurs primarily in the feces, while it is mainly eliminated through urinary excretion in humans. This is due to the molecular weight thresholds for biliary excretion being different in humans and rats. In rats, very little steviol is found beyond the portal or biliary systems, while in humans steviol is quickly converted to its glucuronide, which is a stable detoxification product that is quickly excreted by the kidneys (Geuns and others 2006). Toxicity study Because rebaudioside A and stevioside have similar chemical structures and metabolism studies indicate that steviol glycosides are metabolized into steviol in the human body, the results from studies regarding about the physiological effects and toxicity of stevioside and steviol may be used to evaluate the safety of rebaudioside A (Roberts and Renwick 2008). Steviol equivalents were used in several studies to compare intake and safety limits. Because the molecular weight of rebaudioside A is three times larger than steviol, the safety limitation of rebaudioside A is about three times higher than the one for steviol when expressed by weight. A study conducted on mice, rats and hamsters on the acute toxicity of stevioside showed that the LD50 level of steviol was 5.20 g/kg body weight and 6.10 g/kg body weight for male and female hamsters, respectively. For rats and mice, the LD 50 level was as large as 15.0 g/kg 9 body weight for both genders, which indicates that the hamster was most sensitive to stevioside in this study and that steviol, stevioside and rebaudioside A have no acute toxicity (Toskulkao and others 1997). For comparison, the minimal lethal dose of aspartame in mice, rats and rabbits is greater than 5 g/kg body weight (Molinary 1984). Macroscopic and microscopic examinations showed that there were no changes to the renal or reproductive systems in rats after 90 days at an oral dose of 25,000 and 50,000 ppm rebaudioside A, which suggests rebaudioside A has virtually no subchronic toxicity. Significant weight loss was observed, which the author concluded was not an adverse side effect but was caused by a lower energy density since rebaudioside A was a diet supplement with no calories (Curry and Roberts 2008). Research on Sprague-Dawley rats for 90 days showed there were no treatment-related effects on the general condition and behavior of the animals as determined by clinical observations, functional observational battery, and locomotors activity assessments at doses of 500, 1000, and 2000 mg rebaudioside A/kg bw/day (purity 99.5% treatment) (Nikiforov and Eapen 2008). A study about the toxicity of a Stevia rebaudiana extract to the renal system has shown it induced systematic and renal vasodilation, hypotension and diuresis after 40 and 60 days oral administration in Wistar rats. Because the exact extract composition was not shown in this experiment, it is difficult to conclude whether rebaudioside A, stevioside or another compound caused these effects (Melis 1995). As for the genotoxicity of rebaudioside A and stevioside, several studies were conducted. In vitro, in vivo, mutation, chromosome damage, and DNA strand breakage experiments showed no evidence of genotoxic damage relevant to human health (Brusick 2008). In a study with 10 several mutagenicity tests using bacteria, cultured mammalian cells and mice, stevioside was not shown to be mutagenic (Matsui and others 1996). Nunes and others (2007) showed that 4 mg stevioside/mL in drinking water for 45 days produced DNA breakage in rat blood cells, spleen, liver and brain. Unfortunately, no positive control was provided in this study and the significant elevations of blood cell nuclei number only occurred in week 5, not in the previous 4 weeks. Stomach, colon, liver, kidneys, bladder, lung, brain and bone marrow cells were sampled and tested after 3 and 24 hours of exposure to stevia as high as 2000 mg/kg in mice; these tests produced evidence indicating no increase in DNA damage (Sasaki and others 2002). Studies have been conducted on the mutagenicity of rebaudioside A and consistent negative results were provided (Brusick 2008; Williams and Burdock 2009). Nakajima?s (2000) micronucleus formation experiment on BDF1 mouse bone marrow with 200-2000 mg/kg bw/day for two days suggested rebaudioside A did not have mutagenic toxicity. Rebaudioside A was used in four salmonella strains, where there was no mutagenic response at even the highest level of treatment (Williams and Burdock 2009). Because the stevia plant as a whole has been used historically as an oral contraceptive in Brazil and Paraguay, there have been questions about its effect on fertility. Older studies, which used a crude stevia aqueous extract, had reported effects on the testes eight, reduction in spermatozoa concentration and other fertility defects while more recent studies, which used purified stevioside, have shown different results. Yodyingyuad and Bunyawong (1991) claimed stevioside showed no toxic effect on developing hamsters (2500 mg/kg bw/day). Both female and male rats were treated during three rounds of mating and neither the fertility, number of offspring nor the reproductive tissue was affected (Yodyingyuad and Bunyawong 1991). Usami and others (1995) verified their results about developmental toxicity at lower intakes (1000 11 mg/kg bw/day) of stevioside. Curry and others (2008) showed there were no treatment-related effects of rebaudioside A on either F0 or F1 generations up to 25,000 ppm in Wistar rats on reproductive performance (mating performance, fertility, gestation lengths, estrus cycles, or sperm motility, concentration, or morphology). No developmental defects were noticed in the offspring. Studies have also failed to show any evidence of carcinogenicity. After an oral intake of 85% pure stevioside (600 mg/kg bw/day) for over 24 months, no neoplastic or pre-neoplastic lesions were reported in any Wistar rat tissue (Xili and others 1992). No lesions on any organ or tissue were reported on F334 rats during a 104-week test, which indicated stevioside was not carcinogenic. However, a significant decrease in survival rates was observed in male rats consuming a 5% dose (Toyoda and others 1997). The Joint FAO/WHO Expert Committee on Food Additives (JECFA) used the 970 mg/kg bw/day dose (2.5% dose in male rats) in the previous study to set the temporary ADI for steviol at 12 mg/kg bw/day (Carakostas and other 2008). The reproductive, carcinogenicity, mutagenicity, and general toxicity studies have demonstrated rebaudioside A and stevioside appear safe at even high dietary intake levels. Based on these results and the historical use of stevia in some cultures, its use in food has been approved by several government agencies, as will be discussed later. Health effects Clinical studies about the effect of rebaudioside A on blood pressure and blood sugar levels in healthy humans and patients with hypertension and diabetes have been conducted. Several clinical studies have shown rebaudioside A could offer therapeutic benefits to hypertensive patients. 12 Chan and others (2000) conducted a study with hypertensive patients who were taken off their antihypertensive medications and treated with stevioside (750 mg/day) or placebo for 12 weeks. The results showed that both systolic and diastolic blood pressure decreased significantly in the stevioside group and this effect persisted during the whole year. Ferri and others (2006) showed no effect of 3.75 mg/kg/day (7 weeks), 7.5 mg/kg/day (11 weeks) and 15.0 mg/kg/day (6 weeks) of a crude steviol glycoside extract on the blood pressure of subjects with mild essential hypertension. This might be due to the different intake levels of stevioside and the fact that the second research used crude steviol glycoside instead of one with higher purity. Rebaudioside A?s effect on patients with hypertension made researchers wonder whether it has any effect on the blood pressure of health people. A study on people with normal blood pressure was conducted with rebaudioside A intakes of 1000 mg/day for 4 weeks. The resting seated systolic blood pressure, diastolic blood pressure, mean arterial pressure, heart rate, and 24-hour ambulatory blood pressure responses in healthy humans were not significantly altered as compared to the placebo group (Maki and others 2008a). There were numerous studies about the effect of rebaudioside A on diabetic animals and patients. Tests in Goto-Kakizaki rats with type 2 diabetes and normal Wistar rats showed stevioside could suppress the glucagon level and increase the insulin response, which suggested its potential use in diabetes treatment (Jeppesen and others 2002). Abudula and others (2004) showed rebaudioside A stimulated insulin secretion dose-dependently with the presence of extracellular calcium ion in mice and might serve as a potential type 2 diabetes treatment. In type 2 diabetic patients, 16 weeks of consuming 1,000 mg of rebaudioside A daily did not affect glucose homeostasis or blood pressure (Maki and others 2008b). Barriocanal and others (2008) expanded the sample size to both type 1 and type 2 diabetic patients with 750 mg/day steviol 13 glycosides intake. No significant hemodynamic effects in subjects with or without diabetes mellitus were detected and there was no effect of steviol glycosides on blood lipids (total-, LDL-, HDL-cholesterol). Food uses and approvals Extracts of Stevia rebaudiana have been used for several years for sweetness in Brazil, Japan, China and Korea (Geuns and others 2003). In Japan and Paraguay, stevia was also consumed as a food and medicine (Carakostas and others 2008). Due to the growing concern of caloric intake, rebaudioside A has been used more as a high intensity sweetener. Its use is rapidly growing in the food industry, especially in beverages. It is used as an ingredient in vitamin water zero, carbonated beverages, yogurt, and orange juice beverages for the sweetness. Rebaudioside A can also be used as a table-top sweetener. The Joint Expert Committee on Food Additives (JECFA) approved a temporary 0-2 mg/kg bw/d for steviol intake at the 63rd WHO meeting (WHO 2005). At the 69th meeting, ADI values of 0-4 mg/kg bw/day for steviol intake were approved (FAO/WHO 2008), which is equivalent to 0-12 mg/kg bw/day for rebaudioside A. The Food Standards Australia New Zealand (FSANZ) has completed the evaluation of an application of steviol glycoside in food and allowed its use (FSANZ 2008). At least two petitions seeking authorization to use stevioside and steviol glycoside in foods have been submitted since 1989. FDA issued "no objection" letters for the generally recognized as safe (GRAS) notification of rebaudioside A in late 2008. This GRAS approval was based on rebaudioside A being incorporated ?under the conditions of its intended use? which would be ?largely self-limiting due to its organoleptic properties? (Tarantino 2008). 14 Stability The stability of rebaudioside A has been studied for several decades and is important to understand when using it as a sweetener in foods and beverages. The stability of rebaudioside A is affected by the storage form, time, temperature, pH and light exposure. Chang and Cook (1983) examined the stability of rebaudioside A in various solutions. The concentration of rebaudioside A decreased 31.5% in water at 100 ?C for 48 h and 76-87% in acid solutions after 13 hours, depending upon the type of acid. Degradation was faster in phosphoric acid at pH 2.4 than citric acid at pH 2.6. No significant changes were discovered after 4 months at 4 ?C, 3 months at room temperature (22 ?C) nor 1 month at 37 ?C in either citric or phosphoric acid carbonated beverages. They also evaluated the photo stability of rebaudioside A in citric acid and phosphoric acid carbonated beverages. After storing outdoors for a week (3000 langleys of sunlight exposure) at 10-25 ?C, rebaudioside A decreased by 22% and 18% in the phosphoric and citric acid beverages, respectively. The buffer concentration and pH were not reported for the carbonated beverages. Rate constants were not calculated from the data. Using the degradation data presented by Chang and Cook (1983), kinetic plots can be constructed. Figures 2.2 and 2.3 show the degradation of rebaudioside A in the form of pseudo- zero order and pseudo-first order kinetic plots. From the linear fit based on R2 values, it appears pseudo first order kinetics is more appropriate for modeling the degradation of rebaudioside A. 15 Figure 2.2 Degradation of rebaudioside A at pH 2.6 and 100 ?C modeled using pseudo zero order kinetics. Data from Chang and Cook (1983). Figure 2.3 Degradation of rebaudioside A at pH 2.6 and 100 ?C modeled using pseudo first order kinetics. Data from Chang and Cook (1983). Kroyer (1999) studied the effect of pH on the stability of stevioside. Stevioside was stable from pH 2-10 at 60 ?C for 1 h with only slight losses at pH 2 and 10. Based on the R? = 0.9438 0 20 40 60 80 100 120 0 5 10 15Time (hours) Reba udi osi de A R em aini ng (% ) R? = 0.9968 0.00 1.00 2.00 3.00 4.00 5.00 0 5 10 15 Ln (Re bau dio sid e A Rem ain ing ) Time (hours) 16 structural similarities between stevioside and rebaudioside A, the effect of pH on rebaudioside A stability may be similar to that of stevioside. Clos and others (2008) questioned the photo stability results presented by Chang and Cook (1983) and repeated their experiment. The samples were prepared similarly and stored under similar conditions at slightly higher temperatures (18-34 ?C). They concluded that rebaudioside A stability was not affected by light. However, their data actually show 5 times greater rebaudioside A loss when stored in light than darkness. Prakash and others (2008) reviewed the stability of rebaudioside A in different food environments other than beverages. It was stable in yogurt during pasteurization (190 ?F for 5 min) and fermentation as well as for 6 weeks of storage at 4 ?C. They had also claimed rebaudioside A was very stable when stored as powder as it only had <5% of loss after 24 months. In the process of baking (350 ?F for 20?25 min) and storing (25 ?C and 60% RH) of white cake, no significant losses of rebaudioside A were found. A later study showed the stability of rebaudioside A increased with increasing pH values in a caffeinated soft drink (pH 2.4), a lemon- lime flavored soft drink (pH 2.7) and an energy drink (pH 3.5) at 80 ?C, which agreed with data on stevioside (Kroyer 1999). The highest degradation was observed in caffeinated soft drinks after 72 h of storage, while the lowest was in the energy drinks. They also concluded rebaudioside A was more stable against acid hydrolysis compared to stevioside. By using LC-MS, they claimed the degradation products of rebaudioside A were rebaudioside B and steviolmonoside (Wolwer-Rieck and others 2010b). Prakash and others (2012) studied the degradation products of rebaudioside A in acidic mock beverages, like cola, lemon-lime and root beer soft drinks with pH values from 2.8 to 4.2. The storage temperatures were 5 ?C, 20 ?C, 30 ?C and 40 ?C. Degradation was similar at each of 17 the conditions and occurred after 26-weeks of storage. Excellent mass balance was achieved at all conditions. Rebaudioside A degraded into rebaudioside B and other products faster at lower pH values and high temperatures. The migration of exocyclic double bond between C16/C17 to C15/C16 can form a degradation product and the hydrolysis of the glucose unit at C19 formed rebaudioside B. Chaturvedula and Prakash (2011) studied the degradation of rebaudioside A and stevioside from acid and alkaline hydrolysis. The structures of degradation products were acquired by NMR, high resolution mass spectral (HRMS) data and comparative spectral data. The alkaline hydrolysis of rebaudioside A conducted by NaOH (2.2 mol/L) yielded rebaudioside B from the cleavage of the ?-D-glucopyranosyl unit at the C-19 position while the only product from stevioside was steviolbioside. Acid hydrolysis, applied by H2SO4 (5%), could furnish D- glucose from rebaudioside A and acquire the same degradation products. Objective Numerous studies have been conducted on the short term stability of rebaudioside A under different conditions. However, no study has actually collected kinetic data to calculate the rate constants for rebaudioside A degradation. The objective of this study was to evaluate the long term stability of rebaudioside A in solutions (in terms of its degradation rate constants) as a function of buffer type, buffer concentration, pH and temperature. The effect of light exposure was also studied as a secondary objective. 18 Chapter 3: Materials and Methods Reagents Rebaudioside A was obtained from Sigma-Aldrich (St. Louis, MO). Sodium phosphate dibasic, sodium phosphate monobasic, citric acid and sodium citrate were acquired from Fisher Scientific (Fair Lawn, NJ). HPLC grade acetonitrile was bought from VWR International (Suwanee, GA). Sample Preparation Twelve buffer solutions at different concentrations and pH values were prepared in this study. Six of them were sodium phosphate buffers and the other six were sodium citrate buffers. The buffer concentrations were 0.02 M and 0.1 M, and each was prepared to pH 3, 5 and 7. Each of the 12 buffer solutions was stored at three temperatures to give a total of 36 experimental systems. The phosphate buffer solutions were made using the following procedure. Phosphoric acid (2.88 g/85%) was added to 250 mL water to make a 0.1 M solution. Similarly, a 0.1 M sodium phosphate monobasic solution was made by adding 6.90 g monobasic sodium phosphate to 500 mL water. A 0.1 M dibasic sodium phosphate solution was also made by adding 7.10 g dibasic sodium phosphate to 500 mL water. These solutions were mixed in varying proportions to yield 0.1 M phosphate buffer solutions at pH 3, 5 and 7. The 0.02 M phosphate buffers were made using the same method, but with one-fifth of the buffer salt amounts. 19 The citrate buffer solutions were made by mixing appropriate volumes of citric acid and trisodium citrate solutions to acquire pH 3, 5 and 7 buffer solutions. The 0.02 M citric acid solution was prepared by mixing 1.92 g citric acid and 500 mL water, while the 0.02 M sodium citrate solution was made by dissolving 2.94 g trisodium citrate into 500 mL water. The two 0.1 M solutions were made by adding 14.7 g sodium citrate and 9.6 g citric acid to 500 mL water. Approximately 40 mg of rebaudioside A were dissolved into 100 mL of each buffer solution. The solution was filtered into a 100-mL sterile septum-containing glass bottle using a sterile syringe with a sterile 0.20 ?m nylon filter and a sterile needle. Using a new sterile syringe and needle, 1-2 mL aliquots were aseptically transferred into eighteen 2-mL sterile septum- containing vials per experiment. This protocol minimized microbial contamination of the samples. For the storage stability study, samples were labeled according to their buffer concentration, buffer type, pH and temperature. For example, a 0.1 M phosphate buffer sample at pH 3 and 40 ?C was labeled ?1p340? while a 0.02 M citric buffer sample at pH 7 and 20 ?C was labeled as ?02c720?. The samples used for the photo stability study were labeled according to the buffer concentration, buffer type, pH value and storage condition. For example, 0.1 M phosphate buffer at pH 7 stored under light exposure was labeled ?1p7-L? and 0.1 M citrate buffer at pH 3 stored in dark was labeled ?1c3-D?. The eighteen vials containing samples at each buffer type, buffer concentration, and pH were placed into each of three incubators set at 20, 30 and 40 ?C. Thermometers monitored the internal temperature of the incubators. These samples were stored in darkness and were used to determine the storage stability of rebaudioside A as affected by buffer type, buffer concentration, pH and temperature. 20 The buffer solutions used for the photo stability component were limited to 0.1 M phosphate buffer at pH 7 and 0.1 M citrate buffer at pH 3. These two solutions were selected to represent extremes in the degradation behavior. These solutions contained approximately the same amount of rebaudioside A as the storage stability experimental groups. These samples were all stored at room temperature. The light exposure groups were placed under ambient room light (scattered sunlight, occasional fluorescent lighting) while the dark protected groups were stored in a paperboard box at the same temperature. Sampling Procedure Duplicate samples were removed from storage 9 times for approximately 9 months. For example, the samples in 0.02 M phosphate buffer at 40?C were removed at day 0, 29, 60, 90, 121, 152, 194, 252 and 285. Samples were shaken well before sampling. The exact date of sampling was recorded. An aliquot was removed from each vial using an HPLC syringe, which was injected for analysis. Mold growth was observed in two 20 ?C groups, 02p320 and 1c320, after two months of storage. These solutions were remade; approximately 30 mL were placed into 100-mL sterile bottles due to the unavailability of the 2-mL vials. During the same storage periods, aliquots (1 mL) were removed from these bulk solutions through the rubber septum using a sterile needle and syringe for analysis. Sample Analysis Rebaudioside A concentrations in the experimental solutions were determined using reverse-phase high performance liquid chromatography (HPLC). Two sets of standard solutions were prepared. Standard solution A was made by dissolving 82.1 mg of rebaudioside A in 100 mL deionized water while solution B was made by dissolving 61.1 mg of rebaudioside A in 100 21 mL deionized water. Standard solutions were prepared by serially diluting aliquots these solutions three times. The concentrations of experimental samples were acquired by comparing the peak areas to those from the standard solutions. Standard solutions were analyzed every time experimental samples were tested. The analytical column used was a 250 x 4.6 mm LUNA 5? amino column (Phenomenex, Torrance, CA) with a corresponding guard cartridge. The column was housed in a column heater set at 45?C. The mobile phase consisted of 77.5/22.5 (v/v) acetonitrile/aqueous buffer at pH 7; the buffer consisted of 0.0008 M monobasic sodium phosphate and 0.0015 M dibasic sodium phosphate. The flow rate was set at 2.0 mL/min. The injection volume was 20 ?L, and detection occurred at a wavelength of 210 nm. Data were integrated by a Hewlett-Packard integrator. The retention time for rebaudioside A was around 7.5 min. A sample chromatogram is shown in Figure 3.1. Using the rebaudioside A standard curves every time, the concentrations of rebaudioside A in the experimental solutions were determined. Figure 3.1 Chromatograph of rebaudioside A in 0.1 M phosphate buffer at pH 7 and room temperature after 285 days of dark storage (Rebaudioside A eluted at 7.56 min) 22 Data Analysis Pseudo-first-order rate constants with 95% confidence intervals were calculated for the loss of rebaudioside A using linear least squares analysis as described by Labuza and Kamman (1983). Significant differences between the rate constants were analyzed by testing the homogeneity of regression at p<0.05, as described by Steel and Torrie (1980). 23 Chapter 4: Results and Discussion Rebaudioside A does have the potential to breakdown in solutions during storage, as shown by the degradation profiles in Figures 4.1-4.4. The degradation of rebaudioside A was affected by buffer type, buffer concentration, pH and temperature. From the degradation profiles, in the form of pseudo-first order kinetic plots, the rate constants were determined and are listed in Tables 4.1, 4.2 and 4.3. Table 4.4 shows the time required for the rebaudioside A concentration to decrease by 10%. The 10% loss time was calculated from the rate constants in the former tables. As rebaudioside A degrades, various degradation products are produced. Chang and Cook (1983) reported the formation of rebaudioside B and glucose from the cleavage of the ester linkage of rebaudioside A. Prakash and others (2008) presented a degradation pathway that included rebaudioside B and stevioside as well as a few other derivatives. Wolwer-Rieck and others (2010b) detected rebaudioside B and steviolmonoside as degradation products. Chaturvedula and Prakash (2011) reported the formation of rebaudioside B from alkaline degradation of rebaudioside A. Prakash and others (2012) reported rebaudioside B was formed along with a more prevalent product determined to be 13-[(2-O-?-D-glucopyranosyl-3-O-?-D- gluopyranosyl-?-D-glucopyranosyl)oxy] ent-kaur-15-en-19-oic acid ?-D-glucopyranosyl ester. In the current study, only rebaudioside A concentrations were determined. 24 Figure 4.1 Degradation of rebaudioside A in different buffer solutions at pH 3 and 40?C Figure 4.2 Degradation of rebaudioside A in different buffer solutions at pH 7 and 40?C 2.8 3.0 3.2 3.4 3.6 3.8 4.0 0 50 100 150 200 250 300 0.02 M Citrate 0.1 M Citrate 0.02 M Phosphate 0.1 M PhosphateLn (Rebaudi osi de A conc ent rat ion ) Time (Days) 2.8 3.0 3.2 3.4 3.6 3.8 4.0 0 50 100 150 200 250 300 0.02 M Citrate 0.1 M Citrate 0.02 M Phosphate 0.1 M Phosphate Ln (Rebaudi osi de A conc ent rat ion ) Time (Days) 25 Figure 4.3 Degradation of rebaudioside A at different temperatures in 0.02 M phosphate buffer at pH 7 Figure 4.4 Degradation of rebaudioside A at different temperatures in 0.1 M citrate buffer at pH 3 2.8 3.0 3.2 3.4 3.6 3.8 4.0 0 50 100 150 200 250 300 20?C 30?C 40?C Ln (Rebaudi osi de A conc ent rat ion ) Time (Days) 2.8 3.0 3.2 3.4 3.6 3.8 4.0 0 50 100 150 200 250 300 20?C 30?C 40?CLn (Rebaudi osi de A conc ent rat ion ) Time (Days) 26 Table 4.1 Pseudo-first rate constants (d-1) with 95% confidence limits for rebaudioside A degradation in buffers stored at 20?C pH 3 pH 5 pH 7 0.02 M phosphate 0.0000469?0.000172 aA 0.000301?0.000131 aB 0.000388?0.000109 aB 0.1 M phosphate 0.000226?0.000102 aA 0.000270?0.000115 abA 0.000953?0.000147 bB 0.02 M citrate 0.000138?0.000192 aA 0.000107?0.000186 bcA 0.000247?0.000175 acA 0.1 M citrate 0.0000102?0.000224 aA 0.000109?0.000111 cA 0.0000604?0.000126 cA Different capital letters represent significant differences within the same row (P<0.05). Different lower case letters represent significant differences within the same column (P<0.05). Table 4.2 Pseudo-first rate constants (d-1) with 95% confidence limits for rebaudioside A degradation in buffers stored at 30?C pH 3 pH 5 pH 7 0.02 M phosphate 0.000286?0.0000837 aA 0.000377?0.000168 aAB 0.000547?0.000137 aB 0.1 M phosphate 0.000228?0.0000878 aA 0.000384?0.000136 aB 0.00161?0.000122 bC 0.02 M citrate 0.000332?0.000109 aA 0.000124?0.000113 bB 0.000137?0.0000972 cB 0.1 M citrate 0.000242?0.000121 aA 0.000243?0.000115 abA 0.000133?0.000137 cdA Different capital letters represent significant differences within the same row (P<0.05). Different lower case letters represent significant differences within the same column (P<0.05). 27 Table 4.3 Pseudo-first rate constants (d-1) with 95% confidence limits for rebaudioside A degradation in buffer stored at 40?C pH 3 pH 5 pH 7 0.02 M phosphate 0.000579?0.000253 aA 0.000544?0.000241 aA 0.000802?0.000153 aA 0.1 M phosphate 0.000773?0.000147 abA 0.00138?0.000180 bB 0.00259?0.000221 bC 0.02 M citrate 0.000893?0.000192 bA 0.000229?0.0000995 cB 0.000255?0.000140 cB 0.1 M citrate 0.000930?0.000267 abA 0.000581?0.000188 aB 0.000212?0.000147 cC Different capital letters represent significant differences within the same row (P<0.05). Different lower case letters represent significant differences within the same column (P<0.05). Table 4.4 Time for 10% rebaudioside A concentration decrease (days) pH 3 pH 5 pH 7 20 ?C 0.02 M phosphate 2246 350 272 0.1 M phosphate 466 390 111 0.02 M citrate 763 985 427 0.1 M citrate 10329 967 1744 30 ?C 0.02 M phosphate 368 279 193 0.1 M phosphate 462 274 65 0.02 M citrate 317 850 769 0.1 M citrate 435 434 792 40 ?C 0.02 M phosphate 182 194 131 0.1 M phosphate 136 76 41 0.02 M citrate 118 460 413 0.1 M citrate 113 181 497 28 Effect of pH In phosphate buffer, the rebaudioside A degradation rate constants generally increased as pH increased (Tables 4.1-4.3). However in citrate buffer, rebaudioside A degradation rate constants generally decreased as pH increased at 30 and 40 ?C and were not significantly different at 20?C. Figure 4.5 shows these combined effects of pH and buffer type in 0.1 M buffer. Clearly, in addition to pH, the buffer type is affecting degradation rates, as will be discussed later. Figure 4.5 Rate constants of rebaudioside A degradation in 0.1 M phosphate and citrate buffer at 30 ?C and 40 ?C as a function of pH In the study Chang and Cook (1983) conducted, rebaudioside A had a greater extent of loss in a phosphoric acid solution at pH 2.4 than in a pH 2.6 citric acid solution at 100 ?C for 13 h. The differences in degradation could be due to the different pH values, buffer solution compositions, or a combination of both. Their study demonstrates the importance of controlling multiple variables in order to obtain a clear understanding of the factors affecting the reaction rate. 0 5 10 15 20 25 30 3 5 7 phosphate, 40 ?C citrate, 40 ?C phosphate, 30 ?C citrate, 30 ?C pH Rat e cons tant *1000 0 ( d-1 ) 29 Using data presented by Prakash and others (2012), rebaudioside A degradation rate constants were determined (Table 4.5). At all three temperatures, degradation rate constants decreased as pH increased from 2.8 to 4.2. The buffer solutions they used were made with trisodium citrate and acidified with phosphoric acid. Their results are generally in agreement with those obtained in the current study involving citrate buffer. Table 4.5 Rate constants (d-1) calculated from the data presented by Prakash and others (2012) pH 2.8 pH 3.2 pH 3.8 pH 4.2 20?C 0.0004 0.00007 0.00008 0.00001 30?C 0.0013 0.0006 0.0001 0.00007 40?C 0.0053 0.0025 0.0007 0.0004 Wolwer-Rieck and others (2010b) conducted an accelerated shelf-life study and reported that the concentration of rebaudioside A decreased by 54% in a caffeinated soft drink at pH 2.4 after being stored for 72 h at 80?C while its concentration decreased by 28% in an energy drink at pH 3.5. Although the beverage compositions were not clear, the trend that the stability of rebaudioside A was improved in higher pH environments is again consistent with the citrate buffer data obtained from the current study. Effect of buffer type and concentration In low pH environments (pH 3), buffer type did not affect the degradation rates of rebaudioside A at 20 and 30 ?C. Rate constants for rebaudioside A degradation in the two 0.1 M buffers were also not different at 40?C (Table 4.1-4.3). Slightly faster degradation was observed in 0.02 M citrate buffer than in 0.02 M phosphate buffer, but the general trend was very little effect of buffer type on the rate constants at pH 3. Likewise, buffer concentration (0.02 M versus 30 0.1 M) had little to no effect on the degradation rate constants of rebaudioside A at pH 3. These results are not consistent with a previous study where rebaudioside A stored at 22 ?C and 37 ?C appeared more stable in citric acid than in phosphoric acid beverages (Chang and Cook 1983). The study by Chang and Cook (1983) only had three data points, the number of replicates was not specified, and no statistical analysis was presented making their results questionable. At higher pH values, an effect of buffer type and concentration became apparent (Tables 4.1-4.3). At pH 5 and 7, rebaudioside A broke down significantly faster in phosphate buffer than in citrate buffer in ten out of twelve experimental group comparisons while the two other groups (0.02 M buffer with pH 7 at 20 ?C and 0.1 M buffer with pH 5 at 30 ?C) showed the same trend but were not significant different. In addition, higher concentrations of phosphate buffer yielded larger degradation rate constants at pH 7 whereas the concentration of citrate buffer at pH 7 had no effect on the rate constants. The different effects of buffer type and concentration on rebaudioside A degradation may be due to different hydrolysis mechanisms at the different pH levels. At pH 3, rebaudioside A degradation appears to be occurring via specific acid hydrolysis, where the hydronium ion (H3O+) catalyzes the cleavage, irrespective of buffer type and concentration. Therefore, buffer type and concentration have little effect on the degradation rate constant. However at pH 7, there appears to be some general acid-base catalysis occurring. At pH 7, degradation was faster in phosphate buffer than citrate buffer and faster at higher concentrations of phosphate buffer, meaning that buffer species other than H3O+ or OH- were affecting the reaction. The concentration of the phosphate dibasic anion (HPO4-2) increases as pH increases from approximately pH 5 to 10 (Christian 1980). This anion has been previously linked to enhanced degradation rates for aspartame (Bell and Wetzel 1995), thiamin (Pachapurkar and Bell 2005), 31 and tagatose (Dobbs and Bell 2010). The role of phosphate buffer as a catalyst was also found in the Maillard reaction (Bell, 1997). It appears that the phosphate dibasic anion is also facilitating the necessary proton transfers to hydrolyze rebaudioside A. Although some degradation products have been identified, the exact degradation mechanisms for rebaudioside A have not been described. Current data suggest the effect of buffer on rebaudioside A degradation depends upon the pH. More research is needed to better understand the role of buffer salts on the mechanism of rebaudioside degradation. Effect of temperature As expected, rebaudioside A degradation rate constants increased as temperature increased (Tables 4.1-4.3), except for degradation in 0.02 M citrate buffer at pH 7 and 20 ?C. Because this rate constant was larger than in the same system at 30 ?C, the activation energy could not be reliably detected. Activation energy is the energy needed to be overcome in order for a chemical reaction to occur. Higher activation energy means the reaction is more sensitive to temperature changes. In this study, activation energies for rebaudioside A degradation in different solutions were calculated from the slope of Arrhenius plots (Figure 4.6). The activation energies are listed in Table 4.6 and represent the effect of temperature on the stability of rebaudioside A. 32 Figure 4.6 Arrhenius plots of rebaudioside A degradation in buffer solutions Table 4.6 Activation energy (kcal/mol) for rebaudioside A degradation in solution pH 3 pH 5 pH 7 phosphate 0.02 M 23 5.4 6.6 0.1 M 11.1 14.8 9.1 citrate 0.02 M 17 6.9 * 0.1 M 41.3 15.2 11.5 *unreliable because rate constant at 20 ?C was larger than 30 ?C With the exception of degradation in 0.1 M citrate buffer at pH 3, activation energies ranged from 5-23 kcal/mol (Table 4.6). No definitive trends were observed with respect to buffer type, buffer concentration or pH. Potentially different degradation mechanism (specific acid catalysis at pH 3 vs general acid-base catalysis at pH 7) may explain the lack of observable trends. Our values at pH 3 (11-41 kcal/mol) are similar to the 20-34 kcal/mol calculated from previously published data for rebaudioside degradation in mock beverages using citrate and phosphate buffers at pH 2.8-4.2 (Prakash and others, 2012). Other activation energy values -12.0 -10.0 -8.0 -6.0 0.0030 0.0032 0.0034 0.0036 0.1 M citrate buffer, pH 3 0.02 M phosphate buffer, pH 7 1/T ln (K) 33 include 15-30 kcal/mol for aspartame degradation (Bell and Labuza 1991) and 12-17 kcal/mol for thiamin degradation (Pachapurkar and Bell 2005). Collecting kinetic data at additional temperatures would improve the reliability of the activation energies. Effect of light Light did not appear to play an important role in the degradation of rebaudioside A in phosphate buffer at pH 7 (Table 4.7). On the other hand, rebaudioside A in citrate buffer at pH 3 broke down almost ten times faster under light exposure than that stored in darkness. The results of this study somewhat agree with those of Chang and Cook (1983), who found sunlight enhanced the loss of rebaudioside A. However, they noted the amount of rebaudioside A decreased 18-22%, which is much higher than the amount lost in this study (Table 4.8). The results of this study do not agree with the conclusions made by Clos and others (2008); they claimed rebaudioside A was stable to light exposure. However, as mentioned in the literature review, a closer examination of their data actually reveals that there was five times less rebaudioside A after exposure to light for 1 week. Based on the current data and the two published studies, further research on the effect of light exposure is justified. Table 4.7 Pseudo-first rate constants with 95% confidence limits (d-1) for rebaudioside A stored under light or dark at room temperature 0.1 M phosphate pH 7 0.1 M citrate pH 3 light 0.00118?0.000126 a 0.000443?0.000155 a dark 0.000993?0.000183 a 0.0000547?0.000143 b Different lower case letters represent significant differences within the same column (P<0.05). 34 Table 4.8 Predicted rebaudioside A loss (%) in 0.1 M buffer solutions after 1 week at room temperature Storage environment Percentage loss (%) Phosphate, pH 7-light 0.82% Phosphate, pH 7-dark 0.69% Citrate, pH 3-light 0.31% Citrate, pH 3-dark 0.04% Comparison to other sweeteners Quinlan and Jenner (1990) concluded that sucralose is stable in beverages in regard to temperature, pH and sunlight. There was no loss of sucralose in carbonated cola drinks (pH 2.8) at both 20 ?C and 35 ?C for 9 months, nor did degradation occur in pH 2.7 and 3.0 cola samples after 26 weeks of storage. Sunlight exposure did not affect the stability of sucralose in both cola and lemon-lime beverages. The stability of rebaudioside A found in the current study is similar to sucralose; very little loss of each occurs in acidic beverages at 20 ?C. Although aspartame degrades much faster than rebaudioside A (Bell and Wetzel 1995), there are some similarities with respect to their behavior. Aspartame was more stable in citrate buffer than in phosphate buffer. Higher concentrations of both buffer solutions accelerated the degradation of aspartame. However, aspartame degradation increased as pH increased irrespective of the buffer type while rebaudioside A degradation increased in phosphate buffer but decreased in citrate buffer as pH increased. As described above, rebaudioside A stability is equivalent to or better than that of sucralose or aspartame. Another advantage of rebaudioside A is being naturally-derived. As mentioned in Chapter 2, rebaudioside A is also appropriate for individuals with PKU. Overall, rebaudioside A has several advantages compared to the high intensity sweeteners sucralose and aspartame that should allow its expanding use in foods and beverages. 35 Chapter 5: Summary and Conclusion Rebaudioside A, a high intensity sweetener, is gaining popularity in the food industry due to its natural classification and zero calories. For better usage in food and beverages, its stability should be evaluated. This study provided kinetic data on the storage stability of rebaudioside A as affected by pH, buffer type and concentration, storage temperature and light exposure. Rate constants for rebaudioside A degradation were similar at pH 3, regardless of the buffer type or concentration. As pH increased, the rate constants increased for degradation in phosphate buffer, but decreased or stayed the same in citrate buffer. At pH 7, higher phosphate buffer concentrations led to faster rebaudioside degradation rates. Light exposure appears to enhance rebaudioside A degradation in pH 3 citrate buffer, but not in pH 7 phosphate buffer. Beverage formulators should recognize the combined effects of buffer type, buffer concentration, and pH on rebaudioside A stability to optimize the quality of their product during storage. For example, stability of rebaudioside A in beverages would be enhanced by lower concentrations of citrate buffer in light shielded containers. More studies could be conducted on the light stability of rebaudioside A. Constant temperature and constant light at certain wavelengths could be applied to the experimental groups and the correlation of pH and buffer type could be further studied. More data points at different temperatures could be collected improving the activation energy determination for the degradation of rebaudioside A. Although this study provided information about the storage and photo stability of rebaudioside A, the mechanism was not thoroughly studied and further experiments could be conducted to better understand degradation pathways. 36 References Abudula R, Jeppesen PB, Rolfen SD, Xiao J, Hermansen K. 2004. Rebaudioside A potently stimulates insulin secretion from isolated mouse islets: studies on the dose-, glucose-, and calcium-dependency. Metabolism 53:1378-1381. Barriocanal LA, Palacios M, Benitez G, Benitez S, Jimenex JT, Jimenez N, Rojas V. 2008. A pilot study of repeated exposures in some normotensive and hypotensive individuals and type 1 and type 2 diabetics. Regul Toxicol Pharm 51 (1): 37-41. Bell LN, Labuza TP. 1991. Aspartame degradation kinetics as affected by pH in intermediate and low moisture food system. J Food Sci 56(1): 17-20. Bell LN, Wetzel CR. 1995. Aspartame degradation in solution as impacted by buffer type and concentration. J Agric Food Chem 43: 2608-2612. Bell LN. 1997. Maillard reaction as influenced by buffer type and concentration. Food Chem 59: 143-147. Brusick DJ. 2008. A critical review of the genetic toxicity of steviol and steviol glycosides. Food Chem Toxicol 46: 83-91. Carakostas MC, Curry LL, Boileau AC, Brusick DJ. 2008. Overview: The history, technical function and safety of rebaudioside A, a naturally occurring steviol glycoside, for use in food and beverages. Food Chem Toxicol 46: 1-10. Chabot S, Beaulieu M. 2012. Extraction method for providing an organic certifiable Stevia rebaudiana extract. WO/2012/006742. International patent application PCT/CA2011/050433. 37 Chan P, Tomlinson B, Chen YJ, Liu JC, Hsieh MH, Cheng JT. 2000. A double-blind placebo- controlled study of the effectiveness and tolerability of oral stevioside in human hypertension. Br J Clin Pharmacol 50:215?220. Chang SS, Cook JM. 1983. Stability studies of stevioside and rebaudioside A in carbonated beverages. J Agric Food Chem 31: 409-412. Chaturvedula VSP, Prakash I. 2011. Acid and alkaline hydrolysis studies of stevioside and rebaudioside A. J App Pharm Sci Tech 8: 104-108. Chen TH, Zhang Y, Liu X, Shi Z, Sun J, He B. 1999. Enrichment and separation of rebaudioside A from stevia glycoside by a novel adsorbent with pyridyl group. Sci. China 42: 277-282. Christian GD. 1980. Analytical Chemistry, 3rd edition. New York: John Wiley. p 181. Clos JF, DuBois GE, Prakash I. 2008. Photostability of rebaudioside A and stevioside in beverages. J Agric Food Chem 56: 8507- 8513. Curry LL, Roberts A. 2008. Subchronic toxicity of rebaudioside A. Food Chem Toxicol 46: 11? 20. Curry LL, Roberts A, Brown N. 2008. Rebaudioside A: two-generation reproductive toxicity study in rats. Food Chem Toxicol 46: 21?30. Dobbs CM, Bell LN. 2010. Storage stability of tagatose in buffer solutions of various compositions. Food Res Int 43: 382-386. Fan N, Ammann J, Mabud A. 2007. Monitoring stevioside in Soju by high-performance liquid chromatography and liquid chromatography/mass spectrometry, J AOAC Int 5: 1365- 1372. 38 FAO/WHO. 2008. Summary and conclusions of the sixty-ninth meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA). [Accessed 2012 June 8] Available from: http://www.who.int/foodsafety/chem/jecfa/summaries/summary69.pdf. Ferri LAF, Alves-Do-Prado W, Yamada SS, Gazola S, Batista MR, Bazotte RB. 2006. Investigation of the antihypertensive effect of oral crude stevioside in patients with mild essential hypertension. Phytother Res 20: 732?736. FSANZ. 2008. Steviol glycosides as intense sweeteners. Application A540. Food Standards Australia New Zealand [Accessed 2012 June 8] Available from: http://www.foodstandards.gov.au/_srcfiles/FAR_A540_Steviol_glycosides.pdf. Gardana C, Simonetti P, Canzi E, Zanchi R, Pietta P. 2003. Metabolism of stevioside and rebaudioside A from stevia rebaudiana extracts by human microflora. J Agric Food Chem 51: 6618-6622. Geuns JMC, Augustijns P, Mols R, Buyse JG, Driessen B. 2003. Metabolism of stevioside in pigs and intestinal absorption characteristics of stevioside, rebaudioside A and steviol. Food Chem Toxicol 41: 1599?1607. Geuns JMC, Buyse J, Vankeirsbilck A, Temme EHM. 2007. Metabolism of stevioside by healthy subjects. Proc Soc Exp Biol Med 232: 164?173. Geuns, JMC, Vankeirsbilck A, Temme EH, Compernolle F, Toppet S. 2006. Identification of steviol glucuronide in human urine. J Agric Food Chem 54: 2794?2798. Grenby TH. 1991. Intense sweeteners for the food industry: an overview. Trends Food Sci Tech 2: 2?6. 39 Jaitak V, Bandna SB, Kaul VK. 2009. An efficient microwave-assisted extraction process of stevioside and rebaudioside-A from Stevia rebaudiana (Bertoni). Phytochem Anal 20: 240?245. Jaitak V, Gupta AP, Kaul VK, Ahuja PS. 2008. Validated high-performance thin-layer chromatography method for steviol glycosides in Stevia rebaudiana, J. Pharmaceut. Biomed. Anal 47:790-794. Jeppesen PB, Gregersen S, Alstrup KK, Hermansen K. 2002. Stevioside induces antihyperglycaemic, insulinotropic and glucagonostatic effects in vivo: studies in the diabetic Goto-Kakizaki (GK) rats. Phytomedicine 9:9-14. Kitada Y, Sasaki M, Yamazoe Y. 1989. Simultaneous determination of stevioside, rebaudioside A and C and dulcoside A in foods by high-performance liquid chromatography. J Chromatogr 474: 447-451. Kolb N, Herrera JL, Ferreyra DJ, Uliana RF. 2001. Analysis of sweet diterpene glycosides from stevia rebaudiana: improved HPLC method. J Agric Food Chem 49: 4538-4541. Koyama E, Kitazawa K, Ohori Y, Izawa O, Kakegawa K, Fujino A, Michio U. 2003a. Intro metabolism of glycosidic sweeteners, stevia mixture and enzymatically modified stevia in human intestinal microflora. Food Chem Toxicol 41: 359-374. Koyama E, Sakai N, Ohori Y, Kitazawa K, Izawa O, Kakegawa K, Fujino A, Michio U. 2003b. Absorption and metabolism of glycosidic sweeteners of stevia mixture and their aglycone steviol in rats and humans. Food Chem Toxicol 41: 875?883. Kroyer GT. 1999. The low calorie sweetener stevioside: stability and interaction with food in ingredients. Lebensm-Wissenschaft und-Technologie 32: 509-512. 40 Labuza TP, Kamman JF. 1983. Reaction kinetics and accelerated tests simulation as a function of temperature. In: Saguy I, editor. Computer-aided techniques in food technology. New York: Marcel Dekker. p 71-115. Li J, Chen Z, Di D. 2012. Preparative separation and purification of Rebaudioside A from Stevia rebaudiana Bertoni crude extracts by mixed bed of macroporous adsorption resins. Food Chem 132: 268?276. Liu Y, Di D, Bai Q, Li J, Chen Z, Lou S, Ye H. 2011. Preparative separation and purification of rebaudioside A from steviol glycosides using mixed-mode macroporous adsorption resins. J Agric Food Chem 59: 9629?9636. Maki KC, Curry LL, Carakostas MC, Tarka SM, Reeves MS, Farmer MV, McKenney JM, Toth PD, Schwartz SL, Lubin BC, Dicklin MR, Boileau AC, Bisognano JD. 2008a. The hemodynamic effects of rebaudioside A in healthy adults with normal and low-normal blood pressure. Food Chem Toxicol 46 : 40?46. Maki KC, Curry LL, Reeves MS, Toth PD, McKenney JM, Farmer MV, Schwartz SL, Lubin BC, Boileau AC, Dicklin MR, Carakostas MC, Tarka SM. 2008b. Chronic consumption of rebaudioside A, a steviol glycoside, in men and women with type 2 diabetes mellitus. Food Chem Toxicol 46:47?53. Matsui M, Matsui K, Kawasaki Y, Oda Y, Noguchi T, Kitagawa Y, Sawada M, Hayashi M, Nohmi T, Yoshihira T, Ishidate MJ, Sofuni T. 1996. Evaluation of the genotoxicity of stevioside and steviol using six in vitro and one in vivo mutagenicity assays. Mutagenesis 11: 573?579. Melis MS. 1995. Chronic administration of aqueous extract of Stevia rebaudiana in rats: renal effects, J Ethnopharmacol 47: 12-134. 41 Molinary SV. 1984. Preclinical studies of aspartame in nonprimate animals. In: Stegink LD, Filer LJ, editors. Aspartame: physiology and biochemistry. New York: Marcel Dekker. p 289-306. Nakajima, 2000. Micronucleus test of Rebaudioside A in mice. Test Number 5002 (079-086). Unpublished report of a study conducted at the Biosafety Research Center, Japan. Submitted to WHO by Ministry of Health and Welfare, Japan. Cited in: WHO, 2005. Nakayama K, Kasahara D,Yamamoto F. 1986. Absorption, distribution, metabolism and excretion of stevioside in rats. Shokuhin Eiseigaku Zasshi 27: 1?8. Nikiforov AI, Eapen AK. 2008. A 90-day (oral) toxicity study of rebaudioside A in Sprague- Dawley rats. Int J Toxicol 27: 65-80. Nunes APM, Ferreira-Machado SC, Nunes RM, Dantas FJS, De Mattos JCP, Caldeira-de-Araujo A. 2007. Analysis of genotoxic potentiality of stevioside by comet assay. Food Chem Toxicol 45: 662?666. Pachapurkar D, Bell LN. 2005. Kinetics of Thiamin degradation in solutions under ambient storage condition. J Food Sci 70: 423-426. Pol J, Hohnova B, Hyotylainen T. 2007. Characterization of Stevia rebaudiana by comprehensive two-dimensional liquid chromatography time-of-flight mass spectrometry. J Chromatogr A 1150: 85?92. Prakash I, Clos JF, Chaturvedula VSP. 2012. Stability of rebaudioside A under acidic conditions and its degradation products. Food Res Int 48: 65-75. Prakash I, DuBois GE, Clos. 2008. Development of rebiana, a natural, non-caloric sweetener. Food Chem Toxicol 46: 75-82. 42 Prakash I, DuBois GE, King GA, Upreti M. 2007. Rebaudioside A composition and method for purifying rebaudioside. US patent application 0292582. Quinlan ME, Jenner MR. 1990. Analysis and stability of the sweetener sucralose in beverages. J Food Sci 55: 244-246. Roberts A, Renwick AG. 2008. Comparative toxicokinetics and metabolism of rebaudioside A, stevioside, and steviol in rats. Food Chem Toxicol 46: 31?39. Sasaki YF, Kawaguchi S, Kamaya A, Ohshita M, Kabasawa K, Iwama K, Taniguchi K, Tsuda S. 2002. The comet assay with 8 mouse organs: results with 39 currently used food additives. Mutation Res 519: 103?119. Soejarto D, Compadre CM, Medon PJ, Kamath SK, Kinghorn AD. 1983. Potential sweetening agents of plant origin. II. Field search for sweet-tasting stevia species. Econ Bot 37:71?79. Steel RGD, Torrie JH. 1980. Principles and procedures of statistics: a biometrical approach. 2nd ed. New York: McGraw-Hill. 258 p. Tarantino LM. 2008. Agency Response Letter GRAS Notice No. GRN 000253. CFSAN/Office of Food Additive Safety. [Accessed 2012 June 8]. Available at: http://www.fda.gov/Food/FoodIngredientsPackaging/GenerallyRecognizedasSafeGRAS/ GRASListings/ucm154989.htm. Toskulkao C, Chaturat L, Temcharoen P, Glinsukon T. 1997. Acute toxicity of stevioside, a natural sweetener, and its metabolite, steviol, in several animal species. Drug Chem Toxicol 20: 31-44. Toyoda K, Matsui H, Shoda T, Uneyama C, Takada K, Takahashi M. 1997. Assessment of the carcinogenicity of stevioside in F344 rats. Food Chem Toxicol 35: 597?603. 43 Usami M, Sakemo K, Kawashima K, Tsuda M, Ohno Y. 1995. Teratogenicity study of stevioside in rats [in Japanese]. Bulletin of National Institute of Hygienic Sciences, 113: 31?35. WHO. 2005. Evaluation of certain food additives, WHO Technical Report Series, 928. Joint Expert Committee on Food Additives. [Accessed 2012 June 8]. Available at: http://whqlibdoc.who.int/trs/WHO_TRS_928.pdf. Williams LD, Burdock GA. 2009. Genotoxicity studies in a high-purity rebaudioside A preparation. Food Chem Toxicol 47: 1831-36. Wingard RE, Brown JP, Enderlin FE, Dale JA, Hale RL, Seitz CT. 1980. Intestinal degradation and absorption of the glycosidic sweeteners stevioside and rebaudioside A. Experientia. 36: 519-520. Wolwer-Rieck U, Lankes C, Wawrzum A, Wust M. 2010a. Improved HPLC method for the evaluation of the major steviol glycosides in leaves of Stevia rebaudiana. Eur Food Res Technol 231: 581-588. Wolwer- Rieck U, Tomberg W, Wawrzun A. 2010b. Investigations on the stability if Stevioside and Rebaudioside A in soft drinks. J Agric Food Chem 58: 12216-12220. Xili L, Chengjiany B, Eryi X, Reiming S, Yuengmin M, Haodong S, Zhiyian H. 1992. Chronic oral toxicity and carcinogenicity study of stevioside in rats. Food Chem Toxicol 30: 957? 965. Yodyingyuad V, Bunyawong S. 1991. Effect of stevioside on growth and reproduction. Human Reprod 6: 158?165. 44 APPENDIX Table 1. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 3 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 51.321 50.697 32 49.392 54.225 61 50.650 48.775 92 50.142 52.442 133 51.076 53.288 160 52.226 55.294 184 48.793 47.482 238 50.802 52.974 288 49.719 51.545 Table 2. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 3 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 50.814 50.018 28 50.030 54.672 60 50.590 50.990 89 50.682 51.179 131 50.954 50.222 189 49.271 50.459 222 47.429 48.809 264 47.785 47.667 278 48.903 48.958 45 Table 3. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 3 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 49.078 46.817 29 42.941 41.734 60 41.721 43.886 90 42.107 43.025 121 44.677 46.928 152 39.192 39.938 194 38.268 38.6 252 37.221 37.520 285 34.721 34.407 Table 4. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 5 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 45.313 45.439 29 45.349 45.619 60 45.860 44.393 90 43.031 43.241 121 45.168 44.927 152 43.303 43.601 194 45.131 45.179 252 42.738 43.640 285 44.147 44.506 Table 5. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 5 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 45.697 46.347 29 45.909 45.651 60 44.267 45.097 90 46.147 43.084 121 44.462 43.970 152 44.317 43.684 194 44.089 45.837 252 44.170 42.922 285 41.417 42.563 46 Table 6. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 5 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 46.985 46.130 29 44.880 45.714 60 43.184 44.630 90 48.061 45.916 121 44.313 43.532 152 42.415 41.971 194 42.745 43.053 252 41.837 40.773 285 39.587 36.790 Table 7. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 7 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 46.323 45.451 29 46.975 46.729 60 44.569 44.569 90 44.707 43.241 121 45.302 45.203 152 44.330 47.247 194 46.522 45.439 252 45.422 44.803 285 44.362 44.738 Table 8. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 7 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 47.301 45.845 29 47.778 46.433 60 43.203 45.525 90 44.497 44.192 121 45.908 45.231 152 45.241 45.256 194 47.324 45.246 252 44.269 44.209 285 44.708 43.897 47 Table 9. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 7 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 46.058 47.959 29 44.583 44.21 60 45.441 45.307 90 48.241 46.937 121 44.646 43.711 152 43.763 44.485 194 44.512 44.427 252 45.067 44.355 285 43.143 42.637 Table 10. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 3 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 42.887 42.295 29 43 43.307 60 42.683 41.358 90 41.278 39.89 121 42.01 42.268 152 40.28 40.719 194 42.045 37.03 252 40.321 41.285 285 42.381 42.012 Table 11. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 3 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 41.769 43.006 29 43.954 43.694 60 43.124 40.797 90 40.647 42.131 121 42.106 42.273 152 40.196 40.799 194 40.079 40.662 252 39.671 40.295 285 38.868 39.024 48 Table 12. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 3 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 42.333 41.554 29 40.565 40.833 60 39.733 40.057 90 38.138 41.324 121 40.081 39.231 152 35.413 37.288 194 36.516 35.142 252 35.92 34.757 285 31.557 31.036 Table 13. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 5 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 38.787 37.846 29 38.203 39.994 60 36.924 37.879 90 37.072 41.016 121 37.263 38.801 152 36.416 36.625 194 37.361 37.644 252 36.567 35.703 285 38.978 38.717 Table 14. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 5 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 45.198 45.810 28 46.348 47.422 60 46.661 46.561 89 47.202 46.050 131 48.071 45.584 189 46.325 44.833 222 45.261 46.454 264 44.652 43.316 278 44.99 45.86 49 Table 15. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 5 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 37.889 37.308 29 37.985 38.362 60 36.614 37.303 90 37.601 36.689 121 37.327 38.1 152 36.738 36.746 194 36.022 37.334 252 36.578 36.494 285 34.438 34.959 Table 16. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 7 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 40.106 40.359 29 41.556 41.714 60 41.098 40.134 90 38.293 36.88 121 40.712 40.759 152 39.662 38.843 194 40.061 39.943 252 36.314 38.639 285 38.214 38.109 Table 17. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 7 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 39.753 41.199 29 41.706 41.485 60 39.782 40.042 90 38.547 40.699 121 39.983 40.174 152 40.685 40.41 194 39.366 39.91 252 39.351 39.732 285 39.328 38.665 50 Table 18. Concentration of rebaudioside A in 0.02 M citrate buffer at pH 7 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 40.08 39.721 29 41.407 40.746 60 39.247 39.871 90 38.555 42.142 121 40.343 39.639 152 39.32 39.102 194 39.41 39.426 252 38.936 38.668 285 35.856 37.802 Table 19. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 3 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 43.812 44.894 29 44.696 44.941 60 43.418 44.57 90 43.111 42.305 121 43.578 43.626 152 41.307 41.732 194 43.36 42.998 252 41.132 41.889 285 42.381 42.012 Table 20. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 3 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 43.576 42.781 29 43.464 43.106 60 42.632 42.808 90 41.389 41.668 121 41.393 42.114 152 41.776 41.898 194 40.853 43.061 252 40.904 41.734 285 39.983 39.717 51 Table 21. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 3 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 43.608 44.089 29 41.84 42.744 60 41.636 41.142 90 38.547 42.739 121 41.286 41.244 152 38.169 38.041 194 37.311 36.918 252 36.341 37.198 285 34.859 34.162 Table 22. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 5 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 41.765 42.547 29 42.542 42.266 60 40.127 42.183 90 38.866 41.014 121 40.303 40.829 152 39.894 38.745 194 40.494 40.549 252 39.167 39.626 285 38.978 38.717 Table 23. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 5 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 41.761 40.492 29 42.675 42.109 60 40.21 41.347 90 39.774 40.49 121 39.225 40.439 152 39.036 40.356 194 37.748 41.298 252 37.753 38.187 285 36.209 38.111 52 Table 24. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 5 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 40.157 42.73 29 40.093 39.661 60 39.15 38.032 90 38.129 36.638 121 36.362 36.588 152 34.524 34.163 194 32.171 29.292 252 31.251 29.789 285 27.426 28.17 Table 25. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 7 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 39.126 38.779 29 39.646 39.268 60 37.508 37.499 90 33.912 37.273 121 35.011 35.075 152 32.321 32.969 194 32.928 32.535 252 31.344 30.633 285 30.338 30.607 Table 26. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 7 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 38.09 38.018 29 36.426 35.932 60 33.21 32.97 90 31.515 32.615 121 30.904 30.556 152 28.985 28.725 194 26.631 26.263 252 26.059 25.077 285 23.757 23.796 53 Table 27. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 7 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 37.774 38.377 29 33.446 33.273 60 29.422 31.05 90 27.384 28.407 121 26.867 25.384 152 22.93 22.824 194 21.119 21.698 252 19.831 19.724 285 17.679 16.959 Table 28. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 3 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 51.688 47.741 32 52.169 52.678 61 52.393 50.876 92 53.978 52.165 133 52.555 50.633 160 54.271 52.468 184 51.423 50.227 238 50.550 50.471 288 51.087 49.771 Table 29. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 3 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 50.222 51.168 28 49.814 50.809 60 47.821 48.609 89 50.419 49.822 131 48.273 48.839 189 48.266 48.038 222 47.405 47.000 264 45.910 46.199 278 46.762 47.633 54 Table 30. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 3 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 39.935 39.362 29 38.021 37.983 60 37.131 37.19 90 36.911 39.305 121 36.438 37.53 152 31.603 32.273 194 34.61 34.872 252 35.599 34.611 285 33.465 32.338 Table 31. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 5 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 41.765 42.547 29 42.542 42.266 60 40.127 42.183 90 38.866 41.014 121 40.303 40.829 152 39.894 38.745 194 40.494 40.549 252 39.167 39.626 285 38.978 38.717 Table 32. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 5 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 39.938 40.09 29 41.186 40.357 60 44.495 41.926 90 41.826 39.765 121 39.029 39.431 152 39.502 38.593 194 38.024 38.722 252 38.255 37.372 285 37.183 37.215 55 Table 33. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 5 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 41.203 43.043 29 38.714 39.026 60 37.257 38.001 90 41.052 39.1 121 37.347 38.608 152 33.019 35.978 194 36.323 37.604 252 36.333 36.111 285 34.721 34.139 Table 34. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 7 and 20 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 43.718 45.155 29 45.357 44.167 60 42.599 44.469 90 43.232 43.015 121 42.495 42.444 152 41.144 39.908 194 42.47 41.654 252 39.985 40.91 285 40.107 40.115 Table 35. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 7 and 30 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 44.315 44.026 29 43.579 41.859 60 43.103 42.44 90 41.773 38.172 121 41.11 41.347 152 40.043 40.135 194 39.021 38.937 252 39.321 37.951 285 37.199 36.942 56 Table 36. Concentration of rebaudioside A in 0.02 M phosphate buffer at pH 7 and 40 ?C Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 46.029 43.875 29 40.263 40.559 60 40.316 40.851 90 40.046 38.964 121 39.569 39.487 152 37.401 37.227 194 36.099 35.436 252 35.928 35.821 285 34.101 34.522 Table 37. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 3 and light exposure Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 45.418 46.106 29 44.015 45.424 60 44.462 44.954 90 43.388 41.325 121 40.360 43.093 152 40.781 40.412 194 41.278 41.582 252 40.630 42.389 285 39.074 40.258 Table 38. Concentration of rebaudioside A in 0.1 M citrate buffer at pH 3 and light protected Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 44.243 45.175 29 43.75 43.54 60 43.702 43.577 90 44.628 41.897 121 44.283 44.191 152 46.755 46.026 194 43.54 43.334 252 44.271 43.98 285 42.504 42.771 57 Table 39. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 7 and light exposure Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 38.38 39.067 29 36.099 35.998 60 36.477 36.42 90 35.596 36.248 121 33.211 32.776 152 33.757 31.709 194 30.519 30.724 252 28.163 28.881 285 27.363 27.629 Table 40. Concentration of rebaudioside A in 0.1 M phosphate buffer at pH 7 and light protected Time (Days) Concentration (mg/100 mL) sample 1 sample 2 0 37.566 38.197 29 35.196 35.079 60 35.423 36.122 90 35.792 37.783 121 35.604 34.4 152 31.655 33.072 194 31.542 31.275 252 28.973 29.524 285 28.439 28.34