Article Cite This: J. Nat. Prod. 2019, 82, 1200−1206
pubs.acs.org/jnp
Do Steviol Glycosides Provide Ecological Fitness to Stevia rebaudiana through Impact on Dietary Preference of Plant Pests and Herbivores? Ria R. deGuzman,‡ David J. Midmore,*,§ and Kerry B. Walsh
Downloaded via GUILFORD COLG on July 26, 2019 at 08:00:03 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Central Queensland University, Rockhampton, Queensland 4702, Australia ABSTRACT: The impact of Stevia rebaudiana leaf on feeding preferences of an insect, a mite, and a mammal was investigated. The grasshopper, Valanga irregularis of the Acrididae family, avoided feeding on S. rebaudiana leaf, as evidenced by a decrease in animal weight. Increased mortality on S. rebaudiana feed was ascribed to feeding avoidance to the point of starvation. The extent of red spider mite (Tetranychus urticae) damage was not proportional to leaf steviol glycoside (SG) concentration, a result ascribed to a feeding mechanism that avoids chlorenchyma cells that contain SGs. Guinea pigs (Cavia porcellus) were presented with the choice between a control feed and feed amended to contain 5% sucrose or 0.02%, 4%, or 10% (dry weight) of S. rebaudiana leaf. Feed intake increased (39% above the control) for the diet involving high levels of SG amendment of feed (10% S. rebaudiana leaf). Encouragement of general mammalian herbivory may provide ecological fitness to S. rebaudiana if it is more tolerant of grazing pressure than other plants in its environment. Improvement in feed intake may have commercial implication for use of S. rebaudiana as an additive in stock feeds.
D
to aphids even at the highest concentration used (750 ppm). Acceptance of an in vitro diet containing steviol (500 and 40 ppm, respectively) by Heliothis virescens Fabr. and Aedes aegypti L. (yellow fever mosquito) (unpublished study of Klocke cited in Nanayakkara et al.6) may represent variation in insect responsiveness. However, the in vivo feeding mechanism of aphids involves stylus access to the plant phloem, while mosquitoes feed on floral nectar. Thus, aphids and mosquitoes would not normally access leaf SGs, which are stored in the vacuole of the mesophyll cells.7 Work with an insect that feeds on the whole leaf tissue is, therefore, of interest for investigation of the role of SGs in insect deterrence. To date, only observational accounts have been made on the nonpreference of Epicauta adomaria (Coleoptera) toward S. rebaudiana leaf, as indicated by avoidance of S. rebaudiana leaf disks when mixed with leaf disks of Capsicum, Amaranthus, Emilia, and Lycopersicum.8 Investigation of the feeding behavior of grasshoppers toward S. rebaudiana leaves is interesting, because grasshoppers represent a likely pest of S. rebaudiana and the chewing of grasshoppers would ensure that the SGs in the leaves are consumed. Grasshoppers are generalist herbivores that feed on a wide variety of grass and broadleaf plants.9 Grasshoppers’ feeding avoidance has been reported in the context of a number of leaf secondary metabolites. For example, Schistocerca americana was reported to avoid phenolics
iterpenoid steviol glycosides (SGs) are of commercial interest for their well-demonstrated sweetness. These compounds elicit a response from the “sweet” taste receptors of the mouth approximately 300 times that of sucrose (on a weight basis).1 SGs comprise about 10% of Stevia rebaudiana (Bertoni) Bertoni (Asteraceae) leaf dry weight (dw), conferring a level of sweetness that may act in mammalian or invertebrate herbivory, in terms of either attraction or deterrence. Herbivore deterrence has an obvious ecological benefit to the plant, while attraction may serve to aid seed dispersal or to reduce the fitness of neighbor competitor plant species through associated grazing.2 We are unaware of any published work on herbivory of S. rebaudiana in its native habitat, although cattle grazing of S. rebaudiana in Paraguay has been observed.3 The SGs found in S. rebaudiana all contain steviol as an aglycone backbone, with the number of glucose units attached to the C-13 and C-19 of the molecule differing.4 Of the eight common SGs in S. rebaudiana leaf, the forms in highest concentration are stevioside, rebaudioside A, rebaudioside C, and dulcoside A (e.g., at 9.1%, 3.8%, 0.6%, and 0.3% w/w, dry weight basis, respectively in wild plants in Paraquay5). The different SG forms differ in taste and aftertaste profiles and may thus have different impacts on herbivory. The impact of SGs on aphid (Schizaphis graminum) feeding behavior involving artificial feed testing has been reported.6 Feeding deterrence was noted at low concentrations for steviol (150 ppm), while stevioside deterrence occurred at a higher concentration (650 ppm), and rebaudioside A was acceptable © 2019 American Chemical Society and American Society of Pharmacognosy
Received: November 12, 2018 Published: May 7, 2019 1200
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
Journal of Natural Products
Article
(coumarin, umbelliferone, salicin), alkaloids (gramine, nicotine), and terpenoids (geraniol, abeitic acid, ursolic acid) when presented in semimicro gelatin capsules or gelatin-walled microcapsules or spray-dried on leaves.10 In another study, Locusta migratoria exhibited a 50% reduction in feeding for feeds with alkaloids (nicotine, quinine, and tomatine), phenolics (salicin and umbelliferone), sulfur compounds (sinigrin and allylisothiocyanate), or triterpenoids (azadirachtin) equivalent to a leaf concentration of 56.25), associated with the first days of presentation with a new feed, and three preferences (χ2 > 51.02) for the sucrose-sweetened feed. When given a choice between the sucrose (5%) and the S. rebaudiana (0.02%) feed (i.e., T6), equal sampling of both dishes was observed (data not presented). Preference for the 5% sucrose feed was shown over control feed, but only in some animals. An overall increase in total feed consumption (i.e., by 39% for T4 relative to T1 treatment) occurred with the 10% S. rebaudiana treatment (Figure 4); however, deterrence for the high-concentration (T4, 10% S. rebaudiana) feed was shown over control feed, again only in some animals. The electrophysiological response of Mongolian gerbil taste buds (a close relative of guinea pigs) to different sweet solutions has been documented in terms of the electrical signal strength from the chorda tympani nerve as the taste buds are exposed to different sweet solutions (Table 1).21−24 Sucrose is perceived by gerbils to be sweeter than glucose, while the sweeteners saccharin, stevioside, and rebaudioside A were perceived to be sweeter than sucrose although at a lesser magnitude than human taste perception.25 The response of guinea pig chorda tympani response to saccharin, stevioside, and rebaudioside A is not known, although the response of sucrose has been reported.26
Figure 3. Leaf concentration (% leaf dw) of stevioside (white bar) or rebaudioside A (black bar) in leaves qualitatively evaluated for level of red mite damage (with 1 as least damaged and 5 as most damaged). Bars represent averages with an associated standard error of the mean.
leaf damage by puncturing through the cuticle and epidermis layer with their stylets, allowing selective damage of palisade and spongy mesophyll cells.20 With this feeding mechanism, the spider mites were possibly able to avoid the SGs and their precursors that are found in chlorenchyma tissue.20 Future 1202
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
Journal of Natural Products
Article
link between feeding and SG level in the guinea pig. It is also important to realize that a positive association does not prove the hypothesis, as there may be a correlation between SG level and another compound that impacts feeding preference. Evolutionary Fitness. A range of secondary metabolites in plants have been identified as either insecticidal or feedingdeterrent to insects, with a range of the compounds belonging to the terpenoid family.30 The level of SGs in stevia leaf, at ca. 10% of dw, is significant to the C economy of the plant. Yet this level is far higher than is required to elicit perceived sweetness in mammals when chewing leaves. It has been noted that the amount of insecticidal or deterrent metabolites within plants can be disproportionate to the amount required for feeding deterrence, suggesting either evolutionary unfitness or other roles for the metabolite within the plant. For example, 1,8-cineole, a volatile monoterpenic oil, occurs with nerolidol at up to 90% of the leaf dw of Melaleuca leaves31 but is lethal at minute levels against the rice weevil Sitophilus oryzae and the red flour beetle Tribolium castaneum (23.5 and 15.3 μL L−1 of air, respectively).32,33 Such a situation may also apply with steviol glycosides in S. rebaudiana leaf. Commercial Application in Mammal Feeds. The increase in total feed consumption for 10% S. rebaudianaamended feeds is interesting, with an increase in consumption of the unsweetened control feed as well as the S. rebaudianaamended feed. This result indicates potential for S. rebaudiana amendment in animal feeds to increase feed consumption and growth rate. We are aware of two other reports on the use of S. rebaudiana as an animal feed supplement. One reports on a study of weaning pigs,13 with more consumption of a 5% w/w sucrose feed (0.546 ± 0.021 kg pig−1 d−1) than of the control feed (0.444 ± 0.021 kg pig−1 d−1) or of feed with any of three S. rebaudiana sweetener concentrations (at the highest S. rebaudiana concentration of 334 mg kg−1 feed, consumption was 0.473 ± 0.021 kg pig−1 d−1). The composition of the commercial S. rebaudiana sweetener (“98% stevia purity”) in terms of stevioside and rebaudioside A was not indicated. In comparison, the current feeding trials involved two diet amendments of 3200 and 7900 mg stevioside kg−1 feed, i.e., higher levels of stevioside that promoted total feed consumption. The other report was on feeding chicken either a basal diet or one where S. rebaudiana replaced 10% of the basal diet.34 Although classed as a high-energy fodder, the S. rebaudiana-amended diet did not contribute to apparent metabolizable energy, and the authors suggest such an amended diet would be more suitable for ruminant animals. The authors did not report on the encouragement of feeding when the diet included S. rebaudiana. In another study with a natural sweetness enhancer (haumatin at 500 g per ton of feed) feed consumption by broilers was increased by 4.4% and weight gain by 3.1%.35 In summary, fresh leaves of S. rebaudiana promoted a nonfeeding-deterrent response by grasshoppers, to the detriment of insect growth and mortality. It was noted that grasshoppers may avoid the SGs or their precursors by preferentially feeding on the stem (rather than on the leaves) of S. rebaudiana. The ecological value of feeding deterrence to plant survival and reproductive fitness is obvious. There was no evidence of a link between leaf SG level and spider mite feeding level; this may be due to their feeding on leaf tissues other than the chlorenchyma tissues believed to store SGs. Inclusion of ground S. rebaudiana leaf in rations of guinea pigs at a rate of
Table 1. Summary of Equivalent Ratio of Sucrose to Other Sweet Compounds Based on the Electrophysiological Response of the Chorda Tympani (CT) of the Mongolian Gerbil Based on Literaturea ratio sucrose:glucose 1:0.5 sucrose:saccharin 1:4 sucrose:stevioside 1:117
sucrose:rebaudioside A 1:117
basis glucose solution (0.5 M) elicits only half the CT signal of sucrose solution (0.5 M) 3.9 × 10−2 eliciting a response similar to 9.9 × 10−3 saccharin based on the Kd value of sucrose (0.07) and stevioside (0.0006), which were obtained from a plot of CT responsiveness vs molar concentration of sucrose and stevioside based on the Kd value of sucrose (0.07) and stevioside (0.0006), which were obtained from a plot of CT responsiveness vs molar concentration of sucrose and rebaudioside A
ref 21 23 22, 24
22, 24
a
Kd is the dissociation constant by Beidler’s27 taste theory to summarize the linear relationship between the concentration of the tastant compound and the available site of taste receptors.
Given the lack of available data on guinea pigs, the sweetness equivalence derived from the gerbil electrophysiological tests was used with the assumption that the taste bud receptors of gerbils and guinea pigs are similar. A sweetness equivalence of 117 (Table 1) implies that the treatments of 0.02%, 4%, and 10% S. rebaudiana in amended feeds had a sucrose equivalency of 0.2%, 37%, and 92%. A few fatalities were evident after T1−T3 of the first group of animals. The observed fatalities were not associated with a feeding treatment and were associated with either stress or a dietary deficiency of antioxidants. An autopsy of a fatality associated with one S. rebaudiana feeding treatment revealed inflammation of the mesenteric fat, which was attributed to possible dietary deficiency of antioxidants, a fatty change in the liver consistent with anorexia, and pulmonary edema in the lungs and stomach erosion possibly due to stress. The long duration of the feeding trials (almost 2.5 months), with the social guinea pigs isolated, may have created stress for the animals The hydrolysis product of SG, steviol, has a reported toxicity, with an LD50 of 5.2 g kg−1 body weight reported for male hamsters.28,29 Assuming all stevioside was degraded to steviol upon ingestion by the animals, the daily intake of the highest S. rebaudiana-amended feed (0.037 kg of 10% feed kg−1 body weight) involved a steviol intake of 119 mg kg−1 body weight (calculated from the feed stevioside content and the molecular weight of steviol, 318 g mol−1, and stevioside, 805 g mol−1). This intake is 44 times lower than the LD50 of 5.2 g kg−1 body weight reported for male hamsters.29,29 For these reasons we conclude, therefore, that the fatalities were not associated with the S. rebaudiana feed amendment. The guinea pig feeding trials had some potential limitations: (i) The current trials employed a trial duration of 3 days, with daily measurements. It is possible that daily sampling obscured an initial short-term preference for the S. rebaudiana-mixed feed. Conversely, a longer feeding period has more relevance to a consideration of an ecological role in influencing herbivory behavior. (ii) Other compounds may have been present in the ground S. rebaudiana leaf, additional to the SG. Further work could employ the S. rebaudiana sweetening agents directly (stevioside compared to rebaudioside A) as feed amendments since the negative result (no relationship between feeding preference and SG level) does not disprove the hypothesis of a 1203
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
Journal of Natural Products
Article
Table 2. Designs of Feeding Trials grasshopper feeding trial, each treatment with three replicates grasshopper size/#
acclimatized 3 days
large (0.9 g) and small (0.3 g), three per jar
trema leaves
grasshopper size/#
control 3 days
various, one per jar
trema leaves
control 3 days
test period 3 days
trema leaves
control: trema leaves trema leaves choice: trema and stevia leaves no choice: stevia leaves only grasshopper mortality trial, each treatment with seven replicates test period 3 days
two-week acclimatize period commercial feed
guinea pig size/#
two-week acclimatize period commercial feed
assessment of leaf damage, SG leaf concentration guinea pig feeding trial, each test period treatment with nine replicates
test period 3 days
7-day control
test period 3 days
T1: control feed vs stevia feed (4% w/w)
commercial feed
T2: control feed vs stevia feed (10% w/w)
test period 3 days
8-day control
test period 3 days
T4: control feed vs sucrose feed (5% w/w)
commercial feed
T5: control feed vs stevia feed (0.02% w/w)
10-day control commercial feed 15-day control commercial feed
test period 3 days T3: control feed vs sucrose feed (5% w/w)
measurements daily weight of feed and water intake
test period 3 days
measurements
T6: sugar feed (5% w/w) vs stevia feed (0.02% w/w)
daily weight of feed and water intake
Feeding trials were conducted between March and April 2010, using 20 jars, arranged in a 4 × 5 matrix, in each trial. The jars were maintained in a 12 h/12 h light/dark cycle at 25 °C day/20 °C night. Grasshoppers initially were acclimatized for 3 days (days 0 to 3), three similar-sized per jar, with a supply of trema leaves only. The insects were maintained on a trema-only diet for another 3 days, which served as the control measurement period (day 4−6). A 3-day test period (day 7−9) followed wherein grasshoppers were randomly assigned one of the three diets: (a) control: trema leaves only, (b) choice: trema and S. rebaudiana leaves, or (c) no choice: S. rebaudiana leaves only. After the test period, the grasshoppers were returned to the trema leaf diet for a 5-day post-test measurement period (day 10− 14). Leaf weight was not monitored in the experiment because of the daily variations in the fresh weight of the stem cuttings. Instead, feeding activity was measured through weight comparison of the grasshoppers relative to their starting weight at day 0, with weight assessed at the end of the acclimatization (day 3), control (day 6), test (day 9), and post-test (day 14) periods. In instances where a grasshopper died during a feeding period, only the weights of the live grasshoppers were averaged. Fecal weight was also used as an indirect measure of feeding. The fecal weight for each jar was referenced to the fecal weight on day 3, normalized to the number of grasshoppers. Some mortality was noted during the feeding trials. However, it was not clear if this was associated with ingestion of S. rebaudiana tissue, starvation, or another reason. To characterize the timeline for mortality through starvation, grasshoppers from the feeding test were reacclimatized with trema leaves for 2 weeks to normalize their feeding behavior. Fourteen grasshoppers were then randomly selected and individually placed inside a jar. Grasshoppers were initially given trema leaves for the first 3 days (day 0−3) and then either starved (control group, no food source) or presented only with S. rebaudiana leaves (stevia group) for the next 3 days (day 4−6). Grasshopper weight was measured on day 0, 3, and 6, while fecal weight was measured on day 3 and 6. Change in grasshopper weight was expressed relative to initial weight, while fecal weight was expressed as
10% w/w enhanced overall intake, although there was variation between individual animals. Mammalian feeding encouragement could increase the ecological fitness of S. rebaudiana if it is more tolerant of herbivory than its environmental competitors. However, at least for large herbivores, the reverse appears to be true, given Soejarto’s (2002) remark on the decline of S. rebaudiana in plant communities experiencing cattle grazing.3
■
day 0, 3, 6, grasshopper weight and day 3, 6 feces weight
measurements
small (ca. 450 g), one per cage
large (ca. 900 g), one per cage
measurements daily, grasshopper weight and feces weight
measurements
control: no food source stevia group: S. rebaudiana leaves only spider mite feeding survey
leaf # 29 leaves from two plants
guinea pig size/#
post-test period 5 days
EXPERIMENTAL SECTION
Grasshopper Feeding Trial. Native Australian grasshoppers (Valanga irregularis) were harvested from poison peach trees (Trema tomentosa), subsequently referred to as “trema”, in a coastal vineforest community in Yeppoon, Queensland, Australia (23.1° S; 150.7° E) from March to April of 2010. Lightweight grasshoppers were mostly first to third instar nymphs with an average start weight of 0.3 g (range 0.1−0.4 g). Heavyweight grasshoppers were mostly fourth and fifth instar nymphs with an average weight of 0.9 g (range 0.5−1.3 g). Stem cuttings of trema tree and of an S. rebaudiana plant were placed into water, and the cuttings covered with a plastic bag to increase relative humidity. Stem cuttings of S. rebaudiana were cut from a single plant maintained in a growth chamber (photoperiod 14 h light/10 h dark to inhibit flowering, 80% humidity, 25 °C). The leaf stevioside and rebaudioside A concentration in actively growing leaf tissue and extracted using water was about 10−15% dw and 1−2% dw, respectively, analyzed with HPLC using the procedure described previously.36 In essence, HPLC analysis was performed using an Agilent 1100 equipped with an Agilent Zorbax high pressure Reliance Cartridge guard column (4.6 × 12.5 mm, 5 μm) in series with a Zorbax NH2 column (4.6 × 250 mm). The system was calibrated with stevioside and rebaudioside A standards using concentrations of 0, 0.2, 0.4, 0.6, and 1.25 mM. Grasshoppers were kept in groups of three inside inverted cylindrical plastic jars (12 cm diameter base, 12 cm height), with a perforated top allowing air circulation. A perforated cardboard base allowed leaf cuttings to sit in water throughout the experiment. 1204
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
Journal of Natural Products
Article
standard feed and the 10% S. rebaudiana leaf feed; (iii) T3: a choice between the standard feed and 5% w/w sucrose feed; (iv) T4: a choice between the standard feed and 5% w/w sucrose feed (i.e., a repeat of T3); (v) T5: a choice between the standard feed and 0.02% S. rebaudiana leaf feed; and (vi) T6: a choice between the sucrose feed (as used in T4) and the 0.02% S. rebaudiana leaf feed (as used in T5). Data Analysis. Analysis of variance tests were employed using SPSS v.17.038 with statistical significance reported at a 95% confidence level. In the grasshopper trial, the changes in weight relative to day 0 and the mortality rate between treatments were compared using a one-way ANOVA analysis. In the guinea pig trial, the daily proportion of feed intake from the two feeding dishes was calculated for each animal. A modified chi-square test was used to quantify the normal distribution curve of the feeding behavior of the guinea pigs during the control periods.39 The daily chi-scores for each guinea pig were calculated for all control period (C) days (n = 202), given that a 50:50 sampling ratio from both dishes is expected given nonpreference feeding from the two feeding dishes. These scores were ranked, with the highest 5% regarded as significant deviations from a 50:50 sampling. The lowest value of this group was adopted as a criterion for significant difference from control treatment. The average daily intake of the control feed, the test feed, and the sum of both during the test periods were compared using a one-way ANOVA analysis using SPSS v.17.0.
a ratio relative to grasshopper weight. A summary of the experimental setups is presented in Table 2. Spider Mite Feeding. An infestation of red spider mites (T. urticae) developed on a population of S. rebaudiana plants maintained inside a growth cabinet (600 μmol m−2 s−1, 25 °C, relative humidity of 85%, and 14 h light/10 h dark cycle). To investigate a possible correlation between leaf damage and SG leaf concentration, leaf damage was evaluated on a scale of 1−5 (with 1 being the least damaged). Leaf samples were observed using a scanning electron microscope to investigate the tissues damaged, and leaf SG concentration was evaluated using water extraction and HPLC as described above. A total of 29 leaf samples from two plants were analyzed. Guinea Pig Feeding Trial. C. porcellus preference toward S. rebaudiana leaf was tested through two-choice feed experiments using a standard guinea pig feed amended with crushed S. rebaudiana leaf, a standard feed mixed with sugar, and standard guinea pig feed. This work was approved by the CQUniversity Animal Ethics Committee (Permit A09/05-247). A total of 18 male guinea pigs (C. porcellus) were assigned at random to two groups, with each group having 9 guinea pigs. The animals were initially kept in pairs for 5 days to help transition to the new environment, after which the guinea pigs were placed individually in cages. Adjacent cages were separated by wire mesh that allowed a level of interaction between animals. Each wire-bottomed cage had a dimension of 38 cm × 90 cm × 29 cm height, with an enclosed section and an open section. Feed was presented in the open section in two feeding dishes, positioned on opposite sides of the cage with a water bottle placed centrally. Room temperature was constant at 25 °C, light was on a 12 h light/12 h dark diurnal cycle, and classical music was played at low volume during the light period, as recommended by ethical guidelines.37 Mortalities observed in the first group were autopsied by a veterinarian, with pathological and histological testing of the intestine, lung, heart, and liver tissues. Feed (Cavia Origins, Pet Circle, Sydney, Australia) consisted of pellets comprising fescue, lucerne, oat fiber, wheat, soybean, and oats. The feed pellets and dry S. rebaudiana leaf material was ground through a hammer mill to pass a 2 mm mesh, then blended and reground to produce a 4% and a 10% w/w S. rebaudiana leaf:pellet mix. For the 0.02% blend, a sample of the 4% S. rebaudiana mix was finely milled using a centrifugal grinding mill (0.2 mm mesh size) (Retsch GmbH, Germany) and proportionately mixed into the ground standard feed. The dried S. rebaudiana leaf stock, analyzed as noted above, contained 7.9% stevioside and 1.8% w/dw rebaudioside A, with the 0.02%, 4%, and 10% S. rebaudiana feed treatments containing 0.002%, 0.316%, and 0.79% stevioside, with a sucrose equivalence of 0.2%, 37%, and 92% w/dw sucrose, respectively (given a sucrose to stevoide sweeteness equivalence of 117 to 1, Table 1). For the 5% sucrose feed formulation, sugar was added to the feed by spraying a known volume of 33% w/v sucrose solution onto the ground standard feed, with the feed dried in a forced air oven at 40 °C for 24 h and then passed through a hammer mill before storage to avoid particle size differences between sugar crystals and the ground standard feed. The standard feed control was also ground. The feeding schedule began with a two-week acclimatization phase. After the acclimatization phase, a series of 3-day test periods (T) were imposed, separated by control periods (C) of at least 7 days’ duration. Feed and water were weighed and replaced daily. In the choice experiments, the dish used for control and treatment feeds was randomly chosen every day to avoid development of a spatial preference by the animals. Tests 1 to 3 were conducted on a group of guinea pigs consisting of nine animals (average weight = 450 g). During the subsequent control period, symptoms of weight loss were observed in some guinea pigs, and so a decision was made to discontinue use of these animals. Tests 4 to 6 were conducted on a second group of nine animals (average weight = 600 g). The diet of this second set was augmented by a small ration of vegetable in the nontest periods. The tests periods involved (i) T1: a choice between the standard feed and the 4% S. rebaudiana leaf feed; (ii) T2: a choice between the
■
AUTHOR INFORMATION
Corresponding Author
*Tel: +52 15951203545. E-mail:
[email protected]. ORCID
David J. Midmore: 0000-0001-9946-5008 Present Addresses ‡
Asahi Australia, 2 Beverage Drive, Tullamarine, Victoria 3043, Australia. § School of Policy Agriculture and Development, University of Reading, Earley Gate, Whiteknights Road, Reading, RG6 6AR, UK. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The assistance of Dr. A. Fenning with issues on animal culture, health, and ethics, and A/Prof. S. McKillup for statistics advice is acknowledged. The encouragement of J. Ashton of Sanitarium P/L to undertake this work is appreciated. We thank the Rural Industries Research and Development Corporation and Sanitarium Plc Ltd. for financial support for these studies, and CQU for an International Postgraduate Research Scholarship awarded to R.D.G.
■
REFERENCES
(1) Inglett, G. E. Food Technol. 1981, 35, 37−41. (2) Harborne, J. B. Introduction to Ecological Biochemistry; Academic Press: London, 1982. (3) Soejarto, D. D. In Stevia, the Genus Stevia, Medicinal and Aromatic Plants-Industrial Profiles Series, Vol. 19; Kinghorn, A. D., Ed.; Taylor and Francis: London, 2002; Chapter 2, pp 18−39. (4) Dacome, A. S.; da Silva, C. C.; da Costa, C. E. M.; Fontana, J. D.; Adelmann, J.; da Costa, S. C. Process Biochem. 2005, 40, 3587−3594. (5) Brandle, J. E.; Starratt, A. B.; Gijzen, M. Can. J. Plant Sci. 1998, 78, 527−536. (6) Nanayakkara, N. P. D.; Klocke, J. A.; Compadre, C. M.; Hussain, R. A.; Pezzuto, J. M.; Kinghorn, A. D. J. Nat. Prod. 1987, 50, 434− 441. (7) Brandle, J. E.; Telmer, P. G. Phytochemistry 2007, 68, 1855− 1863. 1205
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
Journal of Natural Products
Article
(8) Metivier, J.; Viana, A. M. J. J. Exp. Bot. 1979, 30, 805−810. (9) Dadd, R. H. Adv. Insect Physiol. 1963, 1, 47−109. (10) Bernays, E. A. J. Chem. Ecol. 1991, 17, 2519−2526. (11) Cottee, P. K.; Bernays, E. A.; Mordue, A. J. Entomol. Exp. Appl. 1988, 46, 241−247. (12) Freeland, W. J. Entomol. Exp. Appl. 1975, 18, 281−289. (13) Munro, P. J.; Lirette, A.; Anderson, D. M.; Ju, H. Y. Can. J. Anim. Sci. 2000, 80, 529−531. (14) Gade, D. W. Geogr. Rev. 1967, 57, 213−224. (15) Jacobs, W. W.; Beauchamp, G. K. Physiol. Behav. 1977, 18, 491−493. (16) Jacobs, W. W. Physiol. Behav. 1978, 20, 579−588. (17) Nolte, D. L.; Mason, R.; Lewis, S. L. J. Chem. Ecol. 1994, 20, 303−308. (18) Ceunen, S.; Geuns, J. M. C. Plant Sci. 2013, 198, 72−82. (19) Fuente, A. L. O. Rev. Cienc. Tecnol. Dir. Invest. UNA 2001, 1, 29−33. (20) Mothes, U.; Seitz, K. A. Acarologia 1982, 23, 149−157. (21) Jakinovich, W., Jr.; Oakley, B. J. Comp. Physiol., A 1975, 99, 89− 101. (22) Jakinovich, W., Jr. Brain Res. 1976, 110, 481−490. (23) Jakinovich, W., Jr. Brain Res. 1981, 210, 69−81. (24) Vasquez, E.; Jakinovich, W., Jr. J. Agric. Food Chem. 1993, 41, 1305−1310. (25) Inglett, G. E. Food Technol. 1981, 35, 37−41. (26) Brouwer, J. N.; Hellekant, G.; Kasahara, Y.; van der Wel, H.; Zotterman, Y. Acta Physiol. Scand. 1973, 89, 550−557. (27) Beidler, L. M. J. Gen. Physiol. 1954, 38, 133−139. (28) Tsokulkao, C.; Chaturat, L.; Temcharoen, P.; Glinsukon, T. Drug Chem. Toxicol. 1997, 20, 31−44. (29) Geuns, J. M. C. In Proceedings of the First Symposium “Safety of Stevioside”.Geuns, J. M. C.; Buyse, J., Eds.; Euprint ed.: KULeuven, 2004; pp 85−127. (30) Mabry, T. J.; Gill, J. E. In Herbivores: Their Interaction with Secondary Metabolites.; Rosenthal, G. A.; Janzen, D. H., Eds.; Academic Press: New York, 1979; pp 501−537. (31) Suzuki, K. Eco-habitat. 1999, 6, 133−141. (32) Lee, B. H.; Choi, W. S.; Lee, S. E.; Park, B. S. Crop Prot. 2001, 20, 317−320. (33) Lee, B. H.; Annis, P. C.; Tumaalii, F.; Choi, W. S. J. Stored Prod. Res. 2004, 40, 553−564. (34) Atteh, J.; Onagbesan, O.; Tona, K.; Buyse, J.; Decuypere, E.; Geuns, J. Arch. Zootec. 2011, 60, 133−136. (35) Cortés Cuevas, A.; Laparra Vega, J. L.; Á vila González, E. Vet. Méx. 2005, 127−133. (36) DeGuzman, R.; Midmore, D. J.; Walsh, K. B. J. Nat. Prod. 2018, 81, 56−70. (37) Animal Research Review Panel, Guideline 21. Guidelines for the Housing of Guinea Pigs in Scientific Institutions; Animal Welfare Branch, NSW Department of Primary Industries, Locked Bag 21, Orange NSW, 2800. https://www.animalethics.org.au/__data/assets/pdf_ file/0012/222510/housing-guinea-pigs-scientific-institutions.pdf (accessed March 21, 2019). (38) SPSS. SPSS Statistics 17.0 Brief Guide; SPSS: Chicago, IL, USA. (39) McKillup, S. Statistics Explained - An Introductory Guide for Life Scientists; Cambridge University Press: Melbourne, 2005.
1206
DOI: 10.1021/acs.jnatprod.8b00958 J. Nat. Prod. 2019, 82, 1200−1206
