- Email: [email protected]
Do low-calorie sweeteners promote weight gain in rodents?
PHB-11192; No of Pages 5 Physiology & Behavior xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Review
Do low-calorie sweeteners promote weight gain in rodents? John I. Glendinning Department of Biology, Barnard College, Columbia University, 3009 Broadway, New York, NY 10027, United States
H I G H L I G H T S • • • • •
The linkage between LCS consumption and elevated body weight in rodents is examined. LCSs promoted weight gain when they were presented in yogurt. The LCS-treated yogurt formulations did not appear to taste sweet to the rats. The elevated weight gain could not be explained solely by increased caloric intake. LCS and yogurt may promote weight gain by modifying the gut microbiota.
a r t i c l e
i n f o
Article history: Received 2 October 2015 Received in revised form 28 January 2016 Accepted 29 January 2016 Available online xxxx Keywords: Low-calorie sweeteners Caloric intake Weight gain Rat Yogurt Gut microbiota
a b s t r a c t Low-calorie sweeteners (LCSs) are used globally to increase the palatability of foods and beverages, without the calories of sugar. Recently, however, there have been claims that LCSs promote obesity. Here, I review the literature linking LCS consumption to elevated body weight in rodents. A recent systematic review found when the LCSs were presented in water or chow, only a minority of the studies reported elevated weight gain. In contrast, when the LCSs were presented in yogurt, the majority of the studies reported elevated weight gain. This review focuses on this latter subset of studies, and asks why the combination of LCSs and yogurt promoted weight gain. First, LCSs have been hypothesized to induce metabolic derangement because they uncouple sweet taste and calories. However, the available evidence indicates that the LCS-treated yogurts did not actually taste sweet to rats in the published studies. Without a sweet taste, the concerns about uncoupling sweet taste and calories would not be relevant. Second, in several studies, the LCS-treated yogurt increased weight gain without increasing caloric intake. This indicates that caloric intake alone cannot explain the elevated weight gain. Third, there is evidence that LCSs and yogurt can each alter the gut microbiota of rodents. Given recent work indicating that changes in gut microbiota can modulate body weight, it is possible that the combination of LCS and yogurt alters the gut microbiota in ways that promote weight gain. While this hypothesis remains speculative, it is consistent with the observed rodent data. In human studies, LCSs are usually presented in beverages. Based on the rodent work, it might be worthwhile to evaluate the impact of LCS-treated yogurt in humans. © 2016 Elsevier Inc. All rights reserved.
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Under what experimental conditions do LCSs increase weight gain? . . . . . . . . . . . 2.1. LCSs, sweet taste and cephalic-phase responses . . . . . . . . . . . . . . . . . 2.2. Is caloric intake greater on the sweet non-predictive diet? . . . . . . . . . . . . 2.3. Is intermittent access to the sweet non-predictive diet necessary for weight gain? . 2.4. Does LCS-treated yogurt alter the gut microbiota in ways that promote weight gain? 3. Are results from the sweet non-predictive diets relevant to humans? . . . . . . . . . . 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
E-mail address: [email protected].
http://dx.doi.org/10.1016/j.physbeh.2016.01.043 0031-9384/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
0 0 0 0 0 0 0 0 0 0
2
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
1. Introduction Low-calorie sweeteners (LCSs) are attractive to consumers because they make foods and beverages taste better, without the caloric or glycemic effects of sugar [8]. Despite widespread usage, LCSs remain controversial. There are persistent claims that LCSs cause adverse health effects, including cancer, neurotoxicity, allergic reactions, elevated caloric intake and obesity (review in [32, 38]). The claims of cancer, neurotoxicity and allergic reactions are contradicted, however, by a large body of empirical work (reviews in [4, 18, 23, 30, 46]) and the fact that LCSs have been approved and recommend for use by regulatory bodies (e.g., U.S. Food and Drug Administration, European Food Safety Authority, World Health Organization) and the Academy of Nutrition and Dietetics [12]. The claim that LCSs increase energy intake and body weight is based on the notion that repeated LCS use disrupts the mechanisms underlying energy regulation [16, 22, 38]. Here, I focus on empirical support for the proposed linkage between LCS consumption and weight gain in rodents. First, I discuss the specific experimental conditions under which LCS have been found to increase body weight. Second, I consider the limitations of the animal studies linking LCSs to overconsumption and weight gain. Third, I examine the extent to which results from the animal studies are relevant to humans. 2. Under what experimental conditions do LCSs increase weight gain? To determine whether LCSs increase body weight in rodents, investigators have used several experimental approaches. In a recent systemic review of this literature, Rogers et al. [31] identified a total of 90 relevant studies, and categorized them according to one of three experimental designs. In design 1, the LCS was added to the rodents' only source of food or water (n = 47 studies). In design 2, the rodents were provided with a standard diet (chow and water) plus continuous access to an LCS-sweetened water, diet or yogurt (n = 21 studies). In design 3, the rodents were offered a standard diet plus intermittent access to an LCS-treated yogurt (n = 15 studies1). Only a small percentage of the studies that used design 1 (9%) or design 2 (24%) reported a significant increase in body weight [31]. In contrast, the vast majority of studies that used design 3 (87%) reported significant increases in body weight. The fact that so few of the studies that used designs 1 or 2 reported weight gain indicates that LCSs are not inherently obesogenic to rodents. Below I examine design 3 in greater detail so as to gain insight into why it was more likely to cause weight gain. The majority of the studies that used design 3 (i.e., [5, 39, 41–43]) were conducted by a group at Purdue University. In most cases, the investigators exposed rats to one of two diets. The “sweet predictive” diet consisted of a maintenance diet (standard chow and water ad libitum) plus a 30 g supplement of low-fat yogurt per day. The yogurt was offered 6 days/week. On 3 of the days, it was sweetened with 20% glucose; and on the other 3 days, it was unsweetened. This diet was called sweet predictive because the sweet taste of the glucose predicted the presence of sugar calories in the yogurt. The “sweet non-predictive” diet was identical, except that the sweetened yogurt contained 0.3% saccharin instead of glucose. This latter diet was called sweet non-predictive because the sweet taste of the saccharin did not predict the presence of sugar calories. As compared with rats on the sweet predictive diet, those on the sweet non-predictive diet gained more weight and (in many cases) ingested slightly but significantly more calories (Fig. 1A, B). Likewise, when the maintenance diet consisted of a high-fat chow sweetened with 20% glucose [5], the rats on the sweet non-predictive diet gained more weight and ingested more kcal/week than rats on the sweet predictive diet (Fig. 1C, D). Even though the results in 1 I excluded 7 studies from this analysis because they involved rats that were either ovariectomized [45] or bred for susceptibility to obesity [44].
Fig. 1A–D have been replicated on several occasions, there are a few studies in which rats on the sweet non-predictive diet did not consume more calories or gain more weight than rats on the sweet predictive diet —e.g., when the maintenance diet consisted of high-fat chow (Fig. 1E and F). A group in Brazil sought to replicate the findings reported in Fig. 1A– D. They used a similar experimental design, with a few changes: the supplemental yogurt was diluted 50% with water, and was provided continuously (5 days a week) [11, 13]. In the Feijó et al. [11] study, the sweet-non-predictive diets were supplemented with yogurt containing saccharin or aspartame, while the sweet predictive diet was supplemented with yogurt containing sucrose. They found that the rats gained significantly more weight on the sweet non-predictive diets, but consumed the same total number of calories from all three diets. One design limitation of the study by Feijó et al. (and the studies by the Purdue University group) is that the investigators simply compared responses to different experimental diets. Without control diets, the investigators lacked a reference point against which to assess body weight changes. As a result they could not determine whether the sweet nonpredictive diet increased body weight, or the sweet predictive diet decreased it. To address this concern, a second study by the Brazilian group [13] offered rats a sweet non-predictive diet (standard chow, water and a dietary supplement of 0.3% saccharin-yogurt) or a control LCS-free diet (standard chow, water and a dietary supplement of unsweetened yogurt). The rats consumed the same number of calories from both diets, but nevertheless gained more weight on the sweet non-predictive diet. Owing to the use of the control LCS-free diet, the authors were able to demonstrate that the combination of LCS and yogurt was necessary to elevate body weight. Taken together, these studies indicate that there is something idiosyncratic about the LCS-treated yogurt, which makes it more obesogenic than the sugar-treated or unsweetened yogurt. Below, I discuss several explanations for this unexpected observation. 2.1. LCSs, sweet taste and cephalic-phase responses Several investigators have focused on the fact that sweet tasting substances in nature (e.g., fruits and honey) typically produce a postingestive spike in blood nutrient levels. To minimize this spike, mammals activate variety of anticipatory (cephalic-phase) responses (or CPRs) [50]. These CPRs are elicited pregastrically by the taste, odor and visual appearance of food [24]. For instance, the sweet taste of sugars and saccharin is known to elicit at least two CPRs in rats: insulin release [3, 48, 49] and thermogenesis [34]. Whereas the cephalic-phase insulin response (CPIR) helps limit blood sugar spikes following a meal, the cephalic-phase thermogenesis helps limit the obesogenic effects of the sugars. When humans ingest an LCS, they experience a sweet taste but no post-ingestive spike in blood nutrient levels [15, 17]. For this reason, LCSs are said to uncouple the sweet taste and post-oral nutritive effects of sugars. Because of this uncoupling, there is no obvious benefit for mammals to generate a CPIR following oral stimulation with LCSs. This reasoning has led several investigators to hypothesize that repeated dietary exposure to an LCS (e.g., in a sweet non-predictive diet) should cause the CPRs to extinguish (review in [38]). A key assumption of this hypothesis is that the LCS-treated yogurt in the sweet nonpredictive diet actually elicits a sucrose-like taste sensation in rats. However, two observations are inconsistent with this assumption. First, when 0.3% saccharin was added to yogurt, it did not stimulate greater intake than unsweetened yogurt [13]. For perspective, when 0.3% saccharin is added to water, it stimulates substantially greater intake than water alone in rats [35]. The most parsimonious explanation for the lack of feeding stimulation by the saccharin-treated yogurt is that the flavor of the yogurt masked the sweet taste of the saccharin. The second observation is that rats exhibit weak to non-existent behavioral attraction to aspartame in water [6, 33]. Accordingly, if aspartame
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
3
Fig. 1. Changes in body weight (A, C, E) and total kcal ingested (B, D, F) over 4 successive weeks in rats maintained on the sweet non-predictive (open circles) or sweet predictive (closed squares) diets. The sweet non-predictive diet consisted of a chow diet supplemented with a saccharin-sweetened yogurt 3 days/week and a non-sweetened yogurt 3 days/week. The sweet predictive diet consisted of a chow diet supplemented with a glucose-sweetened yogurt 3 days/week and a non-sweetened 3 days/week. The chow was standard (left column of panels), high-fat +20% glucose (middle column of panels), or high-fat (right column of panels). The asterisks indicate significant differences (P b 0.05) across diets at specific time points in all panels except B. In panel B, the asterisk indicates a significant main effect of diet on kcal intake (P b 0.05). The data in panels A–B are from Figs. 1 and 3 in [41]; and the data in panels C–F are from Figs. 2 and 3 in [5].
does not produce an attractive (sucrose-like) taste sensation in water, then it is even less likely to do so in yogurt. Taken together, these two observations indicate that neither the saccharin- nor aspartametreated yogurts elicited a sweet taste sensation in the rats. If so, then this would indicate that use of the descriptor “sweet non-predictive diet” is a misnomer in the aforementioned rat studies. More generally, it would indicate that the sweet-taste uncoupling hypothesis is not a relevant conceptual framework for interpreting findings from sweet non-predictive and sweet predictive diets. A related issue is whether repeated dietary exposure to LCSs causes sweet taste-induced CPR to extinguish. One study measured CPIR in rats across 10 successive trials of oral stimulation with 0.15% saccharin (with a 1-h inter-trial interval) [2]. The investigator found that the CPIR magnitude did not change over the trials. Another study provided rats with continuous exposure to a 0.25% saccharin solution for 60 days, and then asked whether this extended period of exposure extinguished the hypoglycemia that typically develops 2 h after oral stimulation with 0.25% saccharin [7]. As predicted, the saccharin-exposed rats displayed less hypoglycemia than water-exposed rats following the 60-day exposure period. The relevance of this latter finding to CPIR is unclear, however, because of the 2 h delay between oral stimulation with saccharin and measurement of blood glucose. Additional studies are needed to determine whether prolonged dietary exposure (e.g., several weeks) to a high preferred concentration of LCS reduces CPIR magnitude in rats or humans. Based on the sweet-uncoupling hypothesis, the Purdue University group predicted that consumption of the sweet non-predictive diet should extinguish CPRs associated with energy regulation ([40]. In support of this hypothesis, they found that rats on the sweet non-predictive diet displayed a larger and more protracted spike in blood glucose (following an oral glucose challenge) than did rats on the sweet predictive diet [42]. This result is difficult to interpret, however, because there was no diet-related change in plasma insulin levels; only a change in GLP-1
levels. Further, in the absence of control diets, one cannot determine whether the sweet non-predictive diet impaired glucose tolerance, or whether the sweet predictive diet enhanced it. A recent study from the Brazil group [13] examined glucose homeostasis in rats maintained one of two diets: the sweet non-predictive diet or an unsweetened control diet. To this end, they measured 12 h fasting blood glucose and insulin levels, beta cell function, and insulin sensitivity. However, they failed to observe any diet-related changes in any of these measures. This indicates that LCSs in the sweet non-predictive diet did not cause metabolic derangement. 2.2. Is caloric intake greater on the sweet non-predictive diet? Many of the studies from the group at Purdue University reported that rats ingest more calories from the non-sweet predictive diet than from the sweet predictive diet. However, the difference in caloric intake was usually quite small (e.g., see Fig. 1B, D); or in some cases, nonexistent (e.g., see Fig. 1F). Likewise, the Brazilian studies reported no difference in caloric intake between rats on (a) the sweet non-predictive diet versus the sweet predictive diet [11], or (b) the sweet non-predictive diet versus a control diet containing plain (unsweetened) yogurt [13]. Taken together, these studies indicate that weight gain on the sweet non-predictive diet is, to varying degrees, uncoupled from caloric intake. 2.3. Is intermittent access to the sweet non-predictive diet necessary for weight gain? The Purdue University group presented the sweet non-predictive diet to rats intermittently (i.e., every other day), whereas the Brazilian group did so continuously (i.e., every day). Despite this difference in presentation schedule, both groups reported that the rats gained more weight on the sweet non-predictive diet than on the sweet predictive
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
4
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
and the unsweetened yogurt control diets. This shows the timing of presentation of the sweet non-predictive diet does not contribute to its obesogenic effects. 2.4. Does LCS-treated yogurt alter the gut microbiota in ways that promote weight gain? A new line of research is examining the impact of LCS consumption on the gut microbiota. For example, dietary exposure to aspartametreated water caused rats to exhibit higher fasting glucose levels and poorer insulin-stimulated glucose disposal than control rats [28]. The investigators attributed the altered glucose regulation to changes in the gut microbiota. This is because consumption of the aspartame solution was associated with changes in both bacterial count and species composition of fecal samples. The fact that aspartame (a dipeptide) caused these changes is remarkable given that it is rapidly degraded to aspartic acid, phenylalanine and methanol by enzymes as soon as it arrives in the small intestine [23]. A second study exposed mice to water treated with one of several LCSs [36]. Afterwards, the investigators conducted oral glucose challenges. They found that exposure to the LCS-treated solutions was associated with impaired glucose tolerance. The authors linked the impaired glucose intolerance to changes in the gut microbiota by showing that consumption of at least one LCS (saccharin) altered the composition of the fecal microbiota. The gut microbiota studies [28, 36] are exciting because they provide a mechanism by which LCSs could alter glucose metabolism and weight gain (e.g., see [27]), independently of caloric intake. For instance, the LCSs could change the gut microbiotia, which in turn could alter nutrient processing and gut signaling [37]. Despite the compelling nature of this new line of work, two caveats should be highlighted. First, when rats were offered the sweet non-predictive diet, the LCSs were presented in yogurt. Given that yogurt itself alters the gut microbiota [26], the impact of LCSs on the gut microbiota may have been influenced by the yogurt. Second, the reports of LCS-induced changes in gut microbiota [28, 36] were based on fecal samples. A recent study found that the composition of the microbiota varies along the length of the gastrointestinal tract, and that fecal samples largely reflect the microbiota in the distal colon [21]. It follows that the microbial communities in fecal samples bear little resemblance to those in the small intestine, which is where most nutrient processing and absorption occurs.
they do not elicit a CPIR [1, 10, 19, 47]. Thus, it would appear that if LCSs elicit CPIRs (and perhaps other cephalic-phase responses) in humans, they are weak and unreliable. If so, then the concern that consumption of LCS-treated substances will cause CPRs to extinguish [38] may be exaggerated. 4. Conclusion This review examined the evidence underlying the claim that cause elevated body weight in rodents. While most of the published studies contradict this claim [31], a subset of the rat studies presented the LCS in yogurt and reported a strong link between LCS consumption and weight gain. Because the rats did not appear to have tasted the LCS in the yogurt, the sweet taste-uncoupling hypothesis does not appear to be a useful framework for analyzing the phenomenon. As an alternative, there is evidence linking changes in the gut microbiota to weight gain in rodents [9, 27]. Given that LCS and yogurt (with active cultures) each alter the gut microbiota in unique ways [26, 36], it is possible that the combination of LCS and yogurt produces idiosyncratic microbiotic changes, which promote weight gain in rodents. Two recent systematic reviews examined the linkage between LCS use and changes in energy intake and body weight in humans [25, 31]. When the authors limited their meta-analyses to studies that used the most rigorous experimental design—randomized control trials (RCTs) —they found no support for this linkage. In fact, the meta-analyses indicated that replacing sugars with LCSs significantly reduced energy intake and body weight in both children and adults. While these metaanalyses provide compelling evidence that LCSs represent an effective tool for helping reduce body weight in humans, it is notable that the RCTs presented the LCSs in a beverage (e.g., soda). In light of the rodent studies, it might be worthwhile to explore the impact of LCS-treated yogurt in humans. Acknowledgements I thank Anthony Sclafani for sharing his insights into the literature on LCS and ingestive behavior, and two anonymous reviewers for their helpful feedback. References
3. Are results from the sweet non-predictive diets relevant to humans? There are several reasons why results from the sweet predictive and non-predictive diets may not be relevant to humans. First, humans typically consume a wide variety of foods, which differ greatly in caloric content. Indeed, some people ingest sugar- and LCS-sweetened substances within the same meal. This type of diet may help people learn how to discriminate foods that differ in caloric content. In contrast, laboratory rodents typically subsist on a monotonous diet, consisting of chow and water for their entire life. This profound interspecific difference in dietary experience raises questions about the relevance of LCS feeding responses in rodents to those in humans [31]. Second, the groups from Purdue University and Brazil have both reported that rats experienced elevated weight gain when the sweet nonpredictive diet contained aspartame- or saccharin-treated yogurt. However, as noted above, it is unlikely that these LCS-containing yogurts actually tasted sweet to the rats. In contrast, most human food products containing LCSs (e.g., beverages and yogurts) are manufactured so that they elicit a salient sweet taste. Thus, the mechanisms stimulating intake of the LCS-treated yogurt in rats likely differ from those stimulating intake of LCS-treated food products in humans. Third, LCSs have been reported to elicit CPIR in rats [2, 29, 48], but their ability to do so in humans is less clear. One study reported that LCSs elicit a weak CPIR in humans [20], but four others reported that
[1] L. Abdallah, M. Chabert, J. Louis-Sylvestre, Cephalic phase responses to sweet taste, Am. J. Clin. Nutr. 65 (1997) 737–743. [2] H.-R. Berthoud, D.A. Bereiter, E.R. Trimble, E.G. Siegel, B. Jeanrenaud, Cephalic phase, reflex insulin secretion: neuroanatomical and physiological characterization, Diabetologia 20 (1981) 393–401. [3] H.-R. Berthoud, E.R. Trimble, E.G. Siegel, D.A. Bereiter, B. Jeanrenaud, Cephalic-phase insulin secretion in normal and pancreatic islet-transplanted rats, Am. J. Physiol. 238 (1980) E336–E340. [4] D. Brusick, J.F. Borzelleca, M. Gallo, G. Williams, J. Kille, A.W. Hayes, F.X. Pi-Sunyer, C. Williams, W. Burks, Expert Panel report on a study of Splenda in male rats, Regul. Toxicol. Pharmacol. 55 (2009) 6–12. [5] T.L. Davidson, A.A. Martin, K. Clark, S.E. Swithers, Intake of high-intensity sweeteners alters the ability of sweet taste to signal caloric consequences: implications for the learned control of energy and body weight regulation, Quart. J. Exp. Psychol. 64 (2011) 1430–1441. [6] J.C. De Francisco, N.K. Dess, Aspartame consumption in rats selectively bred for high versus low saccharin intake, Physiol. Behav. 65 (1998) 393–396. [7] R. Deutch, Conditioned hypoglycemia: a mechanism for saccharin-induced sensitivity to insulin in the rat, J. Comp. Physiol. Psychol. 86 (1974) 350–358. [8] G.E. DuBois, I. Prakash, Non-caloric sweeteners, sweetness modulators, and sweetener enhancers, Annu. Rev. Food Sci. Technol. 3 (2012) 353–380. [9] F.A. Duca, Y. Sakar, P. Lepage, F. Devime, B. Langelier, J. Doré, M. Covasa, Replication of obesity and associated signaling pathways through transfer of microbiota from obese-prone rats, Diabetes 63 (2014) 1624–1636. [10] M. Dušková, M. Macourek, M. Šrámková, M. Hill, L. Stárka, The role of taste in cephalic phase of insulin secretion, Prague Med. Rep. 114 (2013) 222–230. [11] F.D. Feijó, C.R. Ballard, K.C. Foletto, B.A.M. Batista, A.M. Neves, M.F.M. Ribeiro, M.C. Bertoluci, Saccharin and aspartame, compared with sucrose, induce greater weight gain in adult Wistar rats, at similar total caloric intake levels, Appetite 60 (2013) 203–207. [12] C. Fitch, Keim K.S., Position of the Academy of Nutrition and Dietetics: use of nutritive and nonnutritive sweeteners, J. Acad. Nutr. Diet. 112 (2012) 739–758.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx [13] K.C. Foletto, B.A.M. Batista, A.M. Neves, F.D. Feijó, C.R. Ballard, M.F.M. Ribeiro, M.C. Bertoluci, Sweet taste of saccharin induces weight gain without increasing caloric intake, not related to insulin-resistance in Wistar rats, Appetite 96 (2016) 604–610. [15] H.E. Ford, V. Peters, N.M. Martin, M.L. Sleeth, M.A. Ghatei, G.S. Frost, S.R. Bloom, Effects of oral ingestion of sucralose on gut hormone response and appetite in healthy normal-weight subjects, Eur. J. Clin. Nutr. 65 (2011) 508–513. [16] S.P. Fowler, K. Williams, R.G. Resendez, K.J. Hunt, H.P. Hazuda, M.P. Stern, Fueling the obesity epidemic? Artificially sweetened beverage use and long-term weight gain, Obesity 16 (2008) 1894–1900. [17] Y. Fujita, R.D. Wideman, M. Speck, A. Asadi, D.S. King, T.D. Webber, M. Haneda, T.J. Kieffer, Incretin release from gut is acutely enhanced by sugar but not by sweeteners in vivo, Am. J. Physiol. 296 (2009) E473–E479. [18] V.L. Grotz, I.C. Munro, An overview of the safety of sucralose, Regul. Toxicol. Pharmacol. 55 (2009) 1–5. [19] B. Härtel, H.-J. Graubaum, B. Schneider, The influence of sweetener solutions on the secretion of insulin and the blood glucose level, Ernährungs-Umschau. 40 (1993) 152–155. [20] T. Just, H.W. Pau, U. Engel, T. Hummel, Cephalic phase insulin release in healthy humans after taste stimulation? Appetite 51 (2008) 622–627. [21] J.S. Lichtman, E. Alsentzer, M. Jaffe, D. Sprockett, E. Masutani, E. Ikwa, G.K. Fragiadakis, D. Clifford, B.E. Huang, J.L. Sonnenburg, K.C. Huang, J.E. Elias, The effect of microbial colonization on the host proteome varies by gastrointestinal location, ISME J. (2015). http://dx.doi.org/10.1038/ismej.2015.187. [22] D.S. Ludwig, Artificially sweetened beverages: cause for concern, JAMA 301 (2009) 2477–2478. [23] B.A. Magnuson, G.A. Burdock, J. Doull, R.M. Kroes, G.M. Marsh, M.W. Pariza, P.S. Spencer, W.J. Waddell, R. Walker, G.M. Williams, Aspartame: a safety evaluation based on current use levels, regulations, and toxicological and epidemiological studies, Crit. Rev. Toxicol. 37 (2007) 629–727. [24] R.D. Mattes, Physiological responses to sensory stimulation by food: nutritional implications, J. Am. Diet. Assoc. 97 (1997) 406–413. [25] P.E. Miller, V. Perez, Low-calorie sweeteners and body weight and composition: a meta-analysis of randomized controlled trials and prospective cohort studies, Am. J. Clin. Nutr. 100 (2014) 765–777. [26] F. Pace, M. Pace, G. Quartarone, Probiotics in digestive diseases: focus on Lactobacillus GG, Minerva Gastroenterol. Dietol. 61 (2015) 273–292. [27] Palau-Rodriguez, M., Tulipani, S., Queipo-Ortuño, M. I., Mireia Urpi-Sarda, Tinahones, F. J., Andres-Lacueva, C. Metabolomic insights into the intricate gut microbial–host interaction in the development of obesity and type-2 diabetes. Front. Microbiol. 2015, 6: article 1151. [28] M.S.A. Palmnäs, T.E. Cowan, M.R. Bomhof, J. Su, R.A. Reimer, H.J. Vogel, D.S. Hittel, J. Shearer, Low-dose aspartame consumption differentially affects gut microbiotahost metabolic interactions in the diet-induced obese rat, PLoS One 9 (2014), e109841. [29] T.L. Powley, H.-R. Berthoud, Diet and cephalic phase insulin responses, Am. J. Clin. Nutr. 42 (1985) 991–1002. [30] A.G. Renwick, The metabolism of intense sweeteners, Xenobiot. 16 (1986) 1057–1071. [31] P.J. Rogers, P.S. Hogenkamp, K. de Graaf, S. Higgs, A. Lluch, A.R. Ness, C. Penfold, R. Perry, P. Putz, M.R. Yeomans, D.J. Mela, Does low-energy sweetener consumption affect energy intake and body weight? A systematic review, including meta-analyses,
[32] [33] [34]
[35] [36]
[37]
[38] [39]
[40] [41] [42]
[43]
[44]
[45]
[46] [47] [48] [49] [50]
5
of the evidence from human and animal studies, Int. J. Obes. (Lond) (2015). http:// dx.doi.org/10.1038/ijo.2015.177. S.S. Schiffman, Rationale for further medical and health research on high-potency sweeteners, Chem. Senses 37 (2012) 671–679. A. Sclafani, M. Abrams, Rats show only a weak preference for the artificial sweetener aspartame, Physiol. Behav. 37 (1986) 253–256. K. Shinozaki, Y. Shimizu, T. Shiina, H. Morita, T. Takewaki, Relationship between taste-induced physiological reflexes and temperature of sweet taste, Physiol. Behav. 93 (2008) 1000–1004. J.C. Smith, A. Sclafani, Saccharin as a sugar surrogate revisted, Appetite 38 (2002) 155–160. J. Suez, T. Korem, D. Zeevi, G. Zilberman-Schapira, C.A. Thaiss, O. Maza, D. Israeli, N. Zmora, S. Gilad, A. Weinberger, Y. Kuperman, A. Harmelin, I. Kolodkin-Gal, H. Shapiro, Z. Halpern, E. Segal, E. Elinav, Artificial sweeteners induce glucose intolerance by altering the gutmicrobiota, Nature 514 (2014) 181–186. T.D. Swartz, F.A. Duca, T. de Wouters, Y. Sakar, M. Covasa, Up-regulation of intestinal type 1 taste receptor 3 and sodium glucose luminal transporter-1 expression and increased sucrose intake in mice lacking gut microbiota, British J. Nutr. 107 (2012) 621–630. S.E. Swithers, Artificial sweeteners produce the counterintuitive effect of inducing metabolic derangements, Trends Endocrinol. Metab. 24 (2013) 431–441. S.E. Swithers, C.R. Baker, T.L. Davidson, General and persistent effects of highintensity sweeteners on body weight gain and caloric compensation in rats, Behav. Neurosci. 123 (2009) 772–780. S.E. Swithers, T.L. Davidson, Obesity: outwitting the wisdom of the body? Curr. Neurol. Neurosci. Rep. 5 (2005) 159–162. S.E. Swithers, T.L. Davidson, A role for sweet taste: calorie predictive relations in energy regulation by rats, Behav. Neurosci. 122 (2008) 161–173. S.E. Swithers, A.F. Laboy, K. Clark, S. Cooper, T.L. Davidson, Experience with the highintensity sweetener saccharin impairs glucose homeostasis and GLP-1 release in rats, Behav. Brain Res. 233 (2012) 1–14. S.E. Swithers, A.A. Martin, K.M. Clark, A.F. Laboy, T.L. Davidson, Body weight gain in rats consuming sweetened liquids. Effects of caffeine and diet composition, Appetite 55 (2010) 528–533. S.E. Swithers, C.H. Sample, T.L. Davidson, Adverse effects of high-intensity sweeteners on energy intake and weight control in male and obesity-prone female rats, Behav. Neurosci. 127 (2013) 262–274. S.E. Swithers, C.H. Sample, D.P. Katz, Influence of ovarian and non-ovarian estrogens on weight gain in response to disruption of sweet taste-calorie relations in female rats, Horm. Behav. 63 (2013) 40–48. K.R. Tandel, Sugar substitutes: health controversy over perceived benefits, J. Pharmacol. Pharmacother. 2 (2011) 236–243. K.L. Teff, J. Devine, K. Engelman, Sweet taste: effect on cephalic phase insulin release in men, Physiol. Behav. 57 (1995) 1089–1095. K. Tonosaki, Y. Hori, Y. Shimizu, K. Tonosaki, Relationship between insulin release and taste, Biomed. Res. 28 (2007) 79–83. M.G. Tordoff, M.I. Friedman, Drinking saccharin increases food intake and preference-IV. Cephalic phase and metabolic factors, Appetite 12 (1989) 37–56. S.C. Woods, The eating paradox: how we tolerate food, Psychol. Rev. 98 (1991) 488–505.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Review
Do low-calorie sweeteners promote weight gain in rodents? John I. Glendinning Department of Biology, Barnard College, Columbia University, 3009 Broadway, New York, NY 10027, United States
H I G H L I G H T S • • • • •
The linkage between LCS consumption and elevated body weight in rodents is examined. LCSs promoted weight gain when they were presented in yogurt. The LCS-treated yogurt formulations did not appear to taste sweet to the rats. The elevated weight gain could not be explained solely by increased caloric intake. LCS and yogurt may promote weight gain by modifying the gut microbiota.
a r t i c l e
i n f o
Article history: Received 2 October 2015 Received in revised form 28 January 2016 Accepted 29 January 2016 Available online xxxx Keywords: Low-calorie sweeteners Caloric intake Weight gain Rat Yogurt Gut microbiota
a b s t r a c t Low-calorie sweeteners (LCSs) are used globally to increase the palatability of foods and beverages, without the calories of sugar. Recently, however, there have been claims that LCSs promote obesity. Here, I review the literature linking LCS consumption to elevated body weight in rodents. A recent systematic review found when the LCSs were presented in water or chow, only a minority of the studies reported elevated weight gain. In contrast, when the LCSs were presented in yogurt, the majority of the studies reported elevated weight gain. This review focuses on this latter subset of studies, and asks why the combination of LCSs and yogurt promoted weight gain. First, LCSs have been hypothesized to induce metabolic derangement because they uncouple sweet taste and calories. However, the available evidence indicates that the LCS-treated yogurts did not actually taste sweet to rats in the published studies. Without a sweet taste, the concerns about uncoupling sweet taste and calories would not be relevant. Second, in several studies, the LCS-treated yogurt increased weight gain without increasing caloric intake. This indicates that caloric intake alone cannot explain the elevated weight gain. Third, there is evidence that LCSs and yogurt can each alter the gut microbiota of rodents. Given recent work indicating that changes in gut microbiota can modulate body weight, it is possible that the combination of LCS and yogurt alters the gut microbiota in ways that promote weight gain. While this hypothesis remains speculative, it is consistent with the observed rodent data. In human studies, LCSs are usually presented in beverages. Based on the rodent work, it might be worthwhile to evaluate the impact of LCS-treated yogurt in humans. © 2016 Elsevier Inc. All rights reserved.
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Under what experimental conditions do LCSs increase weight gain? . . . . . . . . . . . 2.1. LCSs, sweet taste and cephalic-phase responses . . . . . . . . . . . . . . . . . 2.2. Is caloric intake greater on the sweet non-predictive diet? . . . . . . . . . . . . 2.3. Is intermittent access to the sweet non-predictive diet necessary for weight gain? . 2.4. Does LCS-treated yogurt alter the gut microbiota in ways that promote weight gain? 3. Are results from the sweet non-predictive diets relevant to humans? . . . . . . . . . . 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
E-mail address: [email protected].
http://dx.doi.org/10.1016/j.physbeh.2016.01.043 0031-9384/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
0 0 0 0 0 0 0 0 0 0
2
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
1. Introduction Low-calorie sweeteners (LCSs) are attractive to consumers because they make foods and beverages taste better, without the caloric or glycemic effects of sugar [8]. Despite widespread usage, LCSs remain controversial. There are persistent claims that LCSs cause adverse health effects, including cancer, neurotoxicity, allergic reactions, elevated caloric intake and obesity (review in [32, 38]). The claims of cancer, neurotoxicity and allergic reactions are contradicted, however, by a large body of empirical work (reviews in [4, 18, 23, 30, 46]) and the fact that LCSs have been approved and recommend for use by regulatory bodies (e.g., U.S. Food and Drug Administration, European Food Safety Authority, World Health Organization) and the Academy of Nutrition and Dietetics [12]. The claim that LCSs increase energy intake and body weight is based on the notion that repeated LCS use disrupts the mechanisms underlying energy regulation [16, 22, 38]. Here, I focus on empirical support for the proposed linkage between LCS consumption and weight gain in rodents. First, I discuss the specific experimental conditions under which LCS have been found to increase body weight. Second, I consider the limitations of the animal studies linking LCSs to overconsumption and weight gain. Third, I examine the extent to which results from the animal studies are relevant to humans. 2. Under what experimental conditions do LCSs increase weight gain? To determine whether LCSs increase body weight in rodents, investigators have used several experimental approaches. In a recent systemic review of this literature, Rogers et al. [31] identified a total of 90 relevant studies, and categorized them according to one of three experimental designs. In design 1, the LCS was added to the rodents' only source of food or water (n = 47 studies). In design 2, the rodents were provided with a standard diet (chow and water) plus continuous access to an LCS-sweetened water, diet or yogurt (n = 21 studies). In design 3, the rodents were offered a standard diet plus intermittent access to an LCS-treated yogurt (n = 15 studies1). Only a small percentage of the studies that used design 1 (9%) or design 2 (24%) reported a significant increase in body weight [31]. In contrast, the vast majority of studies that used design 3 (87%) reported significant increases in body weight. The fact that so few of the studies that used designs 1 or 2 reported weight gain indicates that LCSs are not inherently obesogenic to rodents. Below I examine design 3 in greater detail so as to gain insight into why it was more likely to cause weight gain. The majority of the studies that used design 3 (i.e., [5, 39, 41–43]) were conducted by a group at Purdue University. In most cases, the investigators exposed rats to one of two diets. The “sweet predictive” diet consisted of a maintenance diet (standard chow and water ad libitum) plus a 30 g supplement of low-fat yogurt per day. The yogurt was offered 6 days/week. On 3 of the days, it was sweetened with 20% glucose; and on the other 3 days, it was unsweetened. This diet was called sweet predictive because the sweet taste of the glucose predicted the presence of sugar calories in the yogurt. The “sweet non-predictive” diet was identical, except that the sweetened yogurt contained 0.3% saccharin instead of glucose. This latter diet was called sweet non-predictive because the sweet taste of the saccharin did not predict the presence of sugar calories. As compared with rats on the sweet predictive diet, those on the sweet non-predictive diet gained more weight and (in many cases) ingested slightly but significantly more calories (Fig. 1A, B). Likewise, when the maintenance diet consisted of a high-fat chow sweetened with 20% glucose [5], the rats on the sweet non-predictive diet gained more weight and ingested more kcal/week than rats on the sweet predictive diet (Fig. 1C, D). Even though the results in 1 I excluded 7 studies from this analysis because they involved rats that were either ovariectomized [45] or bred for susceptibility to obesity [44].
Fig. 1A–D have been replicated on several occasions, there are a few studies in which rats on the sweet non-predictive diet did not consume more calories or gain more weight than rats on the sweet predictive diet —e.g., when the maintenance diet consisted of high-fat chow (Fig. 1E and F). A group in Brazil sought to replicate the findings reported in Fig. 1A– D. They used a similar experimental design, with a few changes: the supplemental yogurt was diluted 50% with water, and was provided continuously (5 days a week) [11, 13]. In the Feijó et al. [11] study, the sweet-non-predictive diets were supplemented with yogurt containing saccharin or aspartame, while the sweet predictive diet was supplemented with yogurt containing sucrose. They found that the rats gained significantly more weight on the sweet non-predictive diets, but consumed the same total number of calories from all three diets. One design limitation of the study by Feijó et al. (and the studies by the Purdue University group) is that the investigators simply compared responses to different experimental diets. Without control diets, the investigators lacked a reference point against which to assess body weight changes. As a result they could not determine whether the sweet nonpredictive diet increased body weight, or the sweet predictive diet decreased it. To address this concern, a second study by the Brazilian group [13] offered rats a sweet non-predictive diet (standard chow, water and a dietary supplement of 0.3% saccharin-yogurt) or a control LCS-free diet (standard chow, water and a dietary supplement of unsweetened yogurt). The rats consumed the same number of calories from both diets, but nevertheless gained more weight on the sweet non-predictive diet. Owing to the use of the control LCS-free diet, the authors were able to demonstrate that the combination of LCS and yogurt was necessary to elevate body weight. Taken together, these studies indicate that there is something idiosyncratic about the LCS-treated yogurt, which makes it more obesogenic than the sugar-treated or unsweetened yogurt. Below, I discuss several explanations for this unexpected observation. 2.1. LCSs, sweet taste and cephalic-phase responses Several investigators have focused on the fact that sweet tasting substances in nature (e.g., fruits and honey) typically produce a postingestive spike in blood nutrient levels. To minimize this spike, mammals activate variety of anticipatory (cephalic-phase) responses (or CPRs) [50]. These CPRs are elicited pregastrically by the taste, odor and visual appearance of food [24]. For instance, the sweet taste of sugars and saccharin is known to elicit at least two CPRs in rats: insulin release [3, 48, 49] and thermogenesis [34]. Whereas the cephalic-phase insulin response (CPIR) helps limit blood sugar spikes following a meal, the cephalic-phase thermogenesis helps limit the obesogenic effects of the sugars. When humans ingest an LCS, they experience a sweet taste but no post-ingestive spike in blood nutrient levels [15, 17]. For this reason, LCSs are said to uncouple the sweet taste and post-oral nutritive effects of sugars. Because of this uncoupling, there is no obvious benefit for mammals to generate a CPIR following oral stimulation with LCSs. This reasoning has led several investigators to hypothesize that repeated dietary exposure to an LCS (e.g., in a sweet non-predictive diet) should cause the CPRs to extinguish (review in [38]). A key assumption of this hypothesis is that the LCS-treated yogurt in the sweet nonpredictive diet actually elicits a sucrose-like taste sensation in rats. However, two observations are inconsistent with this assumption. First, when 0.3% saccharin was added to yogurt, it did not stimulate greater intake than unsweetened yogurt [13]. For perspective, when 0.3% saccharin is added to water, it stimulates substantially greater intake than water alone in rats [35]. The most parsimonious explanation for the lack of feeding stimulation by the saccharin-treated yogurt is that the flavor of the yogurt masked the sweet taste of the saccharin. The second observation is that rats exhibit weak to non-existent behavioral attraction to aspartame in water [6, 33]. Accordingly, if aspartame
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
3
Fig. 1. Changes in body weight (A, C, E) and total kcal ingested (B, D, F) over 4 successive weeks in rats maintained on the sweet non-predictive (open circles) or sweet predictive (closed squares) diets. The sweet non-predictive diet consisted of a chow diet supplemented with a saccharin-sweetened yogurt 3 days/week and a non-sweetened yogurt 3 days/week. The sweet predictive diet consisted of a chow diet supplemented with a glucose-sweetened yogurt 3 days/week and a non-sweetened 3 days/week. The chow was standard (left column of panels), high-fat +20% glucose (middle column of panels), or high-fat (right column of panels). The asterisks indicate significant differences (P b 0.05) across diets at specific time points in all panels except B. In panel B, the asterisk indicates a significant main effect of diet on kcal intake (P b 0.05). The data in panels A–B are from Figs. 1 and 3 in [41]; and the data in panels C–F are from Figs. 2 and 3 in [5].
does not produce an attractive (sucrose-like) taste sensation in water, then it is even less likely to do so in yogurt. Taken together, these two observations indicate that neither the saccharin- nor aspartametreated yogurts elicited a sweet taste sensation in the rats. If so, then this would indicate that use of the descriptor “sweet non-predictive diet” is a misnomer in the aforementioned rat studies. More generally, it would indicate that the sweet-taste uncoupling hypothesis is not a relevant conceptual framework for interpreting findings from sweet non-predictive and sweet predictive diets. A related issue is whether repeated dietary exposure to LCSs causes sweet taste-induced CPR to extinguish. One study measured CPIR in rats across 10 successive trials of oral stimulation with 0.15% saccharin (with a 1-h inter-trial interval) [2]. The investigator found that the CPIR magnitude did not change over the trials. Another study provided rats with continuous exposure to a 0.25% saccharin solution for 60 days, and then asked whether this extended period of exposure extinguished the hypoglycemia that typically develops 2 h after oral stimulation with 0.25% saccharin [7]. As predicted, the saccharin-exposed rats displayed less hypoglycemia than water-exposed rats following the 60-day exposure period. The relevance of this latter finding to CPIR is unclear, however, because of the 2 h delay between oral stimulation with saccharin and measurement of blood glucose. Additional studies are needed to determine whether prolonged dietary exposure (e.g., several weeks) to a high preferred concentration of LCS reduces CPIR magnitude in rats or humans. Based on the sweet-uncoupling hypothesis, the Purdue University group predicted that consumption of the sweet non-predictive diet should extinguish CPRs associated with energy regulation ([40]. In support of this hypothesis, they found that rats on the sweet non-predictive diet displayed a larger and more protracted spike in blood glucose (following an oral glucose challenge) than did rats on the sweet predictive diet [42]. This result is difficult to interpret, however, because there was no diet-related change in plasma insulin levels; only a change in GLP-1
levels. Further, in the absence of control diets, one cannot determine whether the sweet non-predictive diet impaired glucose tolerance, or whether the sweet predictive diet enhanced it. A recent study from the Brazil group [13] examined glucose homeostasis in rats maintained one of two diets: the sweet non-predictive diet or an unsweetened control diet. To this end, they measured 12 h fasting blood glucose and insulin levels, beta cell function, and insulin sensitivity. However, they failed to observe any diet-related changes in any of these measures. This indicates that LCSs in the sweet non-predictive diet did not cause metabolic derangement. 2.2. Is caloric intake greater on the sweet non-predictive diet? Many of the studies from the group at Purdue University reported that rats ingest more calories from the non-sweet predictive diet than from the sweet predictive diet. However, the difference in caloric intake was usually quite small (e.g., see Fig. 1B, D); or in some cases, nonexistent (e.g., see Fig. 1F). Likewise, the Brazilian studies reported no difference in caloric intake between rats on (a) the sweet non-predictive diet versus the sweet predictive diet [11], or (b) the sweet non-predictive diet versus a control diet containing plain (unsweetened) yogurt [13]. Taken together, these studies indicate that weight gain on the sweet non-predictive diet is, to varying degrees, uncoupled from caloric intake. 2.3. Is intermittent access to the sweet non-predictive diet necessary for weight gain? The Purdue University group presented the sweet non-predictive diet to rats intermittently (i.e., every other day), whereas the Brazilian group did so continuously (i.e., every day). Despite this difference in presentation schedule, both groups reported that the rats gained more weight on the sweet non-predictive diet than on the sweet predictive
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
4
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx
and the unsweetened yogurt control diets. This shows the timing of presentation of the sweet non-predictive diet does not contribute to its obesogenic effects. 2.4. Does LCS-treated yogurt alter the gut microbiota in ways that promote weight gain? A new line of research is examining the impact of LCS consumption on the gut microbiota. For example, dietary exposure to aspartametreated water caused rats to exhibit higher fasting glucose levels and poorer insulin-stimulated glucose disposal than control rats [28]. The investigators attributed the altered glucose regulation to changes in the gut microbiota. This is because consumption of the aspartame solution was associated with changes in both bacterial count and species composition of fecal samples. The fact that aspartame (a dipeptide) caused these changes is remarkable given that it is rapidly degraded to aspartic acid, phenylalanine and methanol by enzymes as soon as it arrives in the small intestine [23]. A second study exposed mice to water treated with one of several LCSs [36]. Afterwards, the investigators conducted oral glucose challenges. They found that exposure to the LCS-treated solutions was associated with impaired glucose tolerance. The authors linked the impaired glucose intolerance to changes in the gut microbiota by showing that consumption of at least one LCS (saccharin) altered the composition of the fecal microbiota. The gut microbiota studies [28, 36] are exciting because they provide a mechanism by which LCSs could alter glucose metabolism and weight gain (e.g., see [27]), independently of caloric intake. For instance, the LCSs could change the gut microbiotia, which in turn could alter nutrient processing and gut signaling [37]. Despite the compelling nature of this new line of work, two caveats should be highlighted. First, when rats were offered the sweet non-predictive diet, the LCSs were presented in yogurt. Given that yogurt itself alters the gut microbiota [26], the impact of LCSs on the gut microbiota may have been influenced by the yogurt. Second, the reports of LCS-induced changes in gut microbiota [28, 36] were based on fecal samples. A recent study found that the composition of the microbiota varies along the length of the gastrointestinal tract, and that fecal samples largely reflect the microbiota in the distal colon [21]. It follows that the microbial communities in fecal samples bear little resemblance to those in the small intestine, which is where most nutrient processing and absorption occurs.
they do not elicit a CPIR [1, 10, 19, 47]. Thus, it would appear that if LCSs elicit CPIRs (and perhaps other cephalic-phase responses) in humans, they are weak and unreliable. If so, then the concern that consumption of LCS-treated substances will cause CPRs to extinguish [38] may be exaggerated. 4. Conclusion This review examined the evidence underlying the claim that cause elevated body weight in rodents. While most of the published studies contradict this claim [31], a subset of the rat studies presented the LCS in yogurt and reported a strong link between LCS consumption and weight gain. Because the rats did not appear to have tasted the LCS in the yogurt, the sweet taste-uncoupling hypothesis does not appear to be a useful framework for analyzing the phenomenon. As an alternative, there is evidence linking changes in the gut microbiota to weight gain in rodents [9, 27]. Given that LCS and yogurt (with active cultures) each alter the gut microbiota in unique ways [26, 36], it is possible that the combination of LCS and yogurt produces idiosyncratic microbiotic changes, which promote weight gain in rodents. Two recent systematic reviews examined the linkage between LCS use and changes in energy intake and body weight in humans [25, 31]. When the authors limited their meta-analyses to studies that used the most rigorous experimental design—randomized control trials (RCTs) —they found no support for this linkage. In fact, the meta-analyses indicated that replacing sugars with LCSs significantly reduced energy intake and body weight in both children and adults. While these metaanalyses provide compelling evidence that LCSs represent an effective tool for helping reduce body weight in humans, it is notable that the RCTs presented the LCSs in a beverage (e.g., soda). In light of the rodent studies, it might be worthwhile to explore the impact of LCS-treated yogurt in humans. Acknowledgements I thank Anthony Sclafani for sharing his insights into the literature on LCS and ingestive behavior, and two anonymous reviewers for their helpful feedback. References
3. Are results from the sweet non-predictive diets relevant to humans? There are several reasons why results from the sweet predictive and non-predictive diets may not be relevant to humans. First, humans typically consume a wide variety of foods, which differ greatly in caloric content. Indeed, some people ingest sugar- and LCS-sweetened substances within the same meal. This type of diet may help people learn how to discriminate foods that differ in caloric content. In contrast, laboratory rodents typically subsist on a monotonous diet, consisting of chow and water for their entire life. This profound interspecific difference in dietary experience raises questions about the relevance of LCS feeding responses in rodents to those in humans [31]. Second, the groups from Purdue University and Brazil have both reported that rats experienced elevated weight gain when the sweet nonpredictive diet contained aspartame- or saccharin-treated yogurt. However, as noted above, it is unlikely that these LCS-containing yogurts actually tasted sweet to the rats. In contrast, most human food products containing LCSs (e.g., beverages and yogurts) are manufactured so that they elicit a salient sweet taste. Thus, the mechanisms stimulating intake of the LCS-treated yogurt in rats likely differ from those stimulating intake of LCS-treated food products in humans. Third, LCSs have been reported to elicit CPIR in rats [2, 29, 48], but their ability to do so in humans is less clear. One study reported that LCSs elicit a weak CPIR in humans [20], but four others reported that
[1] L. Abdallah, M. Chabert, J. Louis-Sylvestre, Cephalic phase responses to sweet taste, Am. J. Clin. Nutr. 65 (1997) 737–743. [2] H.-R. Berthoud, D.A. Bereiter, E.R. Trimble, E.G. Siegel, B. Jeanrenaud, Cephalic phase, reflex insulin secretion: neuroanatomical and physiological characterization, Diabetologia 20 (1981) 393–401. [3] H.-R. Berthoud, E.R. Trimble, E.G. Siegel, D.A. Bereiter, B. Jeanrenaud, Cephalic-phase insulin secretion in normal and pancreatic islet-transplanted rats, Am. J. Physiol. 238 (1980) E336–E340. [4] D. Brusick, J.F. Borzelleca, M. Gallo, G. Williams, J. Kille, A.W. Hayes, F.X. Pi-Sunyer, C. Williams, W. Burks, Expert Panel report on a study of Splenda in male rats, Regul. Toxicol. Pharmacol. 55 (2009) 6–12. [5] T.L. Davidson, A.A. Martin, K. Clark, S.E. Swithers, Intake of high-intensity sweeteners alters the ability of sweet taste to signal caloric consequences: implications for the learned control of energy and body weight regulation, Quart. J. Exp. Psychol. 64 (2011) 1430–1441. [6] J.C. De Francisco, N.K. Dess, Aspartame consumption in rats selectively bred for high versus low saccharin intake, Physiol. Behav. 65 (1998) 393–396. [7] R. Deutch, Conditioned hypoglycemia: a mechanism for saccharin-induced sensitivity to insulin in the rat, J. Comp. Physiol. Psychol. 86 (1974) 350–358. [8] G.E. DuBois, I. Prakash, Non-caloric sweeteners, sweetness modulators, and sweetener enhancers, Annu. Rev. Food Sci. Technol. 3 (2012) 353–380. [9] F.A. Duca, Y. Sakar, P. Lepage, F. Devime, B. Langelier, J. Doré, M. Covasa, Replication of obesity and associated signaling pathways through transfer of microbiota from obese-prone rats, Diabetes 63 (2014) 1624–1636. [10] M. Dušková, M. Macourek, M. Šrámková, M. Hill, L. Stárka, The role of taste in cephalic phase of insulin secretion, Prague Med. Rep. 114 (2013) 222–230. [11] F.D. Feijó, C.R. Ballard, K.C. Foletto, B.A.M. Batista, A.M. Neves, M.F.M. Ribeiro, M.C. Bertoluci, Saccharin and aspartame, compared with sucrose, induce greater weight gain in adult Wistar rats, at similar total caloric intake levels, Appetite 60 (2013) 203–207. [12] C. Fitch, Keim K.S., Position of the Academy of Nutrition and Dietetics: use of nutritive and nonnutritive sweeteners, J. Acad. Nutr. Diet. 112 (2012) 739–758.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043
J.I. Glendinning / Physiology & Behavior xxx (2016) xxx–xxx [13] K.C. Foletto, B.A.M. Batista, A.M. Neves, F.D. Feijó, C.R. Ballard, M.F.M. Ribeiro, M.C. Bertoluci, Sweet taste of saccharin induces weight gain without increasing caloric intake, not related to insulin-resistance in Wistar rats, Appetite 96 (2016) 604–610. [15] H.E. Ford, V. Peters, N.M. Martin, M.L. Sleeth, M.A. Ghatei, G.S. Frost, S.R. Bloom, Effects of oral ingestion of sucralose on gut hormone response and appetite in healthy normal-weight subjects, Eur. J. Clin. Nutr. 65 (2011) 508–513. [16] S.P. Fowler, K. Williams, R.G. Resendez, K.J. Hunt, H.P. Hazuda, M.P. Stern, Fueling the obesity epidemic? Artificially sweetened beverage use and long-term weight gain, Obesity 16 (2008) 1894–1900. [17] Y. Fujita, R.D. Wideman, M. Speck, A. Asadi, D.S. King, T.D. Webber, M. Haneda, T.J. Kieffer, Incretin release from gut is acutely enhanced by sugar but not by sweeteners in vivo, Am. J. Physiol. 296 (2009) E473–E479. [18] V.L. Grotz, I.C. Munro, An overview of the safety of sucralose, Regul. Toxicol. Pharmacol. 55 (2009) 1–5. [19] B. Härtel, H.-J. Graubaum, B. Schneider, The influence of sweetener solutions on the secretion of insulin and the blood glucose level, Ernährungs-Umschau. 40 (1993) 152–155. [20] T. Just, H.W. Pau, U. Engel, T. Hummel, Cephalic phase insulin release in healthy humans after taste stimulation? Appetite 51 (2008) 622–627. [21] J.S. Lichtman, E. Alsentzer, M. Jaffe, D. Sprockett, E. Masutani, E. Ikwa, G.K. Fragiadakis, D. Clifford, B.E. Huang, J.L. Sonnenburg, K.C. Huang, J.E. Elias, The effect of microbial colonization on the host proteome varies by gastrointestinal location, ISME J. (2015). http://dx.doi.org/10.1038/ismej.2015.187. [22] D.S. Ludwig, Artificially sweetened beverages: cause for concern, JAMA 301 (2009) 2477–2478. [23] B.A. Magnuson, G.A. Burdock, J. Doull, R.M. Kroes, G.M. Marsh, M.W. Pariza, P.S. Spencer, W.J. Waddell, R. Walker, G.M. Williams, Aspartame: a safety evaluation based on current use levels, regulations, and toxicological and epidemiological studies, Crit. Rev. Toxicol. 37 (2007) 629–727. [24] R.D. Mattes, Physiological responses to sensory stimulation by food: nutritional implications, J. Am. Diet. Assoc. 97 (1997) 406–413. [25] P.E. Miller, V. Perez, Low-calorie sweeteners and body weight and composition: a meta-analysis of randomized controlled trials and prospective cohort studies, Am. J. Clin. Nutr. 100 (2014) 765–777. [26] F. Pace, M. Pace, G. Quartarone, Probiotics in digestive diseases: focus on Lactobacillus GG, Minerva Gastroenterol. Dietol. 61 (2015) 273–292. [27] Palau-Rodriguez, M., Tulipani, S., Queipo-Ortuño, M. I., Mireia Urpi-Sarda, Tinahones, F. J., Andres-Lacueva, C. Metabolomic insights into the intricate gut microbial–host interaction in the development of obesity and type-2 diabetes. Front. Microbiol. 2015, 6: article 1151. [28] M.S.A. Palmnäs, T.E. Cowan, M.R. Bomhof, J. Su, R.A. Reimer, H.J. Vogel, D.S. Hittel, J. Shearer, Low-dose aspartame consumption differentially affects gut microbiotahost metabolic interactions in the diet-induced obese rat, PLoS One 9 (2014), e109841. [29] T.L. Powley, H.-R. Berthoud, Diet and cephalic phase insulin responses, Am. J. Clin. Nutr. 42 (1985) 991–1002. [30] A.G. Renwick, The metabolism of intense sweeteners, Xenobiot. 16 (1986) 1057–1071. [31] P.J. Rogers, P.S. Hogenkamp, K. de Graaf, S. Higgs, A. Lluch, A.R. Ness, C. Penfold, R. Perry, P. Putz, M.R. Yeomans, D.J. Mela, Does low-energy sweetener consumption affect energy intake and body weight? A systematic review, including meta-analyses,
[32] [33] [34]
[35] [36]
[37]
[38] [39]
[40] [41] [42]
[43]
[44]
[45]
[46] [47] [48] [49] [50]
5
of the evidence from human and animal studies, Int. J. Obes. (Lond) (2015). http:// dx.doi.org/10.1038/ijo.2015.177. S.S. Schiffman, Rationale for further medical and health research on high-potency sweeteners, Chem. Senses 37 (2012) 671–679. A. Sclafani, M. Abrams, Rats show only a weak preference for the artificial sweetener aspartame, Physiol. Behav. 37 (1986) 253–256. K. Shinozaki, Y. Shimizu, T. Shiina, H. Morita, T. Takewaki, Relationship between taste-induced physiological reflexes and temperature of sweet taste, Physiol. Behav. 93 (2008) 1000–1004. J.C. Smith, A. Sclafani, Saccharin as a sugar surrogate revisted, Appetite 38 (2002) 155–160. J. Suez, T. Korem, D. Zeevi, G. Zilberman-Schapira, C.A. Thaiss, O. Maza, D. Israeli, N. Zmora, S. Gilad, A. Weinberger, Y. Kuperman, A. Harmelin, I. Kolodkin-Gal, H. Shapiro, Z. Halpern, E. Segal, E. Elinav, Artificial sweeteners induce glucose intolerance by altering the gutmicrobiota, Nature 514 (2014) 181–186. T.D. Swartz, F.A. Duca, T. de Wouters, Y. Sakar, M. Covasa, Up-regulation of intestinal type 1 taste receptor 3 and sodium glucose luminal transporter-1 expression and increased sucrose intake in mice lacking gut microbiota, British J. Nutr. 107 (2012) 621–630. S.E. Swithers, Artificial sweeteners produce the counterintuitive effect of inducing metabolic derangements, Trends Endocrinol. Metab. 24 (2013) 431–441. S.E. Swithers, C.R. Baker, T.L. Davidson, General and persistent effects of highintensity sweeteners on body weight gain and caloric compensation in rats, Behav. Neurosci. 123 (2009) 772–780. S.E. Swithers, T.L. Davidson, Obesity: outwitting the wisdom of the body? Curr. Neurol. Neurosci. Rep. 5 (2005) 159–162. S.E. Swithers, T.L. Davidson, A role for sweet taste: calorie predictive relations in energy regulation by rats, Behav. Neurosci. 122 (2008) 161–173. S.E. Swithers, A.F. Laboy, K. Clark, S. Cooper, T.L. Davidson, Experience with the highintensity sweetener saccharin impairs glucose homeostasis and GLP-1 release in rats, Behav. Brain Res. 233 (2012) 1–14. S.E. Swithers, A.A. Martin, K.M. Clark, A.F. Laboy, T.L. Davidson, Body weight gain in rats consuming sweetened liquids. Effects of caffeine and diet composition, Appetite 55 (2010) 528–533. S.E. Swithers, C.H. Sample, T.L. Davidson, Adverse effects of high-intensity sweeteners on energy intake and weight control in male and obesity-prone female rats, Behav. Neurosci. 127 (2013) 262–274. S.E. Swithers, C.H. Sample, D.P. Katz, Influence of ovarian and non-ovarian estrogens on weight gain in response to disruption of sweet taste-calorie relations in female rats, Horm. Behav. 63 (2013) 40–48. K.R. Tandel, Sugar substitutes: health controversy over perceived benefits, J. Pharmacol. Pharmacother. 2 (2011) 236–243. K.L. Teff, J. Devine, K. Engelman, Sweet taste: effect on cephalic phase insulin release in men, Physiol. Behav. 57 (1995) 1089–1095. K. Tonosaki, Y. Hori, Y. Shimizu, K. Tonosaki, Relationship between insulin release and taste, Biomed. Res. 28 (2007) 79–83. M.G. Tordoff, M.I. Friedman, Drinking saccharin increases food intake and preference-IV. Cephalic phase and metabolic factors, Appetite 12 (1989) 37–56. S.C. Woods, The eating paradox: how we tolerate food, Psychol. Rev. 98 (1991) 488–505.
Please cite this article as: J.I. Glendinning, Do low-calorie sweeteners promote weight gain in rodents?, Physiol Behav (2016), http://dx.doi.org/ 10.1016/j.physbeh.2016.01.043