Obesity is a worldwide epidemic associated with many metabolic disorders such as insulin resistance, increasing the risk of developing metabolic syndrome (MetS), non-alcoholic fatty liver disease (NAFLD), type 2 diabetes (T2D), and cardiovascular disease (CVD) [1
]. Obesity and associated metabolic disorders have arisen in parallel with the increased adoption of the Western lifestyle which is in part characterized by a diet rich in fat, refined carbohydrates, and generally low in fibers [3
]. Among these dietary factors, added sugars—defined as sugars and syrups added to foods and beverages during preparation or processing—have been particularly pointed out. Excessive consumption of added sugars (>10% of total energy intake) has been demonstrated to play a key role in the obesity pandemic [4
]. Consequently, several public health organizations recommend limiting its intake [6
]. Nevertheless, cutting down on sugar consumption is a daunting task as it is easy to underestimate the amount of added sugars consumed daily. Indeed, two-thirds of packaged foods contain added sugars (e.g., sucrose, glucose, or fructose) to enhance flavor or to extend shelf life [7
]. Sweets and soft drinks typically contain high levels of added sugar but even “so-called” healthier food products, such as granola bars, cereals, and yoghurt, can also contain high amounts of added sugars.
High intake of fructose is currently pointed out as the principal culprit for the rapid rise of metabolic diseases associated with obesity. Indeed, excessive consumption of fructose has been reported to induce hepatic lipid accumulation [8
], dyslipidemia [9
], and insulin resistance [9
]. Fructose has also been shown to be more detrimental than glucose in promoting visceral fat accumulation [11
], liver steatosis [13
], dyslipidemia [11
], and insulin resistance [11
]. These data are concerning, knowing that fructose represents about 50% of all added sugar consumed [7
]. Furthermore, few studies suggest that free fructose may have more severe adverse metabolic effects than bounded fructose when comparing the effect of an equivalent mix of fructose and glucose to sucrose [14
]. However, it is important to point out that most studies showing detrimental effects of fructose have used relatively high doses. In studies using moderate doses more compatible with human consumption, fructose did not alter metabolic health to a greater degree than equivalent doses of glucose or sucrose as reviewed in two recent meta-analyses [16
], suggesting that fructose might be more detrimental than other sugars only when over consumed.
Previous studies have shown that consumption of natural sweeteners compared to refined sugar can improve metabolic health. It was previously reported that honey consumption limits weight gain, improves dyslipidemia, and reduces glycemia in both human [18
] and rats [20
]. Moreover, both maple syrup [22
] and agave syrup [24
] intake has been shown to improve glucose homeostasis compared to sucrose. Yet, honey is mostly composed of an equal mix of fructose and glucose, maple syrup is 97% sucrose and agave 90% fructose. However, in addition to their carbohydrate content, natural sweeteners are also enriched with many other molecules that may confer potential health benefits, such as phytohormones, minerals, and polyphenols [25
], that may counteract the detrimental metabolic effects of the high sugars they contain. Indeed, honey contains about 181 substances (such as polyphenols) that have been reported for their beneficial impacts on hyperglycemia and hypertension in addition to antimicrobial properties [25
]. As for maple syrup, it contains many interesting polyphenols, such as lignans and a unique phenolic molecule named quebecol [26
], as well as phytonutrients and minerals [27
]. Agave syrup also contains polyphenols and phytonutrients, but to a lesser extent [28
]. Molasses has been shown to contain even higher concentrations of polyphenols than the aforementioned natural sweeteners [28
] and has already been considered as a potential supplement for iron deficiency [29
In a previous study, we compared several metabolic parameters following a single ingestion of an equivalent dose of carbohydrate from either glucose or specific natural sweeteners [28
]. We showed the potential benefits of maple syrup, molasses, and agave syrup on glycemic and insulinemic responses, as well as on the secretion of hormones associated with food ingestion such as amylin and gastric inhibitory polypeptide. However, the potential metabolic benefits of chronic intake of low concentrations of natural sweeteners in the context of diet-induced obesity have not yet been investigated. Therefore, the objective of the current study was to compare in a rat model of diet-induced obesity the metabolic impact of low intake (4% of total energy) of chronic isocaloric carbohydrate consumption from sucrose to those from natural sweeteners as well as from fructose, as the data regarding its long-term effect on metabolic disorders remains conflicting considering the high doses that have been used in previous studies.
2. Materials and Methods
The sweeteners used in the current study were previously characterized for their carbohydrate and phytochemical contents [28
]. Maple syrup used in this study was a gift from the Québec Maple Producers. All other sweeteners, molasses (Grandma; Saint-John, NB, Canada), organic brown rice syrup (Sweet Dreams; Richvale, CA, USA), organic blue agave syrup (Wholesome Sweeteners; Sugar Land, TX, USA), golden corn syrup (Crown; Memphis, TN, USA), and pure natural honey (McCormick; London, ON, Canada) were purchased at a local grocery. Fructose was purchased from Sigma-Aldrich (Saint-Louis, MO, USA).
2.2. Experimental Design
Wistar male rats (n = 108, Charles River Inc., Wilmington, MA, USA) weighing 250–300 g were housed two animals per cage in a temperature- and humidity-controlled room (21 ± 2 °C, 35–40%), with a daily 12:12 h light–dark cycle (lights on at 07:00 am). After 1 week of acclimation on a standard chow diet (Rodent Chow #2018 Harlan Teklad, Harlan Laboratories, Indianapolis, IN, USA), animals were randomly assigned to dietary groups, with ad libitum food and deionized water. Rats were fed a high-fat high-sucrose diet (HFHS; Teklad TD.130964 adj. kcal diet (66% kcal from fat; 19.8% lard, 19.8% corn oil; and 19% kcal from carbohydrates with sucrose as the only source of carbohydrate calories)) for 8 weeks. Animals received a daily gavage with an equivalent quantity of 1 g of total sugar (4% of total energy) from either maple syrup, molasses, brown rice syrup, agave syrup, corn syrup, honey, fructose, or a control solution of sucrose. The corresponding intake of 1 g of added sugar in total energy was calculated using the following equation: % of total energy as added sugar = (calories from 1g of carbohydrate (kcal) × 100)/daily total energy intake (kcal). One group of rats received a standard chow diet and daily gavage with sucrose, only as a reference group to determine the impact of the HFHS diet on obesity and metabolic parameters. Body weight gain and energy intake were assessed weekly. Food efficiency was calculated using the following equation: Food efficiency = (Body weight (g)/Food intake (g)) × 100. After 8 weeks of HFHS feeding, animals were fasted for 6 h and anaesthetized in chambers saturated with isoflurane and then sacrificed by cardiac puncture. Blood was drawn in ethylenediamine tetraacetic acid (EDTA)-treated tubes and immediately centrifuged to collect plasma. Liver, pancreas, inguinal, epididymal, and brown fat tissues were carefully collected, weighed, and snap-frozen in liquid nitrogen. The animal protocol met the guidelines of the Canadian Council on Animal Care and was approved by the Animal Care Committee of Laval University (CPAUL-111).
2.3. Indirect Calorimetry
After 6 weeks of diet treatments, 8–10 rats from each group were randomly placed in individual metabolic cages (Columbus Instruments, Columbus, OH, USA) to evaluate their energy expenditure and respiratory exchange ratio by measuring oxygen consumption and carbon dioxide production. Rats had free access to water and their respective diet during 24 h of acclimation prior to 48 h measurements. The temperature of the calorimetric room was set to 28 °C to keep the rats at thermoneutrality. Values were expressed per metabolic body weight, according to Kleiber’s law (body weight0.75
2.4. Oral Glucose Tolerance Test
Three days prior to sacrifice, after a 12 h fast, rats received an oral gavage of 2 g/kg of body weight solution of 50% dextrose. Blood samples were taken from the saphenous vein using EDTA-coated capillary tubes before (0 min), as well as 15, 30, 60, 90, and 120 min after the glucose administration. Glycemia was determined at each time point using a calibrated hand-held glucometer. The plasma was separated by centrifugation and stored at −80 °C until further analysis.
Insulin levels were measured using ultrasensitive rat insulin (80-INSRTU-E 10) enzyme-linked immunosorbent assay (ELISA) kits (Alpco, Salem, NH, USA). The areas under the curve (AUC) for glucose and insulin were calculated using GraphPad Prism 5 for Mac OS X (GraphPad Software, San Diego, CA, USA), using the value at t
= 0 min of each animal as baseline. The homeostasis model assessment of insulin resistance (HOMA-IR) index was calculated based on the following formula: fasting insulinemia (μUI/mL) × fasting glycemia (mM)/22.5. Liver triglyceride (TG) and total cholesterol contents were measured after chloroform–methanol extraction [31
] with colorimetric kits (Randox Laboratories, Crumlin, UK). Cytokines interleukin-1 beta (IL-1β), interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), and the chemiokine regulated on activation normal T-cell expressed (RANTES) were assessed in liver lysates (900 μg/mL protein in phosphate-buffered saline (PBS) containing 1% Nonidet P-40 (NP-40) and 0.01% protease inhibitor cocktail) by using a Bio-plex multiplex and Bio-plex 200 system (Biorad Laboratories, Hercules, CA, USA).
2.6. Liver Histology
Livers sections were fixed in 4% paraformaldehyde and then embedded in paraffin. Paraffin sections of 4 μm were stained with Hematoxylin and Eosin (H&E). Images were obtained by using the AxioObserverZ1 microscope (Zeiss, Jena, Germany) equipped with a 20× objective (numerical aperture 0.8) for qualitative assessment of lipid accumulation.
2.7. Statistical Analyses
The biological effects of the HFHS-sucrose versus chow-sucrose (reference for HFHS-induced obesity) were assessed using Student’s t-tests. The effects of the natural sweeteners and the fructose compared to the sucrose control were assessed using one-way ANOVA with Dunnett post-hoc test. When equivalent variance test failed, non-parametric tests were used (Mann–Whitney test for t-test and Kruskal–Wallis test for ANOVA followed by Dunn’s post-hoc test). Comparisons over the course of the experiment were made using two-way repeated-measures ANOVA with Bonferoni post-hoc test. Significant differences (p < 0.05) between groups were reported. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). Results are presented as mean ± SEM.
While the vast majority of added sugars are consumed as sucrose, composed of equal parts of glucose and fructose, it is mainly the latter that is believed to represent the main culprit for the obesity and T2D epidemic. Furthermore, less is known about the metabolic impact of chronic consumption of natural sweeteners, which are usually considered only for their sugar content and not for the other nutrients and phytochemicals they also contain. In the present study, we evaluated the metabolic impact of the addition of only 4% of the total daily calories consumed as added sugar coming from natural sweeteners (e.g., from maple, brown rice, agave, and corn syrups, molasses, or honey) as compared to an equivalent amount of sucrose. We found that treatment of HFHS-fed obese rats with most of the types of natural sweeteners reduced insulin resistance compared to sucrose-treated HFHS-fed animals. This extends our previous work, where we showed that a single gavage with maple and agave syrups as well as molasses resulted in a lower blood glucose and insulin responses [28
]. We thus propose that daily substitution of refined sucrose with natural sweeteners would generate lower glycemic and insulinemic responses, translating into better metabolic health in the long-term.
Despite the fact that natural sweeteners are often described as just sugar, they actually contain many other nutrients (e.g., vitamins and minerals) as well as many phytochemicals including different classes of polyphenols. The latter molecules have been shown to influence glucose metabolism by several mechanisms. They can inhibit carbohydrate digestion and absorption through inhibiting α-glucosidase [35
], sodium-dependent glucose transporter 1 [37
], and sodium-independent glucose transporter [38
] in the intestine, therefore reducing glucose absorption. Among the natural sweeteners we tested, maple syrup has already been shown to modulate glucose absorption in part due to its polyphenols content. An extract of maple syrup rich in polyphenols (MSX) has been reported to inhibit the activity of α-glucosidase in vitro and to decrease the uptake of glucose in HepG2 cells [40
]. Moreover, another team found that administration of maple syrup compared to sucrose in rats reduced plasma glucose levels [22
] and intestinal glucose absorption [23
]. The reduction of glucose absorption they observed was increased when maple syrup was darker, which was associated with an increased antioxidant activity [23
]. Thus, we can hypothesize that chronic inhibition of glucose absorption might have provided an advantage for these animals treated with natural sweeteners compared to those treated with sucrose. It has also been reported that the metabolic benefits of honey cannot be reproduced by a solution that only contains its equivalent sugar content [41
], which supports the concept that bioactive molecules found in honey and probably other natural sweeteners can counteract the detrimental metabolic effects of added refined sugars. Corn syrup was the only natural sweetener that was not able to significantly reduce insulin resistance compared to sucrose-treated HFHS-fed rats. This lack of effect might be explained by the lower concentration of total polyphenols in corn syrup compared to all the other natural sweeteners tested, as previously reported [28
Polyphenols have also been proposed to protect from NAFLD (reviewed in [43
]). We have previously shown that a polyphenol-rich cranberry extract prevented obesity and hepatic steatosis [45
] and could even reverse liver steatosis independently from its anti-obesity effect [46
]. In the current study, we showed that chronic intake of maple syrup, molasses, agave, and corn syrups was associated with a blunted hepatic inflammation compared to sucrose intake as revealed by a significant and specific reduction of IL-1βprotein level, an important mediator of NAFLD in animal models [47
] Indeed, mice deficient in IL-1β signaling, or in inflammasome components converting the pro-form of IL-1β to its active form, are resistant to the development of high fat diet-induced hepatic steatosis. Importantly, a recent study comparing the metabolic impact of fructose to sucrose also reported that the disaccharide was inducing higher level of IL-1β in kidney, and was more detrimental on glucose metabolism [49
]. It is interesting to put these results using pure maple syrup in line with a previous study evaluating the hepatic effect of 8 weeks supplementation of HF diet with the maple syrup extract, known as MSX, on rats. MSX did not reduce TG or cholesterol accumulation in the liver but mitigated hepatic inflammation caused by the HF diet as shown here suggesting that the anti-inflammatory molecule(s) in maple syrup may be enriched in this extract [50
]. It was showed that 17 genes involved in immune response were up-regulated in the liver, six being related to protection against infection. In line with these findings, MSX has also been reported to help antibiotics to fight against resistant bacteria in isolated cells [51
The present study was also designed to test whether the daily and chronic addition of fructose could indeed exert more deleterious metabolic effects compared to isocaloric amounts of sucrose in a diet induced obesity model. Indeed, in contrast to glucose, it is known that dietary fructose is largely metabolized in the liver where the major part is converted into glucose and then either stored as glycogen or released into the circulation. Consequently, fructose consumption does not immediately stimulate the secretion of insulin [52
]. Moreover, part of the fructose pool in the liver will be converted into lactate and a small remaining proportion will be converted into fatty acids. Although minor, this last pathway could nevertheless be implicated in the development of liver steatosis, particularly in a fructose overconsumption context [53
]. Indeed, studies have shown that high consumption of fructose, representing 25% of the total energy intake, is associated with NAFLD and dyslipidemia [54
]. However, when fructose is consumed at lower doses of 18% or 9% of total calories, it was not associated with a greater risk of MetS and CVD [56
]. Consistent with this, it has recently been demonstrated that fructose at non-excessive doses is mainly converted to glucose in the enterocyte before its absorption [57
]. Fructose is absorbed intact only when it is ingested in an amount that exceeds the capacity of its conversion to glucose in the enterocyte. In the present study, we found that daily chronic fructose administration did not result in greater hepatic steatosis than an equivalent amount of sucrose, as revealed by similar accumulation of lipids in the liver of HFHS-fed obese rats. In contrast to previous studies in rodent models, the added sugars provided by gavage in our study only represented 4% of the total daily energy intake of the rats, and the total fructose consumption, including from the diet, represented ~13% of the total energy intake, which is more relevant to typical human consumption of these added sugars in humans. We found that in these more physiologically and clinically relevant conditions, that fructose is in fact less harmful than an equivalent dose of sucrose. Our findings thus demonstrate that fructose is not metabolized preferentially into fatty acid in the liver compared to glucose when both sugars are given to rats at levels that are more nutritionally relevant to typical human consumption. Nevertheless, it cannot be excluded that the effects we observed are specific to murine metabolism which might differ from the human metabolism. Other limitations of the study should be noted. The duration of the study was 8 weeks and a longer period of treatment with the HFHS diets and the various sweeteners might have resulted in more significant differences. Moreover, the data could result from differences in absorption, as it is known that the presence of both glucose and fructose leads to greater fructose absorption [58
]. Finally, further investigations are needed to identify the mechanism of actions underlying the healthy benefits of the natural sugars. Those benefits were furthermore found compared to an equivalent amount of sucrose and future studies are needed to determine if those natural sweeteners also have health benefits versus other simple sugars or dietary conditions.
The population’s awareness of the recommendation of health agencies to limit intake of total added sugar motivates consumers to find alternative solutions. This is in part explaining the success of the low or no calorie non-nutritive sweeteners (NNS), offering a palatable alternative to caloric sugar in all kinds of products [59
]. Nevertheless, despite this substitution of NNS at the expense of regular added sugar, the prevalence of obesity and associated diseases continue to rise [60
]. Another less harmful alternative might be to improve our understanding of the metabolic impact of natural sweeteners. Indeed, despite the fact that natural sweeteners are rich in sugar and must be consumed with moderation, the present data clearly indicate that these sweeteners should be considered as more than just sugar sources. Importantly, natural sweeteners resulted in improved metabolic health when substituting sucrose even if this represented only 4% of the total daily energy consumed by the animals. Substituting the total intake of refined added sugars by natural sweeteners would not be possible in human as 90% of added sugars intake come from industrial products [7
]. However, 4% of total energy intake in humans is equivalent to only 5 teaspoons of sucrose, which substitution with natural sweeteners is clearly achievable.