The current article describes a detailed protocol for isocaloric 2:1 intermittent fasting to protect and treat against obesity and impaired glucose metabolism in wild-type and ob/ob mice.
Intermittent fasting (IF), a dietary intervention involving periodic energy restriction, has been considered to provide numerous benefits and counteract metabolic abnormalities. So far, different types of IF models with varying durations of fasting and feeding periods have been documented. However, interpreting the outcomes is challenging, as many of these models involve multifactorial contributions from both time- and calorie-restriction strategies. For example, the alternate day fasting model, often used as a rodent IF regimen, can result in underfeeding, suggesting that health benefits from this intervention are likely mediated via both caloric restriction and fasting-refeeding cycles. Recently, it has been successfully demonstrated that 2:1 IF, comprising 1 day of fasting followed by 2 days of feeding, can provide protection against diet-induced obesity and metabolic improvements without a reduction in overall caloric intake. Presented here is a protocol of this isocaloric 2:1 IF intervention in mice. Also described is a pair-feeding (PF) protocol required to examine a mouse model with altered eating behaviors, such as hyperphagia. Using the 2:1 IF regimen, it is demonstrated that isocaloric IF leads to reduced body weight gain, improved glucose homeostasis, and elevated energy expenditure. Thus, this regimen may be useful to investigate the health impacts of IF on various disease conditions.
Modern lifestyle is associated with longer daily food intake time and shorter fasting periods1. This contributes to the current global obesity epidemic, with metabolic disadvantages seen in humans. Fasting has been practiced throughout human history, and its diverse health benefits include prolonged lifespan, reduced oxidative damage, and optimized energy homeostasis2,3. Among several ways to practice fasting, periodic energy deprivation, termed intermittent fasting (IF), is a popular dietary method that is widely practiced by the general population due to its easy and simple regimen. Recent studies in preclinical and clinical models have demonstrated that IF can provide health benefits comparable to prolonged fasting and caloric restriction, suggesting that IF can be a potential therapeutic strategy for obesity and metabolic diseases2,3,4,5.
IF regimens vary in terms of fasting duration and frequency. Alternate day fasting (i.e., 1 day feeding/1 day fasting; 1:1 IF) has been the most commonly used IF regimen in rodents to study its beneficial health impacts on obesity, cardiovascular diseases, neurodegenerative diseases, etc.2,3. However, as shown in previous studies6,7, and further mechanistically confirmed in our energy intake analysis8, 1:1 IF results in underfeeding (~80%) due to the lack of sufficient feeding time to compensate for energy loss. This makes it unclear whether the health benefits conferred by 1:1 IF are mediated by calorie restriction or modification of eating patterns. Therefore, a new IF regimen has been developed and is shown here, comprising of a 2 day feeding/1 day fasting (2:1 IF) pattern, which provides mice with sufficient time to compensate for food intake (~99%) and body weight. These mice are then compared to an ad libitum (AL) group. This regimen enables examination of the effects of isocaloric IF in the absence of caloric reduction in wild-type mice.
In contrast, in a mouse model that exhibits altered feeding behavior, AL feeding may not be a proper control condition to compare and examine the effects of 2:1 IF. For example, since ob/ob mice (a commonly used genetic model for obesity) exhibit hyperphagia due to the lack of leptin regulating appetite and satiety, those with 2:1 IF exhibit ~20% reduced caloric intake compared to ob/ob mice with AL feeding. Thus, to properly examine and compare the effects of IF in ob/ob mice, a pair-feeding group as a suitable control needs to be employed.
Overall, a comprehensive protocol is provided to perform isocaloric 2:1 IF, including use of a pair-feeding control. It is further demonstrated that isocaloric 2:1 IF protects mice from high fat diet-induced obesity and/or metabolic dysfunction in both wild-type and ob/ob mice. This protocol can be used to examine the beneficial health impacts of 2:1 IF on various pathological conditions including neurological disorders, cardiovascular diseases, and cancer.
All methods and protocols here have been approved by Animal Care Committees in The Animal Care and Veterinary Service (ACVS) of the University of Ottawa and The Centre for Phenogenomics (TCP) and conform to the standards of the Canadian Council on Animal Care. It should be noted that all procedures described here should be performed under institutional and governmental approval as well as by staff who are technically proficient. All mice were housed in standard vented cages in temperature- and humidity-controlled rooms with 12 h/12 h light/dark cycles (21–22 °C, 30%–60% humidity for normal housing) and free access to water. Male C57BL/6J and ob/ob mice were obtained from the Jackson Laboratory.
1. 2:1 Isocaloric IF Regimen
2. Pair-feeding (PF) Control Group
NOTE: For an IF experiment in which altered feeding behavior is observed in a mouse model (e.g., hyperphagia in ob/ob mice), it is necessary to have a pair-feeding group as a control for proper calorie-independent comparison to IF.
3. Body Composition Analysis
NOTE: Since long-term IF affects body weight in mice, body composition can be measured at appropriate cycles (e.g., every 3 or 4 cycles) using a body composition analyzer to quantify fat and lean mass in live, non-anesthetized mice.
4. Glucose and Insulin Tolerance Tests
5. Indirect Calorimetry
NOTE: Energy metabolism of IF-treated mice can be further evaluated through indirect calorimetry over a single cycle of IF. This will measure oxygen consumption (VO2), carbon dioxide production (VCO2), respiratory exchange ratio (RER), and heat (kcal/h).
Figure 1 shows the feeding analyses after 24 h fasting and the comparison between 1:1 and 2:1 intermittent fasting. A 24 h fasting period resulted in a ~10% reduction in body weight, which was fully recovered after 2 days of refeeding (Figure 1A). A 24 h fasting period induced hyperphagia during the subsequent 2 days of refeeding (Figure 1B). Nevertheless, the comparison of energy intake between 1:1 alternative day fasting and 2:1 intermittent fasting revealed that the 1 day of the refeeding period in 1:1 IF was not sufficient (~80%) to compensate for the caloric loss by fasting, compared to the AL condition (Figure 1C). On the other hand, 99% of energy intake was fully compensated during 2 days of refeeding in 2:1 IF. This regimen enables examination of the effects of isocaloric IF that are independent of caloric intake difference.
Figure 2 illustrates a schematic timeline for the isocaloric 2:1 IF and PF regimens. To minimize the differences in caloric intake, an observation made in alternate day fasting6,7, this protocol established a new IF regimen comprising of 2 day feeding and 1 day fasting periods (2:1 IF)8, which enabled the examination of the health effects of isocaloric IF in wild-type mice. However, in ob/ob mice, which exhibited hyperphagic behavior, 2:1 IF-treated ob/ob mice showed a 21% caloric intake reduction, compared to ob/ob AL mice15. Since this prevents a proper caloric-independent comparison, a PF control group that maintained the same caloric intake as IF-treated ob/ob mice was used. Briefly, the total amount of food consumed during 2 days of feeding in 2:1 IF mice were divided equally into three daily amounts, then provided to the PF group.
For a comprehensive overview on the metabolic outcomes of 2:1 IF, we compared the effects of AL, IF, and PF in body weight, food intake and body composition in wild-type and ob/ob mice under normal diet (ND) and HFD. Compared to AL, IF treatment led to lower body weight increase in ND-fed and HFD-fed WT mice without significant differences in food intake (Figure 3A,B). Body composition analysis revealed that IF specifically reduced fat mass without changes in lean mass in wild-type mice (Figure 3C). It is possible that a slightly, albeit not significantly, lower accumulated energy intake over 16 weeks of the IF program could result in reduced body weight gain of IF animals. However, IF experiment with the pair-feeding regimen confirmed that the decreased body weight gain by IF was not due to altered energy intake (Figure 3D,E). Unlike wild-type animals, body weight of ob/ob mice subjected to IF (Ob-IF) was lower than that of Ob-AL mice (Figure 3G). This is due to hyperphagia (excessive eating) of ob/ob mice, leading to mildly higher (21%) food intake in AL mice, compared to IF-treated animals (Figure 3H). Therefore, to specifically examine the metabolic effect of IF in a caloric-independent manner, a pair-feeding control group was employed. However, unlike wild-type mice8, Ob-PF mice were indistinguishable compared to Ob-IF mice in body weights and body composition15 (Figure 3I). These results suggest that leptin is likely implicated in isocaloric IF-mediated body weight reduction in mice.
The major metabolic benefit conferred by isocaloric IF is the improved glucose homeostasis. As shown in Figure 4A,B,C,D, HFD-IF mice exhibited a significant improvement in glucose homeostasis. GTT showed that blood glucose is more rapidly cleared in IF-treated mice, while ITT revealed higher insulin sensitivity in HFD-IF mice, compared to HFD-AL or HFD-PF mice. Unexpectedly, despite the failures in IF-mediated weight reduction, Ob-IF animals exhibited significantly improved glucose handling with smaller glucose excursions in GTT, compared to Ob-PF mice (Figure 4E), whereas insulin sensitivity was indistinguishable between Ob-IF and Ob-PF mice (Figure 4F). This improved glucose homeostasis in Ob-IF mice is likely mediated by increases in plasma level of glucagon-like peptide-1 (GLP-1) and glucose-stimulated insulin secretion (data not shown)15. Overall, by using this 2:1 IF protocol and proper caloric-independent PF control, we showed the metabolic benefits of isocaloric IF in wild-type and ob/ob mice.
One of the metabolic effects of IF in wild-type mice is higher total O2 consumption, used to estimate the energy expenditure (Figure 5A,B). This elevation in O2 consumption was found only during feeding period in IF mice, but not fasting period, compared to AL mice. The increased energy expenditure was largely mediated by adipose thermogenesis, such as browning of white adipose tissues and activation of brown adipose tissue (data not shown)8,16. IF-mediated adipose thermogenesis would presumably explain how wild-type mice subjected to IF exhibited the reduced body weight gain with no difference in food intake, compared to AL mice. On the other hand, IF failed to increase O2 consumption in ob/ob mice (Figure 5C-D), and even led to a reduction in energy expenditure during fasting period. Consistently, IF-induced adipose thermogenesis was completely abolished in ob/ob mice (data not shown). These data suggest a possible limitation of IF as it may work differently for individuals with different genetic and environmental backgrounds.
Figure 1: Feeding analyses after 24 h fasting and comparison between 1:1 and 2:1 IF. (A) Daily body weight changes of mice before and after 24 h fasting (n = 10). (B) Daily energy intake before and after 24 h fasting (n = 5 cages; 2 mice per cage). (C) Comparison of energy intake between alternate day fasting (i.e., 1 day feeding/1 day fasting, 1:1 IF) and 2:1 intermittent fasting (i.e., 2 day feeding/1 day fasting). In the 1:1 IF regimen, only ~80% of food intake was compensated during the subsequent 1 day of refeeding compared to food intake over 2 days of feeding. On the other hand, 99% of energy intake was achieved when 2 days of refeeding was given, compared to that over 3 days of feeding. Data are expressed as mean ± SEM. This figure was reproduced with permission from Kim et al.8. Please click here to view a larger version of this figure.
Figure 2: Schematic illustration of the isocaloric 2:1 IF regimen. For PF control, the amount of food consumed during the 2 days of feeding by IF-treated mice is divided into three equal portions, which is then provided daily to PF mice during the next cycle. AL = ad libitum; PF = pair-feeding. Part of this figure was reproduced with permission from Kim et al.8. Please click here to view a larger version of this figure.
Figure 3: Comparison of AL, IF, and PF effects on body weight, food intake, and body composition between wild-type and ob/ob mice. (A,B,C) Body weight, food intake, and body composition in AL or IF-treated wild-type mice under normal diet (ND) or high fat diet (HFD) during 16 weeks of IF regimen. Data are expressed as mean ± SEM. (ND-AL: n = 7; ND-IF: n = 8; HFD-AL: n = 7; and HFD-IF: n = 8); one- or two-way ANOVA with Student-Newman-Keuls post-hoc analysis; **p < 0.01 vs. HFD-AL. (D,E,F) Body weight, food intake, and body composition in PF vs. IF mice fed with high fat diet (HFD) during 12 weeks of IF regimen. (PF: n = 6 and IF: n = 6); two-tailed unpaired Student's t-test; *p < 0.05 vs. HFD-PF; NS = not significant. (G,H,I) Body weight, food intake, and body composition in AL, PF, or IF-treated ob/ob mice fed with normal chow (Ob-AL: n = 4; Ob-PF: n = 7; Ob-IF: n = 6); Ob-AL vs. Ob-PF: *p < 0.05; Ob-AL vs. Ob-IF: *p < 0.05; Ob-PF vs. Ob-IF. Panels A–F were reproduced with permission from Kim et al.8. Panels G–I were reproduced with permission from Kim et al.15. Please click here to view a larger version of this figure.
Figure 4: Improved glucose homeostasis by IF in both wild-type and ob/ob mice. (A,B) Intraperitoneal GTT and ITT in HFD-AL and HFD-IF wild-type mice after 16 weeks of IF regimen. The inset shows area under curve (AUC); *p < 0.05 vs. HFD-AL. (C,D) GTT and ITT in HFD-PF compared to HFD-IF wild-type mice after 12 weeks of IF regimen. The inset shows AUC; *p < 0.05 vs. HFD-PF. (E,F) GTT and ITT in Ob-IF compared to Ob-PF mice after 16 weeks of IF regimen. The inset shows AUC (*p < 0.05 vs. Ob-PF). Panels A–D were reproduced with permission from Kim et al.8. Panels E and F were reproduced with permission from Kim et al.15. Please click here to view a larger version of this figure.
Figure 5: Energy expenditure analysis in IF-treated wild-type and ob/ob mice. (A) Traces of O2 consumption during one cycle of 2:1 IF in wild-type mice (i.e., 1 day fasting followed by 2 days of feeding). (B) Average of O2 consumption per hour during fasting, feeding, and one cycle of 2:1 IF. Data are expressed as mean ± SEM (HFD-AL: n = 6; and HFD-IF: n = 12); *p < 0.05 vs. HFD-AL. (C) O2 consumption traces of ob/ob mice during one cycle of 2:1 IF. (D) Average of O2 consumption per hour during fasting, feeding, and one cycle of 2:1 IF (Ob-PF: n = 7; Ob-IF: n = 6); *p < 0.05 vs. Ob-PF. Panel B was reproduced with permission from Kim et al.8. Panels C and D were reproduced with permission from Kim et al.15. Please click here to view a larger version of this figure.
It has been well-documented that IF provides beneficial health effects on various diseases in both humans and animals8,15,16,17,18,19. Its underlying mechanisms, such as autophagy and gut microbiome, have recently been elucidated. The presented protocol describes an isocaloric 2:1 IF regimen in mice for investigating calorie-independent metabolic benefits of IF against diet-induced obesity and associated metabolic dysfunction. Unlike the alternate day fasting (1:1 IF) protocol that results in a reduction in overall caloric intake6,7, providing 1 more day of refeeding in the 2:1 IF regimen enables maintenance of an isocaloric condition in wild-type mice.
Additionally, compared to 1:1 IF, the 2:1 IF regimen may reduce possible fasting-mediated stress or torpor in mice20 and is also comparable to a popular dietary method, the 5:2 diet2. Although its effects have not been tested, the regimen can be modified by providing additional days of refeeding (e.g., 3:1 or 4:1 IF). Moreover, this protocol presented can be easily adjusted to an hourly-scale called time-restricted feeding (TRF), in which access to food is limited to 8 h per day during the active phase21, which is known to achieve an isocaloric diet regimen and provide metabolic benefits against HFD-induced obesity and diabetes19,21,22.
As shown in the feeding analysis (Figure 1B), hyperphagic behavior immediately after 24 h of fasting decreases gradually in wild-type mice, which enables isocaloric IF. However, this isocaloric condition cannot be attained in ob/ob mice, as they lack leptin signaling-mediated satiety and energy metabolism, leading to a continuous hyperphagic phenotype23,24. Therefore, before performing an IF experiment, it is recommended to examine feeding behavior of the mouse model of interest. To examine the effects of IF using a hyperphagic mouse model (e.g., ob/ob, db/db, Sim1+/-, MC4R-/-)24,25,26, as described in this protocol, employment of a pair-feeding group as an isocaloric experimental control is important for making proper comparisons. It also requires careful planning when testing a mouse model with a hypophagic phenotype (e.g., melanin-containing hormone KO mice)27.
An important factor to consider for IF studies is housing temperature, which affects various physiological and behavioral parameters in mice. Particularly, cold exposure (4–6 °C) significantly increases energy intake to maintain core body temperature28. In contrast, in thermoneutral conditions (30 °C) under which heat gain is balanced by heat loss, reductions in food consumption is markedly reduced8. With respect to metabolic outcomes, cold exposure induces adipose thermogenesis, which is hampered by thermoneutral condition. Therefore, it is expected that housing temperature influences the metabolic phenotypes of IF and appropriate feeding:fasting ratio to achieve isocaloric IF.
Indeed, it has been previously demonstrated that isocaloric 2:1 IF can be achieved in thermoneutral conditions, leading to improved metabolic health in diet-induced obesity and metabolic dysfunction without differences in food intake between IF and AL groups8. However, isocaloric IF may not be achievable with 2:1 ratio at cold temperatures because mice under cold exposure will show a hyperphagic phenotype, which leads to underfeeding in the IF group. Since cold exposure and IF display comparable metabolic outcomes and mechanisms (i.e., adipose thermogenesis and improved glucose homeostasis) that help fight obesity, there is interest in combining these two interventions to maximize metabolic impact. Therefore, to properly test this, performing the feeding analysis before running an IF experiment and utilizing a pair-feeding control group under cold exposure are recommended.
Other factors that may potentially affect the outcomes of IF studies include housing density. Similar to the previous study, which showed reduced food consumption in more densely housed mice29, mice from a cage of five consumed significantly less food than those from a cage of two (unpublished results). In addition, it has been demonstrated that housing density significantly affects ambient temperature, as the temperature inside a cage housing five mice is 1–2 °C higher than those housing one to two mice30. Although this study concluded that housing density did not significantly affect food intake (examined for 5 weeks), in an IF study lasting 12–16 weeks, temperature inside the cage affected by housing density may still influence food intake and energy metabolism. Together, it is important to keep the same number of mice housed in a cage and minimize changing the number per cage over the course of a study.
In summary, this report shows a simple and reproducible protocol for testing isocaloric 2:1 IF in mice. Although the current study is focused on metabolic benefits of IF in diet-induced obesity and metabolic dysfunction, it can be easily adapted to investigate the protective and therapeutic effects of isocaloric IF against other conditions, such as cardiovascular and neurological diseases.
The authors have nothing to disclose.
K.-H.K was supported by the Heart and Stroke Foundation of Canada Grant-in-Aid (G-18-0022213), J. P. Bickell Foundation and the University of Ottawa Heart Institute Start-up fund; H.-K.S. was supported by grants from the Canadian Institutes of Health Research (PJT-162083), Reuben and Helene Dennis Scholar and Sun Life Financial New Investigator Award for Diabetes Research from Banting & Best Diabetes Centre (BBDC) and Natural Sciences and Engineering Research Council (NSERC) of Canada (RGPIN-2016-06610). R.Y.K. was supported by a fellowship from the University of Ottawa Cardiology Research Endowment Fund. J.H.L. was supported by the NSERC Doctoral Scholarship and Ontario Graduate Scholarship. Y.O. was supported by UOHI Endowed Graduate Award and Queen Elizabeth II Graduate Scholarship in Science and Technology.
Comprehensive Lab Animal Monitoring System (CLAMS) | Columbus Instruments | Indirect calorimeter | |
D-(+)-Glucose solution | Sigma-Aldrich | G8769 | For GTT |
EchoMRI 3-in-1 | EchoMRI | EchoMRI 3-in-1 | Body composition analysis |
Glucometer and strips | Bayer | Contour NEXT | These are for GTT and ITT experiments |
High Fat Diet (45% Kcal% fat) | Research Diets Inc. | #D12451 | 3.3 Kcal/g |
High Fat Diet (60% Kcal% fat) | Research Diets Inc. | #D12452 | 4.73 Kcal/g |
Insulin | El Lilly | Humulin R | For ITT |
Mouse Strain: B6.Cg-Lepob/J | The Jackson Laboratory | #000632 | Ob/Ob mouse |
Mouse Strain: C57BL/6J | The Jackson Laboratory | #000664 | |
Normal chow (17% Kcal% fat) | Harlan | #2918 | |
Scale | Mettler Toledo | Body weight and food intake measurement |