Leptin, adiponectin, and ghrelin responses to endurance exercise in different ambient conditions

ABSTRACT

Excessive positive energy balance is a major factor leading to obesity. The ability to alter the appetite-regulating hormones leptin, adiponectin, and ghrelin may help decrease excessive energy intake. Exercise and exposure to extreme temperatures can independently affect these appetite-regulating hormones. PURPOSE: To determine the effect of exercising in different environmental conditions on the circulating concentrations of leptin, adiponectin, and ghrelin. METHODS: Eleven recreationally-trained male participants completed 3 separate 1 h cycling bouts at 60% W_max in hot, cold, and room temperature conditions (33°C, 7°C, 20°C), followed by a 3 h recovery at room temperature. Blood was drawn pre-exercise, post-exercise, and 3 h post-exercise. Hematocrit and hemoglobin were measured to account for change in plasma volume. RESULTS: Leptin concentrations were lower at post and 3 h post-exercise compared with pre-exercise, with and without correction for plasma volume shifts, regardless of temperature (p < 0.05). Adiponectin was higher post-exercise compared with pre-exercise (p = 0.021) but not 3 h post-exercise (p = 0.084) without correction for plasma volume shifts. However, adiponectin concentrations were not different at any time point when plasma volume shifts were accounted for (p > 0.05). Total ghrelin and acylated ghrelin concentrations were not affected at post and 3 h post-exercise compared with pre-exercise, with and without correcting for plasma volume shifts, regardless of ambient temperature (p > 0.05). No differences in leptin, adiponectin, or ghrelin were found between trials (p > 0.05). CONCLUSION: Temperature does not affect the circulating concentrations of appetite-regulating hormones during an acute bout of endurance exercise.

Introduction

Environmental temperature has a large impact on fat mass and metabolic homeostasis. Indigenous populations that live in polar climates have elevated basal metabolic rates and decreased fat mass,^1-3 whereas populations from tropical climates have decreased basal metabolic rates and increased fat mass.^4,5 Thus, there appears to be an effect of temperature on energy balance. However, the relationship between environmental temperature and the circulating concentration of appetite-regulating hormones is less clear. Exercise has been shown to be an effective method for altering the appetite-regulating hormones leptin, adiponectin, and ghrelin.^6-14 Exercise in a hot environment, compared with a room temperature environment appears to attenuate appetite.^15,16 However, exercise in a cold environment may stimulate appetite.¹⁷

Adipose tissue plays an important role in the effect that different environmental temperatures have on appetite.¹⁸ Adipose tissue can act as an endocrine organ, instead of solely as an energy storage depot.^19-22 Adipokines are secreted by adipose tissue and are involved in homeostatic and appetite-regulating signaling in the body. Specifically, leptin²³ and adiponectin²⁴ are adipokines that play a major role in energy homeostasis and appetite regulation.²⁵ Leptin signals the hypothalamus that energy requirements are being met and that no more food intake is required.²⁶ Adiponectin also acts at the hypothalamus but works to stimulate food intake.²⁷ Ghrelin is produced and released from the fundus of the stomach.²⁸ Unlike leptin, ghrelin is an orexigenic or an appetite-stimulating hormone.²⁹ Acylation of ghrelin is essential for appetite regulation.³⁰ The ability to alter these hormones could lead to a better control of appetite and potentially used in treating conditions commonly associated with increased levels of fat mass such as obesity, metabolic syndrome, and type 2 diabetes. Increasing leptin concentrations and decreasing adiponectin and ghrelin may lead to a decreased appetite. This change in hormone concentration could lead to a reduction in caloric intake and thus a more negative energy balance resulting in a subsequent reduction in fat mass.

Exercise is a well-known therapy to combat the conditions of obesity, metabolic syndrome, and diabetes.³¹ Circulating concentrations of leptin generally decrease,^6-8,13,14 whereas adiponectin generally increases in response to an exercise bout.^9-12 It has been demonstrated that an acute exercise bout suppresses plasma levels of acylated ghrelin,³⁰ but has no effect on total ghrelin.³² Thus, exercise induces an increased appetite in an attempt to replace energy reserves. However, there is contradicting evidence that claims exercise has no effect on these hormones.^6,7,33-35 The exact response of these hormones appears to be influenced by intensity, mode, and duration of exercise and may help explain the variability of response in these previous studies.

The ambient temperature may further alter adipokine response related to appetite. Research investigating the relationship between environmental temperature and the cytokines responsible for appetite regulation is limited to few human trials. These research studies show no change in circulating hormone levels. Current study investigates the influence of temperature in conjunction with endurance exercise on circulating hormone levels. To our knowledge, this is the first study attempting to investigate this relationship in a human model. Previously, results in human trials demonstrated conflicting results.^15,16,36 However, in a mouse model, heat exposure generally increased leptin and adiponectin concentration while cold exposure decreased leptin and adiponectin concentration.^37-40

Temperature and exercise both appear to independently affect leptin, adiponectin, and ghrelin. The potential impact of exercising in different environmental temperatures on appetite-regulating hormones is unknown. If there is an effect, however, it may lead to the development of temperature-optimized exercise protocols to control appetite. These temperature-optimized protocols could be implemented in the prevention and treatment of chronic conditions associated with excessive positive energy balance. Thus, the purpose of the current study is to determine the effects of exercise in hot, cold, and room temperature environmental conditions on circulating levels of leptin, adiponectin, and ghrelin in humans.

Materials and methods

Subjects

This study recruited 11 recreationally-trained male subjects (age 25 ± 4, height 178 ± 5 cm, weight 79.4 ± 13.5 kg, body fat 14.7 ± 3.6%, VO_2peak 54.6 ± 12.0 ml· kg⁻¹· min⁻¹, power at VO_2peak 277 ± 41 W). Recreationally-trained was defined as engaging in physical activity at least 1-2 times per week. Due to hormonal differences in men and women, exclusively male subjects were recruited for the study. Participants were between the ages of 19 and 45 and capable of cycling consistently for one hour. Participants were considered “low risk” according to ACSM risk stratification. All participants were to understand and sign an Institutional Review Board approved informed consent form before participating in this study.

Preliminary testing

The initial testing session included an exercise protocol to determine maximal aerobic capacity and collection of participant descriptive data. Specifically, height (Seca North America, Chino, CA), weight (Befour, Saukville, WI), body composition and VO_2peak were assessed. Body composition was assessed using a hydrostatic weighing technique that uses a custom load cell based system (Exertech, Dresbach, MN). Body density was converted to percent body fat using the Siri equation.⁴¹ To determine VO_2peak participants performed a graded exercise test using an electronically braked cycle ergometer (Velotron, RacerMate Inc., Seattle WA). The test began at 95 W and workload was increased by 35 W every 3 min. Participants cycled until volitional fatigue and the highest VO₂ obtained was recorded as the VO_2peak. The maximum workload was measured by adding the highest completed stage (in watts) to the proportion of the last stage multiplied by the 35 W per stage increment. During the experimental trials participants cycled at 60% of this maximum calculated workload.

Experimental protocol

Each participant completed 3 separate trials that were no less than 4 and no more than 7 d apart. Average monthly temperature for the testing period (April–May) was 56.8°F. Participants tested were natives of the area, therefore not recently exposed to prolonged periods of hot or cold climates.

For each trial, the subject cycled in one of the experimental environmental conditions chosen by a randomized, counterbalanced, repeated measures design. The 3 trials consisted of a hot (33°C, 60% humidity), cold (7°C, 60% humidity), or room temperature (20°C, 60% humidity) environment. Participants came into the laboratory following a 12 h overnight fast and having not engaged in exercise within the previous 24 h. A 24 h food consumption log was maintained by the participant and replicated as closely as possible between trials. As subjects initially entered the laboratory, urine specific gravity was tested to ensure consistent hydration between trials (Pocket Refractometer, ATAGO Tokyo, Japan). Prior to exercise, a resting blood draw (∼6 mL) was taken from the antecubital vein of the arm. During the experimental portion of the trial, participants cycled at 60% of the peak aerobic power that was previously determined. The cycling bout was one hour in a temperature and humidity controlled environmental chamber (Darwin Chamber Company, St. Louis, MO). During the exercise bout, subjects were also required to ingest 500 mL of water. Immediately post-exercise, subjects were removed from the chamber, had blood drawn, and began a 3 h recovery period lying in a supine position quietly at room temperature (∼22°C). Participants were not allowed to eat during the recovery. At the end of the 3 h recovery period another blood draw was taken (Fig. 1).

Figure 1.

Schematic representation of the protocol. RPE: rating of perceived exertion, EV: environmental conditions, CT: core temperature, ST: skin temperature.

Measurements

Core temperature, skin temperature, and heart rate

Immediately upon arrival for the experimental trials (∼45 min before exercise), subjects ingested a Jonah Core Body Temperature Capsule (JCBC, Hidalgo Limited, Cambridge, UK). After capsule ingestion, subjects drank 125 mL of water and ate a fiber bar (Fiber One bar, 140 kcal, 4 g fat, 29 g carbohydrates, 2 g protein) to facilitate movement of the capsule out of the stomach and into the small intestine. The capsule transmitted core temperature data and heart rate to an EQ02 LifeMonitor Sensor Electronics Module vest (SEM, Hidalgo Limited, Cambridge, UK) that was worn throughout the exercise and recovery.

Blood draws and plasma levels

Blood draws were taken before exercise, immediately post-exercise, and 3 h post-exercise from the antecubital vein of the arm. Approximately 6 mL of blood was drawn into an Ethylenediaminetetraacetic acid (EDTA) coated vacutainer, (Greiner Bio-One, Monroe, NC) and whole blood was immediately tested for hematocrit (ZIPOcrit, LW Scientific, Lawrenceville, GA) and hemoglobin (HemoCue Cypress, CA) for the calculation of plasma volume shifts that occurred throughout each trial.⁴² The whole blood was then centrifuged at 10,000 X g for 15 min to separate plasma for later analysis. The plasma was aliquoted and stored at −80°C.

Circulating leptin, adiponectin, and ghrelin

Circulating plasma levels of the hormones of interest were quantified by enzyme-linked immunosorbent assays (ELISA). Protocols for each specific hormone were completed according to manufacturer's instructions. Prepackaged ELISA kits from Invitrogen (Life Technologies Corporation, Frederick, MD) were used to measure leptin and adiponectin. Prepackaged ELISA kits from Sigma-Aldrich (St. Louis, MO) were used to measure total ghrelin and kits from LifeSpan BioSciences Inc. (Seattle, WA) were used for acylated ghrelin. Plasma samples were diluted 1:100 for leptin, 1:2000 for adiponectin. No dilutions were used for ghrelin.

Statistical analysis

Environmental temperature, core and skin temperature, heart rate, VO₂, and plasma content of each of the hormones of interest were analyzed with a repeated measures 2-way ANOVA (time x trial). If F-ratio values were found to be significant, a Fisher's protected LSD (least significant difference) post hoc was performed to evaluate where significance occurred. A probability of type I error of less than 5% was considered significant (p < 0.05). All statistical data were analyzed using the Statistical Package for Social Sciences software (SPSS 23.0; Chicago, IL). Data are reported as mean ± SE (standard error).

Results

Core and skin temperature

Core temperature was higher in the hot condition than the cold and room temperature conditions at 50, 55, and 60 min (p < 0.05; Fig. 2). During recovery, core temperature was higher in the hot condition (37.4 ± 0.3°C) compared with the cold (37.1 ± 0.3°C, p = 0.001), and room temperature conditions (37.1 ± 0.2°C, p = 0.038), regardless of time. Skin temperature was highest in the hot condition and lower in the cold condition, compared with the room temperature condition by minute 5 of exercise and persisted to be different throughout the exercise bout (p < 0.001; Fig. 3). During the recovery period, there were no differences in skin temperature between the 3 conditions (hot: 34.2 ± 1.0°C, cold: 33.6 ± 1.1°C, room temperature: 34.4 ± 0.8°C; p = 0.101).

Figure 2.

Core temperature during exercise. *p < 0.05 from room temperature and cold conditions. Data are mean ± SE. RT: room temperature.

Figure 3.

Skin temperature during exercise. *p <0.05 between all conditions. Data are mean ± SE. RT: room temperature.

Oxygen consumption

During exercise, both absolute and relative oxygen consumption were higher in the hot condition (3.0 ± 0.5 L· min⁻¹, 38.5 ± 6.6 mL · kg⁻¹ · min⁻¹) and lower in the cold condition (2.7 ± 0.4 L· min⁻¹, 34.5 ± 5.4 mL · kg⁻¹ · min⁻¹), compared with room temperature (2.8 ± 0.4 L· min⁻¹, 35.7 ± 5.3 mL · kg⁻¹ · min⁻¹; p < 0.05). During recovery there were no differences in absolute (hot: 0.35 ± 0.05 L· min⁻¹, cold: 0.33 ± 0.05 L· min⁻¹, room temperature: 0.34 ± 0.05 L· min⁻¹; p = 0.237), or relative oxygen consumption (hot: 4.5 ± 0.4 mL · kg⁻¹ · min⁻¹, cold: 4.3 ± 0.4 mL · kg⁻¹ · min⁻¹, room temperature: 4.3 ± 0.5 mL · kg⁻¹ · min⁻¹; p = 0.236).

Heart rate

During exercise, heart rate in the hot condition (165 ± 10 bpm) was higher than in the cold (152 ± 8 bpm) and room temperature condition (154 ± 11 bpm; p < 0.001), but no difference was observed between cold and room temperature (p = 0.318) conditions. During recovery, there were no differences in heart rate, but there was a trend toward higher heart rate in the hot condition (82 ± 11 bpm) compared with cold (75 ± 12 bpm) and room temperature (76 ± 10 bpm; p = 0.067) conditions.

Plasma leptin concentrations

Hematocrit and hemoglobin levels were significantly lower in the hot condition, compared with cold and room temperature conditions (p < 0.05).

There were no differences in leptin concentrations between any of the temperature conditions when concentrations were not corrected for plasma volume shifts (p = 0.458; Fig. 4A), or when concentrations were corrected for plasma volume shifts (p = 0.465; Fig. 4B). However, leptin concentrations were lower at post-exercise and 3 h post-exercise when compared with pre-exercise regardless of temperature and with and without correcting for plasma volume shifts (p < 0.05, Fig. 4).

Figure 4.

(A) Leptin concentration not correcting for plasma volume shifts. (B) Leptin concentration corrected for plasma volume shifts. *p < 0.05 from pre-exercise (main effect of time). Data are mean ± SE.

Plasma adiponectin concentrations

There were no differences in adiponectin concentrations between any of the temperature conditions without correcting concentration for plasma volume shifts (p = 0.752; Fig. 5A), or when concentrations were corrected for plasma shifts (p = 0.691; Fig. 5B). Adiponectin plasma concentrations were higher post- exercise when compared with pre-exercise when concentrations were not corrected for plasma volume shifts (p = 0.021; Fig. 5A). There were no differences between the pre-exercise and 3 h post-exercise adiponectin concentrations when uncorrected for plasma volume shifts (p = 0.395; Fig. 5B). When adiponectin concentration was corrected for plasma volume shifts, no differences between any of the time points were observed (p = 0.790; Fig. 5B).

Figure 5.

(A) Adiponectin concentration not corrected for plasma volume shifts. (B) Adiponectin concentration corrected for plasma volume shifts. *p < 0.05 from pre-exercise (main effect of time). Data are mean ± SE.

Plasma ghrelin concentrations

There were no differences in acylated and nonacylated ghrelin concentrations between any of the temperature conditions without correcting concentration for plasma volume shifts (p = 0.917 and p = 0.963; Figs. 6A and 7A), or when concentrations were corrected for plasma shifts (p = 0.939 and 0.792; Figs. 6B and 7B).

Figure 6.

(A) Total ghrelin concentration not corrected for plasma volume shifts. (B) Total ghrelin concentration corrected for plasma volume shifts. Data are mean ± SE.

Figure 7.

(A) Acylated ghrelin concentration not corrected for plasma volume shifts. (B) Acylated ghrelin concentration corrected for plasma volume shifts. Data are mean ± SE.

Discussion

This study proposed that exercising in different environmental temperatures would alter circulating concentrations of appetite-regulating hormones. Based on previous research, it was hypothesized that in a hot environment both leptin and adiponectin would increase compared with a room temperature condition. It was also hypothesized that in a cold environment both leptin and adiponectin would decrease compared with a room temperature environment. However, contrary to the hypothesis, the results of this study did not demonstrate temperature related effects on these hormones. Exercise independent of temperature did stimulate a decrease in leptin and a potential increase in adiponectin, but had no effect on circulating levels of total and acylated ghrelin. This hormonal profile would suggest an increase in appetite after exercise.

Research that has linked changes in leptin and adiponectin concentration to different temperature exposures has used mice and cell line models. Mice subjected to 8°C for 12 h had decreases in leptin concentrations.³⁹ Additionally, mouse preadipocytes exposed to 39°C and 41°C produced incremental increases in leptin when compared with cells exposed to 37°C.⁴³ Similar to leptin in a mouse model, mice that were exposed to chronic heat stress showed elevated adiponectin concentrations,³⁸ whereas, during a 24 h exposure to 4°C adiponectin levels were decreased.³⁷ It is difficult to interpret the effect on appetite when an appetite-reducing hormone (leptin) and an appetite-stimulating hormone (adiponectin) are increased or decreased together. However, in the case of temperature, leptin appears to play a more dominant role since exercise in the heat suppresses appetite,^15,16,36 while exercise in the cold increases appetite^15,16 relative to a room temperature exercise condition. Based on this previous data showing an effect of temperature on hormones and appetite, we were surprised not to see any effect of temperature on leptin and adiponectin using our human exercise model. Furthermore, current findings may be attributed to less influence of temperature on thermoregulatory responses in humans, compared with rodents. Specifically, high body surface area-to-mass ratio and differences in metabolism and thermal tolerance of rats may explain changes in hormone responses compared with those observed in humans.^44,45

Lack of differences in hormone concentrations between temperatures in the current study may be due to the relatively short exposure times incorporated into the current design. In non-human models exposures were 8 to 24 times longer^37,38 and more intense,^37-39,43 than in the current study. Prolonged extreme temperature exposure may cause a greater difference in core body temperature compared with the moderate, short-lived effects observed here. A greater change in core body temperature may be required to induce the changes in hormone concentration that have been observed in non-human models. Ethical limitations using a human model may preclude this effect to be observed and thus may not have practical implications.

The current study did not directly measure appetite, and thus we cannot be certain that the lack of hormonal differences between exercise in different temperatures does not affect appetite. Indeed, previous research studies have shown a change in subjective appetite without a corresponding change in appetite-regulating hormones.^15,16,36 Further mechanistic work is needed to determine the sensitivity of hormonal concentration to predict appetite, as factors such as receptor number and receptor sensitivity may also be important.

Leptin and adiponectin are secreted from adipose tissue, and therefore leptin and adiponectin concentrations are elevated in obese individuals due to the increased adipose tissue mass. Much of the previous research examining appetite-regulating hormones has focused on overweight or obese populations.^7,8,46 The present study used recreationally active, relatively fit, and healthy male participants. These individuals were not having the hormonal imbalances, and/or receptor sensitivity issues associated with excess adipose tissue.^47,48 This population may be able to better regulate homeostasis when adding the additional stress of extreme temperatures and exercise.^48,49 During the hot trial heart rate and oxygen consumption were higher than the other trials. This physiologically more challenging trial likely placed additional stress on the participants. It appears that the current participants were able to adapt to this slight elevation in stress and relative intensity, whereas a less trained population may not be able to withstand these stresses without augmenting appetite-regulating hormones. Thus, it is possible that less trained individuals may show a different hormonal response to exercise in hot or cold environmental conditions than the current participants.

Results from the current study are in agreement with previous research that demonstrates decreases in leptin concentrations,^7,8,13,14,50 and increases in adiponectin concentrations with exercise.^9-12 Previous literature has reported that exercise bouts shorter than 60 min may not alter leptin concentration.⁶ However, the current study using a 60 min cycling bout did result in decreased leptin levels. Not only are changes in leptin concentrations dependent on duration, but also depends upon intensity and exercise mode.^13,35,51-54 Thus, it appears that a 60 min bout of cycling at 60% of maximal workload associated with VO_2max is sufficient to alter leptin and adiponectin concentrations in a recreationally trained healthy population. Furthermore, it has been demonstrated that swimming stimulates appetite and acylated ghrelin concentrations and lowers energy balance when compared with no exercise control.⁵⁵

Blood plasma volume lost during exercise by way of sweating and respiration should be accounted for when analyzing concentrations of circulating hormones. The loss of plasma will increase hormone concentration without a change in absolute hormone level. The current study analyzed the concentrations of leptin and adiponectin with and without considering changes due to plasma volume shifts.⁴² Leptin was decreased post exercise when changes in plasma volume were not considered and when changes in plasma volume were considered. When adiponectin concentrations were uncorrected for plasma volume shifts, adiponectin increased post-exercise. Whereas, correcting for the plasma volume shifts resulted in no changes in adiponectin or ghrelin concentrations. It is uncertain whether an increased concentration of adiponectin due to a reduction in plasma volume is sufficient to cause a physiologic change or if an increase in total adiponectin is needed. No clear standard has been established on the need to correct for plasma volume shifts as there are many cases in which these shifts are ignored, calculated but not corrected for, or calculated and corrected for.⁵⁶ Thus, the current study reports both corrected and uncorrected concentrations so that future research may consider the impact of correcting leptin and adiponectin concentrations for shifts in plasma volume.

In conclusion, these results indicate that in healthy, recreationally-trained human participants, environmental temperature has no effect on the circulating concentrations of leptin, adiponectin or ghrelin immediately post-exercise or 3 h post-exercise. Furthermore, the current protocol seems to be of adequate duration, intensity, and mode to cause a decrease in leptin and increase of adiponectin concentrations. Further research is needed to determine the impact that temperature and exercise have on energy intake in the face of a change in leptin and adiponectin concentrations. Future studies investigating the effects of combined effects of exercise and temperature in clinical populations are warranted. Additionally, chronic exposure to temperature and exercise is needed to determine the therapeutic use of such interventions to treat and prevent conditions associated with high-energy intake.

Abbreviations

ANOVA

analysis of variance

VO_2peak

peak oxygen consumption

W_max

Wattage max

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This publication was made possible by grants from the University Committee on Research and Creative Activity (UCRCA) and the National Institute for General Medical Science (NIGMS) (5P20GM103427), a component of the National Institutes of Health (NIH). The contents of this paper are the sole responsibility of the authors and do not necessarily represent the official views of NIGMS or NIH.