The present study was performed to examine the effects of exercise on the neuronal response to food-related visual cues as well as on various measures of appetite and eating-related behavior. The results of this study demonstrate that a 6-month exercise intervention is associated with attenuation in the response to visual food cues in brain regions known to be important in food intake regulation. The insula, in particular, appears to play an important role in potential exercise-induced weight loss and weight loss maintenance. Measures of eating-related behaviors and appetite, however, were not affected by the exercise intervention even despite modest weight/fat loss, suggesting that exercise may attenuate the changes in ingestive behavior expected with negative energy balance. Alternately, the observation of a neuronal change correlated with body weight and fat mass change in the absence of effects on behavioral measures may suggest that neuronal measures of food intake behaviors are more sensitive than behavioral measures.
As we have consistently shown in previous studies, the neuronal response to food-related visual cues is complex, associated with the activation of a network of brain regions, including the insula, somatosensory cortices, parietal cortices and visual cortex [29
]. The activation of a number of these regions is consistent with increased attention to food cues and enhanced motivation to eat and implicates these regions as important in the regulation of food intake.
It has been previously shown that alterations in energy balance impact the neuronal response to food cues. We, for example, have shown that overfeeding-induced positive energy balance results in attenuation of the response to food cues [29
]. Similarly, the acutely fed state has also been shown to be associated with reduced neuronal response to food cues [42
]. Conversely, prolonged fasting has been shown to result in enhanced response to food cues in brain regions important in motivation, reward and attention [23
]. We are aware of only one published study examining the effects of weight loss in a longitudinal manner on the neuronal response to food cues. In this study, Rosenbaum et al, showed that the weight-reduced state was associated with changes in the neuronal response to visual food stimuli in brain regions known to be important in homeostatic, emotional and cognitive regulation of energy intake, including increased activity in the limbic system and in brain regions important in executive and decision-making functions and decreased activity in brain regions important in homeostasis, emotional and cognitive control, and motor planning [44
]. It might, therefore, be expected that chronic exercise, resulting in weight/fat loss, would result in similar changes in the neuronal response to food cues.
While we show in the present study that relative negative energy balance associated with chronic exercise also impacts similar brain regions, the response appears to be in the opposite direction as compared to diet-related negative energy balance. Specifically, chronic exercise resulted in diminished activation in the insula, parietal cortices, and visual cortex and did not result in increased response in any brain region. The attenuation of these brain regions known to be important in attention, visual processing and motivation, suggest that the salience of the food cues is reduced with exercise. Furthermore, the attenuation of insular response with exercise was associated with weight/fat loss. Other studies have also shown that the insula plays a central role in the regulation of ingestive behavior and may not only be the primary taste cortex but may also relate to the memory of the rewarding aspects of food and eating as well as in representing information about the internal state, ie an integration area [22
]. Interestingly, though, an acute bout of exercise appears to attenuate the effects of chronic exercise on these responses. While most studies suggest that acute exercise does not significantly impact appetite and energy intake [12
], the neuronal response to food may be a more sensitive outcome, suggesting that the acute relative negative energy balance associated with an exercise bout may in fact alter appetitive behavior.
What mechanisms underlie these observed differences in responses to visual food stimuli with chronic exercise? The effects of exercise on leptin action could be important. As has been shown by others [50
], we found greater reductions in leptin concentrations than would be expected for the degree of fat mass loss with the exercise intervention. In spite of this, we did not see increased hunger or drive to eat and in fact found attenuation of brain regions important in motivation to eat, suggesting potential enhanced leptin sensitivity. Animal studies have shown that exercise improves leptin sensitivity as well as altering the anorexic/orexigenic responses to other mediators such as neuropeptide Y and melanin concentrating hormone [53
]. Although one might postulate that these central effects would primarily impact homeostatic-related brain regions, fMRI studies examining leptin deficiency and replacement have shown that it can alter higher brain responses such as in the insula to food stimuli [39
]. Certainly these exercise effects deserve further investigation.
We did not find significant differences in measures of eating-related behaviors or appetite with acute or chronic exercise. These measures are subjective and have inherent variability, so with our small sample size we may simply not have been powered to see an effect. It may be, though, that chronic exercise attenuates appetitive behavior favoring reduced food intake and resultant weight loss and/or weight maintenance. Perhaps the attenuation in the brain response to food cues seen with chronic exercise is responsible for the less than expected appetitive responses. This may be especially true in those with the greatest suppression in insula activity who also had the greatest reduction in fat mass. Also of note, while there was no exercise-based difference in disinhibition, a correlation was observed between the change in disinhibition and change in body weight in response to the exercise intervention. Does exercise “improve” disinhibited behaviors leading to less energy intake? These effects need to be studied further.
A few limitations must also be discussed. The relatively small sample size limits generalizability. Nevertheless, significant effects of exercise on the neuronal response to food cues were still observed. We were not powered to evaluate sex-based effects. As we and others have reported there are important sex-based differences in the neuronal response to food cues [56
]. The effects, though, have been demonstrated in regions not related to those seen impacted by exercise in the present study. The lack of a non-exercise control group raises the concern regarding the potential effects of consecutive testing on the results. Previous fMRI studies suggest that neuronal response to the types of stimuli used in the present study are consistent over time [58
]. Additionally, our observation of no significant differences between baseline and “acute exercise” conditions argues against a substantial effect purely from consecutive testing. Finally, we did not measure the effects of exercise on true ad libitum energy intake which may have impacted the lack of correlation seen between energy intake and other measures.
In conclusion, the results of this study demonstrate that chronic exercise training alters the neuronal response to food cues. A 6-month exercise intervention is associated with attenuation in the response to visual food cues in brain regions known to be important in food intake regulation. The insula, in particular, appears to play an important role in potential exercise-induced weight loss and weight loss maintenance. Measures of eating-related behaviors and appetite, however, were not affected by the exercise intervention even despite modest weight/fat loss, suggesting that exercise may attenuate the changes in ingestive behavior expected with negative energy balance. These findings emphasize the important role of external visual cues in the regulation of food intake and suggest that an intervention such as exercise can alter these responses.