Our study is the first to reveal a differential intake among overweight as compared to lean individuals in response to ATD. Overweight participants increased their sweet calorie intake in response to ATD relative to placebo. Lean individuals actually tended to decrease their sweet calorie intake in response to ATD. On average, overweight participants consumed 123 more calories of sweet-tasting food than their lean counterparts. We also found that overweight participants responded to ATD by choosing to eat sweet foods first, before consuming non-sweet foods. In contrast, lean participants responded to ATD in the opposite manner, that is, they choose to eat non-sweet foods first. These effects on sweet food calorie intake and preference persisted after adjusted for gender, MDD history, and any ATD associated depressive symptoms.
Results of the present study suggest that acutely lowering 5-HT has a selective influence on sweet food consumption. The present findings are consistent with previous research that links sweet, palatable food intake to 5-HT deficiency (Asin et al., 1992
; Wogar et al., 1991
). At first glance, results appear to be at variance with prior studies showing that indirect serotonergic agonists selectively reduce elevated intake of carbohydrates, independent of sweetness (Wurtman et al., 1993
; Wurtman et al., 1981
). However, since the high-carbohydrate snack foods often overconsumed by overweight dysphoric adults also tend to be sweet (Drewnowski, Kurth, Holden-Wiltse, & Saari, 1992
), the discrepancy may be more apparent than real.
The behavioral pattern of reacting to 5-HT deficiency and emotional distress by increasing sweets intake resembles the clinical profile in atypical depression, which is often accompanied by overweight (Paykel, Parker, Rowan, Rao, & Taylor, 1983
). Binge eating is at the extreme end on a continuum of emotionally-triggered eating that can vary considerably in frequency and energy content. In some overweight and obese individuals the 5-HT neuronal system may be dysregulated, similar to what is observed in bulimics (Kaye et al., 2000
). Jimerson and colleagues propose that binge eating increases the plasma tryptophan to large neutral amino acid ratio, increasing brain tryptophan availability and enhancing 5-HT synthesis (Jimerson, Lesem, Kaye, & Brewerton, 1992
). The current findings raise the prospect that serotonin plays a role in the overeating patterns of overweight and obese individuals similar to its involvement in bulimic binge eating.
Intake of sweets has also been shown to release mesolimbic dopamine (DA) in nonhumans (Avena & Hoebel, 2003
; Salamone, Cousins, McCullough, Carriero, & Berkowitz, 1994
), not unlike drugs of abuse. Serotonin neurons, in turn, inhibit mesolimbic DA neurons (Alex & Pehek, 2007
; Rothman & Baumann, 2006
). Reduced 5-HT synthesis via ATD has been shown to diminish serotonergic inhibitory control of the dopaminergic system which stimulates DA release, craving, and compulsive drug-seeking (Cox et al., 2006
). Possibly, via an indirect effect on DA release, ATD may both heighten the incentive salience of sweet foods (Robinson & Berridge, 1993
; Roiser et al., 2006
) and undermine control of the impulse to respond to rewarding treats (Olausson, Engel, & Soderpalm, 2002
The ATD challenge we used in the current study (neurotransmitter challenge combined with psychological negative mood induction) may also have acted as a stressor by perturbing brain neuroregulatory and psychological coping systems thereby provoking emotional discomfort, particularly in populations prone to depression (Spring et al., 2007
). Data from human experiments show that undergoing a stressor increases the reward value of palatable treat foods (Spring et al., 2003
). Dallman and colleagues (M. F. Dallman et al., 2004
; Pecoraro, Reyes, Gomez, Bhargava, & Dallman, 2004
) posit that preference for high energy “comfort foods” following stress has evolved to replenish the energy stores depleted by physiological and behavioral coping responses. Nonhuman data show increased intake of sweet and fatty foods after various stressors (M. F. Dallman et al., 2004
; Dess, Choe, & Minor, 1998
), as well as an association with abdominal obesity (M.F. Dallman, Pecoraro, & la Fleur, 2005
). Our finding that ATD increased overweight participants’ consumption of calories from sweets suggests a need to elucidate common neurobiological mechanisms whereby ATD and other stressors promote increased consumption of sweets leading to overweight and obesity.
This study has several limitations. First, the range of BMI was restricted by exclusion criteria. Differences between the lean and overweight groups might have been greater if more obese individuals (BMI ≥ 35) had participated in the study. Those with class II or greater obesity were excluded because it remains unclear whether the amino acid dosages used in the present ATD protocol are effective in more obese individuals. In addition, self-reported measures of height and weight were used instead of more objective measures. Second, carbohydrate-rich foods were not purely carbohydrate in composition and protein-rich foods were not purely protein, because single nutrient foods rare in nature or in a typical diet. Carbohydrate-rich foods were chosen to be have at least a >6:1 carbohydrate to protein because such foods have insufficient protein to block the insulin-mediated rise in plasma tryptophan after carbohydrate intake (Yokogoshi & Wurtman, 1986
). Illustrativecarbohydrate-rich menu items included potato chips, spaghetti, and M&M candies. Illustrative protein-rich foods were baked chicken, cheese, and roasted nuts. A third limitation is that the first food consumed might not have necessarily been the most immediately desired food. This assumption was based on findings that palatable food cues trigger activation of brain regions that enhance preferences for immediate over delayed rewards (Hariri et al., 2006
; Kelley, Schlitz, & Landry, 2005
; McClure, Ericson, Laibson, Lowenstein, & Cohen, 2007
; McClure, Laibson, Loewenstein, & Cohen, 2004
; Tanaka et al., 2004
), however other factors could have guided the first food consumed. Finally, the sample size was estimated based on the hypotheses of the parent study. The number of participants tested may, therefore, may have been insufficient to detect differences in food consumption across conditions, given that previously observed effects of ATD on food intake have been modest (Oldman et al., 1995
). On the other hand, however, some studies using much smaller sample sizes (N=20) did detect effects of ATD on food intake in recovered bulimics (e.g., (Weltzin et al., 1995
)). The present study was an initial exploration into the effect of ATD on food intake in a non-eating disordered population, larger studies are merited to further explore this effect which will help elucidate the role of serotonin in “comfort eating.”
In conclusion, the present study provides evidence of serotonergic involvement in the food consumption of overweight individuals. Acutely lowering serotonin synthesis by tryptophan depletion heightened the intake of sweet-tasting foods by overweight individuals, regardless of gender, smoking status, history of MDD, or level of depressive symptoms. Whether tryptophan depletion enhanced the incentive salience of the rewarding food (Robinson & Berridge, 1993
) or interfered with the ability to modulate food intake warrants further investigation.