Energy balance is dependent upon a constant equilibrium between the energy intake and energy expenditure. The former is controlled by "appetite" through an intermediate phase termed "satiation" and its termination via "satiety" [1-4]. The role of satiety in controlling food intake begins once food has interacted with the receptors on the tongue and nose. The bolus of food follows the oropharyngeal duct to the stomach and duodenum, onward to the jejuno-ileum and colon to be expelled once digested. Along the transit, satiety signals arise from visceral afferent nerve fibers through the gastro-intestinal tract itself and the blood. At every step there is a cascade of neuromodulators that are postulated to participate, in part or entirely, in satiety (bombesin, CCK, PYY, ghrelin, leptin, insulin, glucocorticoids) [5-9]. Maladjustment in any component may favor increase daily intake and thus contribute to an upward drift in body weight and eventually obesity.

Some authors attribute the rising prevalence of obesity [10-13] simply to ever increasing caloric intake [14] and decreased physical activity while others have implicated over- or under-responses of satiating hormones or maladjusted satiety cues [15,16]. Regardless of the underlying etiology of obesity, this pathology is further complicated by such biobehavioral factors as dietary variety, portion size, snacking, cafeteria foods, increased palatability, and low cost of calorie dense foods [14].

In the 50's, Keys' experiment of semi starvation and refeeding in healthy normal-weight men [17] has provided data supporting the existence of autoregulatory feedback signals linking the state of depletion of fat stores to compensatory mechanisms operating via both food intake and regulatory thermogenesis [18-20]. In the following years, a number of authors have defended the hypothesis that body weight is regulated and that there exists a ponderostat [21-45]. According to that point of view, the major signal responsible for the overall stability of body weight is the set-point. That point of view entails that obesity results from a rise in body weight set-point, an analog signal that is the aim the system tends to achieve. If this were not the case, all defense mechanisms would operate to counter the rise of body weight.

Others have argued against the set-point theory and have instead substituted the term settling point [46,47]. According Davis & Wirtshafter [46] the absence of a set-point would then mean the absence of regulation per se. The same point of view against a set-point also existed in temperature regulation. For Webb [48,49] the system defends heat rather than temperature. Such a concept of regulatory biological systems is actually equal to an absence of any regulation and systems operating as simple steady states. With temperature as well as with body weight, the fact that responses (shivering, sweating, alliesthesia, thermogenesis, hoarding etc.) oppose changes to temperature and to body weight minimize the significance of the settling point hypothesis.

In previous studies, our laboratory has found that negative alliesthesia occurred when normal weight subjects were maintained on a ad libitum bland monotonous diet [50]. These subjects also experienced significant weight loss. In addition, when maintained on a bland diet normal weight subjects' lose weight without reporting a sensation of chronic hunger, a pattern understood as a lowering of the body weight set-point [50]. Inversely, when obese subjects lower their body weight while maintained on a caloric restriction diet it has been shown that satiety was decreased or suppressed [51]. The only difference between normal and obese patients is that the response takes place at a higher body weight in the latter [51].

The gustatory pleasure evoked by a sweet stimulus in subjects maintained under varying weight conditions reveals valuable information about the internal state of the biological system [32,51-53]. The decrease of initial pleasure after repeated ingestion of a sweet stimulus demonstrates an important physiological mechanism termed "negative alliesthesia".

The relationship between obese individuals' alliesthesia and their body weight has scarcely been studied in the clinical setting. The aim of the present clinical study was to compare the onset of negative alliesthesia in obese subjects before- and three months after following a weight loss diet.

Table 1 Participant's demographics and descriptive data analyses. Obesity Class I, II & III according to World Health Organization Classification (World Health Organization, Geneva, 1998). Where Class I = BMI 30.0 to 34.9, Class II = BMI 35.0 to 39.9 (more ...)

Participants were instructed to fast overnight and arrive at our laboratory early the following morning. At each visit, only one participant was taken at a time. Participants completed a general health questionnaire and body weights were recorded before initiating the experimental sessions.

Depending on their BMI, sex, age and physical activity, participants are assigned one of the four diet plans that range from 1400 to 2000 kcal/d (5852 to 8360 kJ/d) in the weight loss phase. During that phase, it is estimated that 50% of the energy is derived from carbohydrates, 25% from protein, and 25% from lipids. When participants reach the goal weight, they enter the "weight maintenance phase" and are taught to increase their caloric intake by 50 kcal /wk, in a minimum of 8 wk, to eventually reach a daily intake of 1800 to 2400 kcal/d (7524 to 10 032 kJ/day) depending on energy needs. During this second phase, diet composition changes slightly, with a decrease in protein and an increase in lipid content (carbohydrates = 50%, protein = 22%, lipids = 27%).

In groups of 50 to 100, participants are instructed on how to record everything they consume in a daily journal designed for that purpose and they are invited to share them with their group leader every week. Weigh-in sessions followed by a 30 to 45 min lecture on various topics (e.g., weight loss, nutrition, motivation) and recipe sampling; take place on a weekly basis. In addition, the commercial program provides support from a dietician through telephone line and Internet.

i. Body Mass Index (BMI). Determined by dividing participant's body weight (kg) / participant's height squared (m²).

ii. Time to reach zero rating for the hedonicity of taste sensation. This variable is defined as the time taken by participant to reach indifference after ingesting sweet stimuli every 3 min. This variable is interpreted as indicative of satiation.

iii. Time to abandon experimental session. This variable is interpreted as full satiety through negative alliesthesia since the participant was free to withdraw whenever she chose.

iv. Amplitude of hedonic rating. The highest (maximal) hedonic rating was the most pleasurable rating reported by participants after ingesting sweet stimulus. In opposition, the lowest (minimal) hedonic rating was the most displeasurable rating reported by participants after ingesting a sweet stimulus.

v. To compare rate of alliesthesia, the Kaplan Meier's method was used as for survival rate.

After following the Minçavi^® weight loss program for three months, mean group body weight was significantly lower (Initial session 94.7 +/- 3.0; 3 mo 89.9 +/- 3.0 kg; Student's paired t = 9.90; p < 0.0001; two tailed; Table 1). Maximal weight loss observed in our participants was 6.8 kg and minimal was 0.9 kg (Figure 1). Expressed as BMI, the weight loss program caused a significant decrease after three months (Initial session 36.8 +/- 1.8; 3 mo 34.9 +/- 1.8 kg/m²; Student's paired t = 10.4; p < 0.0001; two tailed). Body weight loss at 3 mo was 5.2 % ± 0.2 (Table 1)

Figure 1 Individual body weight loss in 10 obese participants at Initial session and 3 mo following a commercial regimen (Minçavi®). Mean group body weight loss showed a statistically significant decrease (Initial session 94.7 +/- 3.0; 3 mo 89.9 (more ...)

When patients repeatedly ingested sweet stimuli and reported their hedonic rating, the entire cohort of fasted participants started with reports of pleasurable sensation and gave positive hedonic rating. Over the course of the session the positive rating fell to zero or became negative, indicating negative alliesthesia. A representative alliesthesia kinetic from one participant before and while on the Minçavi^® diet is shown in Figure 2. The mean time to reach zero rating, or indifference, revealed a statistically significant earlier delay after dieting (Initial session 21.9 +/- 3.8; 3 mo 16.2 +/- 2.4 min; Student's paired t = 2.48; p = 0.0351; two tailed, Figure 3). The solid stimulus was chosen by 7/10 of participants, over the liquid stimulus. Regardless of the solid or liquid chosen, participants still yielded identical kinetic profiles for alliesthesia before and after weight loss.

Figure 2 A representative satiation kinetic from one obese participant (Class I) at Initial session and 3 mo following a commercial regimen (Minçavi®). The curve shows that time to reach zero rating significantly decreased from 21 min at Initial (more ...)

Figure 3 Mean (+/- SE) time to reach zero rating or indifference. The overall time to reach zero rating was significantly decreased after dieting (Initial session 21.9 +/- 3.8; 3 mo 16.2 +/- 2.4 min; Student's paired t = 2.48; p = 0.0351; two tailed).

In order to follow the time course of alliesthesia in our experimental groups, we created survival curve using the product limit method of Kaplan-Meier. This procedure would cancel the fact that not all participants followed the same time course and stayed for some duration. No event times thus were considered missing or censored. The median time to achieve negative alliesthesia was significantly increased after diet (Log-rank chi-square = 4.312, df = 1, p = 0.0378; Figure 4). Diet lowered the median time to achieve negative alliesthesia (Initial session 33 vs. 3 mo 24 min). There was no difference in maximal and minimal median hedonic rating (maximal Initial session +79.5 +/- 11.7 mm; 3 mo +94.5 +/- 9.9 mm; Wilcoxon = 14.5; p = 0.1934 or minimal Initial session -90.0 +/- 14.4 mm; 3 mo -106 +/- 11.1 mm Wilcoxon = 22.0; p = 0.6250). The absence of detecting a significant difference in this endpoint reveals that participant's hedonic rating for sweet stimulus was unchanged after diet.

Figure 4 Kaplan-Meier curve was created to determine the time to achieve an event: satiety. Diet showed a significant increase in the satiety rate at 3 mo vs. Initial session (Log-rank chi-square = 4.312, df = 1, p = 0.0378). The median time to achieve satiety (more ...)

Obese participants enrolled in the present study all lost weight by following the Minçavi^® weight loss program. Although the overall group weight loss was statistically significant, the clinical relevance of such a loss for any individual participant (range 0.9 – 6.8 kg) cannot be ascertained. However, for obese people even moderate weight loss is thought to be beneficial if not physiologically at least psychologically[55,56]. All but one participant met the objective of a 5% decrease in BMI purported by the Minçavi^® program (Figure 1).

Historical data from a controlled clinical trial in morbidly obese participants awaiting bariatric surgery and weight-matched un-operated controls utilizing the same experimental approach over a longer period revealed a similar alliesthesia kinetic [57]. Curve parameters from the present diet study suggest that negative alliesthesia as seen in our participants is comparable to those reported in morbidly obese patients (Historical controls initial session: 25.5 min, vs. Historical controls 3 month: 30 min).

The pleasure aroused by sweet stimuli faded earlier at the three month mark than at the initial session. The relationship of pleasure with intake is obvious as a stimulus that turns toward unpleasantness should be rejected; such negative alliesthesia is suggestive of a satiation process. Earlier onset of negative alliesthesia was observed in each participant who had lost weight. A weak negative correlation was observed when we tested the relationship between amount of weight loss from diet to time to achieve negative alliesthesia (Pearson r = -0.53). Correlation data did however suggest the use of greater patients in future studies utilizing the method. At first visit, all participants yielded a response kinetic, that could be interpreted as three distinct phases, namely; appetite, satiation, and satiety. Because participants all arrived at our laboratory after an overnight fast, their initial rating of hedonicity was positive at every session. Each participant transitioned through to what could be referred to as a "satiation phase", indicative of a zero or indifferent hedonicity rating. The final phase, suggestive of satiety, was the abandonment due to significant displeasure and disgust for the sweet stimulus. Thus, our obese participants displayed a negative alliesthesia pattern similar to that observed in normal weight subjects. It is likely that changes in alliesthesia observed here were linked to that specific diet and that upon returning to their previous dietary habits, subjects would probably return to their baseline response.

Previous experiments have focused mainly on shift in taste preferences in obese and lean individuals after dieting [53,58]. These investigations yielded equivocal results, and assessed the changes in taste correlated with lowered body weight. Our participants did not modify their initial ratings although their weights were lowered from dieting. Such a result would not support a changed taste preference related to body weight in fasted subjects. Cummings, Weigle, Frayo, Breen, Ma, Dellinger, & Purnell, [59] studied the effects of diet-induced weight loss on plasma ghrelin levels. Their subjects showed a significant 17% decrease in BMI after six months. Although the population studied was similar to ours, their experimental approach to weight loss was more drastic than the Minçavi^® diet. Subjects from the Cummings study were initially on a three-month liquid diet (1000 kcal/d) and transitioned to a solid diet for the remainder of the study.

An earlier onset of negative alliesthesia was obvious at three months when participants had lost weight and were still on the Minçavi^® diet. In the light of previous work in animals and in humans cited earlier, these findings could be interpreted has a lowering of body weight set-point. Dietary factors may have played a role in this phenomenon. Since Minçavi^® promotes a diet that is lower in fat and in added sugar, it is likely that the participants have found it to be less palatable than their usual diet. Earlier work has shown that a decrease in palatability could lower one's body weight set-point [50]. Some studies also suggest that diets that are relatively high in protein [60-62] and that have a low glycemic index [63-67] such as the Minçavi^® diet, may promote satiety and weight loss. It is possible such factors, when present in the diet, contribute to lower one's set-point.

Maintaining a lowered set-point, by consuming a sensible diet that promotes satiety and gradual weight loss, may be the key for long-term success, as the body strives to maintain a body weight close to that set-point by reducing food intake and enhancing energy expenditure. Further studies are needed to better understand alliesthesia in obese individuals and its link to body weight set-point.

The author(s) declare that they have no competing interests.

PF carried out the experimental alliesthesia sessions, and data analyses. MC & CG conceived the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.

Figure 5 A Kaplan-Meier curve was created using historical control data obtained from morbidly obese patients (BMI > 50) awaiting bariatric surgery. Subjects did not receive any diet intervention. Participants all received the same test for negative alliesthesia (more ...)

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1471-2458/5/112/prepub

This work was supported by the Conseil de la Recherche en Sciences Naturelles et en Génie (NSERC) Canada. The authors wish to thank Ms. Danielle Dubois, Dtp. and Minçavi^® group leaders for their helpful assistance.