Gain in body weight over a number of years could be achieved through cumulative positive energy balances. These positive balances could come about through adjustments in the various components of energy expenditure or fuel utilization, together with shifts in food selection or eating patterns leading to adjustments in macronutrient intake. This means that many combinations of intake and expenditure could lead to a positive energy balance; these combinations can be called routes to body weight gain. However, these routes are difficult to trace by studying random samples of individuals. Previous investigations have found a clear association between high fat consumption and the occurrence of obesity, and although a high fat intake is a strong behavioural risk factor for weight gain, the relationship does not constitute a biological inevitability. Some normal-weight and lean individuals appear to eat a high-fat diet. To investigate reasons for this we have studied individuals initially defined by particular clusters of dietary characteristics related to fat and carbohydrate consumption. Habitual high-fat (HF) and low-fat (LF) consumers have been termed phenotypes. Various aspects of energy expenditure (physiological and behavioural) and energy intake were measured in these individuals with contrasting profiles. HF phenotypes had high intakes of fatty foods and an overall higher energy intake than LF. However, these groups of young adult males had similar BMIs and percentage body fat. The HF had a significantly higher resting metabolic rate (RMR) and a lower RQ, together with high plasma fasting leptin levels, and a higher sleeping heart rate. In HF individuals the physical activity level was somewhat lower and they had significantly more periods of sedentary behaviour than LF subjects. Although HF individuals appear to be more vulnerable to developing obesity, both phenotypes carry particular risk factors and protective factors for weight gain. The use of phenotypes has allowed the identification of different potential routes to weight gain. Different strategies are required to prevent age-related increase in body weight in these quite different individuals. Obesity: Phenotypes: High-fat: Low-fat Food choices, dietary intake and obesityThere is currently a debate concerning the role played by macronutrients in the development of a positive energy balance and obesity. Most of the argument concerns dietary fat. On one hand, there is considerable evidence that highfat foods either have a disproportionally weak control over appetite, or actually stimulate appetite and lead to a positive energy balance (Lawton et al. 1993). It has been noted that 'there is evidence to indicate that the consumption of a high fat diet undermines the normal mechanisms regulating energy balance in humans' (Prentice & Jebb 1995). In contrast, it has been stated that 'Diets high in fat do not appear to be the primary cause of the high prevalence of excess body fat in society, and reductions in fat will not be a solution ' (Willett, 1998).Some ...

OBJECTIVE: To investigate physiological differences between habitual high-fat (HF) and low-fat (LF) consumers, which could in¯uence the balance between energy expenditure and energy intake, and the potential for weight gain. SUBJECTS: 16 young, lean males (eight HF and eight LF consumers; % energy from fat 44.3 and 32.0, respectively). MEASUREMENTS: Habitual dietary variables (from FFQ), body mass index (BMI), body fat % (measured by impedance), resting metabolic rate (RMR) (indirect calorimetry), substrate oxidation and basal heart rate, postprandial thermogenesis and heart rate in response to a high-fat (low carbohydrate (CHO)) and high-CHO (low fat) challenge. RESULTS: HF and LF (selected for their intake of fat) did not differ signi®cantly in BMI or % body fat. HF had a signi®cantly higher RMR (1624 vs 1455 kcalad) and basal heart rate (66 vs 57 bpm) than LF. Differences in oxygen utilisation and heart rate were maintained over a 180 min period, following the high-fat and high-CHO challenge meals. HF had a signi®cantly lower resting respiratory quotient (RQ) than LF and the differences in average RQ were signi®cant over the 180 min examination period. HF had a signi®cantly lower RQ response to the high fat (low CHO) than to the high CHO (low fat) challenge; this effect was not observed in LF. HF had higher total energy intake than LF and a higher absolute (but not %) intake of protein. CONCLUSION: Signi®cant differences in basal energy expenditure and fat oxidation between habitual HF and LF consumers have been observed. The contributions of energy intake and protein intake (g not %) remain to be determined. In this particular group of subjects (young adult males) a high energy intake characterised by a large fat component is associated with metabolic adaptations which could offset the weight inducing properties of a high-fat diet. These physiological differences may be important when considering the relationship between dietary-fat and obesity.

Objective: To characterise the appetite control in habitual high fat (HF) and low fat (LF) phenotypes. Design: Four treatment conditions for each subject group in a fully repeated 2 6 2 6 2 measures design. Setting: The Human Appetite Research Unit at Leeds University, Psychology Department. Subjects: Eight lean HF (mean % fat intake746.7% daily energy) and eight lean LF (mean % fat intake729.9% daily energy) were recruited from the staff/student population of Leeds University. Interventions: All subjects were provided with either a low (2129 kJ) or high (3801 kJ) energy meal at midday and the capacity for compensation was later measured by nutrient challenge (ad libitum consumption of either high fat or high CHO foods). Satiation and satiety were assessed by changes in energy and nutrient intakes, hunger, fullness and food preferences. Results: The energy and nutrient manipulations gave rise to different levels in the rated intensity of hunger between HF and LF (P`0.01). HF rated their baseline hunger at a higher level than LF, and the nutrient induced changes in hunger had a much greater amplitude. HF consumed signi®cantly more energy from the high fat meals than from the high CHO meals (P`0.05); this effect was not observed in LF. HF ate more energy and a greater weight of the high fat foods but less energy and smaller weight of the high CHO foods than did the LF. HF rated the high fat and high CHO foods equally satisfying, tasty and ®lling, whereas LF indicated a preference for high CHO foods (P`0.05). Conclusions: The appetite control in habitual high and low fat consumers is different. HF`passively overconsume' fat whereas this effect is weak in LF. The HF ate a constant weight of food whereas LF ate a more constant level of energy. HF could not distinguish between high and low fat foods suggesting that they were intrinsically insensitive or`taste adapted' whereas LF were fat sensitive. The clear differences disclosed in response to signals generated by the characteristics of ingested food (weight, energy, nutrient composition, taste) suggest that habitual high and low fat consumers can be regarded as distinct behavioural phenotypes. The different styles of appetite control could arise from: (a) intrinsic physiological differences, or (b) a system which is adapted to deal with a particular type of diet. Sponsorship: This study was supported by the Biotechnology and Biological Sciences Research Council (BBSRC).

OBJECTIVE: To investigate physiological differences which could in¯uence the balance between energy expenditure and energy intake, between habitual high-fat (HF) and low-fat (LF) consumers and the potential for weight gain. SUBJECTS: Ten HF and nine LF consumers, all young, lean males (% energy from fat 45.4 and 31.8, respectively). MEASUREMENTS: Habitual dietary variables (from the food frequency questionnaire, FFQ), body mass index (BMI), % body fat (% BF, measured by impedance), fasting concentrations of plasma leptin, glucose and triglycerides. RESULTS: HF and LF subjects (selected for their fat intake) did not differ signi®cantly in BMI or % BF. HF subjects had signi®cantly higher concentrations of plasma leptin and lower concentrations of plasma glucose than LF subjects. In all subjects, concentrations of fasting plasma leptin correlated signi®cantly with BMI, % BF and fat mass; difference in leptin between groups remained signi®cant when BMI and % BF were used as covariants. Leptin signi®cantly correlated with dietary variables; particularly dietary fat (% energy and g) and inversely with dietary carbohydrate (% energy), but showed no correlation with dietary protein or total energy intake. CONCLUSION: Signi®cant differences in concentrations of fasting plasma leptin have been observed between lean male HF and LF consumers. These ®ndings suggest that the difference in leptin concentrations could be associated with a metabolic adaptation which could help to offset the weight inducing properties of high fat (high energy) diets.

Objective: To characterize taste preferences in habitual high fat (HF) and low fat (LF) phenotypes. Design: Eighteen test solutions to taste for each subject group in a fully repeated 2Â6Â3 measures design. Setting: The Human Appetite Research Unit at Leeds University, Psychology Department. Subjects: Eight lean HF (mean percentage fat intake 43.4% daily energy) and eight lean LF (mean percentage fat intake 32.7% daily energy) were recruited from the staff=student population of Leeds University. Interventions: All subjects were required to taste 18 solutions based on six levels of fat content and three levels of sucrose content. Subjects rated each solution for fat content, sweetness and pleasantness. Results: HF and LF showed significant effect for the fat content of the solution on the perception of creaminess (P < 0.000), and sugar content on the perception of sweetness (P < 0.000). HF and LF did not differ in their taste preferences and no preferred level of fat or sweetness was detected in either group. Conclusions: In young adult males, habitual food selection is not related to taste preference.