Adaptive thermogenesis: Do we need new thinking?

Abstract

Weight loss is known to reduce resting energy expenditure (REE) and non-resting energy expenditure (NREE). Although seen in subjects with normal weight and excess weight, this decrease is variable and independent of the weight loss strategy. Adaptive thermogenesis (AT) refers to fat-free mass (FFM)-independent decreases in energy expenditure (1). AT spares energy, explains the nonlinear weight loss with time, and is related to weight loss maintenance. Weight loss-associated changes in composition of FFM, i.e., in skeletal muscle and organ masses, explained one third of AT (2). With a considerable interindividual variance, AT is considered as an individual trait that is biologically controlled. Low sympathetic nervous system (SNS) activity and decreases in leptin and T3 levels are its proposed determinants. This idea is supported by associations between AT and changes in its endocrine determinants after weight loss and during weight maintenance. However, preventing weight loss-associated decreases in serum T3 concentrations or in SNS activity did not affect REE (2). By contrast, substituting leptin in subjects who were weight reduced and weight stable normalized NREE but was without effect on REE (3). In this issue of Obesity, Rosenbaum and Leibel (4) proposed three models for AT. First, a “mechanical model” related to the “settling” of body weight where weight loss has no effect on slope and intercept of the REE-FFM regression line; second, a “threshold model” where AT is related to the decrease in fat mass below a minimum; and third, a so-called “spring loading model” with an effect of the dynamics of weight loss (4). No model fully explained AT. However, decreases in REE were consistent with the threshold model (with a maximal effect of fat mass on REE at 10% weight loss), whereas NREE declined similarly at each weight loss consistent with the third model. The calculations were performed using data obtained in patients with obesity during maintenance at different weights by controlled diet and physical activity. The major caveat of this protocol is that it is not about regulation of AT but more on control of long-term metabolic adaptations in response to reduced body weight. Since data obtained in a weight-reduced/weight-stable status reflect adaptations to reduced body weight, they do not explain regulation of AT. In fact, recent data showed that AT occurred early, within 3 days of semi-starvation, and is independent of changes in body weight and fat mass (2). AT fully established within 1 week of caloric restriction with no further changes during the following weeks (Figure 1). Thus, REE is regulated during early starvation. Although similar in magnitude, this may differ from long-term adaptations and weight maintenance. Surprisingly, neither leptin nor T3 nor SNS activity further explained AT (2). By contrast, decreases in insulin secretion and fluid retention correlated with AT, suggesting a very early threshold model unrelated to the adipocyte. Changes in resting energy expenditure (REE) and REE adjusted for fat-free mass during a 3-week period of controlled caloric restriction (at 50% of individual energy needs) and physical activity (at about 5,000 steps per day) in 32 young men who were healthy and with normal weight. Data were given separately for each week (i.e., they were calculated as differences between the last and the first day of the respective week). REE decreased during week 1. When compared with week 1, there were no further decreases in REE during weeks 2 and 3. The early decrease in REE corresponds to the nonlinear fall in body weight. The data were taken as evidence that regulation of adaptive thermogenesis occurs early, i.e., during week 1 of caloric restriction. With prolonged starvation, adaptation to reduced body weight rather than regulation of energy metabolism became more and more evident. *P ≤ 0.05; ***P ≤ 0.001. Data are from Müller et al., Am J Clin Nutr 2015;102:807-819 (see Ref. 2). Currently there is some confusion about defining AT in relation to either weight loss or weight loss maintenance (1). We need to differentiate between (i) mass-independent sparing of energy with negative energy balance and (ii) metabolic and endocrine factors related to maintenance of reduced weight. There is first evidence that the two issues are differently explained. Putting the adipocyte at the center of body weight control was a big step forward; this was also a good story. However, as far as the regulation of AT is concerned, this idea has some limitations, i.e., up to now, there is no evidence of an effect of leptin on REE in humans. By contrast, the issue of AT brings us back to a classical and “pre-adipocentric” view on endocrine and metabolic adaptation to starvation (5). From now on, taking both points of view into account will be wise.