nucleus of the solitary tract (NTS) receives information from the oral cavity (e.g.,
gustation) and the gastrointestinal tract and relays information to the paraventricular
nucleus of the hypothalamus and the central nucleus of the amygdala, in part
regulating short-term satiety (Guyenet and Schwartz 2012 ). PET imaging of these
structures shows lower activity under conditions of satiety than hunger (Tataranni
et al. 1999 ).
Peripheral signals from the stomach, intestine, adipose tissue, liver, and pancreas
respond to perceived food cues, ingested food, and the energy status of the body.
Episodic signals regulate short-term (meal to meal) responses while tonic signals
are responsive to long-term energy balance. The adipose tissue secretion of leptin
partially regulates long-term appetite (i.e., days and weeks) (Power and Schulkin
2009 ). The most important regulatory hormones, peptides, enzymes, and neuro-
transmitters are briefly discussed below.
There are intensive efforts tofind biomarkers that can be developed into phar-
maceuticals and other interventions for preventing and managing obesity (Lemmens
et al. 2011 ). However, the concordance of subjective evaluations, observed eating
behaviors, CNS imaging studies (e.g., fMRI, PET), and plasma molecular
biomarkers of hunger and satiety have been low. Blood biomarkers have had little
predictive value for food intake behaviors in humans and self-reports of hunger,
satiety, fullness, caloric density, or intake volume (Flood et al. 2006 ; Kral and Rolls
2006; Livingstone et al. 2000 ; Rolls and Barnett 2000 ; Wansink 2010 ).
A recent review by Lemmens et al. ( 2011 ) concluded that the plasma biomarkers—
glucagon-like peptide-1, peptide tyrosinetyrosine, ghrelin, glucose, and insulin—do
not correlate well with subjective ratings of hunger and fullness. There is, for example,
only weak evidence of an association between high blood glucose concentrations and
lower reported appetite and intake, but a somewhat stronger association of hypo-
glycemia initiating eating. Glucose concentrations influence the secretion of insulin
that has a dual role of promoting energy storage and signaling satiety. As with glucose
concentrations, there is a stronger correlation between insulin concentrations and
appetite and energy intake in lean compared to obese subjects who are already in
positive energy balance. Particularly for humans, glucose and insulin concentrations
may be poor biomarkers of appetite because they are mediated by many other
metabolic and psychophysiological processes (de Graaf et al. 2004 ).
In studies using biomarkers there are many aspects of experimental design that
may account for the lack of agreement among studies and their generally poor
ability to predict feeding behaviors and self-reports about appetite (de Graaf et al.
2004 ; Livingstone et al. 2000 ). These methodological issues have been summarized
by Livingstone et al. ( 2000 : 98): (1) the use of free-living versus laboratory envi-
ronments; (2) the mode of administration (e.g., oral, gastric, intravenous, intrac-
erebral) of exogenous hormones, peptides, enzymes, and neurotransmitters; (3) the
sensitivity and statistical power of the study; (4) subject selection; (5) the use of
subjective ratings of satiety; (6) customary antecedent diet of the subjects; (7) the
formulation of the preload foods which are given prior to the test meal; (8) the time
interval between preload and subsequent test meal(s); (9) the formulation of the test
meal(s); and (10) use of ad libitum versusfixed diet regimens. These design features
210 L.S. Lieberman