Cholecystokinin and Treatment of Meal Size: Proof of Principle


  • Gerard P. Smith

    Corresponding author
    1. Department of Psychiatry, Weill Medical College of Cornell University and New York-Presbyterian Hospital, White Plains, New York.
    Search for more papers by this author

Box 66, New York-Presbyterian Hospital, 21 Bloomingdale Road, White Plains, NY 10605. E-mail:


SMITH, GERARD P. Cholecystokinin and treatment of meal size: proof of principle.


The purpose of this discussion was to review the therapeutic potential of cholecystokinin (CCK)1-8 for decreasing meal size. Parenteral administration and endogenous release of small intestinal CCK-8 decrease meal size reliably in lean and obese men and women with minimal side effects. Experimental analysis has demonstrated that CCK-8 released from the small intestine by ingested food exerts negative feedback control of meal size in animals and humans. The peripheral effect of CCK-8 is mediated through CCK1 receptors and is transmitted to the hindbrain by vagal afferent nerves from the abdomen. The decreased intake occurs without a detectable change in the flavor, pleasantness, or satisfaction of the ingested food. This demonstrates proof of therapeutic principle for CCK-8. Therapeutic efficacy and safety of peripheral CCK-8 for treatment of the abnormally large meals involved in the development of obesity, however, remains to be demonstrated by chronic administration in obese animals and humans.

Obesity occurs when energy intake exceeds energy expenditure for a sufficiently long time. Energy intake is the summation of the size and number of meals. The number of meals is primarily controlled by the diurnal rhythm, learning about the ease of access to foods, and the acquisition of nutrient and flavor preferences (1, 2). As a learned phenomenon, it is amenable to cognitive-behavioral components of a treatment program.

Meal size, on the other hand, is determined by the integration of two effects of food that is being eaten: orosensory positive feedback and post-ingestive negative feedback. Orosensory positive feedback stimulates and maintains eating. As food accumulates in the stomach and empties into the small intestine during a meal, the negative feedback effect of post-ingestive nutrient stimuli gradually increases until the comparator function of the brain judges that it exceeds the potency of the orosensory positive feedback (3, 4). When this occurs, eating stops, the meal ends, and postprandial satiety begins.

These feedback effects are elicited by preabsorptive stimuli in the stomach and small intestine. The adequate stimulus in the stomach is volume. Volume is also effective in the small intestine, but the digestive products of fats, proteins, and carbohydrates are more important (5). Postabsorptive metabolic effects and their neuroendocrine controls, prior experience, and social stimuli also influence meal size, but their effects are achieved by central modulation of the potencies of the positive and negative feedbacks from the gut (3, 4).

Thus, when meal size is larger than normal during the development of obesity, this is because orosensory positive feedback increased, post-ingestive negative feedback decreased, or both occurred (3, 4). The therapeutic corollary of this is that meal size can be reduced by decreasing orosensory positive feedback or increasing post-ingestive negative feedback or both. Decreasing orosensory positive feedback by decreasing the flavors of preferred foods is not an effective strategy in our culture. The optimal therapeutic goal is to reduce meal size by increasing post-ingestive negative feedback without decreasing orosensory positive feedback.

Given that endogenous physiological mechanisms often make good therapies, the mechanisms of post-ingestive negative feedback have been pursued intensively since the report in 1973 that peripheral administration of CCK-8 decreased meal size in rats (6). Although subsequent research has revealed a number of peptides released by ingested food from the small intestine and pancreas that decrease meal size in rodents and humans (7), CCK-8 remains the paradigm of the peptides that act in the periphery.

CCK Is a Molecular Mechanism of Post-ingestive Negative Feedback

The inhibitory effect of exogenous CCK-8 suggested that endogenous CCK-8 released from the upper small intestine was a physiological mechanism of post-ingestive negative feedback. This hypothesis could be tested when specific antagonists of CCK receptors became available. The rationale was that if endogenous CCK-8 was a mechanism of post-ingestive negative feedback, pretreatment with a CCK antagonist should increase meal size. Pretreatment with a CCK1 antagonist increased meal size in numerous experiments, but a CCK2 antagonist did not (8). These results with acute administration of a reversible antagonist have been confirmed by recent reports that transgenic knockout of CCK1 receptors in mice and a spontaneous null mutation of CCK1 receptors in rats result in abnormally large meals (9).

How does the release of a peripheral peptide from the small intestine affect the central controls of eating to decrease meal size? When bilateral lesion of vagal afferent nerves in the abdomen abolished the inhibitory effect of exogenous CCK-8 (10), these nerves became the focus of the search for an answer to that question. Electrophysiological recordings revealed that CCK-8 stimulated single vagal afferent nerves through CCK1 receptors in the low-affinity state (11). This was a direct effect of CCK-8 and not an indirect effect of stimulation of gastric motility by CCK-8 because atropinization blocked the motility effect of CCK-8 but did not change its stimulation of vagal afferent activity.

CCK-8 not only stimulated vagal afferent activity, it also interacted with mechanical stimulation of vagal afferent activity by the volume of a load in the stomach or duodenum in two ways. First, the effects of CCK-8 and of mechanical stimulation were additive in the same afferent fiber. Second, prior stimulation with CCK-8 enhanced the response of the fiber to subsequent mechanical stimulation (11). These synergistic interactions were seen in vagal fibers from the stomach and duodenum. Gastric vagal afferent fibers also integrate the effects of leptin and CCK-8 in neonatal and adult rats (12, 13).

Hepatic vagal afferent nerves are also involved. In contrast to the vagal afferents from the stomach and small intestine, however, CCK-8 is only a weak stimulant of the hepatic vagal afferents, whereas CCK-33, a larger molecular form of CCK, is very potent (14).

All of these results demonstrate that polymodal vagal afferent fibers from the upper gut integrate post-ingestive mechanical and peptide stimuli before processing of this information by the brain (11).

The vagal afferent nerves transmit the neural message produced by CCK stimulation to the nucleus tractus solitarius in the hindbrain where central processing and integration begin (11). Further sites of processing of the vagal input have been identified by c-Fos-like immunoreactivty in cells of the brainstem and hypothalamus (15, 16, 17). Forebrain activation is not necessary for the inhibitory effect of CCK-8, however, because the effect has been demonstrated in adult rats in which the forebrain is disconnected from the hindbrain (18) and in neonatal rats in which the connections between the hindbrain and forebrain are too immature to be stimulated by peripheral administration of CCK-8 (19, 20).

CCK and Therapeutic Reduction of Meal Size

Two modes of administration of CCK-8 are available for therapeutic use. The first is administration of exogenous CCK-8. There are numerous reports over the past 25 years that CCK-8 infused intravenously before a meal decreased meal size under laboratory conditions without significant side effects in lean and obese people (21, 22, 23, 24). The inhibition of intake was obtained without a decrease in orosensory positive feedback assessed by self reports of flavor, satisfaction, or liking. This achieves the major goal of treatment.

There is little information on whether the inhibitory effect of CCK-8 shows tachyphylaxis, an important issue for chronic use. In rodents and monkeys, tachyphylaxis has not been observed over weeks of injection before a test meal (25, 26). When CCK-8 was administered continuously intraperitoneally in rats for 2 weeks, however, tachyphylaxis developed rapidly (27).

The major current limitation to the therapeutic use of CCK-8 is that it has to be administered parenterally because it is inactivated by proteolytic enzymes in the small intestine. An orally active analog of CCK decreased meal size in rats (28), but its effect in humans has not been reported.

The second mode of administration is the release of endogenous CCK from the upper small intestine by calorically trivial stimuli. Proteolytic inhibitors, known to release endogenous CCK, decreased meal size in rats (29) and humans (30). These two results provide a proof of principle but not a robust demonstration. Subsequent to these reports, a potent luminal releasing factor for CCK from the small intestine has been identified in rats and humans (31, 32), but its effect on food intake is unknown.

In summary, exogenous and endogenous CCK-8 decreases meal size in lean and obese women and men. The decreased intake occurs without a detectable change in the flavor, pleasantness, or satisfaction of the ingested food and with minimal side effects. This represents proof of therapeutic principle for CCK-8. Therapeutic efficacy and safety of peripheral CCK-8 for treatment of abnormally large meals involved in the development of obesity, however, remain to be demonstrated by chronic administration in obese animals and humans.

In addition, what we learned from investigating the negative feedback effect of CCK-8 provides a scientific framework in which to investigate the therapeutic potential of other candidate peptides and to develop drugs to stimulate the pattern of vagal afferent stimulation that the brain reads as inhibitory information for the control of meal size.


  • 1

    Nonstandard abbreviation: CCK, cholecystokinin.