Annu. Rev. Physiol. 1997. 59:221–42
`Copyright c(cid:2) 1997 by Annual Reviews Inc. All rights reserved
`
`CHOLECYSTOKININ CELLS
`
`Rodger A. Liddle
`Department of Medicine, Duke University Medical Center and Durham VA Medical
`Center, Durham, North Carolina 27710
`
`KEY WORDS:
`
`gastrointestinal hormone, radioimmunoassay, bioassay, cell culture
`
`ABSTRACT
`Cholecystokinin (CCK) is an important hormonal regulator of the digestive pro-
`cess. CCK cells are concentrated in the proximal small intestine, and hormone
`is secreted into the blood upon the ingestion of food. The physiological actions
`of CCK include stimulation of pancreatic secretion and gallbladder contraction,
`regulation of gastric emptying, and induction of satiety. Therefore, in a highly
`coordinated manner, CCK regulates the ingestion, digestion, and absorption of
`nutrients. CCK is produced by two separate cell types: endocrine cells of the
`small intestine and various neurons in the gastrointestinal tract and central nervous
`system. Accordingly, CCK can function as either a hormone or a neuropeptide.
`This review focuses on the physiology of the CCK cell in the intestine and, in
`particular, on how the CCK cell is regulated to secrete its hormone product. The
`effects of ingested nutrients on the CCK cell and the intracellular messenger sys-
`tems involved in controlling secretion are reviewed. A summary is provided of
`recent studies examining the electrophysiological properties of CCK cells and
`newly discovered proteins that act as releasing factors for CCK, which mediate
`feedback pathways critical for regulated secretion in the intact organism.
`
`Introduction
`Cholecystokinin (CCK) was discovered in 1928 by Ivy & Oldberg based on
`the ability of intestinal extracts to stimulate gallbladder contraction when in-
`fused into dogs (1). In 1943, Harper & Raper recognized that similar intestinal
`extracts also stimulated pancreatic enzyme secretion and proposed the name
`pancreozymin (2). It was not until the active substance was purified and the
`amino acid sequence determined that CCK and pancreozymin were proven to
`be the same hormone that now goes by the name cholecystokinin (3).
`In addition to the two biological actions described above, CCK has several
`other important activities. Among the most notable is its ability to induce
`
`0066-4278/97/0315-0221$08.00
`
`221
`
`-221-
`
`
`
`
`MAIA Exhibit 1027
`MAIA V. BRACCO
`IPR PETITION
`
`

`

`222
`
`LIDDLE
`
`satiety and reduce food intake in experimental animals and humans (4–6). It
`also inhibits gastric emptying and gastric acid secretion and stimulates intesti-
`nal peristalsis. Defining the physiological actions of CCK has been greatly
`facilitated by the development of specific and potent antagonists of the CCK-A
`(A for alimentary origin) receptor. Each of the above actions ascribed to CCK
`arises from endogenous CCK, as shown in studies by which the response to a
`normal meal can be reversed by specific CCK-A receptor antagonists.
`CCK release is stimulated by ingestion of food, with fats and protein the
`most potent secretagogues. CCK secretion is initiated when food leaves the
`stomach and enters the small intestine, and secretion continues until proteins,
`fats, and their metabolites have passed the upper small intestine.
`CCK receptors have been identified on the gallbladder, pancreas, and stom-
`ach (7). Recent evidence indicates that specific CCK-A receptors are also
`present on peripheral autonomic afferent nerves that enable CCK to initiate
`certain neural reflexes. The specific site on which CCK acts to affect organ
`responses is still somewhat unclear. The best evidence for a purely endocrine
`role for CCK is regulation of gallbladder contraction, whereas other effects
`may be either neural or a combination of endocrine and neural actions. Nev-
`ertheless, the importance of CCK in regulating a variety of digestive processes
`must not be underestimated. CCK secreted locally or into the blood binds to
`specific receptors on the gallbladder, pancreas, stomach, or various nerves to
`stimulate gallbladder contraction, pancreatic enzyme secretion, delay gastric
`emptying, and regulate satiety (7). Therefore, in a highly regulated fashion,
`CCK coordinates the ingestion, digestion, and disposal of nutrients (8, 9).
`Peptide Structure
`CCK was originally purified from porcine intestine as a 33-amino acid pep-
`tide (3). The hormone possesses an amidated carboxy-terminal pentapeptide,
`Gly-Trp-Asp-Met-Phe-NH2, that is identical to that of gastrin. The carboxyl
`terminus of CCK is the biologically active portion of the hormone, and be-
`cause of sequence similarity between CCK and gastrin, each hormone can
`interact with the receptor of the other (see receptor characterization below).
`Therefore, gastrin has slight CCK-like bioactivity and CCK possesses some,
`albeit weak, gastrin-like activity. The homology shared by the two peptides ex-
`plains why antibodies raised against CCK often cross-react with gastrin. This
`overlapping cross-reactivity has been a considerable problem in developing
`radioimmunoassays for CCK, particularly because gastrin circulates in the
`blood at concentrations 10–100 times greater than those of CCK (10).
`Since the discovery of CCK, multiple molecular forms have been identified in
`intestine, brain, and the circulation of multiple species (11–20). The octapeptide
`of CCK (CCK-8), consisting of the carboxy-terminal 8 amino acids of CCK,
`
`-222-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`223
`
`Figure 1 Structure of the CCK gene, complementary DNA and preprohormone. (Modified from
`Reference 8.)
`
`is the most biologically potent small peptide of CCK that has been isolated.
`However, amino-terminal extended forms have been extracted from brain and
`intestine of the pig, dog, rat, and human. Using techniques to minimize protein
`degradation, recent studies have demonstrated that the most abundant molec-
`ular form of CCK in these species is CCK-58. Intermediate-sized peptides of
`39, 33, 25, 22, 18, 8, 7, 5, and 4 amino acids have been isolated from several
`species.
`CCK Gene Structure and Expression
`The complementary DNA structures have been determined for rat, mouse, pig,
`and human CCK (21–24). The structure of the rat cDNA is shown schematically
`in Figure 1. The 345-nucleotide mRNA encodes a 115-amino acid precursor
`consisting of a 20-amino acid signal peptide, a 25-amino acid spacer peptide,
`CCK-58, and a 12-amino acid extension at the carboxyl terminus.
`The genes for CCK isolated for mouse, rat, and human reveal remarkable
`conservation (22–25). Each consists of ∼7 kilobases containing three exons,
`the second and third of which encode the prepropeptide. In all species, only a
`single gene encodes CCK. The human CCK gene is located on chromosome 3
`in the 3q12-3pter region (26, 27).
`The mature CCK mRNA is ∼750 bases.
`It is expressed in brain and in-
`testine in mature animals in nearly equivalent amounts. CCK mRNA is most
`abundant in the cerebral cortex and duodenum. In the mouse, at birth, CCK
`mRNA levels are relatively high in the intestine but very low in brain. Over the
`first two weeks of life, levels decrease in the gut and increase in brain. These
`
`-223-
`
`
`
`
`

`

`224
`
`LIDDLE
`
`changes correlate with levels of immunoreactive CCK protein in the respective
`tissues. Intestinal expression of CCK mRNA is modified by diet. CCK mRNA
`levels decline in either fasting rats or rats fed a diet that does not stimulate CCK
`secretion (28). CCK mRNA expression increases with feeding under condi-
`tions that also stimulate CCK secretion. However, it is possible to stimulate
`CCK secretion without affecting CCK gene expression by administration of
`exogenous bombesin (29). Somatostatin has been shown to inhibit CCK gene
`expression (30).
`Structure-Activity Relationships
`To function as a CCK molecule, a peptide must have the sequence Trp-Asp-
`Met-Phe-NH2. However, because this sequence is identical to the the carboxyl
`terminus of gastrin, this peptide does not confer CCK-like specificity. In order to
`bind specifically to CCK receptors, CCK peptides must be extended to 7 amino
`acids. Full potency is not achieved unless the tyrosine residue at position 7
`from the carboxyl terminus is sulfated (31). Although sulfation is unusual for
`hormones, it is critical for biological potency of CCKs. The unsulfated form
`of CCK is ∼1000-fold less active than its sulfated counterpart.
`In contrast,
`sulfation occurs only 50% of the time in gastrin biosynthesis and is not important
`for its biological activity.
`Molecular forms ranging in size from CCK-8 to CCK-33 appear to be of
`similar biological potency on pancreatic acinar cells (32). However, CCK-58
`is somewhat less potent on a molar basis than CCK-8 and is less immunore-
`active with CCK antibodies (33). This diminished activity is likely due to the
`structure of CCK-58, whereby the amino-terminal end shields the biologically
`active carboxy-terminal end of CCK from interacting with CCK receptors or
`C-terminal directed antibodies. Cleavage of CCK-58 with trypsin restores
`both the full biological potency and immunoreactivity to that of CCK-8 (34).
`Although CCK-58 is the major molecular form in intestine, brain, and the cir-
`culation of many species, identification of multiple smaller forms of CCK in
`circulation, even under conditions that preserve CCK-58, suggests that intra-
`cellular processing of CCK occurs to produce small forms of CCK that are
`secreted into the blood.
`CCK Receptors
`CCK exerts its biological effects by binding to specific receptors on its target tis-
`sues. Originally receptors were characterized on pancreatic acinar cells, islets,
`gallbladder, and brain by radioligand binding and autoradiography (31). In the
`pancreas and gallbladder, CCK bound with an affinity that was ∼1,000-fold
`greater than that of either unsulfated CCK or gastrin. These receptors were
`termed CCK-A. In the brain, however, CCK and gastrin bound with similar
`
`-224-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`225
`
`affinities. These receptors became known as CCK-B (B for brain origin).
`Along with these two types of CCK receptors, a third related type, “gastrin
`receptor,” which exhibited binding properties similar to those of the brain CCK
`receptor, was believed to transmit the biological actions of gastrin in the stom-
`ach. However, with recent cloning and expression of the CCK receptor cDNAs,
`it has become clear that there are only two CCK receptors, A and B, and that
`the gastrin receptor is identical to the CCK-B receptor.
`The CCK-A receptor complementary DNA was cloned following purifica-
`tion of the receptor protein from rat pancreatic acinar cells (35). The cDNA
`encodes a seven transmembrane protein typical of G protein-coupled recep-
`tors.
`It consists of 444 amino acids, which when expressed in transfected
`cells, demonstrates high affinity for sulfated CCK and much lower affinity for
`unsulfated CCK or gastrin.
`The CCK-B receptor cDNA was identified by expression cloning from canine
`gastric parietal cells (36). When expressed in transfected cells, the CCK-
`B receptor binds CCK and gastrin with similar affinities and demonstrates a
`ligand-induced increase in intracellular calcium. When the cDNAs for the
`rat and human receptors were cloned, it was demonstrated that CCK receptor
`cloned from brain was identical to the gastrin receptor in the stomach (37). The
`deduced amino acid sequences of the rat CCK-A and canine CCK-B receptors
`are ∼50% homologous.
`CCK Cells
`As is typical of many gastrointestinal hormones, CCK is produced by endocrine
`cells of the intestinal mucosa, which are concentrated in the duodenum and
`proximal jejunum. A gradient of cell density exists such that CCK cell abun-
`dance is greatest in the proximal small intestine and less in the distal intestine.
`Endocrine cells containing CCK are flask-shaped, with their apical surfaces
`oriented toward the lumen of the gut (38, 39) (Figure 2).
`In this position,
`microvillus-like processes come in contact with lumenal contents. Secretory
`granules, which are ∼250 nm in size and contain CCK, are concentrated around
`the basolateral surface of the cell. It is this orientation that allows food or other
`factors within the intestinal lumen to interact with the apical surface of the
`CCK cell and initiate a series of as yet unknown intracellular signaling events
`that ultimately result in secretion of CCK from the basal surface of the cell
`into the blood. According to the Wiesbaden classification, CCK cells, by their
`ultrastructural characteristics, have been officially named I cells (40). Where
`examined in experimental animals or humans, I cells have not been shown to
`contain other gut hormones. Outside of the intestinal tract, CCK is synthesized
`by a subpopulation of pituitary and adrenal meduallary cells (41, 42). The
`function of CCK in these locations is unknown.
`
`-225-
`
`
`
`
`

`

`226
`
`LIDDLE
`
`Figure 2 Schematic diagram of a CCK (I) cell in the upper small intestine. Note that its apical
`surface is exposed to the lumen of the gut where it may sample lumenal contents (9). Somatostatin-
`containing (D) cells may influence CCK release.
`
`CCK is expressed in neurons of the brain and gut and as such is a member of
`the growing family of “brain-gut peptides.” In the central nervous system, CCK
`is distributed in neurons throughout the brain. It is found in highest concentra-
`tions in the cerebral cortex (43, 44). CCK cell bodies are localized in layers
`II–VI and concentrated in layers II and III. In the brain, CCK is colocalized
`with dopamine in neurons that arise in the mesencephalon and project to the
`limbic forebrain. CCK neurons also project to the ventromedial hypothalamus,
`which may be important for central effects of CCK in the regulation of satiety.
`In the peripheral autonomic nervous system, CCK-containing neurons are
`found in the myenteric plexus, submucous plexus, circular muscle layers of
`the distal small intestine and colon, in the pancreas surrounding the islets of
`Langerhans, and in the celiac plexus and vagus nerve (45). In the intestine, CCK
`stimulates release of acetylcholine in postganglionic neurons of Auerbach’s
`and Meissner’s plexi and causes smooth muscle contraction. Because CCK
`stimulates glucagon and insulin release, it is likely that postganglionic cells
`terminating around islets of Langerhans play a role in modifying α and β cell
`function (46).
`
`-226-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`227
`
`CCK Release
`With the development of sensitive and specific radioimmunoassays and bioas-
`says for measuring blood levels of CCK, there is now general agreement that
`fasting levels of CCK in most mammals are ∼1 pM and increase to ∼5–10 pM
`following strong endogenous stimulation (12, 47–50).
`CCK release is stimulated by ingestion of food. Before the development of
`specific in vitro assays for measuring blood levels of CCK, estimates of CCK
`release were made by inference from examination of target tissue responses
`such as pancreatic secretion or gallbladder contraction. Although it was not
`possible to distinguish between hormonal and neural influences, substantial
`insights were gained about the food components that stimulate pancreactic
`secretion. Digestion products of fat and protein were found to be the most potent
`stimulants of secretion (51). Effective stimulation required that triglycerides
`were hydrolyzed to fatty acids, with chain lengths of 9 or more carbons (52,
`53). However, it is important to recognize that not all effects on pancreatic
`secretion are attributable to circulating CCK because other hormones or neural
`CCK also may be involved (54).
`In most species, for protein to be an effective stimulant of CCK release, its
`digestion is required to effectively stimulate pancreatic secretion or gallbladder
`contraction (55–57). Of the amino acids, tryptophan and phenylalanine are the
`most potent stimulants of canine or human pancreatic secretion. Secretion is
`most vigorous when the proximal (relative to the distal) half of the small intes-
`tine is perfused with nutrients. This finding is consistent with the abundance
`of CCK cells in the proximal intestine.
`In the rat, intact protein is an effective stimulant of CCK release. This is
`likely due to the ability of protein to preserve the activity of an endogenously
`produced CCK-releasing factor that is active in the intestinal lumen to modulate
`CCK secretion (58). Although this regulatory system has been best studied in
`the rat, it is also present in other species. Regulation of CCK secretion by
`releasing factors is discussed in more detail below.
`The effects of neural factors on CCK secretion have been a matter of con-
`troversy but clarified somewhat by the application of specific CCK radioim-
`munoassays. Although the neuropeptide bombesin is known to affect pancreatic
`secretion through bombesin receptors on pancreatic acini, only recently has it
`been shown to directly stimulate CCK release (59).
`Vagotomy alters pancreatic and gallbladder responses to dietary stimulation.
`Therefore, indirect estimates of CCK secretion in vagotomized animals have
`been limited because the responsiveness of the pancreas or gallbladder could be
`the result of either changes in CCK release or the effects of vagotomy. Studies
`using radioimmunoassay measurements have shown that circulating CCK levels
`were unaltered in dogs administered oleate, despite impaired pancreatic and
`
`-227-
`
`
`
`
`

`

`228
`
`LIDDLE
`
`gallbladder responses (60). In vagotomized subjects, plasma CCK responses
`were greater following a liquid meal, perhaps because of a more rapid delivery
`of nutrients to the small intestine following vagotomy (61).
`In sum, these
`experiments suggest that vagal innervation does not affect CCK directly, but
`may alter the target tissue responses to stimulation by CCK.
`Regulation of CCK Secretion by Releasing Factors
`The exact mechanisms by which foods cause CCK secretion have yet to be
`elucidated, but it is likely that multiple factors are involved. The location of the
`CCK cell in the intestinal mucosa, with its apical surface exposed to the lumen
`of the gut, provides an opportunity for foods or other factors to interact directly
`with the CCK cell to stimulate hormone secretion. However, it is not known
`if food directly interacts with CCK cells to stimulate secretion. Substantial
`evidence has indicated that at least one mechanism causing CCK release in-
`volves endogenously produced releasing factors that are secreted into the lumen
`of the intestine and under the proper conditions stimulate CCK secretion (62,
`63). The rationale for their existence and description of at least two types of
`CCK-releasing factors are summarized below.
`Experiments in rats in the 1970s demonstrated that inactivation or removal
`of proteolytic activity from the proximal small intestine by either instillation
`of trypsin inhibitors or diversion of bile-pancreatic juice from the intestine
`produced a large increase in pancreatic exocrine secretion (64). In addition,
`intraduodenal infusion of trypsin in the absence of bile-pancreatic juice in the
`intestine suppressed pancreatic enzyme secretion (65). These observations
`suggested that a negative feedback mechanism existed whereby pancreatic se-
`cretion was controlled by the presence or absence of protease activity within
`the lumen of the small intestine.
`This concept has been used to explain how dietary proteins stimulate pan-
`creatic secretion. In rats, intact proteins are more potent stimulants of CCK
`release than hydrolyzed proteins or amino acids, suggesting that the ability of
`dietary protein to stimulate the pancreas is related to its ability to serve as a
`proteolytic substrate (66). In this regard, dietary proteins are similar to trypsin
`inhibitors when instilled into the gut because they both bind trypsin and cause
`pancreatic secretion.
`It was proposed that a trypsin-sensitive releasing factor is secreted into the
`lumen of the intestine and stimulates CCK release (67). In the presence of active
`proteases such as trypsin, this releasing factor would be degraded. However,
`when other proteins or potential substrates for trypsin are present in the gut
`(such as after eating a meal), the putative releasing factor would be protected
`and able to interact with CCK-secreting cells of the intestine (Figure 3).
`
`-228-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`229
`
`Figure 3 CCK release from enteric endocrine cells is regulated by lumenal intestinal CCK-
`releasing factors. It is proposed that both the pancreatic and intestinal peptides are secreted into the
`intestinal lumen and interact directly with the CCK cell through specific cell surface receptors to
`initiate a cascade of intracellular signaling events resulting in CCK release. Under basal conditions,
`trypsin has the potential to degrade the releasing factors. However, food (or trypsin inhibitor), which
`also serves as a substrate for trypsin, may temporarily bind trypsin and prevent the degradation of
`the intestinal-releasing factor or monitor peptide, thus allowing them to stimulate the CCK cell.
`(Modified from Reference 58.)
`
`There is now strong evidence that a feedback mechanism exists in the chicken,
`pig, rat, and human, and probably is present in the dog (64, 68–76). In humans,
`intraduodenal perfusion of phenylalanine not only stimulates pancreatic secre-
`tion, but also CCK release; these effects are blocked by simultaneous infusion
`of trypsin, indicating that negative feedback regulation of CCK is important in
`this species (77).
`The mechanisms regulating the production and secretion of an intestinal
`CCK-releasing factor are largely unknown. It is controversial whether cholin-
`ergic innervation is important for secretion of the releasing peptide. In anes-
`thetized rats, atropine prevents the spontaneous increase in pancreatic secretion
`caused by diversion of bile-pancreatic juice (78); however, in other studies in
`conscious rats, atropine has no effect on the increase in pancreatic secretion
`stimulated by intraduodenal infusion of either dietary protein or trypsin in-
`hibitors (79, 80) or on plasma CCK levels that are appropriately elevated in
`atropine-treated, conscious rats following bile-pancreatic juice diversion (81).
`
`-229-
`
`
`
`
`

`

`230
`
`LIDDLE
`
`Therefore, a precise role for cholinergic regulation of CCK-releasing factor
`secretion remains to be established.
`It has been demonstrated recently that
`somatostatin inhibits the elevation in plasma CCK levels normally seen in bile-
`pancreatic juice-diverted rats by inhibiting the release of a CCK-releasing factor
`(82).
`In
`Regulation of a CCK-releasing factor may also differ among species.
`humans, it is postulated that lumenal stimulation by nutrients such as amino
`acids or fatty acids is necessary to stimulate secretion of the putative CCK-
`releasing factor (69, 83, 84). This situation differs from the rat, in which
`spontaneous secretion of CCK-releasing factor occurs and maximal CCK and
`pancreatic secretion result from removing proteases from the proximal intestine.
`Very recently, two putative CCK-releasing factors have been chemically char-
`acterized. A novel 8169-kDa protein called lumenal cholecystokinin-releasing
`factor (LCRF) was purified from rat pancreatic juice (85). LCRF stimulates pan-
`creatic secretion and CCK release when instilled into the intestine. Moreover,
`this increase in CCK release is blocked by immunoneutralization with LCRF-
`specific antisera. It is possible that diazepam-binding inhibitor (DBI), which
`when extracted from porcine intestine was found to possess CCK-releasing ac-
`tivity, is a second CCK-releasing peptide (86). The relative contributions of
`LCRF and DBI to regulation of CCK secretion following ingestion of a meal
`remain to be determined.
`Characterization of a CCK-Releasing Factor from Pancreatic
`Juice: Monitor Peptide
`In 1986, a 61-amino acid protein purified from pancreatic juice was found to
`stimulate CCK secretion when introduced into the lumen of rat intestine (87).
`This protein was named monitor peptide for its ability to monitor the intraduo-
`denal environment for protein digestion (88, 89). Because of its similarity to
`other pancreatic trypsin inhibitors, monitor peptide is also known as pancreatic
`secretory trypsin inhibitor-I (PSTI-61); it is distinct from a 56-amino acid pan-
`creatic trypsin inhibitor known as PSTI-II (PSTI-56) that shares 66% sequence
`homology with monitor peptide but does not stimulate intestinal CCK release
`and pancreatic secretion (90).
`Monitor peptide is produced by pancreatic acinar cells and is secreted into
`pancreatic juice where, upon reaching the intestine, it stimulates CCK release.
`This mechanism for stimulating the CCK cell requires some level of pancreatic
`secretion to be effective. Therefore, once pancreatic secretion has been initiated,
`monitor peptide in pancreatic juice would tend to reinforce or perpetuate further
`secretion via release of CCK. In contrast, intestinal CCK-releasing factors differ
`from monitor peptide because they are of intestinal rather than pancreatic origin
`and are present in the gut under basal conditions (91). Lumenal-releasing factors
`
`-230-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`231
`
`are believed to mediate much of the CCK secretion that results from nutrient
`ingestion, although it is clear that multiple mechanisms exist to ensure proper
`regulation of CCK release.
`In isolated rat intestinal mucosal cells, monitor peptide stimulates CCK re-
`lease in a concentration- and calcium-dependent manner, suggesting that mon-
`itor peptide has a direct effect on CCK cells and that calcium may serve as an
`intracellular messenger in monitor peptide-mediated CCK secretion (92). The
`ability of monitor peptide to elevate [Ca2+]i and stimulate CCK secretion has
`been used to enrich CCK cells from the intestine by fluorescence-activated cell
`sorting (93). Enriched CCK cells respond to stimulation by monitor peptide,
`dibutyryl cAMP, membrane depolarizing concentrations of potassium chloride,
`and calcium ionophore.
`A putative cell surface receptor that binds monitor peptide has been partially
`characterized in intestinal mucosal cells of the rat jejunum (94). Autoradio-
`graphy of an affinity cross-linked complex of 125I-labeled monitor peptide and
`its binding site identified a potential receptor with a molecular mass of 33 or
`53 kDa in its reduced or nonreduced form, respectively, providing evidence
`that monitor peptide binds directly to receptors on enteric cells of the small
`intestine. A proposed scheme for the regulation of CCK by the two classes of
`lumenal-releasing factors is shown in Figure 4.
`Cellular Regulation of CCK Secretion
`The precise signaling pathways involved with CCK secretion are unknown.
`Studies in isolated canine intestinal mucosal cells enriched in CCK demon-
`strated that forskolin and cAMP analogues stimulate CCK release (95, 96).
`Moreover, forskolin-activated secretion was blocked by somatostatin. Stimu-
`lation of protein kinase C (PKC) by the phorbol ester, β-phorbol-12-myristate-
`13-acetate (PMA), also caused significant CCK secretion, as did exposure to
`L-phenylalanine.
`Recent investigations using the intestinal CCK-containing cell line STC-1
`have been useful for examining the receptors on CCK cells and the potential
`second messenger pathways involved in regulating secretion. The bombesin
`family of neuropeptides, including gastrin-releasing peptide (GRP), have potent
`CCK stimulatory effects in animals (29, 97–99) and in organ preparations (100).
`Although presumed to be a direct effect, the identification of GRP receptors in
`CCK-secreting cells (by radioligand binding and mRNA analysis) established
`that CCK cells possess bombesin-like receptors (101). As a G protein-coupled
`receptor that couples to and signals through the phosphoinositide cascade, gen-
`erating inositol trisphosphate (IP3), bombesin causes an increase in intracellular
`calcium concentration ([Ca2+]i) (101). It is believed that this [Ca2+]i increase
`is important for evoking CCK secretion because hormone release is reduced
`
`-231-
`
`
`
`
`

`

`232
`
`LIDDLE
`
`Figure 4 Proposed model for intracellular signaling in the CCK cell. Because both adenylyl
`cyclase and phosphoinositide cascades exist in CCK cells, it is possible that potential secretagogues
`such as bombesin and intralumenal-releasing factors bind to cell surface receptors and initiate
`a series of signaling events to ultimately cause CCK secretion. Secretion may result through
`(a) activation of phospholipase C-β with subsequent generation of IP3 and diacylglycerol (DAG)
`and subsequent increases in [Ca2+]i and activation of PKC, respectively; (b) activation of adenylyl
`cyclase with subsequent elevation in cAMP concentrations and activation of protein kinase A
`(PKA); or (c) some yet unknown pathway, perhaps involving the activation of calcium channels,
`that appears to be critical for sustained CCK secretion. The locations of the putative receptors
`for LCRF, monitor peptide, and GRP are portrayed on the lumenal and basolateral surfaces of the
`CCK cell, respectively, solely because the corresponding ligands are located in either the intestinal
`lumen or enteric nerves. A calcium influx pathway can be affected by multiple second messengers.
`(Modified from Reference 58.)
`
`-232-
`
`
`
`
`

`

`CHOLECYSTOKININ CELLS
`
`233
`
`by exposure to calcium-free media or treatment with L-type calcium channel
`blockers (59).
`Although maneuvers that increase intracellular cAMP have been shown to
`stimulate CCK release, little is known about the mechanisms for increasing
`cAMP levels in CCK cells. It was recently demonstrated, using STC-1 cells, that
`β-adrenergic receptors on CCK cells effectively coupled to the production of
`cAMP, resulting in CCK release (102). Interestingly, the β-adrenergic agonist
`isoproterenol increased [Ca2+]i, coincident with stimulating hormone secretion.
`Both CCK secretion and the elevation in [Ca2+]i were reduced by treatment with
`the L-type calcium channel blocker diltiazem, indicating that β receptors on
`CCK cells are coupled to the production of cAMP, which may stimulate CCK
`release through a calcium-dependent process.
`The manner by which dietary stimulants affect CCK secretion are largely
`unknown. A study of isolated intestinal cells in vitro indicates that dietary
`proteins which cause CCK release in animals do not have stimulatory effects
`in perfusion models (103). These findings are consistent with the proposal that
`effects of protein on CCK release are mediated by an intestinal-releasing factor.
`However, in native CCK cells and STC-1 cells, L-phenylalanine stimulated CCK
`release and increased cytosolic calcium levels (in a dose-dependent manner)
`(104). Both effects were blocked by diltiazem, indicating that phenylalanine
`stimulates release of CCK by a calcium-dependent process involving a calcium
`influx pathway in the cell.
`The results of these experiments emphasize that influx of extracellular cal-
`cium from outside to inside the cell is a common feature of CCK secretion
`initiated by a variety of mechanisms. The critical role of [Ca2+]i seems to apply
`whether the inciting event is due to monitor peptide, bombesin, isoproterenol,
`or phenylalanine stimulation. The relationship between intracellular cAMP
`and calcium signaling pathways becomes more apparent when it is observed
`that phosphodiesterase inhibitor IBMX not only stimulates CCK release in
`STC-1 cells, but also increases cytosolic calcium (105). Moreover, the calcium-
`calmodulin kinase II inhibitor KN-62 markedly reduces IBMX-stimulated se-
`cretion and [Ca2+]i, suggesting that cAMP may activate diltiazem-sensitive
`calcium channels by a calmodulin-dependent process. This conclusion is sup-
`ported by the finding that the stimulatory effects of forskolin on CCK release
`are similarly blocked by KN-62 (106). These studies provide evidence that
`both the adenylyl cyclase and phosphoinositide signaling cascades are active
`in CCK cells and may be coupled to ligand occupation of cell surface recep-
`tors to initiate intracellular events leading to CCK secretion (Figure 4). The
`two signaling cascades are linked by calcium-calmodulin kinase II activity and
`regulation of intracellular calcium concentration.
`
`-233-
`
`
`
`
`

`

`234
`
`LIDDLE
`
`Role of Ion Channels in CCK Secretion
`In many endocrine cells, hormone secretion is coupled to changes in ion channel
`activity. One of the most apparent relationships is the dependence upon plasma
`membrane potential. Secretion of insulin, thyrotropin, growth hormone, pro-
`lactin, calcitonin, secretin, somatostatin, and cholecystokinin are all induced by
`solutions containing high potassium concentrations, indicating that the resting
`potential of these secretory cells is at least partially determined by the activity
`of potassium channels (107–111).
`Models for studying ion channel activity and endocrine secretion have in-
`cluded tissue preparations, dispersed cells, and clonal popu